Kinetic Interactions of EDDS with Soils. 1. Metal Resorption and

Jan 6, 2009 - Kinetic Interactions of EDDS with Soils. 1. Metal Resorption and Competition under EDDS Deficiency. Theo C. M. Yip, Daniel C. W. Tsang*,...
0 downloads 4 Views 821KB Size
Download PDF
Environ. Sci. Technol. 2009, 43, 831–836

Kinetic Interactions of EDDS with Soils. 1. Metal Resorption and Competition under EDDS Deficiency THEO C. M. YIP,† D A N I E L C . W . T S A N G , * ,‡,§ KELVIN T. W. NG,† AND IRENE M. C. LO† Department of Civil and Environmental Engineering, The Hong Kong University of Science and Technology, Hong Kong, China, and Department of Civil and Natural Resources Engineering, University of Canterbury, New Zealand

Received January 16, 2008. Revised manuscript received November 10, 2008. Accepted November 18, 2008.

Biodegradable EDDS ([S,S]-ethylenediaminedisuccinic acid) is an emerging chelant for enhancing heavy metal extraction. During soil remediation that involves continuous flushing, metal extraction is often limited by the amount of EDDS. Under EDDS deficiency, initial extraction of Zn and Pb followed by resorption was observed in batch kinetic experiments. Speciation calculations indicated that the percentages of ZnEDDS2- and PbEDDS2- in respective dissolved metal concentrations decreased with time, whereas the contribution of CuEDDS2- to total EDDS increased accordingly. This pointed to the metal exchange of newly formed ZnEDDS2- and PbEDDS2- with sorbed Cu on the soil surfaces, rather than with Fe oxides. A portion of displaced Zn and Pb was resorbed on the exchangeable and carbonate fractions, whereas the rest was mainly bound to dissolved organic matter (DOM) and remained in solution. On the other hand, although dissolved Al was the major mineral cation in solution under EDDS deficiency, the resulting competitive effect on metal extraction was marginal because Al readily dissociated from EDDS complexes and predominantly existed as colloidal precipitates, DOM-complexes, or hydrolyzed species. By contrast, under EDDS excess, metal resorption was indiscernible while more significant Al and Fe dissolution influenced the EDDS speciation.

Introduction Heavy metal contamination in soils, resulting from industrial and agricultural activities, is widespread around the world (1-3). To restore metal-contaminated soils, chelating agents have been extensively studied for enhancing in situ (e.g., flushing, phytoextraction) and ex situ (e.g., washing, heap leaching) remediation (4-10). EDTA (ethylenediaminetetraacetic acid) has been the most often proposed chelating agent (3, 11, 12, and references therein), but it is nonbiodegradable in the environment so that residual metalsEDTA complexes would travel in the subsurface with high mobility (13, 14) and possibly lead to adverse health and environmental effects (4, 9). Thus, EDDS ([S,S]-stereoisomer of ethylene* Corresponding author e-mail: [email protected]; fax: 64 3 364 2758; tel: 64 3 364 2394. † The Hong Kong University of Science and Technology. ‡ University of Canterbury. § Former address: Institute for the Environment, The Hong Kong University of Science and Technology, Hong Kong, China. 10.1021/es802030k CCC: $40.75

Published on Web 01/06/2009

 2009 American Chemical Society

diaminedisuccinic acid) has recently emerged as a promising substitute, for it is readily biodegradable in soils, less toxic to plants, fungi, and microorganisms (15), and has insignificant residual effects (16). Although EDDS could be fully degraded in soils without sludge amendment (17-19), an initial lag phase, in the range of 7-32 days, was found to be necessary for the population growth or adaptation of adequate microbes (8, 17, 19). The length of lag phase may vary with the soil type and extent of metal contamination (19). The kinetic interactions of EDDS with soils, therefore, are of primary concern, especially at the early stage of soil remediation where EDDS degradation is minimal. It has been speculated in the literature that the effectiveness of EDDS application may be reduced via two processes. The first one is the resorption (immobilization) of EDDSextracted metals due to iron dissolution by newly formed metalsEDDS complexes, during which the metals would be liberated from EDDS complexes and resorbed on the soils (6, 7). The second is the competition of mineral cations for EDDS, which may reduce the available amount of EDDS for extracting the target metals and influence the metal speciation in solution. Dissolution of Ca, Mg, and Fe were substantial at different pH (6, 7, 10, 20) and Al dissolution was also found to be significant in the latest studies (19, 21). These mineral cations were suggested to be the major competing cations. Most of the previous studies aimed to demonstrate the effectiveness of EDDS application at EDDS-to-metal molar ratios (MR) ranging between 1 and 10 (6, 7, 10, 19, 20); that is, EDDS was applied in sufficient or excessive amounts with respect to total amount of target metals in soils. However, for in situ soil remediation that involves continuous flushing, the concentration of incoming chelant is deficient with respect to the total concentrations of target metals (i.e., MR < 1), until the cumulative amount of applied chelant equals the total amount of metals in soils (at which chelant application is usually ceased) (3, 11). Thus, metal extraction is limited by the amount of EDDS most of the time. The significance of resorption of newly extracted metals and competition of dissolved mineral cations for EDDS is likely to be different under EDDS-deficiency and EDDS-excess scenarios. In this study, therefore, batch experiments were conducted to investigate the 7-day kinetic interactions of uncomplexed (i.e., free) EDDS with soils under EDDS deficiency (MR 0.5) for the case of in situ soil flushing and under EDDS excess (MR 2 and 5) for direct comparison with the case of prolonged flushing or ex situ soil washing. The adsorption of EDDS, metal extraction and temporal distribution change, mineral and organic matter dissolution, and metal and EDDS speciation were analyzed at pH 5.5 and 8, respectively. In a companion paper (22), kinetic interactions of the newly formed metalsEDDS complexes with uncontaminated and contaminated soils were investigated.

