Remediation technologies for metal-contaminated ...

Engineering Geology 60 (2001) 193±207

www.elsevier.nl/locate/enggeo

Remediation technologies for metal-contaminated soils and groundwater: an evaluation C.N. Mulligan a,*, R.N. Yong b, B.F. Gibbs c a

Department of Building, Civil and Environmental Engineering, Concordia University, 1455 de Maisonneuve Blvd. W., Montreal, Quebec, Canada H3G 1M8 b Geoenvironmental Engineering Research Centre, Cardiff School of Engineering, Cardiff University, P.O. Box 917, Newport Road, Cardiff, CF2 1XH, United Kingdom c MDS Pharma Services, 2350 Cohen Street, Montreal, Canada H4R 2N6 Accepted for publication 28 March 2000

Abstract Metals including lead, chromium, arsenic, zinc, cadmium, copper and mercury can cause signi®cant damage to the environment and human health as a result of their mobilities and solubilities. The selection of the most appropriate soil and sediment remediation method depends on the site characteristics, concentration, types of pollutants to be removed, and the end use of the contaminated medium. The approaches include isolation, immobilization, toxicity reduction, physical separation and extraction. Many of these technologies have been used full-scale. This paper will review both the full-scale and developing technologies that are available. Contaminants can be isolated and contained to minimize further movement, to reduce the permeability of the waste to less than 1 £ 10 27 m/s (according to U.S. guidelines) and to increase the strength or bearing capacity of the waste. Physical barriers made of steel, cement, bentonite and grout walls can be used for isolation and minimization of metal mobility. Another method is solidi®cation /stabilization, which contains the contaminants in an area by mixing or injecting agents. Solidi®cation encapsulates contaminants in a solid matrix while stabilization involves formation of chemical bonds to reduce contaminant mobility. Another approach is size selection processes for removal of the larger, cleaner particles from the smaller more polluted ones. To accomplish this, several processes are used. They include: hydrocyclones, ¯uidized bed separation and ¯otation. Addition of special chemicals and aeration in the latter case causes these contaminated particles to ¯oat. Electrokinetic processes involve passing a low intensity electric current between a cathode and an anode imbedded in the contaminated soil. Ions and small charged particles, in addition to water, are transported between the electrodes. This technology have been demonstrated in the U.S. full-scale, in a limited manner but in Europe, it is used for copper, zinc, lead, arsenic, cadmium, chromium and nickel. The duration of time that the electrode remains in the soil, and spacing is site-speci®c. Techniques for the extraction of metals by biological means have been not extensively applied up to this point. The main methods include bioleaching and phytoremediation. Bioleaching involves Thiobacillus sp. bacteria which can reduce sulphur compounds under aerobic and acidic conditions (pH 4) at temperatures between 15 and 558C. Plants such as Thlaspi, Urtica, Chenopodium, Polygonum sachalase and Alyssim have the capability to accumulate cadmium, copper, lead, nickel and zinc and can therefore be considered as an indirect method of treating contaminated soils. This method is limited to shallow depths of contamination. Soil washing and in situ ¯ushing involve the addition of water with or without additives including organic and inorganic acids, sodium hydroxide which can dissolve organic soil matter, water soluble solvents such as methanol, nontoxic cations, complexing agents such as ethylenediaminetetraacetic acid (EDTA), acids in combination with * Corresponding author. Fax: 11-514-848-2809. E-mail address: [email protected] (C.N. Mulligan). 0013-7952/01/$ - see front matter q 2001 Elsevier Science B.V. All rights reserved. PII: S 0013-795 2(00)00101-0

194

C.N. Mulligan et al. / Engineering Geology 60 (2001) 193±207

complexation agents or oxidizing/reducing agents. Our research has indicated that biosurfactants, biologically produced surfactants, may also be promising agents for enhancing removal of metals from contaminated soils and sediments. In summary, the main techniques that have been used for metal removal are solidi®cation/stabilization, electrokinetics, and in situ extraction. Site characteristics are of paramount importance in choosing the most appropriate remediation method. Phytoremediation and bioleaching can also be used but are not as well developed. q 2001 Elsevier Science B.V. All rights reserved. Keywords: Soil; Remediation technologies; Groundwater; Sediments; Heavy metals

