Report on Bioavailability of

                  *••
              •'-  •••

                               >

-------
  Report on Bioavailability of
Chemical Wastes With Respect
    to the Potential for Soil
        Bioremediation
          Eugene L. Madsen, Ph.D.
     Department of Microbiology, Cornell University
              Ithaca, NY

-------
   Although the information in this document has been funded by the U.S. Envi-
ronmental Protection Agency under contract number T28006: QT-DC-99-003260 to
Cornell University, it does not necessarily reflect the views of the Agency and no
official endorsement should be inferred.

   This document has been reviewed by a panel of five external experts: Dr. Kevin
Johnson of Southern Illinois University; Dr. Todd Anderson of Texas Tech Univer-
sity; Dr. Katherine Banks of Purdue University; Dr. Teresa Fan of University of
California, Davis; and Dr. James Shine of Harvard University.
   Acknowledgments

   The author is grateful for the support of the U.S. Environmental Protection
Agency and Dr. Robert Menzer.  During preparation of this report, insightful dis-
cussions with P. Baveye, M. Alexander, J. Pignatello, S. Hawthorne, W. Ball, J.
Kreitinger, and input from four anonymous reviewers occurred. Remarkable pa-
tience and expertise in manuscript preparation was contributed by Patti Durfee.

-------
                             Contents

Executive Summary	1
1     Introduction: Scope and Goals	3
2     Definitions of Biodegradation and Bioremediation, and the Many
      Facets of Bioavailability	 5
      2.1  Biodegradation	 5
      2.2  Bioremediation	5
      2.3  The Susceptibility of Inorganic Materials, Including Metals,
          Nonmetals, and Radionuclides, to Microbe-Based Bioremediation	6
      2.4  Summary	 10
3     Bioavailability	  12
      3.1.  Defining Bioavailability: Nine Definitions	 12
      3.2  Biosensor Technology Offers a Means Towards Direct
          Measurement of Bioavailability	 19
      3.3  Summary	 20
4     A Survey of Field Projects Using Bioremediation To Treat
      Contaminants in Soils	  22
5     Mechanisms of Persistence of Organic Compounds	  31
      5.1  An Evolutionary Perspective on the Persistence of
          Organic Compounds	 31
      5.2  Placing "Bioavailability" Within the Established Framework of
          Persistence Mechanisms	 32
      5.3  Illustrating Mechanisms of Persistence: Soil Organic
          Matter(SOM)	 34
          5.3.1  Summary (Soil Organic Matter Persistence)	 35
6     Paradigms for the Composition and Structure of Soil and the Physical-
      Chemical State of Contaminants Therein 	  37
      6.1  Soil Complexity	38
      6.2  A Thermodynamic Overview of Inorganic Soil Reactions	 38
                                   in

-------
IV
           6.3   Models of Soil Structural Characteristics and Methodological
                Limitations in Environmental Microbiology	39
           6.4   Interactions Between Geosorbents and Contaminants	42
                6.4.1   Inferences of Geosorbent Structure Based on Indirect
                      Observations of Hydrophobic Organic Compounds	 42
                6.4.2.  Direct Observations of Contaminated Geosorbents	 45
                      6.4.2.1   Hydrophobic Organic Compounds	 45
                      6.4.2.2   Inorganic Contaminants	 47
           6.5   Summary	 48
           Uptake of Soil Constituents by Plants and Microorganisms	  50
           7.1   Principles of Soil Solution Chemistry and Uptake by Plants of
                Inorganic Compounds	50
           7.2   Movement of Solutes From Soil Solution to Roots	 51
           7.3   Examples of Nutrient Uptake by Plant Roots	52
                7.3.1.  Phosphorus (P)	 53
                7.3.2.  Iron(Fe)	 53
           7.4   Contrasts Between Uptake Mechanisms in Microorganisms
                and Plants	 54
           7.5   Membrane Transport in Microorganisms	 56
           7.6   Uptake of Insoluble Organic Substrates by Microorganisms: Wood.. 56
           7.7   Phosphorus and Iron Uptake by Bacteria	57
           7.8   Summary	58
           Reviewing the Facts: Examining Relationships Between Contaminant
           Sequestration and Bioremediation	60
           8.1   Ambiguity Is the Rule: AnHistorical Overview of the Impact
                of Solid Surfaces on Microbial Activity	60
           8.2   Selection and Justification of Criteria for Identifying the Highest
                Quality Investigations Pertinent to Bioavailability and
                Biodegradation	62
                8.2.1   Further Scrutiny of Artifacts That May Be Caused by Soil
                      Sterilization and Contaminant Addition in Organic Solvents  ... 67

-------
      8.3  Scrutinizing Selected Investigations Describing the Bioavailability
          of Contaminants and Their Biodegradation	 69
          8.3.1.  Summary	 94
      8.4  Influence of Bioavailability on Phytoremediation of Metal-
          Contaminated Soils 	 95
      8.5  A Synthesis: Evaluating the Relationships Between Bioavailability
          andBioremediationBasedonSections2to8.4ofThisReport	 96
9     Overcoming Constraints on Site Cleanup	  99
10    Conclusions, Implications, and Possible Areas of Future Research	 101
11    Literature Cited	 105

-------
VI

-------
                  EXECUTIVE SUMMARY
    Based on conservative, reasonably thorough and careful evaluation of scientific
studies described in this report, there is no doubt that chemical wastes in soil can be,
and often are, in a state of reduced bioavailability.  An analysis of the literature on
bioremediation research concludes that bioremediation of chemical wastes in soils
and sediments is rarely 100 percent efficient, due at least in part to the reduced bio-
availability of the chemical.  Reduced bioavailability simply means that a chemical
waste's diminished "effective concentration" is proportionately balanced by a lin-
gering reservoir of the chemical waste in soil and sediments. This lingering reservoir
remains in the soil habitat regardless of which combinations of conceptual or actual
sequestration mechanisms (e.g., complexation into bound residues, diffusion into
soil pores, NAPL partitioning) apply.

    Soil is, by definition, a  thermodynamically unstable, kinetically constrained
medium whose chemical composition, (solid, liquid, and gaseous phases) is con-
stantly changing. Thus, the  "nonbioavailable" chemical wastes in this lingering
reservoir are always subject to release into soil  solution where the wastes are
resubjected to a variety of transport and/or transformation processes (e.g., immo-
bilization, biodegradation, uptake by receptors).

    Considerable effort has been expended in investigating the hypothesis that
chemical wastes have diminished bioavailability in soils and sediments. Results of
these efforts have been ambiguous because of the immense diversity in types and
properties of chemical wastes, geosorbents, biota, experimental approaches, and
the idiosyncrasies in mechanisms by whichbiota interact with chemical wastes. In
this report, 17 studies were  selected from the vast literature on bioavailability to
illustrate the range in quality of reported research. Methodologies were  analyzed,
and conclusions were critiqued. From this analysis, criteria for identifying the high-
est quality investigations pertinent to bioavailability are proposed.  These are
based on the realism and environmental relevance of the study, the absence of
experimental artifacts, and the consistency of the results.

    An accurate estimate of risks to human and environmental health posed by
chemical wastes in soils is a crucial step toward:  (1) identifying pragmatic, eco-
nomically feasible environmental cleanup goals; (2) establishing operational defi-
nitions of "treatment" by bioremediation technology; (3) realistically classifying

-------
polluted sites based on planned land-use scenarios; (4) developing public accep-
tance of risk-based contaminant cleanup efforts; (5) developing public acceptance
of cleanup goals that are above the "original, pristine state" of the contaminated
site; and (6) legitimizing the concept of "environmentally acceptable endpoints."

    From a practical, regulatory point of view, establishing the foundation of re-
duced bioavailability is crucial. If the reduced bioavailability of chemical wastes in
soil becomes widely accepted, then proper quantitative measures of bioavailability
reduction could be developed to accurately estimate the risks posed by chemical
wastes in soils and sediments to human health and ecological processes.

-------
                            Section 1

          Introduction:   Scope and Goals
    Commercial, industrial, and military activity, largely in the 19th and 20th centuries,
have led to environmental contamination problems that can threaten human health
and ecosystem function (Alexander, 1999;Madsen, 1998b; National Research Coun-
cil, 1997; Young and Cerniglia, 1995). The degree to which such contaminants are
available to biota has far-reaching scientific, lexicological, and regulatory implica-
tions (e.g., Lee etal., 2000). Bioremediation, the use of organisms (especially naturally
occurring microorganisms or plants) to eliminate these pollution problems has im-
mense potential for meeting society's pollution-control needs (Alexander, 1999;
Madsen, 1998b; National Research Council, 1993,1997; Schwarzenbach etal., 1999;
VanderLelie etal., 2001; Young and Cerniglia, 1995). However, in recent years, evi-
dence has begun to accumulate suggesting that pollution-eliminating processes may
be thwarted by a variety of physical, chemical, or other spontaneous reactions that
diminish the availability of environmental contaminants for uptake and transforma-
tion by biota (Alexander, 1995,1999).

    This report critically evaluates current knowledge of the relationships between
the bioavailability of chemical wastes and their susceptibility to bioremediation.
The context of these relationships is that all terrestrial and aquatic habitats are
located in either the surface or subsurface portions of the landscape. However, due
to this report's focus, soil habitats will be emphasized. Although model systems
using laboratory  incubations of environmental  samples or cultures of individual
microorganisms provide insights about these relationships, the primary focus of
this report is the behavior of chemical wastes and naturally occurring (or inocu-
lated) microorganisms in situ — in real-world contaminated field sites. When
appropriate, information about the nascent field of phytoremediation (plant-medi-
ated decontamination processes) will be included. The chemical wastes of interest
are those that confront society today. These include organic (e.g., petroleum hy-
drocarbons, pesticides, chlorinated  solvents) and inorganic (e.g., metals, radionu-
clides, oxyanions) compounds.  This  report will not attempt to comprehensively

-------
survey the behavior of all contaminants. Instead, compounds will be selected from
both categories (organic and inorganic) to reveal how microorganisms and plants
interact with chemical wastes. There is no doubt that, under favorable conditions,
many contaminants and microorganisms or plants can interact in ways that suc-
cessfully lead to biodegradation and bioremediation. The issue addressed by this
report is how bioavailability influences the degree to which the bioremediation
outcome is successful (according to the standards of regulatory agencies and
society).

    This report is divided into  10 sections, including this introduction. Section 2
will begin by defining biodegradation and bioremediation; because of their novelty,
fundamental  microbial reactions that govern the fate of inorganic materials are
reviewed in detail. Section 3 defines bioavailability. Next, selected case studies of
completed bioremediation field projects are presented—seeking evidence for a
reduction in the efficacy of organic- and inorganic-contaminant bioremediation
that may have been caused by bioavailability limitations. To interpret how and why
bioremediation may be inefficient, Section 5 reviews information about the mecha-
nisms by which organic compounds persist in soils. In Section 6, paradigms for
geosorbents and the physical and chemical state of associated contaminants  in
field sites will be summarized. Section 7 reviews physiological principles describing
the uptake of sequestered organic and inorganic chemicals by microorganisms and
plants. Assisted by recent scientific reviews, Section 8 critically evaluates the meth-
odologies that have  been used to generate  information about the influence  of
bioavailability on bioremediation. Section 9 examines studies designed to enhance
the efficiency of remediating organic compounds in soil by alleviating bioavailability
limitations. The report concludes (Section 10) with recommendations about new
frontiers and areas of future research.

    This document is heuristic—it strives to synthesize new combinations of infor-
mation of the highest quality from a variety of sources. The sponsor of this report (the
US EPA) prescribed that it be a concise compilation of current knowledge, not an
encyclopedic review. In compiling and processing information for this report, difficult
decisions were made to include some while excluding other scientific studies. The
author apologizes for any omissions of crucial pertinent scientific results. Through-
out this report, selected portions of the text will be highlighted in bold because they
figure prominently in the logic that will lead to this report's conclusions.

-------
                          Section 2


        Definitions of Biodegradation and

   Bioremediation, and the Many Facets of

                       Bioavailability



2.1   Biodegradation

   "Biodegradation" is the partial simplification or complete destruction of the
molecular structure of environmental pollutants by complex, genetically regulated
physiological reactions catalyzed largely by microorganisms (Alexander, 1999;
Madsen, 1991;Madsen, 1998b; Young and Cerniglia, 1995); and plants (Bhadrarf
al, 1999;Bizily etal, 1999;BurkenandSchnoor, 1998; Siciliano and Germida, 1998).
Many of these reactions are predictable based on established laws of thermody-
namics applied under environmental geochemical conditions (e.g., temperature,
pressure, and oxidation-reduction potential) that prevail in terrestrial and aquatic
habitats (Madsen, 1998a). Microbial biodegradation is routinely measured by ap-
plying chemical and physiological assays to laboratory incubations of flasks con-
taining pure cultures of microorganisms, mixed cultures, or environmental samples
(soil, water, or sediment; Madsen, 2002). Plant-mediated biodegradation and/or
contaminant accumulation also is routinely documented in appropriately scaled
laboratory test systems (Bhadrae/a/., 1999; Bizily etal., 1999; VanderLelie etal.,
2001; Vangronsveld and Cunningham, 1998).


2.2   Bioremediation

   "Bioremediation" is the intentional use of biodegradation or contaminant-
accumulation processes to eliminate environmental pollutants  from sites where
they have been released. Bioremediation technologies use the physiological po-
tential of microorganisms and plants, as documented most readily in laboratory
assays, to eliminate or reduce the concentration of environmental pollutants in field

-------
sites to levels that are acceptable to site owners and/or regulatory agencies (Burken
and Schnoor, 1998; Madsen, 1998b; National Research Council, 1997; Siciliano and
Germida, 1998). Bioremediation may be approached using in situ technology (ap-
plied directly to contaminated sites) or ex situ bioreactor methodologies (after
contaminants and/or accompanying soil, sediment, or water are removed from con-
taminated sites) (Crawford and Crawford, 1996; Flathmane/a/., 1994; Hinchee and
Olfenbuttel, 1994; Madsen, 1998b;Norrise/a/., 1994;Rittmane/a/., 1994; Skipper
and Turco, 1995; Siciliano and Germida, 1998). The open, complex, poorly defined
settings where bioremediation occurs and the often wide array of possible contami-
nant attenuation pathways may obscure the precise mechanisms that operate dur-
ing field bioremediation operations  (Madsen, 1991; National Research  Council,
1993; National Research Council, 2000).

    Phytoremediation is a technology that seeks to use several  plant-based pro-
cesses to remove, transfer, stabilize, or destroy contaminants in soil, sediment, and
groundwater (Cunningham et al., 1997; Ensley, 2000; EPA, 2001; Van der Lelie et al.,
2001; Vangronsveld and Cunningham, 1998). The mechanisms of phytoremediation
include rhizosphere bio degradation, phytoextraction, phytodegradation, and
phytostabilization. Phytoremediation strategies can use long-lived perennial spe-
cies in a stable system or short-lived annual species that are repeatedly planted and
harvested.


2.3   The Susceptibility of Inorganic Materials, Including
      Metals, Nonmetals, and Radionucleotides, to Microbe-
      Based Bioremediation

    Biological processes can influence inorganic environmental contaminants (Babu
etal, 1992;Brierley, 1990; Chapatwalarfa/., 1995; Hinchee etal,  1995;Kalinetal.,
1991;Lenhardetal., 1995;Lovley, 1993,1995a,b;Lovley,2000;McHaleandMcHale,
1994; Saouteretal., 1995; Summers, 1992; Thompson-Eagle andFrankenberger, 1992;
Videla and Characklis, 1992; and Whitlock, 1990). Unlike organic compounds that are
often susceptible to partial structural alteration or complete detoxification to carbon
dioxide by microorganisms, the majority of inorganic contaminant compounds are
subject only to changes in speciation that may influence contaminant mobility. Mi-
croorganisms or plants may cause precipitation, volatilization, sorption, and solubili-
zation of inorganic compounds. Mechanistically, these reactions can be the result of
direct enzymatic processes such as oxidation, reduction,  methylation, or uptake.
Reaction mechanisms also can be indirect (nonenzymatic)—resulting from produc-
tion of metabolites or biomass that can strongly influence the behavior of inorganic
contaminants via redox, acid/base, coprecipitation, sorption, and other geochemical
means.  In many phytoremediation scenarios, contaminant-attenuation mechanisms
are driven by transpiration-based hydrodynamic interception of contaminants in soil
solution and groundwater. Direct physiological transformation of inorganic com-

-------
pounds by plants is an area of current research, as is plant-microbe interactions in the
rhizosphere. Because the science and technology of microbial processes that influ-
ence inorganic compounds are well developed, microbial-, and not plant-based,
bioremediation is emphasized below. However, some microbial reaction types may
apply to plants, and certainly apply to the rhizosphere.

    One nearly  universal means by which microorganisms have been shown to
reduce concentrations of inorganic contaminants in water (e.g., Cu, Ni, Zn, Cd, Pb)
is by immobilizing aqueous-phase inorganics in microbial biomass and/or microbial
exopolymers (Diels, 1997; Macaskie andBasnakova, 1998). The mechanisms range
from nonspecific  electrostatic sorptive interaction between cationic metals  and
anionic extracellularpolysaccharides (Williams et al., 1998) to highly specific active
transport systems that cause metals to accumulate in high concentrations within
microbial cells (Chen and Wilson, 1997). The utility of these sequestration reactions
for soil is in doubt but they are promising in engineered wastewater treatment
systems where metal-laden water flows over fixed biofilms that can be removed
from the treatment system so that the toxic inorganics can be recovered.

    Many inorganic contaminants, especially metals, become relatively soluble or
mobile at low pH. In contrast to the various bioremediation approaches that rely on
immobilization reactions, the opposite (washing soils and sediments free of inorganic
contaminants) can theoretically be achieved by directing low pH waters through
contaminated sites. The acidification step can be mediated by a variety of microbial
processes that include the oxidation of elemental sulfur (Lovley and Coates, 1997).

    The often highly abundant nontoxic metals, iron and manganese, exist in re-
duced and oxidized states. The oxidized states [Fe(III), Mn(IV)] react chemically to
form oxyhydroxide precipitates that serve as physiological electron acceptors for
anaerobic microbial food chains (Lovley, 1995a; 2000). The end products of Fe- and
Mn-reduction [Fe(II) and Mn(II)] are relatively soluble and may migrate to aerobic
habitats, where reoxidation and precipitation also canbe catalyzed by microorgan-
isms (Ghiorse, 1994). The behavior of many of the toxic metals discussed below is
intimately tied to the microbially mediated cycling of Fe and Mn because the toxic
metals may be immobilized (through coprecipitation and sorptive reactions with
many Fe and Mn oxides) or solubilized [by being reduced via chemical reactions
with Fe(II) and Mn(II)] (Martinez and McBride, 2001). Thus, most problem inor-
ganic compounds (e.g., Pb, Cr, U, Ni, Hg, Cd, Sr) undergo immobilization reactions
via sorption and precipitation.

    Chromium is a metal whose key oxidation states are (VI) and (III) (NRC, 2000). In
aqueous environments, chromium (VI) predominates as the mobile and highly toxic
anions, chromate (CrO42") and dichromate (Cr2O72")- Reduced chromium (III) is less
toxic and less mobile because of precipitation reactions as oxides and hydroxides at
pH 5 and above. A variety of both aerobic and anaerobic microorganisms have been
shown to enzymatically reduce Cr(VI) to Cr(III), but the physiological reasonforthis

-------
ability has not been adequately investigated. Among the hypotheses explaining the
reduction reactions are: survival (i.e., detoxification), cometabolism (i.e., fortuitous
enzymatic reactions), and the use of Cr(VI) as a physiological electron acceptor (to
date, only equivocal evidence for the latter hypothesis has been obtained). Direct
microbial detoxification (reduction) of Cr(VI) is unlikely to be a useful remediation
technology in anaerobic subsurface habitats because the reduction occurs sponta-
neously in the presence of sulfide, Fe(II), and some organic compounds. Although
microbial production of sulfide, Fe(II), and reduced organic compounds is generally
reliable, additional research is required before judging if Cr(VI) reduction has the
potential to serve as a useful bioremediation tool (NRC, 2000).

    Mercury is a toxic metal whose predominant forms include mercuric ion [Hg(II)],
elemental mercury [Hg(O)], and the biomagnification-prone organic mercury com-
pounds, monomethyl- and dimethyl-mercury (NRC, 2000). All transformations of
mercury by microorganisms are considered detoxification reactions that are in-
tended to mobilize mercury away from microbial cells. Most reactions are enzy-
matic, carried out by both aerobes and anaerobes, and involve uptake of Hg(II)
followed by its reduction to volatile forms (elemental Hg(0), methyl-, and dimethyl-
mercury). Mercuric ion [Hg(II)] also forms highly insoluble precipitates with sul-
fide; thus, one indirect microbial detoxification strategy involves the stimulation of
sulfate-reducing microorganisms. Although engineered bioreactors that first re-
duce the mercuric ions and then purge the volatile Hg from water have been de-
signed, no successful applications of this technology  have yet been implemented
in soil environments (NRC, 2000).

    In addition to mercury, microorganisms are capable of methy lating other metals
(Cd,Pb, Sn,Te, Se;Ehrlich, 1996; Semethylation is discussed in detail below). Addi-
tional methy lation reactions may occur as a result of nonbiological transmethylation
by microbially produced methylated donor compounds such as trimethyl tin (Ehrlich,
1996). These donors may react with ionic forms of Pd, Th, Pt, and Au, but the resultant
reduced metals may not be chemically stable. The significance of these unusual metal
methy lation reactions for bioremediation is unknown.

    Arsenic is a toxic element capable of existing in five different valence states [As
(-III), (0), (II), (III), and (V)]. Forms of arsenic range from sulfide minerals (e.g.,
As2S3) to elemental As to arsenic acid to arsenite (AsO2") to arsenate (AsO43") to
various organic forms that include methylated arsenates and trimethyl arsine. Clearly,
the chemical and microbiological reactions of As are complex  (Ehrlich, 1996;
Frankenberger and Losi, 1995). Both anionic forms (arsenite and arsenate) are highly
soluble and highly toxic—interfering with various enzyme functions and oxidative
phosphory lation, respectively. All forms of As are toxic. Microorganisms transform
arsenic for several fundamental physiological reasons: (1) under anaerobic condi-
tions, arsenate [As(V)] can be used as a final electron acceptor; (2) under aerobic
conditions, reduced As (e.g., arsenite) oxidation had been shown to generate en-
ergy (ATP); and (3) under both anaerobic and aerobic conditions, As can be detoxi-

-------
fiedby methylation, oxidation, or reduction mechanisms that mobilize As away from
microbial cells. Engineered bioremediation strategies that rely on mobilizing methy-
lated As from water have been implemented (Frankenberger and Losi, 1995).

    Although selenium is an important and beneficial micronutrient for plants, ani-
mals, humans, and some microorganisms (largely because of its role in some key
amino acids), this element can be toxic at greater than trace concentrations.  In
natural environments, selenium has four predominant inorganic species: Se(VI)
(selenate, SeO42'), Se(IV) (selenite, SeO/'), Se(0) (elemental selenium), and Se(-II)
(selenide) (Ehrlich, 1996; Lovley, 1995b; Frankenberger and Losi, 1995). Like ar-
senic, selenium also  has many volatile organic forms; these include:  dimethyl
selenide, dimethyl diselenide, methane selanone, methane selenol, and dimethyl
selenyl sulfide. Each of these compounds exhibits its own chemical and biochemi-
cal behavior, mobility, and toxicity. The various forms of selenium are transformed
by microorganisms according to their physiological needs and the ambient thermo-
dynamic conditions. Reduced inorganic selenium compounds have been shown to
be oxidized under aerobic conditions, though not linked to microbial growth. Oxi-
dized selenium (selenate) can serve as a final electron acceptor for anaerobic micro-
organisms—resulting in production of selenide and/or elemental Se.  Methylation
of the various Se compounds is thought to be a protective detoxification mecha-
nism that mobilizes Se away from microbial cells. Thus like As, the environmental
fate of Se is governed by complex interactions between chemical and physiological
processes. For instance, anaerobic microbial reduction of selenate and selenite to
insoluble elemental selenium represents an important mechanism for immobilizing
and removing Se from aqueous solution. Furthermore, the various volatile methy-
lated forms of Se are  sufficiently mobile that aerobic deselenification (largely via
dimethylselenide formation) of highly contaminated California soils has been dem-
onstrated in field experiments (Frankenberger and Losi, 1995).

    Oxyanions are water soluble, negatively charged chemical species in which a
central atom is surrounded by oxygen. Nitrate (NO3") is a naturally occurring oxyanion
commonly found at low concentrations in soils, sediments,  and both surface and
groundwaters as a result of the biogeochemical cycling of organic matter. Nitrate is
a readily utilizable form of nitrogen that can be assimilated into amino acids by
plants and microorganisms (NRC, 2000). Although serving to supply nitrogen, an
essential nutrient, nitrate also is a serious health concern for at least two reasons:
(1) it can be a chemical or microbiological precursor for nitrite, which spontane-
ously combines with amino compounds to form highly carcinogenic nitrosamines;
and (2) nitrate can be  reduced to nitrite in stomachs of infants, which can cause the
respiratory stress disease, methemaglobinemia.

    Nitrate is produced from ammonia by nitrifying microorganisms under aerobic
conditions. The major microbial process that destroys nitrate is dissimilatory reduc-
tion to dinitrogengas (denitrification). Genetic, biochemical, physiological, ecologi-
cal, environmental, and engineering aspects of denitrification (Zumft, 1997; Atlas and

-------
10
     Bartha, 1997) have been examined for decades. Nitrate is used as a physiological
     electron acceptor under oxygen-free (anaerobic conditions). The denitrification pro-
     cess is widespread among microorganisms, and occurs reliably in every anaerobic
     habitat with abundant carbon and electron sources. Denitrification may not be appli-
     cable to surface soil because of the presence of O2 and the tendency of nitrate to
     leach. However, denitrification is a well-established bioremediation process and is
     used routinely in sewage treatment plants to curb eutrophication.

         The oxyanions chlorate (C1O3") andperchlorate (C1O4") or their precursors (chlorine
     dioxide, hypochlorite, and chlorite) are produced by a variety of paper manufacturing,
     water disinfection, and both aerospace and defense industries. Although not naturally
     occurring, these highly oxidized forms of chlorine are energetically very favorable physi-
     ological electron acceptors for microorganisms. Compared to denitrification, knowl-
     edge of chlorate and perchlorate biodegradation reactions is quite limited. However,
     laboratory studies using both pure bacterial cultures and environmental samples (soil,
     freshwater sediments, and sewage) have shown that in the presence of common elec-
     tron donors (carbohydrates, carboxylic acids, amino acids, evenH2 andH2S), microor-
     ganisms can grow at the expense of perchlorate and chlorate; thus reducing them to the
     nontoxic chloride ion (Malmqvistetal.,  1991;vanGinkeletal., 1995;Rikkonetal., 1996).
     Furthermore, a bioreactor has recently been engineered to successfully convert chlor-
     ate and perchlorate to chloride (Wallace etal,. 1997).

         Uranium is a radioactive element that releases alpha, beta, and gamma radiation that
     can be toxic. Uranium can exist in the oxidation states of (0), (III), (IV), (V), and (VI),
     though in nature insoluble U(IV)O2 and soluble U(VI)O22+ predominate. Recently, U(VT)
     has been shown to serve as a terminal electron acceptor for anaerobic microorganisms;
     thus, in anaerobic habitats growth-linked reduction (hence immobilization) of U should
     be a reliable process (Lovley, 1995a). Although the reverse process, microbial oxidation
     of U(IV) to U(VT) under aerobic conditions has been demonstrated, this process has not
     been shown to be physiologically beneficial to the responsible microorganisms. Direct
     chemical oxidation of U(TV) by molecular oxygen [creating U(VT)] may also influence the
     robustness of U bioremediation strategies (NRC, 2000).  Another radioactive element,
     plutonium (Pu) also has been shown to be susceptible to microbial transformation. Iron-
     reducing microorganisms were found to reduce insoluble Pu4+ to the more soluble Pu3+;
     thus, rendering the soluble form more susceptible to mobilization (NRC, 2000). These
     microbially mediated oxidation/reduction reactions provide tools  for emerging
     bioremediation strategies for U, Pu, and other radionuclides.


     2.4   Summary

         Biodegradation is a complex series of metabolic processes that simplify the molecu-
     lar structure of organic compounds. Biodegradation is catalyzed largely by microorgan-
     isms. Plants also can cause biodegradation reactions, but  they  are more suited for
     uptake and accumulation reactions.  Inorganic contaminants (metals, non-metals,

-------
                                                                                   11
oxyanions, and radionuclides) cannot be biodegraded, but their environmental mobility
can be altered through oxidation-reduction, sorption, methylation and precipitation
reactions mediated by microorganisms or plants. Each inorganic pollutant features a
unique set of direct and indirect biotic and abiotic reaction pathways that may be
exploitable by bioremediation technologies. Biodegradation, accumulation, and an al-
tered mobility of contaminants can be rigorously documented in laboratory experi-
ments. These processes are the basis for potential site cleanup technology.

-------
                            Section 3


                         Bioavailability



3.1   Defining Bioavailability:  Nine Definitions

    Microorganisms and plants are the catalysts that effect bioremediation. But, as
discussed in Section 1.0, the "bioavailability" of chemical waste in soils, sediments,
and waters may influence, even regulate, bioremediation efficacy.  The term
"bioavailability" means many things in many contexts. This section compiles defi-
nitions of bioavailability from nine recent authoritative sources. The quotes that
appear below are intentionally extensive to reveal both the many facets of the term
"bioavailability" and the biases of the sources.

    The first definition (Southgate et al., 1989) is taken from a 1988 conference,
"Proceedings of Bioavailability 88:  Chemical and Biological Aspects of Nutrient
Bioavailability," organized by the Food Group of Royal Society of Chemistry, The
Working Party on Food Chemistry of the Federation of European Chemical Societ-
ies, and the Federation of European Nutritional Societies.

          "Until relatively recently, it has been a common assumption that
      the nutritional value of foods and diets was more or less synony-
      mous with their nutritional composition as determined by chemical
      analysis. The limitations of this simple picture first became obvious
      in relation to trace element nutrition, but recently many  lines of re-
      search have emphasized the  need to understand and quantify the
      intestinal absorption and  subsequent metabolism of all the major
      nutrient classes...

          Bioavailability has been defined as the proportion of the nutri-
      ent that is digested, absorbed,  and metabolized through normal path-
      ways.. . This definition has two important corollaries. First (bio) avail-
      ability is not a property of the diet of food per se but the response of
      the individual to the diet or food.  Second, observed (bio)availability,
                                   12

-------
                                                                               13
       therefore, represents an integration of the processes whereby an
       ingested nutrient becomes available..."

    This definition was selected because the authors were scholars concerned with
human nutrition—they were not concerned with the scientific and/or political is-
sues of the behavior and fate of environmental pollutants. Yet, the nutritional es-
sence of "bioavailability" establishes themes that will echo throughout this report:
(1) chemical assays of inorganic and organic materials may overestimate what is
actually biologically absorbed and metabolized; and (2) bioavailability is not an
inherent property of substances under examination, rather bioavailability reflects
the response of a biological system to many integrated processes.

    The second definition (Alexander, 1999), from a recent textbook onbiodegrada-
tion and bioremediation, makes the abrupt transition from human nutrition to pol-
lutant chemicals in soils and sediments. Chemically extractable pollutant compounds
are contrasted with biologically available pollutant compounds, as assessed with
biodegradation assays.  Five theoretical mechanisms are advanced, which suggest
that reduced bioavailability of organic soil pollutants for microbial biodegradation
is caused by  reactions between organic pollutants and the constituents of soils
and sediments.

          "The availability of many chemicals is affected by a series of ill-
       defined, often uncharacterized processes. In some of these processes,
       the compound is readily evident and can be easily removed from the
       soil, sediment, or aquifer by conventional extraction procedures. The
       evidence for reduced bioavailability of these compounds is the
       marked decline in the rate of biodegradation. In other processes, the
       compound is still present, but can only be removed from the envi-
       ronmental  sample by highly vigorous extraction techniques.  The
       evidence for reduced bioavailability of such a compound is the marked
       decline in the rate of biodegradation with time or the almost complete
       resistance of the molecule to microbial destruction...

          At this stage in the development of knowledge of a subject that
       is still characterized by confusion, it might be prudent to envision
       three separate categories of molecules of reduced bioavailability.
       These are in addition to compounds that are poorly available be-
       cause they are sorbed to solid surfaces or present within NAPLs in
       the immediate vicinity of microbial cells having the requisite cata-
       bolic enzymes:

       (a) Nonsorbed  compounds in micropores  at some distance from
          cells having the requisite enzymes...,

       (b) Compounds that either enter into,  and are retained within,
          nanopores or that partition into the solids themselves...,  and

-------
14
            (c) Chemicals that complex with humic materials or other environ-
               mental constituents to form molecular species that, although
               containing the parent molecules or metabolites generated from
               them, are in fact new molecular species."

         The third definition (Scow and Johnson, 1997), from a review in the soil sciences
     literature, reiterates that several processes act on organic and metallic pollutants in soils
     to reduce their lexicological and microbiological availability. An analogy between the
     availability of native soil organic material and pollutant (xenobiotic) chemicals is drawn.

               "The term bioavailability designates the state of that fraction of
            a chemical that is available for uptake and/or transformation by liv-
            ing organisms. Although associated primarily with ecotoxicology,
            and usually in reference to organic and metallic pollutants, the term
            bioavailability is also relevant to native organic material. Thus, the
            "problem" of bioavailability, has existed for microorganisms far longer
            than has the presence of xenobiotic chemicals in the environment.