Experimental Section Soil Characteristics. In view of qualitatively similar results obtained for the three soils investigated, this study reports the results of only one soil, which was collected from 25-50 cm below ground surface at Clearwater Bay in Hong Kong, air-dried, and passed through a 2-mm sieve. The soil characteristics are summarized in Table S1 (Supporting Information). Artificial soil contamination allowed direct comparison between the kinetic interactions of EDDS with uncontaminated soils and that with metal-contaminated soils VOL. 43, NO. 3, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

831

in this paper and its companion (22). Contamination by Cu, Zn, and Pb are of particular interest, because these metals are most often encountered (2) and present in fieldcontaminated soils/sediments where EDDS application had been studied (4-10, 15-21). The soils were contaminated by mixing with multimetalcontaining solution (500 mg L-1 Cu, Zn, and Pb, respectively) at a soil-to-solution ratio of 200 g L-1. The multimetalcontaining solution was prepared by dissolving Cu(NO3)2, Zn(NO3)2, and Pb(NO3)2 (analytical-grade, Fisher Scientific) simultaneously in 0.01 M NaNO3 background solution and adjusted to pH 5.5 by 0.1 M HNO3 or NaOH. The soil suspensions were shaken end-over-end at 26 rpm for 7 days and then aged with occasional shaking for an additional 1 month (with regular pH readjustment to 5.5). Afterward the soils were separated by centrifugation (Beckman Auegra 6 centrifuge) at 3500 rpm for 15 min, rinsed with background solution three times to displace entrapped and loosely bound metals, and air-dried. The total metal concentrations of the artificially contaminated soils, measured by acid digestion with HNO3-HCl-HF in a microwave digester (CEM MDS2000), were within the range of reported values of fieldcontaminated soils/sediments, whereas the metal distribution, determined by sequential extractions as described in previous studies (23, 24), indicated a large fraction of weakly bound metals (Table S2). It should be noted that in fieldcontaminated soils, due to aging effect, metals were more strongly bound that their extraction required longer time and higher chelant application rate (4, 11, 25). But efficient metal extraction of field-contaminated soils could also be achieved by EDDS, except for Zn phyllosilicate component, underlining the importance of metal speciation in soils (26). Batch Kinetic Experiments. The soils (0.5 g) were mixed with EDDS solution at a soil/solution ratio of 50 g L-1 in 40-mL polypropylene centrifuge tubes. The EDDS concentrations in solution and corresponding application rates at MR 0 (control), 0.5, 2, and 5 are summarized in Table S3. The EDDS solution was prepared by mixing [S,S]-EDDS (30% Na3EDDS solution, Innospec Ltd., HK) with 0.01 M NaNO3 background solution. The solution pH was maintained at pH 5.5 with 2 mM MES (2-morpholinoethane-sulfonic acid) and pH 8 with 2 mM HEPES [4-(2-hydroxyethyl)-piperazine1-ethane-sulfonic acid], respectively (27-29). MES and HEPES buffers have very weak complexing properties that would not influence EDDS adsorption and complexation. The solution pH varied by at most 0.3 after the experiments. To avoid biodegradation and photodegradation, EDDS solutions contained sodium azide (200 mg L-1) and were kept in dark. The soil suspensions were shaken end-over-end at 26 rpm for 16 different reaction times under room temperature. Samples (1-min to 90-min) were collected using PE syringes and 0.2-µm cellulose acetate filters, while following samples (2-h to 168-h) were separated by centrifugation at 3500 rpm for 15 min (27-29) for subsequent soil analysis. All experiments were at least replicated to ensure reproducibility. Soil samples (2-h, 24-h, and 168-h) were rinsed with ultrapure water and freeze-dried before sequential extractions were carried out to determine the temporal change of Cu, Zn and Pb distribution. Solution samples were acidified with concentrated HNO3 and stored in dark at 4 °C. Dissolved metal concentrations (target metals: Cu, Zn, and Pb; mineral cations: Al, Ca, Mn, and Fe) were determined by inductively coupled plasmaoptical emission spectrometer (Perkin-Elmer Optima 3000XL). The initial kinetics of metal extraction and mineral dissolution was estimated using a first-order equation (r2 > 0.89). Dissolved organic carbon (DOC) concentrations measured by TOC analyzer (Shimadzu TOC-5000A), from which carbon concentration resulting from EDDS and buffer was sub832

