Buffering of Alkaline Steel Slag Leachate across a Natural Wetland

Buffering of high-pH (>12) steel slag leachate is documented across a small, natural calcareous wetland. The alkaline leachate is supersaturated with ...
0 downloads 0 Views 222KB Size
Download PDF
Environ. Sci. Technol. 2006, 40, 1237-1243

Buffering of Alkaline Steel Slag Leachate across a Natural Wetland WILLIAM M. MAYES,* PAUL L. YOUNGER, AND J O N A T H A N A U MO ˆNIER Hydrogeochemical Engineering Research and Outreach, Institute for Research on Environment and Sustainability, University of Newcastle upon Tyne, Newcastle upon Tyne NE1 7RU, U.K.

Buffering of high-pH (>12) steel slag leachate is documented across a small, natural calcareous wetland. The alkaline leachate is supersaturated with respect to calcite upstream of the wetland (SIcalcite values +2.3) and becomes less saturated with progress across the wetland, to SIcalcite values of +0.27 at the wetland outlet. Reduction in pH across the wetland (to around pH 8 at the wetland outlet) was observed to be more pronounced over summer months, possibly due to increased microbial activity, possibly further assisted by greater flow baffling by emergent vegetation. Calculated calcite precipitation rates downstream of the leachate source, estimated from hydrochemical data, flow, and surface area, were on the order of 0.4-15 g m-2 day-1, while direct measurements (using immersed limestone blocks) showed calcite precipitation values in the range 3-10 g m-2 day-1. Precipitation rate was highest in the pH range where the carbonate ion is a dominant constituent of sample alkalinity (pH 9.5-11) and at the locations where wetland biota became established downstream of the leachate emergence. These data provide valuable insights into the potential for using constructed wetlands for the passive treatment of high pH steel slag leachates.

Introduction High-pH (pH 9-13) calcareous leachates are a potential environmental problem from the weathering of various industrial residues, including lime works spoil (1), steelworks slags (2), coal combustion residues (e.g. fly and bottom ash, flue gas desulfurization byproducts (3)), Solvay process waste (4), and cementitous construction and demolition waste (5). These wastes all tend to contain lime (CaO), which hydrolyzes in natural waters to produce calcium hydroxide (Ca(OH)2). The dissociation of Ca(OH)2 in solution liberates the hydroxyl ion (OH-) and elevates solution pH. In terms of ecological impacts, high-pH waters are characterized by the high rates of calcite precipitation as calcareous crystalline crusts (analogous in some ways to natural travertine deposits) which smother benthic and littoral aquatic habitats (6) and reduce light penetration to benthic primary producers. High pH itself can be directly harmful to fish populations (7), while associated water quality impacts of increased chemical oxygen demand, high sulfate loadings (8), salinity (9), and elevations of amphoteric or anionic heavy metals (e.g. arsenic and selenium (8)) have been reported for some alkaline leachates. * Corresponding author phone: +44 01912228599; fax: +44 01912226669; e-mail: [email protected]. 10.1021/es051304u CCC: $33.50 Published on Web 01/18/2006

