Environ. Sci. Technol. 2002, 36, 3991-3996
Mechanisms of Attenuation of Metal Availability in In Situ Remediation Treatments REBECCA E. HAMON,* MIKE J. MCLAUGHLIN, AND GILL COZENS CSIRO Land & Water, PMB 2, Glen Osmond, South Australia, Australia 5064
One suite of in situ technologies for remediating metal contaminated soils involves the addition of reactive materials which lower metal availability. Until now it has been difficult to assess whether the amendment induced decrease in metal availability is due simply to increased sorption of the metal or whether it is the result of surface precipitation or other fixation mechanisms. This has made it difficult to predict the long-term stability of such remedial treatments. Using an isotopic dilution technique coupled with a stepwise acidification treatment, we examined changes in the labile pool of Cd and Zn in a polluted soil amended with either CaCO3, KH2PO4, redmud, or a kaolin byproduct. Fixation of both Cd and Zn was greatest in the KH2PO4 treated soil. The mode of fixation in this treatment was also found to be resistant to soil acidification. The results allowed a clear distinction between three classes of attenuation mechanisms which are hypothesized to increase in their resilience to environmental change as follows: reversible sorption < irreversible “fixation” at constant pH < irreversible “fixation” across a range of pH.
Introduction Significant effort has been undertaken to evaluate different amendment materials for use in in situ remediation of metal contaminated soils (1-5). For example, lime is widely regarded as a key ameliorant for decreasing metal bioavailability and toxicity in soils (6-8), but the effect of lime may be ephemeral if the soil reacidifies (9). In lime-amended soils where metal availability poses risks to humans or the environment, there may therefore be a need to manage soil pH in perpetuity if toxicity is to be avoided. Other materials added to soil, such as phosphorus-, iron-, or manganeserich materials, zeolites, and clays, are designed to remediate soils not only by affecting soil pH but also by binding metals in forms not in rapid equilibrium with the soil solution, with the aim of having a more persistent remediation effect (10, 11). Reductions in metal concentrations in solution can be achieved by increasing adsorption (i.e. increasing the soil: solution partition coefficient or Kd) (11) so that metals are so strongly held that uptake by organisms is limited. Solution metal concentrations can also be reduced by entrapment of metals in crystal lattices, which may be important for materials containing hydrous oxides (12). Finally, precipitation of soluble metals can also be used to minimize environmental risks (13, 14). * Corresponding author phone: 61 8 8303 8489; fax: 61 8 8303 8565; e-mail:
[email protected]. 10.1021/es025558g CCC: $22.00 Published on Web 08/16/2002
2002 American Chemical Society
The mechanisms of metal fixation by materials are not fully known; however, a question often posed by regulators is whether the remediation mechanisms are permanent, or reversible, through time. If in situ remediation materials exert their effect through changes in soil pH only (15), then presumably reacidification of soil (9) will return metal bioavailability to the original toxic level. If remediation materials raise soil pH and sorb metals strongly, then reacidification of soil will raise metal bioavailability but at a reduced level compared to the original soil, depending on the soil and amendment pH buffer capacity and the rate of change in amendment Kd with pH. If remediation materials promote fixation of metals in nonlabile pools in the soil through entrapment in crystal lattices or precipitation, then the attenuation of availability will depend on factors such as solid-solution equilibrium kinetics, or the solubility product (Ksp) of the solid phase formed, and is likely to be more enduring. Isotopic exchange and dilution techniques are useful tools to examine the lability of metals in soils (16-18). When isotopic tracers are added to soil solution, they rapidly exchange with the surface-bound metal that is in direct equilibrium with the soil solution. In contrast, metals tightly held in crystalline mineral precipitates or occluded in mineral lattices are only very slowly exchangeable (19). We describe here a novel application of these isotopic exchange techniques to determine mechanisms controlling metal availability in soil remediated by a range of materials including lime, phosphate, and industrial byproducts. This application has a further, complementary function in that it can be also used for assessing the potential for reversibility of the remediation process.