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 3, 2009

FIGURE 1. Adsorbed EDDS on: (a) uncontaminated soils (• pH 5.5, O pH 8; initial concentration equivalent to MR 2); and (b) contaminated soils at pH 5.5 (EDDS-to-metal molar ratio (MR): • MR 0.5; 9 MR 2; 2 MR 5; adsorption was insignificant at pH 8 and thus not shown). tracted, were 38 and 55 mg L-1 at pH 5.5 and 8, respectively. The EDDS concentrations were too high for fluorescence detection analysis (30) and thus were measured by colorimetric analysis at 670 nm absorption peak of CuEDDS2- using UV-visible spectrophotometer (Milton Roy Spectronic 3000), with the detection range of 0.1-2 mM EDDS (31). Solution samples were diluted where necessary; acetate buffer (pH 5) and copper sulfate were added to 50 and 3 mM, respectively. The absorbance at 670 nm was measured after 72-h incubation at room temperature, which, based on preliminary experiments, was sufficient for added Cu to completely displace other metals from EDDS complexes. Metal and EDDS Speciation. The program Visual MINTEQ version 2.53 (32) was used to calculate the EDDS and metal speciation in solution at different reaction times. Input parameters were the measured EDDS concentrations, dissolved metals concentrations (Al, Ca, Mn, Fe, Cu, Zn, and Pb), DOM concentrations, solution pH, and background ions (Na and NO3-). The stability constants (Table S4) of metalsEDDS complexes, except AlEDDS-, and protonation of EDDS were obtained from Martell et al. (33). The stability constant of AlEDDS- was recently estimated by Koopsman et al. (21) based on known stability constants for complexes of trivalent metals with EDTA and those with EDDS. Dissolved organic matter (DOM) concentrations were calculated by multiplying the DOC concentrations by two (17, 21). The composition of DOM was assumed to be 50% fulvic acid and 50% humic acid (17). The binding of metals to DOM was modeled using the NICA-Donnan model with generic parameters taken from Milne et al. (34). Possible precipitation of Al and Fe oxides (Al(OH)3 and amorphous Fe(OH)3 (21)) in solution was taken into account.

Results and Discussion EDDS Adsorption. In uncontaminated soils EDDS adsorption was fast in the first 4 h and reached apparent equilibrium after 48-72 h (Figure 1a). At equilibrium, adsorbed EDDS

FIGURE 2. Heavy metal extraction at an EDDS-to-metal molar ratio of 0.5: (a) initial 6 h; and (b) 6-168 h (b, O Cu; 9, 0 Zn; 2, ∆ Pb; closed and open symbols represent data at pH 5.5 and pH 8, respectively; error bars represent the standard deviations). was 28 mmol kg-1 (20.1% of total EDDS) and 4 mmol kg-1 (2.8%) at pH 5.5 and 8, respectively. Adsorption is the basis for all surface chemical reactions that are associated with the ligand (e.g., dissolution) (35) and possibly occurs through inner-sphere complexation (ligand exchange), outer-sphere complexation (electrostatic attraction), and/or hydrogen bonding (27, 36-38). Ligand exchange and electrostatic attraction would be enhanced when the soil surfaces become more protonated and more positive with decreasing pH (yet EDDS precipitation predominates at pH < 4). EDDS adsorption is, therefore, well-recognized to be inversely correlated to solution pH (4, 7). Adsorbed EDDS was higher than found in previous studies, in which they were 0.55 mmol kg-1 (17%) at pH 4.7 (21) and 1 mmol kg-1 (5%) at pH 7 (7), respectively. In addition to low soil pH that may facilitate EDDS adsorption (21), this is attributed to the high content of Fe oxides (48.5 g kg-1) with 99.4% in crystalline form (Table S1), which can provide a substantial number of sites for EDDS adsorption. Strong adsorption of EDTA and phosphonates, depending on pH, on various crystalline oxides of Fe and Al has also been reported (27, 28, 39). The significance of crystalline oxides was corroborated by the fact that EDDS adsorption at the same solution pH (buffered) was negligible on two other soils that contained similar amount of amorphous oxides but 6- to 9-fold less crystalline oxides (data not shown). Equilibrium EDDS adsorption on contaminated soils at pH 5.5 was 0, 25, and 27 mmol kg-1 (i.e., 0, 17.9, and 7.8% of initial amount) at MR 0.5, 2, and 5, respectively (Figure 1b). In view of reported ligand adsorption isotherm (27, 28, 39), a larger fraction of EDDS is expected to be adsorbed at low EDDS concentration at MR 0.5. However, the observation of negligible adsorption suggested that the limited amount of EDDS was preferentially used for metal

