Environ. Sci. Technol. 2002, 36, 3749-3754
Slow Formation and Dissolution of Zn Precipitates in Soil: A Combined Column-Transport and XAFS Study ANDREAS VOEGELIN, ANDREAS C. SCHEINOST, KARIN BU ¨ HLMANN, KURT BARMETTLER, AND RUBEN KRETZSCHMAR* Institute of Terrestrial Ecology, Department of Environmental Sciences, Swiss Federal Institute of Technology, Grabenstrasse 3, CH-8952 Schlieren, Switzerland
Recent spectroscopic studies have demonstrated the formation of layered double hydroxides (LDH) and phyllosilicates upon sorption of Zn2+, Ni2+, and Co2+ to clay minerals and aluminum oxides at neutral to alkaline pH and at relatively high initial metal concentrations (>1 mM). The intention of the present study was to investigate whether such phases also form in soil under slightly acidic conditions and at lower metal concentrations. Columns packed with a loamy soil were percolated with aqueous solutions containing 0.1 or 0.2 mM Zn, Ni, Co, and Cd in a 10 mM CaCl2 background at pH 6.5. Metal breakthrough curves indicated a rapid initial sorption step, resulting in retarded breakthrough fronts, followed by further slow metal retention during the entire loading period of 42 days (7000 pore volumes). Total metal sorption and the contribution of slow sorption processes decreased in the order Zn > Ni > Co > Cd. Leaching the reacted soil with 10 mM CaCl2 at pH 6.5 remobilized 8% of the total retained Zn, 15% of Ni, 21% of Co, and 77% of Cd. Subsequent leaching with acidified influent (pH 3.0) remobilized most of the remaining metals. X-ray absorption fine-structure (XAFS) spectroscopy revealed that slow Zn sorption was due to the formation of a Zn-Al LDH precipitate. Although Ni, Co, and Cd concentrations were too low for XAFS analysis, their leaching patterns suggest that part of Ni and Co were also incorporated in solid phases, while most sorbed Cd was still present as exchangeable sorption complex after 42 days. A small but significant percentage of the sorbed metals (2-5%) remained in the soil, even after leaching with more than 3000 pore volumes at pH 3.0, which may suggest micropore diffusion or incorporation into more stable mineral phases.
Introduction The fate of metal contaminants in soils and their potential effects on environmental quality largely depend on sorption reactions at solid-water interfaces of soil minerals and organic matter. Numerous studies with soils and clay minerals have demonstrated that initial rapid adsorption reactions are often followed by further slow sorption processes, resulting in a slow decrease in metal solubility with time (1, * Corresponding author phone: +41-1-6336003; fax: +41-16331118; e-mail:
[email protected]. 10.1021/es010316m CCC: $22.00 Published on Web 08/01/2002
2002 American Chemical Society
2). Such slow sorption processes have been ascribed to several different mechanisms, including micropore or surface diffusion (1, 3-5) and slow formation of surface precipitates, sometimes coupled to mineral weathering reactions (6-10). In acidic soils (pH < 5), fast and reversible cation exchange reactions dominate the sorption of Zn2+, Ni2+, Co2+, and Cd2+ (11). Therefore, reactive transport of these metal cations through uniformly packed soil columns can be modeled quantitatively by cation exchange reactions coupled to a transport model assuming local chemical equilibrium (12). However, slow sorption reactions tend to become more prominent with increasing pH, especially for Zn, Ni, and Co. Development of models for risk assessment and metal transport in soils therefore critically depends on our understanding of the responsible slow sorption mechanisms, including their kinetics and reversibility. Recent studies using X-ray absorption fine structure (XAFS) spectroscopy demonstrated the slow formation of mixed Ni-Al layered double hydroxide (LDH) precipitates upon reaction of Ni with pyrophyllite, kaolinite, gibbsite, and montmorillonite clay (6, 10, 13, 14). Formation of LDH-type precipitates were also observed for Co reacted with kaolinite (9) and Zn reacted with pyrophyllite (15), respectively. In all studies, LDH precipitates were formed at metal concentrations below the solubility products of the respective metal hydroxide phases. However, the thermodynamic properties of the LDH phases are difficult to determine and are therefore not wellestablished. There is also evidence suggesting that Ni-Al LDH precipitates may be slowly transformed into more stable mineral phases, for example, by silica exchange in the interlayers and subsequent aging to Ni-Al phyllosilicates (7). Alternatively, heavy metal-containing phyllosilicates may also be formed directly by nucleation and epitaxial growth on edges of clay minerals, as was shown for Co sorption on hectorite (16). A recent study on Zn speciation in smeltercontaminated soils has confirmed the existence of a Zn phyllosilicate under natural conditions (17). Although CdAl LDH can be synthesized (18), its formation in soils seems less likely because of the significant larger ionic radius of Cd compared to Zn, Ni, and Co. Until now, most studies on the formation of surface precipitates of Zn, Ni, and Co on pure clay minerals or soil clay fractions have been conducted at fairly high metal concentrations (1-3 mM), high pH values (pH ∼ 7.5), and high background electrolyte concentration (∼0.1 M) (6, 8, 10, 13, 15). Roberts et al. also examined Ni sorption to a soil clay fraction at pH 6.0 and 6.8 (8). At pH 6.8, Ni-Al LDH was formed within a few hours, while at pH 6.0 no precipitates were detected even after 72 h. For other metals, pH dependent studies are lacking; therefore, the possible formation of LDH phases under slightly acidic conditions is uncertain. Even in contaminated soils, the solution concentrations of Zn, Ni, Co, and Cd are often far below the concentrations used in most previous laboratory experiments. In addition, the electrolyte concentration of the soil solution is often well below 0.1 M, allowing for a larger contribution of cation exchange to metal sorption. On the other hand, the speciation of metals in soils by XAFS spectroscopy requires sufficiently high sorbed concentrations, which depend also on beamline specific parameters. Using a combination of flow-through column experiments and XAFS spectroscopy may offer a promising new approach for studying the formation of metal precipitates in soil at lower metal concentrations in solution. First, the soil can be reacted in a flow-through column over long time periods with a solution of low metal concentration to achieve a surface loading high enough to be analyzed by VOL. 36, NO. 17, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3749
XAFS spectroscopy. Second, the metal-loaded soil column can be used to assess the remobilization of metals in response to soil chemical changes, such as an increase in electrolyte concentration or a decrease in pH. Therefore, the specific objectives of this study were (1) to investigate the rapid and slow sorption of Zn, Ni, Co, and Cd in a loamy soil at slightly acidic pH and low metal concentrations using a flow-through column, (2) to examine the subsequent release of metals due to cation exchange and soil acidification, and (3) to assess the predominant sorption mechanisms of Zn in the soil based on a combination of column breakthrough curves and XAFS spectroscopy.
Experimental Section Soil. The soil used in this study was collected from the B-horizon (15-25 cm sampling depth) of an acidic forest soil in northern Switzerland (Riedhof soil, aquic dystric Eutrochrept). Several kilograms of soil were collected, airdried, and then passed through a 2-mm screen. To obtain a uniform column packing for transport experiments, an aggregate size fraction between 63 and 400 µm was carefully separated by dry sieving. The composition of this fraction was comparable to the unfractionated soil, with 9 g/kg organic carbon, mixed clay mineralogy (vermiculite, illite, kaolinite, chlorite), and a loamy texture (0.37 g/g sand, 0.47 g/g silt, 0.16 g/g clay). The same soil has been used in a previous study on Cd, Zn, and Ni sorption and reactive transport modeling (12). Column Transport Experiments. Chromatographic glass columns (Omni) with an inner diameter of 1 cm were uniformly packed with 8.0 g of dry soil. The resulting soil columns were 9 cm long, with a pore volume of 4.0 mL and a packing density of 2.0 g of soil/mL of pore volume (i.e., bulk density ∼ 1.13 g/cm3). To ensure complete water saturation, the dry columns were first purged with a stream of CO2 gas to displace the enclosed air. The bottom end of the columns were then connected to an HPLC pump delivering a degassed 0.5 M CaCl2 solution at a rate of 0.5 mL/min. Because of dissolution of entrapped CO2 gas in water and the upward flow direction, complete water saturation of the pore space was achieved within a few pore volumes. In the first set of column experiments, the following sixstep percolation scheme was applied. (Step 1) Preconditioning of the columns by several hundred pore volumes of 0.5 M CaCl2 at a flow rate of 0.5 mL/min. (Step 2) Equilibration with 200 pore