Article pubs.acs.org/est
Importance of Dynamic Soil Properties in Metal Retention: An Example from Long-Term Cu Partitioning and Redistribution Studies Using Model Systems Nadia Martínez-Villegas†,‡,* and Carmen Enid Martínez† †
Department of Crop and Soil Sciences, The Pennsylvania State University, 116 ASI Building, University Park, Pennsylvania 16802, United States ‡ IPICyT, Instituto Potosino de Investigacion Cientifica y Tecnologica, Division de Geociencias Aplicadas, Camino a la Presa San Jose #2055 Col. Lomas 4ta seccion CP 78216, San Luis Potosi, SLP, Mexico S Supporting Information *
ABSTRACT: The effect of initial conditions and reaction pathways in the long term solid-solution partitioning and solidphase distribution of Cu among ferrihydrite, leaf compost (LC), and montmorillonite (K-SWy2) were established using compartmentalized batch reactors by varying the sequence of mixing of the sorbents. Copper was allowed to react with a single solid phase for 30 days (1st equilibration) before introducing the other two solid phases and equilibration for 8 additional months (2nd equilibration). The systems were labeled Fe-Ox, Organic, or Smectitic reflecting the single initial solid phase present during the first equilibration. Total dissolved Cu and total Cu in individual solid phases were determined as a function of time during the first and second equilibrations. Results showed that different initial conditions elicited different dynamic responses where the generation of dissolved organic carbon (DOC) and diffusion of colloidal ferrihydrite seemed to influence the long-term partitioning and distribution of Cu. Trends in total dissolved Cu for the systems at the end of the first equilibration were Fe-Ox > Organic > Smectitic, while at the end of the second equilibration the organic system was the least effective in the removal of Cu from solution (Organic > Fe-Ox ≈ Smectitic). Furthermore, our results indicated Cu redistribution toward organic matter and montmorillonite, with small amounts of Cu retained by ferrihydrite. These results are attributed to reaction pathways where the formation of soluble Cu−organic complexes and colloidal Cu−ferrihydrite, and their subsequent reaction with the solids present in the systems, were operative. The experiments reported herein show dynamic properties dictate Cu reaction pathways in multiphase-multicomponent systems and might help to explain unexpected higher mobility of metals after soil remediation.
■
INTRODUCTION
For example, the remediation of metal-contaminated soils by in situ application of oxides and hydroxides of Mn and Fe, clay layer silicates, and/or organic materials are typically regarded as systems where the addition of strong affinity binding sites is expected to decrease the mobility and bioavailability of metals through chemisorption, complexation, and/or ion exchange reactions. However, field and laboratory experiments often show that metal mobility increases in amended soils1−8 thus indicating predictions based on amendment sorption capacities determined at specific, static bulk conditions are not enough to predict dissolved metal concentrations in soils subjected to changes in their content of soil solids. Metal concentrations in soil solutions might, in these cases, represent the combined
It has long been proposed that sorption processes controlling metal concentrations in soil solution are functions of soil and metal properties. The first underlying concept in this premise is the attainment of equilibrium. The second underlying concept is that sorption processes are controlled by bulk soil properties (i.e., soil pH, organic matter content, texture, mineralogy, and metal concentration) which are invariably regarded as static values. Effects of bulk properties on metal retention have been largely studied at equilibrium using homogeneously mixed batch reactors so that their effect on metal sorption is reasonably well understood. However, soils are open dynamic natural systems where bulk properties should be seen not only as a set of static initial conditions leading to a given equilibrium concentration of metal ions in solution but also as a system subjected to perturbation stimuli that can elicit different dynamic responses. © 2012 American Chemical Society
Received: Revised: Accepted: Published: 8069
December 12, 2011 May 2, 2012 July 3, 2012 July 3, 2012 dx.doi.org/10.1021/es3001932 | Environ. Sci. Technol. 2012, 46, 8069−8074
Environmental Science & Technology
Article
Figure 1. Experimental design used to study Cu solid-solution partitioning and solid-phase redistribution among ferrihydrite, leaf compost, and montmorillonite. (a) System identification and solid phases present and added during the 1st and 2nd equilibrations, respectively. (b) Example of experiment with LC as the solid phase present in the 1st equilibration.