FIGURE 3. Mineral dissolution at various EDDS-to-metal molar ratios (MR): (a) Al; (b) Ca; (c) Mn; and (d) Fe (b MR 0; 9 MR 0.5; 2,∆ MR 2; 1 MR 5; closed and open symbols represent data at pH 5.5 and pH 8, respectively; error bars represent the standard deviations). extraction (next section). The resulting metal-EDDS complexes adsorb less strongly than uncomplexed EDDS (data not shown), in agreement with previous studies of EDTA (27, 28). On the other hand, saturation of available EDDS adsorption sites was reflected at MR 5 as adsorbed EDDS marginally increased compared with that of MR 2, but the maximum adsorption capacity should be determined by adsorption isotherm. Metal Extraction. No sorbed Cu or Pb, and only 20% of sorbed Zn in the contaminated soils was extracted at pH 5.5 in the absence of EDDS, whereas 80-90% of sorbed Cu, Zn, and Pb were extracted when EDDS was in excess (MR 2 and 5) (Figure S1), corroborating the pivotal role of EDDS in enhancing metal extraction. Under EDDS deficiency at MR VOL. 43, NO. 3, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

833

FIGURE 4. Metal and EDDS speciation in solution at an EDDS-to-metal molar ratio of 0.5 during the initial 6 h of extraction: (a) metalsEDDS complex as a percentage of respective dissolved metal concentration, at pH 5.5 (a) and pH 8 (b); and EDDS speciation, at pH 5.5 (c) and pH 8 (d). 0.5, it was not surprising to reveal efficient extraction of Cu (Figure 2), which is expected to be preferentially complexed with EDDS. Nevertheless, at pH 5.5, extracted amounts of Pb and Zn also increased during the first 15 and 180 min, respectively (Figure 2a), after which their extraction efficiency decreased with time. Initial extraction of Zn and Pb, despite comparatively low stability constants of their EDDS complexes, suggested that the preference of metal extraction was more influenced by the metal lability (depending on the total concentration, bonding characteristics, steric distribution, and chemical speciation of metals in soils) than by the selectivity of EDDS surface complexation. However, EDDS-extracted Zn and Pb were subsequently resorbed due to adsorption and metal exchange of newly formed ZnEDDS2- and PbEDDS2- on the soil surfaces (6, 7, 22). The change of Zn and Pb distribution between 2 and 168 h showed that they were mainly resorbed on the exchangeable and carbonate fractions, and to a lesser extent, on the oxide fraction (Figure S2). The molar sum of dissolved concentrations of Cu, Zn, and Pb at both pH 5.5 and 8 exceeded EDDS concentrations in solution by a factor of 1.08-1.15 at MR 0.5 after the first 15 min (Figure S3). Molar excess of extracted metals with respect to EDDS in solution (a ratio of 1.22) was also reported in a recent study (21). In consideration of 1:1 metalsEDDS complexation and negligible metal extraction (except for Zn at pH 5.5) in the absence of EDDS (Figure S1), the observed metal excess in solution was an evidence of metal exchange of metalsEDDS complexes on the soil surfaces. In the presence of 34.8 mmol kg-1 of EDDS, 11.5 and 4.8 mmol kg-1 of Zn and Pb were initially extracted, whereas 29.8 mmol kg-1 of Cu was eventually extracted. This suggests that Cu was extracted not only by uncomplexed EDDS, but also by metal exchange of newly formed ZnEDDS2- and PbEDDS2-. Along with partial liberation of Zn and Pb from EDDS complexes, a portion of Zn and Pb was resorbed on the soils while some remained in solution (explained in a later section) and resulted in the metal excess (Figure S3). 834

