Effects of organic ligands and pH on the leaching of copper from brake wear debris in model environmental solutions

Abstract

Copper leaching from a disc brake wear debris sample was examined in a variety of aqueous solutions to simulate potential leaching processes during rain events and in surface waters. Synthetic rainwater leached 40% of the total copper present in the brake wear debris into solution after 18 h in batch reactors, which was approximately three times more copper than that extracted by the US Environmental Protection Agency's Synthetic Precipitation Leaching Procedure. Formate and acetate were responsible for the enhanced copper leaching, as demonstrated by higher average amounts of leached copper in synthetic rainwater with- versus without the organic acids (40 versus 31% recovery). This observation suggests leaching tests that do not incorporate the appropriate types and concentrations of organic ligands present in rainwater will likely underestimate copper mobilization from brake wear debris during rain events. Leaching of copper from the brake wear debris ranged from 23 to 40% in solutions containing 3 to 15 mg C L⁻¹ dissolved humic substances, and was higher still in solutions containing relatively high concentrations of the synthetic metal chelating agent ethylenediaminetetraacetic acid. Static pH tests demonstrated that copper leaching from brake wear debris is highly pH dependent, with more leaching occurring at lower solution pH values. Leaching rate studies revealed that equilibrium generally was not attained within 48 h in the model solutions, indicating that additional copper can be expected to be released in environments where brake wear debris is exposed to long-term leaching processes.

---

Introduction

Storm water runoff has been recognized as a major non-point source of pollutants in watersheds near urban areas.¹ For example, rainfall runoff from urban roadways often contains appreciable concentrations of metals that have adverse effects on ecological systems and human health.^2,3 A substantial portion of the metals contained in roadway runoff is believed to originate from vehicle component wear (brake and tire) and fluid leakage.^1–3 Although there have been reports of roadway runoff water quality, particularly those examining correlations between metal concentrations and factors such as rainfall intensity and traffic volume,^2–4 we still lack adequate information on important environmental processes such as metal leaching characteristics from specific non-point sources.

Wear debris from automobile brake pads have been identified as a potential non-point source of copper in surface waters. For example, based on field measurements, laboratory studies and simplified estimated annual loadings, Davis et al.¹ concluded that brake wear debris would be a major source of copper in non-industrial urban storm water runoff. Because of its potential contribution to copper impairments in streams and urban watersheds, brake wear debris characterization with respect to loadings, transport, fate, and environmental effects is becoming more of a pressing issue worldwide.

In a companion study,⁵ we conducted two common standardized leaching tests, the US Environmental Protection Agency's (US EPA) Toxicity Characteristic Leaching Procedure (TCLP) and the State of California's Waste Extraction Test (WET), to quantify the rates and extents of copper mobilization from a disc brake wear debris sample. Although results from these two standardized tests are expected to help predict the potential for long-term copper leaching in urban watersheds, additional knowledge of the leaching characteristics by more environmentally relevant solutions is still needed. For example, the chemical composition and component concentrations in rainwater or storm water runoff are highly different from the TCLP and WET leaching solutions. Of common regulatory leaching tests, the US EPA's Synthetic Precipitation Leaching Procedure (SPLP) may be an alternative method to evaluate copper leaching from brake wear debris by storm water runoff. The SPLP simulates the extraction of elements from wastes by percolating rainfall, utilizing a 60∶40 weight percent mixture of sulfuric acid and nitric acid.⁶ However, because of its strictly inorganic composition, the SPLP may underestimate metal leaching by actual environmental solutions. For example, reported chemical compositions of rainwater in urban areas reveal that organic ligands such as formate and acetate are typically present at concentrations ranging up to 100 µM.^7,8

Urban storm water runoff often contains appreciable concentrations of natural and synthetic ligands that can complex metals. It has been reported that concentrations of dissolved organic carbon (DOC) in runoff tend to increase as the rainfall passes through plants and soils before reaching surface waters.⁹ This process thus explains the sharp increase in surface water DOC concentrations generally observed with rainfall events.^10,11 As typical natural organic ligands, humic substances (HS) contain various moieties such as carboxyl, hydroxyl and amino groups that can complex metal cations in various ways.¹² Consequently, the solubility and mobility of toxic metals are strongly influenced by the amounts and types of HS and other naturally-occurring and/or synthetic ligands.

