Hydrogel-Encapsulated Soil: A Tool to Measure ... - ACS Publications

Mar 15, 2010 - AND DAVID B. WATSON. Environmental Science Division, Oak Ridge National. Laboratory, Oak Ridge, Tennessee 37831-6036. Received ...
0 downloads 0 Views 767KB Size
Environ. Sci. Technol. 2010, 44, 3047–3051

Hydrogel-Encapsulated Soil: A Tool to Measure Contaminant Attenuation In Situ BRIAN P. SPALDING,* SCOTT C. BROOKS, AND DAVID B. WATSON Environmental Science Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831-6036

Received December 31, 2009. Revised manuscript received February 25, 2010. Accepted February 26, 2010.

Hydrogel encapsulation presents a novel and powerful general method to observe many water-solid contaminant interactions in situ for a variety of aqueous media including groundwater, with a variety of nondestructive analytical methods, and with a variety of solids including contaminated soil. After intervals of groundwater immersion, polyacrylamide hydrogelencapsulated solid specimens were retrieved, assayed nondestructively for uranium and other elements using X-ray fluorescence spectroscopy, and replaced in groundwater for continued reaction. Desorption dynamics of uranium from contaminated soils and other solids, when moved to uncontaminated groundwater, were fit to a general twocomponent kinetic retention model with slow-release and fastrelease fractions for the total uranium. In a group of Oak Ridge soils with varying ambient uranium contamination (1691360 mg/kg), the uranium fraction retained under long-term in situ kinetic behavior was strongly correlated (r2 ) 0.89) with residual uranium after laboratory sequential extraction of watersoluble and cation-exchangeable fractions of the soils. To illustrate how potential remedial techniques can be compared to natural attenuation, thermal stabilization of one soil increased the size of its long-term in situ retained fraction from 50% to 88% of the total uranium and increased the halflife of that long-term retained fraction from 990 to 40000 days.

Introduction The general concept of allowing hydrogel-encapsulated solids, ranging from engineered or natural materials to solid wastes, to equilibrate with contaminants in a wide range of aqueous environments, from seawater to the human gastrointestinal tract, may allow direct measurements of how much and how fast particular chemical species change on the solids in these natural environments, which are difficult to simulate or to extrapolate from laboratory tests. The overall process by which groundwater contaminants become less available when reacting and aging in soil has been termed “natural attenuation” (1, 2). Such processes are thought and hoped to reduce risks for release of and exposure to many contaminants from hazardous waste sites. Unlike biodegradable organic contaminants, the attenuation of toxic metals and long-lived radionuclides can only increase through strong adsorption to soil or by contaminants reacting * Corresponding author phone: 865-382-4972; fax: 865-576-8646; e-mail: [email protected]. 10.1021/es903983f

 2010 American Chemical Society

Published on Web 03/15/2010

to inherently less soluble forms. Although natural attenuation is being increasingly selected as a preferred alternative for remediation of many hazardous waste sites, the Committee on Intrinsic Remediation of the National Research Council (2) has warned, “... that rigorous protocols are needed to ensure that natural attenuation potential is analyzed properly, and that natural attenuation should be accepted as a formal remedy for contamination only when the processes are documented to be working and are sustainable.” However, development of protocols to measure contaminant attenuation, whether natural or manipulated, particularly for toxic metals and long-lived radionuclides, remains an important and difficult challenge for environmental scientists (3). Present technical approaches have focused on inferences based on measurements of large distribution coefficients for contaminants between soil and groundwater, sequential extraction of contaminants from soil requiring increasingly harsher reagents, slow isotopic dilution rates for spiked soluble contaminants with ambient contaminants, measurements of decreasing contaminant concentrations during groundwater monitoring over many years, and theoretical approaches based on geochemical modeling of plausible contaminant phases (2, 3). Field approaches that repeatedly sample soil and groundwater, measuring soluble and adsorbed contaminant concentrations over prolonged intervals, are the most direct and costly. However, the necessity of costly and destructive field sampling of contaminated soil, with its inherent heterogeneity and resulting large sampling variance, makes small changes and trends in contaminant attenuation extremely difficult to discern. Hydrogel encapsulation of soil offers an approach to study contaminant adsorption and desorption to and from soil in groundwater in the field with significantly less sampling variance, if the same samples could be used continuously. An ideal hydrogel for soil encapsulation should have several desired properties, including quantitative retention of encapsulated solid particles over its interval of use, chemical inertness to contaminant interactions in the concentration ranges of interest, freedom from significant biodegradation or alteration during use, relatively high permeability to contaminants so that encapsulated soil can equilibrate rapidly with dissolved contaminants, and freedom from interference by soil or other candidate solids with its hydrogel-forming reactions. Polyacrylamide is a strong candidate for such a hydrogel matrix, likely having many of these desirable properties. Polyacrylamide has been used to encapsulate ion exchange resins to make diffusion gradient thin films for deployment in natural waters and sediments to measure, after destructive extraction, available and dissolved cations in sediments (4); these investigators have also reported diffusion coefficients of many cations in these hydrogels, usually finding magnitudes only slightly smaller than those in water. We have previously used polyacrylamide hydrogels to encapsulate soils containing added radiotracers (85Sr and 134 Cs), using nondestructive gamma ray spectroscopy to follow their retention when deployed in groundwater for up to six months (5). Polyacrylamide hydrogels have been used as inert in vivo biomaterials (6) and for many applications to diffusively release pharmaceuticals, nutrients, and additives in a variety of aqueous environments (7). To establish the potential of the hydrogel-encapsulation technique as a valid tool to study contaminant interactions with groundwater in situ, our investigation focused on several proof-of-principle objectives using uranium as a model contaminant. The first objective was to establish that uncontaminated soil, when encapsulated in hydrogels and VOL. 44, NO. 8, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

