Assessment of a Sequential Extraction Procedure for Perturbed Lead

to the operational definitions of the extraction procedure. However, when the samples were amended with phosphate, results were drastically changed wi...
5 downloads 0 Views 128KB Size
Download PDF
Environ. Sci. Technol. 2003, 37, 1892-1898

Assessment of a Sequential Extraction Procedure for Perturbed Lead-Contaminated Samples with and without Phosphorus Amendments KIRK G. SCHECKEL,* CHRISTOPHER A. IMPELLITTERI, JAMES A. RYAN, AND THOMAS MCEVOY† United States Environmental Protection Agency, ORD, NRMRL, LRPCD, RCB, 5995 Center Hill Avenue, Cincinnati, Ohio 45224

Sequential extraction procedures are used to determine the solid-phase association in which elements of interest exist in soil and sediment matrixes. Foundational work by Tessier et al. (Tessier, A.; Campbell, P. G. C.; Bisson, M. Anal. Chem. 1979, 51, 844-851) has found widespread acceptance and has been employed as an operational definition for metal speciation in solid matrixes. However, a major obstacle confronting sequential extraction procedures is species alteration of extracted metals before, during, and after separation of solids from solution. If this occurs, the results obtained from sequential extraction do not provide an accurate account of metal speciation within the matrix because the metal forms are altered from their field state. Many researchers dismiss this drawback since several sorption and precipitation processes are believed to occur at time scales much longer than any particular extraction step. This assumption may not be valid. The objectives of this study were to investigate the potential formation of pyromorphite (Pb5(PO4)3Cl) during the sequential extraction steps of Pb-spiked samples with and without calcium phosphate amendments and to examine the differences in the operationally defined distribution of Pb in samples with and without the presence of P. The systems that were examined in the absence of phosphate behaved, for the most part, adequately according to the operational definitions of the extraction procedure. However, when the samples were amended with phosphate, results were drastically changed with a significant shift of extractable Pb to the residual phase. This redistribution was due to pyromorphite formation during the extraction procedure as confirmed by X-ray diffraction and X-ray absorption (XAS) spectroscopies. These results indicate that sequential extraction methods may not be suitable for Pb speciation in perturbed environmental systems (i.e., fertilized agricultural soils or amended contaminated soils) and that rigorous interpretation should be avoided, if not supported by methods to definitively prove metal speciation (e.g., XAS). * Corresponding author telephone: (513)487-2865; fax: (513)5697879; e-mail: [email protected]. † Student intern from Walnut Hills Learning Center, 813 Beecher, Cincinnati, OH 45206. 1892 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 9, 2003

