Effects of Aging and pH on Dissolution Kinetics and Stability of

Alexander S. Parker , Rehab Al Botros , Sophie L. Kinnear , Michael E. Snowden , Kim McKelvey , Alexander T. Ashcroft , Mel Carvell , Andrew Joiner , ...
0 downloads 0 Views 102KB Size
Download PDF
Environ. Sci. Technol. 2002, 36, 2198-2204

Effects of Aging and pH on Dissolution Kinetics and Stability of Chloropyromorphite KIRK G. SCHECKEL* AND JAMES A. RYAN U.S. EPA, ORD, NRMRL, LRPCD, RCB, 5995 Center Hill Avenue, Cincinnati, Ohio 45224

The objectives of this research were to understand the effect of aging time on chloropyromorphite stability by dissolution, to examine physical and chemical alterations of the pyromorphite samples, and to model the kinetic data collected from the dissolution experiments. The results of this investigation indicate that chloropyromorphite formation is kinetically rapid and that its dissolution in acid is thermodynamically stable, ideal conditions for Pb immobilization that has emerged as a potential remediation strategy. In terms of aging prior to dissolution, X-ray absorption fine structure (XAFS) and X-ray diffraction (XRD) spectroscopies were unable to distinguish fundamental differences in progressively aged samples; however, highresolution thermogravimetric analysis (HRTGA) did demonstrate that the thermostability of the chloropyromorphite material increased with increasing residence time. The stirred-flow and batch dissolution studies suggest that the aging process ceased within 24 h and that the dissolution rate of the 1-day aged sample was not significantly different than the 1-year aged specimen. The amount of Pb released peaked at 21% (1-h sample, stirred-flow, pH 2.0) and was as low as 0.17% (1-year sample, batch method, pH 6.0). Postdissolution analyses of chloropyromorphite with XAFS, XRD, and HRTGA revealed no detectable chemical alterations of the pyromorphite samples signifying only release of dissolved Pb to solution and no formation of secondary products during dissolution.

Introduction Published literature has clearly demonstrated that the reaction of a lead (Pb) source as either Pb minerals (anglesite, cerrusite, or galena), goethite adsorbed lead, Pb-contaminated soils, or an in vitro bioavailable assay with a phosphate reserve (apatite or hydroxyapatite) results in the rapid kinetic formation of pyromorphite (Pb5(PO4)3(Cl, OH, F, etc.) (1-7). Chloropyromorphite is the most stable Pb mineral found under normal environmental conditions; thus, other solidphase Pb species should be transformed to chloropyromorphite by a dissolution-precipitation mechanism to immobilize soluble soil Pb in situ by simply adding phosphate to the soil (8). However, the long-term fate of immobilized Pb as precipitated pyromorphite is not fully understood. Crystal dynamics dictate that, as aging time increases, the relative stability of a precipitate increases, thus resulting in reduced solubility and decreased bioavailability (9-11). If this hypothesis is correct, Pb-contaminated soils amended with a * Corresponding author phone: 513-487-2865; fax: 513-569-7879; e-mail: [email protected]. 2198 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 10, 2002