result of metal retention by both the original and the added solids acting under bulk conditions, but conditioned by the dynamic properties of a soil system (i.e., mass transport and kinetics). Studying metal retention in soils as a process that result from dynamic responses dependent on initial conditions and perturbation stimuli would allow for the prediction of metal partitioning and distribution in different soil systems after soil disturbance and remediation. This new paradigm could help answer questions such as (i) what would be the long-term solid-solution partitioning and solid-phase distribution of a metal in a particular soil after the addition of a solid phase?; (ii) would metal redistribution occur after system perturbation?; and (iii) would the perturbation caused by a particular stimulus lead to the same equilibrium? Although several studies on the redistribution of metals have been undertaken,9−16 a general conclusion cannot be drawn yet. While some studies using chemical extractions report metal redistribution occurs from readily available operationally defined fractions toward more stable fractions,12,14−17 others report the opposite.10,11,16 Furthermore, it has been shown that soils tend to return to their original pattern of fractionation at low metal levels.9 The dynamic properties of a soil system depend on the number of state variables and their relationships and dictate the reaction pathways that drive a soil system from an initial condition to equilibrium (whether singular or multiple). Reaction pathways can be perturbed by bulk changes which, depending on the stability (i.e., robustness) and on the attraction (i.e., convergence) of the equilibrium, drive the soil system back to the same equilibrium, to a different equilibrium, or even induce nonmonotonic or oscillatory behaviors. The dynamic properties of a soil system (i.e., generation and resorption of dissolved organic matter, release and retention of inorganic and/or organic colloidal material) and their reaction pathways might be difficult to discern due to the complex nature of soils (variety of solid phases and surface properties) that result in a wide range of state variables and interrelations. Nevertheless, comprehensive dynamic analyses can allow us to
better predict whether a perturbation caused by a particular stimulus (for example, wet−dry cycles, abrupt changes in pH, addition of solid phases) will induce a dynamic response in the system. In this work we study the effects that initial conditions and a perturbation in bulk properties (caused by addition of solid phases) might have on Cu reaction pathways using compartmentalized reactors that allow for analyses of dynamic properties. In our experimental setup, Cu was allowed to react with a single solid phase (ferrihydrite, leaf compost, or montmorillonite) for 30 days (1st equilibration) before introduction of the remaining two solid phases to the systems and reaction for 8 additional months (2nd equilibration) (Figure 1). The sequence of mixing of the sorbents was varied to test its effect, if any, on the reaction pathway and final solubility and distribution of Cu. We show that different equilibriums are reached through different reaction pathways, but in general the partitioning and redistribution of Cu is highly influenced by two dynamic properties: the generation-sorption of DOC and colloidal ferrihydrite.
■
MATERIALS AND METHODS Experimental Design. Copper sorption and redistribution among ferrihydrite (Ferrihydrite), leaf compost (LC), and montmorillonite (K-SWy2) were studied using compartmentalized batch reactors. Copper was added to reactors containing a single solid phase and the systems were equilibrated for 30 days (1st equilibration). The remaining two solid phases were then introduced to each system and the systems were equilibrated for 8 additional months (2nd equilibration) (Figure 1). The systems were labeled “Fe-Ox”, “Organic” and “Smectitic” and indicated the solid phase present during the first equilibration (ferrihydrite, LC, and K-SWy2, respectively). Procedures used in the preparation and in the physical and chemical characterization of the solid phases, as well as in the ́ assemblage of the reactors are detailed in Martinez-Villegas and ́ 18 Briefly, a given amount (0.5 g) of an individual Martinez.
8070
dx.doi.org/10.1021/es3001932 | Environ. Sci. Technol. 2012, 46, 8069−8074
Environmental Science & Technology
Article
0.5 × 10−3 kg), summed, and reported as cumulative mass (mg) of Cu extracted. The DOC extracting solution was obtained by mixing 5 g of LC with 1 L of 0.01 M KNO3 for 24 h in an endto-end shaker. After mixing, the suspension was centrifuged (10 min at 5000 rpm) and the supernatant filtered (0.2 μm membrane). DOC concentration was measured in a Shimadzu TOC-5000A total organic carbon analyzer. The DOC solution was diluted to a concentration of 5 mg DOC L−1. This DOC extracting solution is representative of the solution present in the Fe-Ox and smectitic systems 24 h after the addition of LC to each system. No attempt was made to characterize the composition of the DOC.
