Field assessment of lead immobilization in a contaminated soil after phosphate application

The Science of the Total Environment 305 (2003) 117–127

Field assessment of lead immobilization in a contaminated soil after phosphate application Ricardo Melameda, Xinde Caob, Ming Chenb, Lena Q. Mab,* a

Center for Mineral Technology, Ministry of Science and Technology, Av. Ipeˆ 900, Ilha da Cidade Universitaria, Rio de Janeiro, 21941-590, Brazil b Soil and Water Science Department, University of Florida, Gainesville, FL 32611-0290, USA Received 28 May 2002; accepted 15 September 2002

Abstract A pilot-scale field demonstration was conducted at a Pb-contaminated site to assess the effectiveness of Pb immobilization using P amendments. The test site was contaminated by past battery recycling activities, with average soil Pb concentration of 1.16%. Phosphate amendments were applied at a 4.0 molar ratio of PyPb with three treatments: T1, 100% P from H3PO4; T2, 50% from H3PO4q50% from Ca(H2PO4)2; and T3, 50% from H3PO4q 5% phosphate rock. Soil samples were collected and characterized 220 days after P application. Surface soil pH was reduced from 6.45 to 5.05 in T1, to 5.22 in T2, and to 5.71 in T3. Phosphate treatments effectively transformed up to 60% of total soil Pb from the non-residual fraction (sum of water soluble and exchangeable, carbonate, Fe–Mn oxide, and organic fractions) to the residual fraction relative to the control. In addition, P treatments reduced Toxicity Characteristic Leaching Procedure (TCLP) Pb from 82 mg ly1 to below EPA’s regulatory level of 5 mg ly1 in the surface soil. Scanning electron microscopy-energy dispersive X-ray elemental analysis and X-ray diffraction analysis indicated formation of insoluble chloropyromorphite wPb5 (PO4 )3 Clx mineral in the P-treated soils. Although H3 PO4 is necessary to dissolve meta-stable Pb in soil for further lead immobilization, it should be used with caution due to its potential secondary contamination. A mixture of H3PO4 and Ca(H2PO4 )2 or phosphate rock was effective in immobilizing Pb with minimum adverse impacts associated with pH reduction. 䊚 2002 Elsevier Science B.V. All rights reserved. Keywords: Lead immobilization; Phosphate amendment; Contaminated soil; Field demonstration

1. Introduction Accumulation of heavy metals in soils and their transport through the soil matrix are potential threats to human health, especially to children’s health by ingestion of Pb-contaminated soil (Hettiarachchi et al., 2000; Yang et al., 2001). In this *Corresponding author. Tel.: q1-352-392-1951; fax: q1352-392-3902. E-mail address: [email protected] (L.Q. Ma).

regard, metal bioavailability and mobility are two major concerns. Increasing awareness of the hazard that toxic elements can cause to the environment and to humans makes it necessary to remediate metal contaminated sites. Among the remediation technologies available for contaminated sites, in situ immobilization techniques are of particular interest because they are relatively more costeffective compared to conventional techniques, e.g.

0048-9697/03/$ - see front matter 䊚 2002 Elsevier Science B.V. All rights reserved. PII: S 0 0 4 8 - 9 6 9 7 Ž 0 2 . 0 0 4 6 9 - 2