volumes of 10-2 M CaCl2 adjusted to pH 6.5 (with NaOH). (Step 3) Metal loading with a solution containing 0.94 × 10-4 M ZnCl2, 1.00 × 10-4 M NiCl2, 0.95 × 10-4 M CoCl2, and 0.97 × 10-4 M CdCl2 in a background electrolyte of 10-2 M CaCl2 at pH 6.5. Each column was loaded with 7350 pore volumes, which took about 41 days. (Step 4) Leaching with 67 pore volumes of 10-5 M CaCl2 at pH 6.5 to replace the metal solution. (Step 5) Leaching with 133 pore volumes of 10-2 M CaCl2 at pH 6.5 to remove exchangeable cations. (Step 6) Leaching with 3100 pore volumes of 10-2 M CaCl2 adjusted to pH 3.0 (with HCl) to release acid-extractable metals. Column effluent was sampled using a fraction collector, and Zn, Ni, Co, and Cd in the effluent were measured with flame atomic absorption spectrometry (Varian SpectrAA 220). Columns were sampled after steps 4, 5, and 6, respectively, and the soil was dried at 60 °C. Total metal contents of untreated soil and of the samples taken after steps 4, 5, and 6 were measured using X-ray fluorescence spectroscopy (XRF, Spectro X-Lab 2000). The uncontaminated soil contained 60 ppm Zn (9.1 × 10-4 mol/kg), 30 ppm Ni (5.1 × 10-4 mol/kg), Cd. Leaching with Ca desorbed part of the rapidly sorbed fractions. Leaching at pH 3.0 mobilized most of the remaining metal, but 5% Zn, 3% Ni, 4% Co, and 2% Cd (relative to total sorbed) were not remobilized after 3000 pore volumes of leaching at pH 3.0. effluent concentrations of Cd, Co, and Ni were close to unity after 100-200 pore volumes. The importance of slow metal sorption decreased in the order Zn > Ni > Co > Cd. This sequence coincides with that of the solubility products of the metal hydroxides (lowest log K for M-hydroxide ) M2+ + 2OH-, from ref 19): Zn (-16.7) < Ni (-15.2) < Co (-14.9) < Cd (-14.4). Note, however, that the heavy metal concentration levels in our study were at least 2 orders of magnitude below the solubility limits of the respective metal hydroxides. The extent of rapid sorption, on the other hand, roughly relates to the hydrolysis constants of the respective metal cations, insofar as Zn was most strongly retained and also hydrolyzes more readily (9.0, pK for M2+(aq) + H2O ) MOH+ + H+, from ref 19) than Co (9.7), Ni (9.9), and Cd (10.1). This is in agreement with studies in which adsorption affinities of metals to iron oxides or soil materials were related to the hydrolysis of the respective cations (24, 25). In summary, the congruency between rapid sorption and metal hydrolysis constants and between slow sorption and metal hydroxide formation constants might suggest that rapid sorption was dominated by adsorption processes, whereas slow sorption was primarily due to some type of precipitation process. Metal Release. Leaching of the soil columns with CaCl2 solutions at pH 6.5 followed by leaching at pH 3.0 (steps 4-6) resulted in characteristic metal release patterns (Figure 3). The presented effluent metal concentrations are normalized to the respective influent concentrations during metal loading (10-4 M). During leaching with very dilute CaCl2 solution (step 4, 10-5 M CaCl2, pH 6.5), virtually no heavy metals were removed from the soil column. However, increasing the Ca concentration to 10-2 M (step 5) resulted in an immediate release of Zn, Ni, Co, and Cd. The initial peak concentrations were close to the influent concentrations during loading (i.e., c/c0 ∼ 1), suggesting that an equilibrium between the influent and rapidly adsorbed exchangeable cations in the soil had been attained during metal loading. The tailing after the initial elution peak is most pronounced for Zn, in agreement with its higher adsorption affinity as compared to Ni, Co, and Cd. Decreasing the influent pH to 3.0 (step 6) induced further metal release from the soil column. The peak metal concentrations decreased in the sequence Zn > Ni > Co > Cd, thus following the extent of slow metal sorption during the loading step. The short delay period of ∼20 pore volumes between switching to pH 3.0 influent and onset of metal release likely results from metal readsorption within the soil column. The effluent pH dropped VOL. 36, NO. 17, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3751
FIGURE 4. Spatial distribution of Zn, Ni, Co, and Cd after loading at c0 ∼ 0.2 mM. As in the first set of experiments, the extent of metal retention decreases in the order Zn > Ni > Co > Cd. Within the column, metal retention decreased with increasing distance to the inlet, paralleling the decrease in pH from the column influent (pH 6.5) to the effluent (pH 5.4-5.9).