Another factor of expected importance for metal mobilization is solution pH, because it is a critical parameter in determining the solubility and speciation of metals.¹³ Therefore, one might expect copper leaching from brake wear debris to be highly dependent on pH. A wide range of pH values (3.0 to 8.0) has been reported in rainfall near urban areas, the lower values resulting from acidification of rainfall by anthropogenic input of SO₂ and NO₂ from industrial and traffic activities.^7,14

To date, only a few studies have actually quantified copper leaching from brake wear debris.^1,5 Furthermore, to our knowledge, the importance of environmentally-relevant concentrations of the organic ligands present in rainwater and storm water runoff has not been evaluated. Therefore, the objective of this study was to investigate the short-term copper leaching behavior from brake wear debris in model environmental solutions. In particular, we focused on the role played by organic ligands typically found in rainwater, storm water runoff, and surface waters in this process. Copper leaching in 18 h tests was examined in various aqueous solutions, including the SPLP solution, acidified deionized distilled water (DDW), synthetic rainwater, HS solutions, and solutions with ethylenediaminetetraacetic acid (EDTA). Static pH and kinetic leaching tests were also conducted to examine the effects of solution pH and contact time on the copper leaching process. Finally, because iron was the most abundant element present in the brake wear debris and could have potentially interfered with copper leaching,⁵ its leaching from the wear debris was also examined.

Experimental

Materials and reagents

Details of the generation and characterization of the brake wear debris sample used in this study have been previously reported.⁵ The total copper and iron contents of the brake wear debris were 10.8 and 28.6% (g/g), respectively. All solutions were prepared using DDW (Mega-Pure System MP-6A, Corning, NY) having a resistivity >18.0 MΩ cm. Reagents used to prepare synthetic rainwater solutions were potassium sulfate (K₂SO₄, EM Science), sodium nitrate (NaNO₃, EM Science), magnesium sulfate (MgSO₄, Sigma), ammonium chloride (NH₄Cl, Sigma), calcium chloride (CaCl₂·2H₂O, EM Science), sodium formate (HCOONa, Sigma), and sodium acetate (C₂H₃O₂Na, Sigma). All purchased chemical reagents were of analytical grade or better. The chemical composition of the synthetic rainwater was based on a report of the species in typical rainwater collected in Los Angeles (USA) from 1985 to 1991.⁷ Concentrations of the chemical species contained in the synthetic rainwater are shown in Table 1. To maximize the effects of the organic acids (formate and acetate) on copper leaching, concentrations near the reported maxima were used for the synthetic rainwater in this study.

Table 1 Chemical composition of synthetic rainwater leaching solutions and typical Los Angeles rainwater^a

Species	Synthetic rainwater	Inorganic synthetic rainwater	Reported range of Los Angeles rainwater^b
a Units for all values shown are µM. b Data from Sakugawa et al.^7
Na^+	190	190	1.6–254
K^+	20	20	0.2–46.2
NH4^+	100	100	0.41–617
Ca^2+	50	50	0.6–232
Mg^2+	30	30	0.4–64.2
Cl^−	200	200	1.2–285
NO3^−	100	100	1.9–562
SO4^2−	90	60	3.1–170
Formate	70	0	0.9–71.1
Acetate	20	0	0.2–17.6

---

The SPLP solution was prepared following standard procedures⁶ to obtain leaching solutions of pH 5.0. Two types of HS, a terrestrial HS and an aquatic HS, were utilized for this study. Soil humic acid (SHA) and Suwannee River fulvic acid (SRFA) were obtained from the International Humic Substances Society (IHSS) and used without further purification. Both HS were dissolved in DDW and stored at 4 °C at concentrations of ∼5 g C L⁻¹. General physicochemical characteristics of the HS have been previously reported.¹⁵ The carboxyl group contents of SHA and SRFA, determined by the IHSS, were 7.57 and 10.44 meq g⁻¹ C, respectively. EDTA (EM Science, >99.0%), a synthetic organic metal chelating agent commonly found as a pollutant in surface waters,^13,16 was used for comparison with the HS solutions. EDTA is a tetraprotic acid that can occupy up to six coordinating positions (i.e., hexadentate ligand) of a central metal ion.^13,16 New high density polyethylene (HDPE) plastic bottles (Wheaton, Millville, NJ) were used for each extraction without additional treatment. All other bottles and glassware used were soaked in 10% HNO₃ overnight and then rinsed with copious amounts of DDW.

