Table of Contents
I. Introduction
II. Brief History of XAFS Spectroscopy
III. Recent Studies at SSRL and Their Impacts on Environmental Remediation and Technology

• Reductive Remediation of Cr(VI)-Contaminated Groundwater by Iron-Containing Reactive Barriers
• XAFS Studies of Radionuclides
• Phytoremediation of Cadmium- and Chromium-Contaminated Soils by Plants
• Speciation of Lead and Arsenic in Mine Wastes and Sediments:  Implications for Risk Assessment and Remediation Strategies
• Speciation of Se in Soils at Kesterson Reservoir, California

       Anthropogenic contaminants such as heavy metals and radionuclides enter the biosphere through a number of human activities, including mining, industrial and domestic activities, nuclear weapons production, and the burning of fossil fuels (Figure 1).  In addition to these sources, leakage of HLW and fissile materials stored in repositories could lead to serious contamination problems.  Contaminants are commonly transported and dispersed in the subsurface by groundwater in response to the hydraulic gradient, resulting in contaminant plumes such as shown in Figure 1.  Contaminant species may react strongly with soil and aquifer particles, resulting in retarded or facilitated dispersion, or they may react only weakly or not at all.  In addition, they may react with dissolved complexing agents, become incorporated in relatively insoluble precipitate phases, be transformed abiotically into relatively benign forms, used as energy sources by microorganisms, or be taken up and sequestered by plants, all of which affect the ultimate fate of contaminant species.
 

        Synchrotron-based spectroscopic techniques, such as XAFS (x-ray absorption fine structure) spectroscopy, have proven to be among the most useful tools for characterizing the speciation of metal ions, metal-inorganic anion complexes, and metal-organic complexes in natural samples, the chemical processes that govern their transformations, and their reaction rates.  Synchrotron x-ray sources provide tunable, high intensity (e.g., 10¹¹ photons/sec) x-rays over a broad energy range (> 10³ eV), making possible element-specfic measurements at the low concentrations (1000 ppm to ppb ranges) that are typical of environmental samples.  Such high-intensity x-rays cannot be produced using conventional sealed-tube or rotating anode x-ray generators.  In addition, fluorescence-yield detection enables the collection of XAFS spectra in-situ from water-bearing samples at environmentally relevant temperatures and pressures.  Wet samples are critical to environmental studies since water is ubiquitous in the environment and natural samples are often modified by drying and crushing procedures and vacuum conditions.  XAFS spectroscopy provides qualitative and quantitative information about the oxidation states, coordination environments, and short-range order of metal ions, including metal-oxygen bond distances and coordination numbers.   No other structure characterization techniques provide such element-specific information for metal ions in micro-, poorly crystalline or amorphous solids, liquids, and gases.  Thus, XAFS spectroscopy has begun to play a very important role in MES research.

        The success of XAFS spectroscopy as a tool for characterizing metal-ion speciation in complex natural samples has led to a large increase in demand for synchrotron beamtime, as scientists from a wide range of environmental disciplines such as environmental chemistry/geochemistry, environmental microbiology, plant biology, and soil science, begin to utilize this tool.  For example, at SSRL, the number of eight-hour shifts awarded for MES research has increased from 402 shifts in 1993 to 1,075 shifts in 1996.  Currently, all SSRL beamlines are oversubscribed.  XAFS studies are increasingly being coupled with other synchrotron-based techniques such as micro-XAFS spectroscopy, soft x-ray spectromicroscopy, x-ray standing wave studies, and x-ray scattering methods, which provide complementary information about the speciation and spatial distribution of metal ions in samples.