1. Introduction In the United States, 1200 sites are on the National Priority List (NPL) for the treatment of contaminated soils, indicating the extensiveness of this problem. Approximately 63% of the sites on the NPL include contamination from toxic heavy metals (Hazardous Waste Consultant, 1996). For example, lead was found at 15% of the sites, followed by chromium, cadmium and copper at 11, 8 and 7% of the sites, respectively. Therefore, metal contamination is a major problem. Cadmium, copper, lead, mercury, nickel and zinc are considered the most hazardous and are included on the US Environmental Protection Agency's (EPA) list of priority pollutants (Cameron, 1992). Sources of metals include domestic and industrial ef¯uents, the atmosphere, runoff and lithosphere. Once metals are allowed to pass through the municipal waste treatment facility, the heavy ones return to the environment where they are persistent, cannot be biodegraded and can thus follow a number of different pathways. The metals can adsorb onto the soil, runoff into rivers or lakes or leach in the groundwater, an important source of drinking water. Exposure to the heavy metals through ingestion or uptake of drinking water (particularly where water is reused) and foods can lead to accumulation in animals, plants and humans. Table 1 Global production of metals and the rate of metals reaching the soil (10 3 ton/year) (adapted from World Resources Institute (1992/ 1993) and Nriagu and Pacyna (1988)) Metal

1975

1980

1985

1990

Emissions to the soil in the 1980s

Cd Cu Pb Zn

15.2 6739.0 3432.2 3975.4

18.2 7204.0 3448.2 4030.3

19.1 7870.0 3431.2 4723.1

20.2 8814.0 3367.2 5570.9

22 954 796 1372

This phenomenon can lead to extinction or alteration of plants and animals. Metals can accumulate in the following order, river sediments, bacteria, tubicids and then ®sh and man if one consumes these ®sh. Over the past years, use of metals such as copper, cadmium and zinc have increased substantially (Table 1). Copper is produced more than any other metal, whereas more zinc reaches the soil than any other metal. Lead use has decreased due to toxicity concerns. In Canada, according to the National Pollutant Release Inventory, approximately 13,300 ton of copper, 9500 ton of zinc, 1300 ton of lead and 33 ton of cadmium were released to the air, water and soil (NPRI, 1995). In view of the extensiveness of metals in the environment, this paper describes the fate and transport of selected metals and technologies for remediation that are full-scale and developing. This information will assist in the selection of the appropriate technology for treatment of metal-contaminated soils and sediments.

2. Mobility of metals In its natural form, cadmium is relatively rare and concentrated in argillaceous and shale deposits as greenockite (CdS) or otavite (CdCO3) and is usually associated with zinc, lead or copper in sul®de form (Cameron, 1992). It is a bluish-white soft metal or grayish powder. It is more mobile, though, than zinc at low pH, particularly at pH values between 4.5 and 5.5. Above pH 7.5, cadmium is not very mobile. Its divalent form is soluble but it can also complex with organics and oxides. A natural source of cadmium is volcanoes that can release cadmium into the atmosphere, spreading it over a wide area. It is only in the last twenty years that cadmium has become a concern due to the extensive use in industrial applications including steel plating, pigment stabilization and


nickel±cadmium batteries (Fassett, 1980). The average content of cadmium in soil is less than 1 ppm. In plants, the normal range is 0.005±0.02 ppm with toxic levels between 5 and 30 ppm. Sources of cadmium include alloys, polyvinyl chloride plastic (PVC) manufacture, solders, fungicides, enamels, motor oil, textile manufacturing, electroplating and rubber, sewage sludges and phosphate fertilizers (Matthews, 1984). It enters the environment through industrial ef¯uents and land®ll leaching, spills and leaks at hazardous waste sites, mining and household wastes. Copper is found naturally in sandstones and in minerals such as malachite and chalcopyrite. It is a reddish-brown metal that binds strongly to organic matter and clay minerals, thus decreasing its mobility. The organic matter, however, can be degraded through anaerobic or aerobic means releasing the copper in its monovalent or divalent states, respectively. It has also been demonstrated that biosurfactants can release organically bound copper (Mulligan et al., 1999a). The average content of copper in rural soils is 2 to 100 ppm. Plants can accumulate copper with average contents in the range of 5±30 ppm while toxic levels vary from 20 to 100 ppm. Increased levels of copper are due to uses in fertilizers, building materials, rayon manufacture, pesticide sprays, agricultural and municipal wastes and industrial emissions (Cameron, 1992). Lead is found naturally in soils, most commonly in the form of the ore gelena (PbS) and in smaller quantities in cerussite (PbCO3), anglesite (PbSO4) and crocoite (PbCrO4). It is a bluish-white, silvery or gray metal with a high density of 11.4 g/cm 3. Lead can be found in soils at the surface and organic matter in higher quantities. Sources of lead include lead±zinc smelters, ammunition, solder, glass, piping, insecticides, paints and batteries (Jawarsky, 1978). The divalent form is the most common and is capable of replacing calcium, strontium, barium and potassium in soils. In general, background levels less than 10 ppm are found, and mobility of lead in soils is low. Organic matter can adsorb substantial quantities of lead. Lead is released into the air from burning of wastes and fossil fuels and subsequently lands onto the soil. It also reaches the soil from land®lls and paints. Although not as toxic as cadmium, zinc is quite often associated with this metal. Zinc is a soft, white