Introduction Many sequential extraction schemes are modifications stemming from a procedure developed by Tessier et al. (1). These procedures were originally developed for the examination of aquatic sediments but gained wide acceptance as tools for the speciation of metals in soils. In the mid-1980s, an increasing number of researchers began to question the validity of sequential extraction procedures for metal speciation in solid samples. Among the chief concerns were sampling/preservation procedures (8), chemical properties of the target element and sample chemical composition (9), little specificity for the solid-phase attacked (10), lack of reference standards for quality control (11), and lack of standardized procedures (12-14). While sampling techniques and actual extraction routines can be standardized to some degree, results from sequential extraction procedures may vary widely in complex matrixes such as soils and sediments. Lack of specificity for the fractions attacked by a particular extractant can mobilize elements that are associated with solid phases other than the phase targeted by the extraction. Jouanneau et al. (10) compared two different orders of the same reaction scheme where one order employed a “carbonate” extraction step following the first (“exchangeable”) step and the other order used an “organic matter” extraction in the second step. They analyzed all extracts for Fe, Mn, organic carbon, and CaCO3. They found that the carbonate extraction removed a nonnegligible amount of Mn as did the organic matter extraction and concluded that the sequential extraction schemes were simply not selective enough especially with respect to manganese oxides. Wallman et al. (15) used thermodynamic equilibria data to show that sulfides of Cd, Co, Pb, and Zn are significantly soluble in the acetateexchangeable (steps 1 and 2) and oxalate-reducible (step 3) fractions of a sequential extraction series. They concluded that the dissolution of sulfide species must be accounted for in the interpretation of sequential extraction results. Readsorption of mobilized elements has been demonstrated in single extractions. Rendell et al. (16) examined several different extraction media commonly used for speciation of metals in sediments. They found that re-adsorption of spiked Cu, Pb, and Cd was significant in all of the extractions and concluded that metals in solution may not adequately represent the metals in the sediment fractions attacked. Qiang et al. (12) demonstrated that the presence of humic acids mobilized by an extraction phase targeting metals associated with carbonates significantly mobilized Cu and decreased sorption of Pb. Tipping et al. (17) showed that a hydroxylamine/HNO3 extraction solution resulted in Pb disassociating from manganese oxides, but the system pH allowed readsorption onto iron oxides in the same matrix. Re-adsorption problems are exacerbated by multiple extraction schemes where uncertainty due to re-adsorption is compounded by the lack of specificity of individual target fractions in a particular extraction scheme. Roger (18) studied two sequential extraction schemes (as well as different successions of extractions) and concluded that the sequences generally shifted the metals extracted to the residual phase as metals are successively mobilized and re-adsorbed by the extractions. This causes a tendency to underestimate more mobile forms of metal and to overestimate metals in the more immobile residual phase. Kheboian and Bauer (19) found a significant redistribution of Cu, Pb, and Zn between phases on sequential extractions on model sediments. The significance of element redistribution has been argued (20-23), and it may be that selective sequential extractions allow a more accurate portrayal of metals associated with solid 10.1021/es026160n Not subject to U.S. copyright. Publ. 2003 Am. Chem.Soc. Published on Web 03/25/2003

phases in some samples as compared with others because of sample composition (24). Soils are seldom, if ever, in equilibrium (25). Thus, the assumption of equilibrium is often a convenience to simplify an experimental design, which is typically the case for sequential extraction procedures that operate under the premise of equilibrium. Tessier et al. (1) presuppose that sequential extraction procedures will conceptually work under ideal equilibrium conditions. Immobilization of Pb as pyromorphite via soil amendments (phosphate with inorganic phases of biosolids and/or industrial iron byproducts) is a case-in-point of a nonequilibrated, perturbed environmental system that may lead to erroneous results when examined by sequential extraction procedures (26-37). As a result of such findings, some researchers have declared that the soils are safe (both environmentally and biologically) since pyromorphite is highly insoluble and biologically inert. However, few studies (4, 30, 38-40) have ever applied advanced spectroscopic methods to establish the presence of pyromorphite before the extraction procedure or if the extraction procedure itself causes a release of Pb and P into solution with subsequent formation of pyromorphite. We believe that literature studies reporting formation of pyromorphite in field and laboratory systems based on sequential extraction techniques may be in error as the formation may be a result of the sequential extraction chemicals releasing Pb and P into solution, thus creating favorable conditions for the precipitation of pyromorphite. Pyromorphite formed early in the extraction sequence would tend to pass through the extraction procedure to the residual fraction customarily attributed to insoluble phases such as pyromorphite and galena. As a result, a false assumption is implied that residual forms of Pb dominate in the soil prior to the sequential extraction procedure but, in fact, are probably formed during the extraction process for these perturbed systems. Ryan et al. (30) examined a sequential extraction procedure coupled with X-ray absorption fine structure (XAFS) spectroscopy of Pb-contaminated soil samples incubated with and without hydroxyapatite (HA) for aging periods of 0 and 240 days to investigate pyromorphite formation. They found that for both time periods HA amendments resulted in a substantial increase in the residual fraction of the sequential extraction procedure and confirmed that pyromorphite was the dominant Pb species in the residual fraction by XAFS. The presence of P and Pb in solution will result in the extremely rapid kinetic precipitation of pyromorphite (41, 42). Since pyromorphite formation is exceptionally quick, it is very conceivable that pyromorphite could form in solution if an environmental system that contained independent sources of soluble Pb and P was perturbed from its natural equilibrium state. The 0-day incubation with HA (where HA was added to the soil just prior to beginning the sequential extraction procedure) resulted in 49% of the Pb being accounted for in the residual fraction as compared to 11% without the HA amendment (control). The increase in the extraction residual fraction in the HA-amended 0-day incubated soil as compared to the 0-day non-HA-treated control soil is direct evidence that the chemical extraction procedure itself caused changes in extractability. The 240day incubation sample with HA resulted in 60% of the extracted Pb in the residual phase relative to 15% without HA. These results suggest that the chemical extraction procedure cannot easily be utilized to confirm changes occurring in amended soils; however, it may incorrectly imply that the transformation of soil-Pb to pyromorphite occurred during the incubation period. XAFS analysis confirmed that the resulting Pb species was similar to pure chloropyromorphite (30). Previous studies have interpreted the term re-sorption to include complexation, precipitation, coprecipitation, and