Leaching tests

Because of the small quantity of brake wear debris available,⁵ a low solid-to-liquid ratio (1∶10⁴) was maintained for all leaching experiments. For the 18 h leaching tests, an appropriate mass (0.0025 g) of brake wear debris was weighed out in each bottle-point reactor (30 mL, Wheaton HDPE bottles) before transferring a fixed volume (25 mL) of leaching solution to each bottle. These tests were conducted in triplicate using three separate reactors containing the same leaching solution. All reactor bottles were equilibrated in the dark at room temperature (21 ± 2 °C) on an end-over-end tumbler rotating at approximately 25 rpm. At the end of the 18 h leaching tests, particulate materials were removed from the samples using an acid-washed 0.2 μm nylon membrane syringe filter (Gelman Laboratory), and the first 2 to 3 mL of each filtrate was discarded before collecting a filtered sample for copper and iron analyses. The pH of each filtered sample was measured to monitor pH changes during the leaching process.

The static pH leaching procedure was similar to that used above for the regular 18 h leaching tests, except for maintaining constant solution pH values by periodic adjustment with 0.1 N HNO₃ or 0.1 N NaOH as needed during mixing. Once the pH deviation for each sample was confirmed to be less than ±0.2 from its target value, an additional 18 h mixing period was conducted for a total combined nominal leaching time of 48 h. All static pH leaching tests were conducted with duplicate samples.

For the kinetic leaching studies, an appropriate mass (0.02 g) of brake wear debris was added to 250 mL reactors (Nalgene HDPE bottles) followed by 200 mL of leaching solution. Nominal sampling times for the kinetic studies (0.3, 1, 3, 10, 24, and 48 h) were selected based on preliminary experiments and encompassed the standard SPLP time period of 18 h. At appropriate times, each reactor was vigorously shaken to achieve sample uniformity before taking a fixed volume (12 mL) of total sample (i.e. solution plus particulate brake wear debris). This procedure ensured that a constant solid-to-liquid ratio was maintained throughout the kinetic leaching experiments despite the reduced total sample volume. In addition, the absence of measurable heterogeneity in the particulate brake wear debris⁵ ensured that each sample collected did not skew the later sampling results due to fractionation effects. After sampling was completed, each reactor was then placed back on the tumbler for additional mixing and the samples were filtered and analyzed as described above. All kinetic leaching tests were conducted with duplicate samples.

Analytical methods

All filtered samples were transferred to clean 15 mL plastic bottles (Wheaton HDPE bottle), acidified to pH 3 by adding 0.1 N nitric acid, and then analyzed for total copper and iron by inductively coupled plasma atomic emission spectrometry.⁵ Operational parameters and conditions for the ICP analyses and quality control/assurance procedures have been previously reported.⁵ In all cases, external standard calibration curves were prepared in the corresponding leaching solutions.

Results and discussion

pH change in 18 h leaching tests

Although the initial pH of each leaching solution was nominally 5.0, sample pH values generally increased by the end of the 18 h tests (Table 2) due to the alkaline nature of the brake wear debris. For example, it has been reported that phenolic resins and metal oxides are common materials in brake pads.¹⁷ Final pH values ranged from 5.8 to 6.2 for most of the leaching solutions, with the exception of the concentrated synthetic rainwater (pH = 5.3) and the higher EDTA concentration (pH = 5.0). The high ligand concentrations used in the latter two solutions apparently kept these samples reasonably well buffered. The pH values for the two 15 mg C L⁻¹ HS solutions increased even with this relatively high concentration, but the carboxyl contents of the solutions (i.e., 0.114 and 0.157 meq L⁻¹ for SHA and SRFA, respectively) were still much lower than those of the concentrated rainwater and the 0.01 M EDTA solutions (4.5 and 40 meq L⁻¹, respectively).