       In response to this growth and the need for staff support of MES users, SSRL hired a MES scientific staff member (John Bargar) in late 1996 and is constructing a high-flux beamline (BL 11-2) dedicated to MES research (funded by DOE-BES-Chemical Sciences).  This beamline will provide approximately 3 to 10 times more flux than SSRL beamlines commonly used for MES research over the energy range 5 - 23 keV, which includes the K-absorption edges of transition metals and the L-edges of heavy metals and actinides.  In addition, the beamline will feature an experimental hutch designed for XAFS measurements of toxic and/or radioactive samples, and will be outfitted with a high throughput, 30-element Ge detector for use on compositionally heterogeneous and dilute natural samples.  Thus, it will be possible to study samples at least 50 times more dilute than currently possible with SSRL beamlines commonly used for MES research.  For example, it will be possible to study metal ion adsorption at defect sites (which are typically present at low surface concentrations) on single crystal surfaces.  Such model system studies are essential for defining the chemical processes that determine the behavior of contaminant species.  The amounts of radioactive materials required for analysis will be smaller, resulting in increased safety.  Furthermore, the higher flux will facilitate micro-XAFS measurements at 10 mm length scales, making it possible to determine the spatial distribution of contaminant species in heterogeneous samples.

II. Brief History of XAFS Spectroscopy

        The first x-ray absorption spectrum was observed by de Broglie in 1913 who noticed two sharp and intense features on a photographic film used in a rotating crystal spectrograph.  These features proved to be absorption lines of Ag and Br contained in the photographic emulsion on the film.  Because of the high energies of these absorption lines (25.51 and 13.47 keV, respectively), these spectra were of low resolution and no fine structure was observed.  During this same time period, C.G. Barkla began to systematically examine the phenomena of x-ray absorption and emission.  He discovered that elements absorb most strongly at discrete x-ray energies, referred to as absorption edges, and that these discrete absorption energies are characteristic of the element.  The Cu K-absorption edges shown in Figure 2 illustrate the sharp rise in x-ray absorbance when the x-ray energy matches the binding energy of the 1s electron of Cu.
 

       Fine structure associated with K absorption edges of the elements Mg, Fe, and Cr in condensed matter was first observed by H. Fricke in 1920.  During the next ten years, others examined the K and L absorption edges of additional elements in various phases including gases and aqueous solutions.  Simultaneously, Kronig began developing a theory based on Bragg scattering of the ejected photoelectrons to explain the extended fine structure, which bears his name (Kronig structure).  Considerable effort was spent in developing both long-range and short-range scattering theories for XAFS in the forty years following Kronig's theoretical work.  In the mid-1960's, Lytle and Levy first used short-range scattering theories to derive accurate bond distances from EXAFS (extended XAFS, which is the name given to the XAFS region starting about 30 eV above the absorption edge and extending up to 1000 eV above the edge).  However, the real breakthrough in developing the modern methods of EXAFS analysis occurred in 1971 when Dale Sayers and Edward Stern of the University of Washington and Farrell Lytle of the Boeing Corporation in Seattle, Washington, recognized that Fourier transformation of EXAFS data, with respect to the photoelectron wave vector k, gives peaks at distances corresponding approximately to the distances to nearest neighbors around an absorber.  Fourier transform-derived distances differ from the exact absorber-backscatter distances due to phase-shift effects caused by the interaction of the scattered photoelectron with the electric fields of the absorbing and scattering atoms.  However, accurate absorber-backscatterer distances can be obtained when the EXAFS data are corrected for these phase-shift effects.  Sayers and Lytle performed the first synchrotron-based EXAFS experiment at SSRL in 1974.  This experiment, coupled with the availability of synchrotron light sources to general users, started the modern era of XAFS spectroscopy.  Use of this method in many diverse scientific disciplines has grown over the past 25 years, and XAFS methods (including XANES - x-ray absorption near edge structure - spectroscopy) are now well established.  The experimental methods of XAFS spectroscopy have improved substantially since the first synchrotron-based experiments.  Over the past decade there have been substantial improvements in the stability and availability of high-flux, high-brightness synchrotron sources, as well as in beam lines optics, and x-ray detectors (brightness is a measure of the vertical and horizontal dimensions, divergence, and flux of a synchrotron x-ray beam).  These advances have extended the usefulness of XAFS in MES research by allowing detection and characterizations of increasingly lower concentrations of contaminant species.  High-brightness third-generation synchrotron light sources have provided the capability to determine the spatial distributions of chemical species for a given contaminant element.  Furthermore, first-principles ab-initio multiple-scattering methods have been developed and are available to users through computer codes such as FEFF (Rehr and Albers, 1990; Rehr et al., 1991; Rehr et al., 1992) which can accurately calculate the EXAFS spectrum of an element in any structural environment.  These theoretical advances have moved EXAFS spectroscopy from a semi-quantitative structural method to a fully quantitative one that rivals x-ray scattering for structural studies of elements in certain classes of materials (e.g., amorphous materials and liquids) and surpasses most other structural methods for experimental studies of elements at environmental interfaces.  Efforts are currently underway to extend the usefulness of theoretical codes to XANES regions of XAFS spectra, which are dominated by multiple-scattering processes.  With this next advance in theory, XANES spectroscopy should move from a semi-quantitative to a quantitative method, which will have a significant impact on MES research due to the considerably higher amplitude of the adsorption edge of an element relative to its EXAFS.  This development will permit characterization of contaminant speciation at even lower concentrations than is now possible with EXAFS spectroscopy alone.

