HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Dermatas, D. Articles by Xu, X. Search for Related Content PubMed PubMed Citation Articles by Dermatas, D. Articles by Xu, X. GeoRef GeoRef Citation Agricola Articles by Dermatas, D. Articles by Xu, X. Related Collections Best Management Practices Heavy Metals Soil Pollution

TECHNICAL REPORTS

Heavy Metals in the Environment

Phosphate Treatment of Firing Range Soils: Lead Fixation or Phosphorus Release?

W.M. Keck Geoenvironmental Lab., Stevens Inst. of Technology, Hoboken, NJ 07030, USA

^* Corresponding author (xxu{at}stevens.edu).

Received for publication March 27, 2007.

	ABSTRACT

TOP NOTES ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Summary and Conclusions REFERENCES

 
Phosphate treatment of lead (Pb)-contaminated soils relies on the premise that Pb converts to the thermodynamically stable, insoluble mineral class of pyromorphites. Recent research showed that treatment performance is kinetically controlled and strongly dependent on soil pH; this study employed an acidic phosphate (P) form, monobasic calcium phosphate (MCP), to investigate treatment performance of Pb occurring in an alkaline-buffered and an acidic firing range soil. The results of leaching, X-ray powder diffraction (XRPD), and modeling analyses showed that P and Pb dissolution in the alkaline soil and transformation reactions were kinetically controlled, so that: (i) TCLP (toxicity characteristic leaching procedure) and SPLP (synthetic precipitation leaching procedure) results were poor to marginal even at high MCP dosages; (ii) brushite (Ca(HPO₄)·2H₂O) and cerussite (PbCO₃) persisted in XRPD patterns; and, (iii) geochemical modeling failed to predict leaching and phase assemblages. In the acidic soil, Pb-P reactions promoted further soil acidification, improved TCLP performance, and generated better agreement with the equilibrium-based model; however, SPLP and modeling results showed that Pb concentrations could not be reduced below 15 µg/L mainly due to the low soil pH. The marginal or inadequate Pb immobilization was observed in both soils despite the elevated MCP dosages, which were well in excess of the pyromorphite stoichiometric ratio (P/Pb = 0.6). Additionally, P leaching concentrations and rates were extremely high (>300 mg/L), under both SPLP and deionized (DI) water extraction conditions, and as predicted by thermodynamic equilibrium. The performance and sustainability of phosphate-based treatment therefore seem questionable.

Abbreviations: MCP, monobasic calcium phosphate • TCLP, toxicity characteristic leaching procedure • SPLP, synthetic precipitation leaching procedure • DIW, deionized water • PATF, Picatinny Arsenal Technology Facility • FDR26, Fort Dix Range 26 • HP, hydroxypyromorphite • CP, chloropyromorphite

	INTRODUCTION

TOP NOTES ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Summary and Conclusions REFERENCES

 
FIRING ranges are the second most important source of Pb contamination according to a study by the USGS (2002). In 2001, the U.S. Environmental Protection Agency (USEPA) issued a manual on best management practices (BMPs) for Pb in outdoor shooting ranges, acknowledging the need to minimize Pb release to the environment through range maintenance activities (USEPA, 2001a). Among the BMPs to prevent Pb migration in soils, the USEPA included phosphate application to bind Pb particles; the recommendation was based on a series of treatability studies published between 1993 and 2001, including a study conducted by the USEPA itself (USEPA, 2001a). The premise for using phosphate as a stabilizing agent is to target the formation of lead phosphates, or pyromorphites (Pb₅(PO₄)₃X where X = Cl^–, OH^–, F^–), which are the thermodynamically most stable and most insoluble Pb minerals over a large pH and Eh range (Nriagu, 1974). The term pyromorphite hereafter will generally refer to all three species unless otherwise noted. The corresponding Pb solubility is extremely low under many conditions of environmental significance.

The authors have participated in an extensive investigation on metals contamination in firing ranges operated by the U.S. Department of Defense (U.S. DoD). The characteristics of the investigated sites varied greatly with respect to such factors as the magnitude of Pb concentration, Pb fragment particle size (and distribution), soil pH, soil particle size, and mineralogy. Consequently, the remedial investigation involved the screening of a number of available treatment technologies and management schemes, including phosphate amendments. A literature review focusing on phosphate immobilization of Pb revealed the following issues (Chrysochoou et al., 2007): (i) Pb-P reaction kinetics may be extremely slow, depending on the pH regime of the treatment and the solubility of the initial P and Pb sources; and (ii) the consequences of high unreacted P concentrations are largely unstudied.

Specifically, the review of phosphate-based treatability studies, especially those conducted in situ, showed that a prerequisite to simultaneously promote P and Pb dissolution and pyromorphite formation was the prevalence of acidic conditions in the soil or other media (pH < 5). To meet this requirement, most studies employed acidic phosphate forms (primarily phosphoric acid and fertilizers such as monobasic calcium phosphate (MCP) and triple super phosphate (TSP)) and high P dosages, well in excess of the necessary stoichiometric ratio (P/Pb = 0.6) for pyromorphite formation (Chrysochoou et al., 2007). However, extended X-ray absorption fine structure (EXAFS) studies showed that the rate of pyromorphite formation was slow even under these conditions. Scheckel and Ryan (2004) reported that only 45% of the total Pb was transformed to pyromorphite after 32 mo of curing with excess phosphoric acid (1% wt. P, P/Pb 28). According to Ryan et al. (2004), transformation reactions ceased after 3 mo of curing, when lime (CaO) was applied to the treated plot to reverse soil acidification to neutral pH conditions.