phosphate source should observe decreasing amounts of bioavailable Pb with increasing residence time. Nevertheless, the effect of residence time has not been explored in relation to the formation of pyromorphite and its potential as an in situ remediation technique. The solubility of pyromorphite has been addressed in the literature; however, it is often misrepresented or misinterpreted when cited from original solubility literature. Pioneering research on the solubility of lead phosphates can be traced to Jowett and Price (12), who calculated the solubility of chloropyromorphite to be 10-79.1 at 37 °C. An effort was undertaken to examine the solubility of chloropyromorphite and its potential conversion to hinsdalite (PbAl3(PO4)(SO4)(OH)6) in an alkaline solution containing aluminum and sulfate (13). Employing the Debye-Hu ¨ckel limiting law, Baker (13) determined the solubility of chloropyromorphite to be 10-34.5; however, he was quick to point out that the DebyeHu ¨ ckel limiting law fails when solution ionic strengths are greater than 0.01 and that it was also possible that the solubility studies were not in equilibrium at the time of analysis. An excellent four-paper series by Nriagu (8, 14-16) in the early 1970s discussed the environmental importance and solubility of a variety of lead orthophosphates. One of these papers (15) addressed the conversion of PbHPO4 to chloropyromorphite in a NaCl solution. On the basis of elemental concentrations in solution, the solubility product (Ksp) for chloropyromorphite was calculated to be 10-84.4, as suggested by its Gibbs free energy of formation (∆Gof) (eq 4). This value, as Nriagu suggests, is limited to a very select pH range, and the dissolution of chloropyromorphite is highly dependent on the pKa of the phosphate anion (eqs 1-4). Under most environmental conditions, the Ksp of chloropyromorphite is 10-25.05 (eq 2) (17); however, this value underestimates solubility at low pH (8.0), as noted in the following equations (phosphoric acid pKa’s from Brown et al. (18)).

pH 0-2.12 Pb5(PO4)3Cl + 9H+ f 5Pb2+ +

3H3PO4 + ClpKa1 ) 2.12

Ksp ) 10-18.69 (1)

pH 2.12-7.21 Pb5(PO4)3Cl + 6H+ f 5Pb2+ + 3H2PO41- + ClpKa2 ) 7.21

Ksp ) 10-25.05 (2)

pH 7.21-12.38 Pb5(PO4)3Cl + 3H+ f 5Pb2+ +

3HPO42- + ClpKa3 ) 12.38

Ksp ) 10-46.9 (3)

pH 12.38-14 Pb5(PO4)3Cl + 0H+ f 5Pb2+ + 3PO43- + Cl-

Ksp ) 10-84.4 (4)

Phosphate treatment experiments of Pb-contaminated soils have been undertaken to convert soil Pb to pyromorphite (1, 3, 19-22). One such study was designed by the In-Place Inactivation Natural Ecological Remediation Team (IINERT) workgroup (23) that has, for the past several years, maintained test plots at a Pb-contaminated smelter site (Joplin, MO) in which an array of phosphorus and phosphorus and Fe or 10.1021/es015803g Not subject to U.S. copyright. Publ. 2002 Am. Chem.Soc. Published on Web 04/17/2002

biosolids amendments have been added. Preliminary XANES data (24) suggest that some pyromorphite formed as a result of P amendments; however, little work has been done to examine or predict the potential fate of the formed pyromorphite. The purpose of this study was to investigate the kinetics of dissolution and crystallization dynamics of synthetic chloropyromorphites as influenced by increasing aging times ranging from 1 h to 1 year to better understand what may be happening in the field setting. The objectives of this research were to (1) measure the effect of residence time on the stability of synthetic pyromorphite as influenced by nitric acid (HNO3) at three pH values (2.0, 4.0, and 6.0) via dissolution in stirred-flow and batch reaction vessels, (2) investigate changes in the crystal chemistry with time of the aged synthetic pyromorphites employing spectroscopic (Xray diffraction (XRD), X-ray absorption fine structure (XAFS)) and thermogravimetric (high-resolution thermogravimetric analysis (HRTGA)) techniques as a result of the dissolution reaction, and (3) model and analyze the kinetic data to determine rate constants and kinetic relationships.