               Sorption, insolubility, and  related processes are largely respon-
            sible  for controlling bioavailability of many pollutants to microor-
            ganisms in soils and sediments."

         Definition number four (Baveye and Bladon, 1999), derived from a recent inter-
     national conference on the  environmental behavior of organic  xenobiotic com-
     pounds, evokes the historical, lexicological, and physical-chemical basis for re-
     duced bioavailability. The form of chemicals, Ihe release of chemicals from subsur-
     face environmenls, and Ihe proportion reaching organisms are emphasized.

               "The concepl of biological availability was apparenlly firsl pro-
            posed in 1975 al a Nalional Science Foundation workshop on eco-
            system processes and organic conlaminanls.  and was originally
            based mainly on physical chemislry... Since Ihen, physical chemisls
            and biologisls have  developed independenl interpretations of Ihe
            concepl. Traditionally, chemisls have defined bioavailability in terms
            of Ihe chemical form in which Ihe compound or elemenl of inleresl
            occurs al a given time. Alternatively, definitions derived by biolo-
            gisls  have assumed lhal Ihe chemical form in Ihe bulk phase is rel-
            evanl only to Ihe presence of a biological receptor; Ihus, Ihey have
            defined bioavailability based on Ihe portion of Ihe compound lhal
            could pass into an organism under a given sel of conditions...

               Al Ihis juncture, it is nol necessary for Ihe sake of our argumenl
            to choose belween Ihe approach to bioavailability lhal focuses solely
            on Ihe amounls of xenobiolics lhal cross Ihe organism's external
            boundary.. .or Ihe approach lhal relates Ihese absorbed amounls to
            Ihe capacity of Ihe organism's environmenl lo supply/release Ihe

-------
                                                                                15
       xenobiotics... Both perspectives have merits, which have to be as-
       sessed on the basis of the measurements that one can perform on
       organisms in subsurface materials."

    Definition number five (Maier, 2000) was prepared for the International Society
for Biotechnology. Emphasizing microbial metabolism of organic environmental
pollutants, this definition examines the relationships between "perceived" con-
taminant concentrations and details of microbial metabolism such as nutritional
status, enzyme-induction, energy expenditure, and growth.  Three causes for re-
duced bioavailability are mentioned.

          "Bioavailability can be  defined as the amount of contaminant
       present that can be  readily taken up by microorganisms and de-
       graded. One reason bioavailability as defined here is so important is
       that it governs the rates of biodegradation. Bioavailability controls
       biodegradation because microbial cells must expend energy to in-
       duce  catabolic gene systems used in biodegradation and if the per-
       ceived contaminant concentration is too low, induction will not oc-
       cur. Indigenous soil microbial populations are typically slow grow-
       ing organisms often exposed to nutrient poor environments... The
       presence of contaminants can have a significant impact on the meta-
       bolic status of such cells. Three cases can be envisioned that would
       result from differing bioavailability of contaminants. In the first case,
       biodegradation will not occur because the amount of bioavailable
       contaminant is insufficient to justify the energy expenditure to in-
       duce  biodegradation. In the second case, at low bioavailable con-
       centrations, microbial cells  may biodegrade contaminants but in a
       resting stage rather than a growing stage. While in this case, biodeg-
       radation will occur, it will be limited because new cells are not being
       produced. In the third case, there is enough bioavailable  contami-
       nant to induce biodegradation in a growing stage. It is this case that
       will allow for optimal rates of remediation.

          There are  several constraints that can limit the bioavailability of
       organic compounds in the environment. These are low  aqueous solu-
       bility, sorption, and micropore exclusion."

    The sixth definition (Bosnia et al., 1997) is from a research article focusing on how
mass transfer in soil and aquifers limits the bioavailability of organic compounds.
These authors indicate that reduced bioavailability is a likely explanation for ineffi-
cient biodegradation of contaminants in "old polluted sites." Three causative mecha-
nisms for reduced bioavail-ability are suggested.

          "Particularly in old polluted sites, part of the contaminants ap-
       pear to be inaccessible for biodegradation.. .these observations in-
       dicated a reduced availability of pollutants in soils and sediments

-------
16
            contaminated for a prolonged period of time, pollutant—and not
            nutrient—availability being the obvious cause. The decrease of the
            bioavailability in the course of time is often referred to as "aging" or
            "weathering." It may result from (i) chemical oxidation reactions in-
            corporating them into natural organic matter... (ii) slow diffusion
            into very small pores and absorption into organic matter, or (iii) the
            formation of semi-rigid films around non-aqueous-phase liquids
            (NAPL) with a high resistance toward NAPL-water mass transfer."

         Definition number seven (Alexander, 1997) is derived from a text addressing the
     topic of managing contaminated soils. The author points towards limitations in the
     use of biodegradation as an assay for the bioavailability of organic compounds in
     soils because assimilation, toxicity, and biodegradation are distinctive biological
     processes.

               "A note of caution is required in regard to the term bioavailability.
            It is sometimes  considered as synonymous with toxicity to one or
            another species, sometimes as equivalent to biodegradation by mi-
            croorganisms, and sometimes as synonymous with uptake or as-
            similation. A compound may be assimilated and, although toxic, may
            not cause injury because it is not transported to the tissue, cell or
            intracellular site where the toxicity can be expressed. A chemical may
            be taken up into microbial cells but still not be biodegraded because
            that organism does not contain the requisite catabolic enzymes.
            Uptake or assimilation is thus a better means  of assessing bioavail-
            ability, but because of the few studies of uptake per se and the many
            more of toxicity and biodegradation, the term bioavailability also will
            be used here to include toxicity and biodegradation."

         A recent "state-of-the-science" summary document, addressing the toxicologi-
     cal and environmental impacts of organics, explosives, and metals in soils  [the
     eighth definition (Loehr et al., 1997)], stressed that bioavailability is a measure that
     is dependent on "the receptor, the route of entry, time of exposure, and the matrix
     containing the chemical." This comprehensive definition attributes diminished
     bioavailability of contaminants to "physicochemical processes that lead to seques-
     tration." Three categories of bioavailability are suggested. Furthermore, this defini-
     tion alludes to bioavailability's potential impact on endpoints achieved by remediation
     efforts.

               "Bioavailabilitv is a measure of the potential of a chemical for entry
            into ecological or human receptors. It is specific to the receptor, the
            route of entry, time of exposure, and the matrix containing the chemical.

               It is increasingly recognized that the response of an ecosystem
            or any at-risk population is not controlled by the total concentration
            of a chemical in the media in which a receptor resides, but instead is

-------
                                                                                17
       controlled by only that portion which is biologically available. Thus,
       the definition of "how clean is clean?" or, alternatively, what adverse
       effects may be exhibited by exposure to chemicals in soils, is largely
       determined by the physicochemical processes that lead to seques-
       tration and the biological processes that may lead to chemical re-
       lease and accumulation in an ecosystem. In order to set realistic soil
       or sediment quality limits for regulatory purposes or to establish
       endpoints for remediation processes, the physical, chemical and bio-
       logical mechanisms that govern chemical release and biological up-
       take from soils and sediments must be more fully understood.

       Aspects of bioavailabilitv include:

       •   Physicochemical availability—the fraction of a chemical that is
          not sequestered and rendered inactive in a solid or other stable
          phase;

       •   Direct bioavailability—all or part of the non-sequestered chemi-
          cal that is directly available for entry into specific receptors, de-
          pending upon the route and duration of exposure; and

       •   Organic-induced bioavailability—the fraction of a  chemical,
          perhaps initially considered to be sequestered, that is available
          after processing of the soil/sediment by living organisms.

          Ultimately, bioavailability is defined by field conditions and all
       tests and surrogates for bioavailability must be validated by field
       measurements."

    The final definition (Linz and Nakles, 1997) is taken from a document apparently
aimed at using scientific arguments about the behavior of contaminants and their
bioavailability in soil to influence environmental regulatory policy. Contaminant
release from soil and specific exposure routes for receptors are emphasized. This
lengthy excerpt illustrates that it might be possible to extend agricultural, pharma-
ceutical,  and waste stabilization policies toward establishing "environmentally ac-
ceptable" concentrations of contaminants in soil.

          "The definition of "bioavailability" was also highlighted as impor-
       tant. As defined in this report, the "availability" of contaminants in soils
       has two components: (1) the rate and extent to which the contaminant is
       released from the soil into the surrounding groundwater and soil vapor
       and (2) the rate and extent to which the contaminant is assimilated by
       ecological and human receptors following dermal contact, ingestion, or
       inhalation (i.e., bioavailability). The work group stressed the importance
       of defining "bioavailability" in terms of the specific receptors of interest
       (e.g., microorganisms, humans, animals) and the likely exposure routes

-------
18
            of importance to these receptors (e.g., ingestion of water or soil or der-
            mal contact with water or soil). This matrix of receptors and exposure
            routes identifies the biological systems for which the mechanisms of
            bioavailability should be investigated...

                The work group noted that the concept of environmentally accept-
            able endpoints currently exists within the existing regulatory framework
            as evidenced by the use of TCLP (Toxicity Characteristic Leaching Pro-
            cedure) as a means to determine the hazardous classification of wastes.
            This procedure classifies the potential hazard of a waste based on the
            fraction of the  contaminants that leach from it and not on the total
            contaminant concentration that is present. Furthermore, the US EPA has
            accepted waste stabilization and solidification as acceptable technolo-
            gies for site remediation. This acceptance required the agency to ac-
            knowledge that the relative mobility of the contaminants in the waste
            (without treatment) was reduced following the application of the tech-
            nologies and that this reduction was permanent over time. These con-
            cepts are  consistent with this  document,  namely that hydrocarbons
            remaining in contaminated soil following bioremediation are not mobile.
            In effect, the contaminated material has been "biostabilized."

                Agriculture and other industries integrate information associated
            with the presence of chemicals and their lexicological and environ-
            mental impact:

            •   Agriculture. The methodologies for evaluating the delivery and
               fate of nutrients in soil as well as the fate and effects of pesti-
               cides in soil have beenformally documented (e.g., pesticides are
               regulated by the US EPA under FIFPxA [Federal Insecticide, Fun-
               gicide, andRodenticide Act]). These methodologies and evalua-
               tion techniques may be transferable to the determination of envi-
               ronmentally acceptable concentrations of contaminants in soil.

                Pharmaceutical Industry.  The development of pharmaceutical
                products encompasses many of the issues that must be  ad-
                dressed for contaminants in soil. For example, the development
                of slow-release capsules required an understanding of the rate
                of release of the chemicals to the receptor and the  response of
                the receptor to that dose of chemicals. These  same issues must
                be addressed to make the argument that the release of contami-
                nants from soil that are ingested, inhaled, or contacted dermally
                by humans may be acceptable, even if the concentration of the
                contaminants in soil is not zero. The methodology that has been
                followed by the pharmaceutical industry in  this area may be
                applicable to this environmental issue."

-------
                                                                              19
3.2.  Biosensor Technology Offers a Means Towards  Direct
      Measurement of Bioavailability

    Recent biotechnological developments have opened the possibility of directly
measuring the bioavailable fraction of chemicals in environmental samples. These
developments and their implications are briefly described below.

    Many types of microorganisms have evolved networks of enzymes that con-
tribute to metabolic pathways for the transformation of both organic and inorganic
toxic contaminant chemicals. Regulation of genes coding for such metabolic path-
ways is essential for survival and efficient metabolic function of the host microor-
ganism. In a typical genetic regulatory system, gene expression  (transcription of
mRNA) is activated when the proper inducer compound is sensed in the environ-
ment of the microorganism. This type of transcriptional control is achieved through
the interaction of a transcriptional activator protein with an inducer  compound
(often the toxic chemical itself), as well as with RNA polymerase and the DNA that
codes for the appropriate detoxifying proteins.

    During the past decade, several groups of investigators have used genetic
engineering techniques to combine the genes involved in sensing environmental
contaminants with reporter genes, such as the lux operon, that  triggers a light-
producing bacterial luciferase reaction. Thus, biosensors  of this type can be engi-
neered by placing reporter genes under the  control of transcriptional activators
(along with their corresponding promoters; Willardsone/a/., 1998). Under appro-
priate conditions, a direct correlation between the concentration of a contaminant
chemical and reporter enzyme activity (e.g., light)  can be established.  These
biosensors have been developed for detection of organic (e.g., Heitzer et al., 1994;
Kmgetal., 1991; Willardsone/a/., 1998) and inorganic (e.g., Corbisiere/ al., 1999;
McGrath et al., 1999) chemicals in environmental samples.

    Because whole bacterial cells are often the detection system in the biosensors,
they, by definition, measure the amount of contaminant that is bioavailable. Thus,
to some degree these biosensor techniques provide a means of direct detection of
bioavailable environmental contaminants. Biosensors offer the potential of being
free of the pitfalls characteristic of chemical extraction and analytical procedures.

    However, despite possible advantages, biosensors also have limitations. These
include: (1) responses specific to the membrane structure and uptake system of the
genetically engineered microorganisms (usually E. coli). The diverse, uncharacterized
microorganisms actually active in soil may differ from E. coli in their  membrane
permeability traits; thus, the biosensor may not be a valid surrogate; (2) the bio-
chemistry and specificity of sensor reactions may not be sufficiently understood to
contend with false positive or false negative signals that may occur in complex
geochemical settings; and (3) the physical location of biosensor microorganisms is
unlikely to realistically mimic that of the native soil microorganisms. Thus, much

-------
20
     progress will likely be required before biosensor measurements will supercede the
     indirect bioavailability measures discussed above.


     3.3  Summary

         Bioavailability is a concept that has recently been borrowed from the discipline
     of nutrition and applied to studies of pollutant compounds. Bioavailability is used,
     conceptually, to interpret laboratory- and field-based measurements of the behav-
     ior of environmental pollutants in the presence of biota (microorganisms and plants)
     in soils, sediment, or waters. Contrary to many implicit and explicit views (e.g.,
     definition number 8, above, and the promise of biosensors), bioavailability of chemical
     wastes to soil organisms cannot be directly measured. Instead, bioavailability is an
     emergent trait, a malleable trait. Information about bioavailability emerges from the
     context of real-world or laboratory experimental systems containing biota, contami-
     nant compounds, and a geochemical matrix  (soil or sediment or water).  It is the
     specific, detailed, three-way interactions between system components that allow
     inferences about bioavailabilitv to be drawn. Figure 1 provides a graphical concep-
     tual summary of this definition of bioavailability.

         Figure 1 displays directly measurable system components. Thermodynamics (large
     circle) provides a stage for potential reactions that could occur between contaminants
     (small circle), biota (oval), and soils or sediments (cube). The success of biodegradation
     processes, hence bioremediation efforts (one-way arrow, down), depends on specific
     interactions (two-way arrows) between system components. Inferences about bio-
     availability can be drawn only after system  behavior and output (biodegradation/
     bioremediation) have been observed. Because information about bioavailability is de-
     termined by the experimental conditions, the environmental relevance of experimental
     conditions used in bioavailability assays must be carefully scrutinized.

         Based on the nine definitions listed earlier, there are six hypothesized mecha-
     nisms that reduce the availability of contaminant compounds forbiotic uptake and
     metabolism in soils, sediments, and waters. These mechanisms, which stem from
     interactions between the contaminants and the soils or sediments are as follows:

     (i)  sorption reactions that bind contaminants to solid phase natural surfaces;

     (ii)  partitioning reactions that place contaminants in non-aqueous phase liquids
         (NAPLs)—including NAPLs with "semi-rigid" surface films;

     (iii) spatial  separation of contaminants in micro- or nano-pores;

     (iv) complexation reactions that create new covalent bonds between contaminants
         and humic substances;

     (v)  insolubility of the contaminants; and

-------
                                                                               21
FIGURE 1 Conceptual representation of the relationships between chemical wastes,
biota, and soils, or sediments. The bioremediation outcome is governed by thermo-
dynamic instabilities of unique sets of contaminants in unique combinations with
biota and soil or sediment.  The outcome allows inferences about bioavailability to
be drawn (see text, Section 3.3, for explanation).
(vi) partitioning reactions that place contaminants in natural organic matter.

    A major goal of this report is to scrutinize these six (and related) hypotheses
(see especially, Sections 5,6, and 8).

-------
                            Section 4

         A Survey of  Field  Projects Using
                Biodegradation To Treat
                  Contaminants in Soils
    This section is designed to document the efficacy of bioremediation field projects
and to pose the question: "Does availability of chemical wastes constrainbioremediation
as a site cleanup technology?" A comprehensive summary of the performance of
bioremediation in field projects is beyond the scope of the report. A sampling of repre-
sentative projects is provided in Table I.1

    The first five entries in Table 1 describe completed microbial bioremediation
projects in which soil was contaminated by mixtures of petroleum-related materials.
In some cases, the remediation was implemented on freshly spilled materials (en-
tries 3 and 4). In the other cases, the petroleum was left in contact with soil for many
years (precise duration uncertain) before the active cleanup efforts. The cleanup
endpoints achieved from these remediation activities (800; 700; <50; 10; 100; and
5,409 ppm, respectively, for entries 1-5 of Table 1) were highly variable. One can
make reasonable guesses about causes for the variability. These include: concen-
trations initially loaded, climatic factors, soil characteristics, degree of mixing, vari-
able microbiological populations, and limited availability of contaminants and nu-
trients. Distinguishing between these many possible causes is an overarching
theme of this report.
-A note on sources. Information in Table 1 was not easy to obtain. Unlike peer-reviewed
literature published largely by academicians, the practical results of applied bioremediation
technology are largely reserved for reports in the "grey literature," where experimental
details and rigorous treatment controls are sometimes omitted. Furthermore, the authors of
"grey literature" bioremediation reports may be biased toward favoring the results of their
efforts both to please their clients (owners of contaminated sites), and to support claims of
the effectiveness of bioremediation technology. Quality of data and experimental design
were two of the criteria for selecting entries in Table 1. Another criterion was field realism.
Although academic studies may feature high-quality scientific procedures, these may be
applied to unrealistic "model systems" whose relevance for the field behavior of contaminant
compounds may be suspect.
                                   22

-------
                                                                              23
    Entry #6 in Table 1 reports that phenol-contaminated sediment (phenol is a
water soluble, readily biodegradable compound) was cleaned up by microorgan-
isms to below 1 ppm; while tars failed to fall below 183 ppm after 71 days.

    Studies examining microbial metabolism of pentachlorophenol (PCP)-contaminated
soils also appear in Table 1. Entry #7 reported a PCP decline from 416 to 150 ppm after 13
weeks of land farming. In the land-treatment system described in entry #8, PCP declined
from 55 to 7 ppm after a nearly 8-year mixture of active and passive treatment.

    The next four microbial-related entries in Table 1 (numbers 8-11) describe ef-
forts that attempted to use microbial processes to eliminate polycyclic aromatic
hydrocarbons (PAHs) from soils or sediments. One of these studies  (Entry #9)
revealed that PAHs in coal coking waste were very difficult to analyze:  despite
mixed aeration in a lagoon, sample variability prevented any trends inbiodegrada-
tion from being discerned. In the remaining three studies (Entries 8,10,11) concen-
trations of PAHs declined substantially, though residual  concentrations were ap-
proximately 200 and 159 ppm (for Entries 8 and 11, respectively).

    Phytoremediation technology is relatively new to bioremediation. Consequently,
documented full-scale field studies are still relatively rare. The majority of reports
published to date  (e.g.,  Huang et al., 1997; see publications accompanying
Cunningham et al.,  1997; Ensley, 2000; VanderLeliee/a/., 2001; Vangronsveld and
Cunningham, 1998) focus primarily on the ability of plants to accumulate contami-
nants; whereas case studies on the performance of the technology in significantly
reducing contaminant concentrations in field soils are quite difficult to identify. The
next three entries in Table 1 (numbers 12-14) are among the few field-scale phyto-
remediation efforts reported  to date. Entry #12 reported that total and leachable
lead in a silt loam soil decreased somewhat after treatment with Indian mustard and
sunflower plants (478 ppm of Pb remained). Entry #13 provides a record of how a
planting of corn and white mustard withdrew some metals from soil (based on
tissue analyses) but project performance, in terms of reduction of metals in bulk
soil, was not provided. Entry #14 of Table 1 notes that a commercial effort using
plants to reduce petroleum hydrocarbon concentrations below a regulatory thresh-
old of 500 ppm was successful.

    The final two entries in Table 1 (numbers 15 and 16) are among the many studies
devoted to clean up of chlorinated solvent-contaminated groundwaters. The mi-
crobiological basis for  such bioremediation approaches is either aerobic
cometabolism of TCE (to short-lived compounds) or anaerobic reductive dechlori-
nation of PCE, TCE, and daughter products (DCE and VC) to nontoxic ethene (NRC,
2000). Because chlorinated solvents are very volatile (and surface soils are easily
excavated) bioremediation technology for chlorinated solvents is usually focused
on groundwater (deep) habitats. Entry #15 displays results of a recent  engineered
bioremediation test over a 12 m distance. Steffen et al.'s goal was to prove that
injection of an aerobic bacterium could effectively metabolize  TCE and related

-------
TABLE 1 Survey of Endpoints Achieved by Field BioremediationProjects
Entry
1
2
3
Habitat/
Location
Soil in land farm
Soil adjacent to
petroleum tank
farm
Soil impacted
by New Jersey
fuel spill
Age of
Spill
PD
PD
Fresh
Contaminant
Petroleum
(diesel fuel,
heating oil)
Bunker fuel oil
Diesel fuel
Treatment
Crushing, sieving,
mixing for 1 year
in a lined land
treatment facility
Land farming
(plowing and
irrigation) for
14 weeks
Land treatment
(plowing and
irrigation) for
147 days
Degree of
Cleanup
PAH concentra-
tion declined
from 10,000
to 800 ppm
Total petroleum
hydrocarbons
declined steadily
from 5,000 to
700 ppm
TPH
concentration
dropped from
>700 to <50 ppm
Author's
Interpretation

Successful
project; 150,000
cubic yards of
soil were treated
Complete
remediation
achieved
Reference
cited in
(Angehm
etal, 1998)
(Compeau
etal, 1991)
(Troy etal.,
1993)

-------
TABLE 1 Survey of Endpoints Achieved by Field Bioremediation Projects (Continued)
Entry

4









5



6







Habitat/
Location
Soil surrounding a
fuel oil storage
tank







Soil in Southern
California


Sediment in
surface
impoundment,
Plaque mine, LA




Age of
Spill
Fresh









PD



PD







Contaminant

No. 6 fuel oil









Petroleum
hydrocarbons


Phenol, tars







Treatment

Active
treatment: in situ
mixing, plowing,
aeration for 6
months.
Subsequent
passive treatment
for 27 months


Land treatment:
tilling, plowing


Lined land
farming unit:
mixing, aeration
for 71 days




Degree of
Cleanup
TPH
concentration
declined from
60,000 to
23,700 ppm.
TPH
concenration
declined from
23,700 to
10,100 ppm
TPH declined
from 13,860 to
5,409 ppm in 14
months
Phenol
concentration
declined from
137 to <1 ppm.
Tar concentration
declined from
1,455 to 183
ppm
Author's
Interpretation
Biodegradation
during active
treatment was
more than 6 times
more rapid than
during passive
treatment



Site closure
successfully
achieved

All materials
passed screening
criteria for closure





Reference

(Fogel, 1993)









(Jergere/a/.,
1993)


(Portier and
Christiansen,
1993)






-------
                                                                                             ON
TABLE 1 Survey ofEndpoints Achieved by Field Bioremediation Projects (Continued)
Entiy
7
8




Habital/
Location
Soil in
Minnesota
wood treatment
plant
Soil in creosote-
impacted wood
treating site,
Northeastern US




Age of
Spill
60 years
PD




Contaminant
Pentachloro-
phenol (PCP)
PAHs, PCP




Treatment
Land farming
(plowing and
irrigation) for
13 weeks
Active land
treatment: mix-
ing, aeration for
1 year




Degree of
Cleanup
PCP
concentration
declined from
416 to 150 ppm
Total PAH
concentration
declined from
3,000 to 900
ppm.
PCP declined
from 55 to 16
ppm.
PAHs declined
from -900 to
-200 ppm.
Author's
Interpretation
Rate declined
with
concentration;
17,000 cubic
yards of soil
were treated
Engineered land
treatment was
effective in
eliminating and
stabilizing
contaminants




Reference
(Compeau et
al, 1991)
cited in:
(Loehrand
Webster, 1997)





-------
TABLE 1 Survey ofEndpoints Achieved by Field Bioremediation Projects (Continued)
Entry

9






10














Habitat/
Location
Sediment in
mixed tank
systems




SoilinGeelong,
Vicotria,Australia













Age of
Spill
PD






PD














Contaminant

PAHs in coal
coking waste
lagoon




PAHs














Treatment

Mixing,
aeration for 5
months




Land farming
(mixing and
aeration) for
18 months











Degree of
Cleanup
Total PAH
concentrations
began in the
25 to 50 ppm
range and ended
at 25 to 500
ppm range
Initial PAH
concentration
was between 0.8
and 174 ppm;
average PAH
mass declined
by >l/3








Author's
Interpretation
Much
analytical
variability;
conclusions
evasive


All PAHs below
2 ppm were
persistent and
considered not
bioavailable.
Bioremediation
was limited by:
low bioavailability
low solubility, low
concentration of
low MW PAHs,
desiccation,
elevated pH, and
uneven nutrient
distribution
Reference

(Leavitt et
al, 1991)





(Connolly,
1999)














-------
TABLE 1 Survey of Endpoints Achieved by Field Bioremediation Projects (Continued)
Entiy
11
12

13



Habitat/
Location
Soil and
creosote-rich
sludge in a wood
treating site,
Western MT
Poorly drained silt
loam soil. Ensign-
Bickford
Company,
Simsbury, CT

Twin Cities Army
Ammunition
plant, Arden Hills,
MN


Age of
Spill
PD
PD

PD



Contaminant
PAHs
Lead(Pb)
released during
open burn/open
detonation
activities

Heavy metals,
antimony,
arsenic,
beryllium, lead,
thallium


Treatment
Tillage, nutrient
amendment,
irrigation for
55 months
Indian mustard
and sunflower
were planted in a
1.5 acre area

Corn and white
mustard were
planted to 0.2 acre
plots. Fertilizers
and EDTA were
added to the soil


Degree of
Cleanup
PAHs declined
from 8340 to
159 ppm
Total average Pb
declined from
635 ppm (4/98)
to 478 ppm
(10/98); leaching
assay for Pb
showed a decline
of 0.95 ppm
Plant yields and
metal tissue
content were
reported;
changes in soil
concenterations
were not
reported
Author's
Interpretation

Plants
phytostabilized
the site. Further
treatment is
planned

During first year,
results were less
than anticipated


Reference
cited in:
(Loehrand
Webster, 1997)
Federal
Remediation
Technology
Round Table
(frtr) Web Site.
Blaylock(1997)

Federal
Remediation
Technology
Round Table
(frtr) Web Site



-------
                    TABLE 1 Survey ofEndpoints Achieved by Field Bioremediation Projects (Continued)
Entry
14


15



16



Habitat/
Location
Former Chevron
Bulk Petroleum
processing plant,
Astoria, OR
Shallow sand/clay
aquifer in
Pennsauken, NJ



Deep sand aquifer
at Dover Air Force
Base, Dover, DE



Age of
Spill
PD


PD



PD



Contaminant
Petroleum
hydrocarbons
(TPHs)
VOCs: TCE,
dichloroethane,
vinyl chloride



Tetrachloro-
ethene,
Trichloroethene



Treatment
0.6 acre site was
tilled, fertilized,
and seeded. Plant
species
unspecified
305 and 243
liters of a
biodegrading
aerobic bacterium
were injected into
groundwater;
contaminant loss
was measured in
both test and
control areas
Groundwater
samples were
analyzed for
stable isotope
fractionation to
document
reductive
dechlorinationof
PCEandTCE
Degree of
Cleanup
Contaminant
concentration
range in top 2 feet
of soil dropped
from 170-
l,000ppmto
210-450 ppm
Average
concentrations
fell from 2.2 ppm
to between 0.5
and 0.05 ppm in
the test area



DNAPL
replenishment of
contaminants
prevented full
quantification of
efforts



Author's
Interpretation
Site closure was
obtained after first
growing season
Using several
evaluation
parameters,
removal ranged
from 44 to 78 %



40 % degradation
of TCE in plume
compared to
source area



Reference
US EPA
Remediation and
Characterization
Innovative
Technologies
(REACFUT)
WebSite
Steffene/
al, 1999



Loftaretal.,
2001



Abbreviations: PD - age uncertain; probably decades old; TPH - Total Petroleum Hydrocarbons; PCP - pentachlorophenol; PCE - perchloroethene; TCE - trichloroethene; DCE
- dichloroethene; VOC - volatile organochlorides; VC - vinyl chloride; DNAPL - dense nonaqueous phase liquid; PAH - Polycyclic Aromatic Hydrocarbon

-------
30
     chloroethenes. The bacterium moved out from the injection well, and average con-
     centrations of volatile organochlorides (VOCs) fell during the 2-day treatment.
     Residuals ranged from 0.5 to 0.05 ppm. Entry #16 (Table 1) examined a new tech-
     nique (stable isotope fractionation) as a tool for proving that PCE and TCE were
     anaerobically biodegraded in an aquifer in which a dense nonaqueous phase liquid
     (DNAPL) source of PCE and TCE resided. Under the circumstances, complete
     cleanup could not be assessed—but a significant (40 percent) proportion of the
     TCE was shown to have been metabolized.

         Conclusions about the 16 More-mediation projects shown  in Table 1 are:
     (1) residual contaminants remained in all soils and subsurface waters; (2) mixing
     and aeration accelerated organic contaminant loss; and (3) in some circumstances,
     cleanup goals and site closure were achieved.

         To augment the survey of field bioremediation projects presented in Table  1,
     another recent source of information was consulted. In Appendix D of "Treatment
     Technology for Site Clean-Up: Annual Status Report (Ninth edition)" (EPA, 1999), a
     summary of treatment technology status-report updates was presented.  This site-by-
     site compilation listed  8 years of additions, changes, and deletions of Records  of
     Decisions for treatment technologies used at US EPA Superfund Sites. Of the 373
     changes presented, 62 concerned bioremediation treatment. Of these 62 bioremediation
     projects, 3 were found to be ineffective in bench-scale biodegradation tests and 9
     were abandoned because treatment goals could not be met in the field.

         It is clearfrom the above-mentioned EPAreport on site cleanup technologies and
     from the data in Table 1 as well as from related compilations [e.g., laboratory studies
     of PAH biodegradation in soils (Hughes etal., 1997); field observations of pesticide
     and other chemical persistence in soil (Alexander, 1997)] that both microbial bio-
     degradation and phytoremediation projects are far from 100 percent efficient. In-
     deed, even when the soil described in entry 1 of Table 1 was re-spread onto experi-
     mental field soils, an average of 82 percent of the oily residues remained 21/2 years
     later (Angehrn etal., 1999). The question is "Why?" What causes substances  to
     persist in soil? Certainly bioavailability is one of the candidate explanations for
     ineffective bioremediation, but the influence of bioavailability must be evaluated
     simultaneously with other alternative hypotheses. The next section summarizes
     the past and present understanding of mechanisms by which organic compounds
     resist microbial metabolism.

-------
                           Section 5

    Mechanisms  of Persistence  of Organic
                         Compounds
5.1   An Evolutionary Perspective on the Persistence of
      Organic Compounds

   The Oxford English Dictionary defines "persistence" as "the action or fact of
persisting; firm or obstinate continuance in a particular course in spite of opposition;
continued existence in time or (rarely) in space; endurance; continuous occurrence."
Within a biogeochemical context where concern is focused on the behavior of or-
ganic compounds in soils, sediments, and waters, "persistent" compounds are ones
that, once found, continue to be found. This tendency to persist occurs despite the
fact that soils, sediments, and waters in the real world are dynamic, open systems
where processes such as dilution, volatilization, photolysis, sorption, advection, and
microbial biodegradation all contribute toward the disappearance of organic com-
pounds. Biodegradation is often unique among these processes because enzyme-
catalyzed biochemical rearrangements often convert organic compounds completely
to nontoxic carbon dioxide (Madsen, 1991). Suchmicrobial mineralization reactions
contribute to the growth of microorganisms; thus, such processes are robust.

   Millions (Wackett, 1996; Wackett and Hershberg, 2000) of naturally occurring
organic compounds (synthesized by plants, animals, and microorganisms) are con-
stantly released into the biosphere. These compounds are continuously destroyed
and transported by the above-mentioned processes that prevent persistence. Con-
sequently over broad temporal and seasonal scales, naturally  occurring organic
compounds are generally maintained in steady-state concentrations in soils, sedi-
ments, and waters. A widespread readily biodegradable substance, such as glu-
cose (released by microbial attack of the products of photosynthesis), is turned
over rapidly and occurs in  soil at very low concentrations; while biochemically
resistant soil organic matter recycles relatively slowly and occurs at relatively high
concentrations. Soil organic matter is a "persistent" naturally occurring material. A
                                 31

-------
32
     small subset of naturally occurring organic compounds are toxic (e.g., aflatoxin,
     botulinum toxin), but occur so rarely or are restricted to such rare habitats (e.g.,
     peanuts, anaerobic foods) that these toxins seldom threaten human health and/or
     ecosystem function. In summary, compounds of low (or rare) toxicity in balanced
     dynamic steady-state are the rule for naturally occurring organic compounds.