 2006 American Chemical Society

Remediation options at high-pH sites generally comprise high-cost acid-dosing of leachates (10), reciprocation of leachates over stockpiled or lagooned high-pH residues (11), and natural attenuation, dilution, and dispersal of the leachate (12). More novel approaches to treatment of highpH waters have been suggested, such as aeration with CO2 gas to enhance calcite precipitation (13), but again, these are likely to incur high pumping costs at field scale. The buffering capacities of organic-rich wetland substrates alkaline waters have been widely documented (14, 15), although the specific mechanisms behind the process are understood only in broad terms. They are considered to be due to relatively high levels of CO2 in wetland waters and substrates, formed as a product of both aerobic and anaerobic microbial respiration. High CO2 levels enhance calcite precipitation, a process which consumes the constituents of alkalinity. Cation exchange (16) and humic acid production may also assist in ameliorating high pH in organic-rich substrates. The application of constructed wetlands as elements of passive remediation systems (which might also include aeration cascades and settling lagoon(s) in series with a wetland) at some high pH sites may provide a low-cost, environmentally favorable alternative to the high-cost remedial options that currently dominate. Similar systems have been developed and effectively applied over the past two decades for treating metalliferous and/or acidic mine waters (17, 18). These schemes are characterized by an initial capital outlay but low running costs for infrequent (albeit regular) maintenance. However, there are no design guidelines for the application of such systems to alkaline leachates (e.g. design rates of calcite precipitation) and, as such, little basis exists for sizing calculations for wetlands to treat high-pH waters. The data presented here provide an example of buffering in a small natural wetland which receives high-pH steel slag leachate at the site of a former steelworks. Flow through the system is low (12 at source to around pH 10 at sample location 3. This fall in alkalinity is consistent with the loss of alkalinity constituents (principally carbonate) from solution as precipitated calcite. The other significant anions in the system are chloride and sulfate. Chloride loadings are consistent downstream of the leachate over time and spatially across the sample locations, with values of 20-30 mg s-1 typical. Samples from the leachate source area do however exhibit a slight enrichment in sulfate over other locations with loadings ranging from 50 to 64 mg s-1. Sulfate loadings downstream of the leachate source tend to be slightly lower (ranging from 10 to 59 mg s-1), suggesting either the reduction of sulfate to sulfide in the wetland or the loss of sulfate in precipitated salts (although negative SI values in the range of -1.8 to -2.4 for both anhydrite and gypsum suggest this is unlikely). The characteristic odor of hydrogen sulfide was noticed during site visits, suggesting the former may be of greater importance, despite field Eh readings suggesting the wetland to be predominantly aerobic. Calcite Precipitation. Calculated SIcalcite values offer an indication of the relative saturation of waters with respect to calcite through the system. Although values may vary considerably depending on the geochemical environment, a saturation index (SIcalcite) in excess of +1.5 for calcite is demanded for significant production of homogeneous calcite crystals in solution, whereas heterogeneous precipitation of calcite onto existing solid surfaces can occur at SIcalcite > +0.3 (24). Calcite precipitation is sluggish, at best, in natural systems at values of 0.0 < SIcalcite < +0.3 and of course nonexistent for SIcalcite < 0.0. SI values fall markedly and consistently (see standard deviation values in y error bars of Figure 2) through the system from the source zone (mean value +2.3), where there is a high degree of supersaturation with respect to calcite, to the wetland outlet (mean value +0.28). The values at the outlet are just below those considered sufficient for appreciable calcite precipitation. Calculated Values. On the basis of the assumption that any reduction in Ca2+ loading between consecutive sample points is a result of loss from solution as calcite, estimates of calcite precipitation can be made using the generic treatment wetland sizing equation (eq 1), where the contaminant is Ca2+ and calcite removal is calculated from molar weight ratios. Calculations for precipitation rate between source and the mid-marsh sample point (sites 2 and 3; surface area of 1958 m2) and mid-marsh to wetland outlet (sites 3 and 4; surface area of 905 m2) can therefore be made for each sample visit giving area-averaged calcite precipitation rates (Table 1). Calculated precipitation rates are shown to be quite variable ranging from 1.3 to 13.6 g m-2 day-1 at the upstream sample point. Values calculated between sites 3 and 4 (lower wetland) are consistently lower than the upstream sample point, with values ranging between 0.4 and 1.8 g m-2 day-1. Precipitation rates are generally calculated to be lower in low flow conditions (e.g. 22/5, 30/7, 8/9) with some peak values recorded in higher flow conditions (e.g. 11/5). Leachate quality was observed to change slightly with flow (e.g. peaks in conductivity and measured Ca2+ after high flow events), which may account for some of the higher precipitation rates in high flow (i.e. higher leachate Ca2+ and flow, coupled with consistent outlet chemistry, means greater Ca2+ loss across 1240