Despite the slow rate of Pb conversion to pyromorphite, it may be argued that phosphate-based treatment is successful, as long as the soluble Pb concentrations remain under the regulatory limits (5 mg/L for the toxicity characteristic leaching procedure (TCLP), 15 µg/L for drinking water in the United States). This may be an acceptable philosophy as long as there are no adverse effects from the release of high P concentrations to the environment. Considering that P is the limiting micronutrient associated with eutrophication, it becomes apparent that soluble P concentrations and P leaching also should be minimized, if phosphate treatment is to be considered environmentally sustainable. The USEPA has set the P water quality criteria in surface waters in the range of 8 to 128 µg/L depending on the region (USEPA, 2001b). While P leaching has been mentioned as a potential issue by some studies (Cao et al., 2002; Basta and McGowen, 2004; Conca and Wright, 2006), including the USEPA BMP manual, there are very limited data that associate P release with the pH regime, the source and the amount of P used, and treatment success. Cao et al. (2002) reported that 20% of the total P added (phosphoric acid, P/Pb4) migrated vertically from an in situ treatment plot, but did not provide any P concentration data. Basta and McGowen (2004) treated a Pb, Zn, and Cd contaminated smelter soils with diammonium phosphate ((NH₄)₂HPO₄) and phosphate rock, with P addition as low as 10 g/kg (P/total metals 1/15) and a maximum of 180 g/kg (P/total metals 0.6). Even though the P dosages were very low compared to the majority of treatability studies encountered in the literature (Chrysochoou et al., 2007), and despite the high pH and soil Ca content, as much as 10% of P leached from the treated columns (Basta and McGowen, 2004). The P concentrations were initially very high (>2000 mg/L) and progressively declined to low, but indeterminable, values. Conca and Wright (2006) reported orthophosphate concentrations up to 50 mg/L in the effluent of a permeable reactive barrier using biogenic apatite. Given the elevated P leaching data from these reports, this study draws attention to the leaching potential of P applied at different dosages and geochemical conditions.

Accordingly, two firing range soils were chosen to illustrate the potential range of Pb and P interaction: an alkaline soil with a high Pb concentration, present predominantly as cerussite (PbCO₃), and an acidic soil with a lower Pb concentration and metallic Pb as the primary Pb source. More information with regard to the firing ranges and Pb speciation can be found in Dermatas et al. (2005, 2006) and Dermatas and Chrysochoou (2007). Dermatas et al. (2005) presented the results of a preliminary treatability study using a monobasic calcium phosphate (Ca(H₂PO₄)₂·H₂O or MCP), a soluble, acidic phosphate reagent; the TCLP results yielded relatively high Pb concentrations (>1.5 mg/L) for MCP addition up to 150 g/kg. However, lead hydrogen phosphate (PbHPO₄) surprisingly occurred in one XRPD pattern of the treated soil. The study was therefore repeated using MCP to more closely monitor Pb speciation, this time for two different soils and using a greater variety of tests, including on assessment of P leaching.

	Materials and Methods

TOP NOTES ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Summary and Conclusions REFERENCES

 
Soil Sampling and Characterization
Soil samples were collected from the surface (top 20 cm) of berms (soil banks behind targets that stop bullets) at two U.S. military shooting ranges: the indoor Picatinny Arsenal Technology Facility (PATF) and the outdoor Fort Dix Range 26 (FDR26), both located in New Jersey, USA. For study, the PATF soil samples consisted of the soil fraction passing through the No. 200 U.S. standard sieve (75 µm), and the FDR26 samples consisted of the fraction passing the No. 4 (2 mm) sieve. The PATF soil had an elevated fines content, accumulating Pb as cerussite, whereas the FDR26 soil was largely a clean sand, containing large bullet fragments removable by conventional sieving (Dermatas and Chrysochoou, 2007). The fines fraction of the PATF soil was used for stabilization purposes, as it was the most highly contaminated fraction, from which Pb could not be separated by physical (soil washing) remediation methods that were attempted for the coarser fractions. Soil pH was measured according to ASTM method D4980-89. The total organic and inorganic carbon contents (TOC, TIC) were determined using the Rock-Eval Method (Hetényi et al., 2005). Water-soluble anions were measured using the following procedure (Huerta et al., 2005): deionized (DI) water (10.0 g) was added to 0.5 g of pulverized sample in a plastic digestion tube. The mixture was kept closed and warmed using a hot plate at 85°C for 18 h, and then the tube was left for 1 h for settling purposes. The supernatant was filtered through a 0.45 µm Whatman polypropylene membrane filter and the soluble chloride, nitrate, and sulfate concentrations were measured using ion chromatography (IC 25 Ion Chromatograph, DIONEX, CA) with a 30 mmol KOH mobile phase and a 4 x 250 mm IonPac AS16 column with a detection limit of 0.01 mg/L. Major metals were analyzed by acid digestion (USEPA Method 3050B) and inductive coupled plasma/optical emission spectrometry (ICP/OES) using a Varian Vista-MPX (Varian, CA) spectrometer. Selected characteristics of the two soils are presented in Table 1 .