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 3, 2009

By contrast, at MR 0.5 and pH 8, the reduction in extraction efficiency with time was almost negligible for Zn and notably less for Pb (Figure 2). The extent of Zn and Pb resorption on the exchangeable, carbonate, and oxide fractions correspondingly decreased (Figure S2). Under EDDS excess (MR 2 and 5), Zn and Pb resorption was also not apparent (Figure S1). The varying importance of metal resorption was analyzed with speciation calculation and is reported in a later section. The extraction kinetics (rate coefficients were 2.83 × 10-5 to 1.04 × 10-4 and 7.61 × 10-6 to 3.78 × 10-5 s-1 at pH 5.5 and 8, respectively) of Cu, Zn, and Pb was similar at MR 2 and 5. Within 2 h considerable amounts of exchangeable and carbonate fractions and minor amounts of organic matter and residual fractions were extracted, whereas the oxide fraction was slowly extracted over 168 h (data not shown). Mineral Dissolution. Adsorbed EDDS may substantially enhance mineral dissolution through the destabilization of metal-oxygen bonds (37, 40, 41). Negligible amounts of Al, Fe, and Mn were dissolved in the absence of EDDS, while there was increased dissolution with longer reaction time at MR 0.5, 2, and 5 (Figure 3), indicating that their dissolution was predominantly induced by EDDS. Nevertheless, Ca dissolution resulting from exchangeable Ca and carbonates (42) was invariant with time. Mineral dissolution would reduce the shear strength and aggregate stability of soils (43). A small amount of dissolution (e.g., < 13 g kg-1) was found to mobilize colloids and fine particles (i.e., soil dispersion) and, in turn, cause column clogging in flushing studies (3, 7, 11). However, dissolution rate coefficients of Al, Mn, and Fe (4.17-6.11 × 10-7, 1.11-1.67 × 10-7, and 2.78 × 10-9 to 2.22 × 10-8 s-1, respectively) were significantly slower than those of target metal extraction, reflecting the rate-limited detachment of metals from the mineral structure. In addition, minor extent of increase of dissolution at MR 5 compared with that at MR 2 (data not shown) was in line with the limited availability of EDDS adsorption sites (Figure 1b). Among all minerals, Al dissolution was by far the most significant. Relatively rapid Al dissolution (Figure 3a) appears

to derive from readily available Al sources (e.g., electrostatically bound) (44, 45) besides Al oxides and aluminosilicates, because Al oxides represented only 12.6% of total Al content (Table S1). On the other hand, Mn and Fe dissolution were slow (Figure 3c and d) because of a large crystalline proportion of Mn and Fe oxides (83.8% and 99.4%, respectively; Table S1). Dissolution kinetics of goethite by EDTA, for instance, was six orders of magnitude slower than that of hydrous ferric oxide (29). Therefore, dissolved Al was the major mineral cation that may compete for EDDS under EDDS deficiency, whichwassubsequentlyevaluatedwithspeciationcalculations. Dissolution of Al, Mn, and Fe were enhanced at pH 8 (Figure 3), despite lower EDDS adsorption (Figure 1). The dominant adsorption mechanism of EDDS may change from binuclear at low pH to mononuclear at above-neutral pH and facilitate dissolution, as suggested for the surface complexes of EDTA on crystalline Fe oxides (27, 28). Nevertheless, regardless of the adsorbed forms, newly formed MnEDDS2- and FeEDDS- possibly have a higher tendency to detach from the mineral surfaces at pH 8 than at pH 5.5. Because at pH 8 MnEDDS2- and FeEDDS- are more stable in solution (as reflected by higher stability constants), the metalsoxygen bonds in the oxide structures are more likely to be destabilized so that the metal center can dissociate from the surface to maximize bonding with EDDS (40, 46). Moreover, dissolution of organic matter, as well as DOMbound Al and Fe (47), was enhanced by increasing pH. The observed pH dependence may result from one or a combination of the above effects. Metal and EDDS Speciation. Under EDDS deficiency at MR 0.5, metal speciation in solution indicated that ZnEDDS2and PbEDDS2- as a percentage of respective dissolved metal concentrations decreased with time (Figure 4a and b and Figure S4a and b). As these complexes alone were stable in solution, their dissociation evidenced the metal exchange of newly formed metalsEDDS complexes, by which the metals would be liberated from EDDS complexes and become more likely to be resorbed on soils. High sorption strength of Pb may also facilitate its complex dissociation for metal exchange (48). However, a fraction of liberated Zn and Pb remained in solution due to DOM binding (Figure S5a and b), accounting for the observed metal excess relative to EDDS in solution. Recent studies also illustrated the importance of DOM binding for metal speciation in the presence of EDDS (17, 21). As a result of metal exchange, the contribution of CuEDDS2- to total EDDS increased with time while those of ZnEDDS2- and PbEDDS2- decreased accordingly (Figure 4c and d and Figure S4c and d). The decrease of PbEDDS2- was faster than that of ZnEDDS2-, whereas both percentages decreased more slowly at higher pH. These temporal changes of EDDS speciation corresponded to the observation of faster resorption of Pb than Zn and less metal resorption at pH 8 than at pH 5.5 (Figure 2). The varying significance of metal exchange of metalsEDDS complexes was investigated in the companion paper (22). While dissolved Al concentration was almost equal to 20% of total EDDS, there was insignificant contribution of AlEDDS- to total EDDS, because newly formed AlEDDS- would readily dissociate in solution or undergo metal exchange (22). The predominant forms of Al were colloidal Al(OH)3, Al-DOM, and hydrolyzed species (Figure S5c and d). Thus, the competitive effect of Al on metal extraction was unimportant although there was no uncomplexed EDDS. In addition, limited amounts of dissolved Ca, Mn, and Fe (Figure 3) caused negligible competition for EDDS; Fe was predominantly present as Fe-DOM (and colloidal Fe(OH)3 at pH 8) (Figure S5c and d). Yet, Fe competition for EDDS may be more important in some soils (21), depending on the type of oxides because the individual