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 4, 2006

TABLE 1. Calculated Calcite Precipitation Rate Derived from Hydrochemical, Flow, and Area Data in the Hownsgill Valley, Consetta

date

precipitatn rate between sites 2 and 3 (g m-2 day-1)

precipitatn rate between sites 3 and 4 (g m-2 day-1)

flow (L s-1)

22/01/04 04/02/04 26/02/04 01/04/04 11/05/05 22/05/04 15/06/04 09/07/04 30/07/04 08/09/04 12/10/04 18/11/04

1.65 7.36 8.16 6.15 13.61 1.34 7.34 11.60 4.17 4.85 11.16 11.25

0.361 1.16 1.30 1.79 0.646 0.404 1.28 0.509 0.842

1.0 1.3 1.2 1.7 2.9 0.6 0.9 1.2 0.4 0.5 1.0 1.4

a “ -” indicates no positive value for calcite precipitation; i.e., Ca2+ concentrations increase between sites 3 and 4.

TABLE 2. Calcite Precipitation Rate across the Hownsgill Valley Site Measured Using Limestone Blocks for 21, 11, and 2 day Periods of Immersion precipitatn rate (g m-2 day-1) site

21 days

11 days

2 days

1 2 A 3 4 B

0.92 4.35 9.88 5.72 4.07 1.13

0.77 4.35 6.83 5.79 4.33 1.44

0.83 3.66 8.05 4.39 3.90 1.68

the system). However, this pattern is not entirely consistent as precipitation rates in excess of 10 g m-2 day-1 were recorded at flows as low as 1.0 L s-1. Ideally residence time data would be required to assess variations in precipitation rates under changing flow conditions. Unfortunately, such residence time values could not be accurately quantified in this study as obtaining meaningful water depth measurements (for volume calculations) on each visit was prohibited by the physical nature of the site. As such, judgment should be reserved on changing process rates until precipitation rate data from a more controlled system can be obtained.

Ra )

Qd(Ci - Ct) A

(1)

Here, Ra ) area adjusted contaminant removal rate (g m-2 day-1) , Qd ) mean daily flow rate (m3 day-1), Ci ) mean daily influent contaminant concentration (mg L-1), Ct ) concentration of contaminant in final discharge (mg L-1), and A ) wetland area (m2). Measured Values. Measured values of calcite precipitation rate obtained using the Dreybrodt limestone tablet method are presented in Table 2. The precipitation rates are shown to be relatively consistent over time (i.e. values for different periods of immersion are similar at each site) but vary quite markedly spatially. Although the data presented here represents a small data set from a single site, this consistency is encouraging with respect to using the method for obtaining repeatable estimates of calcite precipitation at a particular location. Peak calcite loadings were consistently recorded at site A (on the margin of pond and wetland) with highest values of 9.88 g m-2 day-1 for the 21-day period of immersion. These peak values were roughly twice those measured at the leachate source (site 2), where precipitation rate was in the

range of 3.66 to 4.35 g m-2 day-1. Precipitation rate is seen to decline downstream through the wetland from site A to the outlet of the system at the confluence with the Dene Burn where precipitation rate was around 1.4 g m-2 day-1. The precipitation rates at the outlet exceed the upstream control rates of 0.8-0.9 g m-2 day-1 at sample location 1.