118

R. Melamed et al. / The Science of the Total Environment 305 (2003) 117–127

excavation, and off-site disposal (Rabinowitz, 1993; Yang et al., 2001). Application of phosphate amendments to soils has been identified as a potentially effective in situ remediation technology (Cotter-Howells, 1996; Hettiarachchi et al., 2000; Ma et al., 1994; Ryan et al., 2001). These amendments are available in various forms, environmental-friendly, and simple to use. Phosphate has been shown to effectively immobilize Pb from aqueous solutions as well as various contaminated soils (Ma et al., 1993, 1994; Ma and Rao, 1997), to reduce plant Pb uptake (Laperche et al., 1997), and to mitigate acid mine drainage by coating the pyrite surface with FePO4 which hinders sulfide oxidation, reducing the transport of heavy metals (Evangelou, 1996). The mechanisms involved in P-induced Pb immobilization include ionic exchange and chemical precipitation. Several studies suggest that a potential retention mechanism of hydroxyapatite for Pb (Suzuki et al., 1984) is ionic exchange with Ca. However, Ma et al. (1993) demonstrated that Pb reacts with hydroxyapatite wCa5(PO4)3OHx in solution, forming stable pyromorphite-type minerals wPb5(PO4)3X; XsF, Cl, Br and OHx, suggesting dissolution of hydroxyapatite followed by precipitation of pyromorphite as the primary mechanism. As such, the solubility of P amendments dictates their effectiveness in Pb immobilization. While chemical precipitation of metals depends on the solubility products of the solids formed, metal sorption involves adsorption, surface precipitation or co-precipitation, and intra-particle diffusion. Melamed et al. (2000) reported that when phosphate rock is used to immobilize Pb at its natural pH 8.7, soluble P concentration is low, resulting in a relatively slow Pb immobilization. However, at pH 3.7, phosphate rock dissolves and Pb immobilization is instantaneous, forming a pyromorphite-type mineral. Although much knowledge about the mechanisms involved in immobilization of Pb using P amendments has been acquired, implementation of this technology in the field has been limited. Thus, a field demonstration of this technology, at a site heavily contaminated with Pb from battery recycling wastes, was undertaken.

The main objective of this pilot-scale field experiment was to evaluate the effectiveness of P amendments on in situ Pb immobilization. The special tasks were (1) to identify formation of less soluble pyromorphite-like minerals; (2) to determine Pb distribution after phosphate treatments; (3) to assess Pb leaching characteristics. This paper reports field data from a contaminated site which was determined on day 220 after P application in February 2000. Results from this experiment provide the evidence on the effectiveness of using P amendments to immobilize Pb in contaminated soil at field scale, narrowing the gap between field and laboratory experiments. 2. Materials and methods 2.1. Site description The contaminated site is located northwest of Jacksonville, Florida. It is a nearly-level area of approximately 4100 m2 (71.4=57.3 m) with surface runoff in a west to southwesterly direction. It was probably exposed to Pb contamination due to its use as battery recycling and as a salvage yard with discharge of waste oil during the 1940s. Based on previous reports, levels of total Pb and TCLP-Pb are elevated but restricted mostly to the surface horizon (Cao et al., 2001). The site is dominated by disturbed urban soils lacking structure. The soil in areas that have not been filled or otherwise disturbed is classified as a Spodosol (sandy siliceous thermic ultic alaquod). Selected chemical and physical properties of composite soil samples from the site are given in Table 1. The soil is neutral, and heavily contaminated with Pb. Mineralogical characterization by XRD revealed the presence of CaCO3 and PbCO3 (cerussite) in the coarse non-magnetic soil fractions () 1 mm), which is related to the slightly alkaline conditions of the site (Cao et al., 2001). 2.2. Experimental plot establishment In our preliminary laboratory experiments this field soils were subject to Pb immobilization optimization by testing P sources and application rates using both batch and column experiments

R. Melamed et al. / The Science of the Total Environment 305 (2003) 117–127

119

Table 1 Selected physical and chemical properties of the soila PHb

CECc OMd Sand Silt Clay cmol kgy1 ...................................%...................................

PeT PbT CuT ZnT ................................g kgy1................................

Soil 6.95"0.19 5.75"0.85 3.91"0.90 87.7"1.37 9.0"1.58 3.35"0.54 0.89"0.16 11.6"1.57 2.64"0.11 1.95"0.83 a

Data represent an average of twelve replicates with a standard deviation; b pH was determined with a 1:1 ratio of soilywater; Cation exchange capacity; dOrganic matter; eTotal concentration.