Experimental Section Chloropyromorphite Synthesis. Precipitation of various aged chloropyromorphites was carried out in 2-L HDPE bottles by the combination of 0.25 M PbCl2 (Fisher Scientific, Pittsburgh, PA) and 0.15 M H3PO4 (Fisher Scientific, Pittsburgh, PA) at 23 ( 2 °C. Aging times included 1 h, 1 day, 1 week, 1 month, 3 months, 6 months, and 1 year. In the 2-L HDPE bottle, 69.53 g of PbCl2 was dissolved in 1 L of Millipore water and purged with nitrogen to inhibit the formation of lead carbonate from atmospheric carbon dioxide as a result of the high initial Pb concentration. The phosphoric acid solution consisted of 9 mL of 86% (v/v) reagent-grade H3PO4 added to 991 mL of Millipore water in a 1000 mL volumetric flask. The 0.15 M phosphoric acid solution was added via a 100 mL buret with the stopcock opened to its fullest flow. The remaining phosphoric acid solution from the 1000 mL volumetric flask was added to the buret with a polycarbonate funnel. The suspension was continuously stirred to form a small vortex to ensure adequate mixing and N2 purged. Once all of the phosphoric acid solution was added to the 2-L HDPE bottle, the pH (under criteria outlined in EPA Method 9040B) was adjusted to 7.0 by the addition of 5 M NaOH (prepared fresh with Millipore water) with the assistance of a pH-stat titrator and nitrogen purging. A pH of 7 was chosen to mimic environmental conditions of established limed field plots of Pb-contaminated soils that have undergone various P amendments to examine the effectiveness of pyromorphite formation as an immobilization technique. Other field conditions such as Ca from liming material, other metals (Cd and Zn) that may substitute in the pyromorphite structure, and competitive anions (SO4 and CO3) were not considered so that only unmodified chloropyromorphite (Pb5(PO4)3Cl) was consistently used in the dissolution studies for comparative results. The suspension was allowed to equilibrate for 3 days (shorter times as necessary to collect aged samples of 1 h and 1 day) on the pH-stat apparatus so that very little or no additional NaOH inputs were required to maintain the pH at 7.0. After 3 days, the bottle was removed from the pH-stat titrator, capped to maintain a nitrogen environment, and placed on a large stir-plate for aging. Periodically, the pH was adjusted to 7.0 with 1.0 M NaOH, while nitrogen purged until the assigned aging period was completed at which time the precipitated pyromorphite material was harvested. The aged material was collected by centrifugation and washed 5 times with Millipore water to removed excess lead, phosphorus, and chlorine. After the washing was complete, the chloropyromorphite was freezedried to stop the aging process.

Macroscopic Dissolution. The dissolution experiments were carried out by a continuous stirred-flow technique and a one-step batch method to compare the influence of reaction products on dissolution kinetics. It was expected that more dissolution would occur in the stirred-flow technique versus the batch method because reaction products that may inhibit further dissolution will be continuously removed in the stirred-flow design. In both dissolution methods, the overall parameters included the use of nitric acid (HNO3) (Fisher Scientific) at pH values of 2.0, 4.0, and 6.0 to simulate environmental conditions one may find in areas ranging from acid mine drainage to agricultural soil settings. The concentration of pyromorphite was the same in both dissolution systems at 10 g L-1 ratio (0.6 g/60 mL in stirred-flow and 0.3 g/30 mL in batch). The stirred-flow dissolution experiments employed a custom-designed stirred-flow dissolution apparatus (60 mL total volume) (Soil Measurements Systems, Tucson, AZ) in which a precision volume piston-flow pump (Fluid Metering, Inc., Syosset, NY) was attached to the inputport of the cell and the out-flow solution was delivered to a fraction collector (Spectra/Chrom, Fisher Scientific) for flame atomic absorption spectroscopy (AAS) (Perkin-Elmer 5100) analysis (EPA Methods 7000A and B). A 47-mm membrane filter (0.45 µm) was fitted in the stirred-flow chamber so that the pyromorphite material remained in the vessel while allowing the acid solution and dissolution products to freely leave. Nitric acid was added to the stirred-flow chamber at a rate of 36 mL h-1 with fraction collection at 20-min intervals resulting in three 12-mL samples/h for a total of 72 samples in a 24-h period. The batch dissolution experiments used a 50-mL polypropylene centrifuge tube in which 30 mL of nitric acid at either pH 2.0, 4.0, or 6.0 was deposited into the tube with the pyromorphite material (0.3 g). The tube containing the aged pyromorphite and nitric acid was capped, sealed, and placed on a reciprocating shaker (120 rpm) for a 24-h reaction period at 23 ( 2 °C. After the reaction time, the tube was centrifuged, 10 mL of the supernatant was filtered with a disposable syringe-filter apparatus (0.45 µm), and the rest discarded. A portion of the filtered supernatant was acidified and diluted at a 1:10 ratio for AAS analysis and data reduction. The final fluid pH was measured and recorded prior to acidification. For both dissolution methods, if the sample pH was not within (0.3 pH units of the initial pH level (1.72.3 for pH 2.0; 3.7-4.3 for pH 4.0; 5.7-6.3 at pH 6.0), the test was discarded and repeated for quality control purposes; however, this was never necessary. The solid pyromorphite samples remaining from the batch dissolution experiments were retained, washed 5 times with Millipore water, and freeze-dried for XAFS, XRD, and HRTGA analyses. The pyromorphite from the batch studies were used for spectroscopic analyses because postexperimental dismantling of the stirred-flow reaction vessel made collection of the pyromorphite samples difficult due to spillage of the pyromorphite suspension and design of the reaction cell. XAFS. Aged chloropyromorphite samples pre- and postdissolution were analyzed by XAFS to determine the local atomic structure surrounding Pb in the crystal phases. For XAFS analysis, the freeze-dried chloropyromorphites were loaded into sample cell holders and sealed with Kapton tape. Pb (13 055 eV) LIII-XAFS data were collected at beamline X-11A using Si(111) monochromator crystals at the National Synchrotron Light Source (NSLS), Brookhaven National Laboratory, Upton, NY. Three to five scans were collected at ambient temperature in fluorescence mode using a Kr-purged Lytle detector with an As filter to reduce scattered radiation and aluminum foil to diminish background fluorescence. Energy was calibrated to the first inflection of a Pb-metal foil standard and was collected simultaneously with the spectrum of each sample. The collected scans for a particular sample were merged, the data were then normalized, and the VOL. 36, NO. 10, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2199