Methods Treatments. Soil was sampled from a USEPA Region 7 CERCLA site subject to extensive contamination with heavy metals entrained in mine tailings that were deposited in the area during the mid 1900s. Metal contents and other soil properties are given in Table 1. The soil was air-dried and sieved < 2 mm before being subdivided and mixed with a range of amendments. The amendments trialled in this experiment are reported on a dry weight basis and were as follows: (a) CaCO3 applied at a rate of 10 g/kg soil, (b) KH2PO4 applied at a rate of 20 g of P/kg of soil. To neutralize the acidity introduced by the phosphate addition, CaO was added giving a final treatment pH of 7.2. The treatment was then leached until the leachate EC fell below 2 dS/m (c) a 1:1 (w/w) mixture of red mud and biosolids applied to the soil (“redmud treatment”) at a final rate of 20 g mixture/kg soil, and (d) a 1:1 (w/w) mixture of kaolin amorphous derivative (KAD) and biosolids applied to the soil (“KAD treatment”) at a final rate of 20 g of mixture/ kg of soil. The redmud was sourced as a byproduct from the aluminum industry and due to its high initial salt content and alkalinity was leached with deionized water until the EC fell below 2 dS/m prior to mixing with the biosolids. The biosolids were sampled from a metropolitan treatment works in Adelaide, South Australia. The KAD is supplied as an ameliorant for water purification and is a nanocrystalline stable aluminosilicate with variable but reproducible types of micro- and meso-porosity. The batch of KAD used in this experiment was obtained from Nanochem Ltd., Queensland and had a cation exchange capacity (for 1 M ammonium exchange) of 232 mequiv/100 g and a BET surface area of 207 m2/g. Other properties of the soil and amendment materials are given in Table 1. The amendments were incorporated VOL. 36, NO. 18, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3991
TABLE 1. Properties of the Soil and the Amendment Materials That Were Added to the Soil pH (H2O) total Cd (mg kg-1)a total Zn (mg kg-1)a carbon (%)b Al ox (mg kg-1)c Al dith (mg kg-1)d Fe ox (mg kg-1)c Fe dith (mg kg-1)d
soil
redmud
KAD
biosolid
5.9 92 18540 0.12 116 97 3230 5920
10.8 6 36 nde 20510 3750 2050 17830
10.9 12 106 nde 58620 40770 1200 1700
6.4 12 642 nde 2700 1600 5800 7380
a Method after ref 31. b Method after ref 32. c Ox indicates oxalate extractable metal, method after ref 32. d Dith indicates citrate/dithionite extractable metal, method after ref 32. e nd indicates not determined.
into dry soil by thorough mixing by hand. Following incorporation, the moisture content of the treatments was increased to field capacity (-12 kPa), and the treatments were left to incubate at room temperature for 4 weeks, after which time they were air-dried and stored dry for 8 months prior to experimentation. Determination of Labile Pools of Metals. Experiment 1. Changes in the available pool of soil Cd and Zn following amendment additions were determined using the isotopic dilution E-value method (20). For this method, the soil samples (5 g) were placed in centrifuge tubes to which was added 50 mL of double deionized (DD) H2O and 2 drops of toluene to minimize microbial activity. The samples were shaken on an end-over-end shaker for 4 d at which time they were spiked with 0.1 mL of solution containing 109Cd (200 kBq/mL) and 65Zn (1500 kBq/mL). The samples were returned to the shaker for a further 3 d. The samples were then centrifuged at 2000g for 15 min, a 10 mL aliquot of the supernatant was filtered through a 0.2 µm membrane filter (Sartorius), and the quantity of cold and radioactive Cd and Zn in the solutions was measured. Unlabeled Cd was measured using a graphite furnace atomic absorption spectrometer; Zn and a range of other elements were measured by inductively coupled plasma atomic emission spectrometry. Activities of radioactive Cd and Zn were assessed using gamma spectrometry, with care taken to correct for the interference of 65Zn on 109Cd. All analyses were performed in triplicate and included true blanks as well as blank solutions spiked with radioisotope so that an accurate determination could be made as to the total amount of radioisotope added to each sample. The labile pool (Ea) of Cd and Zn in each of the samples was determined as follows
Ea ) (Csol/C*sol) × R × (V/W)
(1)
where Csol is the concentration of cold Cd or Zn in solution (µg/mL), C*sol is the concentration of radioisotope remaining in solution after the 3 d equilibration time (Bq/mL), R is the total amount of either radioisotope that was added to each sample (Bq/mL), and V/W is the ratio of solution to sample, which in this case was 10 mL/g. The amount of isotopically exchangeable metal associated specifically with the solid phase (Ee) of each treatment was calculated as follows