Discussion Calculated versus Measured Precipitation Rate. Although comparison between area-averaged and point precipitation rates must be approached with considerable caution, the peak values recorded at the pond-wetland margin with the limestone block method are generally about 30% lower than calculated peak values. This may be expected with the potential error margins and assumptions in the values calculated from hydrochemistry and area estimates, which are likely to overestimate precipitation rate. The assumption that all Ca2+ lost between two sample locations is as precipitated calcite fails to take into account effects of ion pairing (weak cation-anion associations in solutions rich in free ions that can lead to underestimation of measured Ca2+ by up to 29% (25)) and the possible loss of Ca2+ adsorbed to humic substances in the wetland. Further to this, the estimates of flow areas in the terrain mapping also involve an element of subjectivity in the delineation of flow margins. In this study the margins of effective flow were deduced from field surveys of vegetation patterns, ground saturation, topography, and evidence of calcite precipitation. However, the braided nature of the flow among tussocks of vegetation and the overall low flow rate through the system makes surveyed values of surface area no more than rough estimates. Precipitation rate values calculated for the lower wetland (site 3-4) were up to 1 order of magnitude less than those measured directly with limestone blocks. This is a feature of incorporating the estimates of flow area into the calculations. Flow area is particularly difficult to estimate between sample points 3 and 4 due to the dispersed nature of flow among emergent vegetation, and estimates from flow boundaries will tend to overestimate the flow area. As such, the precipitation rates derived from flow and area estimates must be treated with caution. The repeatability of the direct measurements using limestone blocks at the low precipitation rate sites suggests that they provide a better indication of the lower calcite precipitation rates toward the wetland outlet. One notable aspect of the precipitation rate data obtained from direct measurement is the low precipitation rate around the leachate source where SIcalcite values suggest that calcite supersaturation is at its greatest. These patterns are likely to be controlled by the shift in the dominant constituents of alkalinity through the system and the influence of biological activity. Relatively low precipitation rates at the leachate source are a reflection of the dominance of the hydroxyl ion at pH >11 (and low carbonate concentration), and precipitation rate is likely to be limited here by the slow transfer of atmospheric CO2 to solution. As pH falls slightly through the pond, with CO2 transfer into solution, the carbonate ion becomes the dominant component of sample alkalinity (for pH in the range of 9.5-11). It is within this range that the hydrochemical data reveal a general fall in sample alkalinity and electrical conductivity (see Figure 2), indicative of the rapid loss of ionic species (i.e. Ca2+ and CO32-) from solution. This area of highest precipitation rate coincides with the establishment of vegetation and increased biological activity (as evidenced by the colonisation of algae in the waters) on the lagoon/wetland margin (sites A to 3). Given that the precipitation of calcite in the CaCO3-H2O-CO2 system has been shown to be limited by the conversion of HCO3- to CO2 (26), the influence of microbial CO2 as wetland biota establishes could be significant in yielding rates of precipitation far greater than those observed in leachate source areas