background was removed by spline fitting using WinXAS 1.3. The data were then converted to k space, windowed, and Fourier-transformed to convert to R space. Conventional shell-by-shell fitting of the radial structure functions (RSFs) was attempted by theoretical paths for Pb, O, and P backscatter atoms generated from crystallographic data of model compounds using FEFF 7.0. Variables obtained from the fits include the energy phase shift (∆Eo), coordination numbers (N), bond distances (R), and Debye-Waller factors (σ2) that were derived from nonlinear least-squares fitting. Values for R are accurate from approximately (0.02 Å (first shell) to (0.04 Å (second shell), and N values are generally accurate to (1 (25, 26). XRD. Powder XRD patterns were obtained using a Phillips PW3040/00 X’Pert-MPD Diffractometer system with a Cu anode ceramic diffraction X-ray tube operating at 50 kV and 40 mA. The sample platform used in for the XRD experiments was a PW3064 Sample Spinner and was automated in conjunction with a sample (changer) batch program. The pyromorphite samples were back-filled into 32-mm circular sample holders to obtain a flat, randomly oriented surface for analysis. Using the X’Pert Data Collector software, a relative scan method was employed that used the upper diffracted beam path and a goniometer scan axis. Data were collected in a step scan mode from 10 to 75° 2θ at a step size of 0.015° and a rate of 0.50 s per step. The samples were rotated on the spinning sample platform at 2 revolutions/s so that each step of the scan observed a full rotation of the sample. After the raw XRD patterns were collected, X’Pert Graphics and Identity software was employed to conduct a peak search, smooth the raw spectra to a factor of 1, KR2 stripping with a factor of 1, and a search-match of the identified peaks relative to the ICDD library card database. HRTGA. HRTGA was employed to characterize the thermostability attributed to the aged chloropyromorphite samples. Data are presented as the derivative of the weight loss curve versus temperature. Intensity of the peaks for the derivative of the weight loss curves and the associated temperatures for the peaks are indicative of compositional and physical changes of the solid material. We employed a TA Instruments 2950 high-resolution thermogravimetric analyzer to examine approximately 30-50 mg samples from aging and dissolution studies that were previously freezedried. The experiments were run under a N2 atmosphere over a temperature range of 30-810 °C. The analytical highresolution heating method settings included a 5-min isothermal period to allow N2 purging of the furnace and sample, a maximum heating rate of 15 °C min-1 with a resolution of 5.0, and a high-resolution sensitivity of 4.0.