195

metal with bluish tinge. Soil texture, pH, nature of the parent rocks and organic content all affect the natural content of zinc in the soil. Under acidic conditions, zinc is usually divalent and quite mobile. At high pH, zinc is bioavailable due to the solubility of its organic and mineral colloids. Zinc hydrolyzes at pH 7.0±7.5, forming Zn(OH)2 at pH values higher than 8. Under anoxic conditions, ZnS can form upon precipitation, whereas the unprecipitated zinc can form ZnOH 1, ZnCO3 and ZnCl 1. Natural levels of zinc in soils are 30±150 ppm. Levels of 10±150 ppm are normal in plants while 400 ppm is toxic. Sources of zinc include brass and bronze alloys, galvanized products, rubber, copying paper, cosmetics, pharmaceuticals, batteries, televisions, tires, metal coatings, glass, paints and zinc-based alloys (Cameron, 1992). It can enter the environment from galvanizing plant ef¯uents, coal and waste burning, leachates from galvanized structures, natural ores and municipal waste treatment plant discharge. Zinc is commonly found in wastes as zinc chloride, zinc oxide, zinc sulfate and zinc sul®de (Agency for Toxic Substances and Disease Registry, 1995). 3. Metal speciation The term speciation is related to the distribution of an element among chemical forms or species. Heavy metals can occur in several forms in water and soils. Interest has increased in sequential extraction techniques to relate the degree of mobility with risk assessment, (i.e. the more mobile the metal is, the more risk associated with it (Bourg, 1995)) and as a method of designing remediation techniques (Mulligan et al., 1999b). Not only is total metal concentration of interest, but is also now accepted that understanding the environmental behavior by determining its speciation is of paramount importance. Based on this information on contaminated soil, the most appropriate method for soil remediation can be determined. To determine the speciation of metals in soils, speci®c extractants are used. The different extractants solubilize different phases of metals. By sequentially extracting with solutions of increasing strengths, a more precise evaluation of the different fractions can be obtained (Tessier et al., 1979). A soil or

196


sediment sample is shaken over time with a weak extractant, centrifuged, and the supernatant is removed by decantation. The pellet is washed in water and the supernatant removed and combined with the previous supernatant. A sequence of reagents is used, following the same procedure, until ®nally, mineral acid is used to extract the residual fraction. Heavy metal concentrations are then determined in the various extracts by atomic absorption or other means. Numerous techniques and reagents have been developed and have been applied to soils (Shuman, 1985), sediments (Lum and Edgar, 1983), sludge-treated soils (Petrozelli et al., 1983) and sludges (Lakanen and Ervio, 1971). These methods are not standardized and the results can even vary with the same reagents, pH, temperature, extractant strength and solid sample to volume of extractant ratio. None of the extractions is completely speci®c. However, the extractants chosen attempt to minimize solubilization of other fractions. To extract the exchangeable fraction, ammonium acetate, barium chloride or magnesium chloride at pH 7.0 are generally used (Lake, 1987). They cause the displacement of the ions in the soil or sediment matrix bound by electrostatic attraction. Pickering (1986) showed that magnesium chloride leached low quantities of other sul®des, organic matter, aluminum and silicon. The carbonate phase (calcite and dolomite) is extracted at pH 5.0 with sodium acetate acidi®ed with acetic acid. This solubilizes the carbonates, releasing carbonate-entrapped metals. Organic matter, oxides or clay components are not solubilized. The reducible phase (iron and manganese oxides) is extracted with hydroxylamine hydrochloride with acetic acid at pH 2.0. The hydroxylamine hydrochloride reduces the ferrous and manganese hydroxides to soluble forms. Other components such as organic matter and clay components are not solubilized to any great extent (Tessier et al., 1979). Hot hydrogen peroxide in nitric acid is used to oxidize the organic matter and solubilize sul®des. The oxidized organic matter then releases metals that are complexed, adsorbed and chelated. These agents are used so that the silicates are not affected by this treatment. In the ®nal step, the silicates and other materials are dissolved by strong acids at high