III. Recent Studies at SSRL and Their Impacts on Environmental Remediation and Technologies

        The ability to probe the speciation, spatial distribution, and reactivity of metal ion contaminants in natural samples at SSRL has been utilized by a growing number of scientists to investigate the reaction of metal ions including arsenic, cadmium, chromium, lead, selenium, and uranium with environmentally important oxides and plants.  Highlights from some of these investigations are described in the sections that follow.

Reductive Remediation of Cr(VI)-Contaminated Groundwater by Iron-Containing Reactive Barriers  Cr(VI) is one of the most widespread and toxic anthropogenic contaminants in groundwaters in the US, present at nuclear weapons complex facilities, DOD installations, and numerous industrial areas because of its common use in electroplating metals and its presence in metal components used in nuclear fuel rods.  Current EPA and DOE regulations call for remediation of Cr-contaminated soils and groundwater.  The scale of Cr(VI) contamination and its toxicity dictate that cost-efficient and effective remediation technologies be developed.  One type of proposed environmental remediation strategy is the construction of in-situ permeable reactive barriers (PRBs), which utilize natural-gradient groundwater flow to transport contaminated water through a reactor system (in its simplest form this is a trench filled with inexpensive reactive solid material).  Since contaminants are remediated in-situ in PRBs (i.e., the groundwater is remediated in place), they are energy-efficient, conservative of groundwater resources, and have significantly lower operations and maintenance costs than alternative groundwater remediation technologies such as pump-and-treat methods.  It has been proposed that zero-valent iron (e.g., iron pellets) could be an effective PRB reactant material for treatment of Cr(VI)-contaminated groundwaters because iron can reduce Cr(VI) to Cr(III), followed by precipitation of Cr(III) hydroxide phases or adsorption to particulate surfaces.  However, few studies have been performed to confirm the reduction of Cr(III) by iron or to characterize the reaction products that occur in zero-valent iron PRBs.  Hence, little is known about the reactivity and viability of zero-valent iron PRBs to retard Cr(VI) over long periods of time (Shoemaker et al., 1995).  Several mechanisms of PRB inhibition and failure have been acknowledged.  It is of particular concern that reaction of Cr(VI) with iron particles will result in the formation of Fe(III)-oxide coatings on the iron particles, which, when sufficiently thick, could arrest further Cr(VI) reduction, i.e., the iron particle surfaces would become passivated by a non-conductive coating.

