If you need any help send an email to: help@igsoa.net

Ground Water Issue

Metal Attenuation Processes at Mining Sites

Richard T. Wilkin

Background

The EPA Regional Ground Water Forum is a group of EPA scientists representing Regional Superfund and Resource Conservation and Recovery Act Offices (RCRA). The Forum is focused on exchanging information related to ground-water characterization, monitoring, and remediation. The application of monitored natural attenuation (MNA) for inorganic contaminants in ground water is a topic of concern to the Forum. The purpose of this Issue Paper is to provide scientists and engineers responsible for assessing remediation technologies with background information on MNA processes at mining-impacted sites. Some of the key issues concerning the application of natural attenuation for inorganic contaminants are discussed, such as the geochemical mechanisms responsible for attenuation, attenuation capacity, monitoring parameters, and evaluating whether attenuated metal and metalloid contaminants will remain immobile.

Introduction

Acid mine drainage (AMD) is a major source of water contamination in metal-mining and coal-mining districts worldwide. The causes of AMD are well known. They relate to the natural weathering of mine wastes and rocks enriched in metal sulfide minerals. Environmental impacts include the destruction of aquatic life and habitats and contamination of drinking water resources. The most common reactions that lead to the production of AMD involve the chemical and biological oxidation of metal sulfides contained in mine waste heaps, active or abandoned mine workings, or in tailings piles left over from the processing of sulfide ores. The iron sulfides: pyrite, marcasite and pyrrhotite, are perhaps the most common sources of AMD production, because they are ubiquitous in metal sulfide ores and because they generally are not the target of ore beneficiation processes. Numerous variables factor into the assessment of potential AMD releases, including the quantity of reactive sulfides, grain size distribution and grain morphology, bacterial activity, moisture content, and the availability of dissolved oxygen or other oxidants (e.g., Jambor et al., 2000; Lowson, 1982; Nordstrom and Southam, 1997; Rigby et al., 2006; Williamson and Rimstidt, 1994).

Acid mine drainage may form via the interaction of surface water or ground water with materials enriched in metal sulfides, such as tailings piles or the underground workings of deep mines (Figure 1). Production of AMD may occur during mine operations and may continue for many years after mines are closed and tailings dams are decommissioned from operation; consequently, evaluation of AMD is often a long-term proposition which usually adds up to high costs for site characterization, monitoring and cleanup. Estimates of the number of sites in the United States affected by AMD vary widely from 200,000 to over 550,000 (U.S. Environmental Protection Agency, 2004). Estimated costs to cleanup contamination at AMD sites are equally difficult to assess. One hundred and fifty-six hardrock mining sites were on or had the potential to be on the National Priorities List (NPL) for cleanup under the Comprehensive Environmental Response, Compensation and Liability Act (CERCLA), with potential cleanup costs of up to $24 billion dollars (U.S. Environmental Protection Agency, 2004). Mine sites are frequently remotely located, which further adds to the costs of site characterization, remediation, and monitoring.

In some cases, especially where ore host rock is capable of reacting with acidic drainage, metal concentrations may attenuate over time and space. A primary control on the process of metal attenuation at mining-impacted sites is acid neutralization (Al et al., 2000; Berger et al., 2000). Neutralizing capacity in sulfide ore tailings is predominantly from carbonate minerals (calcite and dolomite) because most non-carbonate minerals associated with metalliferous deposits are extremely slow to react and affect pH (e.g., Jambor et al., 2000). As pH increases, aqueous metal species tend to precipitate as hydroxide, oxyhydroxide, or hydroxysulfate minerals (Nordstrom, 1982). In addition, as pH increases dissolved metals may adsorb onto surfaces of these newly formed minerals and/or other surfaces present in the environment, such as organic matter due to decreasing competition with protons, decreased surface potential, and increased hydrolysis of metal ions at circum-neutral pH.