adsorption (16, 18). The work presented here refines the phenomenon of re-sorption by focusing on precipitation of Pb during the selective sequential extraction process. The objectives of the experiments were to investigate the potential formation of pyromorphite during sequential extraction steps and to examine the effects of the presence of P on the distribution of Pb in the operationally defined fractions. These experiments aim to illustrate that sequential extractions redistribute Pb from more soluble forms to residual phases (usually considered more recalcitrant forms of metal) by precipitation processes. If this hypothesis is correct, sequential extraction experiments on Pb-contaminated soils with significant sources of potentially soluble phosphorus (i.e., granular fertilizer, remediation amendments, runoff sources, etc.) will tend to overestimate the overall stability of the Pb in the system.

Experimental Section The experimental approach focused on the removal of Pb from quartz sand matrixes spiked with Pb minerals/chemicals [lead acetate (Pb(CH3CO2)2‚3H2O) highly soluble; anglesite (PbSO4) moderately soluble; cerussite (PbCO3) moderately soluble; galena (PbS) very low solubility; and pyromorphite (Pb5(PO4)3Cl) very low solubility] by a sequential extraction procedure based on that of Tessier et al. (1). In one set of experiments, the Pb-spiked solid matrixes contained calcium phosphate (CaHPO4‚2H2O) amendments while the other experiment did not. Sample Preparation. Samples were prepared from the PbSO4 > PbCO3 > PbS ≈ pyromorphite) followed trends based on known solubility data (42, 54, 55). Likewise, we anticipated the steps (fractions) of the sequential extraction procedure to variably induce solid-phase solubility as exchangeable (F1) < carbonate (F2) < iron/manganese oxide (F3) < organic matter (F4) < residual (F5). These assumptions were confirmed in Figure 1 and Table 1 of Supporting Information for the sequential extraction of the Pb-spiked sand samples without calcium phosphate to examine the fractionation of just the Pb minerals/compounds with recoveries from 91% to 103%. Since the quartz sand was absent of organic matter, iron and manganese oxides, and any other potential sorbent phase aside from SiO2, it was not expected to observe Pb in the extraction fractions associated with iron/manganese oxides and organic matter although the chemistry of these extraction steps may result in some Pb being solubilized. While there was no evidence of Pb associated with the organic matter fraction in any of the experiments, the iron/manganese oxide fraction step had some Pb extracted primarily from the pyromorphite and galena samples and could be as a result of either oxidative dissolution of galena or desorption of Pb from exterior crystal metal sites from reaction with the extraction solution. In terms of the pure compounds individually, lead acetate and anglesite were predominantly extracted in the exchangeable phase, and cerrusite was primarily removed in the carbonate step. The majority of pyromorphite and galena was removed in the residual extraction step, as expected, but 33-40% of the extracted Pb is distributed among the exchangeable, carbonate, and iron/manganese oxide fractions indicating that relatively stable minerals that would normally be associated with residual phases can undergo desorption/ dissolution in the milder extraction solutions. The fact that Pb was removed during the iron/manganese oxide step is a