(where biological activity appears negligible). The onset of biological activity may also assist in elevating calcite precipitation through the provision of more potential precipitation nuclei. The presence of algae has been shown to assist calcite precipitation in other alkaline leachate impacted surface waters (27). The low precipitation rates at source and the high initial pH suggest the necessity for preliminary settlement and/or aeration of alkaline leachate prior to wetland treatment. This would facilitate an initial drop in pH to levels at which aquatic macrophytes could successfully colonize (in the region of pH 10.5 (28)). This system design is similar to passive remediation systems targeting acidic and/or metalliferous mine waters for which engineering specifications are now well-developed (17, 18). Future Applications. The smothering of benthic habitats by rapid calcite precipitation at sites in receipt of alkaline leachates has been widely documented (29, 30), with accounts typically documenting deleterious effects on aquatic flora and fauna. Little attention however has been previously paid to the hydro-geochemistry (in particular rates of calcite smothering) in surface water courses affected by high pH waters. The data provided in this study, although limited to one site, have allowed quantification of rates of calcite precipitation in a calcareous high pH environment. Quantification of calcite precipitation rate from alkaline leachates is a key design parameter in formulating sizing estimates of potential passive remediation facilities. First-order estimates of treatment system size can be calculated on the basis of the measured precipitation rates and hypothetical scenarios of leachate composition and flow (see Table S3, Supporting Information). Through rearranging eq 1, wetland area can be estimated using flow rates and influent Ca2+ concentrations typical of alkaline leachate sources quoted in the literature (1, 4, 28, 31). A target final Ca2+ concentration of 30 mg L-1 is set for all scenarios. Data from the Consett site suggests that within the range of 30 mg L-1 Ca2+, pH and SIcalcite fall to levels at which precipitation rate is low or negligible. Maximum wetland areas in the order of 30 00050 000 m2 are calculated for high flow and high Ca2+ removal, which are similar to sizes quoted for successful mine water remediation wetlands (18). The lowering of pH and reduction in calcite precipitation represents just one aspect of what can be a complex environmental problem at some alkaline leachate sites. Although trace metal loadings were low at this site, elevations of some metal oxyanions and amphoteric metals may be exhibited by certain alkaline leachates, in particular those derived from coal combustion residues (32, 33) and steel slags (34). Further field data concerning the nature of trace metal loadings present in the full range of alkaline leachate types would however be desirable to assess the extent of the problem. The abundant data from laboratory leaching tests on various alkaline residues (5, 32, 33) rarely give a good indication of the products of natural weathering (3, 32, 33, 35). Much of the focus on trace metal leaching from high pH residues has been centered on coal combustion ashes, with the suggestion that it is trace elements found in surface associations (e.g. Mo, As, Cu, Zn, and Pb) that are most potentially leachable (33). However, field data do demonstrate that the low permeability and consequent low infiltration rates through fine-grained fly ash mounds often results in minimal actual impact of trace metals in leachates on recipient water courses (33). Where there are elevated trace metals in alkaline leachates at high pH values such as those encountered in this study, the sorption of trace metals onto calcite through adsorption and subsequent recrystallization (e.g. Zn, Co, Ni) or surface precipitation (e.g. Cd, Mn (35)) and the formation of secondary mineral phases (e.g. hydroxides and carbonates) may also provide mechanisms for VOL. 40, NO. 4, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

1241

limiting the mobility of some metals. Further study would be needed to identify the chemical composition and behavior of such secondary phases in various alkaline leachates, to understand the mobility and long-term fate of trace elements in any proposed treatment wetlands. Another issue relating to the maintenance of constructed wetland systems is the accumulation of calcite in the system. The precipitation rates measured here are similar to loadings of ferric hydroxide precipitates documented for mine water treatment wetlands (17, 18), suggesting that similar frequencies of maintenance (e.g. the dredging of preliminary settlement ponds every 2-10 years) may be applicable to steel slag leachate treatment systems. Such maintenance frequencies are well within the range for economically viable passive remediation systems. Should there be potential for the significant accumulation of trace metals in the calcite sludge then judgment on the cost of sludge disposal (most likely via landfill after dehydration of the sludge) would need to be balanced against the environmental benefits of the treatment system. These decisions are common to any water remediation system (passive or active) when treatment results in removed (immobilized) metals being concentrated in residues requiring disposal to land. The data provided here provide a promising foundation upon which to base the future development of passive treatment systems to ameliorate some of the problems posed by alkaline steel slag leachates (and, potentially, other similar leachates). Future research should however endeavor to encompass (1) controlled pilot-scale studies of buffering for a range of leachate types, (2) further data on leachate constituent, flow, and calcite precipitation rate for a range of leachate sources (e.g. coal combustion residues, lime spoil, and C&D waste), with particular reference to trace metal loadings and their fate in wetland systems, and (3) a detailed assessment of microbial populations and performance in high-pH media.

Acknowledgments This research was funded by ENTRUST through the Mineral Industry Research Organization (MIRO) under Project RC 174. Patrick Orme is thanked for undertaking the laboratory chemical analyses. The assistance of Andrew Daugherty in undertaking the terrain mapping and Fred Beadle in preparing the limestone blocks is also gratefully acknowledged. The authors also acknowledge the contribution of the three reviewers whose comments have improved earlier drafts of the manuscript.