3047

placed into contaminated groundwater, could take up a contaminant like uranium, and the process could be followed over time using a nondestructive analytical technique on the same specimens when repeatedly retrieved and replaced. The second objective was to determine if hydrogel-encapsulated contaminated soil, when immersed in uncontaminated groundwater, could be followed for in situ changes in contaminant concentrations over time using the same nondestructive analytical and recycled sampling protocol. The third objective was to determine if imposed soil contaminant treatments could be compared with untreated soil by following the contaminant retention over time using this same analytical/sampling approach. Our fourth objective was to establish that the polyacrylamide hydrogel matrix remained inert to contaminant chemical interactions and remained stable to chemical and microbiological degradation for several years so that changes in contaminant retention could to be followed for the necessary prolonged intervals to monitor and model natural and imposed contaminant attenuation. Our final objective was to determine if hydrogel encapsulation might also work generally to follow contaminant dynamics for a broad range of different soils and other relevant or environmentally useful solids, including limestone, hydrated Portland cement paste, activated charcoal, one specific uranium-adsorbing resin, and one uranium-coprecipitating chemical species (calcium phytate).

Materials and Methods Hydrogel-encapsulated soil samples were prepared from soil slurries [1:1.0-1.1; weight (g):volume (mL)] prepared from a common polyacrylamide electrophoresis gel-forming recipe (nominally 14% acrylamide, 0.3% N,N′-methylene-bis-acrylamide, 0.04-0.06% tetramethylethylenediamine (TEMED), and 0.002-0.06% ammonium persulfate [all wt./vol]) using disposable plastic syringes as inexpensive molds to form the final encapsulating homogeneous hydrogels. These hydrogel reagents and disposable syringe/molds are available from most laboratory chemical suppliers at a combined cost of less than $1.00/hydrogel specimen. We have termed these hydrogel-encapsulated soil samples polymer-encapsulated leaching capsules (PELCAPs) as a convenient acronym. Samples of uranium-contaminated soil were collected as entrained solids from groundwater monitoring wells within the U.S. Department of Energy, Oak Ridge Field Research Center (FRC). The FRC consists of an extensively monitored shallow groundwater plume contaminated with uranium, thorium, nitrate, and acidity (8) through past operations of the Oak Ridge Y-12 site’s S-3 ponds, which disposed of acidic radioactive wastewaters between 1950 and 1990. In 1990, the ponds, but not the contiguous residual groundwater plume, were neutralized, denitrified, and capped with a multilayer infiltration barrier. Initially, 2.5 cm diameter X-ray fluorescence (XRF) sample holders were employed as hydrogel molds and permanent sample holders in the field so that hydrogel standard materials and samples could be presented in identical geometry to the relatively inexpensive ($38000) Niton XLp700 portable analyzer (9) for elemental analyses (10). Considering this investigation entailed over 10000 nondestructive XRF analyses with minimal sample handling beyond the initial PELCAP preparation, the amortized unit analytical costs were quite low. All subsequent PELCAP preparations were performed using the smaller 1.5 cm diameter × 3 cm high right cylinders, prepared in and expressed from 10 mL capacity polypropylene disposable syringe molds. The small window of the Niton XRF analyzer performed identically during elemental analyses when exposed to the length of this small cylinder as when using the flat face of the larger 2.5 cm cell-contained geometric samples. Known additions of 10 routinely detect3048