FIGURE 3. Effluent metal concentrations during leaching with 10-5 M CaCl2, pH 6.5 (step 4; 0-67 pore volumes), 10-2 M CaCl2, pH 6.5 (step 5; 67-200 pore volumes), and 10-2 M CaCl2, pH 3.0 (step 6; 200-3150 pore volumes). Concentrations are normalized to influent metal concentrations during loading (c0 ) 10-4 M). Desorption by Ca2+ is comparable for all metals, whereas peak concentrations induced at pH 3.0 reflect the extent of slow metal sorption: Zn > Ni > Co > Cd. to 3.0 only after the major portion of all four metals was eluted from the soil material. The amounts of metals released by CaCl2 leaching at pH 6.5 (step 5) and pH 3 (step 6) were determined by numerical integration of the breakthrough curves. The total amounts sorbed during the loading step (step 3) was determined as the sum of released metals and residual metal contents in the soil column at the end of step 6. Figure 2 compares the amounts of metals sorbed rapidly and slowly during the loading step with the amounts of metals released by Ca exchange at pH 6.5 and by acidification to pH 3. For all metals, the rapidly sorbed amounts were readily mobilized by Ca exchange, indicating that cation exchange was the dominant sorption mechanism. In contrast, the fractions of metals sorbed by slow processes were not exchangeable with Ca but were released within ∼600 pore volumes by acidification to pH 3. This suggests that either formation of strong innersphere sorption complexes or a solid precipitate were responsible for the slow sorption process. Of the total sorbed amounts, approximately 5% of Zn, 3% of Ni, 4% of Co, and 2% of Cd remained in the soil, even after leaching with 3000 pore volumes of acidified solution (see Supporting Information), possibly indicating an incorporation into more stable minerals. Metal Distribution in Soil Columns. In a second set of experiments, slightly higher metal concentrations of ∼0.2 mM were applied for metal loading. Breakthrough and release patterns were similar to those observed at the lower loading concentrations of ∼0.1 mM (Figures 1 and 3, see Supporting Information for sorbed and released amounts). To determine the metal distribution in the column after loading (step 3), the column was cut into three sections. Metal concentrations in the soil were highest in the inlet region and decreased toward the outlet of the column (Figure 4). The concentration gradients of the four metals followed the extent of slow sorption (Figure 2), meaning that they most likely resulted from the slow sorption reaction. The contribution of rapid adsorption to metal retention in the three sections can be estimated from the amounts leached with CaCl2, which are again similar for all metals. While the columns were fed with 3752
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 17, 2002
solutions adjusted to pH 6.5, effluent solution pH was always slightly lower, at pH ∼5.8 (Figure 1), probably due to proton release during sorption. Considering that metal concentrations in solution during slow sorption were virtually the same in the entire column (effluent concentrations close to influent concentrations), the observed metal distribution most likely reflects the pH dependence of the slow sorption process. At pH ∼5.8-6.5, the effective cation exchange capacity of the soil can be estimated as 0.076-0.085 molc/kg (12), corresponding to ∼0.042 mol/kg of adsorbed bivalent cations. The sum of Zn, Ni, Co, and Cd retained in the inlet section (∼0.061 mol/kg) considerably exceeds this amount, notably in the