c

(Cao et al., 2001). Sources of P included soluble KH2PO4, NH4H2PO4, H3PO4 and less soluble Ca(H2PO4)2, as well as phosphatic clay and phosphate rock. Results of these initial screening tests revealed that a mixture of Ca(H2PO4)2qCaCl2q H3PO4 displayed high immobilization efficiency. The results also demonstrated that application of phosphate rock or phosphatic clay alone was effective for Pb immobilization only at the low concentration range of Pb due to the limitation of the high P application rate required. Phosphoric acid was the most effective in reducing Pb extracted using Toxicity Characteristic Leaching Procedure (TCLP) to below regulatory levels, which was attributed to its effectiveness in enhancing the solubility of meta-stable cerussite (PbCO3), the main form of Pb at the site, and allowing the precipitation of pyromorphite, a geochemically more stable phase (Cao et al., 2001). The drawback of adding H3PO4 may take a higher risk of potential eutrophication due to its high mobility. The use of soluble H3PO4 in combination with Ca(H2PO4)2 or with phosphate rock was thus a rational step to take. H3PO4 addition would fulfill the need to solubilize cerussite, precipitating the readily available Pb, while the phosphate rock would supply a slow continuous source of phosphate ions, while minimizing the decrease in soil pH. Further studies, focusing on P application rates, indicated that among various molar ratio of PyPb tested, 4.0 was sufficient to achieve the goals for this site (Cao et al., 2001). The experimental plots were established in the highly contaminated zone of the site. Each plot was an approximately area of 4 m2, which were circled by high-density polyethylene geomembrane liner of 2.5 mm in thickness to prevent flooding out of or into the plots. The plots were separated by 1.5 m in distance from each other to avoid

possible inter-plot contamination. Phosphate amendments were applied to three plots at a molar ratio of 4.0 PyPb with three treatments. The total amount of P added was calculated for the surface soil of 0–20 cm depth. To pre-acidify the soil, half of the amount of P was initially applied to the three plots on 17 Febuary 2000 as a CaCl2q H3PO4 mixture in 25 l of water and sprayed uniformly in each area. The plots were then covered with a plastic sheet to maintain moisture contents in the surface layers and to prevent leaching from rainfall. The addition of equivalent amount of CaCl2 was to provide adequate Cl needed for the formation of least soluble chloropyromorphite wPb5(PO4)3Clx. On 27 March 2000, 40 days after the first application, the second half of the P amendments was applied as H3PO4 in T1, Ca(H2PO4)2 in T2 and 5% phosphate rock in T3, and mixed to a depth of 20 cm by a shovel. Please note, 5% phosphate rock was mixed with soil in T3 instead of 50% of the remaining P. Additional phosphate was added as phosphate rock to compensate for its low solubility with a PyPb molar ratio of 4.5 in T3 treatment. The plot without P addition was set as the control (T0). After the second P application, the plots were exposed to air. 2.3. Sampling procedure Three composite soil samples were collected from each of the plots 220 days after first P application, using a soil probe of 2-cm diameter at 6 depths (0–10, 10–20, 20–30, 30–40, 40–60 and 60–80 cm). After air-dried, soils were passed through a 2-mm sieve. Subsoil samples were digested with HNO3 yH2O2 hot block digestion procedure (USEPA Method 3050a).

120

R. Melamed et al. / The Science of the Total Environment 305 (2003) 117–127

2.4. Soil Pb fractionation and TCLP test Sequential extraction was performed on soil samples using the methodology of Tessier et al. (1999). The extractions were carried out in 40 ml centrifuge tubes with 1 g of soil. The procedure separated the lead into five operationally-defined fractions: water soluble and exchangeable (WE), carbonate bound (CB), Fe–Mn oxides bound (FM), organic matter (OM) and residual (RS). A reference soil material SRM 2710 (NIST, Gaithersburg, MD) was used to compare Pb recovery based on sequential extraction with certified values. Lead recovery was satisfactory with 103"10%. The Toxicity Characteristics Leaching Procedure (TCLP) was used to evaluate the efficiency of P amendments on lead toxicity (USEPA, 1992). 2.5. Chemical analysis Lead was analyzed by an atomic absorption spectrophotometer equipped with a graphite furnace (Perkin Elmer SAMMA 6000, Norwalk, CT). Other elements were analyzed with a multi-channel inductively coupled plasma spectrophotometer (Thermo Jarrel Ash ICAP 61-E, Franklin, MA). Total P was measured colorimetrically with a Shimadzu 160 U spectrometer using the molybdate ascorbic acid method (Olsen and Sommers, 1982). Soil pH was measured with a 1:1 ratio of solid to DI water. Soil organic matter was determined using the Walkley–Black procedure (Nelson and Sommers, 1982). Cation exchange capacity (CEC) was determined using the method of Rhoades (1982). 2.6. X-ray diffraction (XRD) analysis The clay fraction (-2 mm) of selected soil samples were separated by centrifugation and were analyzed using XRD. Samples were scanned from 2 to 608 2u with CuKa radiation on a computercontrolled diffractometer (Philips Electronic Instruments, Inc., Mahwah, NJ) equipped with stepping motor and graphite crystal monochromator.