For further information contact Richard T. Wilkin (580) 436-8874 [wilkin.rick@epa.gov] at the Ground Water and Ecosystems Restoration Division of the National Risk Management Research Laboratory, Office of Research and Development, U.S. Environmental Protection Agency, 919 Kerr Research Drive, Ada, Oklahoma 74820.

2

In the context of mining sites, natural attenuation refers to the observed reduction of contaminant concentrations and/or contaminant mass flow rates as contaminants are transported downgradient from their source. At many mining sites that require cleanup of contaminated ground water, monitored natural attenuation is not expected to be relied upon as a sole remedy. The global magnitude of the acid drainage problem is clear evidence that in most cases natural processes are incapable of ameliorating the acidity and metal contamination produced by oxidizing sulfide minerals. However, monitored natural attenuation may be an effective strategy to augment more active approaches of remediation (summarized below). In addition, natural attenuation processes often tend to spread contaminants out in space away from source zones via various mineral-water reactions. Therefore, it is important to recognize natural attenuation processes from the perspective of tracking contaminant transport and fate in the environment. This Issue Paper provides remedial project managers and other state or private remediation managers and their technical support personnel with background information on the various physical, chemical, and biological processes of natural attenuation that may occur at mining sites. This background information is necessary for preparing sampling plans to support site characterization, remedy selection, and post-remedial monitoring efforts.

Treatment Strategies for AMD

Strategies for dealing with AMD include source control and chemical or biological treatment of contaminated ground water. Preventing the formation or the migration of AMD from source zones is generally a favorable option, but is often difficult to accomplish effectively due to the large aerial extent of tailings areas and the large volumes of materials involved. Consequently, cleanup efforts are usually focused on directly treating impacted ground water and surface water (U.S. Environmental Protection Agency, 2006). Remediation methods at mining sites can be generally divided into "active" and "passive" approaches (Johnson and Hallberg, 2005). Perhaps the most straightforward and common active approach to treat acidic effluents is through addition of alkaline materials to raise pH, increase the rate of ferrous iron oxidation, and cause the removal of metals and metalloids present in solution via mineral precipitation or surface adsorption processes. A number of alkaline materials have been used for active treatment, including lime, calcium carbonate, sodium carbonate, sodium hydroxide, and magnesium oxide. Use of these materials can lead to effective controls on the release of acidic drainage and dissolved metals. The cost of maintaining direct treatment facilities is often high. Large volumes of a low-density sludge result from the reaction between alkaline compounds and acidic effluents. Moreover, the sludge itself becomes an environmental concern, both in terms of disposal and the potential release of contaminants through subsequent leaching (Jambor et al., 2000).

Figure 2 presents a general classification of passive AMD treatment systems, which can be broadly grouped into chemical and biological systems (Neculita et al., 2007). Treatment systems that rely largely on abiotic chemical processes include open limestone channels, anoxic limestone drains, and successive alkalinity-producing systems. In open limestone channels, acidic water flows over crushed limestone or some other alkaline agent. The goals of such applications are to generate alkalinity, neutralize pH, and remove soluble aluminum, iron, and manganese via mineral precipitation (Mukhopadhyay et al., 2007; Ziemkiewicz et al., 1997). The treatment strategy with anoxic limestone drains is to neutralize acid-mine drainage using limestone while maintaining iron in a reduced state (as ferrous iron)

Figure 1. Impacts of mine workings and mine wastes on ground water and surface water. As oxygenated waters interact with sulfide minerals, pH decreases and metals are mobilized in the absence of acid-neutralizing rocks and overburden.3

to avoid iron oxidation and the consequent precipitation of hydrous ferric oxides on the limestone surfaces. Such surface precipitation or armoring by iron and aluminum precipitates greatly reduces the effectiveness of the neutralizing material. In its simplest form, an anoxic limestone drain (ALD) is a buried, limestone-filled trench that intercepts AMD before it is exposed to atmospheric O2 (Cravotta and Trahan, 1999; Hedin et al., 1994). ALD designs are enclosed to minimize gas flux in contrast to systems such as limestone channels that are open to the atmosphere. In this way CO2 gas is retained in the subsurface channel which leads to enhanced calcium carbonate dissolution and alkalinity production. After acidic waters pass through the ALD, effluents are exposed to the atmosphere and hydrous ferric oxides are produced by the oxidation of ferrous iron to ferric iron. In successive alkalinity-producing systems, both limestone and organic matter are used in vertical flow systems to provide alkalinity production, sulfate reduction, and metal removal (Keplar and McCleary, 1994).