presence of 10 mM CaCl2. This also points toward a precipitation rather than an adsorption process causing slow metal sorption. Zn XAFS Spectroscopy. Figure 5 shows Zn K-edge XAFS spectra of the soil samples from the first set of experiments collected after loading (loaded), after subsequent removal of exchangeable cations (leached), and from the second set of experiments collected in three sections after loading (inlet, middle, outlet), together with reference spectra (Zn-LDH, sorbed Zn, aqueous Zn). Sample descriptions and fit results are provided in Table 1. The near-edge region (XANES), the k3-weighted χ function of the EXAFS region, and its Fourier transform (from left to right) all indicate that a solid Zn phase has formed during metal sorption. The first shell was fit with 6 ( 1 oxygen atoms at a radial distance of 2.05-2.07 Å, both coordination number and distance being consistent with Zn in octahedral coordination with O (Table 1). The second shell was fit with 2-3 Zn atoms at 3.06-3.1 Å (Table 1). The distances and coordination numbers for O and Zn match those found by Ford and Sparks (15) for Zn-Al LDH surface precipitates on pyrophyllite and those of our own Zn-Al LDH (Table 1). Aqueous Zn2+ has a similar first shell, indicating that Zn is surrounded by six water molecules in a symmetrical octahedral arrangement, but a second shell is missing. A similar spectrum would be expected for Zn sorbed as an outer-sphere complex, because the hydration sphere is maintained. Zn sorbed to the same soil at pH 5.5, however, shows a weak second shell, which could be fit with 1 Al at 2.99 Å. This indicates, that Zn is sorbed as an innersphere sorption complex, most likely to aluminol groups on edge surfaces of phyllosilicates or amorphous Al hydroxides. While the distances and coordination numbers of the column samples are in line with formation of a Zn-Al LDH precipitate, other phases such as hydrozincite, Zn hydroxide nitrate, or Zn-containing phyllosilicates may have a similar second shell configuration (15). Furthermore, backscattering from Zn cannot easily be distinguished from that of Fe or Mn. Therefore, the low second shell coordination numbers and the observed distances could also indicate the formation
FIGURE 5. Zn-XAFS spectra of Zn-reacted soil samples (loaded, leached, inlet, middle, outlet) and reference samples (Zn-LDH, sorbed Zn, aqueous Zn).
TABLE 1. Zn-K XAFS Fit Results (S02)0.9) of Column Soil Samples and Reference Substances samples label
description
loaded
after metal loading (step 3) (0.1 mM Me2+, pH 6.5, Figure 3) leached after leaching with CaCl2 (step 5) (0.1 mM Me2+, pH 6.5, Figure 3) inlet after metal loading (step 4) (0.2 mM Me2+, pH 6.5, Figure 4) middle after metal loading (step 4) (0.2 mM Me2+, pH 6.5, Figure 4) outlet after metal loading (step 4) (0.2 mM Me2+, pH 6.5, Figure 4) sorbed Zn Zn sorbed to soil (pH 5.5) aqueous Zn aqueous Zn2+ (1 M Zn(NO3)2) Zn-LDH Zn-Al LDH (synthesized) Zn-Al LDH on pyrophyllite from ref 15 a
Coordination number.
b
Zn-O CNa
R [Å]b
Zn-metal σ2 [Å2]c
CN
R [Å]
σ2 [Å2]
∆E0 [Å]d χ2res %e
6.2
2.05
0.0091
2.3 Zn
3.06
0.0064
0.5
11.6
6.6
2.05
0.0093
2.2 Zn
3.08
0.0068
0.9
10.2
7.0
2.07
0.0104
2.9 Zn
3.09
0.0075
1.3
5.2
7.0
2.06
0.0117
2.0 Zn
3.11
0.0064
1.3
7.1
5.7
2.06
0.0078
1.7 Zn
3.09
0.0067
-0.1
21.9
6.8 5.8 6.0 5.7-6.2
2.08 2.07 2.07 2.02-2.06
0.0080 0.7 Al 2.99 0.0011 0.5 7.3 0.0089 -0.4 6.8 0.0092 3.9 Zn 3.10 0.0084 1.0 10.5 0.010-0.011 1.1-3.1 3.09-3.10 0.006-0.011 0.2-0.7
Radial distance. c Debye-Waller factor.