         Industrial chemicals are subject to the same forces that govern the concentra-
     tions of naturally occurring organic compounds. However, two additional factors
     need to be considered when attempting to understand the environmental fate and
     persistence of industrial chemicals. First, some industrial chemicals exhibit newly
     synthesized, novel molecular structures that are unlike naturally occurring com-
     pounds. These may not be  susceptible to  enzymatic attack and/or may not have
     provided sufficient selection pressure to allow the assembly of efficient metabolic
     pathways in microbial systems. In the absence of well-integrated enzymatic sys-
     tems, microorganisms might not attack novel synthetic chemical structures and
     persistence could result. Second, industrial chemicals are often released nonuniformly
     in time and space, often at high concentrations. This patchy, discontinuous distri-
     bution can often overwhelm naturally occurring processes that might otherwise
     successfully maintain a dynamic balance between production and destruction. At
     high concentrations, even readily biodegradable compounds can noticeably per-
     sist in soils,  sediments, and waters. Regardless of the tendency to persist, indus-
     trial chemicals that are nontoxic (e.g., polyethylene) may not threaten human health
     or ecosystem function, but the combined traits of persistence and toxicity in an
     industrial chemical (e.g., PCBs, dioxin) can pose serious environmental dilemmas.


     5.2   Placing "Bioavailability" Within the Established
           Framework of Persistence  Mechanisms

         In an insightful review written more than a quarter century ago, M. Alexander
     (1973) prepared a list of factors contributing to the persistence (recalcitrance) of
     organic compounds. This list was as follows:

     I.   Property of Organic Molecule

         a. Chemical resists attack by all existing enzymes (proper enzymes have not yet
           evolved),

         b. Molecule unable  to pass through cell wall hence unavailable for enzyme
           induction and intracellular attack,

         c. Site on molecule for enzymatic attack obscured by intramolecular folding,

         d. Molecule composed of a variety of subunits connected by a variety of
           linkages (thus it is improbable for all of the proper enzymes to be coincident
           in time and space),

-------
                                                                              33
    e. Molecule does not yield energy or carbon for microbial growth (cometabolism
      may occur, however), and/or

    f. Molecule or its products are toxic.

n.  Property of environment

    a. Environment exceeds the tolerance range of environmental factors affecting
      survival and activity of organisms ordinarily responsible for mineralization
      of the compounds. Factors include temperature, pH, moisture, light, salinity,
      toxins, redox potential, and hydrostatic pressure,

    b. Environment lacks growth factors or nutrients essential for microbial activ-
      ity,
    c. Environment inactivates enzymes that would otherwise attack the molecule,

    d. Environment is sufficiently voluminous to reduce the concentration of the
      molecule below a threshold required for enzyme induction and/or microbial
      activity,

    e. Environment renders the molecule inaccessible to enzymatic attack by shield-
      ing the molecule in microsites on solids or coating the molecule with inert
      substances,

    f. Environment contains resistant organic substances which form complexes
      with the molecule, thus rendering the molecule resistant, and/or

    g. Environment contains metallic cations which form complexes with the mol-
      ecule, thus rendering the molecule resistant.

    Readers of this report may perhaps be surprised to note that "bioavailability"
was not explicitly mentioned in the above listing. Instead, the hypotheses for ex-
plaining resistance of organic compounds emphasized properties of the organic
molecule and of the environment.

    The last four items listed above (II. d, e, f, g under properties of environment)
are manifestations of interactions between contaminants and soil or sediment. More-
over, threshold concentration, shielding in microsites, organic complexation, and
metallic complexation fully encompass the six bioavailability-based mechanisms
identified in Section 3.3.

    All six hypothesized mechanisms by which bioavailability controls biodegra-
dation (sorption, NAPL-partitioning, micro- or nanopores, absorption into native
organic matter, organic complexation, and insolubility) stem from interactions
between the contaminant and the soil or sediment. Therefore, progress in compre-
hending the influence of bioavailability on bioremediation requires understand-
ing the fundamentals of interactions between contaminants and soils and sedi-

-------
34
     ments (Section 6 of this report) and the mechanisms of biotic uptake of contami-
     nants (Section 7 of this report).


     5.3   Illustrating Mechanisms of Persistence: Soil Organic
           Matter (SOM)

        To facilitate the task of interpreting data describing the persistence of chemical
     wastes in soils (Section 8 of this report), it may be instructive to first review our
     understanding of the  persistence of a ubiquitous, nontoxic naturally occurring mate-
     rial, soil organic matter (SOM). Because of its agronomic significance in maintaining
     soil physical conditions and nutrient status, decades of scholarly research have been
     devoted to understand why SOM resists microbial decomposition processes (e.g.,
     Alexander, 1973; Allison 1965; Brady and Weil, 1999; Stevenson, 1994; Stouletal.,
     1981; van Veen and Paul, 1981). As will become clear in Section 6, ever-improving
     models are needed to understand soil systems. A synopsis of the current model for
     SOM is presented below.

        Organic matter in soil is composed of recognizable plant material, unrecogniz-
     able plant material in various stages of decay, soil biota (microorganisms, soil fauna),
     and the persistent material, humus. Humus is divided into two pools  (McBride,
     1994; Stevenson, 1994): biochemically familiar plant-derived compounds such as
     polysaccharides, polypeptides, and  lignins; and humic substances (amorphous
     polymeric compounds that are inherently resistant to decomposition). In the three-
     dimensional soil matrix, the bulk of soil organic matter occurs as water insoluble
     forms that have been traditionally characterized based on their extractability. Four
     main "associations"  between the inorganic matrix of soil and soil organic matter
     have been identified  (Stevenson, 1994): (1) insoluble macromolecular complexes;
     (2) macromolecular complexes bound together by di- and tri-valent cations such as
     Ca2+, Fe3+, and AP+; (3) organic-clay mineral complexes linked by polyvalent cat-
     ions (clay-metal-humus), H-bonding, and other ways; and (4) organic substances
     intercalated within the interlayers of expanding-type clay minerals.

        Soil organic matter does not accumulate indefinitely in well-drained soils. In-
     stead, it reaches equilibrium levels governed by several "not entirely satisfactory
     explanations" (Stevenson, 1994):  by the ability of dark, high molecular weight
     humic substances to resist microbial attack; by nutrient limitations on humic sub-
     stance synthesis by  microorganisms; and by the  degree of protection afforded
     both pools of humus (familiar classes of compounds and humic substances) through
     associations with polyvalent cations and clays.

        The dynamic state of soil organic matter must be recognized. When plant mate-
     rials  reach the soil, the nine main classes of plant-derived molecules are decom-
     posed by microorganisms in the following order of decreasing rates: sugars, starches,
     simple proteins > crude protein > hemicelluloses > cellulose > fats, waxes > lignins.

-------
                                                                              35
Stevenson (1994) has reviewed the considerable information on soil decomposition
processes obtained using 14C-labeled plant residues. For soils of the temperate
zone, approximately one-third of the plant-derived carbon remains behind in soil
after the first growing season, mostly as a mixture of "labile" and "stable" compo-
nents of humus. Although attempts to measure the age of the stable organic matter
fractions in soil have been made (Stout et al., 1981; Stevenson, 1994), absolute age
determinations are often thwarted by the  dynamic nature of humus: the standing
pool of humus is continuously decomposed while  new humus is re-synthesized.
Nonetheless, mean residence times of the stable pool of modern humus have been
estimated at several hundred to somewhat over 1,000 years. According to Stevenson
(1994), the mean residence time of the "labile" pool of plant carbon (that remaining
after the first season of decomposition) is initially short, but then approaches that
of the stable, native humus carbon.

    The biochemical processes by which high molecular weight humic substances
form are not fully understood. Sposito (1989) suggests four stages in the transformation
of plant biomass into humus:  (1) decomposition of biomass (including lignin) into
simple  organic compounds; (2) microbial metabolism of the simple compounds; (3)
cycling of C, H, N, O between organic matter and microbial biomass; and (4) microbially
mediated polymerization of the cycled organic compounds. The principal humus-form-
ing compounds involved in stages (3) and (4) are believed to be phenolic high molecular
weight structures, especially lignin constituents,  which are converted to a reactive
class of compounds containing quinones that readily condense randomly (Flaig, 1975,
as cited in Stout et al, 1981; Sposito, 1989; Stevenson,  1994). A key trait of humic
substances is their dissimilarity to the biomolecules from which they are derived.

    Soil management practices have long been known to influence SOM persis-
tence. When either forest or prairie soils  are brought under cultivation, a general
decline in SOM occurs (Brady and Weil, 1999; van Veen and Paul,  1981). The
decline in SOM through cultivation and cropping practices can only partially be
attributed to a reduction in the added quantity of plant residues for humus synthe-
sis. It is thought that loss of SOM occurs because cultivation aerates the soil (thus,
stimulating microbial respiratory activity) and exposes previously inaccessible or-
ganic matter (such as in micropores) to microbial attack (Stevenson, 1994)].


5.3.1    Summary (Soil Organic Matter Persistence)

     Key facts related to SOM persistence include:

(i)  Plant-derived biomass (consisting  of many familiar classes of organic  struc-
    tures) is continually replenished in the soil system;

(ii)  Microbial processes convert dead biomass to two humus pools—recognizable
    (e.g., polysaccharide, polypeptide, and lignin) and nonrecognizable (humic
    substances);

-------
36
     (iii) Complexation reactions by metallic soil cations influence both pools of hu-
         mus—thereby rendering them resistant to microbial attack;

     (iv) Both pools of humus associate with clay minerals (inert substances)—thereby
         rendering the humus resistant to microbial attack;

     (v) The "humic substances" pool of SOM also resists enzymatic attack because it
         is composed of a variety of unusual molecular subunits randomly connected
         by diverse chemical linkages; and

     (vi) SOM persistence can be partially overcome by simple mixing of soil—this is
         thought to boost aeration and accessibility of SOM to microbial attack.

         Items (iii)-(v), above, correspond to items II(g), II(e), and I(d) in Alexander's
     (1973) scheme for explaining persistence (Sections.2). In addition, entries I(a)-I(d)
     of Alexander's (1973)  scheme contribute, mechanistically, to the persistence of
     humic substances. Therefore, SOM provides an illustration of the many mecha-
     nisms of persistence and their ability to act in concert. Another crucial point about
     the origin of SOM is that approximately one-third of the initial mixture of naturally
     occurring biodegradable plant materials entering soil resists biodegradation in the
     first season of decay. Furthermore, much of this residue enters the stable humus
     pool with turnover time of hundreds to more than 1,000 years. To the degree that
     plant-derived compounds resemble chemical wastes, lessons from the persistence
     of SOM can guide inquiry into the behavior and bioavailability of chemical wastes
     in soil.

-------
                          Section 6

      Paradigms for the Composition and
      Structure of Soil* and  the Physical-
  Chemical State of Contaminants Therein
   Ideally, in all remediation (including bioremediation) scenarios, the reactions of
both inorganic and organic contaminants in real-world sites would be known prior
to proposing management and technical cleanup procedures. These reactions are
governed by: (1) properties of the contaminants (these are of wide-ranging com-
plexity because they may occur as mixtures whose individual components may
have multiple  reaction pathways); (2) properties of the soils or sediments
(geosorbents*;  as discussed  throughout this section, especially Section 5.4,
geosorbents are extremely complex); and (3) interactions between contaminants
and geosorbents. In reality, however, site management strategies employ an empiri-
cal, iterative, observational approach in which new information is constantly gath-
ered to improve and adjust issues tied to the hazards of contaminants and their
removal/detoxification (National Research Council, 2000).

   What follows are several  discussions designed to foster an awareness of the
compounded complexities of contaminant-geosorbent interactions.  This aware-
ness will later  (Section 8) be brought to bear on the major focus of this report
(bioavailability and bioremediation).
* Note:  Information presented in this section will treat the terms "soil," "sediment," and
"geosorbent" (Luthy, et al., 1997) as synonyms. It will be assumed that the facts and
principles described here apply to all naturally occurring porous geological matrices.
                                37

-------
38

     6.1   Soil Complexity

         Scholarly inquiry into the intrinsic properties of soil (pedology), separate from
     their impacts on plant growth (edaphology), developed significantly in Europe,
     Russia, and the United States in the 19th century (Brady and Weil, 1999). At least
     two complementary approaches to soil science have progressed simultaneously
     since  then: field approaches to natural history and soil genesis; and laboratory
     approaches (chemical, biological, mineralogical, and physical determinations) ap-
     plied  to soil samples. Despite advances in both approaches throughout the 20th
     century, McBride (1994) has written that "much of soil science is empirical rather
     than theoretical in practice. This fact is a result of the extreme complexity and
     heterogeneity of soils, which are impossible to fully describe or quantify by simple
     chemical or physical models."

         Soils are natural bodies, whose lateral and vertical boundaries usually occur as
     gradients between mixtures of materials of atmospheric, geologic, aquatic, and/or
     biotic origin. Soils are  open systems subject to fluxes in energy (e.g., sunlight,
     wind) and materials (e.g., aqueous precipitation, erosion, deposition, and inputs of
     organic compounds from activities of plants, human beings, and other animals).
     Furthermore, soil's intrinsic complexity stems from its nature as an assemblage of
     solid, liquid, gaseous, organic, inorganic, and biological constituents whose chemical
     composition and random three-dimensional structure  have not been completely
     characterized. In addition to physical complexity, the microbial (bacteria, fungi,
     algae, protozoa, and viruses) physiological processes in soil and their multitude of
     interactions are dauntingly complicated. Compounding the challenge of under-
     standing in situ soil processes is the fact that abiotic reactions (e.g., precipitation,
     dilution, hydrolysis; Section 6.2) also must be considered when attempting to un-
     derstand soil geochemistry. Furthermore, in a field setting, plants and animals also
     effect geochemical change.

         Attention also must be paid to the fact that soil properties described above are
     subject to dynamic changes in time and space. No field setting is homogeneous or
     static. Regarding spatial inhomogeneity, the physical, chemical, nutritional, and
     ecological conditions for soil biota undoubtedly vary from the scale of micrometers
     to kilometers. Regarding temporal variability, in situ processes that directly and
     indirectly  influence fluxes of material into, out of, and within soil are dynamic.
     Climate-related influences (such as temperature, sunlight, evaporation, and precipi-
     tation) are probably major variables that cause temporal, variations inbiogeochemi-
     cal processes in soil.


     6.2   A Thermodynamic Overview  of Inorganic  Soil Reactions

         Lindsay (1979) provided a unifying thermodynamic overview of soil in which
     dissolved substances  in soil solution are in constant dynamic equilibria with six

-------
                                                                               39
independent chemical influences: solid mineral phases; exchangeable ions and sur-
face adsorption; nutrient uptake by plants; soil air; organic matter and microorgan-
isms; and water flux. The mineral phases of soil (typically 90 percent of the  solid
matter) have been described as "rock on its way to the ocean" (Lindsay, 1979).
Primary minerals (the parent material from which soils are derived) were often formed
under conditions of high pressure and temperature. At the earth's surface, subject to
oxidative and hydrolytic weathering, the primary minerals become secondary miner-
als as ionic species in solution are leached away and the remaining mineral structures
seek lower free energy levels in their atomic arrangements. Soils contain numerous
minerals, some of which are crystalline, while others are amorphous or meta-stable.
These minerals both respond to and control the dynamic pool of dissolved constitu-
ents in soil solution. A detailed discussion of soil mineralogy and equilibria is beyond
the scope of this report (for this see, Dixonet al., 1977; Lindsay, 1979; McBride, 1994;
Sposito, 1987); but it is critical to appreciate that soil and sediment habitats are in
constant chemical transition, albeit at rates that are slow in human terms. Many of
the mineral components are thermodynamically unstable, and this instability is
compounded by additional reaction pathways imposed by (microbiological processes—
especially those driven by plant-derived carbonaceous materials added via photosyn-
thesis (see Sections 5.3 and 6.1).


6.3  Models of Soil Structural  Characteristics and
      Limitations in Environmental Microbiology

    This section will review information describing  the three-dimensional arrange-
ments of soil components. Ladd et al. (1996) have reviewed relationships between
soil components and the biological activity occurring therein. These authors em-
phasized the vastness in the range of scale of soil constituents (nine orders of
magnitude, from atoms to rocks) and the hierarchical features  of soil aggregates
that form the three-dimensional fabric of soil. Broadly, six size-based categories of
aggregation were described (Ladd etal., 1996): (1) amorphous minerals develop at
the  nanometer-to-angstrom scale; (2) clay microstructure colloids form at 10~7m;
(3)  "quasicrystals," "domains," and "assemblages" form (10~7 to 10~5 m) between
clay, silt, and smaller particles; (4) macroaggregates (0.1 to 250 um) occur between
sand, silt, and smaller particles; (5) macroaggregates (250  um to 25 mm) occur
between gravel, sand, and smaller particles; and (6) clods (> 25 mm) occur between
rocks, gravel and smaller particles.

    It is the aggregation, the aggregate behavior, of  soil that contributes to its com-
plexity. Tisdall and Oades (1982) insightfully presented a schematic model of the
aggregate organization of soil (Figure 2). Emphasized in Figure 2 are both the hierar-
chical scales of soil aggregates and the mechanistically crucial binding agents re-
sponsible for aggregate formation. It is clear in Figure 2 that the various components
of SOM (Section 5.3, this report) play a major role in creating soil structure. Roots,

-------
40
     FIGURE 2 Models, at five different scales, of soil components and their contribu-
     tion to soil structure (from Tisdall & Oades, 1982).

-------
                                                                              41
hyphae, plant debris, fungal debris, bacteria, and humic materials are specifically
mentioned in Figure 2 because of their structural contributions to the soil matrix.

    Documentation of "soil micromorphology" (or "soil fabric"; Ringrose-Voase
and Humphreys, 1994) has played a major role in establishing and reinforcing the
type of model of soil aggregate organization shown in Figure 2. Microscopic proce-
dures applied, whenever possible, to intact soil samples (Foster, 1993; Ladd et al.,
1996) include: transmitted light through soil thin sections, transmission electron
microscopy (TEM), scanning electron microscopy (SEM), electron microprobe analy-
sis (BMP), and environmental SEM. These approaches have provided direct obser-
vations  of intimate associations in soil aggregates of solid surfaces, root hairs,
fungi, bacteria, extracellular poly saccharides, clay films (cutans), humic substances,
and cellular debris. Such "ultrastructural"  studies have revealed that the soil biom-
ass occupies only 0.001 percent of the soil volume. Microorganisms, though present
in large numbers (approximately 109 cells per gram) are neither uniformly nor ran-
domly  distributed but,  as  revealed by TEM  of soil sections, have  been found
clumped near or within cellular residues or in micropores (Ladd et al., 1996). Elec-
tron microscopy/EMP approaches also have been applied to controlled model sys-
tems, and provided important insights into the  mechanisms by which microorgan-
isms can transform (i.e., precipitate and/or dissolve) inorganic materials (e.g., War-
ren and Ferris, 1998). However, one well-recognized limitation of high-resolution
soil microscopy is that each image surveys such a small soil volume that accruing
information truly representative of bulk soil remains a challenge (Foster, 1993).

    Within the discipline  of environmental microbiology, there has been long-
standing interest in understanding small-scale relationships between microorgan-
isms and their habitats. Microscopic environmental microbiological inquiries into
the identity and biogeochemical activity of microorganisms have been conducted.
The goals of these inquiries have been approached  by using various combinations
of confocal scanning laser microscopy (for three-dimensional images), autoradiog-
raphy (to locate both added model radioactive substrates and to document their
uptake by microorganisms), fluorescent antibody staining of individual microor-
ganisms, and probing of naturally occurring microorganisms with fluorescent-la-
beled oligonucleotides that hybridize with 16S rRNA molecules (Amann, et al.,
1995;AmannandKuhl, 1998; Ghiorsee/a/., 1996;Leee/a/., 1999;Ouverneyand
Fuhrman, 1997,1999). Ina now slightly dated review, Madsen (1996) summarized
the accumulating attempts to discover both the in situ biogeochemical activities of
microorganisms in soil  and their identity. Conclusions were that methodological
limitations of sample preparation, incubation, and analysis have prevented achiev-
ing the goal of knowing "who, what, when, and where" of microorganisms in soil
habitats (Madsen, 1996). A recent exception to this statement is the discovery that
anaerobic methane oxidation in deep sea sediments can be carried out by  an inti-
mate association between sulfate-reducing and methanogenic bacteria (Hinrichs et
al., 1999;Boetiuse/a/., 2000, Orphan et al., 200 \). Existing methods for growing

-------
42
     and describing microorganisms in soil typically overlook at least 90 percent of
     those that are detectable microscopically (Amann etal., 1995; Madsen, 1998). Thus,
     not only has the complexity of the soil habitat prevented its full chemical charac-
     terization, but there may be a vast, undiscovered diversity of microorganisms,
     perhaps with novel metabolic properties, residing in soil. A complete census of
     naturally  occurring microorganisms has never been achieved in any habitat
     (Madsen, 1998). Moreover, microorganisms responsible for biogeochemically sig-
     nificant processes in natural habitats have almost never been identified (Madsen,
     1998).


     6.4   Interactions  Between Geosorbents  and Contaminants

        There  is a broad diversity of physical and chemical forces that governs the
     interactions between geosorbents (soils and sediments)  and  contaminants. These
     interactions range from colloidal and electrochemical phenomena to acid-base and
     sorptive reactions (e.g., McBride,  1994; Sposito, 1989). Given the perhaps "impos-
     sible" complexity of soil and sediment systems (Sections 6.1 and 6.3 of this report), it
     is likely that an accurate and comprehensive understanding of soil reactions will only
     be approached by combining the results of many types of inquiry. Physical, chemical,
     mineralogical, and microbiological characterization of extracted soil constituents can
     augment direct microscopic observations of soil micromorphology (Section 6.3). As
     was emphasized, conceptually, in Sections 3.1 and 3.3. the central issues of this report
     focus on the details of specific interactions between specific geosorbents and spe-
     cific contaminants. Considering the wide-ranging properties of potential geosorbents
     (e.g., types of organic matter, types of minerals, and their respective abundances) and
     potential contaminants (e.g., organic, inorganic, soluble, insoluble, ionic, hydropho-
     bic),  a thorough examination of  the many  possible permutations of geosorbent-
     contaminant interactions is beyond the scope of this report. Nonetheless, both or-
     ganic and inorganic contaminants will be discussed below. A particular class of or-
     ganic contaminants, hydrophobic organic compounds, will serve to illustrate how
     inferences  of geosorbent structure can be drawn from indirect observations of con-
     taminant behavior. Examples of direct observations of organic and inorganic contami-
     nants in soils also will be presented.


     6.4.1   Inferences of Geosorbent Structure Based on Indirect
             Observations of Hydrophobic Organic Compounds

        In a recent critical review on sequestration of hydrophobic organic compounds
     (HOCs) by geosorbents (Luthy et al, 1997), a group of experts stated that "cur-
     rently there are no direct observational data revealing the molecular-scale loca-
     tions in which nonpolar organic compounds accumulate when associated with
     natural soils or sediments. Hence, macroscopic observations are used to make

-------
                                                                               43
inferences about sorptive mechanisms and the chemical factors affecting the se-
questration of HOCs by geosorbents." Many such macroscopic observations are
derived from laboratory operations that: (1) react soil or sediment with a solution of
known composition at fixed temperature and pressure for a prescribed period of
time; and (2) chemically analyze the reacted soil, the soil solution, or both to deter-
mine their compositions (Sposito, 1989). For decades, these types of experiments
have been carried out on combinations of HOCs and field-derived or artificial
geosorbents. By analyzing the rates and extents of sorption and desorption, while
varying key components of the geosorbent [e.g., organic matter,  mineral surfaces,
and non-aqueous phase liquid (NAPL) content], many investigators have indepen-
dently developed theories about geosorbent properties. These have often involved
mathematical models that explicitly and quantitatively define a variety of "sites" for
sorption and desorption reactions based on equilibrium and kinetic interactions
between contaminants and geosorbents. These types of analysis have given rise
to many theories that contribute explanations for contaminant behavior (e.g., "two-
site model," "two site-two region model," "gamma model of continuous distribu-
tion of rate constants," "organic matter diffusion model," "sorption-retarded pore
diffusion model"; seeBrusseaue/a/., 1991; Connaughtone/a/.,  1993; Pignatello
and Xing, 1996; Scow,  1997; Young and Weber, 1995). These models often evoke
physical characteristics of sorbents such as "intraparticle nanopores" (Pignatello
and Xing, 1996) and "glassy versus rubbery organic matter" (Graber and Borisover,
1998; Pignatello and Xing, 1999; White and Pignatello,  1999).

    The recent consensus critical review mentioned above (Luthy et al., 1997) has
summarized the "state-of-the-science" for HOC-geosorbent interactions. Figure 3
graphically presents current ideas. Please note that there is a great deal of similarity
between Figure 3 (from Luthy etal.,1997) and Figure 2 (the soil model presented by
Tisdall and  Oades, 15  years earlier). The components of Luthy et al's.  (1997)
geosorbent model (Figure 3) are: mineral-phase geosorbent (the particles); amor-
phous sorbent organic matter ("SOM" in Figure 3); dense SOM; clay particles and/
or oxide coatings; NAPLs (fresh, and aged or weathered); mesopores; micropores;
combustion residue (e.g., soot) and water or gas in pores. Each of the model compo-
nents was carefully scrutinized and justified based on a variety of physical, chemi-
cal, and contaminant-behavior techniques.

    As summarized by Luthy et al. (1997), the key measurement parameters that pro-
vide clues to geosorbent properties are: sorption-kinetics (is equilibrium reached rap-
idly or slowly? is there evidence for hysteresis?); extent of equilibrium partitioning of
the contaminant between aqueous and solid phases (is this partitioning linear or nonlin-
ear?); heats of sorption (low or high?); activation energy (low or high?); competition
(do other sorbates influence sorbtion reactions?); sorbate characteristics (do molecular
steric effects influence sorption?); and extractability (can solvents alter geosorbent
behavior by removing key components?). Based on a systematic examination of the
patterns of HOC partitioning behavior in the presence of many types of geosorbents,

-------
44
            !
                                    WATER OR CIAS    .'
                                                        '
                                                   *   ,
                                                   .iiiaaiiiTiiifc   Js»^?^*.»:'«e.!s   »-y^^-tll;:;::;
                                                     M
                               •t           v/^^^m /a
     CON         N     IH.: E
                                           "'A''       ^
                                           ^riH- vCJM '""
     FIGURE 3 Conceptual model of geosorbent domains developed by Luthy et al,
     (1997). The circled letters refer to representations of sorption mechanisms (de-
     scribed in text, Section 6.4.1.). The geosorbent domains include different forms of
     sorbent organic matter (SOM), combustion residue paniculate carbon such as soot,
     and anthropogenic carbon including Nonaqueous Phase Liquid (NAPLs).
     Luthy et al. (1997) presented five hypothetical mechanisms for explaining how HOCs
     interact with geosorbents. Each of these five mechanisms is represented graphically in
     Figure 3 as a capital letter from A to E. Case "A" is absorption into amorphous or "soft"
     natural organic matter or NAPL. Case "B" is absorption into condensed or "hard"
     organic polymeric matter or combustion residue (e.g., soot). Case "C" is adsorption
     onto water-wet organic surfaces. Case "D" is adsorption to exposed, water-wet mineral
     surfactants (e.g., quartz). Case "E" is adsorption into microvoids ormicroporous miner-
     als (e.g., zeolites) with porous surfaces at water saturation.

         The reader should note the correspondence of the five hypothetical sorption
     mechanisms (letters A to E in Figure 3) mentioned earlier with the six hypoth-
     esized mechanisms that reduced bioavailability from Section 3.3. The consensus

-------
                                                                             45
critical review by Luthy et al. 1997) concluded by recommending that "the disci-
pline is in need of a more sophisticated understanding of...mechanisms at the
microscale." Some progress in this area has been recently made (see below).


6.4.2   Direct Observations of Contaminated Geosorbents

    Direct, small-scale observations of biological and geological materials have led
to profound discoveries in virtually all areas of scientific endeavor (e.g., histology,
mineralogy, biochemistry, and the engineering and material sciences) since the
invention of the microscope (Goldstein et al., 1992). The trend in recent decades
has been in at least two directions: toward greater resolution (e.g., atomic force
microscopy, angstrom resolution in x-ray crystallography of proteins) and toward
combining spectroscopic with visual analyses so that both the spatial arrangement
and the composition of materials can be simultaneously integrated.

    Although new technical advances in microscopic characterization of environ-
mental materials yield new information, methodological limitations are slow to be
fully overcome. As an example, Myneni etal. (1999) recently studied the macromo-
lecular structure of humic substances  using high-resolution spectromicroscopy at
the Advanced Light Source at Lawrence Berkeley Laboratory. These investigators
achieved heretofore unattainable image resolution that demonstrated the influence
of solution chemistry, mineralogy, and source of origin on the size and shape of
humic substances isolated from aquatic and soil environments. Though such de-
tails of the macromolecular structure of humic substances were new, the true three-
dimensional arrangement of humic substances in intact soil matrices remains virtu-
ally unexplored.


6.4.2.1   Hydrophobic Organic Compounds

    During  preparation of this report, several articles were encountered that di-
rectly examined organic contaminants in geosorbent matrices. Examples follow.

    Gillette et al. (1999) used microprobe two-step laser desorption/laser ionization
mass spectrometry (uL2MS) to identify and characterize trace PAHs on geosorbents.
The geosorbents examined included soil spiked with PAHs and also bothbiotreated
and field-derived contaminated samples (soil and sediment). PAH occurrence was
mapped across the surface of individual particles at 40 um increments. Results
indicated that sorption phenomena were heterogeneous: individual geosorbent
particles seemed to be composed of different subparticle-size regions having differ-
ent affinities for PAHs.

    Ghosh et al. (2000) continued the  uL2 MS studies but added Fourier transform
infrared (FTIR) micro-spectroscopy and scanning electron microscopy /wave length
dispersive x-ray spectroscopy (SEM/WDX) to their inquiry. These multiple assays

-------
46
     were used to examine PAH contaminants across sections of sediment particles
     derived from Milwaukee Harbor. PAH concentrations in coal- and wood-derived
     particles were found to be several orders of magnitude higher than on silica par-
     ticles. Surface analyses revealed that PAHs were associated with organic-rich areas
     on sand particles, and not associated with bare silica regions.  Cross-sectional
     analyses of particles (silica and coal) showed that PAHs were far more abundant (30
     to 100 times) on exterior, compared to interior, regions of these particles. Thus, near-
     surface sorption processes (<5 um)  and not deeper absorption processes were
     indicated by this study. Ghosh et al. (2000) also fractionated the Milwaukee sedi-
     ment according to particle size and density. This showed that, although organic
     material (coal and wood) only comprised 5 percent of the total sediment weight, this
     fraction contained 62 percent of the PAHs. When PAH desorption from the various
     sediment fractions was measured,  the coal/wood organic material was found to
     release less than 10 percent of the bound PAHs to Tenax resin. This contrasted with
     more than 80 percent release of the sorbed PAHs from the silt/clay inorganic frac-
     tion. Ghosh et al. (2000) suggested that the PAHs associated with wood and coal
     may not be bioavailable.

         In a followup study, Ghosh et al. ( 2001) used desorption kinetics, thermal
     program desorption-mass spectrometry (TPD-MS), and mathematical modeling to
     further characterize the behavior of PAHs present in the Milwaukee Harbor sedi-
     ments. Desorption activation energies and rates were calculated for a range  of
     PAHs. The model that best described desorption was one in which the PAHs were
     located like a rind on the outer regions of sorbent particles. The authors concluded
     that PAHs associated with clay/silt particles desorb rapidly and are characterized
     by low desorption activation energies. In  contrast, PAHs associated with coal-
     derived material desorb at much slower rates and are characterized by high desorp-
     tion activation energies. The study concluded that "PAHs associated with media
     having large activation energies may thus comprise the unavailable fraction in
     sediments and the PAHs may pose less risk than PAHs in clay and silt fraction."

         Two other studies  will be highlighted here to illustrate recent experimental
     attempts to explore contaminant-geosorbent interactions. Unlike Gillette etal. (1999)
     and Ghosh et al. (2000,2001) who largely examined geosorbents that were already
     contaminated, the other studies began with uncontaminated geosorbents and added
     model PAH compounds.

         Guthrie et al. (1999) implemented a study designed to determine the structural
     composition and molecular interactions of pyrene with soluble and insoluble organic
     matter fractions of sediments. Sediments treated with a biocide and untreated sedi-
     ments were incubated with [13C]-pyrene in aerated microcosms over a 60-day period.
     These investigators used pyrolysis-GC/MS and 13C-NMR to follow subsequent in-
     teractions with naturally occurring organic fractions (humic acid and humin).  The
     unique spectroscopic probing, 13C-NMR, allowed Guthrie et al. (1999) to determine
     that the added pyrene remained intact throughout the experiment. Yet, during the 60-

-------
                                                                              47
day period, the pyrene became increasingly resistant to solvent extraction, while
becoming associated with the humic acid and humin sediment fractions. Furthermore,
pyrolysis GC/MS indicated that the pyrene-organic matter association involved
"adsorbtion or encapsulation (not covalent binding)" and that this association was
enhanced in the presence of viable microorganisms (Guthrie etal., 1999).

    Schultz et al. (1999) designed a study to elucidate the relationships between
soil organic matter (SOM) structure and sorption behavior of phenanthrene, an-
other PAH. Pyrolysis-GC/MS was used to directly characterize the heat-labile mo-
lecular components of SOM in five geosorbents—three surface soils and two sub-
surface sediments. Principal components analysis of the pyrolyzed fragmentation
patterns allowed clear distinctions to be drawn between the types of SOM in each
geosorbent. Sorption and desorption parameters were evaluated for the geosorbents,
and correlation analyses were performed between the sorptive properties and the
soil organic matter components. Phenanthrene thermal desorption profiles also
were determined for each geosorbent; these suggested the presence of both min-
eral and organic-matter components in geosorbents that impede release of PAHs.