Results and Discussion Aging of Chloropyromorphite. The synthesis of chloropyromorphite for this study was nearly instantaneous with the formation of white crystals the moment the phosphoric acid solution was added to the reaction vessels. Visually, the white crystals began to change to a yellowish-green tint as aging time progressed, with the most noticeable color change occurring between the 3-month and 6-month samples due to increased crystallinity of the structure. Figure 1 shows data for XAFS, XRD, and HRTGA analyses for aged pyromorphite prior to dissolution. The XAFS data (solid line) in Figure 1a indicate relatively little difference between the 1-h and 1-year sample, and XAFS fits (dotted line) to the RSFs represent the data well. The fit data (Table A, Supporting Information) show that six O atoms surround Pb in an octahedral coordination environment noted by three O atoms at 2.62 Å and the other three O atoms at 2.83 Å in the first shell. Second nearest-neighbor data indicate that three additional O atoms are positioned at approximately 3.01 Å followed by three P atoms at about 3.42 Å (Table A, Supporting 2200

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 10, 2002

FIGURE 1. Pb-LIII XAFS spectra (a) (raw data, solid line; fit data, dotted line), XRD patterns (b), and HRTGA derivative weight loss curves (c) of aged synthetic chloropyromorphite. Information). These local structural data are consistent with those of natural pyromorphite (27). The data do not indicate any significant changes as aging time increased from 1-h to 1-year. These same chloropyromorphite samples were also examined by XRD (Figure 1b) and, again, do not suggest any alteration in the chemistry or crystallinity of the pyromorphite phases as aging time increases. Each sample was positively identified as crystalline, synthetic chloropyromorphite with

FIGURE 2. Rate of Pb removal from chloropyromorphite as a function of aging for stirred-flow (left) and batch (right) dissolution studies using HNO3 at pH 2.0 (a and b), 4.0 (c and d), and 6.0 (e and f). no detectable secondary products. It is worthwhile noting that the intensity of counts for the 1-h samples equates well with the intensity of the 1-year aged pyromorphite in that crystallinity was rapidly achieved. HRTGA was successfully employed to show that aging of the pyromorphite crystals did change with increasing residence time (Figure 1c). Typically with HRTGA analysis, as a crystal becomes more stable with aging due to Ostwald ripening or loss of waters of hydration, the less weight loss one would expect with increasing physical stability. This was observed for the HRTGA data collected with respect to the aged chloropyromorphites. The magnitude of the derivative peaks were much greater at earlier aging times, particularly the 1-h and 1-day samples, and subsequently diminished as aging time proceeded to 1 year suggesting, perhaps, an increase in Ostwald ripening of the chloropyromorphite crystals (28-30). The weight loss events are attributed to the thermodegradation of structural phosphate to P2O5. In summary, as aging time increased from

1 h to 1 year, XAFS and XRD were unable to distinguish chemical or physical differences relative to aging for the chloropyromorphite samples. However, HRTGA was able to demonstrate that the thermostability of pyromorphite increased with increasing aging time. Additionally, these analytical techniques were unable to identify other Pb phases; however, this is not to say that they may be present in microquantities undetectable by the techniques employed in this study. Macroscopic Dissolution of Chloropyromorphite and Kinetics. The experiments conducted in this section were designed to observe and compare data as influenced by aging time, pH of dissolution agent, and dissolution method (stirred-flow vs batch). These data are assembled in Figure 2 and are presented as the percent amount of Pb remaining in the solid phase versus time. Regardless of nitric acid pH or dissolution method, the obvious point that stands out in Figure 2 is that, in every case, the 1-h aged pyromorphite VOL. 36, NO. 10, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2201