Table 2 Comparison of 18 h leaching results^a

Leaching solution^b	Cu recovery (%)^c	Fe recovery (%)^c	Final pH
a Based on total Cu and Fe contents of 10.8 and 28.6%, respectively, by mass in the brake wear debris. b All solutions had a nominal initial pH of 5.0. c Statistically similar values in each column, based on t-tests of the means for α = 0.05 using a two-tailed t distribution, are denoted by the same superscript number. d pH adjusted to 5 with 0.1 N HNO3. e Concentrations 50× higher than Table 1 values.
SPLP	13.15 ± 0.46^1	0.034 ± 0.008^1	6.2
DDW^d	18.24 ± 1.67^2	0.025 ± 0.005^1	6.0
Inorganic synthetic rainwater	30.74 ± 0.56^3	0.026 ± 0.004^1	5.8
Synthetic rainwater	40.07 ± 1.66^4,5	0.034 ± 0.002^1	5.8
Concentrated inorganic synthetic rainwater^e	34.20 ± 1.55^3,4	0.039 ± 0.012^1	6.1
Concentrated synthetic rainwater^e	71.67 ± 1.57^6	1.21 ± 0.06^2	5.3
SHA (3 mg C L^−1)	22.69 ± 0.28^7	0.152 ± 0.024^3	6.0
SHA (15 mg C L^−1)	39.91 ± 0.37^5	0.538 ± 0.077^4	6.0
SRFA (3 mg C L^−1)	23.89 ± 0.83^7	0.078 ± 0.006^5	6.0
SRFA (15 mg C L^−1)	39.07 ± 1.39^4,5	0.345 ± 0.050^4	5.8
EDTA (8 × 10^−5 M)	55.00 ± 1.57^8	1.02 ± 0.29^2	6.0
EDTA (1 × 10^−2 M)	89.62 ± 2.40^9	43.54 ± 0.91^6	5.0

---

Copper leaching in model rainwater solutions

Table 2 presents the 18 h copper leaching results for model rainwater solutions. The SPLP solution, designed by the US EPA to simulate the leaching expected due to percolating rainfall, showed the least amount of copper leaching among all the model rainwater solutions tested. The lower degree of copper leaching observed with the SPLP solution is likely due in part to its higher final pH value, although it should be noted that this value was statistically equivalent to the final pH measured for DDW (p value of 0.0532 based on a t-test of the means for α = 0.05 using a two-tailed t distribution). These results appear to be consistent with a previous report¹⁸ that lower leaching of arsenic, chromium and copper from a contaminated soil was observed with SPLP solutions than with DDW adjusted to the same pH values.

The synthetic rainwater solution leached approximately two and three times more copper from the brake wear debris than DDW and the SPLP solution, respectively. The higher leaching with the synthetic rainwater was statistically significant as demonstrated by the respective p values of 0.0115 and 0.00028 based on t-tests of the means for α = 0.05 using a two-tailed t distribution. The most plausible explanation for the enhanced leaching by synthetic rainwater over DDW and the SPLP solution is that it contained formate and acetate, two organic acids typically found in urban rainwater samples. Further confirmation of their ability to enhance copper leaching from the brake wear debris can be seen by comparing the average leaching results for the synthetic rainwater with and without organic acids present (40 versus 31%, p value of 0.0074, Table 2). The combined results above suggest that leaching tests^1,19 that have used DDW or aqueous solutions containing only inorganic anions (e.g., SPLP) likely have underestimated actual copper leaching from brake wear debris during rainfall events and/or in storm water runoff.