2.7. Scanning electron microscopy and energy dispersive X-ray analysis Selected soil clay samples analyzed by XRD were further examined using a scanning electron microscope (SEM, JSM-6400yTN500, JEOL, USA), equipped with energy dispersive X-ray elemental spectrometry (EDX). Air-dried soil day samples were mounted on carbon stubs and then carbon coated. 3. Results and discussion 3.1. Lead concentrations in soil profiles Due to the extreme heterogeneity of the contaminated site, it is difficult to compare the temporal changes in soil Pb concentrations for different treatments. This was apparent since Pb concentrations in all four experimental plots were different. Furthermore, Pb concentrations varied significantly even within the same experimental plot depending on the precise location of soil sampling. Total Pb concentrations in the surface soil (0–10 cm) were 10 907, 5965, 15 919 and 3762 mg kgy1 for T0, T1, T2 and T3, respectively (Fig. 1a). In general, Pb concentrations were the greatest in the subsurface at a depth of 10–20 cm in all soil profiles, reaching concentrations of 3.1, 1.7, 2.2 and 1.2% in T0, T1, T2 and T3, respectively (Fig. 1a). Elevated Pb concentrations in the subsurface of the control soil may suggest downward migration of Pb in the soil profile. It is possible that after approximately 60 years, lead in the soil was leached to a depth of 10–20 cm. A five-year study of land application of municipal sludge to a forest soil has shown that most metal movement seems to be limited to the upper 5 cm of soil; however, repeated applications in the following year increased metal leaching to the underlying soil (Harris and Urie, 1986). In most contaminated soils, metals do not appear to leach downward in significant quantities in the short run, because of their strong interactions with the soil. However, in

R. Melamed et al. / The Science of the Total Environment 305 (2003) 117–127

121

Fig. 1. Distributions of Pb concentration (a), P concentration (b), and pH (c), in the profile soils (0–80 cm) taken 220 days after first P application. T0, the control; T1, H3PO4 alone; T2, H3PO4qCa(H2PO4), and T3, H3PO4qphosphate rock.

the long run, metals can leach downwards in a soil due to their complexation with solubilized organic matter especially in an alkaline environment where organic matter is more soluble (Mar-

schner and Wilczynski, 1991). This may be true at the demonstration site where soil organic matter (3.91%) and pH (6.95) are much higher than those of typical Florida soils (Chen et al., 1999). More

122

R. Melamed et al. / The Science of the Total Environment 305 (2003) 117–127

than 2000 mg Pbykg was detected in 30–40 cm deep soils of T1 and T2 plots. Therefore, this demonstration site was confronted with both spatial and vertical Pb contamination. It could also be seen from Fig. 1a that a high heterogeneity for Pb distribution occurred at this site, which may limit the effective evaluation of P treatments. However, this limitation could be overcome by using the same molar ratio of 4.0 PyPb during P application. 3.2. P concentrations in soil profiles Since the treatment plots were exposed to air, soil P enrichment and leaching may be an environmental concern for in situ Pb immobilization using P amendments. Soluble P sources (e.g. H3PO4, KH2PO4) pose a high risk of enhanced P eutrophication, while less soluble P sources (e.g. apatite, phosphate rock) pose less. Hettiarachchi et al. (2001) reported that Bray-1 extracted much more P in P-treated soils using triple super phosphate than in soils using phosphate rock. Therefore, it is important to evaluate P distribution when phosphate was used to immobilize Pb in a field test from the viewpoint of potential secondary contamination of P and P utilization efficiency. As expected, most P applied was concentrated in the surface soil (0–10 cm; Fig. 1b). The P retained at the surface (0–10 cm) was 45.1, 54.3 and 73.5% of the total added P for T1, T2 and T3, respectively. Downward migration of P was observed in this soil with low buffer capacity, which should be favorable for Pb immobilization occurrence in the subsurface soil. Generally, P concentrations declined sharply with depth, with less P moving down the soil profile in T3 than in T1 and T2. Phosphorus retained in the whole profile (0–80 cm) accounted for 86.3%, 88.5% and 94.2% of the total P applied for T1, T2 and T3, respectively. The fact that less P moved down the soil profile and less P was lost from T3 than the other two treatments suggested that T3 posed the least eutrophication risk among the three treatments and it provided more P to react with Pb in the soil.