Batch experiments using TCLP were conducted following USEPA Method 1311 with minor modifications. Specifically, 3 g of treated material (air-dried for 24 h immediately after sampling) was placed in 60 mL HDPE bottles and mixed with 60 g of extraction fluid #1 (5.7 mL of glacial acetic acid in 500 mL reagent water per 64.3 mL of 1 mol/L sodium hydroxide solution, diluted to a volume of 1 L, pH = 4.93 ± 0.05). The selection of the leaching fluid was based on the pH and buffering capacity of the soil. The mixture was tumbled at 30 rpm for 18 h.

Batch SPLP experiments were conducted following USEPA Method 1312. SPLP and DI water extractions were essentially identical to the TCLP conditions except that the extraction fluids and tumbling times were different. The SPLP leaching fluid was a 60:40 w/w mixture of sulfuric and nitric acids (or a suitable dilution) with a pH of 4.20 ± 0.05. DI water was prepared by flowing tap water through a DIONEX system so that the conductivity was less than 0.006 mho/cm. Samples were tumbled at 30 rpm for 72 h (4x TCLP time requirement) to promote equilibrium conditions.

All leachates were filtered with a 0.45 µm Whatman polypropylene membrane filter and the filtrate pH was measured using a pH meter (Denver Instrument UB-10). The concentrations of metal cations including Pb and Ca were determined by using ICP–OES (Varian Vista-MPX, Varian, CA). Phosphate and chloride concentrations in the SPLP and deionized water (DIW) leachates were also measured with ion chromatography as previously described. All samples were run in duplicate.

Mineralogical Analyses
Untreated soils, selected treated samples, and filter residues of the leached samples were collected for mineralogical analysis by XRPD. Samples were air-dried for 24 h and manually pulverized to pass a No. 400 U.S. standard sieve (38 µm). Step-scanned XRPD data were collected by a Rigaku DXR 3000 diffractometer using Bragg-Brentano geometry. The diffractometry was conducted at 40 kV and 40 mA using diffracted beam graphite-monochromator with Cu radiation. The data were collected in the range of two-theta values between 5° to 65° with a step size of 0.02° and a count time of 3 s per step. The XRPD patterns were analyzed with the Jade software, Version 7.1 (MDI, 2005) and reference to the patterns of the International Centre for Diffraction Data database (ICDD, 2002) and the Inorganic Crystal Structure Database (Fachinformationszentrum, 2006).

Geochemical Modeling
The results for the TCLP and SPLP leaching tests for the two samples were modeled using Visual Minteq, version 2.51 (Gustafsson, 2004). Total metal analyses (and TIC data for PATF soil) from Table 1 were used as input, and the estimated soluble chloride concentration was corrected for the 20:1 liquid-to-solid ratio. Sulfate was not used in the input (constituent of the SPLP solution), as anglesite is generally significantly more soluble than lead phosphates and the precise SPLP concentration cannot be reliably estimated. Calcium and phosphate inputs were calculated based on the corresponding MCP dosages. The model was run at 25°C using the Debye-Hückel equation for ionic strength correction and with the pH held constant at the final values measured in the respective leachate.

The purpose of the modeling exercise was primarily to compare the theoretical soluble Pb concentration, as controlled by pyromorphite (the desired end product), with the actual concentrations. Sorption processes and the role of colloids and dissolved organic matter (DOM) were therefore not included in the model. Also, sorption is considered to be metastable with respect to pyromorphite formation, especially at the high P dosages applied in the experiment. It was therefore assumed that sorption would not be an important immobilization mechanism at thermodynamic equilibrium (at least, that is what the P-related literature advocates). Furthermore, the role of carbonates was investigated in the case of the FDR26 soil (no TIC data), but it was determined that the presence of carbonate (even in unrealistically high concentrations at that soil pH) did not substantially change the phase assemblages or dissolved Pb and P concentrations.

	Results and Discussion

TOP NOTES ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Summary and Conclusions REFERENCES

 
pH and Leaching Tests
pH
The dissolution of MCP can release H⁺ ions and promotes soil acidification, depending on the initial soil pH and soil buffering capacity. The amount of H⁺ released depends on the type of MCP dissolution (congruent or incongruent), which is a function of soil pH. If MCP dissolves congruently, which can occur below pH 4.7, it produces Ca²⁺ and H₂PO₄^– ions and no acidity is released. However, MCP dissolution is significantly more complex at higher pH values; phase equilibria predict incongruent dissolution to form hydroxyapatite (HA) or Ca₅(PO₄)₃(OH), but kinetics favor the formation of brushite (CaHPO₄·2H₂O) at room temperature (Martin and Brown, 1997). Both the formation of hydroxyapatite and brushite result in the production of acidity, the quantity of which depends on soil pH. Below pH6.3, MCP dissolution to form brushite can be written as:

[1]

Above pH6.3, HPO₄^2– prevails and the reaction results in the production of two protons per mol dissolved MCP and formed brushite. At equilibrium conditions, HA formation results in even higher release of acidity (Eq. [2]). Finally, the formation of pyromorphite at any pH or of other lead phosphates (PbHPO₄ or Pb₃(PO₄)₂) that form at acidic pH also lead to proton release from H₂PO₄^2–.