d
of inner-sphere sorption complexes, where Zn atoms share corners with two adjacent Fe or Mn octahedra in Fe or Mn (hydr)oxide. The latter option can easily be discarded, because the multiple scattering peaks at ca. 6 Å in the Fourier transforms indicate the existence of a long-range structure (i.e., a three-dimensional solid phase). Other solid phases than Zn-Al LDH can also be excluded, because the truncated oscillation at 8 Å-1 of the χ functions, together with an apparent second shell coordination number of < 4, is a unique feature of LDH phases, which does not exist for pure metal hydroxides or phyllosilicates (10). Therefore, the XAFS results unequivocally confirm the formation of Zn-Al LDH during metal sorption at pH ∼ 6.5. The average local structure around Zn is not changed by removing exchangeable Zn with Ca (Table 1, leached). This is in line with the small amount of Zn removed during step 5 (8%), which would not change the second shell coordination number significantly. However, there is a significant decrease of the second shell coordination numbers from the column inlet to the outlet (Table 1, samples inlet, middle, outlet from the experiment with c0 ∼ 0.2 mM). Because the multiple scattering feature at 8 Å-1 is largely maintained, indicating that the Zn-to-Al ratio does not change, the decreasing second shell coordination number indicates an increasing fraction of adsorbed Zn with increasing distance from the inlet. Discussion of Metal Sorption and Release Mechanism. The column experiments provided clear evidence for both rapid and slow sorption processes for Zn, Ni, Co, and Cd.
Phase shift. e Fit error.
The relative importance of slow processes decreased in the order Zn > Ni > Co > Cd. Column release experiments showed that the rapidly sorbed metal cations remained Ca exchangeable, while the slowly sorbed metal fractions were only released upon acidification to pH 3. Several possible mechanisms may contribute to slow metal sorption in soils: (1) diffusion into microporous solid phases, (2) surface diffusion from low- to high-affinity adsorption sites, and (3) precipitation of new solid phases. In our experiments, the observed slow process was not dominated by micropore diffusion, because acidification to pH 3 resulted in rapid metal release from the soil material. Surface diffusion to highaffinity sites would be possible; however, XAFS spectroscopy clearly revealed the formation of a Zn-Al LDH solid phase in the soil columns at pH ∼ 6.5. Because the general leaching pattern was very similar for all metals, we expect that Ni, Co, and possibly to a small extent Cd were also bound in a similar precipitate. Formation of Zn-Al LDH on pyrophillite (15) and of Ni-Al and Co-Al LDH in various different mineral suspensions has been observed previously but only at pH 7-8 and higher metal concentrations (10). Formation of CdAl LDH has not been observed in soil or clay systems, although it has been synthesized in the laboratory (18). The observed sequence of slow metal sorption at pH ∼ 6.5 and subsequent metal release at pH 3.0 coincides with the stability sequence of the respective metal hydroxides. For Zn, Ni, and Co, the stability sequence of metal hydroxides again follows the stability of M-Al LDH phases (26). This might suggest that VOL. 36, NO. 17, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3753
Ni and Co indeed reacted in analogy to Zn, either by forming one mixed Zn-Ni-Co-Al hydroxide phase or by each metal forming a separate precipitate. The ratio of metals sorbed by slow processes (Zn/Ni/Co/Cd ) 31:12:6:1) is closely reflected by the ratio of peak release concentrations during leaching at pH 3 (Zn/Ni/Co/Cd ) 30:14:5:1). This indicates that a mixed solid phase with a single solubility may have been formed rather than separate phases all having the same solubility or dissolution kinetics. However, additional research is needed to clarify the possible formation and stability of mixed LDH phases containing Zn, Ni, and Co. The combination of flow-through soil columns and XAFS spectroscopy seems to be a promising approach for investigating the formation and stability of such precipitates at low metal concentrations in soils.
Acknowledgments The authors gratefully acknowledge the support by the staff of beamline X11A, NSLS. Financial support of this research was provided by the Swiss Ministry of Science and Education (Project BBW-Nr. 97-0116) in the framework of the EU project FAMEST (ENV4-CT97-0554).