6.4.2.2   Inorganic Compounds

    Inorganic contaminants [such as heavy metals, metalloids, radionuclides, and
oxyanions (e.g., nitrate, chlorite)] exhibit a wide spectrum of properties (e.g., oxida-
tion state, speciation) that allow them to react with inorganic or organic geosorbent
components, orboth. Many sources of information describing behavior, reactivity,
mineralogy, and other aspects of inorganic chemical contaminants have been pub-
lished (e.g., Adriano, 1992; Alleman and Leeson,  1999; Alloway, 1995; National
Research Council, 2000; Vandegrift et al., 1992; Vangronsveld and Cunningham,
1998). Efforts to characterize inorganic contaminants in soil have progressed sig-
nificantly in recent years. Many direct microscopic and analytical procedures have
been applied to many combinations of geosorbents and contaminants. Selected
examples are presented below.

    During uranium processing, aqueous, solid, and airborne radioactive wastes
were released to soil at the US Department of Energy's Fernald, Ohio, site. Morris et
al. (1996) examined contaminated soil samples using a combination of x-ray absorp-
tion, optical luminescence, and Raman vibrational spectroscopies, along with ancil-
lary techniques such as energy dispersive scanning electron microscopy and pow-
der x-ray diffraction. The objective was to ascertain the oxidation state, the chemical
form, and the physical state (surface precipitate, secondary mineral or absorbate) of
the uranium. The procedures used had experimental advantages of:  not requiring
invasive sample preparation; a spatial resolution range from 1 cm2 to less than 100
um2; and spectroscopic techniques  that applied to both amorphous samples and
submolecular coatings. The x-ray absorption spectroscopy provided definitive evi-
dence that the bulk of the oxidation state distribution of uranium (75-95 percent)

-------
48
     favored the hexavalent species. Furthermore, the uranium minerals that were identi-
     fied, similar to autunite and schoepite, often appeared as platey tabular grains rang-
     ing in size from 10 to 100 um or in association with geothite or quartz. Considerable
     weathering (especially oxidation and both phosphate and hydroxide precipitation) of
     the initially released uranium had occurred. Additional uranium minerals also were
     noted, and their photodecomposition properties and field distribution suggested that
     uranium binding by organic ligands also may have occurred at the Fernald site.

         Welter et al. (1999)usedacombinationofXAFS (X-ray Absorption Fine Struc-
     ture  Spectroscopy), XANES (X-ray Absorption Near Edge Structure), and SEM/
     EDX (Scanning Electron Microscopy/Energy-Dispersive X-ray fluorescence spec-
     troscopy) to examine the chemical speciation of Pb  in two soil samples from a
     battery manufacturing plant in Hanover, Germany. Pb was heterogeneously distrib-
     uted in the soil, concentrations ranged from between 50 and 140 g/kg. To contend
     with the  inhomogeneity, the soil samples were mixed and ground prior to being
     analyzed. By comparing XAFS soil signals to those of combinations of authentic
     Pb minerals, Welter et al. (1999) were able to quantify the amount of Pb carbonate,
     Pb oxide, and Pb sulfate in the two soil samples. SEM revealed individual Pb miner-
     als in the soil matrix.

         Galvez-Clouthier and Dube (1998) implemented a study designed to document
     the associations between heavy metal contaminants (Pb, Zn, Cu, Cd) and natural
     sediment constituents in the Lechine Canal, Quebec. X-ray diffraction, TEM (Trans-
     mission Electron Microscopy) and other geochemical measures revealed that the
     sediments consisted mainly of silt- and clay-sized fractions composed of feldspar,
     kaolinite, chlorite, calcite, and dolomite, as well as minor amounts of Fe minerals,
     amorphous metal oxides, and organic matter. Each of these constituents bound
     heavy metals to varying degrees, as assessed by sequential chemical extractions of
     residual contaminants from oxide-, carbonate-, organic-, and exchangeable-phases
     of the sediments. The heavy metal partitioning patterns for varying sediment size
     fractions also were evaluated. Results indicated that no particular mineral phase or
     size fraction accumulated particular heavy metals. However, significant concentra-
     tions of the heavy metals had accumulated in the sediments over the last century,
     and these posed a high risk for metal  release into the water column.


     6.5   Summary

         Section 6 of this report has presented a glimpse into the facts, principles, and
     challenges of understanding soil and sediments, and their interactions with organic
     and inorganic chemical contaminants. Of the multitude of geosorbents, contami-
     nants, and their interactions, only a few were highlighted here. Many of the long-
     standing questions about real-world geosorbent matrices remain unanswered: they
     are of complex composition, in complex three-dimensional arrangements, that occur
     heterogeneously in open biogeochemical systems. Layered onto geosorbent com-

-------
                                                                              49
plexity is the biology and biochemistry and ecology of the biota that dwell in soils
and sediments. Also, layered onto this are many potential reaction pathways of
organic and inorganic chemical wastes. There is simply much more to be learned
about geosorbents and contaminants (Sections 6.0 and 6.3), especially because
geosorbent composition, contaminant composition, and the geochemical context
forbiogeochemical reactions are all likely to be site specific.

   Despite the above seemingly "impossible" complexities (Section 6.1), both
geosorbents and contaminants abide by predictable laws of thermodynamics.  We
also have robust chemical, mineralogical, and biochemical principles to constrain
the potential reactions (Section6.2). Furthermore, a growing battery of new micro-
scopic and spectroscopic procedures (Sections 6.3, 6.4) are being applied at an
accelerating rate to real-world contaminated geosorbents. These procedures have
already provided  new information that delivers new insights and distinguishes
between competing hypothetical models for geosorbents and their dynamic reac-
tions with chemical waste materials. For instance, microscale spectroscopic assays
suggest that relatively shallow  regions of the "soot"  component of sediments
feature prominently in governing the release and availability of hydrophobic or-
ganic contaminants. Additional progress is likely to be made via an iterative dia-
logue between hypothesis refinement, application of new analytical technologies,
and an accruing database that catalogs the properties of geosorbent matrices and
the reactions of contaminants therein.

-------
                           Section 7

  Uptake of Soil Constituents by  Plants and
                      Microorganisms
   This section will proceed from the general to the specific, from text book prin-
ciples to recent research articles. After the pivotal role of soil solution in all soil
processes is presented, then detailed uptake mechanisms of inorganic and organic
compounds by plants and microorganisms will be discussed. Bioavailability issues
will be emphasized by focusing on factors that limit accessibility of insoluble materi-
als for transport across the extracellular surfaces of plant roots and microorganisms.


7.1   Principles of Soil  Solution Chemistry and  Uptake by
      Plants of Inorganic Compounds

   The complexity of the soil habitat was discussed in Section 6.0 of this report
(see especially Sections 6.1,6.2,6.5). Text in Section6.2 drew onLindsay's (1979)
observation that "the soil solution is the focal point. The liquid phase that com-
pletely envelops the solid phases.. .is the medium from which plants absorb their
nutrients."

   Regarding inorganic, mineral phase-forming soil components, Lindsay (1979)
further stressed that "two very important parameters influence the availability of an
element to plants. These are: (1) the intensity factor, which is the concentration of
an element in soil solution, and (2) the capacity factor, which is the ability of solid
phases in soils to replenish that element as it is depleted from solution." Soils may
have a high bulk content of a given nutrient (such as iron) and therefore have high
capacity. However, the solution-phase concentration of nutrients is regulated by
their dominant mineral form. Often, the most thermodynamically stable mineral phase
is the least soluble phase and, if the equilibrium concentration of the stable phase
is below the critical concentration for root uptake, plant nutrient deficiency will
result. Lindsay (1979) emphasized that nutrient concentrations found in soil solu-
                                 50

-------
                                                                               51
tion are influenced by other key reservoirs, especially solid-phase sorption sites
and organic matter. Nonetheless, "mineral phases ultimately control the level of
nutrients in solution" (Lindsay, 1979). Thus, characterization of soil minerals and
understanding their stability and reactions with inorganic contaminants is crucial
for predicting the composition of soil solution. This view was reinforced by Sposito
(1989) who stated that "from the perspective of soil chemistry, the bioavailability of
an element is determined by competition among plant root systems, the soil solu-
tion, and solid-phase particles." Sposito (1989) also emphasized that aqueous chemi-
cal complexation reactions influence mineral nutrient bioavailability to plants.
Sposito (1989) presented data depicting a direct relationship between the free-ion
species of metals in solution and uptake of metals by plants.  Sposito (1989) con-
cluded that:  "A chemical element is bioavailable, if it is present as, or can be
transformed readily to, the free-ion species." Uncharged metal-organic complexes
also are generally bioavailable.

    As will be discussed below, the above principles of soil chemistry set the stage
for understanding nutrient bioavailability in soil. But each component of soil solu-
tion has its own set of reaction pathways, solubility products, and spectrum of
chemical species. Furthermore, plants and microorganisms sometimes feature  un-
usual physiological adaptations for obtaining specific nutrients from soil solution
(especially organic iron-binding agents, the siderophores). Thus, sometimes elabo-
rate interactions between soil chemistry and soil biota add to the challenge of
understanding contaminant bioavailability issues.


7.2   Movement of Solutes From Soil Solution to Roots

    Plant uptake of ionic nutrients requires that they come into contact with the root
surface (Havlin et al., 1999). There are three primary mechanisms by which nutrient
ions in soil could reach the root surface (Barber, 1995; Marschner, 1995): (1) mass flow
of ions in solution; (2) diffusion of ions in soil solution; and (3) root interception of
solid-phase ions. Havlin et al. (1999) present data depicting the relative significance
of each of these three mechanisms for the uptake of 12 nutrients by corn plants.
Unique soil chemical reactions govern each nutrient; therefore, the relative contribu-
tion of each uptake mechanism is highly variable. For instance, Havlin et al. (1999)
indicate that, for corn, root interception accounts for 1 percent of nitrogen (N) uptake,
3 percent of phosphorus (P) uptake, excessive (171 percent) uptake of calcium, and 11
percent of ironuptake. In contrast, diffusion-based uptake forthe same four elements
was 0 percent, 94 percent, 0 percent, and 37 percent, respectively.

    Mass flow (convection) of nutrients into plant roots occurs when dissolved
substances are transported in the flow of water to roots that results from transpira-
tional uptake by the plant. The amount of nutrients reaching roots by mass flow is
determined by the rate of water consumption by plants and the concentrations of
dissolved soil solution nutrients.

-------
52
         Diffusion-based uptake of nutrients by roots from soil solution is governed by
     concentration gradients. A high requirement by a plant for a nutrient results in a
     large concentration gradient, favoring a high rate of ion diffusion from soil solution
     to the root surface. According to Havlin et al. (1999) and Marschner (1995), most P
     and potassium (K) uptake in corn is governed by diffusion to the root surface over
     distances of 0.02 cm and 0.2 cm, respectively.

         Of the three nutrient uptake mechanisms, root interception is most germane
     for providing insight into how plants can contend with limited bioavailability. Root
     interception relies on direct physical contact between plant root surfaces and soil
     solids. Havlin et al. (1999) postulate that root interception occurs via a "contact
     exchange" mechanism in which overlapping oscillation volumes between ions (e.g.,
     H+) attached to root hair surfaces exchange with ions held on the surfaces of clay
     particles and organic matter. Root interception is enhanced both by the growth of
     new roots and by mycorrhizal infection (caused by a fungal symbiont whose hy-
     phae link root tissue to soil pores) because these processes allow exploitation of
     greater soil volumes.

         Thus far, only movement of soil components to the root surface has been
     discussed. As a rule, there is great discrepancy between the mineral, nutrient con-
     centrations in soil solution, and the mineral nutrient requirements of plants; there-
     fore, the mechanisms by which plants bring minerals into their tissues must be
     selective. The selective uptake mechanisms used by plants can include (Marschner,
     1995): unique cation-binding properties of rhizodermal cell walls; and ion channels,
     transmembrane pumps, or protein carriers residing in tonoplasts and/or plasma
     membranes within plant cells. Marschner (1995) has discussed plant physiological
     and anatomical mechanisms that confer such selectivity. Details are beyond the
     scope of this report. Nonetheless,  it should be recognized that results of solute-
     uptake studies performed on both lower and higher plants demonstrate the follow-
     ing characteristics (Marschner,  1995): (1) selectivity—preferential uptake and/or
     exclusion; (2) accumulation—concentrations of solutes can be much higher in the
     cell sap than in soil solution; and (3)  genetic variability—distinct differences in
     solute-uptake traits among different plant species.


     7.3  Examples of Nutrient Uptake by Plant Roots

         This subsection of the report  directs  readers towards two examples of the
     detailed mechanisms by which plant physiology and soil-solution chemistry inter-
     act to govern solute uptake by  plant roots. The first example, phosphorus, has a
     fixed oxidation state (+5) but occurs  in anionic forms in soil solution that form
     insoluble minerals (e.g., hydroxy-apatite). These anionic forms can bind and/or
     coprecipitate toxic metals such as Ni, Pb, and U. The second example, Fe, under-
     goes redox reactions, and has several competing biotic and abiotic reaction path-
     ways in the soil habitat.

-------
                                                                               53

7.3.1.  Phosphorus (P)

    According to Barber (1995), the phosphorus content of soils can vary from 0.02
to 0.5 percent, with an average of approximately 0.05 percent. The phosphorus pool
is divided into four general categories: (1) P as ions and compounds in soil solu-
tion; (2) P adsorbed onto surfaces of inorganic constituents; (3) P minerals, both
crystalline and amorphous; and (4) P as a component of soil organic matter. In soil
solution, the dominant chemical species are usually either H2PO4" or HPO42", de-
pending upon ambient pH. This solution-phase P equilibrates rapidly (labile P) and
slowly (nonlabile P) with the adsorbed P pool. Barber (1995) indicates that contro-
versy exists over the amount of P that may be adsorbed on the surfaces of such soil
constituents as iron and aluminum oxides versus the amount that is precipitated as
discrete mineral forms. Many potential P minerals may exist in soil. In basic soils
(above pH 7), calcium phosphates (e.g., fluoro-apatite, hydroxy-apatite) should be
dominant, while in acid soils (pH<7) iron (e.g., strengite) aluminum phosphates
(e.g., variscite) are the dominant forms. Solubilities of the pure crystalline minerals,
and their pH dependence can be used to predict the behavior of P in soil. However,
kinetic barriers to reaching equilibrium in soil, formation in impure minerals, and
metastable states limit the usefulness of such predictions (Barber, 1995).

    One-half or more of the total P in many  surface soils  may be in the organic
form—principally as esters of orthophosphate (e.g.,  inositol phosphates,  phos-
pholipids, nucleic acids) (Havline/a/., 1999; Barber, 19995). The release of organic
P into soil solution (where it can react with other soil constituents or move to the
root and be absorbed) is controlled by the rate of soil organic matter decomposi-
tion. Barber (1995) states that, intemperate climates, an organic matter mineraliza-
tion rate of approximately 2 percent per year is to be expected.


7.3.2.  Iron(Fe)

    In contrast to phosphorus, iron is cationic and undergoes oxidation/reduction
reactions. In soil, iron can exist in the ferrous (Fe2+) and ferric (Fe3+) states. This
variability in oxidation state, combined with a wide range of complexation and
precipitation reactions (all dependent on pH, redox,  and  concentrations of soil
anions such as hydroxide and carbonate), contributes to the intricacies of iron's
behavior in soil (McBride, 1994; Sposito, 1989). In fact, there are more than nine
different possible iron soil minerals (Schwertmann and Taylor, 1977). Iron minerals
commonly found in soils are the oxides or hydroxides (goethite, hematite,
lepidocrocite, maghematite, magnetite, and amorphous forms). According to Barber
(1995), the latter is possibly the most significant form in supplying iron for uptake
by the  plant. Although Fe2+ is more soluble than Fe3+, both oxidation states have
such a strong tendency to form insoluble minerals so that total Fe in soil solution is
insufficient to meet plant nutritional requirements. Clearly mechanisms besides

-------
54
     simple uptake of iron from soil solution exist—otherwise, plants grown on almost
     all soils would be Fe deficient (Havlin et al., 1999).

         There are three primary mechanisms by which plant roots actively increase the
     availability of iron in soil (Hartwig and Loepert, 1993): (1) exudation of protons to
     locally solubilize Fe at low pH, (2) release of reducing agents that convert highly
     insoluble Fe3+ to more soluble Fe2+, and (3) by exudation of siderophores (organic
     Fe-chelating agents that bind and mobilize Fe3+). Plants have been classified as
     either Fe efficient or Fe inefficient. The Fe-efficient plants respond to iron defi-
     ciency by releasing protons and reductants into the rhizosphere where the  root
     hairs absorb the iron, primarily as Fe2+ (Barber, 1995). After transport to the protoxy-
     lem and metaxy lem, the Fe2+ can be oxidized back to Fe3+, chelated with citrate and
     moved into the xylem for transport to the plant shoot (Barber, 1995). According to
     Tagaki (1993), the scheme just described is "Strategy I" for overcoming iron defi-
     ciency; it can involve  enzymatic reduction of Fe3+ to Fe2+ and is prevalent in an-
     giosperms (seed producing plants).

         The third mechanism for enhancing iron uptake involves production of
     siderophores in the roots of grasses. InTagaki's (1993) scheme, these are "Strategy
     II"-type plants. The siderophores are organic chelating agents with such high iron-
     binding affinities that they solubilize the Fe3+ from soil minerals. Once in solution,
     movement of the iron-chelate complex is via mass flow or diffusionback to the root
     tip region (Barber, 1995), where the Fe is released from the siderophore  by being
     reduced to the ferrous form, and absorbed by the plant root. Transporter-mediated
     internalization of Fe-siderophore complexes (RoemheldandMarschner, 1986) also
     has been reported. After being freed following uptake by the root, the siderophore
     is released  to the  soil  solution where it can again solubilize ferric iron  from the
     mineral phase.


     7.4  Contrasts  Between Substrate Uptake Mechanisms in
           Microorganisms and Plants

         Soil is inhabited by many types of microorganisms [bacteria, fungi, algae, pro-
     tists (e.g., protozoa) and viruses]. Most of the discussion below will be restricted to
     bacteria because of their relevance to bioremediation. There are many major struc-
     tural, evolutionary, and physiological distinctions between plants and microorgan-
     isms (e.g., Madigane/a/., 2000; Lengelere/a/., 1999). Only afew of these distinc-
     tions are  relevant to the uptake of substrates in the soil habitats. Implicit in the
     previous  discussion of root-uptake mechanisms was the fact that  the materials
     brought into the plant tissue from soil are used in assimilative metabolism,  e.g.,
     incorporated into  plant structural elements (e.g., cell walls, enzymes, and other
     cytoplasmic components). Microorganisms have assimilative nutritional needs that
     are similarto plants. However, unlike plants, the majority of soil microorganisms are
     nonphotosynthetic—their physiologies rely on uptake and metabolism of organic

-------
                                                                               55
carbon, not CO2. This means that soil microorganisms must bring both inorganic
and organic substances across cell walls and cell membranes to grow.

    Microorganisms live in intimate contact with soil solids. This direct contact
with soil solids (analogous to the "root interception" uptake mechanisms for plants)
is an important means by which bacteria and fungi acquire insoluble organic and
inorganic substances from the extracellular milieu. In addition, microorganisms (es-
pecially bacteria) also have dissimilatory nutritional needs—the primary example of
this is in the generation of ATP using many different final electron acceptors. Like
plants, aerobic microorganisms use oxygen, a sparingly water-soluble  gas that
readily diffuses across cell membranes, in respiratory processes to generate ATP.
Because solution-phase oxygen readily passes into cells, one would not expect
enhanced-transport systems for oxygen to be present in the microorganisms. Spe-
cific adaptations aimed at overcoming solid-phase bioavailability limitations of the
highly soluble alternative electron acceptors  (e.g., nitrate and sulfate,  used by
anaerobic microorganisms) also would not be expected. However,  in anaerobic
settings, alternative bacterial final electron acceptors can include the highly in-
soluble oxides of iron and manganese. The mechanisms by whichbacteria metabo-
lize these  latter insoluble compounds are particularly germane to bioavailability
issues.

    Because of the importance of solid phase biogeochemical processes, the mecha-
nisms of electron transfer by microorganisms to poorly soluble minerals for anaero-
bic respiration is subject to intense, ongoing study (Lovley 2000; Newman and
Kolter, 2000; Seeliger et al., 1998). Currently accepted potential mechanisms (Lovley,
2000) include extracellular electron shuttles either involving naturally occurring
humic substances (Lovley etal., 1996) orbacterially excreted quinones (Newman
and Kolter, 2000) and direct enzymatic reduction of solid oxides at the cell surface
via membrane-bound enzyme systems that span the inner and outer cell surfaces.

    Thus,  it is important to recognize that in bacteria (prokaryotes), the  electron
transport chain used in ATP generation is located in the plasma membrane at the
periphery of the cell (Ehrlich, 1995; Lenglere/a/., 1999). This contrasts with eukary-
otic cells (such as plants) in which electron transport occurs internally in special
organelles called mitochondria. Because of their specialized cellular architecture,
bacteria endowed with appropriate oxido-reductases (enzymes that transfer hydro-
gen atoms or electrons) in their cell envelope are able to oxidize or reduce insoluble
substrates that cannot be taken into these cells (Ehrlich, 1995; Lengler et al., 1999).
Such substrates include elemental sulfur, iron sulfide, iron oxides, and manganese
oxides. Because the essential enzymes that recognize and act on insoluble sub-
strates are located in exterior regions (e.g., the periplasmic space, the plasma
membrane, or the outer membrane), bacteria are uniquely capable of metabolic
activities that render extracellular materials "bioavailable." Utilization of endog-
enous (quinones) and exogenous (humic) substances also can facilitate physi-
ological reactions between bacteria and "nonbioavailable" substrates.

-------
56

     7.5.  Membrane Transport in Microorganisms

         [The following discussion is derived largely from Lengeler et al., 1999.] Cell
     membranes define the internal metabolic system of living cells. Membranes act as a
     physical barrier between microorganisms and their environment. Membranes also
     carry out essential metabolic, sensory, and reproductive functions. Lengeler et al.
     (1999) state that:  "The cytoplasmic membrane of bacterial cells consists of aphos-
     pholipid bilayer, which functions as a permeability barrier for most solutes. Polar
     solutes (e.g., carbohydrates) and charged molecules (ions, carboxylic acids, amino
     acids) have a very low rate of passive flux across lipid bilayer membranes. In addi-
     tion to some small  solutes and molecules (such as water, ethanol, ammonia, or
     oxygen), only apolar (hydrophobic) compounds (e.g., phenylamine, glycerol, or
     fatty acids) pass across lipid bilayers. Bacteria, however, need to transport solutes
     at high rates across the cell wall and cytoplasmic membrane for growth and metabo-
     lism. The solute transfer across bacterial membranes is mediated by specific mem-
     brane proteins called transporters, transport systems, carriers, or, in analogy to
     enzymes, permeases. By means of these proteins, the transfer rate across bacterial
     membranes can be  significantly increased... The presence  and activity of carrier
     systems in a relatively impermeable membrane is the reason for observed concen-
     tration gradients of solutes across the cell membrane. These range from 10-30 fold
     (external/internal Na+ and internal/external K+) to more than 10,000 fold (external/
     internal free Ca2+), up to a 200,000-fold accumulation of some solutes in the cyto-
     plasm (e.g., maltose and particular amino acids)."

         Lengeler et al. (1999) have further summarized the four key mechanisms of
     solute transport into prokary otic cells: (1) diffusion (with or without facilitation by
     a permease); (2) secondary transport, in which a permease combines with an elec-
     trochemical  gradient to  adjust concentrations across the membrane; (3) primary
     transport, in which solute transport is directly coupled to chemical or photochemi-
     cal reactions; and (4) group translocation, in which the solute consumes ATP
     through phosphorylation reactions inside the membrane. Diffusion-based trans-
     port applies  to gases (O2, CO2, NH3), small molecules (water, ethanol), and also
     hydrophobic (lipid soluble) molecules such as aliphatic (e.g., butanol) or aromatic
     molecules (e.g., benzene). A number of solutes are membrane permeable  in their
     uncharged form (e.g., protonated organic acids) but do not pass through mem-
     branes if charged.


     7.6.  Uptake of Insoluble Organic Substrates by  Micro-
           organisms: Wood

         High molecular weight organic polymers such as cellulose and lignin are in-
     soluble. These would appear to be "unavailable" for metabolism, yet they are wide-
     spread in nature and are biodegradable. Lengeler et al. (1999) have described some

-------
                                                                               57
general physiological adaptations that allow microorganisms to exploit the diversity
of naturally occurring polymeric organic substrates. The microbial world features a
wide variety of usually hydrolytic extracellular enzymes that cope with crystallinity,
low solubility, and the association of polymers with solid-phase materials.

    Cellulose is the most common organic substrate in nature. It is a linear polymer
of 100 to 100,000 glucose units linked by B-1,4 bonds. Cellulose chains form in-
tramolecular and intermolecular hydrogen bonds that allow rigid insoluble fibrils to
form. When attacked, this insoluble substrate is eventually converted to soluble
glucose monomers and dimers (cellobiose) that, after cell entry, are intracellularly
metabolized. Attack of cellulose occurs through the action of a battery of extracel-
lular exoenzymes. These enzymes fall into two general categories: endoglucanases
(that sever the linear cellulose polymer, presumably by random attack at sites within
the chains) and exoglucanases that sequentially remove dimeric cellobiose units
from one end of the chain) (Bayer and Lamed,  1992). According to Lengeler et al.
(1999), the binding of exoenzymes to surfaces of the microbial cells, the attachment
of the enzyme to the cellulose, and possibly the uptake of polymeric substrate into
the cell periplasm (between cell wall and cytoplasmic membrane) minimizes both
dilution of the enzymes and losses of hydrolyzed cellulose subunits to competing
microorganisms.

    The anaerobic bacterium, Clostridium thermocellum, has evolved a highly
specialized multienzyme cellulase complex, termed a cellulosome. This extremely
high molecular weight structure (>2 million daltons) is positioned at the interface
between the microorganism and the cellulosic substrate. Strong adhesion between
C. thermocellum  and the cellulases is required prior to cellulose degradation (Bayer
andLamed, 1992).

    Natural structures like wood are highly complex—consisting of many types of
polymers and monomers in a rigid matrix often bound by lignin. Lignin molecules are
not susceptible to hydrolytic cleavage (Lengeler et al., 1999). Rather, lignin metabo-
lism (rare in bacteria, more common in fungi) requires an oxidizable substrate (e.g.,
glucose) and O2 to form hydrogen peroxide that activates ligninase enzymes. Thus, a
mingling of different microorganisms, each with their own unique properties, contrib-
utes to successful biodegradation of naturally occurring, wood-derived materials.


7.7.   Phosphorus and Iron Uptake by Bacteria

    Phosphorus is a growth-limiting element  under many environmental condi-
tions—soil solution concentrations are in the micromolar range, while intracellular
concentrations are approximately 1,000-fold higher (Lengeler et al., 1999). Inor-
ganic P (orthophosphate) is the preferred P source in bacteria, as its presence
suppresses expression of genes involved in the acquisition of alternative (organic)
P sources. Inorganic P can enter the bacterial cell either through low-affinity (per-

-------
58
     mease-based diffusion) or high-affinity (ATP-dependent) transport systems. The
     latter are activated by inorganic P limitation.

         Microorganisms (especially bacteria and fungi) use mechanisms much like plants
     to meet their metabolic demand for insoluble iron in soil. According to Hofte (1993), with
     only a few exceptions, all aerobic and facultative anaerobic microorganisms that have
     been critically examined produce iron siderophores. These fall into three categories
     based on molecular structure: catechols, hydroxamates, and carboxylic acids. The
     overall physiological function of microbial siderophores resembles that described for
     plants (Section 7.3.2); however, some details are distinctive. In an example described by
     Page (1993), ferric iron was scavenged from soil; the siderophore-bound iron was rec-
     ognized by the cell that produced the siderophore through specific receptor proteins
     located in the cell membrane; the ferri-siderophore was transported into the cell and
     deferrated by mechanisms that involved iron reduction or ligand hydrolysis.


     7.8.   Summary

         Aqueous-phase soil solution represents a conduit for dynamic equilibrium and
     transport processes that are governed by a combination of organic, inorganic,
     biochemical, and mineralogical reactions. The soil physical and chemical environ-
     ment establishes the context for metabolism, growth, reproduction, and nutrient
     uptake by plants and microorganisms. Plants and microorganisms dwell in this soil
     habitat whose chemical composition is vastly different from their own.

         To acquire key inorganic nutrients, plant roots combine three general uptake
     mechanisms (mass flow in solution, diffusion, and solid-phase interception) with
     sometimes highly selective  membrane transport carriers (active and passive) to
     regulate the intracellular composition of cellular constituents. Microorganisms also
     feature unique structural and physiological traits that allow them to contend with
     the soil environment. Microorganisms differ from plants in several major ways.
     Most soil microorganisms have heterotrophic nutrition—thus, they must acquire
     both organic and inorganic materials from soil. Also, in prokaryotic cell architec-
     ture, membrane-associated electron transport systems reside near the cell periph-
     ery. This allows some prokaryotes to use insoluble soil components such as Fe and
     Mn oxides as physiological  final electron acceptors. Endogenous (quinones) and
     exogenous (humic) substances also can facilitate extracellular electron transport
     reactions for bacteria. Another key characteristic of bacteria and fungi is their
     ability to produce extracellular enzyme systems that allow insoluble substrates,
     such as cellulose and lignin, to be degraded to constituent components prior to
     their uptake and intracellular metabolism. One remarkable ability, shared by micro-
     organisms  and plants, is an  adaptation that allows them to contend with the un-
     availability of the essential nutrient, Fe. Both plants and microorganisms acquire
     otherwise insoluble Fe3+ by producing highly specialized, organic chelating agents
     (siderophores) that solubilize mineral Fe3+ and mobilize it to the cell surface. At the

-------
                                                                               59
cell surface, the Fe-siderophore complex is destabilized so that, as the Fe is utilized
intracellularly, the siderophore cycles back in soil solution to bind more Fe.

    Key insights into the bioavailability of chemical wastes in soil may be achieved
from fundamental knowledge of the mechanisms by which microorganisms and
plants bring soil constituents into their cytoplasm. Highly specialized physiologi-
cal mechanisms for cellular uptake of some essential soil nutrients have evolved in
both plants and microorganisms. Because much microbial metabolic diversity re-
mains undiscovered (see Section 6.3), the full spectrum of microbial uptake systems
in soil has almost certainly not been described. To the degree that chemical wastes
behave as essential nutrients, one would predict that efficient contaminant uptake
could be achieved. Conversely, to the degree that the properties of chemical wastes
differ from essential nutrients, one would predict low-efficiency uptake by plants
and microorganisms.

    Apolar hydrophobic organic compounds in soil should readily pass through cell
membranes; hence, they face no biological transport barriers for cellular uptake. In
contrast, efficient uptake of polar organic and inorganic contaminant compounds can
have a low rate of passive flux into cells. Mechanisms that enhance the uptake (e.g.,
permease or ATP-driven) of polar contaminant compounds could theoretically in-
crease rates of transport into plants and microorganisms, but enhanced uptake would
not be expected unless the contaminants were fortuitously recognized by the uptake
systems.

    Plants and microorganisms have developed mechanisms for uptake of materials
from solid phases in soils; thus, entry into soil solution (by dissolution, desorption,
or exchange from soil solids) is not required for acquisition, accumulation, and/or
metabolism of chemical wastes by plants and microorganisms. For those who want
to use bioremediation processes for soil cleanup,  solid-phase substrate uptake
capabilities may provide grounds for optimism. However, a less optimistic argu-
ment also can be made. The  soil habitat poses many constraints on the uptake of
soil constituents by plants and microorganisms. The evolution of elaborate solid-
phase uptake mechanisms by soil inhabitants can be taken as long-term evolution-
ary evidence for bioavailability limitations in soil. The  subject  of bioavailability
limitations and their alleviation will be presented in Section 9 of this report.

-------
                          Section 8

        Reviewing the Facts: Examining
      Relationships Between Contaminant
       Sequestration and  Bioremediation
   A recent computerized search of the peer-reviewed literature, using "bioavail-
ability" and "bioremediation" as key words, retrieved hundreds of references that
were published over the last two decades. If the goal of this report, "a concise
compilation of current knowledge," is to be achieved, clearly much of this literature
must be passed over, while focusing on influential research of the highest quality.
But, as will become evident in Section 8.1, even a select subset of the literature on
relationships between biodegradation of organic compounds and bioavailability pro-
vides a confusing mass of seemingly conflicting data and interpretations. This sec-
tion aims to sift through, evaluate, and weigh the significance of key reports. Back-
ground information accrued in Sections 2-7 of this report and arguments developed in
Section 8.2 will serve as evaluation criteria.