Biological passive treatment systems for AMD include bioreactors and constructed wetlands (Figure 2). Sulfate-reducing passive bioreactors have received recent attention as promising technologies for AMD treatment (e.g., Alvarez et al., 2007; Annachatre and Suktrakoolvait, 2001; Costa and Duarte, 2005; Drury, 1999; Dvorak et al., 1992; Johnson and Hallberg, 2005; Neculita et al., 2007; Steed et al., 2000). The advantages of this technology are high metal removal capacity, stable sludge, and low operation costs. The chemical basis for treatment of AMD by sulfate reducing bacteria involves microbially-mediated sulfate reduction coupled to organic matter oxidation. Sulfide precipitation is the desired mechanism of contaminant removal, but other mechanisms including adsorption and precipitation of metal hydroxides occur in passive bioreactors (Neculita et al., 2007). The efficiency of sulfate-reducing bioreactors is primarily controlled by the organic carbon source (Coester et al., 2006; Gibert et al., 2004; Prasad et al., 1999; Zamzow et al., 2006). Solid-phase testing suggests that organic substrates with high protein contents or low lignin contents (e.g., manure) are better capable of supporting bacterial activity and sustaining contaminant removal (Coester et al., 2006; Gibert et al., 2004).

Artificial wetlands and biological treatment systems have been used since the mid-1980s for the treatment of AMD. The effectiveness of these applications has been variable and generally difficult to predict (Barton and Karathanasis, 1999; Wieder, 1989; Wildeman and Updegraff, 1997), although recent process-based modeling efforts are beginning to provide insight into the functioning of these systems that feedback into system design refinements and operational improvements (Whitehead et al., 2005). Both aerobic and anaerobic processes contribute to contaminant removal in constructed wetland systems. The oxidation of ferrous iron to ferric iron and the subsequent precipitation of hydrous ferric oxides is a dominant process that is effective in removing iron and other metals from AMD (Brenner, 2001). Metal accumulation in these systems also occurs through precipitation of metal sulfides via the activity of sulfate reducing bacteria that consume natural organic matter and sulfate and produce reactive sulfide for metal removal (Webb et al., 1998).

Permeable reactive barriers (PRBs) and monitored natural attenuation (MNA) for treatment of ground water impacted by AMD can be classified either as chemical or biological passive treatment systems based on process function (Figure 2). The application of PRBs involves the excavation of a trench or pit in the flow path of contaminated ground water. The excavated volume is then filled with reactive materials that are permeable to allow flow of contaminated ground water and reaction to remove dissolved contaminants via chemical or biological processes. Reactive materials that have been shown to be effective in increasing pH and removing metals include mixtures of organic carbon, limestone, and zero-valent iron (Benner et al., 1999; Gibert et al., 2003; Ludwig et al., 2002; Shokes and Möller, 1999; Waybrant et al., 1998; Wilkin and McNeil, 2003). Organic carbon-based PRB systems take advantage of anaerobic microbiological processes within the PRB to generate alkalinity and remove metals as sulfides. In zero-valent iron PRB systems, a variety of abiotic and biotic metal uptake processes are important in neutralizing acidity and removing metals from solution. Similarly, MNA involves both biologically mediated processes and abiotic geochemical processes that are presented in the following sections. Note that many of the documented natural attenuation processes are actually strategically enhanced in designed remediation systems, such as bioreactors.