FIGURE 1. Sequential extraction results of 2000 ppm Pb-spiked samples. good indication that extraction procedures are not as well defined as some may hope since iron/manganese oxides were not present in our samples. The Pb-spiked samples without calcium phosphate in one-on-one combinations (Figure 1) show that they essentially behaved additively based on each individual component. For example, the lead acetateanglesite sample (containing equal amounts of both components that were individually extracted in the exchangeable fraction) was 94% extracted in the exchangeable phase. The anglesite-galena combination also showed scaled attributes from each mineral as examined separately. These trends were observable throughout the data shown in Figure 1. The mixture of all the Pb minerals/compounds exhibited amounts of extractable Pb that correlated well to the sum of the individual extraction experiments. In short, the extraction sequence performed suitably for Pb in these laboratorycreated samples. Figure 2 and Table 2 of Supporting Information show the results of the sequential extraction procedure for the Pbspiked samples with the calcium phosphate amendment. Total Pb recovery ranged from 90% to 103%. These results are drastically different than those present in Figure 1 for the same samples sequentially extracted without phosphate. First, the presence of phosphate totally eliminates any indication of an exchangeable phase. Furthermore, Pb observed in the residual phase increased dramatically. This was the result of Pb extracted during the exchangeable step reacting with P from the calcium phosphate amendment to form pyromorphite (Figures3 and 4). Figure 3 shows the k3-weighted χ-functions of lead acetate plus calcium phosphate samples before extraction and after steps 1 (MgCl2 extraction solid) and 4 (residual fraction) of the sequential extraction procedure used in this study. The data are compared to reference spectra of lead acetate and chloropyromorphite. The dry sample prior to subjection to the sequential extraction procedure illustrates that Pb was

present as lead acetate relative to the standard. However, upon extraction of the dry sample containing lead acetate and calcium phosphate with MgCl2 during the first extraction step, the solid-phase speciation of Pb changed significantly to mimic the curve of the chloropyromorphite standard. This information alone substantially supports our hypothesis that application of sequential extraction procedures to perturbed Pb-contaminated systems may not accurately predict Pb speciation based on operational definitions with respect to chemical extractants. The results for the residual fraction in Figure 3 show the formation of pyromorphite for the lead acetate and calcium phosphate sample. These trends were observed wholly throughout the sample set (data not shown). XRD data of lead acetate only and lead acetate plus calcium phosphate samples before extraction and after steps 1 (MgCl2 extraction solid) and 4 (residual fraction) of the sequential extraction procedure are presented in Figure 4. The solid vertical lines in Figure 4 represent key characteristic peaks of chloropyromorphite, and the dashed lines are indicative of brushite (CaHPO4). Beginning with the samples that did not contain phosphate (top two curves of Figure 4), one sees the distinctive XRD pattern of amorphous silica, thus illustrating that noncrystalline lead acetate cannot be discerned and that pyromorphite does not form after the MgCl2 extraction step or any extraction step (data not shown). However, data for the lead acetate plus calcium phosphate samples yield different results. The lead acetate plus CaHPO4 sample nonreacted (dry) within the extraction method shows XRD peaks not present in the lead acetate only sample and are identified as brushite. Upon reaction with MgCl2 in the first sequential extraction step, the peaks for brushite disappear and new XRD peaks emerge that align well with the chloropyromorphite markers (solid lines). Further extraction of the lead acetate plus CaHPO4 sample to the residual fraction also demonstrated that pyromorphite was still present, and it appeared that the peaks were becoming VOL. 37, NO. 9, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

1895

FIGURE 2. Sequential extraction results of 2000 ppm Pb-spiked samples amended with 1000 ppm of calcium phosphate.