        In order to understand the fundamental controls on Cr(VI) reduction by iron, the reaction of Cr(VI) with Fe(II,III)-oxides, which are expected to thinly encrust iron particles exposed to groundwater, was studied in a collaborative research effort at SSRL (Peterson et al., 1996; Peterson et al., 1997a; Peterson et al., 1997b) by Maria Peterson and Gordon Brown of the Department of Geological & Environmental Sciences, Stanford University, in collaboration with  Art White of the US Geological Survey (Menlo Park, CA) and Carol Stein of the University of Washington (Seattle).  XAFS spectroscopy was used to characterize the oxidation state and local structure of chromium reacted in simplified synthetic models and in chromium-contaminated soil samples from Sandia National Laboratory and from the U.S. Naval Research Weapons Lab in Keyport, Washington.  Samples were also characterized by high-resolution transmission electron microscopy and batch kinetic studies.

        The results of these studies indicate that that natural Fe²⁺-containing iron oxide phases such as magnetite (Fe₃O₄) do reduce Cr(VI) to Cr(III), as indicated by x-ray absorption near-edge structure (XANES) spectroscopy.  The usefulness and sensitivity of Cr K-edge XANES structure to chromium oxidation state and local coordination geometry are illustrated in Figure 3, which shows a comparison of Cr(VI) and Cr(III) XANES spectra.  The characteristic features in these spectra can be used to quantify the ratio of Cr(III):Cr(VI) in dilute samples.  XAFS is the only technique capable of providing such definitive oxidation state information from dilute samples under environmentally relevant conditions, and these experiments could not have been done without using a synchrotron light source.  An electron-transfer reaction at the magnetite-aqueous solution interface is implicated by these findings, which induces a transformation of Fe₃O₄ surfaces to g-Fe₂O₃.  This reaction halts when the thickness of the g>-Fe₂O₃ layer reaches about 15 Å on the magnetite surface.  Figure 4 is a TEM image of the Cr(III)-bearing g-Fe₂O₃ passivating layers that form on the surfaces of magnetite particles.  Such passivation reactions are also expected to occur in zero-valent iron PRBs and could lead to their failure after relatively short time periods.  Zero-valent iron PRBs have also been proposed for reductive remediation of radioactive Tc(VII) and U(VI) in groundwater at US DOE sites, and reductive dechlorination of halogenated hydrocarbons such as trichloroethylene (TCE), which is a widespread organic contaminant in the U.S.  Studies of these reduction reactions on iron oxides are currently underway as part of a Stanford University-PNNL collaboration funded by the Environmental Management Science Program.  Since oxidative passivation of the iron particle surfaces could occur in all of these systems, the results of Peterson et al. indicate that design and selection of materials for PRBs must account for these reactions in order to achieve cost-effective, long-term remediation.  These findings are also relevant to technological products, such as film substrates and nanoparticle ultracapacitors, in which the surface properties of Fe(II,III) oxides are engineered and utilized.
 

Figure 4.  TEM images of magnetite before and after reaction with Cr(VI) (left), and XANES    spectra of Cr(VI) reacted with magnetites that have increasing (bottom to top) amounts of surface oxidation (right).  (a) Fe₂O₃, (b) 9 month-old magnetite, (c)-(e) 7 month-old magnetite, (f) 1 month-old magnetite.  From Peterson et al. (1996;1997c).  The presence of Cr⁶⁺ pre-edge features in spectra (a) - (e) indicate that Fe(III)-oxide coatings on magnetite inhibit reduction of Cr⁶⁺ to Cr³⁺.

        In a related study, Steve Conradson and others have used XAFS spectroscopy and x-ray scattering methods to examine the effects of aging on Pu alloys that comprise the pits of nuclear weapons, a major effort in the science-based stockpile stewardship program.  These studies have shown that significant changes in the density of these samples with time is related to major changes in the local coordination environment around Pu in these alloys caused by radiation damage.