8.1  Ambiguity Is the Rule: An Historical Overview of the
     Impact of Solid Surfaces on Microbial Activity

   There is  a long, substantive history of investigations that have examined how
microorganisms respond to the presence of solids—both as particles suspended in
an aqueous milieu and as an unsaturated porous geochemical matrix. Much of this
early scholarly information was summarized by Marshall (1976). This excerpt from
his book appeared under the heading "Effect of soil particles on microbial metabo-
lism" within the chapter entitled "Nonspecific interfacial interactions in microbial
ecology: terrestrial ecosystems,"

         "Adsorption of organic substrates to soils depends on the nature
      of the particulate matter, the organization of the fabric, the clay types,
      and the cation status of the soils, as well as on the concentration and
                                60

-------
                                                                                61
       molecular structure of the substrate. The availability (emphasis by
       Madsen) of substrates to soil microorganisms may be enhanced or
       reduced by the presence of soil particulate matter. This may depend
       on buffering effects.. .or on direct sorption phenomena. If the sub-
       strates are sorbed, then availability will depend on substrate location
       relative to that of the appropriate decomposer microorganisms, on
       whether extracellular enzymes are involved and whether these are
       sorbed, and on the configuration and juxtaposition of the substrates
       and enzymes in the sorbed state. Some contradictory results have
       been reported on the metabolism of specific organic substrates in
       different soils, but these probably reflect the complexity of soils as
       microbial ecosystems." (Marshall, 1976)

    Marshall's (1976) comment about "some contradictory results" will appear as a
euphemism when compared to summary statements published 14 years later by van
Loosdrecht etal. (1990). In this influential review entitled "Influences of interfaces on
microbial activity," van Loosdrecht et al. (1990) state: "Although there appears to be a
qualitative consensus that surfaces do influence bacterial metabolism, the experimental
observations are not always consistent; neither has a general explanation been ad-
vanced for this influence. The lack of experimental consistency is at least partly due to
the great variation in experimental design with respect to the nature of the solid phase,
the bacteria, the substrate, sterility, and other experimental conditions. The relevant but
disparate literature is summarized in Table 1, wherein it is confirmed that, although
generally surfaces do exert an influence, no systematic trends can be observed."

    Table 1 of van Loosdrecht et al. (1990) was entitled, "Summary of the literature
on the influence of solid surfaces on microbial behavior."  Within this table, the
following categories of observations appeared: "increased growth rate," "decreased
growth rate," "increased assimilation and decreased respiration rates,"  "increased
respiration," "decreased respiration," "increased adhesion of active cells," "higher
activity of attached cells," "decreased substrate utilization," "lower substrate affin-
ity," "change in pH optimum," "difference in fermentation pattern," "increase in
productivity," "decreased mortality," and "no effect."

    van Loosdrecht et al. (1990) offered the following concluding remarks:  "The
presence of surfaces may positively or negatively (or not at all) affect microbial sub-
strate utilization rates and growth yields. The results often depend on the nature of the
organism, the kind and concentration of substrate, and the nature of the solid surface.
In interpreting the effect of surfaces on bioconversion processes, all possible physical
and chemical interactions (e.g., diffusion, ad- and desorption, ion-exchange reactions,
conformation changes, etc.) of a given compound and its possible metabolites with a
given surface have to be considered before general conclusions can be drawn."

    Particularly germane to this report, Mihelcic  et al. (1993) published a review
entitled "Bioavailability of sorbed- and separate-phase chemicals." This compilation

-------
62
     was motivated by the view that "Hydrophobic organic compounds may be sorbed to
     soils and sediments or present in a separate phase (e.g., oil and coal tar). Conse-
     quently, the effectiveness of bioremediation of soil contaminated with organics may
     be affected by physicochemical processes that control phase partitioning between
     solid and liquid, and subsequent solute accessibility to microorganisms. Thus, the
     bioremediation of soils contaminated with hydrophobic solutes, may depend on the
     rate and extent of desorption from a solid surface or dissolution from a separate
     phase." Mihelcic et al. 's (1993) assessment of the literature began with the above
     premise and thoroughly compiled information addressing: field and laboratory ob-
     servations, associations of microorganisms with surfaces, microbial utilization of
     sorbed and separate-phase substrate, effects of surfactants on biodegradation, and
     models that combine sorption and biodegradation. Throughout their review, Mihelcic
     etal. (1993) emphasized the complexity of soil-based experimental systems, potential
     experimental artifacts, the range of microbial adaptations to biodegradation reactions,
     equivocal experimental results, and the need to carefully evaluate data.

         The first paragraph of Mihelcic et a/.'s (1993) summary section reads as fol-
     lows: "The investigation of the physical, chemical, and biological parameters affect-
     ing the biodegradation of sorbed- and separate-phase contaminants in soil-water
     systems is admittedly complex and presents an investigator with numerous experi-
     mental challenges. The analysis of the data reported from both natural and engi-
     neered systems clearly shows that the results obtained by any individual investiga-
     tor are highly dependent on the target substrate being examined, the identity and
     concentration of the organism(s), and the nature of the sorbent. To date, there seem
     to be few common principles that govern the rate of degradation of a selected pollut-
     ant However, the diversity of the sorbent-water systems examined and the variations
     in experimental design employed by investigators renders the existing studies vir-
     tually impossible to compare. Models to predict experimental observations are so-
     phisticated in their approach to describing mass transfer kinetics yet are simplistic
     in their approach to describing biodegradation and cell growth."

         Thus, ambiguity is the rule when examining the influence of solid surfaces on
     microbial activity and the relationship between bioavailabilitv and biodegradation
     in soils. It is within this established, vet confusing, scholarly context that this
     report aims for progress.


     8.2   Selection and Justification  of Criteria for Identifying the
            Highest Quality Investigations Pertinent to
            Bioavailability and Bioremediation

         This report has been designed to equip a reader with tools to decipher a vast and
     ambiguous literature onbioavailability and biodegradation. Previous sections of this
     report  serve as a prelude that establishes the foundations for this eighth section:
     Section 2 defined bioremediation and how both microorganisms and plants can influ-

-------
                                                                              63
ence soil contaminants; Section 3 defined bioavailability; Section 4 surveyed results
of bioremediation field projects; Section 5 examined mechanisms of persistence, as
illustrated especially by soil organic matter; and Sections 6 and 7 of this report have
provided a glimpse into what is and is not known about geosorbent-contaminant
interactions and the mechanisms by which microorganisms and plants acquire con-
taminants in soil.  This section of the report directly addresses the relationship(s)
between the bioavailability of chemical wastes in soil and their bioremediation.

    Previous sections of this report have emphasized the multifaceted nature of
"bioavailability." It is not an inherent property of substances under examination.
Rather, it reflects the response of experimentally defined biological systems to many
integrated processes. Bioavailability is an emergent, malleable trait that is inferred
from detailed, three-way interactions between geosorbents, biota, and chemical wastes.
The "malleable" aspects of bioavailability stem from the fact that when scientific
investigators design their experiments, intentional or unintentional choices are made.
Each choice influences the outcome of the experiments. Which geosorbent will be
used (e.g., model system or environmental sample, agricultural or industrial soil,
freshwater sediment, of high organic matter orhigh clay, contaminated or pristine)?
Which chemical waste will be studied (e.g., organic or inorganic; soluble or insoluble;
volatile; toxic; hydrophobic or hydrophilic, freshly added or historically in place)?
Which of the many possible biota (e.g., pure microbial cultures or mixed cultures,
monocotyledonous or dicotyledonous plants, soil enrichments, microbial communi-
ties from contaminated or pristine soil) will be included? How will the geosorbents,
chemical wastes, and biota be pretreated, mixed, and incubated? Are the experimental
choices for system components and incubations realistic? How environmentally sig-
nificant are the incubation conditions, the biota, and the physical/chemical states of
the contaminants? What artifacts may be created in the data-generating systems or by
the experimental designs and treatments? Can broad, sweeping principles and con-
clusions be drawn from particular sets of specific, narrowly defined experiments?

    The above types of questions and their answers begin to impinge on philo-
sophical issues about the scientific method and what it can achieve. These philo-
sophical concerns are clearly beyond the scope of this report. Nonetheless, based
on the literature reviews cited in Section 8.1, it is clear that, to some degree, the
outcome of experiments examining relationships between bioavailability and bio-
degradation are arbitrary—virtually any outcome can be achieved when particu-
lar combinations of geosorbent, contaminant, and microorganisms or plants are
selected. Given this possibility, interpretation of the multitude of reports must be
grounded in the following characteristics:

(1)  realism and environmental relevance,

(2)  absence of experimental artifacts, and

(3)  consistency of results.

-------
64
         Without such a foundation, many of the existing studies on how bioavailability
     influences biodegradation provide confusing or misleading results. Moreover,
     these criteria should help identify investigations whose results apply to society's
     real-world problems in real-world situations.

         Table 2 lists the three characteristics selected for evaluating reports to be scruti-
     nized in this section of this report (see Section 8.3 and Table 4). To identify investiga-
     tions of the highest quality (realism, absence of artifacts, consistency; first column of
     Table 2), seven specific criteria for evaluating experimental procedures will be applied
     (middle column, Table 2). The first criterion is that the investigation begin with field
     samples and field observations of the persistence of contaminants. This  guarantees
     the environmental relevance of a study. The second criterion recognizes that hypoth-
     esis testing to explain field persistence often requires manipulation of geosorbents
     and/or biota in the laboratory—but this should be done in a realistic manner, with
     minimal departure of experimental system from field conditions. The third criterion
     (realistic biota) recognizes that the microbial world is extremely diverse and respon-
     sive to laboratory-imposed conditions (Madsen, 1996,1998), thus, misleading assays
     of biodegradation and/orbioavailability may occur if the biotic component of experi-
     mental systems is unlike that found in real-world settings. The fourth criterion ac-
     knowledges that exploring the basic physiological, biochemical, and genetic details
     of biological processes requires the use of single organisms. This is the major strength
     of pure-culture studies. Yet, single-organism studies need to be tempered and inter-
     preted with realism—extrapolation from pure cultures in the laboratory to unknown
     complex microbial communities is usually unwise.

         The seventh criterion in Table 2 simply expects that meritorious investigations
     build on previous results in a way that clearly makes progress toward answering
     significant scientific and technological questions. Given the many  independent
     variables that must  be  confronted in biodegradation and bioavailability  studies
     (e.g., Sections  3.1, 6.1, 8.1), discrepancies should be expected; nonetheless, the
     validity of results from studies that contradict one another (especially if generated
     by the same laboratory) should be questioned and not be considered  of the 'high-
     est quality."

         Entries 5 and 6 in the central column of Table 2 describe criteria for reducing the
     credence given to biodegradation/bioavailability investigations based on artifacts
     that may be imposed on the geosorbent matrix under study. This is, admittedly, a
     controversial issue because compromise is often unavoidable in experimental pro-
     cedures. If an experimental method is imperfect, do the imperfections erase all valid-
     ity of the results? The answer is "probably not." Nonetheless, it is indisputable that
     the physical, chemical, and biological status of geosorbent matrices  drastically
     influence the results of biodegradation and bioavailability assays occurring therein.
     Therefore, for  the purposes  of this review, it is desirable to judge  the highest-
     quality investigations as those that avoid drastic alteration of the geosorbent ma-
     trix. Useful, supportive, confirmatory information can certainly be generated from

-------
           TABLE 2  Criteria for Identifying the Highest Quality Investigations Pertinent to Bioavailability
                                  andBioremediation (See Section8.2 for Rationale)
  Desired Characteristic
           Criteria Applicable to
   Bioavailability/Biodegradation Reports
                     Rationale
Realism and environmental
relevance
Absence of experimental
artifacts
1.  Initial data should be obtained from field
   observations of contaminants
                           2. Subsequent (e.g., laboratory-based) data
                              should be obtained without drastic chemical
                              or physical alteration of geosorbents or
                              contaminants
                              Biota used in bioassays should represent
                              those that are native to the field site
4.  When pure cultures are examined, their
   relevance and the experimental conditions
   selected (such as physiological status)
   should be defended for their environmental
   relevance

5.  Avoidance of soil sterilization procedures
1.  There is such an abundance of real-world contamina-
   tion problems, that contrived, laboratory fabrications
   of contamination scenarios may be superfluous.

2.  After documenting the field behavior of contaminants,
   hypothesis testing may need to occur via sample
   incubation in the laboratory. When brought in from
   the field, neither the geosorbents nor contaminants
   should be unduly altered, unless warranted by
   experimental objectives.

3.  When the experimental objective is to reveal
   biodegradation and/or bioavailability characteristics of
   real-world soil microbial communities, then appropri-
   ate, representative biota should be used in the
   experiments.

4.  If refined hypotheses warrant examination of pure
   microbial cultures, then care should be taken to relate
   experimental variables of pure-culture manipulations
   back to specific hypotheses and their relevance to
   field conditions.

5.  Chemical and physical treatments that allow
   sterile (abiotic) controls to be examined drastically
   alter the properties of geosorbents.
                                                                                                                                      ON
                                                                                                                                      (Ji

-------
          TABLE 2 Criteria for Identifying the Highest Quality Investigations Pertinent to Bioavailability
                           andBioremediation(See Section 8.2 for Rationale) (Continued)
                                                                                                                                    ON
                                                                                                                                    ON
  Desired Characteristic
           Criteria Applicable to
   Bioavailability/Biodegradation Reports
                     Rationale
Consistency of results
                          6. Avoidance of contaminant addition via flash
                             evaporation from volatile carrier solvents
7. Data from one set of experiments should
   support and not conflict with another
   subsequent similar set of experiments
6.  When contaminants are artificially added to
   geosorbents in highly volatile organic solvents, the
   rapid evaporation of the solvent may leave the
   contaminants in a physical/chemical state and/or
   location that is unlike that of contaminants in field
   soils. Also, carrier solvents are likely to remain in the
   geosorbents, unless they are subject to extensive
   measures that encourage evaporation.

7.  Although variability in physical, chemical, and
   biological properties of real-world samples has be-
   come an accepted means of "explaining" conflicting
   experimental results, resorting to such excuses is
   undesirable. Unresolved inconsistency may detract
   from understanding biodegradation and bioavailability

-------
                                                                                67
sterile (entry #5) or spiked (entry #6) geosorbent matrices—but in this report, such
investigations will be considered secondary, not primary, information sources.


8.2.1.  Further Scrutiny of Artifacts That May Be Caused by
        Soil Sterilization and Contaminant Addition in Organic
        Solvents

    Soil sterilization (entry #5, central column of Table 2) is required forthe prepara-
tion of abiotic controls of many experimental designs (Brock, 1978). Yet "sterilization
procedures result in some degree of alteration of soil chemical and physical proper-
ties" (Wolf and Skipper, 1994). Measures that have been used to produce abiotic
controls include chemicals that inhibit microbial metabolism (e.g., acid, azide, mercu-
ric chloride) and sterilization (e.g., by autoclaving or y-irradiation). All of these treat-
ments do far more than simply eliminate biological activity. Cawse (1975) summarized
much of the early literature on the influence of autoclaving and y-irradiation proce-
dures on soil properties and the activity of reinoculated microorganisms. Influences
include: solubilization of organic matter, release of electrolytes, formation of inhibi-
tory substances, lower pH, and many detailed alterations in the chemical composition
of soil solution. In preparing abiotic controls using low organic matter subsurface
sand, Ball (1989) found that overnight heating or overnight cooling within an auto-
clave tended to increase the sand's sorptive properties, but the effect couldbe avoided
if the sand samples were cooled rapidly after autoclaving. Autoclaving of this same
sand also was found to increase its affinity for binding hexachloroethane (Criddle et
al., 1986; MacKay etal., 1986). Xia (1998) also observed varying effects of autoclav-
ing on phenanthrene  sorption by several different subsurface materials. In a study
designed to address PAH bioavailability in soils, Sandoli etal. (1996) discovered that
y-irradiation-induced changes in geosorbent properties prevented 14C-phenanthrene
mineralization in the presence of otherwise active microorganisms. No evidence for
toxicity was evident in these y-irradiated soils. Although gamma irradiation was found
to increase the geosorbent's affinity for phenanthrene by 32 percent, a full explana-
tion of the y-irradiation-induced attention of the geosorbent was not obtained (Sandoli
etal., 1996). Unfortunately, the impacts of inhibitors and sterilization procedures on
physical, chemical, structural, and sorptive properties of soils are probably as vari-
able as the properties of soils themselves.

    Discrepancies in the properties of "freshly added" versus "field contaminated"
pollutants in  geosorbent matrices have been noted in many investigations (e.g.,
Alexander, 1999;Burforde/a/., 1993 ;Hatzinger and Alexander, 1995; Steinberg et
al., 1987). However, seldom have the two distinctive issues pertinent to these
discrepancies been discussed. The issue that has  perhaps received greatest atten-
tion to date is that the physical, chemical, and biological behavior of freshly added
contaminants may change over time. Therefore, freshly added contaminants may
not mimic the behavior of "aged material." It has been postulated that the freshly

-------
68
     added contaminant pool may not have had sufficient time to diffuse into micro- or
     nano-pores or be absorbed into organic matter (e.g., Alexander, 1995; Luthy et al.,
     1997; Reid et al., 2000). Thus, an "aging effect" has been used to argue for kineti-
     cally constrained, time-dependent sequestration of environmental contaminants.

         The second related but distinctive issue for freshly added contaminants (entry
     #6, central column of Table 2) is that the way they are added to geosorbents may
     leave them in locations and in physical/chemical states that differ markedly from
     their field-derived analogs. When small volumes of contaminant-containing vola-
     tile solvents are added ("spiked") to soil and allowed to evaporate, then crystals of
     the contaminant solute may be left behind in the soil matrix. It is unlikely that these
     crystals,  even if later distributed uniformly throughout the soil via mechanical
     mixing, will initially interact with the soil matrix in ways that mimic the interactions
     found in field sites (unless the field sites happen to be  contaminated via similar
     flash-evaporation-type procedures; S. Hawthorne, personal communication). Not
     only are the flash-evaporated spiked contaminants likely to be in states and asso-
     ciations and locations unlike those in the field, but complete removal of the carrier
     solvent requires heat and vacuum because the solvent's effective boiling point can
     be raised in small pores (J. Pignatello, personal communication). Without such
     directed effort to remove residual solvents from spiked soil, they may have unan-
     ticipated effects on soil organic matter (e.g., swelling that may be irreversible; J.
     Pignatello, personal  communication), contaminants, and/or microbial processes.
     Burford et al. (1993)  carried out an extensive comparison of the extraction efficien-
     cies of freshly added, deuterated PAHs with corresponding nondeuterated PAHs
     from historically contaminated soil and sludges. Burford etal 's. (1993) data consis-
     tently showed that freshly added PAHs (dissolved in dichloromethane and flash
     evaporated) were far more extractable than the native PAHs. The discrepancy be-
     tween freshly added and long-aged PAHs was greatest for the low molecular weight
     PAHs (e.g., naphthalene). Burford et al. (1993) concluded that "it is experimentally
     impossible to reproduce the environmental conditions that occur during deposi-
     tion of pollutants in real-world samples."

         This comment by  Burford etal. (1993) is particularly germane to the interpretation
     of data describing "aging" of contaminants in geosorbents. Flash-evaporated surro-
     gate contaminants may be dispersed in laboratory-incubated soil experiments. How-
     ever, once in place, the contaminants will undergo dissolution, diffusion, partitioning,
     and other reactions governed by a combination of contaminant properties, the soil,
     and the incubation conditions. The many investigations describing "aging effects,"
     noted above, have documented time-dependent alteration in biological and/or chemi-
     cal properties of the  contaminants. These investigations have not shown that the
     time-dependent (perhaps asymptotic)  change approaches the same endpoint that is
     found in field-contaminated soils. After all, the real-world weathering processes (wa-
     ter infiltration, freeze/thaw, wind, sunlight, etc.) that occur in uncontrolled field sites
     are vastly different from the typical conditions found in laboratory incubations. It is

-------
                                                                              69
possible that contaminants spiked with geosorbents and subsequently aged in the
laboratory eventually attain the same physical, chemical, and biological properties as
those found in the field. Indeed, "field contaminants" probably occur in a broad
spectrum of states, and are almost certainly neither uniform nor static. However, it is
reasonable to presume that both the rate at which the "field state" is asymptotically
approached, and the asymptote itself, are influenced by the condition of the geosorbent
and the incubation. To the degree that these two variables deviate from real-world
conditions, the state of laboratory-aged contaminants also may deviate from those
found in real-world soils. Currently, there is not enough information addressing the
above-described uncertainties about sterilization, spiking, and aging procedures
and their impact on contaminants in geosorbents. Until the relationships between
field- and laboratory-contaminant disposition in soils are better understood, it will
remain a challenge to interpret laboratory-based experiments that attempt to test
hypotheses about field processes using sterile, spiked soils.


8.3  Scrutinizing Selected  Investigations Describing the
      Bioavailability of Contaminants and Their
      Biodegradation

    To some degree, M. Alexander has become an advocate  for incorporating
bioavailability considerations into regulatory and/or lexicological interpretation of
environmental contamination. Therefore, it seems appropriate to examine and scruti-
nize his recent, relevant publications. In 1995 and 1997, Alexander published critical
reviews that advanced the idea that contaminant bioavailability in soil may be the
most  relevant criterion for "assessing toxicity, determining risk, and establishing
meaningful regulations for clean up of sites containing hazardous wastes." Alexander
presents six lines of evidence to support the proposition that organic compounds
become increasingly sequestered (less bioavailable) as their residence times in soils
increase. These lines of evidence appear in Table 3, along with alternative interpreta-
tions of the "evidence." The alternative interpretations (second column) for evidence
presented in the first column of Table 3 simply apply the background knowledge and
principles that have been established in earlier sections of this report. Although all of
the lines of evidence are plausible, they also need to be carefully reexamined for the
possibility of alternative interpretations and also for the quality of supporting experi-
mental data. As indicated in the second column of Table 3, there are many potential
weaknesses in the experiments supporting the six lines of evidence, including: (1)
many factors besides diminished bioavailability can contribute to the long-term per-
sistence of pesticides in soils; (2) observations in the field combined with laboratory
experiments (using soils that have been contaminated for extended periods to exam-
ine the cause of contaminant persistence) may be the most robust type of evidence,
given the quality of the data; (3) laboratory aging studies may be rich in artifacts,
especially if sterile, spiked soils are used (see Section 8.2.1); (4) chemical extraction
experiments that clearly demonstrate distinctions between aged and freshly contami-

-------
     TABLE 3 Lines of Evidence for Time-Dependent Decline inBioavailability of Organic Soil Contaminants
                         (fromAlexander, 1995,1997) and Possible Alternative Interpretations
Evidence
Alternative Interpretations
(i)  A "hockey stick"-like profile of concentration versus time for
    pesticides in field soils. The initial rapid loss of pesticides,
    followed by a plateau, suggests increased sequestration over
    time.
(ii)  Laboratory studies of field samples contaminated with
    pesticides and PAHs in which native soil microorganisms were
    able to biodegrade a freshly-added, but not an "aged" pool of
    these compounds.

(iii) Laboratory "aging" studies in which the degree of biodegrada-
    tion by a microbial inoculum was governed, inversely, by the
    duration of prior contact between organic chemicals and sterile
    soil.

(iv) Chemical extraction assays showing that the degree of
    sequestration of organic compounds in soil is proportional to
    the duration of compound-soil contact.


(v)  Analysis of the kinetics of contaminant desorbtion from soils
    that suggested the presence of recessed diffusion-limited
    binding sites within pores of soil particles or within organic
    matter phases.
(1)  Most data were from chlorinated pesticides that are not readily
    biodegradable because they do not serve as carbon and energy
    sources for microorganisms. Alternative hypotheses for
    explaining persistence of pesticides (such as lack of cosubstrate,
    simple first order kinetics, nutrient limitations, other limiting
    environmental factors) were discussed in Sections 4.0 and 5.2.
    Without considering all alternative hypotheses, the
    "bioavailability" hypothesis cannot be adequately tested.

(ii)  This evidence can be very convincing (see Table 4)
(iii) Conditions for laboratory "aging" may be artifactual, especially
    if sterile soils and contaminants are added via flash-evaporated
    organic solvents (see Section 8.2).


(iv) This evidence may or may not be convincing (see Section
    8.2.1). A change per se is not adequate. The change in
    contaminant extractability must be toward an environmentally
    relevant endpoint.

(v) This evidence may or may not be very convincing. The test
    systems are highly complex and variable, hence, susceptible to
    many alternative interpretations (see Sections 6.3 and 6.5).
    Validity depends on details of specific procedures in specific
    investigations.

-------
     TABLE 3 Lines of Evidence for Time-Dependent Decline inBioavailability of Organic Soil Contaminants
                 (fromAlexander, 1995,1997) and Possible Alternative Interpretations (Continued)
Evidence
Alternative Interpretations
(vi) Bioassay s showing that chemicals residing in soil have reduced
    toxicity and this reduction may be proportional to the
    chemical's soil residence time.
(vi) This evidence may or may not be very convincing. The test
    systems are highly complex and variable, hence, susceptible to
    many alternative interpretations (see Sections 6.3 and 6.5).
    Validity depends on details of specific procedures in specific
    investigations, the bioassay, and the specific mechanism of
    toxicity.

-------
72
     nated soils are insightful, though scrutiny of spiking and other procedures is war-
     ranted (see Section 8.2); (5) macroscopic observations of the interactions between
     contaminants and geosorbents have many varied conceptual interpretations (Sec-
     tions 6.1 and 6.5); and (6) diminished contaminant toxicity in the presence of soil
     solids has many alternative explanations, depending on the bioassay, the soil, and
     the mechanism of toxicity.

         The quality of experimental data (e.g., field observations, type of model sys-
     tems, potential experimental artifacts; general conformity to the seven criteria in
     Table 2) is of crucial importance in evaluating all "lines of evidence" for or against
     a reduction of contaminant bioavailability in soils. Another major consideration is
     whether or not "statistically significant" differences between experimental treat-
     ments reported in the peer-reviewed literature have real-world ramifications. Bavey e
     and Bladon (1999) pointed out that small, though statistically significant short-term
     treatment differences, may be moot when extrapolated to real-world timeframes of
     months to years.

         A final, key element to include is provided by an organic substance whose
     persistence is well established—soil organic matter. How can the behavior of soil
     organic matter enlighten our understanding and expectations for the behavior of
     chemical wastes in soil?

         Table 4 contains a compilation of some of the most influential and/or recent
     experimental studies that have been published to date on the relationship between
     bioavailability of organic compounds and bioremediation. The columns in Table 4
     (Goals, Compound, Geosorbent, Approach/Methods/Data, Conclusions, Method-
     ological Weaknesses) provide a format to prioritize and evaluate a study using the
     criteria established in Table 2.

         The first entry in Table 4 (Steinberg et al., 1987) is perhaps the most convincing
     study to document the relationship between an aged organic chemical waste and its
     availability for bioremediation. Unlike the majority of entries in Table 4, the study by
     Steinberg et al., (1987) suffers from no obvious methodological weaknesses—it sat-
     isfies all seven criteria in Table 2 for high-quality data. Steinberg et al. (1987) began
     withfield observations that the fumigant, ethylene dibromide (EDB), had persisted in
     field soils for an unexpectedly long time. Subsequent laboratory experiments used
     chemical, physiological, and physical measurements to prove that aged, field-applied
     EDB was notbiodegradedby the native soil microbial community; yet, freshly added
     radiolabeledEDB was readily metabolized. Furthermore, measurements of the rate of
     release of the aged EDB from soil revealed that it was kinetically retarded and that
     release could be accelerated by pulverizing the soil. Entrapment  of EDB in soil
     micropores was inferred. The study by Steinberg et al. (1987) did not use sterile soils
     or flash-evaporated "spiked" soils. Moreover, the study was thorough, internally
     consistent, and provided convincing data supporting logical arguments for a causal
     relationship between diminished bioavailability and diminished biodegradation.

-------
TABLE 4 Survey of Laboratory and Field Studies Aimed at Discerning Mechanistic Relationships Between the Biodegradation
                               of Organic Compounds and Their Bioavailability in Soil
Goals
1. Explain
anomalous
persistence of
volatile,
biodegradable
pesticide in
field soils



















Compound
1,2-Ethylene
Dibromide
(EDB),agas
phase-applied
fumigantwith
aqueous
solubility of
3,370 ppm


















Geosorbent
Four tobacco
farm soils
treated with
EDB in the field
0.9 to 19 years
prior to analysis




















Approach/
Methods/Data
Use chemical and
microbiological assays
to assess iflong-aged
(native) EDB was
distinctive from
freshly added EDB

Freshly added EDB
was volatile; aged
EDB was not

Aged EDB did not
readily equilibrate
with aqueous
solution. Rates of
release of aged EDB
from soil into
aqueous solution
were temperature
dependent.
Diffusion-based
modeling of EDB
release to aqueous
solution suggested
2-3 decades to reach
50% of predicted
equilibrium values.
Conclusions
Residual EDB was
highly resistant to
both mobilization
andmicrobial
degradation

There is slow
exchange between
"native," aged EDB
and the freshly
added chemical

Pulverization of the
soil drastically
accelerated chemical
release of EDB
from soil
Entrapment of EDB
in micropores was
inferred






Methodological*
Weaknesses


























Reference
Steinberg et al,
1987

























-------
TABLE 4 Survey of Laboratory and Field Studies Aimed at Discerning Mechanistic Relationships Between the Biodegradation
                         of Organic Compounds and Their Bioavailability in Soil (Continued)
Goals










2. Construct,
operate, and
manageafield-
scale, prepared
bed, land
treatment unit
for 2 soils;
assess
performance
based on
contaminant
mobility,
toxicity,
concentration;
determine
treatment to
Compound










Crude oil (aged in
field 30 to 40
years)












Geosorbent










Clay soil from
Texas gas storage
facility












Approach/
Methods/Data
The biodegradation
tests showed that
freshly added 14C-
EDB was readily
converted to a
combination of
14CO2 (46 %) and
cell material (54 %),
while the long-aged
EDB was not
Use soils aged in the
field; treat with
tillage, nutrients,
water (clay soil, 55
weeks or silty sand,
34 weeks)

In clay soil, long
contaminated with
crude oil, total
petroleum hydrocar-
bons (6,400 to
11,000 ppm) show-
ed no decline and
leachability declined
from 4.4 to 1.3 ppm
Conclusions










After field treatment,
laboratory
incubations (in
which oxygen and
nutrients were added
and viable
petroleum-degrading
microbial popula-
tions were found)
failed to stimulate
biodegradation

This, and absence
of inhibitory salts
or metals, suggests
that bioavailability
Methodological*
Weaknesses

























Reference










Oliverae/
o/.,1998














-------
TABLE 4 Survey of Laboratory and Field Studies Aimed at Discerning Mechanistic Relationships Between the Biodegradation
                         of Organic Compounds and Their Bioavailability in Soil (Continued)
Goals
reach
environmen-
tally accept-
able endpoint

















3. Use labora-
tory protocols
to quantify
Compound









Diesel fuel (aged
in field 3 years)










Crude oil (aged
in field 30 to 40
years)
Geosorbent









Silty sand from
Texas power
plant









Clay soil from
Texas gas storage
facility
Approach/
Methods/Data
total petroleum
hydrocarbons
(TPHs). Field
leachate ranged
from 0.1 to 75
ppm, and only a
trace of toxicity
(microtox assay)
was found
In the silty sand
contaminated with
diesel fuel, TPH
declined from 1 ,1 00
to 160 ppm,
aqueous release
(TCLP) ranged from
1.5 to 3. 2 ppm, field
leachate ranged from
1.1 to 7.8 ppm, and
only a trace of
toxicity was found.
Build on results of
Oliverarfa/.(1998)
by using laboratory
Conclusions
limited microbial
activity







"Active, prepared
bed treatment of
about 15 weeks
reached an
environmentally
acceptable
endpoint"





No degradation of
hydrocarbons
occurred, nor could
Methodological*
Weaknesses
























Reference





















Bergetal.,
1998


-------
                                                                                                                               ON
TABLE 4 Survey of Laboratory and Field Studies Aimed at Discerning Mechanistic Relationships Between the Biodegradation
                         of Organic Compounds and Their Bioavailability in Soil (Continued)
Goals
rate of release
of petroleum
hydrocar-
bons;
investigate
impacts of
bioremediation
on hydrocar-
bon release
from soils;
explore
possibility
that biore-
mediation is
limited by
hydrocarbon
release








Compound





















Diesel fuel (aged
in fields 3 years)


Geosorbent





















Silty, sand from
Texas power
plant

Approach/
Methods/Data
measurements of
the rate of release
(ROR) of
petroleum
hydrocarbons from
bioremediated field
oils and their
residual fraction

The ROR
procedure used
XAD2 resins as a
constant sink for
contaminants in
long-term aqueous-
phase desorption
studies. The released
contaminants were
subsequently
analyzed by GC/
MS




Conclusions
releases of
hydrocarbon be
quantified; "Lack of
biodegradation was
related to inability
of chemicals to be
released from the
soils"

"This indicated low
bioavailability of
hydrocarbon"









Bioremediation by
land treatment
diminished the
mass of hydrocar-
Methodologicar
Weaknesses





















Data analysis
yielded some
anomalous
(impossible)
Reference


























-------
TABLE 4 Survey of Laboratory and Field Studies Aimed at Discerning Mechanistic Relationships Between the Biodegradation
                         of Organic Compounds and Their Bioavailability in Soil (Continued)
Goals





4. Determine the
limits and
extent of
hydrocarbon
biodegradation
toxicity and
leachabilityin
petroleum
contaminated
soil









Compound





Three different
weights of crude
oil, artificially
weathered at 5 %
(wt/vol) freshly
added












Geosorbent





Two soils, one
low, one high in
organic matter,
initially
uncontaminated













Approach/
Methods/Data





Implement an
8-11 month
bioremediation
treatment (of moist,
nutrient-amended
soil) and then
perform chemical
analyses, toxicity,
and leachability
assays. Depending
on soil, weight of oil
and hydrocarbon
chain length,
between 10 % and
88 % of the initial
oil was biodegraded.
BTEX was almost
always below
detection
Conclusions
bons released from
soil; an initial rapid
release was
followed by a
slow-release phase
Biotreatment
reduced mobility
and eliminated
toxicity of residual
hydrocarbons