FIGURE 3. Normalized Pb(LIII)-EXAFS k3-weighted χ functions of lead acetate sequential extraction and reference (chloropyromorphite and lead acetate) samples. more defined, perhaps as a result of aging of the pyromorphite structure (56). As noted with the XAFS experiments, similar results were observed for the other samples (data not shown). The other combinations exhibited in Figure 2 (carbonatecerrusite, iron/manganese oxide-pyromorphite, and residual fractions-galena, pyromorphite) show an increased concentration of Pb in the residual phase with the P amendment than the samples without P (Figure 1). With P present, acetate and anglesite also had Pb concentrations increase in the carbonate fraction, cerrusite showed a decrease in Pb extracted in the carbonate step, and Pb in the carbonate fraction remained roughly the same for galena and pyromorphite. For the iron/manganese oxide fraction, Pb increased drastically for the acetate, anglesite, and cerrusite samples and slightly decreased in the galena and pyromor1896

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 9, 2003

FIGURE 4. XRD patterns of lead acetate sequential extraction samples with chloropyromorphite (solid lines) and calcium phosphate (dashed lines) identification markers. phite samples. Every sample observed an increase in residual Pb. It should be noted that increases in residual-phase Pb occurred for pyromorphite (9% increase) and galena (14% increase) with the P amendment. This shift toward residual Pb was due to the formation of pyromorphite (Figures 3 and 4). The overall distributions of Pb in the samples in Figure 2 illustrate a transformation of the compounds or mix of compounds at the beginning of the extraction into more stable, recalcitrant forms. The results in Figure 2 suggest that the presence of P in the extractions leads to changes in the form of the Pb into a more homogeneous species regardless of the parent material. While sorption and

coprecipitation reactions cannot be entirely eliminated from consideration, we confidently attribute the shift of Pb to the residual phases to the formation of pyromorphite (Figures 3 and 4). The aim of this research is to alert users of sequential extraction procedures to be aware of precipitation processes as well as re-sorption phenomena. The research presented here shows quite clearly that the results of a sequential extraction on identical materials is severely altered by the presence of P (as either a fertilizer input or soil amendment). Phosphorus chemistry is quite complex, and P is ubiquitous in soils as well as sediments. Application of sequential extraction procedures for Pb-, Cd-, Zn-, Co-, Cr-, and Nicontaminated soil and sediment samples, especially for those with large sources of P or oxalate, should be prudent since these metals have been shown to rapidly form solid metalphosphates and metal-oxalate complexes (2, 4, 26, 30, 5760). Extreme caution should be employed in interpreting data from the sequential extraction of Pb from any soil or sediment. Ideally, one would support and verify sequential extraction data with other speciation techniques (e.g., molecular-level spectroscopic techniques such as X-ray absorption fine structure and X-ray diffraction spectroscopies) (2, 4, 6, 30, 61). This would allow the researcher to examine the sample in situ to identify metal species without physicochemical alteration of neither the target analyte nor constituents, such as P, that will react with the target analyte.

Acknowledgments The U.S. EPA has not subjected this manuscript to internal policy review. Therefore, the research results presented herein do not, necessarily, reflect Agency policy. Mention of trade names of commercial products does not constitute endorsement or recommendation for use. T.M. is grateful for the opportunity to participate in the EPA-UC High School Apprenticeship Program. We appreciate the excellent peer reviews that improved the quality of this manuscript.

Supporting Information Available Tables 1 and 2 listing the numerical values and percent recoveries of data presented in Figures 1 and 2, respectively. This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) Tessier, A.; Campbell, P. G. C.; Bisson, M. Anal. Chem. 1979, 51, 844-851. (2) Scheinost, A. C.; Kretzschmar, R.; Pfister, S. Environ. Sci. Technol. 2002, 36, 5021-5028. (3) Bunzl, K.; Trautmannsheimer, M.; Schramel, P. J. Environ. Qual. 1999, 28, 1168-1173. (4) Calmano, W.; Mangold, S.; Welter, E. Fresenius J. Anal. Chem. 2001, 371, 823-830. (5) Gruebel, K. A.; Davis, J. A.; Leckie, J. O. Soil Sci. Soc. Am. J. 1988, 52, 390-397. (6) Morin, G.; Ostergren, J. D.; Juillot, F.; Ildefonse, P.; Calas, G.; Brown, G. E., Jr. Am. Miner. 1999, 84, 420-434. (7) Roberts, D. R.; Scheinost, A. C.; Sparks, D. L. Envion. Sci. Technol. 2002, 36, 1742-1750. (8) Rapin, F.; Tessier, A.; Campbell, P. G. C.; Carignan, R. Environ. Sci. Technol. 1986, 20, 836-840. (9) Martin, J. M.; Nirel, P.; Thomas, A. J. Mar. Chem. 1987, 22, 313341. (10) Jouanneau, J. M.; Latouche, C.; Pautrizel, F. Environ. Technol. Lett. 1983, 4, 509-514. (11) Fiedler, H. D.; Lupez-Sanchez, J. F.; Rubio, R.; Rauet, G.; Quevauviller, P. H.; Ure, A. M.; Muntau, H. Analyst 1994, 119, 1109-1114. (12) Qiang, T.; Xiao-Quan, S.; Jin, Q.; Zhe-Ming, N. Anal. Chem. 1994, 66, 3562-3568. (13) Quevauviller, P.; Rauret, G.; Muntau, H.; Ure, A. M.; Rubio, R.; Lopez-Sanchez, J. F.; Fiedler, H. D.; Griepink, B. Fresenius J. Anal. Chem. 1994, 349, 808-814.