Ee ) (Csol/C*sol) × (R - C*sol) × (V/W)
(2)
In this paper, “nonlabile” or “fixed” metal is defined as that pool which is nonisotopically exchangeable after up to 7 days of soil:isotope contact (i.e. nonlabile metal (mg/kg) ) total soil metal - Ea). Experiment 2. Subsamples of each of the treatments were titrated with HCl to determine the amount of acidity to add to provide a series of 6 pH levels for each treatment (or 8 pH 3992
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 18, 2002
levels for the unamended soil) which would range from the original starting pH down to a level below the starting pH of the unremediated soil. On day 0, samples (5 g) were placed in centrifuge tubes to which was added DDH2O, 2 drops of toluene, and an aliquot of dilute HCl as determined from the titration, giving a final solution volume of 50 mL. The samples were placed on an end-over-end shaker for 4 days at which time the solution pH was assessed, and solutions were spiked (day 4) with isotope for the E-value procedure. The Ea- and Ee-values were determined as above, except aliquots of solution were extracted from the samples at days 7 and 11 for both E-value and pH assessment and day 18 for pH assessment only.
Results and Discussion Experiment 1. Effect of Amendments on Labile Pools of Cd and Zn. The soil pH was increased by the addition of amendments, and the solution concentration of both Cd and Zn markedly decreased compared to the unamended soil (Table 2), with the magnitude of decrease being in the order KH2PO4 > lime > KAD > redmud for both metals. A decrease in the most available (i.e. solution) fraction of metals can be due to (a) an increase in the amount of metal exchangeably sorbed to the solid-phase through either an increase in the number of sorption sites or an increase in the strength of sorption (Kd); however, such metal would still form part of the bioavailable pool (21, 22) and/or to (b) an increase in the amount of metal fixed in nonexchangeable solid-phase pools, but in this case the metal may be removed from the bioavailable pool (18) therefore constituting a more effective mechanism for reduction of metal availability at a given soil metal concentration. The data presented in Table 2 show that in each of the treatments the amount of exchangeably sorbed Cd (Ee) was significantly increased and % labile Cd significantly decreased compared to the unamended soil, revealing that there was a net movement of metal to both the sorbed and fixed pools following addition of the amendments. In contrast to Cd, the amount of exchangeably bound solid-phase Zn (Ee) was found to decrease (Table 2) following addition of all amendments except lime. This indicates that for each of the treatments except lime, there was a net movement of Zn to the fixed pool only, whereas for the lime treatment, there was again a net movement of metal to both the sorbed and fixed pools. Experiment 2. Stability of Equilibrium Following Acidification. The pH values of the solutions assessed in the samples at days 4 and 18 were found to be consistent between the two measurements (Figure 1). This implies that changes in soil chemistry as a result of the addition of acid had reached their new state of equilibrium within 4 days for all treatments. Activities of 109Cd and 65Zn measured in solution at days 7 and 11 were also found to be equivalent (Figure 2 a and b), confirming that a 3 day incubation with isotope was sufficient time to allow the rapid phase of isotopic exchange to be completed for all of the treatments at all pH levels. Effect of pH on Metal Solubility and Sorption. As would be expected, concentrations of Cd and Zn in solution increased with decreasing sample pH for all treatments (Figure 3 a and b). While not apparent from the log presentation of data in Figure 3, the concentration of metals in solution from all except the KH2PO4 treatment increased linearly with a decrease in pH from 7 to at least 5.5. In the unamended soil, which was the only treatment subject to acidification to pH levels < 4.5, the solution metal concentrations appeared to tend toward a plateau at the lowest pH levels. Additional experimentation is required to determine whether this plateau is the result of the acid labile pool of soil metals being fully solubilized at the lower pHs, or whether it is due to retardation of further metal dissolution from the soil because products of the acid dissolution were not