Background on MNA for Inorganic Contaminants

The term "monitored natural attenuation" refers to the long-term examination of natural processes with the objective that such processes will reach site-specific remedial objectives. MNA can be applied in conjunction with other cleanup

Figure 2. Treatment systems for acid mine drainage.4

approaches, such as source removal, source control, or plume control (Rügner et al., 2006). To be considered as an acceptable option, MNA would be expected to achieve site remedial objectives within a time frame that is reasonable compared to that possible by using other more active remediation technologies as described above. Natural attenuation processes include a variety of physical, chemical, and biological processes that can act to reduce the mass, mobility, volume, or concentration of contaminants in ground water. Attenuation processes important at mining sites include pH buffering and acid neutralization, adsorption at the mineral-water interface, mineral precipitation, and dilution/dispersion (Figure 3).

Figure 3. Processes of natural attenuation at mine-impacted sites.

EPA’s Office of Research and Development is preparing a technical resource document for the application of MNA to inorganic contaminants in ground water (see Reisinger et al., 2005; U.S. Environmental Protection Agency, 2007). The technical resource document presents a four-tiered assessment of MNA as a viable remediation option for selected metal, metalloid, and radionuclide contaminants encountered in ground water. Components of the tiered approach include 1) demonstrating contaminant sequestration mechanisms, 2) estimating attenuation rates and the 3) attenuation capacity of aquifer solids, and 4) evaluating potential reversibility issues. The technical resource document is intended to provide a tiered decision-making approach for determining whether MNA is likely to be an effective remedial approach for inorganic contaminants in ground water at a particular location. EPA expects that users of this document will include EPA and State cleanup programs and their contractors, especially those individuals responsible for evaluating alternative cleanup methods for a given site or facility. A decision-making approach is provided for evaluating MNA as a possible cleanup method for contaminated ground water. Emphasis is placed on developing a more complete understanding of the site through development of a conceptual site model that includes an understanding of the attenuation mechanisms, the geochemical conditions governing these mechanisms, and indicators that can be used to monitor attenuation progress (U.S. Environmental Protection Agency, 2007).

A tiered decision-making approach is an appropriate and cost-effective way to screen out sites unsuitable for MNA while collecting the most relevant data at sites that might be amenable to this technology. Conceptually a tiered assessment of MNA seeks to progressively reduce site uncertainty as MNA-specific data is collected. MNA for inorganics and radionuclides is most effectively implemented through four tiers that require progressively greater information on which to assess the reasonableness of MNA:

Tier I. The plume is not threatening public health, is stable, and some direct evidence of contaminant attenuation exists.

Tier II. The attenuation capacity of the site exceeds the estimated mass of contaminant at the site.

Tier III. There is strong evidence that attenuation mechanism(s) will prevail over long periods of time.

Tier IV. A record of decision including a long-term monitoring plan and other site closure considerations is developed.

MNA Processes at Mining Sites

Based upon the tiered approach presented above, assessments of MNA at mining sites must demonstrate that chemical, physical, or biological processes are occurring to mitigate migration of contaminants, that the capacity of the MNA process exceeds the mass of contaminants in the source, and that the attenuation processes are sustainable over long periods of time. As noted previously, these conditions are frequently not met at mining sites. Although research findings clearly show that attenuation of contaminants does indeed occur at mining sites (Table 1), the documented mechanisms of attenuation are either rate- or capacity-limited so that contaminants are only partially attenuated or attenuation occurs over longer flow paths than are acceptable from a site cleanup perspective. It is equally clear, however, that many of the attenuation processes important at mining sites are long-lived, so that a sound understanding of the factors that control transport and fate of metals in ground water and across the ground water/surface water interface can benefit site cleanup efforts.

Acid Neutralization

A primary control on the process of metal attenuation at mining sites is acid neutralization. Many factors affect the acid neutralization capacity of a system, including the type, abundance and reactivity of metal-bearing sulfides in the ore and waste rock, permeability of the mine workings or mine tailings, and the ability of the host or surrounding rocks to consume acidity. Methods are available to predict whether or not materials will be acid-generating (e.g., U.S. Environmental Protection Agency, 1994). These methods provide a numerical accounting with respect to prediction of acid production and neutralization potential. In general, materials containing elevated concentrations of carbonate minerals or that have elevated inorganic carbon to total sulfur ratios are the most effective in neutralizing acidity.