(14) Usero, J.; Gamero, M.; Morillo, J.; Gracia, I. Environ. Int. 1998, 24, 487-496. (15) Wallmann, K.; Kersten, M.; Gruber, J.; Forstner, U. Int. J. Environ. Anal. Chem. 1993, 51, 187-200. (16) Rendell, R.; Batley, G. E.; Cameron, A. J. Environ. Sci. Technol. 1980, 14, 314-318. (17) Tipping, E.; Hetherington, N. B.; Hilton, J.; Thompson, D. W.; Bowles, E.; Hamilton-Taylor, J. Anal. Chem. 1985, 57, 19441946. (18) Roger, B. Environ. Technol. Lett. 1986, 7, 539-546. (19) Kheboian, C.; Bauer, C. F. Anal. Chem. 1987, 59, 1417-1423. (20) Belzile, N.; Lecomte, P.; Tessier, A. Environ. Sci. Technol. 1989, 23, 1015-1020. (21) Kim, N. D.; Fergusson, J. E. Sci. Total Environ. 1991, 105, 191209. (22) Bauer, C. F.; Kheboian, C. Anal. Chem. 1988, 60, 1477. (23) Tessier, A.; Campbell, P. G. C. Anal. Chem. 1988, 60, 14751476. (24) Nirel, P. M. V.; Morel, F. M. M. Water Res. 1990, 24, 1055-1056. (25) Sparks, D. L.; Scheidegger, A. M.; Strawn, D. G.; Scheckel, K. G. In Mineral-Water Interfacial Reactions; Sparks, D. L., Grundl, T. J., Eds.; ACS Symposium Series 715; American Chemical Society: Washington, DC, 1999; pp 108-135. (26) Seaman, J. C.; Arey, J. S.; Bertsch, P. M. J. Environ. Qual. 2001, 30, 460-469. (27) Maskall, J. E.; Thornton, I. Water Air Soil Pollut. 1998, 108, 391409. (28) Kabala, C.; Singh, R. R. J. Environ. Qual. 2001, 30, 485-492. (29) Barona, A.; Aranguiz, I.; Elias, A. Environ. Pollut. 2001, 113, 79-85. (30) Ryan, J. A.; Zhang, P. C.; Hesterberg, D.; Chou, J.; Sayers, D. E. Environ. Sci. Technol. 2001, 35, 3798-3803. (31) Basta, N. T.; Gradwohl, R.; Snethen, K. L.; Schroder, J. L. J. Environ. Qual. 2001, 30, 1222-1230. (32) Lima, M. C.; Giacomelli, M. B. O.; Stupp, V.; Roberge, F. D.; Barrera, P. B. Quim. Nova 2001, 24, 734-742. (33) Ma, Y. B.; Uren, N. C. Geoderma 1998, 84, 157-168. (34) Ma, L. Q.; Rao, G. N. J. Environ. Qual. 1997, 26, 788-794. (35) Cao, X.; Ma, L. Q.; Chen, M.; Singh, S. P.; Harris, W. G. Environ. Sci. Technol. 2002, 36, 5296-5304. (36) Cao, R. X.; Ma, L. Q.; Chen, M.; Singh, S. P.; Harris, W. G. Environ. Pollut. 2003, 122, 19-28. (37) Melamed, R.; Cao, X.; Chen, M.; Ma, L. Q. Sci. Total Environ. (in press). (38) Hillier, S.; Suzuki, K.; Cotter-Howells, J. Appl. Geochem. 2001, 16, 597-608. (39) Ostergren, J. D.; Brown, G. E.; Parks, G. A.; Tingle, T. N. Environ. Sci. Technol. 