sample stood apart as the most soluble. Aging of chloropyromorphite from 1 day to 1 year did not promote a significant increase in stability suggesting that crystal aging was accomplished within 24 h. As one would expect, the pH of the nitric acid dissolution agent did play a role in the amount of Pb released from the pyromorphite surface and followed the trend that as pH decreased the amount of Pb released increased. This was evident in comparing pH 2.0 (Figure 2b) versus pH 4.0 (Figure 2d) versus pH 6.0 (Figure 2f) for the dissolution of the 1-month aged sample using the batch method, which resulted in 5.0%, 1.0%, and 0.5% of total Pb being dissolved, respectively. Removal of dissolution reaction products in the stirred-flow experiments enhanced pyromorphite dissolution relative to the batch studies, and the difference in Pb released between these two methods was more pronounced as pH increased as seen when comparing stirred-flow versus batch at pH 2.0 (21% vs 17%), 4.0 (11% vs 6%), and 6.0 (11% vs 4%) for the 1-h aged sample. The macroscopic dissolution data show that, aside from the 1-h aged sample, the aging process was rapid and that Pb release during dissolution with nitric acid was limited because of the stability of chloropyromorphite. While the data collected from the stirred-flow studies was useful in determining kinetic rate constants, it does not represent natural environmental systems. The batch method, which is a suspension of reactants and products of a given chemical mixture, more closely mimics the natural environment. With this in mind and relating back to the P-treated field plots, once pyromorphite forms in contaminated soils and sediments, it should remain immobilized and stable in these environments under near neutral pH, which is the case for the IINERT study that limed the field plots to pH 7.0. As previously mentioned, the dissolution data collected from the stirred-flow experiments were used to determine kinetic rate constants and were modeled against an array of kinetic models (zero-third-order, Elovich, power function and parabolic rate law) to determine the best fits in terms of standard error based on linear least-squares analysis. The dissolution rate for all experiments followed first-order kinetics. With over 63 stirred-flow dissolution experiments conducted for this research, the amount of data was abundant, and a table summary of first-order rate coefficients at the various aging times and three pHs are presented in Table B of Supporting Information. A brief summary of Table B provided two trends related to these first-order rate constants. First, as aging time increases, the rate of dissolution decreases (Figure A1, Supporting Information), indicating a slight increase in stability with aging time. Second, as pH increases, the rate of dissolution decreases (Figure A2, Supporting Information), noting that proton concentration plays an important role in chloropyromorphite dissolution. Combining all of these observations in Figure 3, one observes that as both aging time and pH increase, the dissolution rates decrease so that the 1-h sample at pH 2.0 is the least stable pyromorphite specimen (20% Pb released), while the 1-year chloropyromorphite dissolving at pH 6.0 is the most stable sample (0.6% Pb released) in terms of potential Pb availability. The curves in Figure 3 illustrate the point that the 1-h sample (first data point) was the least stable but quickly stabilized in terms of a steady kinetic rate constant as aging time increased noted by the near-zero slope of the curves past the first point. Also noteworthy in Figure 3 is the placement of the curves with respect to pH. The alignment of pH 2.0 data is set apart from the pH 4.0 and 6.0 data curves, which lie nearly on top of each other. This may be influenced by the pKa of the phosphate anion at these pH values, which is identical for pH 4.0 and 6.0 (pKa ) 7.21) and lower for pH 2.0 (pKa ) 2.12). XAFS, XRD, and HRTGA Dissolution of Chloropyromorphite. A concern that arose from the macroscopic 2202