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 8, 2010

FIGURE 1. Retention of uranium by pulverized Oak Ridge soil and limestone in PELCAPs during over two years of in situ leaching by immersion in minimally contaminated groundwater. For immersion intervals of 0 days, in minimally impacted groundwater about two miles down gradient at the same site. Inset depicts typical soil (left) and limestone (right) PELCAPs contained in 2.5 cm diameter XRF sample holders. able elements in various hydrogel geometries and at various moisture/hydrogel contents were prepared for calibration of instrument-reported element concentrations to known element concentrations. The smaller hydrogel cylinders were deployed without individual containers using stacked samples within 1.9 cm inner diameter polypropylene mesh tubes. These mesh tubes allowed suites of replicated hydrogel samples to be deployed within groundwater monitoring wells with g2.5 cm diameters. See the Supporting Information for detailed experimental methods on XRF calibrations, discussion of other XRF-detected elements, PELCAP preparation methods, site characteristics, sampling location maps, in situ immersion techniques, and laboratory leaching procedures.

Results and Discussion Initially, uncontaminated soil and pulverized limestone were used to prepare PELCAPs which were suspended for 73 days in groundwater within a 50 year old contaminated plume near the residual S-3 wastewater disposal ponds at the U.S. DOE Oak Ridge Y-12 site (8). This contaminated groundwater remains high in dissolved uranium (g65 mg/L), strongly acidic (pH e 3.1), and high in nitrate (g9000 ppm). The resulting maximum adsorbed uranium concentration in the immersed soil PELCAPs (about 500 mg uranium/kg, Figure, right axis) was similar to the ambient uranium concentrations in other soils collected as entrained solids during groundwater sampling from nearby monitoring wells. Maximum uranium concentrations in neat PELCAPs (without encapsulated solids) were similar to the groundwater concentrations likely because the neat hydrogels were about 85% water when prepared and swelled to >90% water after immersion. The resulting maximum uranium concentration in the pulverized limestone after adsorption was much higher (about 10000 mg/kg, Figure 1, left axis), which likely resulted from in situ neutralization of the acidic uranium-bearing groundwater within the encapsulated limestone. This accumulation of uranium on this pulverized limestone was similar to uranium concentrations found on limestone fragments sampled from a location further down gradient at this site (11). The solubility of uranium decreases markedly as pH increases from 3 to 7 (8), and the resulting high uranium concentration on this encapsulated limestone powder likely resulted from accumulation of precipitate as the acidic groundwater was neutralized by the limestone. In contrast, the encapsulated soil has little buffering capacity compared to limestone and

TABLE 1. Best-Fit Model Parametersa model parameter

model units

Oak Ridge soil (Figure 1)

pulverized limestone (Figure 1)

untreated soil (Figure 2)

thermally stabilized soil (Figure 2)

starting uranium concentration (C0) standard deviation (C0) slow-release fraction (N) standard deviation (N) slow kinetic constant (k1) standard deviation (k1) fast kinetic constant (k2) standard deviation (k2) model correlation standard error (Ct)

mg/kg mg/kg fraction fraction day-1 day-1 day-1 day-1 R2 mg/kg

499 90 0.12 0.13 1.0 × 10-6 1.2 × 10-3 4.6 × 10-3 8.4 × 10-4 0.989 19

9810 180 0.77 0.04 1.9 × 10-3 1.3 × 10-4 1.9 × 10-2 6.5 × 10-3 0.994 252

1192 50 0.50 0.03 7.0 × 10-4 1.8 × 10-4 0.080 0.017 0.944 55

1400 22 0.88 0.02 1.7 × 10-5 2.5 × 10-5 0.22 0.09 0.805 22

a Parameters were determined using a two-site kinetic model for uranium retention by PELCAPs containing soil or limestone during up to two years of in situ leaching in non-impacted or end-state groundwater at the DOE Field Research Center, Oak Ridge, Tennessee.