The inorganic anions present in the synthetic rainwater did not appear to have a major effect on the leaching of copper from the brake wear debris. For example, although a factor of 50 increase in the concentration of ions in the inorganic synthetic rainwater increased copper leaching from 30.74 to 34.20%, the increase was not statistically significant (p value = 0.0850; Table 2). However, when the formate and acetate concentrations were also increased by a factor of 50, copper leaching increased dramatically from 40.07 to 71.67% (p value = 0.00092). This result corroborates the predominant role that organic ligands play in copper leaching from brake wear debris versus the inorganic anions present in rainwater. However, it should be noted that the increase in copper leaching observed with the concentrated synthetic rainwater may also be attributed in part to the lower final pH value (i.e., 5.3 versus 5.8), because previous studies have shown dramatic differences in metal leaching with changes in pH.^18,20 The influence of pH on copper leaching from brake wear debris will be addressed later.

Despite its high abundance in the brake wear debris sample, iron was generally leached only in very small amounts (0.02 to 1.2%) by the SPLP and other model rainwater solutions (Table 2). Therefore, little to no competition between copper and iron for the organic ligands would be expected because of the relatively low dissolved iron concentrations present.

Copper leaching in HS and EDTA solutions

The two different HS showed similar copper leaching abilities at both the lower and higher concentrations tested (3 and 15 mg C L⁻¹, respectively; Table 2). This finding was unexpected, because the higher carboxyl group content of SRFA versus SHA should have provided more copper binding ligands and, thus, higher average copper leaching values. For both HS, the amount of copper leached did not exceed 40% at the higher concentration tested. However, the higher HS concentration did result in nearly a factor of 2 increase in copper leaching, indicating that large HS concentration variations in storm water runoff and surface waters will likely influence copper release from brake wear debris. Such temporal variations in dissolved organic carbon concentrations in runoff are known to be highly dependent on precipitation volumes and hydrological pathways.²¹

EDTA is a strong metal chelating agent that can form very stable and soluble complexes with metals.^13,16 It is a common synthetic organic ligand often found in natural waters because of its widespread use and resistance to degradation.^13,16,22 For example, Bedsworth and Sedlak²² reported EDTA concentrations in effluent samples from water pollution control plants in the San Francisco Bay area ranging from 0.1 to 2 µM. Xue et al.¹⁶ reported EDTA concentrations of 20–60 nM in the Glatt River (Switzerland). Although EDTA concentrations in storm water runoff are likely to be low to non-existent,²² its presence in many surface waters worldwide suggests an assessment be made of its ability to mobilize copper from brake wear debris. From Table 2, it can be observed that more than 50% of the total copper present in the brake wear debris was mobilized by the relatively high EDTA concentration of 80 µM, suggesting that copper leaching from brake wear debris may be facilitated by EDTA in natural waters. At the very high EDTA concentration of 10 mM, most (∼90%) of the copper was leached from the brake wear debris.

Similarly to the model rainwater results, only a small fraction of iron was leached from the brake wear debris by HS and the lower EDTA concentration tested (0.1–0.5 and 1%, respectively). Only at the very high EDTA concentration was a large amount of iron (44%) leached into solution. The results observed here for copper versus iron leaching are reasonably consistent with the findings of our previous study,⁵ in which copper leaching ranged from a factor of 3 to 10 times higher than iron in standard extraction solutions.

Copper leaching in static pH tests

Static pH leaching tests covering a wide range of pH values were performed with the SPLP, synthetic rainwater with and without organic acids, and the two HS solutions to investigate its effects on copper mobilization from the brake wear debris (Fig. 1). In all five solutions, copper leaching was strongly dependent on pH, with higher copper leaching generally occurring at lower pH values. The observed decrease in copper mobilization with increasing pH can be explained by its effect on total copper solubility. For example, dissolved copper concentrations are typically limited by the solubility of a solid phase such as copper oxide (tenorite) and/or copper chloride hydroxide (atacamite) at pH values above 6.¹³

Fig. 1 Copper leaching (% of total presence) from the brake wear debris as a function of pH. For all samples, pH was maintained constant for at least the last 18 of a total of 48 h. (a) SPLP and synthetic rainwater solutions. (b) HS solutions (3 mg C L⁻¹).