TABLE 2. Metal Partitioning between Different Labile Pools in Response to the Different Amendment Treatmentsb soil treatment unamended lime KH2PO4 redmud KAD
solution pH 6.4 7.7 7.0 6.8 6.8
solution metal (mg kg-1)
Ee value (mg kg-1)
labile metal (%)a
Cd
Cd
Zn
Cd
Zn
632 (2) 662 (3) 224 (6) 601 (1) 535 (32)
3.2a
5.9a 3.6b 1.2c 3.6b 3.1b
2.2 (0.04) 0.19 (0.003) 0.003 (0.001) 0.58 (0.02) 0.27 (0.02)
Zn 465 (2.0) 10 (0.2) 0.64 (0.04) 60 (2.0) 36 (2.0)
a Expressed on a basis of total soil metal content as shown in Table 1. letter are significantly different (p < 0.05).
b
0.75 (0.02) 2.0 (0.01) 1.4 (0.1) 1.2 (0.05) 1.4 (0.1)
2.4b 1.5c 1.6c 1.8c
Standard error in brackets, numbers in columns followed by a different
FIGURE 1. A comparison of the solution pH measured at 4 and 18 days following addition of acid. removed from the system due to the experiment being conducted in a batch system. The concentration of metals in the KH2PO4 treatment increased exponentially with decreasing pH. However, irrespective of the sample pH, the metal concentration in solution was maintained at a level several orders of magnitude lower in this treatment compared to the others (Figure 3a,b). A comparison of the treatments shows that across the range of pH 7 to 5.0, the Cd concentration in solution decreased according to the following ranking: unamended ) lime > redmud > KAD . KH2PO4. For solution Zn at pH levels between 6.0 and 5.0, the ranking followed the order unamended ) lime ) redmud > KAD . KH2PO4. However, at pH levels between 6.5 (corresponding to the original unamended soil pH) and 6.0, the Zn concentration in solution from the lime amended soil appeared to be up to two times greater than concentration in the unamended soil at an equivalent pH. This was also reflected by a significantly greater Ea-value obtained for Zn in the lime vs unamended soil over the same pH range (see below). Though the errors associated with these data are insignificant (see Figure 5 b below), it is possible that this observation may be an artifact of the experimental procedures. However, the consequences of it being a real effect are critical, as it suggests that the common practice of liming to reduce metal availability may in fact exacerbate the problem if extreme care is not taken to ensure that the soil pH is permanently maintained at a high level once lime has been applied. A simple simulation of Zn, Ca, and CO3 interactions was performed using GEOCHEM (23). The concentrations of each species used for input data corresponded to the estimated acid labile pool of each of the species in either the unamended or lime treated soil. For the simulation, the Ea value for Zn in the lowest pH treatment of the unamended soil represented the acid labile
FIGURE 2. (a and b) A comparison of the activity of Cd or Zn measured in solution at 3 (day 7) and 7 (day 11) days following the addition of the radioactive spike. pool of this element, the concentrations of Ca measured in solutions from the lowest pH treatment of the unamended or lime treated soil were used to represent low and high Ca, respectively. The carbonate concentration used in the simulation was the amount of carbonate introduced to the soil with the lime, and all simulations were conducted assuming an atmospheric pCO2. The results of the simulation are shown in Figure 4 and suggest that the presence of Ca can indeed lead to higher concentrations of soluble Zn across the pH range (6-6.5) at which solution Zn was also found to be higher in the lime amended compared to unamended soil. Hence the observation of a higher Ea-value for Zn in the limed treatment compared to the unamended soil upon acidification may be a real effect, attributable to the influence of Ca on Zn-carbonate solubility. Characterization of Attenuation of Metal Availability by Amendments. For many years there has been an unresolved question as to whether the increased partitioning of metals to the soil solid-phase in response to raising the soil pH involves enhancing sorption to readily exchangeable sites or whether the increased partitioning is the result of precipitation reactions (24, 25). We suggest that an investigation of the effect of pH on the size of the isotopically exchangeable pool (Ea-value) can be used to address this VOL. 36, NO. 18, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3993
FIGURE 3. (a and b) Changes in the concentration of metal in solution with respect to pH, different soil treatments are as follows: unamended soil, 0; lime amended, b; redmud amended, 1; KAD amended, 3; KH2PO4 amended, O.