temperatures. This fraction is usually used to complete the mass balances for the metals. In natural soils, Kabata-Pendias (1992) demonstrated that the speciation of trace metals depends on the physical and chemical characteristics of the soil. Soil pH, redox, organic, carbonate, clay and oxide contents all in¯uence metal speciation and mobility. Simple and complex cations are the most mobile, exchangeable cations in organic and inorganic complexes and are of medium mobility while chelated cations are slightly mobile. Metals in organic or mineral particles are only mobile after decomposition or weathering, and precipitated metals are mobile under dissolution conditions (e.g. change in pH). Kabata-Pendias (1992) also showed the speciation of trace metals such as zinc, copper, cadmium and lead. Zinc and cadmium are mostly organically bound, exchangeable and water soluble. Copper is mainly organically bound and exchangeable, whereas, lead is slightly mobile and bound to the residual fraction. Chlopecka (1992) showed, however, that the cadmium and zinc speciation in soils depended signi®cantly on the application of sewage sludge on the soil. Fertilizer addition, water and air pollution can also affect speciation. Recently, sequential extraction techniques have been studied as a tool in various applications. Yong et al. (1993) examined sequential extraction to obtain a better appreciation of the ability of clay soil barriers to contain contaminants in land®ll barriers. The effect of soil pH, constituents and heavy metal types were evaluated. In a study by Ramos et al. (1994), sequential extraction techniques were used to evaluate the mobility of cadmium, zinc, lead and copper in contaminated soil in a national park. Cadmium was found to be the most mobile and would likely be the most bioavailable. Ravishankar et al. (1994) evaluated several sludges for Al, Cu, Fe, Mn and Zn speciation to predict bioleaching processes. They concluded that more stabilized sludges contained higher contents of organically bound metals and that sludges vary considerably making bioleaching prediction dif®cult. A potential method to determine if the heavy metals can be removed by remediation techniques (such as soil washing) or to predict removal ef®ciencies is to determine speciation with selective extractive techniques. It is believed that exchangeable, carbonate


and reducible oxide fraction may be amenable to soil washing techniques (Li et al., 1995). Removal of the organically and residually bound fraction may not be economical to recover or necessary, due to lack of bioavailability. Gombert (1994) used sequential extraction to determine if cesium, cobalt and chromium could be removed by soil washing. Since less than 20% was extracted after dissolving 20% of the soil mass, soil washing was abandoned as an option. Mulligan et al. (1999a) demonstrated that sequential extraction techniques could be used prior to soil washing, to design and monitor the remediation process. Copper associated with organic matter and zinc associated with oxides were successfully removed by biosurfactant/hydroxide solutions and biosurfactant/ acid solutions, respectively. Clearly, more work is needed in this ®eld.

4. Remediation techniques 4.1. Isolation and containment Contaminants can be isolated and contained, to prevent further movement, to reduce the permeability of the waste to less than 1 £ 10 27 m/s (as required by the U.S. EPA) and to increase the strength or bearing capacity of the waste (USEPA, 1994). Physical barriers made of steel, cement, bentonite and grout walls can be used for capping, vertical and horizontal containment. Capping is a site-speci®c proven technology to reduce water in®ltration. Synthetic membranes can be used for this purpose. Vertical barriers reduce the movement of contaminated groundwater or uncontaminated groundwater through a contaminated area. To prevent the transport of contaminants past the barrier, the barrier should extend to a clay or bedrock layer of low permeability. If this cannot be done, a groundwater extraction system would be required to avoid the passage of contaminants below the barrier (Rumer and Ryan, 1995). Slurry walls, grout or geomembrane curtains, and sheet pile walls are employed. Slurry walls are the least expensive and are thus the most common. Although there are many variations, a vertical trench is always constructed under a slurry such as bentonite and water. Horizontal barriers within the soil (trenches or

197

wells) are under development and have not been demonstrated as effective but are potentially useful in restricting downward movement of metal contaminants by acting as underlying liners without the requirement for excavation. Grout injection by either vertical boring or horizontal drilling and block displacement are the main types of horizontal barriers. There have been problems with soil compaction and vertical boreholes can increase the likelihood of contaminant migration. Solidi®cation/stabilization technologies are very common in the United States (Conner, 1990) as they contain the contaminants, not the contaminated area like physical barriers. Solidi®cation is physical encapsulation of the contaminants in a solid matrix while stabilization includes chemical reactions to reduce contaminant mobility. Some metals such as arsenic, chromium (VI) and mercury are not suitable for this type of treatment since they do not form hydroxides that are not highly soluble. Liquid monomers that polymerize, pozzolans, bitumen, ¯y ash and cement are injected to encapsulate the soils. Soils can be treated in situ or after excavation. However, there are few vendors of in situ processes while many exist for ex situ processes since mixing in situ is dif®cult to evaluate. For ex situ processes, in-drum, in-plant or area mixing processes are used. Smallscale pilot plants can treat up to 100 ton per day of contaminated soil whereas 500±1000 ton of soil per day can be stabilized in large plants (Smith et al., 1995). In situ solidi®cation/stabilization techniques are preferred since labor and energy costs are lower but site conditions such as bedrock, large boulders, clays and oily patches may cause mixing problems. In situ processes are most suitable for shallow contamination since conventional equipment including draglines, backhoes, clamshell buckets and vertical auger mixers are used (Jasperse and Ryan, 1992). Augers vary from 1 to 3 m in diameter and can reach depths of 13 m. US regulatory agencies are receptive to this technology since the contaminants are not transferred to another medium which would also require treatment. In the UK, an in situ process called the Colmix process (slurry including cement, slag-based grout and lime) was used to immobilize heavy metals and ammonium at the ICI Explosives' land®ll in Scotland (Wheeler, 1995).