Mixing of mine effluents with ambient ground water and surface water dilutes the dissolved contaminants and can result in pH increases. In surface waters, dilution and neutralization can occur over spans of meters to many kilometers. Dilution and neutralization are often tied to seasonal 5

variations in flow patterns and volumes. Field studies have documented more effective attenuation of mine effluents in dry seasons compared to wet seasons (Webster et al., 1994). In addition, field studies and laboratory column testing results indicate that mineral assemblages present in tailings piles, underlying aquifers, and receiving surface waters play a pivotal role in controlling pH (Blowes and Ptacek, 1994; Jurjovec et al., 2002; Morin et al., 1988; Sánchez-Espańa et al., 2005). Mineral phases important in buffering pH are calcite/siderite, aluminum hydroxides, iron hydroxides, and aluminosilicates.

Role of Secondary Minerals

The oxidation of iron sulfides in mine wastes results in the release of iron, sulfate, acidity, and metals to solution. High aluminum and silica concentrations are also commonly encountered in mine effluents and are the result of weathering of aluminosilicate minerals at low pH. Oxidation and hydrolysis reactions can subsequently lead to the precipitation of a wide array of hydroxide, sulfate, and/or hydroxysulfate minerals depending on geochemical and biogeochemical conditions (Nordstrom and Alpers, 1999). These secondary minerals play important roles in attenuating contaminants from mine effluents (e.g., Accornero et al., 2005; Casiot et al., 2005; Fukushi et al., 2003; Gault et al., 2005; Jamieson et al., 1999; Lee et al., 2002; Lee et al., 2005; Levy et al., 1997; McCarty et al., 1998; McGregor et al., 1998; Moncur et al., 2005; Munk et al., 2002; Sánchez-Espańa et al., 2005; Sidenko and Sherriff, 2005; Webster et al., 1998; Zänker et al., 2002). Some of the common secondary minerals found in association with the weathering of mine wastes, their typical pH range of formation, and documented contaminant associations are listed in Table 1.

Table 1. Secondary Minerals Formed from Acid Mine Waters and Contaminant Associations

Mineral Phase

Formula

Typical pH range of formation

Examples of contaminant associations at mine-impacted settings

Reference

Hydroxides

Goethite

FeOOH

2-4

Sorption/coprecipitation of Pb (up to 21 wt%), As (up to 7.7 wt%), Zn (up to 4.6 wt%), and Cu (up to 2.5 wt%)

Lee et al. (2005)

HFO, hydrous ferric oxide

~Fe5HO8•4H2O

>5

Sorption/coprecipitation of As-rich ferrihydrite; As/Fe=0.02-0.1, with 10-30% As(III)

Casiot et al. (2005)

Gibbsite

Al(OH)3

>5-6

Sorption in the general order of Pb>Cu>Zn>Ni with increasing pH

Munk et al. (2002)

Hydroxysulfates

Alunite

KAl3(OH)6(SO4)2

4-6

Precipitation of Al

Accornero et al. (2005)

Jarosite

KFe3(OH)6(SO4)2

2-5

Coprecipitation with As(V) replacing sulfate in the jarosite structure

Gault et al. (2005)

Schwertmannite

Fe8O8(OH)6(SO4)

2-4

Coprecipitation of Cu, Ni and Zn

Sidenko and Sherriff (2005)

Sulfates

Gypsum

CaSO4•2H2O

>3

"Hardpan" precipitate

Moncur et al. (2005)

Anglesite

PbSO4

>3-4

Precipitation at pH ~3; nanoparticles

Zänker et al. (2002)

Melanterite

FeSO4•7H2O

<2

Coprecipitation with Zn and Cu; temporary metal removal in a highly soluble phase

Jamieson et al. (1999)