site energy and local coordination environment of the detachable metal center determine the dissolution rate (40, 41). In the presence of EDDS excess (MR 2 and 5), Cu, Zn, and Pb were entirely complexed with EDDS (data not shown). Therefore, Zn and Pb that were liberated during metal exchange were recomplexed with EDDS in solution, accounting for little metal resorption under EDDS excess (Figure S1). On the other hand, despite noticeable contributions of AlEDDS- and FeEDDS- to total EDDS (data not shown), about 20% (at MR 2) and 80% (at MR 5) of EDDS was uncomplexed and available for metal extraction. Therefore, the competition of dissolved mineral cations was more influential in EDDS speciation than in metal extraction. Engineering Implications. During continuous in situ flushing, EDDS-deficiency and EDDS-excess scenarios probably exist at different times. Initial extraction of Zn and Pb under EDDS deficiency signifies the preferential extraction of labile metals on the soil surfaces. Metal extraction is affected by, particularly when EDDS is of limited amount at pH 5.5, metal exchange of newly formed metalsEDDS complexes and subsequent metal resorption. This effect, however, is indiscernible under EDDS excess. On the other hand, the competitive effects of dissolved mineral cations depend on the mineral structure and the amount of EDDS. Although Al dissolution is relatively fast under EDDS deficiency, dissolved Al has weak competition for EDDS and low tendency to reduce metal extraction efficiency. When EDDS is in excess, dissolved Al and Fe are of substantial amounts, especially at pH 8, that influence the EDDS speciation in solution, which, in turn, affect the bioavailability and mobility of EDDS in the subsurface.

Acknowledgments We thank the Research Grant Council of Hong Kong for providing financial support under General Research Fund with project account 616608 for this research study.

Supporting Information Available Soil characteristics (Table S1); sequential extraction (Table S2); EDDS application rates (Table S3); stability constants of EDDS complexes (Table S4); metal extraction (Figure S1); metal distribution in soils (Figure S2); dissolved EDDS and metal concentrations (Figure S3); metal and EDDS speciation in solution (Figures S4 and S5). This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) Li, X. D.; Huang, C. Environment impact of heavy metals on urban soil in the vicinity of industrial area of Baoji city, P.R. China. Environ. Geol. 2007, 52, 1631–1637. (2) Peters, R. W. Chelant extraction of heavy metals from contaminated soils. J. Hazard. Mater. 1999, 66, 151–210. (3) Sun, B.; Zhao, F. J.; Lombi, E.; McGrath, S. P. Leaching of heavy metals from contaminated soils using EDTA. Environ. Pollut. 2001, 113, 111–120. (4) Vandevivere, P. C.; Hammes, F.; Verstraete, W.; Feijtel, T.; Schowanek, D. Metal decontamination of soil, sediment, and sewage sludge by means of transition metal chelant [S,S]-EDDS. J. Environ. Eng. 2001, 127, 802–811. (5) Kos, B.; Lestan, D. Induced phytoextraction/soil washing of lead using biodegradable chelate and permeable barriers. Environ. Sci. Technol. 2003, 37, 624–629. (6) Tandy, S.; Bossart, K.; Mueller, R.; Ritschel, J.; Hauser, L.; Schulin, R.; Nowack, B. Extraction of heavy metals from soils using biodegradable chelating agents. Environ. Sci. Technol. 2004, 38, 937–944. (7) Hauser, L.; Tandy, S.; Schulin, R.; Nowack, B. Column extraction of heavy metals from soils using the biodegradable chelating agent EDDS. Environ. Sci. Technol. 2005, 39, 6819–6824. (8) Finzgar, N.; Zumer, A.; Lestan, D. Heap leaching of Cu contaminated soil with [S,S]-EDDS in a closed process loop. J. Hazard Mater. 2006, B135, 418–422. VOL. 43, NO. 3, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