3.3. pH in soil profiles Application of P amendments significantly impacted the pH of this sandy soil with relatively low buffer capacity and CEC (5.75 cmolykg) (Table 1 and Fig. 1c). Among the three treatments, T1 promoted the greatest decrease in soil pH at the surface, and T3 the least. The surface soil pH decreased from 6.45 (T0), to 5.71 (T3), to 5.22 (T2) and to 5.05 (T1), with -1.5 pH unit reduction. For typical Florida soils, an average pH of 5.04 is expected (Chen et al., 1999). The least reduction of pH in T3 plot was consistent with the results of Hettiarachchi et al. (2001), whereby H3PO4 reduced soil pH from 7.0 to 5.2, while phosphate rock had little effect even at the high level of P (2500 mgPykg) on five metal-contaminated soils and mine wastes. Furthermore, Pinduced pH reduction is mostly limited to the top 30 cm of soil, an indication of limited movement of P in the soil profile. At subsurface depths down to 30 cm, soil pH increased with depth (Fig. 1c). However, soil pH decreased with depth from 30 to 80 cm in all treatments, including the control soil, ranging from 4.9 to 5.36, which is a typical pH range for Florida soils. Reduction of soil pH was expected with addition of H3PO4. It has been reported that inducing soil acidic conditions will promote the solubility of Pb compounds, leading to effective Pb immobilization via formation of Pb pyromorphite (Yang et al., 2001). A field scale study suggested that adequate triple super phosphate (32 gykg) should be applied in order to reduce soil pH to levels that allowed effective reduction of bioavailable Pb (Brown et al., 1999). Dissolution of the initial Pb phase has been reported to be the limited factor in the formation of pyromorphite at pH values between 5 and 8 and conversion of PbO to pyromorphite was found to be most rapid at pH 5 (Laperche et al., 1996). However, care should be exercised when applying H3PO4 due to enhanced mobility of P and other heavy metals (Cao et al., 2001). 3.4. Formation of pyromorphite-like mineral Formation of a chloropyromorphite wPb5(PO4)3Clx mineral in the lead contaminated

R. Melamed et al. / The Science of the Total Environment 305 (2003) 117–127

123

Fig. 2. X-ray diffraction patterns of the surface soils (0–10 cm) in the control and P-treated plots on 220 days after first P application. T0, the control; T1, H3PO4 alone; T2, H3PO4qCa(H2PO4 ), and T3, H3PO4 qphosphate rock. CS, Cerussite; ClP, Chloropyromorphite.

soil after P application was confirmed by XRD data (Fig. 2). The chloropyromorphite mineral was observed after 220 days of P application for T1 and T2 treatments, but not in T3 treatment, as ˚ (Fig. 2). indicated by no peak at 2.96 and 2.86 A However, soil samples taken at later dates (330 and 480 days later) showed formation of chloropyromorphite in soil samples taken from T3 (Cao et al., 2001). Also, SEM element maps documented a Pb–P association in both surface (0–10 cm; Fig. 3a) and subsurface (30–40 cm; Fig. 3b) samples of T3. The EDS analysis showed that the Pb particles contained significant amount of Ca, P, Pb, Fe, Cl, Cu and Zn (data not shown), indicating the occurrence of chloropyomorphite. Similarly, SEM-EDS analysis of soils from T1 and T2 showed association of Pb with P. Coupled with formation of chloropyromorphite, H3PO4-induced dissolution of cerrusite (PbCO3), the main form of Pb at the site (Fig. 2), was also evident as indicated by the disappearance of the ˚ for soil samples taken from T1, peaks at 3.56 A

T2 and T3 (Fig. 2). Our data are consistent with the hypothesis that P-induced Pb immobilization was mainly through a dissolution-precipitation mechanism (Ma et al., 1993). 3.5. Fractionation of lead in soil profiles Chemical fractionation was carried out to understand lead transformations in the soil profile induced by P-treatment. Chemical fractionation has been extensively used to assess diagenetic processes, mobility and bioavailability of heavy metals in soils. Although the geochemical phases in sequential extraction schemes are operationally defined by the reagents used, it is of general consensus that the water soluble and exchangeable, carbonate bound, Fe–Mn oxides and organic bound phases (together as non-residual) are more bioavailable than the residual phase (Ma and Rao, 1997). Lead fractionation from the control soil at six different depths indicated that Pb was primarily