would not be expected to result in a similar pH-dependent decrease in uranium solubility. Although uranium soil reactions are a complex mixture of cation exchange, adsorption, precipitation/coprecipitation, and valence (VI/IV) change (8, 11), all interacting with soil properties (mineralogy, acidity, texture, surface area, hydrous oxide content, exchange capacity, etc.), the magnitude of uranium concentrations of hydrogel-encapsulated solids does not provide any direct evidence for the uranium chemical speciation during either uptake or subsequent release. However, such uptake and release kinetics for uranium to and from hydrogel-encapsulated soils can provide some insight into likely retention mechanisms, particularly the slower kinetic constraints on desorption and dissolution. Following uranium uptake from the contaminant plume, all PELCAPs were moved to a minimally impacted groundwater spring (pH about 7, uranium e0.01 mg/L, and nitrate e1 mg/L) approximately 2 miles down gradient but within the site’s shallow groundwater flow path. Detailed groundwater chemical analyses for this spring water, as well as those for the contaminated groundwater employed for uranium uptake, are presented in the Supporting Information. This groundwater has the characteristics of end-state or nonimpacted groundwater at this site. Using repeated XRF analyses of the PELCAPs at intervals over the course of more than two years during immersion in this minimally contaminated groundwater, the in situ retention of uranium by soil was followed (Figure 1). Considering the long interval of uranium leaching from these PELCAPs (Figure 1), it seems important to note initially the durability and persistence of the polyacrylamide hydrogel material itself over its two and one-half years of immersion in groundwater. Both neat and solid-containing PELCAPs maintained essentially constant hydrated fresh weights during this deployment interval. Neat PELCAPs, although faintly colored with iron oxides in their final condition, remained optically clear on visual inspection. Thus, the hydrogel appeared quite resistant to biodegradation or as a medium for microbial growth and thus useful to retain encapsulated solids for prolonged intervals in groundwater environments. The polyacrylamide hydrogel also behaved inertly to its uranium contamination. Uranium concentrations in neat PELCAPs rapidly approached uranium concentrations of the contaminated groundwater in which they were immersed. When these resulting acidic uraniumcontaminated neat PELCAPs were moved to the neutral groundwater spring to initiate uranium release, rapid in situ acid neutralization of their contained acidic water appeared to limit uranium solubility at the resulting near neutral pH of the spring; however, even this limited uranium precipitate underwent a rapid dissolution compared to PELCAPs con-

taining either soil or limestone (Figure 1). In laboratory tests, acidic uranium solutions, when spiked into PELCAPS either before and after acrylamide polymerization, were found to rapidly diffuse uranium into either ambient tap water or dilute nitric acid (Supporting Information). Thus, the polyacrylamide matrix appeared inert to uranium reactivity, although uranium transport from the hydrogel did exhibit a comparatively short diffusive delay. When ambient contaminated soils were encapsulated in polyacrylamide (as described later), their subsequent uranium leaching behavior also appeared to be free of polymer reactivity other than the short diffusive delay. Previously, we noted a similar lack of hydrogel polymer interactions when using soil spiked with 85 Sr and 134Cs radiotracers (5). The encapsulation of cation exchange resins into polyacrylamide hydrogels has also been reported to be free of polymerization artifacts affecting the cation exchange properties of the resins (4). Uranium retention kinetics by soil or other solids were modeled (lines in Figure 1) simply as the sum of two components of the total soil uranium: a fast-release fraction and another comparatively slow-release fraction, each following independent first-order exponential rate equations (12). The retained uranium concentration (Ct) in the soil after any number of days (t) of in situ leaching was modeled as the sum of these two fractions using eq 1 -k1t

Ct ) N × C0 × e

-k2t

+ (1 - N) × C0 × e

(1)

where N is the fraction of total uranium in slow-release form, C0 is the total starting uranium concentration (mg/kg), and k1 and k2 are the kinetic constants (day-1) for retention of the slow and rapid release forms, respectively. The parameters in the eq 1 (Table 1) of uranium retention by soil and by limestone in Figure 1 were optimally fit using an iterative solver (Excel) to minimize the difference between modeled and observed values of Ct over time for a biphasic kinetic model (12). The standard deviations of the four model coefficients (C0, N, k1, and k2) and the standard error of the modeled uranium concentration (Ct) for the four materials in Table 1 were calculated using an Excel macro (SolvStat.xls) for nonlinear regression analysis (12). This simple twofraction model was devised merely to summarize and compare uranium release behavior to groundwater among different soils or materials, by different soil treatments, or under differing groundwater conditions. By moving contaminated soil from regions of maximum contaminant plume characteristics to regions with minimally impacted or endstate groundwater characteristics, we observed quantitative behavior of the natural attenuation of uranium by site soil (Figure 1). This capability to move small hydrogel-encapsulated contaminated soil samples among various groundVOL. 44, NO. 8, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

3049

FIGURE 2. In situ retention of uranium by untreated (naturally attenuated) soil and by thermally stabilized soil in PELCAPs during two years of in situ leaching in non-impacted or end-state groundwater at the DOE Field Research Center, Oak Ridge, Tennessee. Inset is a typical uncontained soilencapsulating PELCAP cylinder of 1.5 cm diameter and 3 cm height.