FIGURE 4. Modeled changes in the amount (% of total) of various phases in simple system containing similar quantities of Zn and CO3 as the lime-amended soil in the presence of high (filled symbols) or low (open symbols) Ca. Circles indicate Zn2+, triangles indicate precipitated ZnCO3. question and thus also provide greater information on mechanisms responsible for the attenuation of metal availability by different in situ remediation materials and hence their potential reversibility (26). As shown in Figure 3, the 3994
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 18, 2002
FIGURE 5. (a and b) Changes in the labile pool (Ea) of Cd or Zn with respect to pH, different soil treatments are as follows: unamended soil, 0; lime amended, b; redmud amended, 1; KAD amended, 3; KH2PO4 amended, O. metal concentrations in solution were greatly affected by pH for all of the different treatments. If these changes were related solely to readjustment of metal partitioning between solution and exchange sites in the solid-phase (i.e. adsorption), then the absolute size of the pool of labile metals would not be affected and hence there would be no effect on Ea-values of either increasing or decreasing the pH. However, if an increase in pH does lead to irreversible sorption/precipitation of metals, the size of the labile pool would therefore be decreased, and this would be measured by a decrease in the Ea-value. Conversely, if metals are held within irreversibly sorbed/precipitated (i.e. nonexchangeable) pools and a decrease in pH causes dissolution of such pools, then the size of the labile pool, and hence the Ea-value, would increase upon acidification. The effects of pH on the Ea-values for Cd and Zn in the different treatments are shown in Figure 5a,b. In the unamended soil, the Ea-values for both Cd and Zn increased sharply with a decrease in pH demonstrating that acidification led to liberation of a large pool of nonexchangeable metal in this soil. As discussed above, increasing the soil pH through liming led to a decrease in the size of the labile pool for both metals, illustrating that the decrease in solution metal concentration was at least in part the result of irreversible (at constant pH) sorption/precipitation reactions. However, apart from the discrepancy between pH 6.0-6.5 for Zn as noted above, acidification of the lime treatment gave similar Ea-values for both metals as the unamended soil. This indicates that the effect of lime on decreasing the labile pool of elements is transient, that is, application of lime resulted in transfer of metals into a discrete nonlabile phase that is
Sustainability of Metal Fixation by the Different Materials. In terms of assessing the overall effectiveness of in situ remediation treatments, it is important to know not only how much metal is removed from the bioavailable pool but also how persistent the attenuation of availability is. The environmental factor likely to have the largest effect on availability of metals which exist in cationic forms, such as Cd and Zn, is soil acidification (9, 28, 29); however, other factors, such as changes in redox status, should also be explored. Soil acidification can occur on a broad scale, such as through aerial deposition or in high-rainfall forest ecosystems, or can be localized, such as in the rhizosphere (29, 30). Hence, if attenuation of metal availability by the remediation