[2]

Because kinetics control the dissolution rate and the amount of dissolved MCP and the formation of calcium and lead phosphates, and because pH shifts are strongly dependent of the soil buffering capacity, it is difficult to theoretically predict soil pH changes on MCP addition. The experimental pH values for the two firing range soils are shown in Fig. 1 .

View larger version (12K): [in this window] [in a new window]   Fig. 1. Soil pH in PATF and FDR26 soils for 1 and 28 d of curing.

Lead Leaching
The post-tumbling pH and Pb concentrations in the TCLP, SPLP, and DIW tests of the PATF and FDR26 samples are shown in Fig. 2 . Only TCLP and SPLP data are shown for the FDR26 samples, as the DIW test results were practically identical with the SPLP data set.

View larger version (28K): [in this window] [in a new window]   Fig. 2. Leachate pH and dissolved Pb in the (a) PATF-TCLP, (b) PATF-SPLP, (c) PATF-DIW, (d) FDR26-TCLP, (e) FDR26-SPLP tests.

TCLP-Pb concentrations in both untreated soils were significantly higher than the USEPA regulatory criteria of 5 mg/L, at 483 mg/L and 218 mg/L for the PATF and FDR26 soils, respectively. The TCLP-Pb leachability ratio (TCLP-Pb concentration times the L/S ratio of 20, divided by the total Pb in the solid) was 27 and 87% for the PATF and FDR26 soils, respectively. As noted previously, Pb in the PATF soil fines occurred predominantly as cerussite, while metallic Pb was the prevalent species in FDR26. The difference between the two soils and their Pb speciation reflects the management practices of the two firing ranges: the PATF range featured the use of high velocity weapons (rifles) and water spraying for dust suppression, which caused extensive bullet fragmentation and rapid Pb weathering, exacerbated by the carbonaceous nature of the dredged sand used for berm construction (Dermatas and Chrysochoou, 2007). Cerussite accumulated in the fines fraction of the PATF soil, which was tested in this study. Conversely, the use of clean sand as berm material and the firing of lower velocity weapons in the FDR26 facility caused partial bullet fragmentation, with metallic Pb persisting as the primary Pb species, and cerussite only occurring as an outer surface crust on bullet fragments. Cerussite is largely insoluble at pH > 6 (Zhang and Ryan, 1999) which is consistent with the low TCLP-Pb leaching ratio of the untreated PATF soil even at a slightly lower TCLP pH (5.4). Metallic Pb in the untreated FDR26 soil leached rapidly due the low initial pH and the high acidity imparted by the TCLP test. Similar observations hold for the SPLP results of the untreated soils; the SPLP-Pb leachability ratio for PATF was <0.01% due to the high post-tumbling pH (8.5), while the respective ratio for the FDR26 soil was 1.3% (pH 6.2). Lower Pb leachability under SPLP conditions is expected, as SPLP was designed to simulate in situ leaching on infiltrating acid rain, imparting significantly less acidity compared to the TCLP solution. Thus, SPLP results are considered to be more relevant for the assessment of Pb leaching in firing range soils. DIW extraction results were found to be very close or identical to SPLP data for both soils, both in terms of pH and soluble Pb and P, in untreated and treated samples. The reason for the similar behavior is attributed to the acidity imparted through uptake of atmospheric CO₂, which equilibrates pure DIW to pH 5.7 (Snoeyink and Jenkins, 1980). Hence, DIW extraction is also deemed suitable for evaluation of in situ leaching conditions, especially in aerated environments.

The incremental addition of MCP to the PATF soil progressively decreased the TCLP-Pb concentration. The 25 and 50 g/kg MCP dosages proved inadequate to suppress TCLP-Pb concentration below 5 mg/L at 28-d curing, even though the MCP dosing exceeded the stoichiometric ratio for pyromorphite formation by a factor of 2x to 4x, respectively (see Table 2). The 75 g/kg MCP dose decreased TCLP-Pb to 3 mg/L, with nominal reductions thereafter, consistent with the observations of Dermatas et al. (2005). The fact that higher MCP dosages did not significantly improve treatment performance is indicative of the kinetically inhibited reactions in the PATF soil system; in other words, the excess phosphate remained largely unreacted (as brushite and potentially as unreacted MCP) during the 28-d testing period.

The behavior of soluble Pb in the SPLP and DIW tests of the treated PATF soil was, to some extent, different compared to TCLP. In both cases, the dissolved Pb concentration at 1-d curing was equal to or higher than the control, suggesting that the SPLP-pH decrease enabled Pb solubilization, while phosphate dissolution lagged. MCP performance improved by 28 d in both the SPLP and DIW tests, but remained above the USEPA SPLP regulatory criteria of 15 µg/L for all MCP dosages. Moreover, the increase in the MCP dosage did not significantly alter soluble Pb, nor did it yield substantially lower concentrations compared to the untreated soil. If pyromorphite was the solubility controlling phase for Pb in the treated sample, soluble Pb should have been significantly lower. According to Zhang and Ryan (1999), CP maintains Pb concentrations at <10 µg/L above pH 6, while cerussite yields concentrations up to two orders of magnitude higher. Cerussite persisted in the XRPD patterns of the PATF soil (see mineralogical analysis section), and no lead phosphate was identified, verifying that Pb solubility control was maintained by cerussite. Hence, Pb leaching in the PATF soil was bound to exceed the regulatory standards, as long as cerussite continued to control its dissolution. It is unknown, if and when pyromorphite will dominate Pb solubility control under the alkaline conditions of the PATF soil.