FIGURE 3. Relation between modeled uranium concentrations in the slow-release fraction (N × C0, eq 1) and the concentrations of uranium remaining in the soil after the water and 0.1 N CaCl2 laboratory sequential extractions of seven uraniumcontaminated hydrogel-encapsulated soils. water locations is one of their most advantageous characteristics, allowing questions to be addressed concerning how a contaminant might behave under many alternate or changing in situ groundwater conditions. Other soil samples from the FRC, which had been exposed to ambient uranium contamination within the groundwater plume beginning in the 1950s when the source radioactive wastewater seepage ponds were constructed, behaved quite similarly when encapsulated and subjected to in situ leaching by immersion in minimally impacted groundwater (Figure 2, untreated soil; Table 1; and six other encapsulated soil samples from the FRC, see Supporting Information). The in situ retention behavior of uranium by natural attenuation in these several soils behaved similarly in the magnitudes of both their modeled fractions and kinetic constants. The magnitude of the modeled uranium fraction in the slow release form, for the seven soil samples carried through extensive long-term in situ leaching profiles, was wellcorrelated (r2 ) 0.89, Figure 3) with the fraction retained after laboratory sequential extractions with water and dilute calcium chloride (13). This correlation indicates that laboratory sequential extraction leached similar amounts of uranium from these soils, although the in situ leaching of 3050

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 8, 2010

PELCAPs has the advantage of being a more direct measure of soil uranium retention. Although the laboratory sequential leaching provided a good estimate of the size of the in situ long-term retained fraction, it did not provide any estimate of the in situ retention half-lives of the long-term fractions, which varied widely among soils (from 39 to >600000 days). In situ retention behavior of uranium by these soils did exhibit a significant contrast with the previously studied behavior of spiked 85Sr and 134Cs in hydrogels (5). Cesium retention by soil was essentially complete and irreversible, exhibiting no detectable leaching during six months of deployment. In contrast, strontium, which was retained largely by cation exchange mechanisms, was weakly retained in situ with >90% leaching within 90 days. As discussed above, a significant fraction of the uranium in the present soils was cation exchangeable like strontium in laboratory tests (Figure 3), although much of the residual uranium (the slow release fraction) appeared to remain slowly leachable but not as irreversibly retained as cesium. In Figure 2, the uranium retention behavior of one potential soil remedial alternative is contrasted with the natural attenuation behavior of the untreated soil. There are many potential remedial methods to improve the retention of uranium by soil, and thermal stabilization (12) was selected here solely to illustrate how hydrogel encapsulation can be used to quantitatively compare natural attenuation with alternative remedial approaches. Two significant changes in uranium availability were imposed on the soil by this simple thermal treatment (heating overnight at 1000 °C): a significant increase in the fraction of the total soil uranium in the slowrelease form (from 0.50 to 0.88 of the total) and a very large decrease in the kinetic constant (from 7 × 10-4 to 1.7 × 10-5 day-1) for that slow-release fraction (Table 1). Expressing these uranium kinetic release constants in terms of halflives, the slow-release constant was increased from 2.7 to 111 years by thermal stabilization. It is apparent from inspection of Figure 2 that an effective immobilization treatment, like thermal stabilization, should exhibit a large and comparatively unleachable contaminant fraction. The model fitting merely provides a precise and objective method to quantify and compare contaminant performance among various soils, treatments, contaminants, or under various aqueous test environments where hydrogels might be deployed. The utility of the hydrogel encapsulation technique for measuring the in situ retention of contaminant elements by soil, when coupled with a nondestructive analytical method like XRF, seems to have several general, useful, and practical extensions. First, hydrogels can be used to contain many other solids of interest besides contaminated soil. Uranium uptake by and release from such materials as Portland cement paste, activated charcoal, ion exchange resins, other site soils, and geologic materials is illustrated in the Supporting Information. Although the Oak Ridge FRC site soils contained few other XRF-detectable elemental contaminants (except thorium) above normal background concentrations (Supporting Information), other potential contaminants would include toxic metals and radionuclides in materials such as soil, fly ash, waste sludge, landfill solids, various nuclear waste forms, groundwater-reactive barrier material, minerals, precipitates, or microbial biomass, which do contain XRFdetectable concentrations of contaminating elements (e.g., Pb, As, Hg, Se, Cd, Cr, Cu, Ba, Zn, Mn, and Th). Second, the range of nondestructive or minimally destructive analytical techniques for encapsulated materials could be expanded to include X-ray absorption and diffraction, nuclear magnetic resonance, neutron scattering, neutron activation, or laserinduced breakdown spectroscopy to target other contaminants or to lower detection limits from those of XRF. Third, hydrogel-encapsulated materials could be deployed in diverse