Phytoremediation of Cadmium- and Chromium-Contaminated Soils by Plants  In cases where pollutants are restricted to near-surface environments such as soils, marshes, and rivers there is great interest remediating the contaminated soils and waters using plants that hyperaccumulate heavy metals and transform them into benign chemical forms, a process known as phytoremediation.  For example, there is much interest in constructing wetlands planted with phytoremediating crops for use in treating contaminant-laden industrial effluents.  The contaminant-bearing plant matter can be harvested and treated if volatilization of the metal does not occur.  In order for a plant to be useful for phytoremediation, it should have high biomass and be capable of absorbing metal ions through its roots and accumulating them in its shoots and leaves, which can then be harvested.  The key advantages of this technology are its energy efficiency and low labor requirements, which dramatically reduce remediation costs.  Competing remediation technologies, such as soil washing, are energy consumptive, rely heavily upon labor and infrastructure, and typically are not as effective as desirable.  Molecular-scale studies of metal-ion uptake, transport, and fixation within such plants, including genetic engineering of plants with specific affinities for certain metals, should facilitate the development of effective phytoremediation crops.    Furthermore, since plants can provide the major route of metal-ion intake to the human diet, studies of metal-ion uptake by plants will be useful in establishing the potential hazards of growing crops in soils treated with metal-bearing composted sewage or contaminated through other means.  XAFS spectroscopy is a particularly useful technique for studying metal-ion speciation in plant tissues because it can be performed at very low metal ion concentrations (ca 10 mM) in situ, without harsh extraction treatments or desiccation of plant materials.  Moreover, micro-XAFS studies of hyperaccumulating plants can reveal the spatial distribution of different metal-ion species, indicating the specific site of accumulation and transformation.  Such information is critical for establishing the mechanisms of heavy metal uptake, transport, transformation, and/or fixation by plants.
 

Figure 5.  XANES spectra of Cr6+ in aqueous nutrient solution and Cr3+ in shoot and root tissue.  Cr³⁺ in aqueous solution is presented at the bottom to show the shape of the Cr³⁺ edge.  From Lytle et al. (1997).

        Arsenic most commonly occurs in the oxidation states (0), (III), and (V) in natural samples.  As(III) is the most toxic form to humans and is more readily transported in groundwater than the (0) and (V) oxidation states.  Selective chemical extractions, which have traditionally been used to determine the oxidation states of “surface-bound” arsenic, cannot distinguish between arsenic in relatively insoluble precipitates, mineral grain coatings, fine-grained particles, and adsorbed arsenic.  With renewed interest by the U.S. Environmental Protection Agency in lowering the maximum As levels in drinking water, it is important to obtain more accurate information on the speciation of As in contaminated soils and mine tailings than is possible with selective extractions.  XAFS spectroscopy is well suited to this problem, since it can provide information on the oxidation state and local coordination environments of arsenic in dilute natural samples.  In the case of arsenic-contaminated mine samples, XANES analyses showed that the dominant forms of arsenic are As(V) and As(0) (Foster et al., 1997).  The more toxic and mobile As(III) species was not detected in this work.  Moreover, As(V) was found in both crystalline arsenate phases and as chemisorbed species on mineral surfaces.  The stability of these crystalline species and the amount of chemisorption may fluctuate substantially on diurnal and annual time scales, due to fluctuations in the pH and composition of groundwater.  Risk assessment models for arsenic contamination in these environments must account for these species in order to accurately predict arsenic transport and toxicity.  Furthermore, effective remediation strategies must be capable of removing or neutralizing As(0) and As(V) in solid phases, and surface-bound arsenic.

        In another set of investigations, mine tailings in Leadville, CO were examined to characterize surface Pb species (Ostergren et al., 1996).  Previous work by the authors and graduate student John Ostergren in the Department of Geological & Environmental Sciences, including electron probe micro-analysis (EPMA), x-ray photoelectron spectroscopy (XPS), and “selective” chemical extractions, have indicated that surface species may account for up to 50% of the total lead in these tailings samples.  Our EXAFS results confirm that both of the Leadville tailings contain significant amounts of lead adsorbed to oxide surfaces.  However, the EXAFS results also indicate that MgCl₂ sequential extractions, which are commonly assumed to be effective at removing weak surface-bound lead species, resulted in a redistribution of lead species.  These results indicate that use of MgCl₂ extractions to quantify adsorbed lead in for predictive risk assessment models could lead to substantial error.