835

(9) Tandy, S.; Schulin, R.; Nowack, B. Uptake of metals during chelant-assisted phytoextraction with EDDS related to the solubilized metal concentration. Environ. Sci. Technol. 2006, 40, 2753–2758. (10) Polettini, A.; Pomi, R.; Rolle, E.; Ceremigna, D. A kinetic study of chelant-assisted remediation of contaminated dredged sediment. J. Hazard. Mater. 2006, B137, 1458–1465. (11) Tsang, D. C. W.; Zhang, W.; Lo, I. M. C. Copper extraction effectiveness and soil dissolution issues of EDTA-flushing of artificially contaminated soils. Chemosphere 2007, 68, 234–243. (12) Zhang, W.; Tsang, D. C. W.; Lo, I. M. C. Removal of MDF and Pb from contaminated soils by EDTA- and SDS-enhanced washing. Chemosphere 2007, 66, 2025–2034. (13) Friedly, J. C.; Kent, D. B.; Davis, J. A. Simulation of the mobility of metalsEDTA complexes in groundwater: the influence of contaminant metals. Environ. Sci. Technol. 2002, 36, 355–363. (14) Tsang, D. C. W.; Lo, I. M. C.; Chan, K. L. Modeling the transport of metals with rate-limited EDTA-promoted extraction and dissolution during EDTA-flushing of copper-contaminated soils. Environ. Sci. Technol. 2007, 41, 3660–3667. (15) Grcman, H.; Vodnik, D.; Bolta, S. V.; Lestan, D. Ethylenediaminedisuccinate as a new chelate for environmentally safe enhanced lead phytoextraction. J. Environ. Qual. 2003, 32, 500– 506. (16) Luo, C. L.; Shen, Z. G.; Li, X. D.; Lou, L. Q. EDDS and EDTAenhanced phytoextraction of metals from artificially contaminated soil and residual effects of chelant compounds. Environ. Pollut. 2006, 144, 862–871. (17) Tandy, S.; Ammann, A.; Schulin, R.; Nowack, B. Biodegradation and speciation of residual SS-ethylenediaminedisuccinic acid (EDDS) in soil solution left after soil washing. Environ. Pollut. 2006, 142, 191–199. (18) Wang, G.; Koopmans, G. F.; Song, J.; Temminghoff, E. J. M.; Luo, Y.; Zhao, Q.; Japenga, J. Mobilization of heavy metals from contaminated paddy soil by EDDS, EDTA, and elemental sulfur. Environ. Geochem. Health 2007, 29, 221–235. (19) Meers, E.; Tack, F. M. G.; Verloo, M. G. Degradability of ethylenediaminedisuccinic acid (EDDS) in metal contaminated soils: Implications for its use soil remediation. Chemosphere 2008, 70, 358–363. (20) Polettini, A.; Pomi, R.; Rolle, E. The effect of operating variables on chelant-assisted remediation of contaminated dredged sediment. Chemosphere 2007, 66, 866–877. (21) Koopmans, G. F.; Schenkeveld, W. D. C.; Song, J.; Luo, Y.; Japenga, J.; Temminghoff, E. J. M. Influence of EDDS on metal speciation in soil extracts: Measurement and mechanistic multicomponent modeling. Environ. Sci. Technol. 2008, 42, 1123–1130. (22) Tsang, D. C. W.; Yip, T. C. M.; Lo, I. M. C. Kinetic interactions of EDDS with soils. 2. MetalsEDDS complexes in uncontaminated and metal-contaminated soils. Environ. Sci. Technol. 2009, 43, 837–842. (23) Tsang, D. C. W.; Lo, I. M. C. Competitive Cu and Cd sorption and transport in soils: A combined batch kinetics, column, and sequential extraction study. Environ. Sci. Technol. 2006, 40, 6655–6661. (24) Tsang, D. C. W.; Zhang, W.; Lo, I. M. C. Modeling cadmium transport in soils using sequential extraction, batch, and miscible displacement experiments. Soil Sci. Soc. Am. J. 2007, 71, 674– 681. (25) Kim, C.; Lee, Y.; Ong, S. K. Factors affecting EDTA extraction of lead from lead-contaminated soils. Chemosphere 2003, 51, 845– 853. (26) Kirpichtchikova, T. A.; Manceau, A.; Spadini, L.; Panfili, F.; Marcus, M. A.; Jacquet, T. Speciation and solubility of heavy metals in contaminated soil using X-ray microfluorescence, EXAFS spectroscopy, chemical extraction, and thermodynamic modeling. Geochim. Cosmochim. Acta 2006, 70, 2163–2190.