TCLP treatment performance was improved in the FDR26 soil (Fig. 2d). Approximately 15 g/kg MCP (P/Pb 5) was adequate to satisfy the TCLP-Pb criteria, and the concentrations decreased further with increased phosphate addition, reaching 0.3 mg/L at 45 g/kg MCP dosage. The SPLP-Pb results also yielded significant Pb reductions, which were enhanced with increasing MCP dosage. However, no dosing rate reduced SPLP-Pb below 15 µg/L after 28 d; 26 µg/L was the lowest concentration achieved by the 45 g/kg MCP dosage. The DIW extraction results were again essentially identical to the SPLP results, both in terms of post-tumbling pH and the corresponding soluble Pb concentrations (data not shown). Continued Pb reductions with increasing phosphate addition in all leaching tests indicate that MCP dissolution was enhanced in the FDR26 soil, as expected by the acidic pH regime. Consequently, reaction kinetics for lead phosphate formation were more favorable. The reason why the SPLP concentrations did not reach the desirable levels was probably that the pH was too low to optimize Pb solubility, even if pyromorphite (either HP or CP) did form. This issue will be discussed below (geochemical modeling). This has important implications for treatment design, as it indicates that restoration of soil pH to neutral values is necessary to ultimately suppress soluble Pb concentrations, after Pb-P reactions are essentially complete. Recent in situ studies using phosphoric acid as the phosphate source already considered this requirement by adding lime (CaO) to the treated soil (Ryan et al., 2004; Yang and Mosby, 2006). The timing of lime application is crucial for treatment success, as it essentially arrests further Pb-P interactions.

It should also be noted that the apparent efficacy of P-induced Pb removal is a function of the assessment test. The strong acidity imparted during the TCLP test initiates Pb-P reactions, i.e., MCP dissolution and lead phosphate precipitation. Conversely, the limited acidity of the SPLP and DIW tests prevents the attainment of the pH conditions (5) which favor optimal Pb-P interactions, suggesting reduced efficacy versus the TCLP results. The influence of the testing procedures on Pb speciation has previously been documented; Scheckel et al. (2003) reported that Pb and phosphate reacted during the sequential extraction test (SET) to form pyromorphite. Similarly, Dermatas et al. (2006) documented that the enhanced release or immobilization of Pb species during TCLP depended on the soil mineralogy, initial Pb speciation, and buffering capacity. Consequently, the outcomes of the leaching tests are not necessarily representative of the prevailing speciation in the treated solids or in situ. Nondestructive techniques, such as XRPD and EXAFS, are better suited to determine the actual Pb speciation in phosphate-treated samples.

Phosphorus Leaching
Inorganic P concentrations were measured in the SPLP leachates (Fig. 3 ), conditions taken to be more reflective of in situ conditions as compared to the TCLP. P concentrations in the DIW leachates were essentially identical with the SPLP data in both the PATF and FDR26 samples (data not shown).

View larger version (14K): [in this window] [in a new window]   Fig. 3. Dissolved P concentrations in the (a) PATF-SPLP, (b) FDR26-SPLP tests.

View this table: [in this window] [in a new window]   Table 4. Phosphate leachability ratio based on the synthetic precipitation leaching procedure (SPLP) test results.

Mineralogical Analyses
Figures 4a,b show the XRPD patterns of the untreated and 100 g/kg MCP-treated PATF soil at 28 d, along with the respective TCLP, SPLP, and DIW residues (Fig. 4c,d,e ). The mineralogy of the untreated PATF soil fines was composed of quartz (SiO₂), feldspars (albite (Ca,Na)(Si,Al)₄O₈), micas (muscovite (K,Na)(Al,Mg,Fe)₂(Si₃Al)O₁₀(OH)₂), and small amounts of calcite (CaCO₃) and cerussite (PbCO₃). The same phases persisted in all treated samples and the residues. The only new phase identified in the treated sample (Fig. 4b) was brushite (primary peak occurring at 211.4°), consistent with the incongruent MCP dissolution at alkaline pH. The brushite peak remained apparently unchanged in the SPLP and DIW residues (Fig. 4d,e), but was significantly reduced in the TCLP residue (Fig. 4c), consistent with the acidity of the TCLP test that enhances Pb and P dissolution and Pb-P precipitation reactions. The cerussite peak (primary peak at 224.8°) diminished in the treated samples and filter residues, reaching an apparent minimum in the TCLP residue.

View larger version (26K): [in this window] [in a new window]   Fig. 4. X-ray powder diffraction (XRPD) patterns of PATF soil, (a) untreated, (b) treated with 100 g/kg MCP, 28 d curing, (c) 100 g/kg MCP, 28 d, TCLP (d) SPLP, (e), and DIW residues. Key: Q: Quartz, F: Feldspar, Ce: Cerussite, Br: Brushite.