124

R. Melamed et al. / The Science of the Total Environment 305 (2003) 117–127

associated with the carbonate fraction in the leadcontaminated soil (50–70%, Fig. 4), which is consistent with the XRD data (Fig. 2). The presence of cerussite may be attributed to the fact that the soil had relatively high pH (6.95) (Table 1). All P treatments were able to modify the partitioning of Pb from the non-residualypotentially available fraction to the residualyless-available fraction (Fig. 4). In this respect, the major transformation was a decrease of the carbonate-bound Pb (up to 40%), while the residual fraction increased significantly (up to 60%) relative to the control, at the surface soil. This was consistent with the designed strategy of dissolving cerussite with phosphoric acid, and the precipitation of Pb as chloropyromorphite (Fig. 2). Previous investigations indicated a significant reduction of soil Pb in exchangeable fraction and increase in residual fraction upon P addition (Berti and Cunningham, 1997; Ryan et al., 2001). The increase of Pb in the residual fraction result from formation of chloropyromorphite. Ma and Rao (1997) reported that 99.9% of Pb pyromorphite was associated with the residual fraction. All the non-residual extractants are ineffective to dissolve Pb from pyromorphite. Fe–Mn oxide bound Pb also decreased (;10%) at the soil surface after P applications. This decrease of Pb in the Fe–Mn oxide bound fraction may be attributed to the dissolution of the oxides and release of sorbed Pb, as a consequence of low soil pH (Hayes and Katz, 1996) caused by phosphoric acid additions, the precipitation of P with soluble Pb, andyor to chloropyromorphite formation from Pb adsorbed on oxide surfaces (Zhang et al., 1997). The efficiency of P treatments to immobilize Pb generally decreased with depth, reflected by the progressive decrease of the residual fraction. However, this fraction was still enhanced up to 20%, at a soil depth of 30–40 cm, compared to the control (Fig. 4). The total Pb in the soil profile was highest at a depth of 10–20 cm in all plots

Fig. 3.

Fig. 3. Elemental maps of a surface (0–10 cm) soil clay sample (a) and a subsurface soil clay samples at depth of 30–40 cm (b) collected from the T2 plot treated with H3PO4q Ca(H2PO4), using scanning electron microscopy coupled with energy dispersive X-ray analysis.

R. Melamed et al. / The Science of the Total Environment 305 (2003) 117–127

125

Fig. 4. Lead fractionation in the soil profiles (0–80 cm) collected 220 days after first P application. T0, the control; T1, H3PO4 alone; T2, H3PO4qCa(H2 PO4 ), and T3, H3 PO4 qphosphate rock. WE, water-soluble and exchangeable; CB, carbonate; FM, Fe– Mn oxide; OM, organic matter; RS, residual.

(Fig. 1a). Thus, the consumption of P in the surface layer (0–10 cm) may have decreased efficiency of Pb immobilization at 10–20 cm. In fact, levels of P were much higher at the surface than in the 10–20 cm layer (Fig. 1b). The proportion of exchangeable Pb and Fe–Mn oxide bound Pb increased with soil depth (Fig. 4). The increase in exchangeable Pb was from approximately 5% at the surface to 20% at 60–80 cm in all plots. Fe–Mn oxide bound Pb increased from approximately 15% at the surface to 30% at 60– 80 cm in T1, while this fraction actually increased by approximately 5% in T0. In general, these changes could result from decreased pH in the treated plots (Fig. 1c), promoting Pb leaching and P attenuation. However, the complexity of the system does not allow definite conclusions from the field data. Although metals associated with Fe–Mn oxides are generally considered as bioavailable, we emphasize that this fraction plays a significant role in the attenuation of metal leaching (Hayes and Katz, 1996). As the metal adsorption phenomena become important at lower depths, the system pH becomes critical. The adsorption of metals on oxides decreases as the system pH is decreased, producing a sigmoidal function (Sposito, 1984).