aqueous environments where their in situ dissolution or metamorphosis needs to be understood, including rivers, lakes, oceans, waste lagoons, waste treatment facilities, unsaturated soils, and gastrointestinal tracts.

Acknowledgments This research was funded by the U.S. Department of Energy, Office of Science Biological and Environmental Research, Environmental Remediation Sciences Program (ERSP). Oak Ridge National Laboratory is managed by UT-Battelle, LLC, for the U.S. Department of Energy under Contract DE-AC0500OR22725.

Supporting Information Available Detailed materials and methods for PELCAP preparation, deployment, XRF analytical methods, calibration and internal standardization methods for other elements, PELCAP laboratory leaching, in situ leaching profiles of uranium from 14 additional soils and solid materials and their modeled parameters, uranium diffusion tests with neat polyacrylamide, comparative sequential extraction of 10 spiked elements from neat and soil-encapsulating PELCAPs, X-ray attenuation measurements and modeling by PELCAP matrices, and internal distribution of uranium within weathered PELCAPs. This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) MacDonald, J. A. Evaluating natural attenuation for groundwater cleanup. Environ. Sci. Technol. 2000, 34, 346A–353A. (2) National Research Council. Natural Attenuation for Groundwater Remediation; National Academy Press: Washington, DC, 2000.

(3) Brady, P. V.; Brady, M. V.; Borns, D. J. Natural Attenuation CERCLA, RBCA’s, and the Future of Environmental Remediation; Lewis Publishers: Boca Raton, FL, 1998. (4) Zhang, H.; Davison, W. Performance characteristics of diffusion gradients in thin films for the in situ measurement of trace metals in aqueous solution. Anal. Chem. 1995, 67, 3391–3400. (5) Spalding, B. P.; Brooks, S. C. Permeable environmental leaching capsules (PELCAPs) for in situ evaluation of contaminant immobilization in soil. Environ. Sci. Technol. 2005, 39, 8912– 8918. (6) Yang, T. Recent application of polyacrylamide as biomaterials. Recent Pat. Mater. Sci. 2008, 1, 29–40. (7) Burey, P.; Bhandari, B. R.; Howes, T.; Gidley, M. J. Hydrocolloid gel particles: Formation, characterization, and application. Crit. Rev. Food Sci. Nutr. 2008, 48, 361–377. (8) Watson, D. B.; Kostka, J. E.; Fields, M. W.; Jardine, P. M. The Oak Ridge Field Research Center Conceptual Model. 2004. http:// public.ornl.gov/nabirfrc/other/frcconmod.pdf. (9) Foley, G. J. Innovative Technology Verification Report. Field Portable X-Ray Fluorescence Analyzer. Niton XL Spectrum Analyzer; EPA/600-R-97/150; Environmental Protection Agency: Washington, DC, 1998. (10) U.S. Environmental Protection Agency. Method 6200 Field Portable X-ray Fluorescence Spectrometry for the Determination of Elemental Concentrations in Soil and Sediment. In Test Methods for Evaluating Solid Waste, Physical/Chemical Methods; SW-846; Environmental Protection Agecny: Washington, DC, 2007. (11) Phillips, D. H.; Watson, D. B.; Kelly, S. D.; Ravel, B.; Kemner, K. M. Deposition of uranium precipitates in dolomitic gravel fill. Environ. Sci. Technol. 2008, 42, 7104–7110. (12) Billo, E. J. Excel for Chemists: A Comprehensive Guide, 2nd ed; Wiley-VCH: New York, 2001. (13) Spalding, B. P. Fixation of radionuclides in soil and minerals by heating. Environ. Sci. Technol. 2001, 35, 4327–4333.

ES903983F

VOL. 44, NO. 8, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

3051