Residuals were not
toxic or leachable,
or biodegraded
further









Methodological*
Weaknesses
modeling
parameters from
the contaminant
release curves

Both contamina-
tion and
bioremediation
events were
conducted in the
laboratory












Reference





Salanitroe/a/.,
1997

















-------
TABLE 4 Survey of Laboratory and Field Studies Aimed at Discerning Mechanistic Relationships Between the Biodegradation
                         of Organic Compounds and Their Bioavailability in Soil (Continued)
Goals











5. Test the
hypothesis
that slow
desorption is
the cause of
limited
biodegrada-
tion; and if
this is the
case,
investigate
prediction of
bioremediation
Compound











Fifteen PAHs











Geosorbent











Two PAH-
contaminated
sediments from
Amsterdam
Harbor







Approach/
Methods/Data
Biotreatment caused
drastic reduction in
leachableBTEX,
eliminated toxicity
to earthworms, and
diminished
inhibition of seed
germination
Plant growth assays
of toxicity revealed
stimulation by oil
Desorption kinetics
of 15 PAHs were
determined before
and after biotreat-
ment via bioreactors
(4 months) or land
farming (2 years)
Desorption kinetics
were measured by
trapping on Tenax
resins
Biotreatment
Conclusions











For readily
metabolized PAHs,
the extent of
possible PAH
degradation could
be roughly
predicted from the
initially desorbing
fractions

However, a pool of
PAHs was
desorbed but not
Methodological*
Weaknesses























Reference











Cornelissene/a/.,
1998











-------
TABLE 4 Survey of Laboratory and Field Studies Aimed at Discerning Mechanistic Relationships Between the Biodegradation
                         of Organic Compounds and Their Bioavailability in Soil (Continued)
Goals
based on
desorption
kinetics







6. Test
hypotheses
for explaining
the long
persistence of
naphthalene
in sediments





Compound










Naphthalene
(sparingly
soluble,
31 ppm)








Geosorbent










Sandy, organic
matter-rich
surface
sediments








Approach/
Methods/Data
reduced the rapidly
desorbing PAH-
fraction

A pool of non-
degrading PAHs
also desorbed; but
this pool remained
unchanged by
biotreatment
Use chemical and
microbiological
assays to attempt
to identify factors in
field that prevent
biodegradation; GC
analysis of field
samples, 14CO2
production from
14C- naphthalene
added to laboratory-
incubated samples;
Conclusions
metabolized,
probably due to
microbial factors,
not lack of
bioavailability





In laboratory
incubations,
naphthalene
metabolism was
oxygen-limited, yet
H-£>2 addition in
the field did not
stimulate naphtha-
lene loss in situ; no
evidence for
nutrient limitation
was obtained; soil
Methodological"
Weaknesses










y-irradiated soils
were used in the
aging study; the
laboratory aging
period was
relatively short (4
weeks); chemical
analysis of field
samples
introduced much
variability in an
attempt to
Reference










Madsen et al.,
1996











-------
                                                                                                                               oo
                                                                                                                               o
TABLE 4 Survey of Laboratory and Field Studies Aimed at Discerning Mechanistic Relationships Between the Biodegradation

                         of Organic Compounds and Their Bioavailability in Soil (Continued)
Goals










7. Compare
bioavailability
and desorp-
tion kinetics
ofweathered
(aged) and
recently added
herbicide







Compound










Simazine
(chlorinated
triazine herbicide
that is not a
microbial carbon
and energy
source; sparingly
soluble in water,
3.5 ppm)






Geosorbent










Agricultural soil
receiving
simazine
annually for
more than 20
years









Approach/
Methods/Data
amendment of
sample, with
nutrients, aging of y-
irradiated soils
followed by
mineralization
assays, addition of
FLp2 to field
sediments followed
by GC analysis
Sorption coefficient
for aged simazine
was 15 times that of
freshly added;
simazine concentra-
tion in soil solution
in field soils showed
that the aged
material was far
below aqueous
equilibrium
concentration;
freshly added
simazine was toxic
to sugar beet
Conclusions
slurries enriched in
sorbed naphthalene
metabolized the
aged substrate; no
clear evidence to
distinguish oxygen
limitation from
sequestration as the
cause of persistence

Pesticide aging
diminished
bioavailability, as
measured by
microbial degrada-
tion and plant
uptake

Desorption kinetics
of aged simazine
were slow com-
pared to the
recently added
chemical

Methodological*
Weaknesses
document
biodegradation zw
situ; different
enrichment
cultures reacted
differently to aged
substrates



Investigators did
not clearly
describe the
means by which
biodegradation of
freshly added 14C-
simazine was
discerned from the
total simazine
pool

Sorption
coefficients for
14C simazine were
measured with
Reference










Scribneretal.,
1992














-------
TABLE 4 Survey of Laboratory and Field Studies Aimed at Discerning Mechanistic Relationships Between the Biodegradation
                         of Organic Compounds and Their Bioavailability in Soil (Continued)
Goals







8. Determine the
effect of aging
time in soils
on the
biodegra-
dability and
extractability
of organic
compounds







Compound







Phenanthrene
(sparingly soluble
1.3 ppm) and 4-
nitrophenol
(readily soluble in
water)










Geosorbent







3 soils of varying
properties
(agricultural soil,
muck soil,
aquifer sand)











Approach/
Methods/Data
seedlings, aged
simazine was not





Using y-irradiated,
sterile, spiked soils,
the contact time
between geosorbent
and the two 14C-
labeled test contami-
nants was varied
between 0 and 31 5
days; then pure
cultures of bacteria
were inoculated into
the geosorbent and
14CO2 evolution
from the contami-
nants was
measured;
phenanthrene was
Conclusions







"The data show
that phenanthrene
and 4-chlorophenol
added to soil
become increasingly
more resistant with
time to degradation
and extraction."
Rates and extents of
mineralization were
inversely propor-
tional to the aging
duration



Methodological*
Weaknesses
DMSO as carrier
for the simazine
(though simazine
soil solution data
were genuine,
carrier-free, field
measurements)
y-irradiation of
soil; phenan-
threne contami-
nant was added
in an organic
solvent; pure
cultures were
used in
biodegradation
tests







Reference







Hatzinger and
Alexander, 1995















-------
TABLE 4 Survey of Laboratory and Field Studies Aimed at Discerning Mechanistic Relationships Between the Biodegradation
                         of Organic Compounds and Their Bioavailability in Soil (Continued)
Goals







9. Use a
nonsterile soil
system to
assess
biodegradabil-
ity (bioavail-
ability) and
extractability
ofdifferen
Compound







Phenanthrene

Geosorbent







Soil from Sierra
Nevada foothills

Approach/
Methods/Data
spiked into soil
using dichloro-
methane as a carrier,
4-chlorophenol was
added in water;
simultaneous to
14CO2 evolution
assay, the labeled
compounds were
extracted from soil
using butanol and
soxhlet procedures;
sonicationofsoil
aggregates slightly
enhanced mineral-
ization
Age (0 to 600
hours) phenan-
threne in a nonsterile
soil with little
biodegradation
activity, then add
active inoculum, and
analyze 14CO2
evolution kinetics,
Conclusions







Both the hexane
extracts and
mineralization
curves showed a
rapid (200 h) de-
cline in the
availability of
phenanthrene; mild
extraction
Methodological*
Weaknesses







Short, aging
periods; phenan-
threne was added
in methylene-
chloride organic
solvent
Reference







Schwartz and
Scow, 1999


-------
TABLE 4 Survey of Laboratory and Field Studies Aimed at Discerning Mechanistic Relationships Between the Biodegradation
                         of Organic Compounds and Their Bioavailability in Soil (Continued)
Goals
tiallyaged
phenanthrene




10. Examine
competitive
displacement
ofaged
phenan-
threne by
pyrene in
chemical
extraction and
biodegrada-
tion assays








Compound






Phenanthrene
















Geosorbent






Agricultural soil
and peat soil
















Approach/
Methods/Data
as well as phenan-
threne extractability




Mineralization and
chemical extractabil-
ity ofaged
radiolabeled
phenanthrene (3 to
123 days) in two
sterile soils, with
and without added
pyrene were
measured

Soils were sterilized
by y-irradiation;
14C-phenanthrene
was coated onto the
wall of a flask and
then allowed to
equilibrate through
the aqueous phase
with soil.
Conclusions
procedures reflect
bioavailability;
sorption processes
may prolong the
presence of a
chemical in soil
Aging diminished
rate and extent of
mineralization;
pyrene addition
enhanced phenan-
threne mineraliza-
tion; phenanthrene
extractability and
Kd values also were
altered by pyrene;
the pyrene
influences may have
been caused by
competitive
displacement of the
phenanthrene from
glassy organic
matter


Methodological*
Weaknesses






y-irradiation of
soil; a pure
culture was used
inbiodegradation
tests














Reference






White et al, 1999

















-------
TABLE 4 Survey of Laboratory and Field Studies Aimed at Discerning Mechanistic Relationships Between the Biodegradation
                         of Organic Compounds and Their Bioavailability in Soil (Continued)
Goals



11. Model
coupled
processes of
desorbtion
and
mineralization
using two
distinctive
model
organisms


12. Determine
how
microbial
activity
influences
associations
between soil
organic
matter and
Compound



Naphthalene (a
sparingly soluble
hydrophobic
compound; 31
ppm)







Pyrene (solubility
0.13 ppm in
water)






Geosorbent



Agricultural
soils — dried,
ground, and
autoclaved








A forest soil and
an inoculum from
a pyrene-
contaminated
Superfund site




Approach/
Methods/Data
Pyrene was added in
methanol as a carrier
(<0.1% by volume)
Naphthalene
mineralization by
two different
bacterial cultures
was examined in
aqueous solution
containing varying
amounts of added
soil



The distribution of
14C- pyrene and
byproducts were
determined by
sequential soil
extraction
procedures from
forest soil, with or
without a pyrene-
Conclusions



For one bacterium,
sorption limited the
rate and extent of
slurry-phase
naphthalene
mineralization. This
was not the case for
the other organism,
which metabolized
a significant portion
ofsorbed
naphthalene
Over time, the
extractability of
pyrene and pyrene
products decreased
to a greater extent in
themetabolically
active soil

Biological activity
Methodological*
Weaknesses



Soils were air-
dried, ground, and
autoclaved. 14C-
naphthalene was
added to soil
slurries in acetone
as initial carrier;
pure cultures were
examined



Methanol was
used as the pyrene
solvent

It is difficult to
interpret the
meaning of
extractability
changes in such a
Reference



Guerinand
Boyd, 1992










Guthrie and
Pfaender, 1998








-------
TABLE 4 Survey of Laboratory and Field Studies Aimed at Discerning Mechanistic Relationships Between the Biodegradation

                          of Organic Compounds and Their Bioavailability in Soil (Continued)
Goals
contami-
nants









13. Describe the
potential
release and
subsequent
degradation
of bound 14C
residues from
soil
previously
treated with
atrazine



Compound











Atrazine, water
soluble
chlorinated
triazine herbicide
that was not a
ready carbon and
energy source for
soil micro-
organisms





Geosorbent











Agricultural soil













Approach/
Methods/Data
degrading inoculum
or a metabolic
inhibitor.
Bioavailability was
assessed by 14CO2
evolution from
radiolabeled pyrene
added to soil in
constantly flushed
incubation
chambers
Form 14C-atrazine
bound residues in
soil via a 1-year
laboratory
incubation;
exhaustively extract
with methanol to
remove residual 14C;
chemically
characterize the
residual 14C in soil;
inoculate the soil
with two different
atrazine-degrading
Conclusions
may reduce
biotoxicity. The
amount of 14CO2
released from
freshly added and
aged (270 days)
pyrene was similar




During an 84-day
incubation, atrazine
and four metabo-
lites were released
from the initially
soil-bound 14C
atrazine; different
bacteria released
hydroxyatrazine to
different extents;
soil bound residues
can be released


Methodological*
Weaknesses
complex
experimental
system








Soil sample
prepared by
exhaustive
extraction with
methanol; use of
pure cultures








Reference











Khan and Behki,
1990












                                                                                                                                oo
                                                                                                                                (Jl

-------
                                                                                                                              oo
                                                                                                                              ON
TABLE 4 Survey of Laboratory and Field Studies Aimed at Discerning Mechanistic Relationships Between the Biodegradation

                         of Organic Compounds and Their Bioavailability in Soil (Continued)
Goals





14. Examine the
fate of soil-
aged
nonextract-
able atrazine
with respect
to biodegra-
dation,
microbially
facilitated
release, and
abiotic
desorption





Compound





Atrazine (water
soluble)










Geosorbent





Agricultural silt
loam










Approach/
Methods/Data
bacteria with and
without a glucose
amendment; assess
metabolism of the
bound 14C
After 3 months of
aging atrazine in
sterile soil, the
extractable atrazine
was removed. Next,
the soil containing
residual atrazine
was subjected to
mineralization and
desorption tests
Almost no (<2.5%)
atrazine mineraliza-
tion occurred
Viable microorgan-
isms did not
enhance release of
previously bound
atrazine
Conclusions





Bound residues are
subject to
desorption and
release from soils

Rates of atrazine
mineralization were
exceeded, in nearly
all cases, by the
desorption rates —
suggesting no mass
transfer limitations






Methodological*
Weaknesses





y-irradiated soils
were used to age
(bind) the atrazine
in soil

Test substrate
(atrazine) was not
appreciably
metabolized, thus
the bioavailability
assay was not
robust





Reference





Johnson eta!.,
1999











-------
TABLE 4 Survey of Laboratory and Field Studies Aimed at Discerning Mechanistic Relationships Between the Biodegradation
                         of Organic Compounds and Their Bioavailability in Soil (Continued)
Goals
15. Compare
biodegrada-
tion of
sorbed- and
solution-
phase
compound in
soil


16. Assess the
feasibility of
using
chemical
extraction
procedures
to estimate
bioavailability
of pesticides
to earth-
worms



Compound
2,4-D (a rea-
sonably water
soluble
compound; 900
ppm)





DDT, DDE,
DDD (sparingly
soluble chlori-
nated insecticide
or byproduct)









Geosorbent
Two agricultural
soils and one
clay, all sterilized
by autoclaving






Seven agricultural
soils of varying
properties and
pesticide
application
histories








Approach/
Methods/Data
Three separate
mathematical
models were used
to evaluate the
mineralization of
2,4-D added to soil
and metabolized by
an inoculated pure
culture

Determine pesticide
concentrations via
soxhlet extraction/
analysis; measure
earthworm uptake
of pesticides; co-
rrelate earthworm
uptake with CIS
membranes and
25% tetrahydrofu-
ran extracts; age
pesticide (0 to 924
d)iny-irradiated
soils; pesticides were
Conclusions
The model that
best fit the data
presumed that
sorbed 2,4-D was
unavailable





Worms assimilated
3% to 66% of the
pesticides; this
uptake correlated
well with chemical
extraction assays.
Correlation
coefficients were
0.921 or higher for
the C 1 8 membrane
and 0.83-09 .48 for
the tetrahydrofuran
extraction

Methodological*
Weaknesses
Soils were
autoclavedand
ground; a pure
culture was used;
2,4-D was freshly
added to soil with
methanol as initial
carrier. A 5-h
incubation time
was used
y-irradiation of
soil; addition of
pesticides in
organic solvent;
incompletely
vented








Reference
Ograme/a/.,
1985








Tang et al, 1999














-------
                                                                                                                              oo
                                                                                                                              oo
TABLE 4 Survey of Laboratory and Field Studies Aimed at Discerning Mechanistic Relationships Between the Biodegradation

                         of Organic Compounds and Their Bioavailability in Soil (Continued)
Goals




17. Assess the
decline in
bioavailability
as a result of
aging of
pesticides












Compound




DDT, DDE,
ODD, dieldrin















Geosorbent




Three agricultural
soils of varying
properties and
pesticide
application
histories












Approach/
Methods/Data
added in
dichloromethane
solvent allowed to
volatilize for 48 h
Use an earthworm
uptake bioassay to
measure pesticide
bioavailability in
field soils and
laboratory-aged
soils; pesticide-
uptake assays
involved exposing
worms to soil, then
harvesting,
extraction, and GC
analyses; field aged
soils (49 and 30-
year periods) were
used; laboratory-
aged soils (0, 90,
1 90 days) were y-
irradiated and
pesticides were
Conclusions




Aging materials
reduces
bioavailability
Soil extraction by
Tenax resin
correlated well with
earthworm uptake
for three of four
pesticides










Methodological*
Weaknesses




y-irradiation of
soil; addition of
pesticides in
organic solvents,
incompletely
vented
In many
instances, the
data showed that
aging had no
effect on
bioavailability







Reference




Morrison et al,
2000
















-------
  TABLE 4 Survey of Laboratory and Field Studies Aimed at Discerning Mechanistic Relationships Between the Biodegradation
                           of Organic Compounds and Their Bioavailability in Soil (Continued)
Goals









Compound









Geosorbent









Approach/
Methods/Data
added in organic
solvents; chemical
analysis of the
pesticides adsorbed
to Tenax resin was
examined as a
possible surrogate
for earthworm
uptake
Conclusions









Methodological*
Weaknesses









Reference









* Departure from the 7 (seven) quality criteria in Table 2

-------
90
         The next two entries in Table 4 were conducted by R. Loehr's research group at
     the University of Texas. Like Steinberg etal. (1987), afield approach was taken, and
     subsequent hypotheses were sequentially tested to try to explain the persistence
     of organic soil contaminants subjected to bioremediation processes; no method-
     ological weaknesses could be discerned. For Olivera et al. (1998) and Berg et al.
     (1998), the contaminants were petroleum hydrocarbons (crude oil and diesel fuel)
     that had been in the field for either 3-4 decades or 3 years before being field
     bioremediated (land treatment) for either 55 or 34 weeks. Chemical analysis of the
     field-bioremediated soils was used to monitor process effectiveness. Simultaneously,
     the mobility (teachability) of the residual hydrocarbons and their toxicity were
     measured. The behavior of the different hydrocarbons (aged in two different field
     soils for different periods) was not uniform: land treatment reduced residual hydro-
     carbons in the soil contaminated for 3 years, but not for 3 to 4 decades. Nonethe-
     less, low hydrocarbon teachability and toxicity were documented for both soils.
     This suggested  to the authors that "environmentally acceptable endpoints had
     been reached." Furthermore, alternative hypotheses for explaining the field persis-
     tence of the hydrocarbons (e.g., nutrient limitations, lack of viable microbial popu-
     lations) were tested and dismissed. Thus, diminished bioavailability was implicated
     as the cause for persistence of hydrocarbons in biotreated field soils.

         Another study addressing bioavailability and biodegradation of petroleum hydro-
     carbons in soils was that of Salanitro et al. (1997; entry number four of Table 4). The
     approach was to freshly contaminate soil with crude oil so that the naturally occurring
     microbial populations could biodegrade it. Then chemical analyses, as well as leachabil-
     ity and toxicity (earthworm and seed germination) assays, were performed to character-
     ize the residual hydrocarbons. This study's only (minor) weakness was that it was done
     in a laboratory—it did not include true field observations of persistence orbiodegrada-
     tionactivity. Nonetheless, Salanitro etal's. (1997) study met most of the criteria in Table
     2 (real-world microbial populations, absence of artifacts possibly associated with ster-
     ilization and flash-evaporated spikes, and consistency). The conclusion seemed sound:
     concomitant with the persistence of hydrocarbon residuals in soils, the toxicity and
     mobility of petroleum hydrocarbons were substantially reduced by biotreatment.

         Cornelissen (1998), the  fifth entry in Table 4, gathered field-contaminated,
     poly cyclic aromatic hydrocarbon (PAH)-rich sediments from Amsterdam Harbor,
     subjected them to bioremediation (either bioreactor or land farming), and measured
     desorption kinetics of the PAHs before and after biotreatment. This  study had no
     obvious methodological weaknesses. It found that some PAH compounds per-
     sisted regardless of their desorption from sediments. However, for the biodegrad-
     able PAH pool, biotreatment was found to effectively degrade the  otherwise rapidly
     leachable (releasable) PAHs. Cornelissen et al. (1998) concluded that the biode-
     gradable PAH pool was predictable, based on the chemically measurable "rapidly
     desorbing fraction." This implied that the rapidly desorbing fraction was bioavailable
     and that the nondesorbing fraction was not.

-------
                                                                                91
    Entry number six in Table 4 (Madsen et al., 1996) began with the goal of testing
several hypotheses for the persistence of a readily biodegradable PAH compound,
naphthalene, in a coal tar-contaminated study  site. Cores removed from the site
proved that naphthalene was present in anaerobic freshwater sediments. Further-
more, laboratory incubations proved that the native microbial community could readily
convert freshly added 14C-labeled naphthalene to 14CO2 under aerobic conditions and
that these microorganisms were not nutrient limited. The cause of contaminant per-
sistence was narrowed to two hypotheses: lack of bioavailable naphthalene or lack of
a final electron acceptor (O2). To address the former hypothesis,  irradiated sterile
sediments were aged in the presence of 14C- naphthalene  and mineralization was
assayed using inocula from the sediment. However, the maximum aging period was
unrealistically short (4 weeks). In addition, two types of inocula (enriched on either
solid- or  liquid-phase naphthalene) responded differently to the aged naphthalene.
Malleability of the microbial community was indicated because sediment microorgan-
isms pre-enriched on soluble naphthalene did not easily mineralize the aged naphtha-
lene, while unenriched sediment microorganisms mineralized aged and imaged naph-
thalene equally. To test the oxygen limitation hypothesis in the field, H2O2 was added
to small volumes of enclosed sediments. However, over 3 weeks, the H2O2 treatment
did not stimulate naphthalene loss from soil. Thus, Madsen et al. 's study (1996) was
equivocal. Lack of field stimulationby the H^ amendment suggested a bioavailability
limitation, but it was possible that the H2O2 treatment was ineffective. Similarly, the
experiments using aged, irradiated sediments showed that diminished bioavailability
associated with aging was dependent on the inoculum.

    To understand field persistence of the herbicide, simazine, Scribnere/a/. (1992;
seventh entry in Table 4) used chemical and bioassay methods to compare the
partitioning behavior, biodegradability, and toxicity of freshly added versus  aged
simazine in soil. When simazine was freshly added in the field, it was found in the
soil solution, but its fractional equilibrium concentration was otherwise very low in
field pore waters. Moreover, freshly added, but not aged, simazine was both biode-
gradable and toxic to plant seedlings. This study presented laboratory data [much
like that of Steinberg et al. (1987)] contrasting a constant pool of aged simazine that
was not biodegraded in soil by the naturally occurring microflora with a declining
pool of 14C-freshly added simazine; however,  the methods leading to such data
were not clearly  described. Overall, this study's results supported the view that an
aged, field-applied pesticide was neither mobile nor bioavailable. The case was not
flawless, however, especially because simazine is not a readily utilized carbon and
energy source for soil microorganisms.

    Hatzinger and Alexander (1995; entry #8 of Table 4) published a pivotal series of
experiments that yielded clear, inverse relationships between the extractability and
biodegradability of phenanthrene and 4-nitrophenol that had been aged in sterile
(irradiated) soil  up to 315  days prior to  chemical and biodegradation analyses.
Unfortunately, using the criteria established in Table 2, the work of Hatzinger and

-------
92
     Alexander (1995) has diminished merit because no field observations were made,
     sterile soil was used, the phenanthrene was added with a carrier organic solvent,
     and pure cultures were used in the mineralization bioassay.

         The ninth entry in Table 4 (Schwartz and Scow, 1990) examined the behavior of
     a three-ring PAH compound, phenanthrene, which was  added to a soil whose
     native microflora had unusually low phenanthrene biodegradation activity. After as
     long as 600 hours (25 days), chemical extractions and a phenanthrene-degrading
     microbial inoculum were used to assess the influence of aging on phenanthrene
     availability. Data showed that, although both the amount of hexane-extractable and
     biodegradable phenanthrene diminished with aging, the bacteria were able to ac-
     cess a pool of phenanthrene unavailable to the hexane extractant. Thus,  some
     support for the diminished bioavailability of aged phenanthrene was garnered. But
     the results were somewhat ambiguous and the methods did not fully conform to the
     criteria of Table 2 (see column 6 of Entry #9 in Table 4).

         A collaborative effort by laboratories operated by M. Alexander and J. Pignatello
     (White etal., 1997; Entry #10, Table 4) sought to test hypotheses about binding sites
     for phenanthrene and how adding another PAH compound (pyrene) could influence
     phenanthrene's chemical and biological behavior. Aging was found to diminish both
     the rate and extent of phenanthrene mineralization. Moreover, the added pyrene was
     found to boost both extractability of the phenanthrene and its extent of mineralization
     by an inoculated bacterium unable to metabolize pyrene. Unfortunately, applying the
     criteria in Table 2, the work by White et al. (1999) was not field based and both
     irradiated soil and a pure culture were used in the experiments.

         In 1992, Guerin and Boyd (entry #11 of Table 4) published an often-cited study
     illustrative of how pure cultures  of bacteria can reveal contrasting physiological
     capabilities. This study failed to conform to any of the seven criteria of quality
     shown in Table 2 [no field observations, unrealistic laboratory incubation condi-
     tions, altered geosorbents, soil sterilized by autoclaving, the test compound (naph-
     thalene) had a co-solvent (acetone), and two pure bacterial cultures were used].
     Nevertheless, diversity and malleability of the  microbial world became apparent
     when one culture was found to metabolize only  aqueous-phase naphthalene; while
     the other metabolized a significant portion of the sorbed chemical.

         In another laboratory study to determine the fate of PAHs in soil, Guthrie and
     Pfaender (1998; entry #12 of Table 4) used the four-ring compound, pyrene, to assess
     how microbial activity influenced the conversion of 14C-labeled pyrene that was
     added to several operationally established humic acid fractions of soil. Viable soil
     microorganisms were found to enhance incorporation of the 14C-labeled pyrene into
     humic acid fractions. However, aging the pyrene for 270 days did not significantly
     diminish its susceptibility to mineralization by  soil microorganisms. The work by
     Guthrie and Pfaender (1998) was not conducted in the field and used several ap-
     proaches that failed to conform with the criteria in Table 2.

-------
                                                                                93
    Khan and Behki (1990) and Johnson et al. (1999) rigorously tackled the issue of
incorporating soil-applied pesticides into humic substances and their potential subse-
quent release (entries #13 and #14, respectively, in Table 4). Khan and Behki (1990)
formed bound 14C-atrazine residues in an agricultural soil by incubating the herbicide
for 1 year and then used ethanol extraction to ensure all unbound forms of atrazine were
removed. Then Khan and Behki (1990) added microbial cultures to the soil and used
chemical analyses to follow microbial metabolism and extractability of the soil-bound
residue during an 84-day incubation. Atrazine and atrazine metabolites (indicative of
microbial metabolism) were released inppm amounts. Johnson et al. (1999) essentially
repeated the work of Khan and Behki (1990) with methodological modifications that
included aseptic preparation of bound residues (in y-irradiated soil) during 3 months of
aging. Data from Johnson et al. (1999) showed that bound residues were extractable
and desorbable from soil but viable microorganisms did not enhance extractability.
Thus, the studies by Khan and Behki (1990) and Johnson et al. (1999) displayed both
consistency and inconsistency: bound residues were released in both studies, but the
role of microorganisms in these releases was not reproduced. Neither of the studies
conformed to the quality criteria in Table 2.

    Entry #15 in Table 4 (Ogram et al., 1985) is an often-cited report that used a
combination of mathematical modeling and careful manipulation of soils and pure
bacterial cultures to test the  hypothesis that sorption of 2,4-D to soil protects the
herbicide frombiodegradation. Ogram et al. used sterile soil (by autoclaving), a pure
bacterial culture, a co-solvent for the 2,4-D, and a brief (5 hours) incubation time for
the experiments.  Clearly, these experimental conditions failed to conform with the
"field relevant" criteria of Table 2. It is noteworthy, however, that on numerous occa-
sions Ogram et al.'s (1985) paper has been cited to support arguments that sorbed
substrates may be unavailable forbiodegradation. However, in the original publica-
tion, the authors were  very careful to advise readers not to misuse the findings.
Ogram et al. (1985) stated that "results...for a single strain of bacteria degrading 2,4-D
in a rigorously controlled  environment...may not be applicable to conditions one
would expect to find in situ, where mixed populations may be degrading the pesticide
for a much longer time than the 5-hour incubation period used here."

    The final two entries in Table 4 (#16,17) examined the influence of the residence
time of agricultural chemicals (DDT and related compounds) onbioavailability, as-
sessed via earthworm uptake instead of biodegradation. Tang et al.'s (1995, entry #16
in Table 4) objective was to calibrate chemical extraction techniques with pesticide
bioavailability to earthworms in soil. These researchers were able to build an experi-
mental design around soils to which DDT had been applied 30 and 49 years prior to
the study. Several other combinations of pesticides and agricultural soils also were
prepared in the laboratory using sterile (irradiated) soil aged in the presence of sol-
vent-delivered, flash-evaporated pesticides. Although Tang et al's. (1999) focus was
on correlating chemical extraction and earthworm uptake measurements, data pre-
sented sometimes failed to support the hypothesis that bioavailability diminishes

-------
94
     over time. For instance, for a Lima loam soil, aseptic aging of DDT, DDE, and ODD led
     to increased pesticide bioavailability to earthworms. In the study by Morrison et al.
     (2000; final entry of Table 4), chemical extraction and earthworm uptake procedures
     were  again applied to aged and unaged pesticides in soils. In several instances,
     inconsistent and conflicting data indicated that long residence times of insecticides
     in soils did not lead to decreases in earthworm uptake from soil. Thus, Tang et al.
     (1999) and Morrison et al.  (2000) provide examples of reports that are of secondary
     significance for this report. Although some of the data were obtained from field-
     contaminated soil, irradiated soils, flash-evaporated pesticide delivery, and inconsis-
     tent and conflicting data (see Table 2) detracted from their results.


     8.3.1.   Summary

         Of the hundreds of research articles published in the last two decades on
     bioavailability and biodegradation, details of only 17 were presented in Section 8.3
     and Table 4. These were chosen because they were influential and/or recent and
     representative of the varying approaches, objectives, and the quality of experimental
     designs and methodologies published to date. Each of the studies described in Table
     4 had its own merits—otherwise publication would not have occurred. Furthermore,
     although none of the studies was designed to specifically address the needs of this
     report, roughly the first half of the entries in Table 4 conformed reasonably well with
     this report's quality criteria (field realism, environmental relevance, absence of arti-
     facts, consistency)  set forth in Table 2.

         Steinberg et al. (1987) showed that field-aged and freshly added EDB pools in
     soil were chemically and biologically distinctive, and that pulverization enhanced
     desorption of aged EDB; thus, micropore entrapment was indicated as a cause of
     reduced EDB bioavailability. Studies by Oliverae/al. (1998) and Berg et al. (1998)
     showed that petroleum hydrocarbon residues were persistent after land  farming
     biotreatment had reduced teachability and toxicity and that additional biodegrada-
     tion was likely prevented by reduced hydrocarbon bioavailability; thus, "environ-
     mentally acceptable endpoints" had been achieved by bioremediation treatment.
     Although laboratory-based, the study by Salanitro et al. (1997) also demonstrated
     that biotreatment of freshly added petroleum hydrocarbons in  soil reduced the
     mobility and toxicity, hence the bioavailability, of residual contaminants. Cornelissen
     et al's. (1998) investigation showed that a portion of the rapidly desorbing (i.e.,
     bioavailable) PAH fraction from harbor sediments corresponded to the readily bio-
     degradable fraction. The combined field and laboratory study by Madsen et al.
     (1996) narrowed the cause of  naphthalene persistence to be limitations in either
     bioavailability or oxygen.  Scribner et al's. (1992) investigation [much like that of
     Steinberg et al. (1987)] showed that the chemical andbiological properties of freshly
     added and aged simazine  were distinctive, and an aging-related reduction in si-
     mazine bioavailability was  the cause. Thus, the above investigations strongly sup-

-------
                                                                             95
port the hypothesis that chemical assays of materials in soil can overestimate
what is actually biologically absorbed and metabolized. Several bioassays (biodeg-
radation, uptake, and toxicity) and chemical extraction assays indicated that the
reduction in contaminant bioavailability in soil was a time-dependent phenomenon
(freshly added contaminants were distinctive from aged contaminants). Such ob-
servations conform with the various operational definitions for bioavailability limi-
tations described in Sections 3.1 and 3.3 of this report.

    The entries that appeared in the latter half of Table 4 largely failed to conform with
the quality criteria of Table 2. Nevertheless, each study contributed to broad schol-
arly information about the potential influences of geosorbents, time, reactions, and
biota on biodegradation and pollutant behavior and/or toxicity in soil.


8.4   Influence of Bioavailability on Phytoremediation of
      Metal-Contaminated  Soils

    One key characteristic of plant-based, cleanup technologies is that plants can
have major physical and hydrologic impacts on contaminated sites (Ensley, 2000;
Schnoor, 2000). Not only  do growing plant roots explore and penetrate the three-
dimensional volume of contaminated soil, but the plant vascular system withdraws
water from the soil depths and delivers it to the atmosphere. Transpirational de-
mand at leaf surfaces is the driving force for the flow of water from soil solution to
the root surface and into and through the plant biomass. As discussed in Section
7.2 of this report, the relative contribution of mass flow (versus diffusion versus
root interception) to nutrient uptake varies with each plant species. Nonetheless,
because all plants transpire, they all establish moisture gradients within the soil
matrix, induce advective mass transfer of water, and convey solutes both through
soil and into their tissues (Schnoor, 2000). If solutes, including metals, are in soil
solution, they assuredly will be conveyed to the plant (see Section 7.1). It is for this
reason that substantial research efforts have been directed at boosting access—
boosting bioavailability—of metals to  plants.