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 10, 2002

FIGURE 3. Relationship between aging time and derived first-order dissolution rate coefficients as influenced by pH via stirred-flow dissolution. dissolution studies was whether some of the released Pb was forming secondary reaction products such as cerrusite (PbCO3) or adsorbing onto the pyromorphite surfaces. If these secondary products were being formed, even in minor amounts, they may have been distinguishable with XAFS, XRD, or HRTGA; however, their nondetection does not suggest nonpresence due to possible technique detection limits. XAFS, XRD, and HRTGA investigations of the chloropyromorphite phases after dissolution in nitric acid at pH 2.0, 4.0, and 6.0 were conducted with the spectroscopic dissolution data for the 1-month aged sample presented in Figure 4 for each technique. Although not shown, similar results were obtained for the other aging times. Figure 4a shows the XAFS results (solid line) and theoretical fits (dotted line) after dissolution with nitric acid compared to the untreated (predissolution) 1-month aged chloropyromorphite samples. The fit data (Table C, Supporting Information) show that six O atoms surround Pb in an octahedral coordination environment noted by three O atoms at 2.62 Å and the other three O atoms at 2.85 Å in the first shell. Second nearest-neighbor data indicate that three additional O atoms are positioned at approximately 3.01 Å followed by three P atoms with a 3.43 Å interatomic bond distance (Table C, Supporting Information). It can be concluded from the dissolution XAFS analysis that there was no alteration in the local structural environment for the 1-month chloropyromorphite sample after dissolution with nitric acid and that there were no evident secondary reaction products formed. The 1-month chloropyromorphite dissolution samples were also examined by XRD (Figure 4b), for which the data do not suggest any alteration in the chemistry or crystallinity of the initial pyromorphite phases and are exact matches to the chloropyromorphite ICDD library cards used in the peakmatching fitting procedure. Emergence of crystalline secondary phases was not detected by XRD. HRTGA provided similar results (Figure 4c) as the XAFS and XRD analyses in that there were no distinguishable differences in the untreated 1-month chloropyromorphite sample in comparison to samples that underwent dissolution as indicated by no change in the magnitude or position of the weight-loss derivative peaks.

In Situ Immobilization of Pb The potential impact of stabilization remediation methods to safely sequester metal contaminants in the natural environment holds great promise but must be managed carefully and intelligently. Research in the immobilization

to examine the stability of immobilized contaminants. The results of this work and of others (1-3, 5, 8, 20, 32) demonstrate that pyromorphite formation can be easily accomplished by the reaction of available Pb and P. However, it is inevitable that one of these components will be ratelimiting under normal environmental conditions and should be considered in adopting this remediation strategy to a Pbcontaminated soil. As suggested from this study, pyromorphite formation in a soil or sediment system can produce stable crystalline precipitates that achieve a low entropy state quickly with aging as demonstrated by its rapid formation as examined by XAFS, XRD, and HRTGA (Figure 1). Such chemical and physical attributes of chloropyromorphite suggest that its persistence would endure most environmental conditions, thus making Pb-immobilization via phosphorus an ideal remediation mechanism as seen by its dissolution characteristics with nitric acid (Figures 2 and 3) as well as XAFS, XRD, and HRTGA confirmation (Figure 4).

Acknowledgments The U.S. EPA has not subjected this manuscript to internal 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. This manuscript benefited from constructive reviews by S. Al-Abed, G. Hettiarachchi, C. Impellitteri, and three anonymous reviewers.

Supporting Information Available Tables A and C list the structural fit parameters of the XAFS data. Figure A and Table B show fits of the kinetic data to the first-order reaction model and calculated forward kinetic rate coefficients, respectively. This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited

FIGURE 4. Pb-LIII XAFS spectra (a) (raw data, solid line; fit data, dotted line), XRD patterns (b), and HRTGA derivative weight loss curves (c) for dissolution of 1-month aged synthetic chloropyromorphite with HNO3 from batch experiments. area has, for the most part, only focused on the initial and intermediate formation steps of the solid phase and often does not explore the long-term fate as a result of aging for the sequestered contaminant (9, 31). It is not prudent to assume that once a metal of interest is complexed in a crystalline phase that it is forever removed from transport and bioavailability reactions, and it is, therefore, necessary