More copper was leached from the brake wear debris by the formate- and acetate-containing synthetic rainwater versus the SPLP solution over most of the pH range examined (Fig. 1a), consistent with results from the 18 h leaching tests discussed above. The importance of the two organic acids is highlighted by the higher copper leaching observed in their presence versus their absence for most pH values. It is only for pH values less than 4 that the two curves collapse together, indicating that the organic acids do not enhance copper leaching at these low pH values. This observation is consistent with the ionization state of the two organic acids (i.e., pK_a values of 4.75 and 3.75 for acetate and formate, respectively), because H⁺ competition for COO⁻ binding sites becomes stronger with decreasing pH. In contrast with the 18 h leaching tests, however, the SPLP and inorganic synthetic rainwater solutions gave relatively similar copper leaching results over the majority of the pH range tested.

More copper leaching from the brake wear debris was observed with the SRFA versus SHA solutions over the pH range examined, although the differences were not always significant. This observation differs slightly from the 18 h leaching tests, which showed similar copper leaching values for the two HS. Unlike the model rainwater solutions, appreciable amounts of copper were still being leached into solution by the two HS at pH values above 8. In general, the degree of HS ionization increases with increasing pH, thus resulting in greater negative charges on the molecules and increased electrostatic potentials that attract cations present in solution.²³ Carboxyl and phenolic groups are considered to be the predominant HS functional groups responsible for complexing metals.

Static pH test results for iron revealed that it was leached only under relatively acidic conditions (pH < 5.0), although the same general trend (i.e., increased metal leaching with decreasing pH) that was observed with copper also applied to iron (Fig. 2). Leached iron amounts only reached ∼15% of the total iron present in the brake wear debris at the lowest pH value tested (pH 3). This observation suggests that although iron is the most abundant element present in the brake wear debris, it likely does not interfere with copper leaching under solution conditions similar to those examined here.

Fig. 2 Iron leaching (% of total presence) from the brake wear debris as a function of pH. For all samples, pH was maintained constant for at least the last 18 of the total 48 h. (a) SPLP and synthetic rainwater solutions. (b) HS solutions (3 mg C L⁻¹).

Copper leaching in kinetic studies

Copper leaching was monitored as a function of time for the SPLP, synthetic rainwater, and SRFA solutions (Fig. 3). Measured pH values for all samples sharply increased from the initial pH by the first sampling time (20 min) and then slowly increased thereafter (Fig. 3a). Consistent with the results above, copper leaching at all observation times was highest with the synthetic rainwater solution (Fig. 3b). For all three solutions, copper was released rapidly within the first few hours followed by slower leaching at longer contact times. Although 80 to 90% of the copper in solution at 48 h was leached within the first 24 h, it was apparent that very slow copper leaching continued beyond the longest sampling time tested. The environmental implications of our observed leaching rate results are not fully known, but one can reasonably speculate that such temporal effects likely contribute to “first flush” phenomena in watersheds^4,24 and also suggest that brake wear debris may be an ongoing low-level non-point source of copper in environments where it is exposed to long-term leaching processes.

Fig. 3 Changes in pH (a) and copper leaching (b, % of total presence) from the brake wear debris as a function of time in kinetic leaching tests.

Conclusions

An appreciable amount (40%) of the copper present in our brake wear debris sample was released into a synthetic rainwater solution after 18 h. This value was higher than the leaching observed with either DDW or an SPLP solution. Our results demonstrate that organic ligands and chelating agents present in rainwater, urban storm water runoff, and/or surface waters enhance the leaching of copper from brake wear debris. Irrespective of the particular leaching solution used, copper mobilization from the brake wear debris showed a very high dependence on pH, with the highest leaching occurring under more acidic conditions. Despite its high abundance in the brake wear debris, iron was not leached to a great extent in most of the solutions tested. The overall results of this study suggest that monitoring important solution chemistry parameters such as pH and the presence of potential metal binding organic ligands in natural waters and storm water runoff will allow for a better understanding of the environmental fate of copper resulting from brake wear debris loadings to watersheds.