198


A

Graphite and glass frit starter path Electodes

Contaminated soil area

B

Subsidence

Electrodes

Natural soil

C

Backfill over monolith

Vitrified monolith Fig. 1. Diagram showing steps in the vitri®cation process for metal, including (A) insertion of electrodes and placement of graphite and glass frit starter path to initiate vitri®cation, (B) subsidence of the soil during vitri®cation and (C) placement of back®ll over vitri®ed monolith.

Vitri®cation is a solidi®cation/stabilization process requiring thermal energy. It involves insertion of electrodes into the soil which must be able to carry a current, and then to solidify, as it cools (Fig. 1). Toxic gases can also be produced during vitri®cation. Full-scale applications exist for arsenic, lead and chromium contaminated soils. Mixed wastes can also be treated in this manner. High clay and moisture contents and debris can affect the ef®ciency of the process. These solidi®cation/stabilization processes are suitable for contamination in shallow depths and of large volume. Leaching of the contaminants must, however, be carefully monitored as is the case for vitri®cation, the formation of a glassy solid. 4.2. Mechanical separation The aim of the size selection processes is to remove the larger, cleaner particles from the smaller, more

polluted ones. Characterization in terms of particle size and contaminant level in each fraction is the most important parameter in determining the suitability of this process. Bench-scale tests to evaluate the separation technique are also valuable. To accomplish this, several processes are used. They include: hydrocyclones which separate the larger particles greater than 10±20 mm by centrifugal force from the smaller particles, ¯uidized bed separation which removes smaller particles at the top (less than 50 mm) in the countercurrent over¯ow in a vertical column, by gravimetric settling and ¯otation which is based on the different surface characteristics of contaminated particles. These methods have been used in mineral ore processing. Addition of special chemicals such as frothers or ¯otation agents and aeration causes these contaminated particles to ¯oat. Magnetic separation, which is based on the magnetic properties of metals, can also be used to separate these from ferrous materials. Physical separation techniques are becoming more common and applications will continue to increase as they can be used to remove metal contamination in a particular form or in combination with other processes, to reduce the volume of soil to be treated by other methods. Most times, offthe-shelf equipment from the mining industry (hydrocyclones, centrifuges, screens, etc.) can be used. 4.3. Pyrometallurgical separation Pyrometallurgical processes use high temperature furnaces to volatilize metals in contaminated soil. Temperatures of 200±7008C are used to evaporate the contaminants. After volatilization, metals are then recovered or immobilized. These methods are most applicable to mercury since it is easily converted to its metallic form at high temperatures. Other metals including lead, arsenic, cadmium and chromium may require pretreatment with reducing or ¯uxing agents to assist melting and provide a uniform feed. This type of treatment is usually performed off site due to a lack of mobile units and is most applicable to highly contaminated soils (5±20%) where metal recovery is pro®table. The soil then must be concentrated by physical or soil washing processes prior to pyrometallurgical processes. Mercury, however, can be easily recovered at lower concentrations (Smith et al., 1995). Other valuable metals such as gold or platinum can also be recovered


from low soil concentrations. Rotary kilns, arc furnaces or rotary hearth furnaces are the main types of equipment used in this process. They usually produce a slag with a high concentration of heavy metals that can then be recovered (USEPA, 1996a,b). The technologies include ¯uidized bed thermal desorption, high-vacuum retort, Pittsburgh Mineral and Environmental Technology's (PMET's) thermal recovery process, Remedial Technology Group's (Farragut, Tennessee) thermal screw processor and X-Traxe thermal desorption system (Hazardous Waste Consultant, 1996). Mercury levels from 1 to 228 mg/kg were obtained after treatment of soil contaminated with 1300± 34,000 mg/kg of mercury at a US EPA SITE demonstration of the X-Traxe process. Rust/OHM has a full-scale unit that can remediate 10 ton/h of soil for a 20,000± 100,000 ton of contaminated soil. Some of the metals remain in the solid residues which will have to be properly disposed of. Pretreatment is often necessary to reduce the volume of soil to be treated, produce a uniform feed and increase metal recoveries. 4.4. Chemical treatment Chemical treatment by reductive as well as oxidative mechanisms may be used to detoxify or decrease the mobility of metal contaminants (Evanko and Dzombak, 1997). This method is commonly used for wastewater treatment. Oxidization reactions which detoxify, precipitate or solubilize metals involve addition of potassium permanganate, hydrogen peroxide, hypochlorite or chlorine gas. Neutralization reactions are performed to adjust the pH of acidic or basic soils. Reduction reactions are induced through the addition of alkali metals such as sodium, sulfur dioxide, sul®te salts and ferrous sulfate. Sometimes chemical treatment is used to pretreat the soil for solidi®cation or other treatments. For example, chemical reduction of Cr(VI) is performed during solidi®cation/stabilization. Oxidation is less commonly used with solidi®cation/stabilization. These reactions are, however, not speci®c and there is therefore, a risk of converting other metals into more toxic or mobile forms. Arsenic is most applicable for chemical oxidation since As(V) is less toxic than As(III). Co-precipitation of high concentrations of As(V) and Fe(III) forms FeAsO4 while low concentrations of As(V) co-precipitate with FeHO2 with high concentrations of Fe(III) to form arsenic ferrihydride, a product