Speciation of Se in Soils at Kesterson Reservoir, California  Selenium occurs naturally in sediments and soils in many parts of the Western U.S.  When such soils are irrigated for agricultural purposes, this element becomes soluble and is transported in agricultural drainage waters to ponds and reservoirs where it concentrates.  Plants and animals that grow in these waters can bioaccumulate selenium (up to 3000 ppm), leading to selenium toxicity in the food chain.  One result of this concentration process was discovered in the early 1980's by government scientists at the Kesterson National Wildlife Refuge in Merced County, California.  Wildlife, particularly waterfowl, died or were born deformed after they had ingested high levels of selenium.  Similar problems have since been documented at nine other sites comprising some 1.5 million acres of farmland in eight Western states.  This problem has major financial and health implications in the San Joaquin Valley of California, a vast area that produces a major portion of the nation's vegetables, fruits, and other crops.  If farmers are prevented from draining irrigation waters in this region, which has been proposed as one solution to this problem, selenium salts will rapidly build up in soils, making production of crops impossible.  About 500,000 acres of farmland in the San Joaquin Valley - one quarter of the Valley's agricultural acreage - are at stake, with an annual crop production worth about $500 million.  Other proposed solutions to this problem include bacterial reduction, immobilization, and removal of selenium from drainage waters or the use of drainage waters to irrigate land on which salt-resistant plants are grown, such as cotton, Eucalyptus trees, and atriplex.  None have been adopted and some skeptics doubt that a viable solution, which satisfies environmental, financial, and political constraints, will be found.

       One of the keys to understanding and potentially solving this problem is knowledge of the speciation, or chemical forms, of selenium in affected soils and groundwaters and the chemical processes that cause selenium(VI) to transform into more benign species.  XAFS studies of selenium speciation in soils from the Kesterson Reservoir region have been performed by SSRL scientists, including Ingrid Pickering and Gordon Brown, Jr. in collaboration with Tetsu Tokunaga, a soil scientist at Lawrence Berkeley National Laboratory (Pickering et al., 1995; Tokunaga et al., 1996).  This work has shown that selenium in the Kesterson soils occurs  in the elemental form, which is insoluble and immobile.  Furthermore, the elemental state was shown to be stabilized against reoxidation under oxic conditions, such as occur during droughts, by organic matter.  These results indicate that any solution to this problem must account for the presence of both oxidized and reduced forms of selenium in the soils.  The natural occurrence of elemental selenium in Kesterson soils suggests that remediation strategies based on reduction of oxidized selenium species to Se(0), such as microbial treatment of agricultural drainage waters, could be successful solutions to this problem.

        Many studies have suggested that the primary pathway for selenium reduction in contaminated soils involves microbial activity.  However, a recent XAFS study by Satish Myneni, Tetsu Tokunaga (LBNL) and Gordon Brown (Stanford University) conducted at SSRL has shown that rapid reduction of Se(VI) can be achieved abiotically when Se(VI) in solution reacts with "green rust" (Fe(II)₄Fe(III)₂(OH)₁₂SO₄ • 3H₂O), a common Fe(II)-containing phase in the near surface environment (Myneni et al., 1997).  This discovery provides a heretofore unknown alternative pathway for the reduction of selenium and other redox-sensitive oxoanions such as As(V)O₄^3- and Cr(VI)O₄^2-).


IV. REFERENCES

Foster A. L., Brown G. E., Jr., Tingle T. N., and Parks G. A. (1998) Quantitative arsenic speciation in mine tailings using x-ray absorption spectroscopy. Am. Mineral. 83, 553.