(1) Hettiarachchi, G. M.; Pierzynski, G. M.; Ransom, M. D. Environ. Sci. Technol. 2000, 34, 4614-4619. (2) Manecki, M.; Maurice, P. A.; Traina, S. J. Soil Sci. 2000, 165, 920-933. (3) Traina, S. J.; Laperche, V. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 3365-3371. (4) Zhang, P. C.; Ryan, J. A. Environ. Sci. Technol. 1998, 32, 33183324. (5) Zhang, P. C.; Ryan, J. A.; Yang, J. Environ. Sci. Technol. 1998, 32, 2763-2768. (6) Zhang, P. C.; Ryan, J. A. Environ. Sci. Technol. 1999, 33, 618624. (7) Zhang, P. C.; Ryan, J. A. Environ. Sci. Technol. 1999, 33, 625630. (8) Nriagu, J. O. Geochim. Cosmochim. Acta 1974, 38, 887-898. (9) Scheckel, K. G.; Scheinost, A. C.; Ford, R. G.; Sparks, D. L. Geochim. Cosmochim. Acta 2000, 64, 2727-2735. (10) Scheckel, K. G.; Sparks, D. L. J. Colloid Interface Sci. 2000, 229, 222-229. (11) Scheckel, K. G.; Sparks, D. L. Soil Sci. Soc. Am. J. 2001, 65, 685694. (12) Jowett, M.; Price, H. I. Trans. Faraday Soc. 1932, 28, 668-681. (13) Baker, W. E. Am. Mineral. 1964, 49, 607-613. (14) Nriagu, J. O. Inorg. Chem. 1972, 11, 2499-2503. (15) Nriagu, J. O. Geochim. Cosmochim. Acta 1973, 37, 367-377. (16) Nriagu, J. O. Geochim. Cosmochim. Acta 1973, 37, 1735-1743. (17) Lindsay, W. L. John Wiley and Sons: New York, 1979; pp 211220. (18) Brown, T. L.; LeMay, J. H. E.; Bursten, B. E. Chemistry, The central science, 6th ed.; Prentice Hall: Englewood Cliffs, NJ, 1994. (19) Yang, J.; Mosby, D. E.; Casteel, S. W.; Blanchar, R. W. Environ. Sci. Technol. 2001, 35, 3553-3559. (20) Cotter-Howells, J. D.; Champness, P. E.; Charnock, J. M. Mineral. Mag. 1999, 63, 777-789. (21) Pearson, M. S.; Maenpaa, K.; Pierzynski, G. M.; Lydy, M. J. J. Environ. Qual. 2000, 29, 1611-1617. (22) Ryan, J. A.; Zhang, P. C.; Hesterberg, D.; Chou, J.; Sayers, D. E. Environ. Sci. Technol. 2001, 35, 3798-3803. VOL. 36, NO. 10, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2203

(23) Remediation Technologies Development Forum, 2001; Vol. 1995. (24) Scheckel, K. G.; Yang, J. Agronomy Abstract 2001. (25) Morin, G.; Ostergren, J. D.; Juillot, F.; Ildefonse, P.; Calas, G.; Brown, G. E., Jr. Am. Mineral. 1999, 84, 420-434. (26) O’Day, P. A.; Rehr, J. J.; Zabinsky, S. I.; Brown, G. E., Jr. J. Am. Chem. Soc. 1994, 116, 2938-2949. (27) Botto, I. L.; Barone, V. L.; Castiglioni, J. L.; Schalamuk, I. B. J. Mater. Sci. 1997, 32, 6549-6553. (28) Ford, R. G.; Sparks, D. L. Environ. Sci. Technol. 2000, 34, 24792483. (29) Kahlweit, M. Scr. Metall. 1976, 10, 601-603.

2204

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 10, 2002

(30) Wei, M.; Ruys, A. J.; Milthorpe, B. K.; Sorrell, C. C. J. Biomed. Mater. Res. 1999, 45, 11-19. (31) Ford, R. G.; Scheinost, A. C.; Scheckel, K. G.; Sparks, D. L. Environ. Sci. Technol. 1999, 33, 3140-3144. (32) Ponizovskii, A. A.; Mironenko, E. V. Eurasian Soil Sci. 2001, 34, 371-381.

Received for review November 21, 2001. Revised manuscript received March 7, 2002. Accepted March 12, 2002. ES015803G