199

Source of contaminant Ground surface Treatment wall

Remediated groundwater

Groundwater

Impermeable layer

Fig. 2. Permeable reactive treatment barrier placed in the groundwater to remove metal contaminants.

that is resistant to acid and neutral leaching (Robins, 1992). Mercury, lead, selenium and silver are also applicable for reduction. These chemical treatments can be performed in situ by injection into the groundwater but have the potential to introduce further contamination. 4.5. Permeable treatment walls Permeable barriers (Fig. 2) that contain a reactive substance (physical, chemical or biological or a combination) are being evaluated for reducing the mobilization of metals in groundwater at contaminated sites. Various materials have been investigated and include zeolite, hydroxyapatite, elemental iron and limestone (Vidac and Pohland, 1996). Preliminary results have shown that elemental iron can be used for chromium reduction and limestone for lead precipitation. The advantages of this technique are that it is in situ, a wide variety of contaminants can be treated and ¯ow control can be used. Further research is required in the areas of matching the contaminant with the media in the barrier, optimization of the ¯ow and retention time through the barrier, and methods of regenerating the media. 4.6. Electrokinetics Electrokinetic processes involve passing a low intensity electric current between a cathode and an anode imbedded in the contaminated soil (Fig. 3). Ions and small charged particles, in addition to water, are transported between the electrodes. Anions move towards the positive electrode and cations towards the negative. An electric gradient initiates

200


4.7. Biochemical processes

AC/DC Converter

Clean purge well

Extraction/processing of fluids

Extraction well

Anode +

Cathode – Groundwater level ˛ ˇ ˛ ˇ ˛

ˇ ˛˛

Metal-contaminated soil Fig. 3. Electrokinetic process for soil remediation. Cations move towards the cathode and anions towards the anode. Buffer solutions are added and removed by the purge and extraction wells.

movement by electromigration (charged chemicals movement), electro-osmosis (movement of ¯uid), electrophoresis (charged particle movement) and electrolysis (chemical reactions due to electric ®eld) (Rodsand and Acar, 1995). Buffer solutions are used to maintain the pH at the electrodes. The metals can be removed by electroplating or precipitation/coprecipitation at the electrodes, using ion exchange resins or recovering the metals by pumping the waste to the surface (Smith et al., 1995). The process can be used in situ or with excavated soil. Metals as soluble ions and bound to soils as oxides, hydroxides and carbonates are removed by this method. Other non-ionic components can also be transported due to the ¯ow. Large metal objects, rocks, foundations, rubble or other obstacles can interfere with the process (Acar and Gale, 1995). Unlike soil washing, this process is effective with clay soils of low permeability. It is mainly applicable for saturated soil with low groundwater ¯ow rates. Demonstrations of this technology have been performed, such as the Lasagnae technology, but are limited. In Europe, this technology is used for copper, zinc, lead, arsenic, cadmium, chromium and nickel. Electrode duration and spacing is site-speci®c and may need to be optimized, in addition to the type of pore ¯uid used.