1999, 33, 1627-1636. (40) Song, Y.; Wilson, M. J.; Moon, H. S.; Bacon, J. R.; Bain, D. C. Appl. Geochem. 1999, 14, 621-633. (41) Nriagu, J. O. Geochim. Cosmochim. Acta 1974, 38, 887-898. (42) Scheckel, K. G.; Ryan, J. A. Environ. Sci. Technol. 2002, 36, 21982204. (43) Zhang, P. C.; Ryan, J. A. Environ. Sci. Technol. 1998, 32, 33183324. (44) Zhang, P. C.; Ryan, J. A. Environ. Sci. Technol. 1999, 33, 618624. (45) Zhang, P. C.; Ryan, J. A. Environ. Sci. Technol. 1999, 33, 625630. (46) Zhang, P. C.; Ryan, J. A.; Bryndzia, L. T. Environ. Sci. Technol. 1997, 31, 2673-2678. (47) Strawn, D. G.; Sparks, D. L. Soil Sci. Soc. Am. J. 2000, 64, 144156. (48) Ponizovskii, A. A.; Mironenko, E. V. Eurasian Soil Sci. 2001, 34, 371-381. (49) Pearson, M. S.; Maenpaa, K.; Pierzynski, G. M.; Lydy, M. J. J. Environ. Qual. 2000, 29, 1611-1617. (50) Raposo, J. C.; Sanz, J.; Zuloaga, O.; Olazabal, M. A.; Madariaga, J. M. Talanta 2002, 57, 849-857. (51) Jackson, M. L. Soil Chemical AnalysissAdvanced Course, Department of Soils, University of Wisconsin, 1956. (52) Anderson, J. U. Clays Clay Miner. 1963, 10, 380-388. (53) Lavkulich, L. M.; Wiens, J. H. Soil Sci. Soc. Am. Proc. 1970, 34, 755-758. (54) Brown, T. L.; LeMay, J. H. E.; Bursten, B. E. Chemistry, The Central Science, 6th ed.; Prentice Hall: Englewood Cliffs, NJ, 1994. (55) Nriagu, J. O. Geochim. Cosmochim. Acta 1973, 37, 367-377. (56) Scheckel, K. G.; Ryan, J. A. Abstr. Pap. Am. Chem. Soc. 2001, 222, 39-GEOC. (57) Valsami-Jones, E.; Ragnarsdottir, K. V.; Putnis, A.; Bosbach, D.; Kemp, A. J.; Cressey, G. Chem. Geol. 1998, 151, 215-233. VOL. 37, NO. 9, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

1897

(58) Sarret, G.; Vangronsveld, J.; Manceau, A.; Musso, M.; D’Haen, J.; Menthonnex, J. J.; Hazemann, J. L. Environ. Sci. Technol. 2001, 35, 2854-2859. (59) Anderson, P. R.; Christensen, T. H. J. Soil Sci. 1988, 39, 15-22. (60) Mielke, H. W.; Gonzales, C. R.; Smith, M. K.; Mielke, P. W. Sci. Total Environ. 2000, 246, 249-259.

1898

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 9, 2003

(61) La Force, M. J.; Fendorf, S. Soil Sci. Soc. Am. J. 2000, 64, 16081615.

Received for review September 16, 2002. Revised manuscript received January 30, 2003. Accepted February 14, 2003. ES026160N