Supporting Information Available Table of total amounts of loaded metals and relative sorbed and released fractions. This material is available free of charge via the Internet at http://pubs.acs.org.
Literature Cited (1) Bru ¨ mmer, G. W.; Gerth, J.; Tiller, K. G. J. Soil Sci. 1988, 39, 37-52. (2) Ford, R. G.; Scheinost, A. C.; Sparks, D. L. Advances in Agronomy; Academic Press: San Diego, 2001; Vol. 74, pp 41-62. (3) Axe, L.; Anderson, P. R. J. Colloid Interface Sci. 1997, 185, 436448. (4) Axe, L.; Trivedi, P. J. Colloid Interface Sci. 2002, 247, 259-265. (5) Scheinost, A. C.; Abend, S.; Pandya, K. I.; Sparks, D. L. Environ. Sci. Technol. 2001, 35, 1090-1096. (6) Scheidegger, A. M.; Lamble, G. M.; Sparks, D. L. J. Colloid Interface Sci. 1997, 186, 118-128.
3754
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 17, 2002
(7) Ford, R. G.; Scheinost, A. C.; Scheckel, K. G.; Sparks, D. L. Environ. Sci. Technol. 1999, 33, 3140-3144. (8) Roberts, D. R.; Scheidegger, A. M.; Sparks, D. L. Environ. Sci. Technol. 1999, 33, 3749-3754. (9) Thompson, H. A.; Parks, G. A.; Brown, G. E. J. J. Colloid Interface Sci. 2000, 222, 241-253. (10) Scheinost, A. C.; Sparks, D. L. J. Colloid Interface Sci. 2000, 223, 167-178. (11) Kretzschmar, R.; Voegelin, A. In Heavy Metals Release in Soils; Selim, H. M., Sparks, D. L., Eds.; Lewis Publishers: Boca Raton, FL, 2001; pp 55-88. (12) Voegelin, A.; Vulava, V. M.; Kretzschmar, R. Environ. Sci. Technol. 2001, 35, 1651-1657. (13) Scheckel, K. G.; Sparks, D. L. J. Colloid Interface Sci. 2000, 229, 222-229. (14) Scheinost, A. C.; Ford, R. G.; Sparks, D. L. Geochim. Cosmochim. Acta 1999, 63, 3193-3203. (15) Ford, R. G.; Sparks, D. L. Environ. Sci. Technol. 2000, 34, 24792483. (16) Schlegel, M. L.; Manceau, A.; Charlet, L.; Chateigner, D.; Hazemann, J. L. Geochim. Cosmochim. Acta 2001, 65, 41554170. (17) Manceau, A.; Lanson, B.; Schlegel, M. L.; Harge, J. C.; Musso, M.; Eybert-Berard, L.; Hazemann, J.-L.; Chateigner, D.; Lamble, G. M. Am. J. Sci. 2000, 300, 289-343. (18) Vichi, F. M.; Alves, O. L. J. Mater. Chem. 1997, 7, 1631-1634. (19) Smith, R. M.; Martell, A. E. Critical Stability Constants; Plenum Press: New York, 1976; Vol. 4. (20) Lytle, F. W.; Sandstrom, D. R.; Marques, E. C.; Wong, J.; Spiro, C. L.; Huffman, G. P.; Huggins, F. E. Nucl. Instrum. Methods Phys. Res. 1984, 226, 542-548. (21) Ressler, T. J. Synchrotron Radiat. 1998, 5, 118-122. (22) FEFF Project, 7.02 ed.; Department of Physics, University of Washington: Seattle, WA, 1996. (23) Taylor, R. M. Clay Miner. 1984, 19, 591-603. (24) Kinniburgh, D. G.; Jackson, M. L. In Adsorption of inorganics at solid-liquid interfaces; Anderson, M. A., Rubin, A. J., Eds.; Ann Arbor Science: Ann Arbor, MI, 1981; pp 91-160. (25) Abd-Elfattah, A.; Wada, K. J. Soil Sci. 1981, 32, 271-283. (26) Boclair, J. W.; Braterman, P. S. Chem. Mater. 1999, 11, 298-302.
Received for review December 5, 2001. Revised manuscript received June 5, 2002. Accepted June 5, 2002. ES010316M