The XRPD analysis of the FDR26 soil revealed the presence of only quartz, muscovite and kaolinite (Al₂Si₂O₅(OH)₄); the low soil Pb concentration (5,000 mg/kg) did not allow the identification of any Pb minerals, and high quartz intensities obscured other peaks. Of all the treatments, a small metallic Pb peak (231.27° or d-spacing 28.6 nm) was identified in the 28-d sample cured with 15 g/kg MCP (Fig. 5 ). This may be the result of the individual specimen containing a metallic Pb fragment, whose dissolution was rate-limited. No other Pb or P minerals could be identified in any of the treatments. That said, the presence of metallic Pb nevertheless indicated that reactions were not 100% complete within the FDR26 soil, despite the favorable pH regime and high aqueous concentrations of P. The absence of brushite and other P minerals suggests near-complete MCP dissolution, even though they may occur below the XRPD detection limit (1–3% depending on the phase assemblage) or in the amorphous fraction.

View larger version (14K): [in this window] [in a new window]   Fig. 5. X-ray powder diffraction (XRPD) pattern of FDR26 soil treated with 15 g/kg MCP at 28 d curing.

View this table: [in this window] [in a new window]   Table 5. Geochemical modeling results for FDR26 soil (6 and 30 g/kg MCP).

Modeled and measured concentrations were also divergent for the SPLP data set. Modeling underestimated the Pb and P concentrations, and overestimated the soluble calcium and chloride concentrations. The predicted precipitates were CP, HA, and calcite, compounds inconsistent with the presence of brushite and cerussite in the XRPD patterns. These results were again supportive of the notion that the reactions in the PATF soil were kinetically limited. The chloride-free SPLP model failed to produce results due to high ionic strength. Overall, the agreement of the model and the experimental data was extremely poor, indicating that equilibrium calculations cannot predict the prevailing conditions in phosphate-amended alkaline soils for a period of 28 d, as the rate of Pb and P transformation reactions are apparently slow.

Convergence between the experimental and the modeled results improved for the simulations involving the FDR26 soil, indicating that the treated soil was closer to equilibrium conditions. Similar trends were obtained for all dosages, with the model underestimating the dissolved Pb concentration by approximately one order of magnitude. The Pb solubility-controlling phase for the TCLP conditions was predicted as PbHPO₄ with only trace amounts of CP, and the omission of chloride from the model (and thus CP from consideration) did not change the dissolved Pb concentration significantly. The modeled SPLP results for the FDR26 soil provided the closest agreement with the measured concentrations. The predicted Pb concentration was within the same order of magnitude, with the 30 g/kg MCP model providing the best comparison (30 µg/L predicted vs. 39 µg/L measured). It should be noted that the higher actual concentrations may be a result of processes such as colloidal or organically bound dissolved Pb, both of which were omitted from the model. The Ca concentrations were also in excellent agreement, indicating that HA potentially precipitated at the higher SPLP-pH. Overall, these results suggest that acidic conditions in the FDR26 soil promoted Pb and P reactions at a faster rate compared to the alkaline PATF soil, so that experimental results were close to those predicted at equilibrium.

The P_in concentrations were, surprisingly, overestimated by the model for all MCP dosages, suggesting that high P concentrations are not only a result of a transitional dissolution process, but are theoretically possible and in fact predicted at thermodynamic equilibrium.

Regardless of treatment performance from a Pb perspective, an important finding of this study was the significant leaching potential for P under all tested conditions. The SPLP and DIW phosphate concentrations were high for both soils and all dosages, ranging from 1 to 40 mg/L in the PATF soil and from 18 to 331 mg/L in the FDR26 soil at 28 d. High P leaching potential was predicted by thermodynamic equilibrium, and is a result of the excess phosphate dosages required for an adequate Pb fixation.

The differences between the PATF and the FDR26 SPLP modeling results confirm that pH plays an important role in this respect. At neutral and alkaline pH, P solubility is suppressed by the theoretical precipitation of HA at equilibrium. Under acidic conditions, P concentrations remain elevated despite precipitation of HA. The question then arises, how should P release to the environment be weighed against the treatment performance, i.e., is the goal of overall environmental protection achieved by treating Pb-contaminated soils with phosphate?

The two objectives of Pb immobilization and tolerable P leaching appear to be at odds. This study demonstrates that the following conditions are necessary for successful Pb immobilization by phosphate: (i) the soil has to be naturally acidic or poorly buffered to allow pre-acidification and promote Pb-P reaction kinetics; (ii) phosphate must be applied in dosages that greatly exceed the pyromorphite stoichiometric ratio to yield satisfactory Pb fixation and conversion rates; and (iii) the restoration of soil pH to neutral values is likely required once P application has been completed. This last issue points toward the need for pH adjustment, or more pragmatically, non-phosphate based or traditional stabilization/solidification (S/S) processes (Conner, 1990) which would obviate the need for the use of phosphate altogether (Dermatas and Meng, 1996, 2003).