3.6. TCLP lead in soil profiles Without P treatments (T0), TCLP-extractable Pb concentrations in the surface soils (0–10 cm) far exceeded 5 mgyl critical level of hazardous waste (USEPA, 1995). Similar to the distribution of total Pb, the highest concentration of TCLPextractable Pb was observed at 10–20 cm. Even at the 20–30 cm profile, TCLP-Pb still exceeded the critical level. This is possibly because most of the Pb was within the carbonate fraction (Fig. 4), which would readily dissolve in the acidic TCLPsolution (Berti and Cunningham, 1997). Phosphate amendment was effective in reducing the TCLP Pb to below the critical level in the surface soil samples (Fig. 5). These results are of great significance with respect to the disposal of the soil, because they show that P amendments can amend the soil to a material that would be considered non-hazardous. Although P treatments did reduce TCLP-Pb within the subsurface soils (10–20 cm), TCLP-extractable Pb concentrations in T1 and T3 were still higher (up to 60 mgyl) than the critical level of 5 mgyl except in T2. This may be due to less soluble P available for Pb to form adequate amount of pyromorphite (Fig. 1b).

126

R. Melamed et al. / The Science of the Total Environment 305 (2003) 117–127

Fig. 5. Lead concentrations in the soil profiles using Toxicity Characteristic Leaching Procedure 220 days after first P application. T0, the control; T1, H3PO4 alone; T2, H3PO4qCa(H2PO4), and T3, H3PO4qphosphate rock.

4. Conclusions The efficiency of in situ P-induced Pb immobilization technology is site specific and depends primarily on the nature and extent of the contamination and on the type of soil at the site. The type and rate of the P amendment, and the appropriate application management to be utilized require careful scrutiny, and as such, preliminary laboratory studies to assess the mechanisms involved with P applications and ultimately their efficiencies are of fundamental importance. The results of the field pilot-scale study, at this particular site, indicate that P amendments were efficient in transforming more bioavailable Pb (non-residual) into a less-bioavailable form (residual). The P-induced formation of pyromorphite in the field was evidenced by both XRD and SEMEDX data. Although H3PO4 is needed to catalyze the dissolution of meta-stable Pb, making it available for further immobilization reactions, its use should be taken with caution. Phosphoric acid decreased soil pH, especially for low-buffering sandy soils, and consequently may cause leaching of heavy metals. Thus, low pH and other heavy metals leaching may be potential drawbacks of its indiscriminate utilization. On the other hand, a mixture of H3PO4 and calcium phosphate or rock phosphate had excellent efficiencies, and both

treatments had less impact on soil pH. A strategy, which could work better than the one used in this study, would be to invert the sequences of P application, i.e. to add calcium phosphate and phosphate rock first and then apply the phosphoric acid, thus producing the dissolution of cerussite and the more soluble P amendments at the same time. Effective remediation technology entails minimizing both leaching and bioavailability. Acknowledgments This project was supported in part by the Florida Institute for Phosphate Research (Contract 97-01148R) and the Ministry of Science and Technology of Brazil. The authors are very thankful to Mr Thomas Luongo for his help in sample analysis. References Berti WR, Cunningham SD. In-place inactivation of Pb in Pbcontaminated soils. Environ Sci Technol 1997;31:1359 – 1364. Brown SL, Chaney R, Berti B. Field test of amendments to reduce the in situ availability of soil lead. In: Wenzel WW, editor. The 5th International Conference on the Biogeochemistry of Trace ElementsVienna, Austria, 1999. p. 506 –507. Cao X, Ma LQ, Singh SP, Chen M, Harris WG, Kizza P. Field demonstration of metal imobilization in containated soils using phosphate amendments. Gainesville, FL: Florida Institute of Phosphate Research, 2001.