Acknowledgements

We greatly appreciate the assistance and suggestions provided by the Brake Pad Partnership Steering Committee (Kelly Moran, Sarah Connick, Jim Trainor, Pat Thesier, Roger Dabish, Tim Merkel, Jim Pendergast, and Michael Endicott). Additional thanks go to Phil Bobel, Geoff Brosseau and Tom Barron for initiating this study and providing overall management. Finally, we wish to acknowledge the Brake Manufacturers Council Product Environmental Committee for their generous donation of the brake wear debris sample. Funding for this work was provided by the City of Palo Alto, the US Environmental Protection Agency, the Bay Area Stormwater Management Agencies Association, and the San Francisco Estuary Project. Additional support was received from the National Science Foundation (Grant 9996441), US Department of Agriculture (SC-1700133 and NRCS-69-4639-1-0010) and a Switzer Foundation Environmental Leadership Grant (C-2001-0814) provided through the San Francisco Foundation. Any opinions, findings, conclusions or recommendations expressed in this publication are solely those of the authors and do not necessarily reflect the views of any of the individuals, associations or other entities mentioned above.

References

1. P. Davis, M. Shokouhian and S. Ni, Chemosphere, 2001, 44, 997.
2. N. Hewitt and M. B. Rashed, Sci. Total Environ., 1990, 93, 375.
3. J. J. Sansalone and S. G. Buchberger, J. Environ. Eng., ASCE, 1997, 123, 134.
4. D. Drapper, R. Tomlinson and P. Williams, J. Environ. Eng., ASCE, 2000, 126, 313.
5. J. Hur, S. Yim and M. A. Schlautman, J. Environ. Monit., 2003, 5, 837.
6. US Environmental Protection Agency, Synthetic Precipitation Leaching Procedure (Method 1312), in Test Methods for Evaluating Solid Waste, Physical/Chemical Methods (SW-846), Office of Solid Waste and Emergency Response, US Government Printing Office, Washington, DC, 3rd edn., 1986.
7. H. Sakugawa and I. R. Kaplan, Atmos. Environ., 1993, 27B, 203.
8. G. B. Avery, J. D. Willey and R. J. Kieber, Atmos. Environ., 2001, 35, 3927.
9. W. H. McDowell and G. E. Likens, Ecol. Monogr., 1988, 58, 177.
10. C. Volk, L. Wood, B. Johnson, J. Robinson, H. W. Zhu and L. Kaplan, J. Environ. Monit., 2002, 4, 43.
11. P. Westerhoff and D. Anning, J. Hydrology, 2000, 236, 202.
12. J. Stevenson, Humus Chemistry: Genesis, Composition, Reaction, John Wiley & Sons, Inc., NY, 1994.
13. W. Stumm and J. J. Morgan, Aquatic Chemistry: Chemical Equilibria and Rates in Natural Waters, John Wiley & Sons, Inc., NY, 3rd edn., 1996.
14. S. Hlawiczka, B. Dyduch and J. Fudala, Water, Air, Soil Pollut., 2003, 142, 151.
15. J. Hur and M. A. Schlautman, Environ. Sci. Technol., 2003, 37, 880.
16. H. B. Xue, L. Sigg and F. G. Kari, Environ. Sci. Technol., 1995, 29, 59.
17. P. J. Blau, Compositions, Functions, and Testing of Fraction Brake Materials and Their Additives, ORNL/TM-2001/64, Oak Ridge National Laboratory, Oak Ridge, Tennessee, 2001.
18. Y.-C. Jang, T. G. Townsend, M. Ward and G. Bitton, Bull. Environ. Contam. Toxicol., 2002, 69, 808.
19. J.-H. Hsu and S.-L. Lo, Water Sci. Technol., 2000, 42, 247.
20. J. Wu, L. J. West and D. I. Stewart, J. Hazard. Mater., 2002, B94, 223.
21. H. Frank, S. Patrick, W. Peter and F. Hannes, Biogeochemistry, 2000, 50, 137.
22. W. W. Bedsworth and D. L. Sedlak, Environ. Sci. Technol., 1999, 33, 926.
23. M. A. Ramos, S. Fiol, R. Lopez, J. M. Antelo and F. Arce, Environ. Sci. Technol., 2002, 36, 3109.
24. C. N. Hewitt and M. B. Rashed, Water Res., 1992, 26, 311.

---