   Mobilization of metals from soil into soil solution and plants is particularly
amenable to experimentation because  organic chelating agents such as ethylene-
diaminetetraacetic acid (EDTA) have been used for decades for similar purposes.
Phytochelators (Khan et al., 2000) also have been described. Many greenhouse
experiments have demonstrated the effectiveness of added chelating agents in
enhancing potential phytoremediation (Blaylock et al., 1997). For example, when
EDTA and HEDTA were added to pot-grown Helianthus annuus, uptake and trans-
location of Cd and Ni from contaminated soil was substantially increased (Chen
and Outright, 2001). Also, Ebbs and Kochian (1998) demonstrated that an EDTA
amendment increased Zn  accumulation by Indian mustard planted in metal-con-
taminated soil.

-------
96
        Although entry of metals into  soil solution is an important aspect of
     phytoremediation strategies, defining soil and plant factors that govern metal uptake
     also is a major area of current research (McGrath et al., 2001). In this regard,
     "nonbioavailable" forms of metal have begun to be identified. For example, Stanhope
     et al. (2000) used isotope dilution techniques in pot studies of Indian mustard to
     show that, despite general mobilization of Cd, Zn, Pb, Ni, and Cu from sludge-con-
     taminated soil, EDTAfailed to facilitate plant access to a nonlabile pool of soil Cd. In
     related studies, researchers have begun to examine the time-dependence of extract-
     ability (Martinez and McBride, 20001), animal toxicity (Lock and Janssen, 2001), and
     plant uptake (Pedersene/a/., 20001;  Stacey etal., 2001) of metal contaminants. Re-
     sults of such studies appear to be delivering a message much like that from microbial
     biodegradation studies of organic soil contaminants (see Sections 8.1 through 8.3 of
     this report): the relationships between metals, soils, and plants seem to vary substan-
     tially depending on the type, composition, and design of the experimental system.


     8.5   A Synthesis:  Evaluating the Relationships Between
           Bioavailability and Bioremediation Based on Sections 2
           to 8.4 of This Report

        This section presents a series of statements that blend lessons from the mecha-
     nistic studies, discussed in Sections 8.1 to 8.4 of this report, with the fundamen-
     tals of biodegradation, bioremediation, bioavailability, persistence of organic com-
     pounds, paradigms of geosorbent-contaminant interaction, and mechanisms of
     contaminant uptake by biota that were presented in Sections 2 to 7 of this report.

     •  Microorganisms and plants offer a broad array of physiological and/or physi-
        cal mechanisms for eliminating and/or binding organic and inorganic con-
        taminants in soil and sediment  matrices (Section 2).

     •  Soil and sediment exist in a state of kinetically constrained thermodynamic
        disequilibrium in which combinations of biochemical and chemical reactions,
        as well as gas-, liquid-, and solid-phase transport processes, cause the bio-
        geochemical cycling of both organic and inorganic compounds (Section 6).

        Naturally occurring, plant-derived organic compounds constantly cycle
        through the soil habitat—partially entering a pool of persistent humic sub-
        stances thought to be protected from microbial attack by a combination of
        complex molecular structure, resistance to enzymatic digestion, insolubility,
        failure to enter the microbial cell, possible enzyme inactivation, and complex-
        ation (masking) (Section 5).

     •  Although the details of how and why chemicals persist in soil are governed by
        each chemical's molecular structure, a survey of soil bioremediation field
        projects (for petroleum hydrocarbons, pentachlorophenol, PAHs, metals; Sec-

-------
                                                                         97
tion 4) indicates that these organic and inorganic contaminants also routinely
persist in soil, though simple mixing and aeration frequently accelerate or-
ganic contaminant loss.

Six specific mechanisms have been hypothesized as the "cause" of diminished
bioavailability, hence, chemical waste persistence in soils:  (1) sorption to
surfaces, (2) partitioning into NAPL phases, (3) micropore- and nanopore-
separation, (4) complexation, (5) insolubility, and (6) partitioning into organic
matter (Section 3).

Bioavailability can seldom be measured directly. Instead, it emerges from the
specific, detailed, three-way interaction between biota, chemical waste com-
pounds, and a geochemical matrix under study (Section 3).

Soil biota (microorganisms and plants) feature a diverse array of anatomical
and physiological adaptations (e.g., solid-phase uptake, diffusion-accelerated
uptake, active transport, extracellular enzymatic digestion, extracellular scav-
enging systems) that allow them to successfully contend with the adverse
nutritional conditions of soil (Section 7). Such physiological and evolutionary
adaptations should not be underestimated when interpreting data or making
predictions related to the availability of chemical wastes for bioremediation in
soil. Many metabolic capabilities of soil biota (especially microorganisms)
probably have yet to be discovered (Section 6.3).

All six hypothesized "mechanisms of diminished bioavailability" (Section 3)
involve a spatial separation of the chemical substrate from intra- or extra-
cellular metabolism. To some degree, direct physical contact between cells (or
roots) and the metabolized chemical can counteract spatial separation. How-
ever, even excretion of extracellular enzymes cannot counteract the size dif-
ferential between molecules that could diffuse into soil nano- or micropores
(Ball and Roberts, 1991; Newman and Thomasson, 1979) and microbial cells
(~lum). Thus, regardless of the stage of sophistication of mathematical and
conceptual models of geosorbent matrices (Section 6.4), there is no doubt that
some portions of chemical waste in soil can be—and are—inaccessible to
plants and microorganisms.

Inaccessibility of chemical wastes in  soil, synonymous with diminished
bioavailability, is observed as discrepancies between modest or undetectable
bioassay responses (uptake, toxicity, or biodegradation) to chemicals, com-
pared to ready analytical quantification of the chemicals in the same soil
samples (Section 8.3).

The prevalence of ambiguous, even conflicting, findings in the existing litera-
ture on biodegradation and bioavailability (Sections 8.1 to 8.4) is probably a
reflection of: (1) the simultaneous variability in the methods, test chemicals,
approaches, and experimental systems that have been devised by numerous

-------
98
         concerned investigators, and (2) the concomitant shifts in prevailing mecha-
         nisms (e.g., direct contact, extracellular enzymes, diffusion, active uptake) by
         which the biotic components in these investigations access and metabolize the
         tested chemicals.

-------
                            Section 9

  Overcoming Constraints on Site Cleanup
    The propensity for contaminated soil to remain contaminated poses a major
challenge for the proper management of contaminated sites, especially site cleanup
technologies (National Research Council, 1997). Several recent publications on site
cleanup techniques, particularly bioremediation measures, have reviewed current
knowledge on efforts  to overcome constraints on contaminant removal  (e.g.,
Alexander, 1999; Maier, 2000; Volkeringe/a/., 1997; Scow and Johnson, 1997; West
and Harwell, 1992). According to Maier (2000), surfactants (biosurfactants and
synthetic surfactants), cosolvents, and thermal treatment can aid in the physico-
chemical removal of organic contaminants from porous media.

    All such contaminant-mobilizing strategies have  advantages and disadvan-
tages, and none have yet been found to be fully satisfactory. The reasons for this
absence of completely satisfactory performance should be no surprise to the reader
of this report. Soils and sediments are highly complex  geochemical matrices and,
given the variable properties of both contaminants and geochemical matrices, the
impact of geosorbent treatment techniques is not likely to be consistent. The task
of mobilizing chemical waste contaminants from geosorbents might be pursued for
"pump and treat" purposes (simply to wash the matrix) or for bioremediation pur-
poses (increasing contaminant availability for biodegradation reactions). Regard-
less of the goal, understanding the science of mobilizing contaminants out of soils
and sediments  is analogous to understanding the science of soil bioremediation
(e.g., Figure 1;  Section 3.3). Thermodynamically governed, complex interactions
between the geosorbent matrix and the contaminants cannot be escaped. If Figure
1 was to be adapted to address mobilization of contaminants from soils, the label on
the oval in Figure 1 should be replaced by "contaminant mobilization treatment."
Then the performance of the system  in the revised Figure 1 would be assessed, not
on bioremediation, but simply soil washing efficacy. The "contaminant mobilization
treatments" examined to date have often been plagued  with implementation diffi-
culties that range from high cost to  inefficiency, secondary chemical effects, and
                                  99

-------
100
     secondary biological effects (toxicity, degradation of water quality, inhibition of
     biodegrading microorganisms).

         Because potential soil amendments feature a diversity of chemical structures and
     their corresponding properties, there are still many promising approaches for
     geosorbent decontamination efforts to be devised and tested. Noordman et al. (1998)
     have explained that aquifers contaminated with hydrophobic organic contaminants
     can theoretically be remediated using dissolved organic matter, cyclodextrins (e.g.,
     McCray andBrusseau, 1999), organic cosolvents, or surfactants (Guha etal., 1998).
     Rhamnolipid biosurfactants may be superior to the alternatives because they are
     naturally occurring compounds with low environmental impact (Noordman et al.,
     1998; Torrens et al., 1998). Surfactant treatment of soils and sediments occurs via
     emulsification, micellar solubilization, or facilitated transport mechanisms. Volkering
     et al. (1997) reviewed many efforts to link surfactant-mediated mobilization of con-
     taminants to bioremediation: results ranged from stimulation to inhibition of both
     desorption and biodegradation of polluting compounds. Volkering et al. (1997) con-
     cluded that "no general trends can be found... Therefore, more research is necessary
     to make the application of surfactants a standard tool in biological soil remediation."

         Regarding phytoremediation of metal-contaminated soil (see Section 8.4 of this
     report), there is no doubt  that when chelating agents boost the concentration of
     metals in soil solution, metal uptake by plants will be enhanced. However, subtle
     secondary effects of such  amendments on soils, plants, and real-world field sites
     have yet  to be fully elucidated.

-------
                          Section 10

          Conclusions, Implications, and
       Possible Areas of Future Research
   This section focuses all prior portions of this report upon key facts, prin-
ciples, and their implications.

   Three Scientific Conclusions:

1.  Based on a conservative, reasonably thorough and careful evaluation of scien-
   tific studies described in this report, there is no doubt that chemical wastes in
   soil can be, and often are, in a state of reduced bioavailability.

2.  Reduced bioavailability simply means that a chemical waste's diminished "ef-
   fective concentration" is proportionately balanced by a lingering reservoir of
   the chemical waste in soil and sediments. This lingering reservoir remains in the
   soil habitat regardless of which combinations of conceptual or actual seques-
   tration mechanisms (e.g., complexation into bound residues, diffusion into soil
   pores, NAPL partitioning) apply.

3.  Soil is, by definition, a thermodynamically unstable, kinetically constrained me-
   dium whose chemical composition, including solid, liquid, and gaseous compo-
   nents, is constantly changing. Thus, the "nonbioavailable" chemical wastes in
   this lingering reservoir (point #2) are always subject to release into soil solution
   where the wastes are resubjected to a variety of transport and/or transformation
   processes (e.g., immobilization, biodegradation, uptake by receptors).

   Implications of Bioavailability:

1.  Considerable effort has been expended in investigating the hypothesis that chemi-
   cal wastes have diminished bioavailability in soils and sediments. Results of these
   efforts have been ambiguous because of the immense diversity in types and prop-
   erties of chemical wastes, geosorbents, biota, experimental approaches, and the
   idiosyncrasies in mechanisms by whichbiota interact with chemical wastes.
                                 101

-------
102
     2.   From a practical, regulatory point of view, establishing the foundation of reduced
         bioavailability is crucial. If the reduced bioavailability of chemical wastes in soil
         becomes widely accepted, then proper quantitative measures of bioavailability
         reduction could be developed to accurately estimate the risks posed to human
         health and ecological processes by chemical wastes in soils and sediments.

     3.   An accurate estimation of risks to human and environmental health posed by
         chemical wastes in soils (point #2) is, itself, a crucial step toward:  (i) identifying
         pragmatic, economically feasible environmental cleanup goals;  (ii) establish-
         ing operational  definitions of "treatment" by bioremediation technology; (iii)
         realistically classifying polluted sites based on planned land-use scenarios;
         (iv) developing public acceptance of risk-based contaminant cleanup efforts;
         (v) developing public acceptance of cleanup goals that are above the "original,
         pristine  state" of the contaminated site, and (vi) legitimizing the concept of
         "environmentally acceptable endpoints."

         Possible Areas of Future Research:

     1.   Microbe- and plant-based bioremediation technologies are not able to remove
         100 percent of contaminants from polluted sites. Thus, managing residual con-
         taminants in real-world sites will require progress in the science of contaminant
         behavior, in risk assessment, and in site  management polices.

     2.   Several major questions remain about the reservoirs of "nonbioavailable" chemi-
         cal wastes in soil:

          A.  Is it possible to determine the rates of soil processes in the field that
              convert "nonbioavailable" chemical wastes to available forms?

          B.   What methods are appropriate?

          C.   Can these  appropriate methods be applied in regionally representative
              field sites containing representative classes of organic and inorganic chemi-
              cal wastes?

          D.  Can the results of item #2C be used to formulate region-specific, site
              management guidelines?

          E   When released into soil solution, what are the relative probabilities and
              rates of the chemical wastes becoming:

              - resequestered?

              - biodegraded?

              - transported to susceptible receptors?

          F.   If the released chemical wastes are readily mineralized (if organic) or immo-
              bilized (if inorganic), would "natural attenuation" of these originally

-------
                                                                                103
        nonbioavailable contaminants offer effective protection of human health
        and the environment?

3.   Several major questions remain about the true, field state of solid, liquid, amor-
    phous, and metastable organic and inorganic constituents in soils and sedi-
    ments:

     A. What are the chemical forms and locations of soil constituents?

     B.  What types of surface chemical and/or other reactions predominate?

     C.  What effects do these reactions have on sequestration and/or release of
        chemical wastes?

     D.  At what rates (minutes, days, decades, centuries) do these reactions occur?

4.   The types of basic scientific questions raised in item #2 also apply to contami-
    nants and their reactions in soils and sediments.

5.   The questions raised in item #2 apply to  soil biota—especially to the soil
    microbial community—because so much of its diversity and potential physi-
    ological activities are uncharacterized.

6.   Would modeling of the processes in items  #2-5 provide predictive, heuristic
    tools to regulators, the public, and scientists?

7.   Can the risks of release of initially "nonbioavailable" soil contaminants to soil
    solution be estimated in a meaningful way that accounts for:

     A. Critical pathways at each contaminated site that could lead to exposure to
        humans or other key receptors?

     B.  An integration of known mechanisms that contribute to both the re-re-
        lease and re-sequestration of soil contaminants?

8.   Can knowledge of the mechanisms that lead to persistence of chemical wastes
    in soils (i.e., resistance to biodegradation and/or physical/chemical immobiliza-
    tion) be  understood well enough to manage the mechanisms at field sites in
    ways that either enhance or counteract sequestration reactions, as appropriate
    for site management goals (e.g., Verstraete and Devliegher, 1997)?

9.   Can researchers be convinced to use methodologies and experimental designs
    that avoid artifact-laden experimental results and resist the temptation to pub-
    lish treatment differences whose practical, long-term implications are moot in
    the real world?

10.  Do we need to anticipate and measure potentially adverse, secondary effects of
    bioremediation technology? These effects could result in nutrient limitations and/
    or inadvertent release from soil of undesirable metals or chemical metabolites.

-------
104
     11. Research efforts to address bioavailability and bioremediation data have pro-
         duced abroad spectrum of information about possible reactions that might occur
         between geosorbents, contaminant wastes, and biota. This substantial roster of
         possibilities needs to be narrowed to identify what actually does happen in real-
         world field sites. Can the true field behavior of contaminants be systematically
         sampled and interpreted so as to enable robust predictions of contaminant be-
         havior based on climate, region, contaminant class, and soil type?

-------
                           Section 11

                       Literature Cited
1.     Adriano, D. C. (ed.). 1992. Biogeochemistry of Trace Metals. Lewis Publish-
      ing, Boca Raton, FL.

2.     Alexander, M.  1973. Nonbiodegradable and other recalcitrant molecules.
      Biotechnol. Bioeng. 15:611-647.

3.     Alexander, M.  1995. How toxic are toxic chemicals in soil? Environ. Sci.
      Technol. 29:2713-217.

4.     Alexander, M. 1997. Sequestration and bioavailability of organic compounds
      in soil, pp. 43-136. In: Linz, D. G, and D. V Nakles, (eds.) Environmentally
      Acceptable Endpoints in Soil: Risk-Based Approach to Contaminated Site
      Management Based on Availability of Chemicals in Soil. American Acad-
      emy of Environmental Engineers, Annapolis, MD.

5.     Alexander, M. 1999. Bio degradation andBioremediation. 2nd ed. Academic
      Press, New York, NY.

6.     Alleman, B.C., and A. Leeson. 1999. Bioremediation of metals and inorganic
      compounds, Fifth International In situ and On-site Symposium 5(4). Battelle
      Press, Columbus, OH.

7.     Allison, F. E. 1965. Soil Organic Matter and Its Role in Crop Production.
      Elsevier, Amsterdam.

8.     Alloway, B. J. (ed.).  1995. Heavy Metals in Soils, 2nd ed. Blackie Academic
      and Professional, London.

9.     Amann, R. I., W. Ludwig, andK.-H. Schleifer.  1995. Phylogenetic identifica-
      tion and in situ detection of individual microbial cells without cultivation.
      Micmbiol. Rev.  59:13-169.
                                  105

-------
106
     10.    Amann, R., and M. Ktihl. 1998. In situ methods for assessment of microor-
           ganisms and their activities. Curr. Opin. Microbiol. 1:352-358.

     11.    Angehrn, D., M. Schluep, R. Galli, and J. Zeyer. 1999. Movement and fate of
           residual mineral oil contaminants in bioremediated soil. Environ. Toxic. Chem.
           18:2224-2231.

     12.    Angehrn, D., R. Galli, and J. Zeyer. 1998. Physicochemical characterization of
           residual mineral oil contaminants in bioremediated soil. Environ. Toxicol.
           Chem. 17:2168-2175.

     13.    Atlas, R.M., and R. Bartha 1997. Microbial Ecology: Fundamentals and
           Applications, 4th ed. Benjamin Cummings Publishing Co., Menlo Park, CA.

     14.    Babu, G. R. V, J. H Wolfram, and K. D. Chapatwala. 1992. Conversion of
           sodium cyanide  to carbon dioxide and ammonia by immobilized cells of
           Pseudomonasputida. J. Ind. Microbiol. 9:235-238.

     15.    Ball, W. P. 1989. Equilibrium sorption and diffusion rate studies with haloge-
           nated organic  chemicals and sandy aquifer material. Ph.D. Dissertation.
           Stanford University.

     16.    Ball, W. P., and P. V. Roberts.  1991. Long-term sorption of halogenated or-
           ganic chemicals by aquifer material. 2. Intraparticle diffusion. Environ. Sci.
           Technol. 25:1237-1249.

     17.    Barber, S. A. 1995. Soil Nutrient Bioavailability: A Mechanistic Approach.
           John Wiley & Sons, Inc., New York, NY.

     18.    Baveye, P., andR. Bladon. 1999. Bioavailability of organic xenobiotics in the
           environment: a critical perspective, pp. 227-248. In: Baveye, P., J.-C. Block,
           and V  V. Goncharuk (eds.). Bioavailability of Organic Xenobiotics in the
           Environment. Kluwer Academic Publishing, Boston, MA.

     19.    Bayer, E. A., and R. Lamed. 1992.  The cellulose paradox:  pollutant par
           excellence and/or a reclaimable natural resource? Biodegrad. 3:171-188.

     20.    Berg, M. S., R. C. Loehr, and M. T. Webster.  1998. Release of petroleum
           hy drocarbons from bioremediated soils. J. Soil Contain. 7:675-695.

     21.    Bhadra, R., R. J.  Spanggord, D. G. Wayment, J. B. Hughes, and J. V Shanks.
           1999. Characterization of oxidation products of TNT metabolism in aquatic
           phytoremediation systems ofMyriophyllum aquaticum. Environ. Sci. Technol.
           33:3354-3361.

     22.    Bizify, W. P., C. L. Rugh, A. O. Summers, andR B. Meagher. 1999. Phytoremediation
           of methylmercury pollution: MerB expression mArabidopsis thaliana confers
           resistance to organomercurials. Proc. Natl.Acad. Sci. 96:6808-6813.

-------
                                                                              107
23.    Blaylock, M. J. 1997. Phytoremediation of lead-contaminated soil at a
      brownfield site in New Jersey—a cost-effective alternative. In: IBC's 2nd
      Annual International  Conference on Phytoremediation, Seattle, WA. In-
      ternational Business Communications: Southborough, MA.

24.    Blaylock, M. I, D. E. Salt, S. Dushenkov, O. Zakharova, C. Gussman, Y Kapulnik,
      B. D. Ensley, and I. Raskin. 1997. Enhanced accumulation of Pb in Indian
      mustard by soil -applied chelating agents. Environ. Sci. Technol. 31:860-865.

25.    Boetius, A., K. Ravenschlag, C. J. Schubert, D. Rickert, F. Widdel, A. Gieseke,
      R. Amann, B. B. Jorgensen, U.  Witte, and O. Pfannkuche. 2000. A marine
      microbial consortium apparently mediating anaerobic oxidation of methane.
      Nature 401:623-626.

26.    Bosnia, T. M. P., P. J. M. Middeldorp, G Schraa, and A. J. B. Zehnder. 1997.
      Mass transfer limitation of biotransformation: Quantifying bioavailability.
      Environ. Sci. Technol. 31:248-252.

27.    Brady, N. C., and R. R. Weil. 1999.  The Nature and Properties of Soils,
      Twelfth Edition. Prentice Hall. Upper Saddle River, NJ.

28.    Brierley,  C. L.  1990.  Bioremediation of metal-contaminated surface and
      groundwaters. Geomicrobiol. 8:201-224.

29.    Brock, T. D. 1978. The poisoned control inbiogeochemical investigations, pp.
      717-725. In: W. E Rrumbein (ed.)EnvironmentalBiogeochemistry and'Micro-
      biology, Volume 3. Methods, Metals, and Assessment. Ann Arbor Science, Inc.

30.    Brusseau, M. L., R. E. Jessup, andP. S. C. Rao. 1991. Nonequilibriumsorption
      of organic chemicals—elucidation of rate-limiting processes. Environ. Sci.
      Technol. 25:134-142.

31.    Burford,M. D., S.B.Hawthorne, and D.J. Miller.  1993. Extraction rates of
      spiked versus native PAHs from heterogeneous environmental samples us-
      ing supercritical fluid extraction and sonication in methylene chloride. Anal.
      Chem. 65:1497-1505.

32.    Burken, J. G, and J. L.  Schnoor. 1998. Predictive relationships foruptake of
      organic contaminants by hybrid poplar trees. Environ. Sci. Technol. 32:3379-
      33385.

33.    Cawse, PA. 1975. Microbiology and biochemistry of irradiated soils, pp.
      213-267. In: Paul, E. A. and A. D. McLaren (eds.). Soil Biochemistry, Volume
      3. Marcel Dekker, New York, NY.

34.    Chaney, R. L., M. Malik, Y-M. Li, S. L. Brown, E. R. Brewer, J. S. Angle, and A.
      J. M. Baker. 1997. Phytoremediation of soil metals. Curr. Opin. Biotechnol.
      8:279-284.

-------
108
     35.    Chapatwala, K. D., G. R. V. Babu, E. R. Armstead, E. M. White, and J. H.
           Wolfram. 1995. Akinetic study on the bioremediation of sodium cyanide and
           acetonitrile by free and immobilized cells of Pseudomonas putida. Appl.
           Biochem. Biotechnol. 51-52:717-726.

     36.    Chen, H., andT. Cutright. 2001. EDTAand HEDTAeffects on Cd, Cr, andNi
           uptake by Helianthus annuus. Chemosphere 45:21-28.

     37.    Chen, S., and D. B. Wilson. 1997. Genetic engineering of bacteria and their
           potential forHg-2+bioremediation. Biodegrad. 8:97-103.

     38.    Compeau, G. C., W D. Mahaffey, andL. Patras. 1991. Full-scale bioremediation
           of contaminated soil and water, pp. 91-109. In: G. S. Sayler, R. Fox ,and J. W.
           Blackburn (eds.). Environmental Biotechnology for Waste Treatment. Ple-
           num Press, New York, NY.

     39.    Connaughton,  D. K, J. R. Stedinger, L. W. Lion, and M. L. Shuler. 1993.
           Description time-varying desorption kinetics: Release of naphthalene from
           contaminated soils. Environ. Sci.  Technol. 27:2397-2403.

     40.    Corbisier, P., D. van der Lelie, B. Borremans, A. Provoost, V de Lorenzo, N. L.
           Brown, J. R. Lloyd, J. L. Hobman, E. Csoregi, G. Johansson, andB. Mattiasson.
           1999. Whole cell-and protein-based biosensors forthe detection of bioavailable
           heavy metals in environmental samples. Anal. Chem.Acta. 387:235-244.

     41.    Cornelissen, G, H. Rigterink,  M. M. A. Ferdinandy, andP. C. M. VanNoort.
           1998. Rapidly  desorbing fractions of PAHs in contaminated sediments  as a
           predictor of the extent of bioremediation. Environ. Sci. Technol. 32:966-970.

     42.    Crawford, R. L., andD. L. Crawford (eds.). 1996. Bioremediation: Principles
           and Applications. Cambridge University Press, New York, NY.

     43.    Criddle,C. S.,PL.McCarty,M. C.EUiott, and IF. Barker. 1986. Reduction of
           hexachloroethane to tetrachloroethylene in groundwater. J. Cont. Hydrol.
           1:133-142.

     44.    Cunningham, S. D., J. R. Shann,  D. E. Crowley, and T. A. Anderson. 1997.
           Phytoremediationof contaminated water and soil, pp. 2-17. In: Kruger, E. L.,
           T. A. Anderson, and J. R. Coates (eds.). Phytoremediation of Soil and Water
           Contaminants. American Chemical Society, Washington, DC, 318 pp.

     45.    Diels, L. 1997.  Heavy metal bioremediation of soil, pp. 283-295. In: Sheehan,
           J. (ed.). Methods in Biotechnology,  2. Bioremediation protocols. Humana
           Press, Inc., Totowa, NJ.

     46.    Dixon, J. B., S. B. Weed, J. A. Kittrick, M. H. Mlford, and J. L. White (eds.). 1977.
           Minerals in Soil Environments. Soil Science Society of America. Madison, WI.

-------
                                                                              109
47.    Ebbs, S. D., and L. V. Kochian. 1998. Phytoextraction of zinc by oat (Avena
      saliva), barley (Hordeum vulgare), and Indian mustard (Brassicajuncea).
      Environ. Sci. Technol. 32:802-806.

48.    Ehrlich, H. L. 1996. Geomicrobiology, Third ed. Marcel Dekker, New York,
      NY

49.    Ensley, B. D. 2000. Rationale foruse of phytoremediation, pp. 3-11. In: Raskin,
      L, andB. D. Ensley (eds.). Phytoremedititon of Toxic Metals Using Plants to
      Clean Up the Environment. John Wiley & Sons, New York, NY, 304 pp.

50.    EPA. 2001. Treatment Technology for Site Cleanup: Annual Status Report.
      TenthEdition. EPA-542-R-01-004. US EPA. Washington, DC.

51.    Flaig, W 1975. Biochemistry of soil organic matter. FA O Soils Bull. 27:31 -3 5.

52.    Flathman, P. E., D. E. Jerger, and J. H. Exner.  1994. Bioremediation: Field
      Experience. Lewis Publishers, Boca Raton, FL.

53.    Fogel, S. 1993. Full-scale bioremediation of No. 6 fuel soil-contaminated soil:
      6 months of active and 3 years of passive treatment, pp. 161-175. In: Flathman,
      P. E., D. E. Jerger, and J. H. Exner (eds.). Bioremediation Field Experience.
      Lewis Publishers, Boca Raton, FL.

54.    Foster, R. C. 1993. The ultramicromorphology of soil biota in situ in natural
      soils:  a review, pp. 381-393. In: A. J. Ringrose-Voase, andG. S.Humphreys
      (eds.). SoilMicromorphology: Studies in Management and Genesis.  Elsevier
      Science, New York, NY.

55.    Francis, A. J. 1999. Bioremediation of radionuclide and toxic metal contami-
      nated soils and wastes, pp. 239-271. In: Adriano, D. C. (ed.). Bioremediation
      of Contaminated Soils. Soil Sci. Soc. Amer, Madison, WI.

56.    Frankenberger, W. T, Jr. and M. E. Losi. 1995. Applications of bioremediation
      in the cleanup of heavy metals and metalloids, pp. 173-210. In: Skipper, H. D.
      and R. F.  Turco  (eds.). Bioremediation Science and Applications. SSSA,
      Special Publication Number 43, Madison, WI.

57.    Galvez-Cloutier, R., and J.-S. Dube. 1998. An evaluation of freshwater sedi-
      ments contamination: the lachine canal sediments case, Montreal,  Canada.
      Part II: Heavy metal paniculate speciation study. Water, Air, and Soil Pollut.
      102:281-302.

58.    Ghiorse, W. C. 1994. Iron and manganese oxidation and reduction, pp.  1079-
      1096. In: Weaver, R. W, S. Angle, and P. Bottomley (eds.). Methods of Soil
      Analysis, Part 2, Microbiological and Biochemical Properties. ASA-SSSA,
      Madison, WI.

-------
110
     59.    Ghiorse, W. C., D. N. Miller, R. L. Sandoli, andP. L. Siering. 1996. Applications
           of laser scanning microscopy for analysis of aquatic microhabitats. Micro.
           Res. Tech. 33:73-86.

     60.    Ghosh, U., J. W. Talley, and R. G. Luthy. 2001. Particle-scale investigation of
           PAH desorption kinetics and thermodynamics  from sediment. Environ. Sci.
           Technol. 35:3468-3475.

     61.    Ghosh, U., R. G Luthy, J. S. Gillette, andR. N. Zare. 2000. Microscale location,
           characterization, and association of polycyclic aromatic hydrocarbons on
           harbor sediment particles. Environ. Sci. Technol. 34:1729-1736.

     62.    Gillette, J. S., R. G. Luthy, S. J. Clemett, and R. N. Zare. 1999. Direct observa-
           tion of polycyclic aromatic hydrocarbons on geosorbents at the subparticle
           scale. Environ. Sci. Technol. 33:1185-1192.

     63.    Goldstein, J. I., D. E . Newbury, P. Echlin, D. C. Joy, A. D. Romig, Jr., C. E.
           Lyman, D. Fiori, andE. Lifshin. 1992. Scanning electron microscopy andX-
           ray microanalysis:  a text for biologists. Materials Scientists, and Geolo-
           gists. Seconded. Plenum Press, New York, NY.

     64.    Graber, E. R., and M. D. Borisover. 1998. Evaluation of the glassy/rubbery
           model for soil organic matter. Environ. Sci. Technol. 32:3286-3292.

     65.    Guerin, W. K, and S. A. Boyd. 1992. Differential bioavailability of soil-sorbed
           naphthalene to two bacterial species. Appl. Environ. Microbiol. 58:1142-
           1152.

     66.    Guha, S., P. R. Jaffe, and C. A. Peters. 1998. Bioavailability  of mixtures of
           PAHs partitioned into the micellar phase of a nonionic surfactant. Environ.
           Sci. Technol. 32:2317-2324.

     67.    Guthrie, E. A., and F. K. Pfaender. 1998. Reduced pyrene bioavailability in
           microbially active soils. Environ. Sci. Technol. 32:501-508.

     68.    Guthrie, E. A., J. M. Bortiatynski, J. D. H. VanHeemst, J. E. Richman, K. S.
           Hardy, E. M. Kovach, andP G. Hatcher. 1999. Determination of [13C]pyrene
           sequestration in sediment microcosms using flash pyrolysis—GCMS and
           "CNMR.E'wv/row. Sci. Technol. 33:119-125.

     69.    Hartwig,R. C.,andR.H.Loeppert. 1993. Evaluation of soil iron, pp. 465-482.
           In: L. L. Barton and B. C. Hemming (eds.). Iron Chelation in Plants and Soil
           Microorganisms. Academic Press, Inc., New York, NY.

     70.    Hatzinger, P. B., andM. Alexander. 1995. Effect of aging of chemicals in soil
           on their biodegradability and extractability. Environ. Sci. Technol. 29:537-
           545.

-------
                                                                              Ill
71.    Havlin, J. L., J. D. Beaton, S. L. Tisdall, and W. L. Nelson. 1999. Soil Fertility
      and Fertilizers: An Introduction to Nutrient Management, 6th ed. Prentice
      Hall, Upper Saddle River, NJ.

72.    Heitzer, A., K. Malachowsky, J. E. Thonnard, P. R. Bienkowski, D. C. White,
      andG. S. Sayler. 1994. Optical biosensor for environmental on-line monitor-
      ing of naphthalene and salicy late bioavailability with an immobilized biolu-
      minescent catabolic reporter bacterium. Appl. Environ. Microbiol. 60:1487-
      1494.

73.    Hinchee, R. E., andR. F. Olfenbuttel. 1994. In situ Bioremediation: Applications
      and Investigations for Hydrocarbon and Contaminated Site Remediation.
      Battelle Press, Columbus, OH.

74.    Hinchee, R. E., J. L. Means, and D. R. Burris (eds.). 1995. Bioremediation of
      Inorganics. Battelle Press, Columbus, OH.