solid phase (Ferrihydrite, LC, or K-SWy2) and 5 mL of 0.01 M KNO3 were added to a compartment (dialysis bag) and placed in a 1000 mL polypropylene container with 975 mL of 0.01 M KNO3. After 24 h, a 0.5 mL aliquot of a 100 mg L−1 Cu(NO3)2 solution was added to each reactor to yield a concentration of 0.051 mg Cu L−1. The systems were equilibrated for 30 days (1st equilibration) after which the remaining two solid phases (0.5 g each, in separate dialysis bags with 5 mL of 0.01 M KNO3) were introduced to each reactor for the second equilibration lasting 8 additional months (total volume of 990.5 mL). A schematic of the experimental design is shown in Figure 1. The experimental design permitted the study of Cu solidsolution partitioning and redistribution among three common soil solids (iron oxides, organic matter, and layer silicates) while testing the influence of the initial sorbent and added solid phases on the final results. Ten reactors per system were prepared initially (a total of 30 reactors). Three reactors (one per system) were disassembled at specific time intervals (1, 7, 15, and 30 days after the start of the first equilibration; and at 1 day, 15 days, and 1, 2, 4, and 8 months after the start of the second equilibration) for analyses of the solution and solid phases as described below. The pH was kept between 6 and 6.5 by addition of small amounts of 0.01 M HNO3 or 0.01 M KOH using a pH-stat device. The rate of Cu diffusion through the dialysis bags was tested and indicated that Cu diffusion was complete within 2 h without any sorption of Cu to the bag. Preliminary experiments also showed that Cu was not sorbed by the container or by the membrane. All chemicals used were trace grade. The solid phases (ferrihydrite, leaf compost, and montmorillonite) used in this study are model solids and their reactivity and spatial arrangement could differ substantially from oxides, organic matter, and aluminosilicate clays present in soil aggregates. Analyses of the Solution and Solid Phases. Solutions in all reactors were analyzed for total dissolved Cu (after acidification) by differential pulse anodic stripping voltammetry (DPASV) using a Metrohm 797 VA Computrace with 813 Autosampler (Ed = 0.02 Vvs. Ag/AgCl(satKCl),td = 120s,v = 15 mVs−1).18 The concentration of Cu in each solid phase (for second equilibration only) was determined by inductively coupled plasma mass spectrometry (ICP-MS) (for Ferrihydrite) or by DPASV (for LC and K-SWy2) analyses of Aqua ́ ́ 18 Regia digests as described in Martinez-Villegas and Martinez. For ease of comparison, all results are reported as mass (mg) of Cu. The mass of Cu in solution was calculated by multiplying total dissolved Cu concentrations (mg L−1) by the volume of solution in the reactor (990.5 mL). The mass of Cu in each solid phase was calculated by multiplying the concentration of Cu (mg Cu/kg solid phase) by 0.5 × 10−3 kg. Desorption Study. The potential for Cu desorption from individual solid phases (ferrihydrite, leaf compost, and montmorillonite) was determined at the end of the first equilibration, that is, after Cu reaction with a single solid phase for 30 days. Half a gram (0.5 g) of ferrihydrite, LC, or K-SWy2 was mixed with 20 mL of extracting solution (0.01 M KNO3 or 5 mg DOC L−1) for 2 h in an end-to-end shaker. After each extraction, the suspensions were centrifuged (10 min at 5000 rpm) and the supernatants filtered (0.2 μm membrane), acidified to pH < 2, and analyzed for total Cu by DPASV. The solids were extracted eight consecutive times by adding fresh solution (KNO3 or DOC) and following the same procedure. Copper concentrations were multiplied by the volume of the extracting solution over the mass of solid phase (20 × 10−3 L/
■
RESULTS AND DISCUSSION Total Dissolved Cu: First and Second Equilibrations. The single solid phases (ferrihydrite, LC, or K-SWy2) that comprise each system in the first equilibration have distinct surface areas and effective cation exchange capacities18 that resulted in different amounts of total dissolved Cu (Figure 2a).
Figure 2. Total dissolved Cu during the (a) 1st and (b) 2nd equilibrations for the Fe-Ox, organic, and smectitic systems.