treatment is not resistant to acidification, the contaminated soil will continue to pose a potential ecological risk and the remediation treatment cannot be said to be sustainable without continued management inputs. Of the materials tested here, only phosphate and KAD demonstrated a remediation effect for both Cd and Zn that was resistant to acidification. We suggest that the isotopic dilution method coupled with a stepwise acidification treatment may be a useful way to screen for remediation treatments which will be effective and robust over both the short- and longerterm. FIGURE 6. (a and b) Changes in the amount of metal exchangeably adsorbed to the soil solid-phase (Ee) with respect to pH, different soil treatments are as follows: unamended soil, 0; lime amended, b; redmud amended, 1; KAD amended, 3; KH2PO4 amended, O. only present at high pH rather than promoting fixation in a pH independent nonlabile pool. Figure 4 suggests that from pH 7.5 to 6, dissolution of a carbonate fraction of metals is likely to have been responsible for the increase in Ea-values with acidification in these two treatments. In contrast, the Ea-values for Zn in the KAD treatment and Cd in the KAD and redmud treatments were lower than either the lime or unamended treatments irrespective of pH. This demonstrates that application of red mud or KAD led, respectively, to fixation of Cd or both Cd and Zn in nonlabile forms that were also partially resistant to acid dissolution and therefore suggests that the effect of these materials on attenuation of metal lability was the result of mechanisms in addition to simply a pH effect. The Ea-values obtained for Cd and Zn in the KH2PO4 treatment were significantly lower than in all other treatments (Figure 5a,b). The size of the labile pool of Cd and Zn increased with acidification to pH 5.6 and pH 6.0, respectively, at which point there was no further increase in the Ea-values, signifying no further dissolution of a nonlabile fraction of metals. Interestingly, pH 6 corresponds to the point at which carbonate solubility is unlikely to have any further control on the solubility of Zn in solution (Figure 4), hence it is possible that the increase in Ea-value upon acidification to this pH level was due to dissolution of a carbonate fraction of metals. Below pH 6, there were increases in solution concentration of metals in this treatment; however, these were the result of desorption of exchangeably sorbed metals and not because of dissolution of a nonlabile pool of metals (Figure 5a,b). While both the Ea-value and solution concentrations of metals were significantly lower than in the other treatments, the quantity of both Cd and Zn exchangeably adsorbed to the solid phase was dramatically higher in the KH2PO4 treatment compared to the others at pH values below 6.5 (Figure 6 a and b). This may have been due to an increase in surface negative charge induced by greater phosphate sorption at lower pH (27) leading to a greater (reversible) sorption of cationic metals to the solid phase at pH values below 6.5.
Acknowledgments The authors thank Dr. Sally Brown and Mark Doolan for providing the soil, Verity Ferguson for preparation of the soil amendment treatments and Dr. Enzo Lombi for constructive comments on the manuscript. This project was supported in part through funding provided by the International Lead/ Zinc Research Organization.