836

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 3, 2009

(27) Nowack, B.; Sigg, L. Adsorption of EDTA and metal-EDTA complexes onto goethite. J. Colloid Interface Sci. 1996, 177, 106– 121. (28) Nowack, B.; Lutzenkirchen, J.; Behra, P.; Sigg, L. Modeling the adsorption of metalsEDTA complexes onto oxides. Environ. Sci. Technol. 1996, 30, 2397–2405. (29) Nowack, B.; Sigg, L. Dissolution of Fe(III) (hydr)oxides by metalsEDTA complexes. Geochim. Cosmochim. Acta 1997, 61, 951–963. (30) Tandy, S.; Schulin, R.; Suter, M. J. F.; Nowack, B. Determination of [S,S′]-ethylenediamine disuccinic acid (EDDS) by high performance liquid chromatography after derivatization with FMOC. J. Chromatogr., A 2005, 1077, 37–43. (31) Vandevivere, P. C.; Saveyn, H.; Verstraete, W.; Feijtel, T. C. J.; Schowaneh, D. R. Biodegradation of metal-[S,S]-EDDS complexes. Environ. Sci. Technol. 2001, 35, 1765–1770. (32) Gustafsson, J. P. Visual MINTEQ Version 2.53; Department of Land and Water Resources Engineering, Royal Institute of Technology: Stockholm, Sweden, 2008. (33) Martell, A. E.; Smith, R. M.; Motekaitis, R. J. NIST Critically Selected Stability Constants of Metal Complexes, Version 8.0; National Institute of Standards and Technology: Gaithersburg, MD, 2004. (34) Milne, C. J.; Kinniburgh, D. G.; van Riemsdijk, W. H.; Tipping, E. Generic NICA-Donnan model parameters for metal-ion binding by humic substances. Environ. Sci. Technol. 2003, 37, 958–971. (35) Mcardell, C. S.; Stone, A. T.; Tian, J. Reaction of EDTA and related aminocarboxylate chelating agents with CoIIIOOH (heterogenite) and MnIIIOOH (Manganite). Environ. Sci. Technol. 1998, 32, 2923–2930. (36) Bargar, J. R.; Persson, P.; Brown, G. E., Jr. Outer-sphere adsorption of Pb(II)EDTA on goethite. Geochim. Cosmochim. Acta 1999, 63, 2957–2969. (37) Nowack, B. Environmental chemistry of aminopolycarboxylate chelating agents. Environ. Sci. Technol. 2002, 36, 4009–4016. (38) Guclu, K.; Apak, R. Modeling the adsorption of free and heavy metal complex-bound EDTA onto red mud by a nonelectrostatic surface complexation model. J. Colloid Interface Sci. 2003, 260, 280–290. (39) Nowack, B.; Stone, A. T. Adsorption of phosphonates onto the goethite-water interface. J. Colloid Interface Sci. 1999, 214, 20– 30. (40) Sposito, G. The Surface Chemistry of Natural Particles; Oxford University Press: New York, 2004. (41) Blesa, M. A.; Morando, P. J.; Regazzoni, A. E. Chemical Dissolution of Metal Oxides; CRC Press: Boca Raton, FL, 1994. (42) Loeppert, R. H.; Suarez, D. L. Chapter 15: Carbonate and gypsum In Methods of Soil Analysis, Part 3: Chemical Methods; Sparks, D. L., Ed.; Soil Science Society of America, Inc.: Madison, WI, 1996. (43) Kay, B. D.; Angers, D. A. Soil structure In Soil Physics Companion; Warwick, A. W., Ed.; CRC Press: Boca Raton, FL, 2002. (44) Jardine, P. M.; Taylor, D. L. Fate and transport of ethylenediaminetetraacetate chelated contaminants in subsurface environments. Geoderma 1995, 67, 125–140. (45) Mayes, M. A.; Jardine, P. M.; Larsen, I. L.; Brooks, S. C.; Fendorf, S. E. Multispecies transport of metal-EDTA complexes and chromate through undisturbed columns of weathered fractured saprolite. J. Contam. Hydrol. 2000, 45, 243–265. (46) McBride, M. B. Environmental Chemistry of Soils; Oxford University Press: New York, 1994. (47) Tipping, E. Cation Binding by Humic Substances; Cambridge University Press: New York, 2002. (48) Kent, D. B.; Davis, J. A.; Joye, J. L.; Curtis, G. P. Influence of variable chemical conditions on EDTA-enhanced transport of metal ions in mildly acidic groundwater. Environ. Pollut. 2008, 153, 44–52.

ES802030K