That is, different initial conditions led to different equilibriums. The Smectitic system containing only K-SWy2 resulted in the least total dissolved Cu after 30 days (0.022 mg), followed by the Organic system containing only leaf compost (0.030 mg). The Fe-Ox system containing only ferrihydrite resulted in the highest amount of total dissolved Cu (0.049 mg) with minimal Cu retention (Figure 2a). During the first equilibration, the dynamic properties dictating the reaction pathways that drove these three different systems to equilibrium were mass transport and sorption kinetics, which are directly derived from the initial composition of each system. In the case of the organic system, a seemingly slow transport of Cu from bulk solution to the leaf compost allowed for the generation of dissolved organic carbon (12 mg/L of DOC at the end of the first equilibration) (Supporting Information (SI) Figure S1) that complexed Cu in solution and prevented organic matter from removing the highest amount of Cu (Figure 2a), as it is typically observed in short-term conventional studies using homogeneously mixed batch reactors.19 Copper sorption in the Fe-Ox and Smectitic systems was not affected by the presence of DOC during the first equilibration. However, minimal Cu removal in the Fe-Ox system can be explained by the passage of Cu containing ferrihydrite colloids less than 0.2 μm in size through the dialysis bag and membrane filter. Cu in these Cuferrihydrite colloids is determined as total dissolved Cu in acidified solutions by DPASV. In fact, DPASV-labile concentrations of Cu in nonacidified solutions were lower than the concentrations in acidified solutions (data not shown), thus 8071
dx.doi.org/10.1021/es3001932 | Environ. Sci. Technol. 2012, 46, 8069−8074
Environmental Science & Technology
Article
systems (Figure 2b) and was lower than at the end of the first equilibration in all systems. Thus, the perturbation stimuli caused by the addition of solid phases drove the systems to a new different equilibrium. System efficacy in Cu removal from solution followed the trend Smectitic ≈ Fe-Ox > Organic, and hence the systems with inorganic phases present initially (Smectitic and Fe-Ox systems) were more effective at removing Cu from solution than the Organic system. This behavior cannot be readily explained by the solid phases themselves because all systems had similar final or total state properties or variables. This behavior can however be explained by the higher concentration of DOC present in the Organic system (SI Figure S1) when ferrihydrite and K-SWy2 were introduced for the second equilibration, that is, soluble Cu−organic complexes formed during the first equilibration remained in solution and did not sorb onto the mineral phases (SI Figures S1−S3).18 In a study using a similar two-step equilibration (21 days each), McLaren et al.13 concluded the final sorption was highly influenced by the identity of the initial sorbent. Although our study does not necessarily contradict their findings it indicates soil dynamic properties (in this case the generation of DOC), and not solely the solid phases, determine total dissolved Cu concentrations in multiphase-multicomponent systems where a wide range of potential interactions and processes can occur through a wide range of reaction pathways. It is only recently that studies are taking into account both the distribution of the metal fractions in the solid phase and the speciation in solution.6,7,18,21,22 Copper Redistribution among Soil Solids: Second Equilibration. Figure 3 shows the distribution of Cu among ferrihydrite, LC, and K-SWy2 as a function of time during the second equilibration for the Fe-Ox, Organic, and Smectitic systems. This figure shows the individual contribution of each solid phase to the system reaction pathway drawn in Figure 2b. In the Fe-Ox system (Figure 3a and b), the mass of Cu sorbed by ferrihydrite during the second equilibration remained small (0.0073−0.01 mg Cu) while the dissolved Cu present at the end of the first equilibration (Figure 3a) was sorbed by the LC and K-SWy2 during the second equilibration. In the organic system, the mass of Cu retained by the LC decreased after addition of ferrihydrite and K-SWy2, thus indicating system disturbance caused desorption of previously sorbed Cu (Figure 3c and d). The newly desorbed Cu and that remaining in solution at the end of the first equilibration redistributed toward the LC and K-SWy2. A small mass of Cu ( Organic > FeOx when single solid phases (K-SWy2, LC and ferrihydrite, respectively) were equilibrated for 30 days (1st equilibration, Figure 2a). After system disturbance, the solid phase present in the first equilibration released a portion of its sorbed Cu, with the exception of ferrihydrite which had sorbed minimal amounts of Cu. This newly desorbed Cu, and the Cu that remained in solution at the end of the first equilibration, was again sorbed by the solid phases present during the second equilibration (i.e., ferrihydrite, LC, and K-SWy2). Once a second pseudo steady-state was established after equilibration for 8 months, system efficacy in Cu sorption was Smectitic ≈ Fe-Ox > Organic. Using compartmentalized reactors we show the dynamic aspects of a system, including the generation of DOC and the potential release and diffusion of metal-containing (nano)colloids to solution, are key factors in the trajectory that determines the final solid-solution partitioning (i.e., solubility) and solid-phase distribution of Cu. The dynamic properties of a system (i.e., soluble organics, colloidal particles) affect sorption onto soil solids. Although multiple pseudoequilibriums were reached through different reaction pathways, it is clear that for all systems the final partitioning and redistribution of Cu was determined by organic matter (solid and dissolved) and that montmorillonite and ferrihydrite played a minor role. In the presence of three soil bulk constituents, Cu redistributed between the LC and K-SWy2, but not toward ferrihydrite. These results are in agreement with recent evidence that suggest metal solubility in soil systems is controlled by organic matter and silicate clays rather than by iron oxides18,23,24 despite the capacity of iron oxides to retain metals.12,15,17,25 Furthermore, these results are attributed to reaction pathways where the generation-sorption of dissolved Cu−organic complexes and Cu-ferrihydrite colloids were operative. In natural soil systems, reaction pathways converging to equilibrium might be largely influenced by dissolved colloids (organic and inorganic), and their dynamics.