Techniques for the extraction of metals by microbiological means are rather limited at this time. The main methods include bioleaching and oxidation/ reduction reactions. Bioleaching involves Thiobacillus sp. bacteria under aerobic and acidic conditions (pH 4) at temperatures between 15 and 558C, depending on the strain. Leaching can be performed by direct means, oxidation of metal sul®des to produce sulfuric acid, which then can desorb the metals on the soil by substitution of protons. Indirect leaching involves conversion of Fe 21 to Fe 31 which in turn oxidizes sulfur minerals to Fe 21 producing acidity. Several options are available for bioleaching including heap leaching, bioslurry reactors and in situ processes. Anoxic sediments are more suitable for treatment since the bacteria can solubilize the metal compounds without substantially decreasing the pH. Soils require lower pH values to extract the metals since they have already been exposed to oxidizing conditions. For both heap leaching and reactors, the bacteria and sulfur compounds are added. In the reactor, mixing is used and pH can be controlled more easily. Leachate is recycled during heap leaching. Copper, zinc, uranium and gold have been removed by Thiobacillus sp. in biohydrometallurgical processes (Karavaiko et al., 1988). Several feasibility studies have indicated that contaminated soils can be remediated by Thiobacilli (Tichy et al., 1992). Sludges from anaerobic processes that contain metal sul®des could be treated in this manner (Blais et al., 1992). Another leaching technique that has potential for remediation of metal-contaminated soil is through the production of citric and gluconic acids by the fungus Aspergillus niger can produce citric and gluconic acids. They can act as acids (pH 3.5) and chelating agents for the removal of metals such as copper from oxide mining residues (Mulligan et al., 1999c). Inexpensive carbon substrates will be required to decrease the costs of this process. Biosorption is a biological treatment method which involves the adsorption of metals into biomass such as algal or bacterial cells that can be dead or alive. If large-scale, inexpensive production techniques for the biomass are developed, this heavy metal treatment is promising (Hazardous Waste Consultant, 1996). This method is mainly applicable for removal of


low concentrations of metals in water. Therefore, the cells could potentially be placed in permeable barriers for adsorption of metals in groundwater. Microorganisms are also known to oxidize and reduce metal contaminants. Mercury and cadmium can be oxidized while arsenic and iron can be reduced by microorganisms. Cr(VI) can be oxidized to Cr(III) that is less mobile and toxic. Bacteria such as Bacillus subtilis and sulfate reducing bacteria in the presence of sulfur can perform this reaction. Sulfate-reducing bacteria (SRB) form metal (Me) sul®des that are insoluble as shown in the following reactions 2 1 CH3 COOH 1 SO22 4 ! 2HCO3 1 HS 1 H

H2 S 1 Me21 ! MeS 1 2H1 As can be seen from the equations, sulfate, low redox conditions and an electron donor such as methanol are required. Oxygen should not be there and nutrients must be added. Stimulating sulfate reduction can increase pH also and form metal hydroxides and oxides that precipitate and do not migrate in soils and groundwater. A project proposed by the Flemish Institute for Technological Research for NATO/CCMS (1999). It involves bioprecipitation of the heavy metals (zinc, cadmium, arsenic, lead, chromium, nickel, and copper) within biological reactive zones or biowalls. The ®rst site contains zinc (0±150 mg/l), cadmium (0.4±4 mg/l), and arsenic (20±270 mg/l) with high concentrations of sulfate (400±700 mg/l),

201

ideal for SRB activity. The second site has copper (up to 92 mg/l), chromium (up to 78 mg/l), zinc (up to 8.3 mg/l), nickel (up to 3.5 mg/l) with high levels of sulfate (up to 3000 mg/l). Results will be available in the coming years. Biomethylation involves the addition of a methyl (±CH3) group to a metal such as arsenic, mercury, cadmium or lead. The methylated forms are more mobile and can migrate into the groundwater. Although, methylation increases volatility, it is not likely that methylation of metals such arsenic will be performed for remediation since the by-products are more toxic. They are currently under development and not commercially available. Volatilization of selenium from contaminated agricultural soils has shown some promise (Thompson-Eagle and Frankenberger, 1990) Therefore, these processes could be used for soil and sediment treatment. However, this form of the metal may dif®cult to control in gas emissions. Another process (called mercrobes) has been developed and tested in Germany at concentrations greater than 100 ppm. Between 95 and 99% of the mercury was reduced in laboratory tests (Hazardous Waste Consultant, 1996). Since the mobility is in¯uenced by its oxidation state, these reactions can affect the contaminant mobility. 4.8. Phytoremediation

Atmosphere

Phytovolatilization

Plant

Phytoaccumulation

Soil

Phytostabilization

Plants such as Thlaspi, Urtica, Chenopodium,

Groundwater level Mechanisms for phytoremediation of metals Fig. 4. Schematic diagram showing the mechanisms of the phytoremediation process for metal uptake.