	Summary and Conclusions

TOP NOTES ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Summary and Conclusions REFERENCES

 
Phosphate treatment of Pb-contaminated soils has been widely proposed and accepted as a successful way to immobilize Pb as pyromorphite. However, the efficacy and sustainability of P application remains unresolved: (i) the phosphate treatment of alkaline and acidic soils pose serious challenges for optimizing Pb-P interactions; and (ii) the risk of secondary eutrophication associated with the significant P leaching potential is unknown.

Challenges associated with Pb immobilization by P application were demonstrated using an alkaline firing range soil (PATF) having high Pb contamination (35,000 mg/kg Pb), and an acidic firing range soil (FDR26) with lower Pb concentrations (5,000 mg/kg Pb). An acidic phosphate salt, monobasic calcium phosphate (MCP) was used, recognizing that soil acidification is necessary to promote Pb-P reactions. The selected MCP dosages all exceeded the stoichiometric requirement for complete Pb transformation to pyromorphite (P/Pb0.6), corresponding to P/Pb molar ratios between 1.1 and 14.7.

Monitoring the pH of the two soils over 28 d curing showed that it is extremely difficult to decrease the pH of a well buffered soil (PATF). Conversely, the pH of the poorly buffered acidic soil (FDR26) decreased by one to two pH units on addition of only 45 g/kg MCP. In both cases, the end result proved to be undesirable, for different reasons.

The prevalence of alkaline conditions in the PATF soil caused a slow rate of conversion reactions on MCP addition. MCP dissolved incongruently to form brushite (Ca(HPO₄)·2H₂O), as predicted at pH > 5 and as evidenced by XRPD analyses. This implies that only limited acidity and phosphate were released by MCP dissolution, the remaining P bound in brushite. Furthermore, Pb dissolution from cerussite (PbCO₃) in the PATF soil was limited, as evidenced by its persistence in the treated soil and the residues at 28 d. Equilibrium-based geochemical modeling failed to predict TCLP and SPLP results for the PATF soil, both in terms of concentrations and phase assemblages, as reactions were kinetically controlled and thus far from equilibrium. It is concluded that Pb concentrations were not controlled by lead phosphate formation in the PATF soil for all applied dosages. This agrees with previous studies indicating that cerussite remained the solubility-controlling phase under alkaline conditions in the presence of hydroxyapatite.

MCP performed better in terms of TCLP results for the acidic FDR26 soil. The fast release of acidity by MCP was evidenced by the quick drop in soil pH, which reached 3.0 at 45 g/kg MCP. At this pH, MCP dissolved congruently, and no evidence of brushite formation was found above the XRPD detection limit (1–3%). The accelerated phosphate release caused a significant decrease in TCLP-Pb concentrations.

However, the SPLP-Pb concentrations exceeded 15 µg/L for all dosages. According to geochemical modeling, the soil (and SPLP) pH was too low to maintain soluble Pb levels below the drinking water standard, even though its solubility was controlled by lead phosphate. Metallic Pb, nevertheless, persisted in the XRPD pattern of the 15 g/kg MCP treated sample at 28 d, showing that Pb-P reactions were not necessarily complete. In general, the presence of metallic Pb fragments in firing range soils constitutes a problem for rapid transformation reactions.

Significant P leaching was observed under all tested conditions. Even though recent in situ studies were geared to satisfying these requirements, the fate of excess P has not been addressed in the field. Treatment design should leverage not only Pb, but also P immobilization, to avoid secondary contamination. Accordingly, the path forward appears to be to quantify P leaching and conduct speciation studies under in situ conditions.

	ACKNOWLEDGMENTS

 
This work was funded in part by U.S. Army Dod Tacom-Ardec, Range Safe Program, Picatinny Arsenal and the New Jersey Dep. of Environmental Protection (NJDEP). The statements and opinions expressed in this paper are those of the authors only; they do not necessarily represent the views of the sponsoring agencies.

	NOTES

TOP NOTES ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Summary and Conclusions REFERENCES

 
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher.

	REFERENCES

TOP NOTES ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Summary and Conclusions REFERENCES

 