Figure 3. Distribution of Cu among ferrihydrite, leaf compost, and montmorillonite as a function of time during the 2nd equilibration. Left panels (a, c, e) show the results for the solid phase present during the 1st equilibration, whereas right panels (b, d, f) show the results for the solid phases introduced in the 2nd equilibration, per system. Open symbols (left panels) represent the mass of Cu (calculated from solution data) in each solid phase at the end of the 1st equilibration.
panels), limited desorption hysteresis was observed in our systems. To test the potential extent of Cu desorption, the solids of the first equilibration (30 days samples) were extracted eight consecutive times with a 0.01 M KNO3 solution and with a 5 mg/L DOC solution. Variable amounts of Cu were desorbed from ferrihydrite, LC, and K-SWy2 (Figure 4). DOC desorbed the greatest amounts of Cu from K-SWy2, whereas KNO3 was
■
ASSOCIATED CONTENT
* Supporting Information S
Three Figures and their interpretation. Figure S1 shows the concentration of DOC as a function of time during the 2nd equilibration. Figure S2 shows dissolved Cu species (total, labile, and free) as a function of DOC present in the 2nd equilibration. Figure S3 presents the concentration of total organic carbon sorbed onto the mineral solid phases as a function of DOC for the 2nd equilibration. This material is available free of charge via the Internet at http://pubs.acs.org.
Figure 4. Cumulative Cu desorbed from ferrihydrite, LC, and K-SWy2 at the end of the 1st equilibration. 8073
dx.doi.org/10.1021/es3001932 | Environ. Sci. Technol. 2012, 46, 8069−8074
Environmental Science & Technology
■
Article
(17) Lim, T. T.; Tay, J. H.; Teh, C. I. Contamination time effect on lead and cadmium fractionation in a tropical coastal clay. J. Environ. Qual. 2002, 31 (3), 806−812. (18) Martínez-Villegas, N.; Martínez, C. E. Solid- and solution- phase organics dictate copper distribution and speciation in multicomponent systems containing ferrihydrite, organic matter, and montmorillonite. Environ. Sci. Technol. 2008, 42 (8), 2833−2838. (19) McLaren, R. G.; Crawford, D. V. Studies on soil copper. II. The specific adsorption of copper by soils. J. Soil Sci. 1973, 24 (4), 443− 452. (20) McBride, M. B. Toxic metals in sewage sludge-amended soils: has promotion of beneficial use discounted the risks? Adv. Environ. Res. 2003, 8 (1), 5−19. (21) Wang, Q. Y.; Zhou, D. M.; Cang, L.; Li, L. Z.; Wang, P. Solid/ solution Cu fractionations/speciation of a Cu contaminated soil after pilot-scale electrokinetic remediation and their relationships with soil microbial and enzyme activities. Environ. Pollut. 2009, 157 (8−9), 2203−2208. (22) Zhou, D. M.; Wang, Q. Y.; Cang, L. Free Cu2+ ions, Cu fractionation and microbial parameters in soils from apple orchards following long-term application of copper fungicides. Pedosphere 2011, 21 (2), 139−145. (23) Terzano, R.; Spagnuolo, M.; Janssens, K.; Vekemans, B.; De Nolf, W.; Falkenberg, G.; Fiore, S.; Ruggiero, P. In Geochemical Forms of Heavy Metals in an Industrial Polluted Soil: A Critical Evaluation by Combined Synchroton Spectromicroscopy Techniques and Bulk Extraction Methods, 18th World Congress of Soil Science, Philadelphia, PA USA, July 9−15, 2006; International Union of Soil Sciences: Philadelphia, PA, 2006. (24) Weng, L. P.; Temminghoff, E. J. M.; van Riemsdijk, W. H. Contribution of individual sorbents to the control of heavy metal activity in sandy soil. Environ. Sci. Technol. 2001, 35 (22), 4436−4443. (25) Dzombak, D. A.; Morel, F. M. M. Surface Complexation Modeling; John Wiley & Sons, Inc.: New York, 1990.