R. Melamed et al. / The Science of the Total Environment 305 (2003) 117–127 Chen M, Ma LQ, Harris WG, Hornsby A. Background concentrations of 15 trace metals in Florida soils. J Environ Qual 1999;28:1173 –1181. Cotter-Howells J. Lead phosphate formation in soils. Environ Pollut 1996;93:9 –16. Evangelou VP. Pyrite oxidation inhibition in coal waste by PO4 and H2O2 pH-buffered pretreatment. Int J Surf Min Reclamation Environ 1996;10:135 –142. Harris AR, Urie DH. Heavy metal storage in soils of an aspen forest fertilized with municipal sludge. In: Co DW, Henry CL, Nutter WL, editors. The forest alternative for treatment and utilization of municipal and industrial wastes. Seattle, WA: University of Washington Press, 1986. p. 168 –176. Hayes KF, Katz LE. Application of X-ray absorption spectroscopy for surface complexation modeling of metal ion sorption. In: Brady PV, editor. Physics and Chemistry of Mineral Surfaces. CRC Press, 1996. Hettiarachchi GM, Pierzynski GM, Ransom MD. In situ stabilization of soil sead using phosphorous and manganese oxide. Environ Sci Technol 2000;34:4614 –4619. Hettiarachchi GM, Pierzynski GM, Ransom MD. In situ stabilization of soil lead using phosphorus. J Environ Qual 2001;30:1214 –1221. Laperche V, Logan TJ, Gaddam P, Traina S. Effect of apatite amendments on plant uptake of lead from contaminated soil. Environ Sci Technol 1997;31:2745 –2753. Laperche V, Traina SJ, Gaddam P, Logan TJ. Chemical mineralogical characterization of Pb in a contaminated soil: reactions with synthetic apatite. Environ Sci Technol 1996;30:3321 –3326. Ma LQ, Rao GN. Effects of phosphate rock on sequential chemical extraction of lead in contaminated soils. J Environ Qual 1997;26:788 –794. Ma QY, Traina SJ, Logan TJ. In situ lead immobilization by apatite. Environ Sci Tech 1993;27:1803 –1810. Ma QY, Traina SJ, Logan TJ, Ryan JA. Effects of aqueous Al, Cd, Cu, Fe(II), Ni and Zn on Pb immobilization by hydroxyapatite. Environ Sci Technol 1994;28:1219 –1228. Marschner B, Wilczynski AM. The effect of liming on quantity and chemical composition of soil organic matter in a pine forest in Berlin, Germany. Plant Soil 1991;137:229 –236. Melamed R, Self P, Smart R. Kinetics and spectroscopy of Pb immobilization on rock phosphate at constant pH. In: Singhal RK, Mehrotra AK, editors. Environmental Issues and

127

Management of Waste in Energy and Mineral Production. Calgary, Alberta, Canada: A.A. BalkemayRotterdamyBrookfield, 2000. p. 301 –306. Nelson DW, Sommers LE. In: Page AL, Miller RH, Keeney DR, editors. Total Carbon, Organic Carbon, and Organic Matter. In Methods of Soil Analysis, Part 2 chemical and microbiological properties, vol. 9. Madison, Wisconsin: ASA, 1982. p. 539 –577. Olsen SR, Sommers LE. Phosphorus. In: Page AL, Miller RH, Keene DR, editors. Methods of Soil Analysis, Part 2 chemical and microbiological properties, vol. 9. Madison, Wisconsin: ASA, 1982. p. 403 –427. Rabinowitz MB. Modifying soil lead bioavailability by phosphate addition. Bull Environ Contam Toxicol 1993;51:438 – 444. Rhoades JD. Cation exchange capacity. In: Page AL, Miller RH, Keeney DR, editors. Methods of Soil Analysis, Part 2 chemical and microbiological properties, vol. 9. Madison, Wisconsin USA: ASA, 1982. p. 149 –157. Ryan JA, Zhang P, Hesterberg D, Chou J, Sayers DE. Formation of chloropyromorphite in a lead-contaminated soil amended with hydroxyapatite. Environ Sci Technol 2001;35:3798 –3803. Sposito G. The surface chemistry of soils. New York: Oxford University Press, 1984. Suzuki T, Ishigaki K, Miyake M. Synthetic Hydroxyapatites as Inorganic Cation Exchangers. J Chem Soc Farady Trans 1984;80:3157 –3165. Tessier A, Campbell PGC, Bisson M. Sequential extraction procedure for the speciations of particular trace metals. Anal Chem 1999;51:844 –851. USEPA 1992. Toxicity characteristic leaching procedure (method 1311, rec.0). July 1992. In Test methods for evaluating solid waste, physicalychemical methods (i. SW846, ed.). Office of Solid Waste, Washington, DC. USEPA 1995, Test methods for evaluation of solid waste. Vol. IA, Laboratory mannaul physicalychemical methods, SW 846, 3ed. 40CFR Part 403 and 503. US Gov. Print. Office, Washington, DC. Yang J, Mosby DE, Casteel SW, Blancher RW. Lead immobilization using phosphoric acid in a smelter-contaminated urban soil. Environ Sci Technol 2001;35:3553 –3559. Zhang P, Ryan JA, Bryndzia LT. Pyromophite formation from goethite adsorbed lead. Environ Sci Technol 1997;31:2673 – 2678.