Supporting Information Available More detailed data concerning hydrochemistry at the site (Tables S1, S2; Figure S1) and sizings of potential wetland treatment systems (Table S3). This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) Andrews, J. E.; Gare, S. G.; Dennis, P. F. Unusual isotopic phenomena in Welsh quarry water and carbonate crusts. Terra Nova 1997, 9, 62-66. (2) Koryak, M.; Stafford, L. J.; Reilly, R. J.; Magnuson, M. P. Impacts of steel mill slag leachate on the water quality of a small Pennsylvania stream. J. Freshwater Ecol. 2002, 17, 461-465. (3) Carlson, C. L.; Adriano, D. C. Environmental impacts of coal combustion residues. J. Environ. Qual. 1993, 22, 227-247. (4) Effler, S. W.; Whitehead, K. A. Tributaries and Discharges. In Limnological and Engineering Analysis of a Polluted Urban Lake: Prelude to Environmental Management of Onondaga Lake, New York; Effler, S. W., Ed.; Springer-Verlag: New York, 1996; pp 97-199. (5) Townsend, T. G.; Jang, Y.; Thurn, L. G. Simulation of construction and demolition waste leachate. J. Environ. Eng. 1999, 125, 10711081. 1242

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 4, 2006

(6) Effler, S. W.; Brooks, C. M.; Addess, J. M.; Doerr, S. M.; Storey, M. L.; Wagner, B. A. Pollutant loadings from Solvay waste beds Lower Ninemile Creek, New York. Water, Air, Soil Pollut. 1991, 55, 427-444. (7) Haslam, S. M. River Pollution-an Ecological Perspective; John Wiley and Sons: London, 1990. (8) Spencer, L. L. S.; Drake, L. D. Hydrogeology of an alkaline fly ash landfill in Eastern Iowa. Groundwater 1987, 25, 519-526. (9) Effler, S. W.; Doerr, S. M.; Owens, E. M. Effect of reduced salinity input on river stratification and dissolved oxygen. Water, Air, Soil Pollut. 1997, 95, 45-58. (10) Forbes, A. M.; Magnuson, J. J.; Harrell, D. M. Effects of habitat modifications from coal ash effluent on stream macroinvertebrates: a synthesis. In The Warmwater Streams Symposium: A National Symposium on Fisheries Aspects of Warmwater Streams, Knoxville, TN, Mar 1980; Krumholz, L. A., Ed.; American Fisheries Society, Allen Press: Lawrence, KS, 1981. (11) Cairns, J.; Dickson, K. L.; Crossman, J. S. The biological recovery of the Clinch River following a fly ash pond spill. In 25th Industrial Waste Conference Proceedings; Purdue University: West Lafayette, IN, 1972; pp 182-192. (12) Matthews, D. A.; Effler, S. W. Decreases in pollutant loading from residual soda ash production waste. Water, Air, Soil Pollut. 2003, 146, 55-73. (13) Schramke, J. A. Neutralization of alkaline coal fly ash leachates by CO2(g). App. Geochem. 1992, 7, 481-492. (14) Ponnamperuma, F. N. The chemistry of submerged soils. Adv. Agron. 1972, 24, 29-96. (15) Mitsch, W. J.; Gosselink, J. G. Wetlands, 3rd ed.; van Nostrand Reinhold: New York, 2000. (16) Sakata, M. Movement and neutralization of alkaline leachate at coal ash disposal sites. Environ. Sci. Technol. 1987, 21, 771-776. (17) Younger, P. L.; Banwart, S. A.; Hedin, R. S. Mine Water: Hydrology, Pollution, Remediation; Kluwer Academic: Dordrecht, The Netherlands, 2002. (18) PIRAMID Consortium. Engineering guidelines for the passive remediation of acidic and/ or metalliferous mine drainage and similar wastewaters. European Commission 5th Framework RTD Project No. EVK1-CT-1999-000021 “Passive in-situ remediation of acidic mine/industrial drainage” (PIRAMID), University of Newcastle upon Tyne, Newcastle upon Tyne, U.K., 2003. (19) Sheffield, P. M. Investigation and remediation of alkaline leachate from the former Consett Steelworks site. Unpublished M.Sc. thesis, Department of Civil Engineering, University of Newcastle upon Tyne, 1996. (20) Parkhurst, D. L.; Appelo, C. A. J. User’s guide to PHREEQC-A computer program for speciation, batch-reaction, one-dimensional transport, and inverse geochemical calculations; U.S. Geological Survey Water-Resources Investigations Report 994259; USGS: Washington, DC, 1999. (21) Dreybrodt, W.; Buhmann, D.; Michaelis, J.; Usdowski, E. Geochemically controlled calcite precipitation by CO2 outgassing: Field measurements of precipitation rates in comparison to theoretical predictions. Chem. Geol. 1992, 97, 285-294. (22) Younger, P. L. 1995. Hydrogeology. In Robson’s Geology of North East England; Johnson, G. A. L., Ed.; Natural History Society of Northumbria: Newcastle, U.K., 1995; pp 353-362. (23) Warren, C. J.; Dudas, M. J. Weathering processes in relation to leachate properties of alkaline fly ash. J. Environ. Qual. 1984, 13, 530-538. (24) Ford, D. C.; Williams, P. W. Karst Geomorphology and Hydrology; Chapman and Hall: Cambridge, U.K., 1991. (25) Wigley, T. M. L. Ion pairing and water quality measurements. Can. J. Earth Sci. 1971, 8, 468-476. (26) Dreybrodt, W.; Eisenlohr, L.; Madry, B.; Ringer, S. Precipitation kinetics of calcite in the system CaCO3-H2O-CO2: The conversion to CO2 by the slow process H+ + HCO3- f CO2 + H2O as a rate-limiting step. Geochim. Cosmochim. Act. 1997, 61 (18), 3897-3904. (27) Womble, R. N.; Driscoll, C. T.; Effler, S. W. Calcium carbonate deposition in Ca2+ polluted Onondaga Lake, New York, USA. Wat. Res. 1996, 30, 2139-2147. (28) Mayes, W. M. Limestone extraction and wetland environments: hydrological, hydrochemical and ecological interactions. Unpublished Ph.D. thesis, University of Newcastle upon Tyne, 2003. (29) Effler, S. W. The impact of a chloro-alkali plant on Onondaga Lake and adjoining systems. Water, Air, Soil Pollut. 1987, 33, 85-115.