AUTHOR INFORMATION
Corresponding Author
*Phone: (+52) 444-834-2000 x7272; fax: (+52) 444-834-2010; e-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This study was funded by the NRI-USDA (2003-3510713650). N.M.-V. was partially supported by CONACyT under fellowship No. 154889.
■
REFERENCES
(1) Bergkvist, P.; Berggren, D.; Jarvis, N. Cadmium solubility and sorption in a long-term sludge-amended arable soil. J. Environ. Qual. 2005, 34 (5), 1530−1538. (2) Hooda, P. S.; Alloway, B. J. Sorption of Cd and Pb by selected temperate and semi-arid soils: Effects of sludge application and ageing of sludged soils. Water, Air, Soil Pollut. 1993, 74, 235−250. (3) McGrath, S. P.; Cegarra, J. Chemical extractability of heavymetals during and after long-term applications of sewage-sludge to soil. J. Soil Sci. 1992, 43 (2), 313−321. (4) Sadovnikova, L.; Otabbong, E.; Iakimenko, O.; Nilsson, I.; Persson, J.; Orlov, D. Dynamic transformation of sewage sludge and farmyard manure components 0.2. Copper, lead and cadmium forms in incubated soils. Agric., Ecosyst. Environ. 1996, 58 (2−3), 127−132. (5) Lavado, R. S.; Rodriguez, M. B.; Taboada, M. A. Treatment with biosolids affects soil availability and plant uptake of potentially toxic elements. Agric., Ecosyst. Environ. 2005, 109 (3−4), 360−364. (6) Murtaza, G.; Haynes, R. J.; Naidu, R.; Belyaeva, O. N.; Kim, K.R.; Lamb, D. T.; Bolan, N. S. Natural attenuation of Zn, Cu, Pb and Cd in three biosolids-amended soils of contrasting pH measured using rhizon pore water samplers. Water, Air, Soil Pollut. 2011, 221 (1−4), 351−363. (7) Tichy, R.; Nydl, V.; Kuzel, S. Increased cadmium availability to crops on a sewage-sludge amended soil. Water, Air, Soil Pollut. 1997, 94 (3−4), 361−372. (8) Munksgaard, N. C.; Lottermoser, B. G. Effects of wood bark and fertilizer amendment on trace element mobility in mine soils, Broken Hill, Australia: Implications for mined land reclamation. J. Environ. Qual. 2010, 39 (6), 2054−2062. (9) Han, F. X.; Banin, A.; Kingery, W. L.; Triplett, G. B.; Zhou, L. X.; Zheng, S. J.; Ding, W. X. New approach to studies of heavy metal redistribution in soil. Adv. Environ. Res. 2003, 8 (1), 113−120. (10) Inaba, S.; Takenaka, C. Changes in chemical species of copper added to brown forest soil in Japan. Water, Air, Soil Pollut. 2005, 162 (1−4), 285−293. (11) Jalali, M.; Khanlari, Z. V. Redistribution of fractions of zinc, cadmium, nickel, copper, and lead in contaminated calcareous soils treated with EDTA. Arch. Environ. Contam. Toxicol. 2007, 53 (4), 519−532. (12) Lu, A. X.; Zhang, S. Z.; Shan, X. Q. Time effect on the fractionation of heavy metals in soils. Geoderma 2005, 125 (3−4), 225−234. (13) McLaren, R. G.; Williams, J. G.; Swift, R. S. Some observations on the desorption and distribution behaviour of copper with soil components. J. Soil Sci. 1983, 34 (2), 325−331. (14) Perez-de-Mora, A.; Madrid, F.; Cabrera, F.; Madejon, E. Amendments and plant cover influence on trace element pools in a contaminated soil. Geoderma 2007, 139 (1−2), 1−10. (15) Saha, J. K.; Mandal, B. Redistribution of copper in alfisols under submergence. II. Applied copper. Commun. Soil Sci. Plant Anal. 2000, 31 (9−10), 1121−1127. (16) Soumare, M.; Tack, F. M. G.; Verloo, M. G. Redistribution and availability of iron, manganese, zinc, and copper in four Malian agricultural soils amended with solid municipal waste compost. Commun. Soil Sci. Plant Anal. 2007, 38 (19−20), 2567−2579. 8074
dx.doi.org/10.1021/es3001932 | Environ. Sci. Technol. 2012, 46, 8069−8074