Literature Cited (1) Mench, M.; Vangronsveld, J.; Didier, V.; Clijsters, H. Environ. Pollut. 1994, 86, 279-286. (2) Vangronsveld, J.; Sterckx, J.; Van Assche, F.; Clijsters, H. J. Geochem. Explor. 1995, 52, 221-229. (3) Boisson, J.; Mench, M.; Sappin-Didier, V.; Solda, P.; Vangronsveld, J. Agronomie 1998, 18, 347-59. (4) Chaney, R. L.; Brown, S. L.; Stuczynski, T. I.; Daniels, W. L.; Henry, C. L.; Li, Y. M.; Siebielic, G.; Malik, M.; Angle, J. A.; Ryan, J. A.; Compton, H. In Innovative Clean-Up Approaches: Investments in Technology Development, Results and Outlook for the Future; USEPA: in press. (5) Berti, W. R.; Cunningham, S. D. Environ. Sci. Technol. 1997, 31, 1359-1364. (6) Andersson, A.; Nilsson K. O. Ambio 1974, 3, 198-200. (7) King, L. D.; Morris, H. D. J. Environ. Qual. 1972, 1, 425-429. (8) Bisesar, S. Sci. Total Environ. 1989, 84, 83-90. (9) Doerge, T. A.; Gardner, E. H. Soil Sci. Soc. Am. J. 1985, 49, 680685. (10) Chlopecka, A.; Adriano, D. C. Sci. Total Environ. 1997, 207, 195206. (11) In Interactions at the Soil Colloid-Solution Interface; De Boodt, M. F., Bolt, G. H., et al., Eds.; Kluwer: Dordrecht, The Netherlands, 1991. (12) Manceau, A.; Charlet, L.; Boisset, M. C.; Didier, B.; Spadini, L. Appl. Clay Sci. 1992, 7, 201-223. (13) Nriagu, J. O. Geochim. Cosmochim. Acta 1974, 38, 887-898. (14) Ma, Q. Y.; Traina, S. J.; Logan, T. J. Environ. Sci. Technol. 1993, 27, 1803-1810. (15) Oste, L. A.; Dolfing, J.; Ma, W. C.; Lexmond, T. H. Environ. Toxicol. Chem. 2001, 20, 1339-1345. (16) Tiller, K. G.; Honeysett, J. L.; de Vries, M. P. C. Aust. J. Soil Res. 1972, 10, 165-182. (17) Smolders, E.; Brans, K.; Foldi, A.; Merckx, R. Soil Sci. Soc. Am. J. 1999, 63, 78-85. (18) Hamon, R. E.; McLaughlin, M. J.; Naidu, R.; Correll, R. Environ. Sci. Technol. 1998, 32, 3699-3703. (19) McAuliffe, C. D.; Hall, N. S.; Dean, L. A.; Hendricks, S. B. Soil Sci. Soc. Am. Proc. 1947, 12, 119-123. (20) Young, S. D.; Tye, A.; Carstensen, A.; Crout, N. R. Eur. J. Soil Sci. 2000, 51, 129-36. VOL. 36, NO. 18, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3995
(21) Nakhone, L. N.; Young, S. D. Environ. Pollut. 1993, 82, 73-77. (22) Tiller, K. G.; Wassermann, P. In Isotopes and Radiation in SoilPlant Relationships Including Forestry; 1972; pp 517-529. (23) Parker, D. R.; Norvell, W. A.; Chaney, R. L. In Chemical Equilibrium and Reaction Models; SSSA Spec. Publ. 42. SSSA and ASA; Loeppert, R. H., et al., Eds.; Madison, WI, 1995; pp 253-269. (24) Bru ¨ mmer, G.; Tiller, K. G.; Herms, U.; Clayton, P. M. Geoderma 1983, 31, 337-354. (25) Tiller, K. G. In Contaminants and the Soil Environment in the Australasia-Pacific Region; Naidu, et al., Eds.; Kluwer Academic Pubs.: 1996; pp 1-27. (26) Ford, R. G.; Scheinost, A. C.; Sparks, D. L. Adv. Agron. 2001, 74, 41-62. (27) Kuo, S.; McNeal, B. L. Soil Sci. Soc. Am. J. 1984, 48, 1040-1044.
3996
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 18, 2002
(28) McBride, M. B.; Sauve´, S.; Hendershot, W. H. Eur. J. Soil Sci. 1997, 48, 337-346. (29) Hesterberg, D. Agric. Ecosyst. Environ. 1998, 67, 121-133. (30) Hamon, R. E.; Lorenz, S. E.; Holm, P. E.; Christensen, T. H.; McGrath, S. P. Plant Cell Environ. 1995, 18, 749-756. (31) McGrath, S. P.; Cunliffe, C. H. J. Sci. Food Agric. 1985, 36, 794798. (32) Rayment, G. E.; Higginson, F. R. Australian Laboratory Handbook of Soil and Water Chemical Methods; Inkata Press: 1992; pp 1-330.
Received for review February 3, 2002. Revised manuscript received July 2, 2002. Accepted July 10, 2002. ES025558G