75.    Hinrichs, H.-U., J. MHayes, S. P. Sylva, P. G. Brewer, andE. F. DeLong. 1999.
      Methane-consuming archaebacteria in marine sediments. Nature 398:802-
      805.

76.    Hofte, M. 1993. Classes of microbial siderophores, pp. 3-26. In: Barton, L. L.
      and B. C. Hemming (eds.). Iron Chelation in Plants and Soil Microorgan-
      isms. Academic Press, Inc., New York, NY.

77.    Huang, J. W., J. Chen, and S. D. Cunningham. 1997. Phytoextraction of lead
      from contaminated soils, pp. 283-298. In: Kruger, E. L., T. A. Anderson, and J.
      R. Coats (eds.). Phytoremediation of Soil and Water Contaminants. Ameri-
      can Chemical Society, Washington, DC.

78.    Huesemann, M. H.  1995. Predictive model for estimating the extent of petro-
      leum hydrocarbon biodegradation in contaminated soils. Environ. Sci.
      Technol. 29:7-18.

79.    Hughes, J. B., D. M. Beckles, S. D. Chandra, and C. H. Ward. 1997. Utilization
      of bioremediation processes for the treatment of PAH-contaminated sedi-
      ments. J. Indus. Microbiol. Biotechnol. 18:152-160.

80.    Jerger, D. E., P. M.  Woodhull, P. E. Flathman, and J. H. Exner. 1993. Solid-
      phase bioremediation of petroleum hydrocarbon-contaminated soil: labora-
      tory treatability study through site closure, pp. 177-193. In: Flathman, P. E.,
      D. E. Jerger, and J. H. Exner (eds.). Bioremediation Field Experience. Lewis
      Publishers, Boca Raton, FL.

81.    Johnson, S.  E., J. S. Herman, A. L. Mills, and G. M. Hornberger.  1999.
      Bioavailability and desorption characteristics of aged, nonextractable atra-
      zine in soil. Environ. Toxicol. Chem. 18:1747-1754.

-------
112
     82.    Kalin, M., J. Cairns, and R. M. McCready. 1991. Ecological engineering meth-
           ods for acid mine drainage treatment of coal wastes. Resources Conserv.
           Recycl. 5:265-276.

     83.    Khan, A. G, C. Kuek, T. M. Chaudhry, C. S. Khoo, and W. J. Hayes. 2000. Role
           of plants, mycorrhizae and phytochelators in heavy metal contaminated land
           remediation. Chemosphere 41:197-207.

     84.    Khan, S. U., andR. M. Behki. 1990. Effects of Pseudomonas species on the
           release of bound 14C residues from soil treated with [14C]atrazine. J. Agric.
           FoodChem. 38:2090-2093.

     85.    King, J. M. H., R M. Digrazia, B. Applegate, R. Burlage, J. Sanseverino, R Dunbar,
           F. Larimer, and G S. Sayler. 1991. Rapid, sensitive bioluminescentreportertech-
           nology for naphthalene exposure andbiodegradation. Science 249:778-781.

     86.    Ladd, J. N., R. C. Foster, R Nannipieri, J. M. Oades. 1996. Soil structure and
           biological activity, pp. 23-78. In: G. Stotzky, and J-M. Bollag (eds.). Soil Bio-
           chemistry, Volume 9. Marcel Dekker, Inc., New York, NY.

     87.    Leavitt, M. E., D. A. Graves, and C.A.Lang. 1991. Evaluation of bioremediation
           in a coal-coking waste lagoon, pp. 71-84. In: Sayler, G. S.,R. Fox, andJ. W.
           Blackburn (eds.). Environmental Biotechnology for Waste Treatment. Ple-
           num Press, New York, NY.

     88.    Lee, B.-G, S. B. Griscom, J.-S. Lee, H. J. Choi, C.-H. Koh, S. N. Luoma, and
           N. S. Fisher. 2000. Influences of dietary uptake and reactive sulfides on metal
           bioavailability from aquatic sediments. Science 287:282-284.

     89.    Lee, N., P. H. Nielsen, K. H. Andreasen, S. Juretschko, J. L. Nielsen, K.-H.
           Schleifer, and M. Wagner. 1999. Combination of fluorescent in situ hybrid-
           ization and microautoradiophy—a new tool for structure-function analyses
           inmicrobial ecology. Appl. Environ. Microbiol. 65:1289-1297.

     90.    Lengeler, J. W., G. Drews, and H. G. Schlegel. 1999. Biology of the Prokary-
           otes. Blackwell Scientific, Maiden, MA.

     91.    Lenhard, R.  J., R. S. Skeen, and T. M. Brouns.  1995. Contaminants at U.S.
           DOE sites and their susceptibility to bioremediation, pp. 157-172. In: Skipper,
           H. D. and R. F. Turco (eds.). Bioremediation  Science and Applications.
           SSSA, Special Publication Number 43, Madison, WI.

     92.    Lindsay, W. L. 1979. Chemical Equilibria in Soils. John Wiley & Sons, New
           York, NY

     93.    Linz, D. G, andD. V Nakles. 1997. Next steps, pp. 467-486. In: Linz, D. G and
           D. V. Nakles (eds.). Environmentally Acceptable Endpoints in Soil. Ameri-
           can Academy of Environmental Engineers.

-------
                                                                               113
94.    Lock, K., and. C. R. Janssen. 2001. Ecotoxicity of zinc in spiked artificial soils
      versus contaminated field soils. Environ. Sci. Technol. 35:4295-4300.

95.    Loehr, R. C., and M. T. Webster. 1997.  Effects of treatment on contaminant
      availability, mobility, andtoxicity, pp. 137-386. In: D. G. Linz andD. V Nakles
      (eds.). Environmentally Acceptable Endpoints in Soil. American Academy
      of Environmental Engineers.

96.    Loehr, R. C. 1999. Overview, pp. 10-17. In: Anderson, W. C., R. C. Loehr, andB.
      P. Smith (eds./ Environmental Availability of Chlorinated Organics, Explo-
      sives, and Metals in Soils. American Academy of Environmental Engineers
      Publishing, Annapolis, MD.

97.    Lollar, B. S., G F. Slater, B. Sleep, M. Witt, G. M. Klecka, M. Harkness,and J.
      Spivack. 2001. Stable carbon isotope evidence for intrinsic bioremediation of
      tetrachloroethene and trichloroethene at Area 6, Dover Air Force Base.
      Environ. Sci. Technol. 35:261-269.

98.    Lovley, D.  R. 1993. Dissimilatory metal reduction. Annu. Rev. Microbiol.
      47:263-290.

99.    Lovley, D. R. 1995a. Bioremediation of organic and metal contaminants with
      dissimilatory metal reduction. J. Ind. Microbiol. 14:85-93.

100.   Lovley,D. R. 1995b. Microbial reduction of iron, manganese, andothermet-
      ?As.Adv.Agron. 54:176-231.

101.   Lovley, D. R. 2000. Fe(III) andMn(IV) reduction, pp. 3-30. In: Lovley, D. R.
      (ed.) Environmental Microbe-Metal Interactions. ASM Press, Washing-
      ton, DC, 3 95 pp.

102.   Lovley, D. R., and J. D. Coates. 1997. Bioremediation of metal contamination.
      Curr. Opin. Biotechnol. 8:285-289.

103.   Lovley, D.  R., J. D. Coates, E. L. Blunt-Harris,  E. J. P.  Phillips, and J. C.
      Woodward. 1996. Humic substances as electron acceptors for microbial res-
      piration. Nature 382:445-448.

104.   Luthy, R. G, G R. Aiken, M. L. Brusseau, S. D. Cunningham, P. M. Gschwend,
      J. J. Pignatello, M. Reinhard, S. J. Traina, W. J. Weber, Jr., and J. C. Westall.
      1997. Sequestration of hydrophobic organic contaminants by geosorbents.
      Environ. Sci. Technol. 31:3341-3347.

105.   Macaskie, L. E., and G Basnakova. 1998. Microbially-enhancedchemosorption
      of heavy metals: a method for the bioremediation of solutions containing long-
      lived isotopes of neptunium and plutonium. £wv/row. Sci.  Technol. 32:184-187.

-------
114
     106.   Mackay, D. M, W. P. Ball, and M. G. Durant. 1986. Variability of aquifer
           sorption properties in a field experiment on groundwater transport of or-
           ganic solutes: methods and preliminary results. J. Cont. Hydrol. 1:119-132.

     107.   Madigan, M. T., J. M. Martinko, and J. Parker. 2000. BrockBiology of Micro-
           organisms, 9th ed. Prentice Hall, Upper Saddle River, NJ.

     108.   Madsen, E. L. 1991. Determining in situ biodegradation: facts and chal-
           lenges. Environ. Sci. Technol. 25:1662-1673.

     109.   Madsen, E. L. 1996. Acritical analysis of methods for determining the com-
           position and biogeochemical activities of soil microbial communities in situ,
           pp. 287-370. In: G. Stotzky, and J-M (eds.). Bollag. Soil Biochemistry, Volume
           9. Marcel Dekker, Inc., New York, NY.

     110.   Madsen, E. L. 2002. Methods for determining biodegradability, pp. 903-913.
           In: Hurst,  C. J., et al. (eds.). Manual of Environmental Microbiology,  2nd
           Ed. American Society for Microbiology, Washington, DC.

     111.   Madsen, E. L. 1998a. Epistemology of environmental microbiology. Environ.
           Sci. Technol. 32:429-539.

     112.   Madsen, E. L. 1998b. Theoretical and applied aspects of bioremediation:  The
           influence of microbiological processes on organic compounds infield sites,
           pp. 354-407. In: Burlage, R., R. Atlas, D. Stahl, G Geesey, and G Sayler (eds.).
           Techniques in Microbial Ecology. Oxford University Press, New York, NY.

     113.   Madsen, E. L., C. L. Mann, and S. Bilotta. 1996. Oxygen limitations and aging
           as explanation for the persistence of naphthalene in coal-tar contaminated
           surface sediments. Environ. Toxicol. Chem. 15:1876-1882.

     114.   Maier, R. 2000. Bioavailability and its importance to bioremediation. In: Valdes,
           J. J. (ed.). International Society for Environmental Biotechnology: Environ-
           mental Monitoring and Biodiagnostics. Kluwer Academic Publishers, pp. 59-
           78.

     115.   Malmqvist,A.,T Welander, andL. Gunnarsson. 1991. Anaerobic growth of
           microorganisms with chlorate as an electron  acceptor. Appl.  Environ.
           Microbiol. 57:2229-2232.

     116.   Marschner, H.1995. Mineral Nutrition of Higher Plants, Second ed. Aca-
           demic Press, New York, NY.

     117.   Marshall,K. C. 1976. Interfaces in Microbial Ecology. Harvard University
           Press, Cambridge, MA.

     118.   Martinez, C. E., andM. B. McBride. 2001. Cd, Cu, Pb, andZncoprecipitates
           in Fe oxide formed at different pH; aging effects  on metal  solubility  and
           extractability by citrate. Environ. Toxicol. Chem. 20:122-126.

-------
                                                                              115
119.   McBride, M. B. 1994. Environmental Chemistry of Soils. Oxford University
      Press. New York, NY.

120.   McCray, J. E., andM. L. Brusseau. 1999. Cyclodextrin-enhanced/'n situ flush-
      ing of multiple-component immiscible organic liquid contamination at the field
      scale: analy sis of dissolution behavior. Environ. Sci. Technol. 32:89-95.

121.   McGrath, S. P., B. Knight, K. Killham, S. Preston, and G. I. Paton. 1999. As-
      sessment of the toxicity of metals in soils amended with sewage sludge
      using a chemical speciation technique and a lux-based biosensor. Environ.
      Tax. Chem. 18:659-663.

122.   McGrath, S.  P., F. J. Zhao, and E. Lombi. 2001. Plant and rhizosphere pro-
      cesses involved in photoremediation of metal-contaminated soils. Plant and
      Soil232:201'-214.

123.   McHale, A. P., and S. McHale. 1994. Microbialbiosorption of metals: poten-
      tial in the treatment of metal pollution. Biotechnol.Adv.  12:647-652.

124.   Meagher, R. B. 2000. Phytoremediation of toxic elemental and organic pollut-
      ants. Curr. OpinPlantBiol. 3:153-162.

125.   Mihelcic,J.R.,D.R.Lueking,R. J.Mitzell,andJ.M.Stapleton. 1993.Bioavailability
      of sorbed- and separate-phase chemicals. Biodegradation 4:141-153.

126.   Morris, D. E., P. G. Allen, J. M. Berg, C. J. Chisholm-Brause, S. D. Conradson,
      R. J. Donohoe, N. J. Hess, J. A. Musgrave, and C. D. Tait. 1996. Speciation of
      uranium in Fernald  soils by molecular spectroscopic methods: Characteriza-
      tion of untreated soils. Environ. Sci. Technol. 30:2322-2331.

127.   Morrison, D. E., B. K. Robertson, andM. Alexander. 2000. Bioavailability to
      earthworms of aged DDT, DDE, DDD, and dieldrin in soil. Environ. Sci.
      Technol. 34:709-713.

128.   Myneni, S. C. B, J. T. Brown, G. A. Martinez, and W. Meyer-Use. 1999. Imag-
      ing of humic substance macromolecular structures in water and soils. Sci-
      ence 286:1335-1337.

129.   National Research Council (NRC). 1993. In situ Bioremediation: WhenDoes
      It Work? National Academy Press, Washington, DC.

130.   National Research Council (NRC). 1997. Innovations in Ground Water and
      Soil Cleanup. National Academy Press, Washington, DC.

131.   National Research Council (NRC). 2000. Natural Attenuation for Ground-
      water Remediation. National Academy Press, Washington, DC.

132.   Newman, A. D. C., and A. J. Thomasson. 1979. Rothamsted studies of soil struc-
      ture III: pore size distribution and shrinkage processes. J. SoilSci. 30:415-439.

-------
116
     133.   Newman, D. K., andR. Kolter. 2000. A role for excreted quinines in extracellu-
           lar electron transfer. Nature 405:94-97

     134.   Noordman, W. H., W. Ji, M. L. Brusseau, andD. B. Janssen. 1998. Effects of
           rhamnolipid biosurfactants on removal of phenanthrene from soil. Environ.
           Sci. Technol. 32:1806-1812.

     135.   Norris, R. D., R. E. Hinchee, R. Brown, P. L. McCarty, L. Semprini, J. T. Wilson,
           D. G. Kampbell, M. Reinhard, D. J. Bouwer, R. C. Borden, T. M. Vogel, J. M.
           Thomas, and C. H. Ward. 1994. Handbook of Bioremediation. Lewis Pub-
           lishers, Boca Raton, FL.

     136.   Ogram, A. V, R. E. Jessup, L. T. Ou, andP. S. C. Rao. 1985. Effects of sorption
           onbiological degradation rates of (2,4-dichlorophenoxy) acetic acid in soils.
           Appl. Environ. Microbiol. 49:582-587.

     137.   Olivera, F. L., R. C. Loehr, B. C. Coplin, H. Eby, and M. T. Webster. 1998.
           Prepared bed land treatment of soils containing diesel and crude oil hydro-
           carbons. J. SoilCont. 7:657-674.

     138.   Orphan, V J., C. H. House, K-U. Hinrichs, K. D. McKeegan, andE. F. DeLong.
           2001. Methane-consuming archaea revealed by directly coupled isotopic
           and phylogenetic analysis. Science 293:484-487.

     139.   Ouverney, C. C., and J. A. Fuhrman. 1997. Increase in fluorescence intensity
           of 16S rRNA in situ hybridization in natural samples treated with chloram-
           phenicol. Appl. Environ. Microbiol. 63:2735-2740.

     140.   Ouverney, C. C., and J. A. Fuhrman. 1999. Combined microautoradiography—
           16S rRNA probe technique for determination of radioisotope uptake by spe-
           cific microbial cell types in situ. Appl. Environ. Microbiol. 65:1746-1752.

     141.   Page, W. J.  1993. Growth conditions for the demonstration of siderophores
           and iron-repressible outer membrane proteins in soil bacteria, with an em-
           phasis on free-living Diazotrophs, pp. 75-110. In: Barton, L. L. and B. C.
           Hemming (eds.). Iron Chelation in Plants and Soil Microorganisms. Aca-
           demic Press, Inc., New York, NY.

     142.   Pedersen,  M. B., C. Kjaer,  and  N.  Elmegaard.  2000. Toxicity and
           bioaccumulation of copper to black bindweed (Fallopia convolvulus) in
           relation to bioavailability and the age of soil contamination. Arch. Environ.
           Contam. Toxicol. 39:431-439.

     143.   Pignatello, J. J., and B. Xing. 1996. Mechanisms of slow sorption of organic
           chemicals to natural particles. Environ. Sci. Technol. 30:1-11.

-------
                                                                              117
144.   Pignatello, J. I, andB. Xing. 1999. Pignatello andXing's comment on "Evalu-
      ation of the glassy /rubbery model for soil organic matter." Environ. Sci.
      Technol. 33:2837-2838.

145.   Portier, R. J., and J. A. Christiansen. 1993. Closure of an RCRA surface im-
      poundment by employing a modified biological treatment approach, pp. 225-
      233. In: Flathman, P. E., D. E. Jerger, and J. H. Exner (eds.). Bioremediation
      Field Experience. Lewis Publishers, Boca Raton, FL.

146.   Reid, B. J., K. C. Jones, andK. T. Semple. 2000. Bioavailability of persistent
      organic pollutants in soils and sediments—a perspective on mechanisms,
      consequences and assessment. Environ. Poll. 108:103-112.

147.   Rikkin, G. B., A. G. M. Kroon, and C. G VanGinkel. 1996. Transformation of
      (per)chlorate into chloride by a newly isolated bacterium: reduction and
      dismutation. Appl. Microbiol. Biotechnol. 45:420-426.

148.   Ringrose-Voase, A. J., and G S. Humphreys (eds.). 1994.  SoilMicromorphol-
      ogy: Studies in Management and Genesis. Elsevier Science, New York, NY.

149.   Rittmann, B. E., E. Seagren, B. A. Wrenn, A. J. Valocchi, C. Ray, and L. Faskin.
      1994. In Situ Bioremediation, Second ed. Noyes Publications. Park Ridge,
      NJ.

150.   Roemheld, V,andH. Marschner. 1986.Evidenceforaspecificuptakesystem
      for iron phytosiderophores in roots of grasses. Plant Physiol. 80:175-180.

151.   Salanitro, J. P, P. B. Dorn, M. H. Huesemann, K. O. Moore, I. A. Rhodes, L. M.
      Rice-Jackson, T. E. Vipond, M. M. Western, and H. L. Wisniewski.  1997.
      Crude oil hydrocarbon bioremediation and soil ecotoxicity assessment.
      Environ. Sci. Technol. 31:1769-1776.

152.   Sandoli, R. S., W. C. Ghiorse, and E. L. Madsen. 1996. Regulation of microbial
      phenanthrene mineralization in sediments by sorbent-sorbate contact time,
      inoculum, and gamma irradiation-induced sterilization artifacts. Environ.
      Toxicol. Chem. 15:1901-1907.

153.   Saouter, E., M. Gillman, andT. Barkay. 1995. An evaluation of mer-specified
      reduction of ionic mercury as a remedial tool of a mercury-contaminated
      freshwater pond. J. Ind. Microbiol. 14:343-348.

154.   Schlegel, H. G, and H. W. Jannasch. 1992. Prokaryotes and their habitats, pp.
      75-125. In: Balows, A., H. G Triiper, M. Dworkin, W. Harder, andK.-H. Schleifer
      (eds.). The Prokaryotes, Second edition. Springer-Verlag, New York, NY.

155.   Schlesinger, W. H. 1991. Biogeochemistry: An Analysis of Global Change.
      Academic Press, Inc., New York, NY.

-------
118
     156.   Schnoor, J. L. 2000. Phytostabilization of metals using hybrid poplar trees,
           pp. 133-150. In: Raskin, I. andB. D. Ensley (eds.). Phytor erne diaiton of Toxic
           Metals. Wiley & Sons, Inc., New York, NY.

     157.   Scholz-Muramatsu, H., etal. 1995. Isolation and Characterization of Dehalo-
           spirillum multivorans gen., sp. nov., a tetrachloroethene-utilizing, strictly
           anaerobic bacterium. Arch. Microbiol.  163:48-56.

     158.   Schultz, L. K, M. Young, andR. M. Higashi. 1999. Sorption-desorptionbehav-
           iorof phenanthrene elucidated by pyrolysis—gas chromatography-mass spec-
           trometry studies of soil organic matter. Environ. Toxicol. Chem. 18:1710-1719.

     159.   Schwartz, E., and K.M. Scow. 1999. Usingbiodegradation kinetics to mea-
           sure availability of aged phenanthrene to bacteria inoculated into soil. Environ.
           Toxicol. Chem.  18:1742-1746.

     160.   Schwarzenbach, R. C., R. W. Scholz, A. Heitzer, B. Staubli, andB. Grossmann.
           1999. Aregional perspective on contaminated site remediation—fate of ma-
           terials and pollutants. Environ. Sci. Technol. 33:2305-2310.

     161.   Schwertmann, U., andR. M. Taylor. 1977. Iron oxides, pp. 145-180. In: Dixon,
           J. B., S. B. Weed, J. A. Kittrick, M. H. Milford, and J. L. White (eds.). Minerals
           in Soil Environments. Soil Science Society of America. Madison, WI.

     162.   Scow, K. M., and C. R. Johnson. 1997.Effectof sorption on biodegradation
           of soil pollutants, pp. 1-56. In: D. L. Sparks  (ed.). Advances in Agronomy,
           Volume 58, Academic Press, Inc., New York, NY.

     163.   Scribner, S. L., T. R. Benzing. S. Sun, and S. A. Boyd. 1992. Desorption and
           bioavailability of aged simazine residues in soil from a continuous corn field.
           J. Environ. Qual.  21:115-120.

     164.   Seagren, E.A., B.E. Rittmannand A.J. Valocchi. 1994. Quantitative evaluation
           of the enhancement of NAPL-pool dissolution by flushing and biodegrada-
           tion. Environ. Sci. Technol. 28:833-839.

     165.   Seeliger, S., R. Cerd-Ruwisch, andB. A. Schink. 1998. Aperipasmic and extra-
           cellular c-type cytochrome of Geobacter sulfurreducens acts as a ferric iron
           reductase and as an electron carrier to other acceptors or to partner bacteria.
           J. Bacterial. 180:3686-3691.

     166.   Semprini, L. 1997a. In Situ Transformation of Halo genated Aliphatic Com-
           pounds under Anaerobic Conditions, pp. 429-450. In: Ward, C.  H., J. A.
           Cherry and M. R. Scalf (eds.). Subsurface Restoration, Ann Arbor Press,
           Inc., Chelsea, MI.

     167.   Semprini, L. 1997b. Strategies forthe aerobic co-metabolism of chlorinated
           solvents. Curr. Opinion Biotechnol. 8: 296-308.

-------
                                                                               119
168.   Semprini, L., P. K. Kitanidis, D. H. Kampbell, and J. T. Wilson. 1995. Anaero-
      bic transformation of chlorinated aliphatic hydrocarbons in a sand aquifer
      based on spatial chemical distributions. Water Resour. Res. 31:1051-1062.

169.   Semprini, L., P. V Roberts, G. D. Hopkins, and P. L. McCarty.  1990. A field
      evaluation of in-situ biodegradation of chlorinated ethenes: Part 2, results of
      biostimulationandbiotransformationexperiments. Ground Water 28:715-727.

170.   Sharma, P., andP. L. McCarty. 1996. Isolation and characterization of a facul-
      tative aerobic bacterium that reductively dehalogenates tetrachloroethene
      to c/'s-l,2-dichloroethene.^4p/>/. Environ. Microbiol. 62:761-765.

171.   Siciliano, S. D., and J. J.  Germida. 1998. Mechanisms of phytoremediation:
      biochemical and ecological interactions between plants and bacteria. Environ.
      Rev. 6:65-79.

172.   Skipper, H. D., andR. F. Turco. (eds.). 1995. Bioremediation:  Science and
      Applications. Soil Science Society of America, Madison, WI.

173.   Smatlak, C. R., J. M. Gossett, and S. H. Zinder. 1996. Comparative kinetics of
      hydrogen utilization for reductive dechlorination of tetrachloroethene and
      methanogenesis in an anaerobic enrichment culture. Environ. Sci. Technol.
      30:2850-2858.

174.   Southgate, D., I. Johnson, and G R. Fenwick (eds.). 1989. Nutrient Availabil-
      ity: Chemical and Biological Aspects. Royal Society of Chemistry, Cam-
      bridge, UK.

175.   Sposito,G  1989.77zeC/ze/w/^qf&/&.OxfordUniversityPress,NewYork,NY

176.   Stacey, S., G. Merrington, and M. J. McLaughlin. 2001. The effect of aging
      biosolids on the availability of cadmium and zinc in soil. Europ. J. Soil Sci.
      52:313-321.

177.   Stanhope, K. G, S. D. Young, J. J. Hutchinson, and R. Kamath. 2000. Use of
      isotopic dilution techniques to assess the mobilization of nonlabile Cd by
      chelating agents in phy to remediation. Environ. Sci. Technol. 34:4123-4127.

178.   Steffan, R. J., K. L. Sperry, M. T. Walsh, S. Vainberg, and C. W Condee. 1999.
      Field-scale evaluation of in situ bioaugmentation for remediation of chlori-
      nated solvents in groundwater. Environ. Sci. Technol. 33:2271-2281.

179.   Steinberg,  S. M., J. J. Pignatello, andB. L. Sawhney. 1987. Persistence of 1,2-
      dibromoethane in soils:  entrapment in intraparticle micropores. Environ.
      Sci. Technol. 21:1201-1208.

180.   Stevenson, F. J. 1994. Humus Chemistry, Second Edition. Wiley & Sons, New
      York, NY

-------
120
     181.   Stout, ID., K.M. Goh,andT. A. Rafter. 1981. Chemistry and turnover of naturally
           occurring resistant organic compounds in soil, pp. 1 -73. In: Paul, E. A. and J. N.
           Ladd (eds.). Soil Biochemistry, \bl. 5. Marcel Dekker, Inc., New York, NY.

     182.   Summers, A. 0.1992. The hard stuff: metals in bio remediation. Curr. Opin-
           ion Biotechnol. 3:271-276.

     183.   Takagi, S-1.1993. Production of phytosiderophores, pp. 111-131. In: Barton,
           L. L. and B. C. Hemming (eds.). Iron Chelation in Plants and Soil Microor-
           ganisms. Academic Press, Inc., New York, NY.

     184.   Tang, J., B. K. Robertson, and M. Alexander. 1999. Chemical-extraction meth-
           ods to estimate bioavailability of DDT, DDE, and DDD in soil. Environ. Sci.
           Technol. 33:4346-4351.

     185.   Thompson-Eagle, E. C., and W. T Frankenberger, Jr. 1992. Bioremediationof
           soils contaminated with selenium. Adv. Soil Sci. 17:261-310.

     186.   Tisdall, J. M., and J. M. Oades. 1982. Organic matter and water-stable aggre-
           gates in soil. J. Soil Sci. 3 3:141 -163.

     187.   Torrens, J.L., D.C.Herman, andR. Miller-Maier. 1998.Biosurfactant(rhamno-
           lipid) sorption and the impact on rhamnolipid-facilitated removal of cadmium
           from various soils under saturated flow conditions. Environ. Sci. Technol.
           32:776-781.

     188.   Troy, M. A., S. W. Berry, and D. E. Jerger. 1993. Biological land treatment of
           diesel fuel-contaminated soil: Emergency response through closure, pp. 145-
           160. In: Flathman, P. E., D. E. Jerger, and J. H. Exner (eds.). Bioremediation
           Field Experience. Lewis Publishers, Boca Raton, FL.

     189.   US EPA. 1999. Treatment technologies for site cleanup: annual status re-
           port. 9th ed. EPA-542-R99-00, Number 9, April 1999.

     190.   VanderLelie, D., J.-P Schwitzbuebel, D. J. Glass, J. Vangronsveld, and A.
           Baker. 2001. Assessing phytoremediation's progress in the United States
           and Europe. Environ. Sci. Technol. 35:447-452.

     191.   van Ginkel, C. G, C. M. Plugge, and C. A. Stroo. 1995. Reduction of chlorate
           with various energy substrates  and inocula under anaerobic conditions.
           Chemosphere 31:4057-4066.

     192.   VanLoosdrecht, M. C. M., J. Lyklema, W. Norde, and A. J. B. Zehnder. 1990.
           Influence of interfaces on microbial activity. Microbiol. Rev. 54:75-87.

     193.   Vandegrift, G. K, D. T. Reed, and I. R. Tasker (eds.). 1992. Environmental
           Remediation: Removing Organic and Metal Ion Pollutants. American Chemi-
           cal Society, Washington, DC.

-------
                                                                               121
194.   Vangronsveld, I, and S. D. Cunningham. 1998. Introduction to the Con-
      cepts, pp. 1-15. In: Vangronsveld, J. and S. D. Cunningham (eds.). Metal-
      Contaminated Soils: In Situ Inactivation and Phytorestoration. Springer-
      Verlag, New York, NY.

195.   Vangronsveld, J., and S. D. Cunningham. 1998. Metal-Contaminated Soils:
      In situ Inactivation and Phytorestoration. Springer-Verlag, New York, NY,
      265pp.

196.   vanVeen, J. A., and E. A. Paul 1981. Organic carbon dynamics in grassland
      soils. 1. Background information and computer simulation. Can. J. SoilSci.
      61:185-201.

197.   Verstraete, W., and W. Devliegher. 1997. Formation of non-bioavailable or-
      ganic residues in soil: Perspective for site remediation. Biodegradation 7:471-
      485.

198.   Videla, H. A., and W. G. Characklis. 1992. Biofouling and microbially influ-
      enced corrosion. Int. Biodeterior. Biodegrad. 29:195-212.

199.   Volkering, K, A. M. Breure, and W. H. Rulkens. 1998. Microbiological aspects
      of surfactant use for biological soil remediation. Biodegradation 8:401 -417.

200.   Wackett L. P., and C. L. Hershberg. 2000. Biocatalysis and Biodegradation.
      American Society for Microbiology, Washington, DC, 228 pp.

201.   Wackett, L. P. 1996. Co-metabolism:  is the emperor wearing new clothes?
      Curr. OpinBiotechnol. 7:321-325.

202.   Wallace, W,  S. Beshear, D. Williams, S. Hospadar, and M. Owens. 1998.
      Perchlorate reduction by a mixed culture in an up-flow anaerobic fixed bed
      reactor. J.Indust.Microbiol. Biotechnol. 20:126-131.

203.   Warren, L. A., and F. G. Ferris. 1998. Continuum between sorption and pre-
      cipitation of Fe(III)  on microbial surfaces. Environ. Sci. Technol. 32:2331-
      2337.

204.   Welter, E., W. Calmano, S. Mangold, and L. Troger, 1999. Chemical speciation
      of heavy metals in soils by use of XAFS spectroscopy and electron micro-
      scopical techniques. Fresenius J. Anal. Chem. 364:238-244.

205.   West, C. C., and J. H. Harwell. 1992. Surfactants and subsurface remediation.
      Environ. Sci. Technol. 26:2324-2330.

206.   White, J. C., and J. J. Pignatello.  1999. Influence of bisolute competition on
      the desorption kinetics of polycyclic aromatic hydrocarbons in soil. Envi-
      ronment. Sci. Technol. 33:4292-4298.

-------
122
     207.   White, J. C., M. Hunter, J. J. Pignatello, andM. Alexander. 1999. Increase in
           bioavailability of aged phenanthrene in soils by competitive displacement
           withpyrene. Environ. Toxicol. Chem. 18:1728-1732.

     208.   Whitlock, J. L. 1990. Biological detoxification of precious metal processing
           wastewaters. Geomicrobiol. J. 8:241-249.

     209.   Willardson, B. M., J. F. Wilkins, T. A. Rand, J. M. Schupp, K. K. Hill, P. Keim,
           and P. J. Jackson. 1998. Development and testing of a bacterial biosensor for
           toluene-based environmental contaminants. Appl. Environ. Microbiol. 64:1006-
           1012.

     210.   Williams, C. J., D. Aderhold, and R. G J. Edyvean. 1998. Comparison between
           biosorbents  for the removal of metal ions from aqueous solutions. Water
           Res. 32:216-224.

     211.   Wolf, D. C., andH. D. Skipper. 1994. Soil sterilization, pp. 41-51. In: Weaver, R.
           W, S. Angle, P. Bottomley, D. Bezdicek, S. Smith, A. Tabatabai, and A. Wollum
           (eds.). Methods of Soil Analysis, Part 2. Microbiological and Biochemical
           Properties. Soil Science Society of America, Inc., Madison, WI.

     212.   Xia, G. 1998. Sorption behavior of nonpolar organic chemicals on natural
           sorbents  (polyanyi theory). Ph.D.  dissertation. Johns Hopkins University.

     213.   Young, L. Y, and C. E. Cerniglia. 1995. Microbial Transformation and Deg-
           radation of Toxic Organic Chemicals. Wiley Liss, Inc., New York, NY.

     214.   Young, T. M., and W. J. Weber. 1995.  A distributed reactivity model for
           sorptionby soils and sediments. 3. Effects of diagenetic processes on sorp-
           tionenergetics. Environ. Sci. Technol. 29:92-97.

     215.   Zumft, W. G.  1977. Cell biology and molecular  basis of denitrification.
           Microbiol. Mol. Biol. Rev. 61:533-616.

-------
United
Environmental
Office of
National Ciller for Environmental         4*70 1 R)
Washington, OC
www.epa.Qov/ncer

Official Business Only
Penalty for Private
§300

-------

-------