• Basta, N.T., and S.L. McGowen. 2004. Evaluation of chemical immobilization treatments for reducing heavy metal transport in a smelter-contaminated soil. Environ. Pollut. 127:73–82.[CrossRef][Medline]
• Cao, X., L.Q. Ma, M. Chen, S.P. Singh, and W.G. Harris. 2002. Impacts of phosphate amendments on lead biogeochemistry at a contaminated site. Environ. Sci. Technol. 36:5296–5304.[Medline]
• Chrysochoou, M., D. Dermatas, and D.G. Grubb. 2007. Phosphate application to firing range soils for Pb immobilization: The unclear role of phosphate. J. Hazard. Mat. 144:1–14.[CrossRef][Web of Science][Medline]
• Conca, G., and J. Wright. 2006. An apatite II permeable reactive barrier to remediate groundwater containing Zn, Pb, and Cd. Appl. Geochem. 21:1288–1300.[CrossRef]
• Conner, J.R. 1990. Chemical fixation and solidification of hazardous wastes. Van Nostrand Reinhold, New York.
• Dermatas, D., and M. Chrysochoou. 2007. Lead particle size and its association with soil fractions in firing range soils: Implications for treatment. Environ. Geochem. Health 29:347–355.[CrossRef][Web of Science][Medline]
• Dermatas, D., and X. Meng. 1996. "Stabilization/solidification (S/S) of heavy metal contaminated soil by means of a quicklime based-treatement approach. p. 499–513. In T.M. Gilliam and C.C. Wiles (ed.) Stabilization and solidification of hazardous, radioactive, and mixed wastes. ASTM STP 1240, Vol. 3. ASTM, Philadelphia, PA.
• Dermatas, D., and X. Meng. 2003. Utilization of fly ash for stabilization/solidification of heavy metal contaminated soils. Eng. Geol. 70:377–394.[CrossRef]
• Dermatas, D., N. Menounou, M. Dadachov, P. Dutko, G. Shen, X. Xu, and V. Tsaneva. 2005. Lead leachability in firing range soils. Environ. Eng. Sci. 23(1):86–99.
• Dermatas, D., G. Shen, M. Chrysochoou, D.G. Grubb, N. Menounou, and P. Dutko. 2006. Pb speciation versus TCLP release in firing range soils. J. Hazard. Mat. 136:34–46.[CrossRef][Web of Science][Medline]
• Gustafsson, J.P. 2004. Visual MINTEQ version 2.51. KTH (Royal Inst. of Technology), Sweden.
• Fachinformationszentrum. 2006. Inorganic Crystal Structure Database. Fachinformationszentrum, Karlsruhe, Germany. Available at http://icsdweb.fiz-karlsruhe.de/ (verified 10 Oct. 2007).
• Hetényi, M., T. Nyilas, and T.M. Toth. 2005. Stepwise Rock-Eval pyrolysis as a tool for typing heterogeneous organic matter in soils. J. Anal. Appl. Pyrolysis 74:45–54.[CrossRef]
• Huerta, V.D., M.L.F. Sánchez, and A. Sanz-Medel. 2005. Qualitative and quantitative speciation analysis of water soluble selenium in three edible wild mushrooms species by liquid chromatography using post-column isotope dilution ICP–MS. Anal. Chim. Acta 538(1–2):99–105.
• International Centre for Diffraction Data. 2002. Powder Diffraction File, PDF-2 Database Release. International Centre for Diffraction Data, Newtown Square, PA.
• Martin, R.I., and P.W. Brown. 1997. Phase equilibria among acid calcium phosphates. J. Am. Ceram. Soc. 80:1263–1266.
• MDI. 2005. Jade Version 7.1. Material's Data Inc., California, USA.
• Nriagu, J.O. 1974. Lead orthophosphates-IV: Formation and stability in the environment. Geochim. Cosmochim. Acta 38:887–898.[CrossRef][Web of Science]
• Ryan, J.A., K.G. Scheckel, W.R. Berti, S.L. Brown, S.W. Casteel, R.L. Chaney, M. Doolan, P. Grevatt, J. Hallfrisch, M. Maddaloni, and D. Mosby. 2004. Reducing children's risk to soil Pb: Summary of a field experiment. Environ. Sci. Technol. 38:8A–24A.[Medline]
• Scheckel, K.G., C.A. Impellitteri, J.A. Ryan, and T. McEvoy. 2003. Assessment of a sequential extraction procedure for perturbed lead-contaminated samples with and without phosphorus amendments. Environ. Sci. Technol. 37:1892–1898.[Medline]
• Scheckel, K.G., and J.A. Ryan. 2004. Spectroscopic speciation and quantification of lead in phosphate-amended soils. J. Environ. Qual. 33:1288–1295.[Abstract/Free Full Text]
• Snoeyink, V.L., and D. Jenkins. 1980. Water chemistry. John Wiley & Sons, New York. p. 165.
• Yang, J., and D. Mosby. 2006. Field assessment of treatment efficacy by three methods of phosphoric acid application in lead-contaminated urban soil. Sci. Total Environ. 366:136–142.[Medline]
• Zhang, P., and J.A. Ryan. 1999. Transformation of Pb(II) from cerussite to chloropyromorphite in the presence of hydroxyapatite under varying conditions of pH. Environ. Sci. Technol. 33:625–630.
• USEPA. 2001a. Best management practices for lead at outdoor shooting ranges. EPA-902-B-01-001. USEPA, Washington, DC.
• USEPA. 2001b. Nutrient criteria development; Notice of ecoregional nutrient criteria. Federal Register, Vol. 66, No. 6. USEPA, Washington, DC.
• USGS. 2002. Lead in December 2001, Mineral industry survey. Minerals Information Publications Services, Reston, VA.

This article has been cited by other articles:

				R. A. Moseley, M. O. Barnett, M. A. Stewart, T. L. Mehlhorn, P. M. Jardine, M. Ginder-Vogel, and S. Fendorf Decreasing Lead Bioaccessibility in Industrial and Firing Range Soils with Phosphate-Based Amendments J. Environ. Qual., October 23, 2008; 37(6): 2116 - 2124. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Dermatas, D. Articles by Xu, X. Search for Related Content PubMed PubMed Citation Articles by Dermatas, D. Articles by Xu, X. GeoRef GeoRef Citation Agricola Articles by Dermatas, D. Articles by Xu, X. Related Collections Best Management Practices Heavy Metals Soil Pollution