Lytle, C.M.,Lytle, F.W., Yang, N., Qian, J.H., and Terry, N. (1997) Phyto-conversion of Cr⁶⁺ to Cr³⁺ by water hyacinths (Iechhornia crassipes Solmsp.).  Science (in submission).

Morris D.E., Allen P.G., Berg J.M., Chisholm-Brause C.J., Conradson S.D., Donohoe R.J., Hess N.J., Musgrave J.A., and Tait C.D. (1996) Speciation of uranium in Fernald soils by molecular spectroscopic methods: Characterization of untreated soils. Environmental Science and Technology 30, 2322-2331.

Myneni, S.C.B., Tokunaga T.K., Brown, G.E., Jr. (1997) Abiotic selenium redox transformations in the presence of Fe(II,III) oxides. Science 278, 1106-1109.

Ostergren J.D., Tingle T.N., Brown G.E., Jr., and Parks G.A. (1996) Lead speciation in selected mine wastes from Leadville, CO: chemical extraction, EPMA, XPS, and XAFS investigations. Abstracts with Programs,  Geological Society of America, Annual Meeting, A-519.

Peterson M., Brown G. E., Jr., and Parks G. A. (1996) Direct XAFS evidence for heterogeneous redox reactions at the aqueous chromium/magnetite interface. Colloids and Surfaces A 107, 77 - 88.

Peterson M. L., Brown G. E., Jr., and Parks G. A. (1997a) Quantitative determination of chromium valence in environmental samples using XAFS spectroscopy. In Aqueous Chemistry and Geochemistry of Oxides, Oxyhydroxides, and Related Materials (ed. J.A. Voigt, T.E. Wood,, B.C. Bunker,  W.H. Casey, L.J. Crossey), pp. 75-80, Materials Research Society.

Peterson M. L., Brown G. E., Jr., Parks G. A., and Stein C. L. (1997b) Differential redox and sorption of Cr(III/VI) on natural silicate and oxide minerals: EXAFS and XANES results. Geochim Cosmochim Acta 61, 3399-3412.

Peterson M. L., White, A.F., Brown G. E., Jr., Parks G. A. (1997c)  Surface passivation of magnetite (Fe₃O₄) by reaction with aqueous Cr(VI): XAFS and TEM results. Environ. Sci. Technol. 31, 1573-1576.

Pickering I. J., G.E. Brown J., and Tokunaga T. K. (1995) Quantitative speciation of selenium in soils using s-ray absorption spectroscopy. Environ. Sci. Technol. 29, 2456-2459.

Rehr J. J. and Albers R. C. (1990) Scattering-matrix formulation of curved-wave multiple-scattering calculations of x-ray absorption fine structure. Phys. Rev. B 41, 8139-8149.

Rehr J. J., Mustre de Leon J., Zabinsky S. I., and Albers R. C. (1991) Theoretical x-ray absorption fine structure standards. J. Amer. Chem. Soc. 113, 5135-5140.

Rehr J. J., Zabinsky S. I., and Albers R. C. (1992) High-order multiple-scattering calculations of x-ray absorption fine structure. Phys. Rev. Lett. 69, 3397.

Salt D. E., Prince R. C., Pickering I. J., and Raskin I. (1995) Mechanisms of cadmium mobility and accumulation in indian mustard.  Plant Physiology 109, 1427-1433.

Shoemaker S. H., Greiner J. F., and Gillham R. W. (1995) Permeable Reactive Barriers. In Assessment of Barrier Containment Technologies (ed. R. R. Rumer and J. K. Mitchell), pp. 301-353. National Technical Information Service.

Stern, E.A., and Heald, S. M. (1983)  Basic principles and applications of EXAFS.  In Handbook on Synchrotron Radiation (ed., E. Koch), pp. 955 - 1014.  North Holland Publishing, Amsterdam.

Tokunaga T., Pickering I. J., and Brown G. E., Jr. (1996) Selenium transformations in ponded sediments. Soil Sci. Soc. Am. J. 60, 781 - 790.

Last Updated August 5, 1998 by Trinh Van