(30) Koryak, M.; Haberman, M. H.; Stafford, L. J.; Reilly, R. J.; Hoskin, R. H. The impact of airport deicing runoff on water quality and aquatic life in a Pennsylvanian Stream. J. Freshwater Ecol. 1998, 13, 287-298. (31) Mayes, W. M.; Large, A. R. G.; Younger, P. L. The impact of groundwater from a dewatered Magnesian limestone quarry on an adjacent wetland. Environ. Pollut. 2005, 138, 444-455. (32) Eary, L. E.; Dhanpat Rai; Mattigod, S. V.; Ainsworth, C. C. Geochemical factors controlling the mobilization of inorganic constituents from fossil fuel combustion residues: II. Review of minor elements. J. Environ. Qual. 1990, 19, 202-214. (33) Spears, D. A.; Lee, S. Geochemistry of leachates from coal ash. In Energy, Waste and the Environment: a Geochemical Perspective; Special Publication 236; Giere´, R., Stille, P., Eds.; Geological Society: London, 2004; pp 619-639.

(34) Simmons, J.; Ziemkiewicz, P.; Courtney Black, D. Use of steel slag leach beds for the treatment of acid mine drainage. Mine Water Environ. 2002, 21, 91-99. (35) Donahoe, R. J. Secondary mineral formation in coal combustion byproduct disposal facilities: implications for trace metal sequestration. In Energy, Waste and the Environment: a Geochemical Perspective; Special Publication 236; Giere´, R., Stille, P., Eds.; Geological Society: London, 2004; pp 641-658.

Received for review July 6, 2005. Revised manuscript received November 29, 2005. Accepted December 8, 2005. ES051304U

VOL. 40, NO. 4, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

1243