202


Polygonum sachalase and Alyssim have the capability to accumulate cadmium, copper, lead, nickel and zinc and can be therefore be considered as an indirect method of treating contaminated soils (Baker et al., 1991). This method is limited to shallow depths of contamination. Rhizo®ltration, the adsorption by plant biomass, can be used to remediate metalcontaminated groundwater. Phytoextraction involves uptake of metals by trees, herbs, grasses and crops and can be used for soil treatment. Phytostabilization is a process to excrete components from the plants to decrease the soil pH and form metal complexes. The plants will have to be isolated from wildlife and agricultural lands. The climatic conditions and bioavailability of the metals must be taken into consideration when using this method. Once contaminated, the plants will have to be disposed of in an appropriate fashion. Some techniques include drying, incineration, gasi®cation, pyrolysis, acid extractions, anaerobic digestion, extraction of the oil, chlorophyll ®bers from the plants (Bolenz et al., 1990) or disposal since plants are easier to dispose of than soil. Phytoremediation will be most applicable to shallow soils with low levels of contamination (2.5± 100 mg/kg) for polishing. Phytoremediation ®eld demonstrations (Fig. 4) are summarized in Table 2. The main disadvantage of this method is that longer times are required compared to other methods. More research is needed to enhance the extraction of the metals by the plants through genetic breeding or other methods and how to correlate bioavailability with metal uptake. Crop plants that grow fast may be viable for phytoremediation. 4.9. In situ treatment (soil ¯ushing) Extracting solutions are in®ltrated into soil using surface ¯ooding, sprinklers, leach ®elds, basin in®ltration systems, surface trenches, horizontal drains or vertical drains. Water with or without additives is employed to solubilize contaminants. The ef®ciency of the extraction depends on the hydraulic conductivity of the soil. High permeability gives better results (greater than 1 £ 10 23 cm/s). The solubility of pollutants and if the pollutant was originally solubilized in water or not affects removal ef®ciencies. Prior mechanical mixing of the soil can disturb the in®ltration of the extractant. Understanding the

chemistry of the binding of the contaminant and the hydrogeology of the site are very important (USEPA, 1987). Since water solubility is the controlling removing mechanism, additives are used to enhance ef®ciencies. In an analysis of a test site, it was determined that 400 years would be required to treat a site with water alone compared to 4 years with chemical enhanced ¯ushing (AAEE, 1993). The research in this area is still quite limited, particularly where metal removal is concerned (USEPA, 1987) but chemical enhanced ¯ushing has potential for a wide variety of metals. Little handling of soil is required. Chemical enhanced ¯ushing includes addition of organic and inorganic acids, sodium hydroxide which can dissolve organic soil matter, water soluble solvents such as methanol, displacement of toxic cations with nontoxic cations, complexing agents such as EDTA, acids in combination with complexation agents or oxidizing/reducing agents. Soil pH, soil type, cation exchange capacity (CEC), particle size, permeabilities and contaminants all affect removal ef®ciencies. High clay and organic matter contents are particularly detrimental. Once the water is pumped from the soil, it must be extracted and then treated to remove the metals in wastewater treatment facilities or reused in the ¯ushing process. Several technologies exist such as sodium hydroxide or sodium sul®de precipitation, ion exchange, activated carbon adsorption, ultra®ltration, reverse osmosis, electrolysis/ electrodialysis and biological means (Patterson, 1985). Metal recovery and recycling must be improved. Large-scale treatment has been done mostly for organic removal and is limited to metals. Full-scale treatment was performed at United Chrome, a chrome plating plant (Corvallis, OR), for removal of Cr(VI) (USEPA, 1996a,b). Water was used as the ¯ushing solution with three methods of in®ltration, in®ltration basins, injection wells and an in®ltration trench. The site had a low permeability silty soil and treatment was performed for both shallow and deep aquifers. A clay aquitard was also ¯ushed indirectly by using deep injection wells. Signi®cant removal of chromium was achieved, in addition to hydraulic containment of the plume. Levels of chromium were reduced to 18 from 2000 mg/l. At another site, a 30,000 m 3 volume has been successfully treated in The Netherlands to decrease the cadmium content in

USDA Salinity Lab, Riverside, CA

Brown, 1995

Schnoor, 1997

Phytotech

Phytotech

Rutgers University

Reference


203

90% of the soil from 10 to less than 1 mg/kg with dilute hydrochloric acid (pH 3) (Urlings, 1990). More demonstrations are needed in this area, in addition to developing more understanding into the mechanisms for solution, metal recovery and use of non-toxic additives.

Phytovolatilization of re®nery wastes and agricultural soils San Francisco, CA

Pennsylvania

Brassica sp.

Uptake of Zn and Cd rapid but soil dif®cult to decontaminate Se is partly taken-up and volatilized but soil dif®cult to decontaminate Thlaspi spp.

Sun¯owers and mustard Rocky Flats, CO

Phytoextraction of 200 £ 300 ft brown®eld Rhizo®ltration of land®ll leachate Phytoextraction of mine wastes Trenton, NJ

Indian mustard Brassica juncea

Rhizo®ltration of energy wastes Ashatabula, OH

Sun¯owers Helieanthus anuus

90% reduction 137Cs, 90Sr in 2 weeks, 8000 X normal concentration in roots 95% removal of U in 24 h (350 ppb to ,5 ppb) In one season, reached below action level In progress Rhizo®ltration near nuclear disaster Chernobyl, Ukraine

Sun¯owers Helieanthus anuus

Application Location

Table 2 Phytoremediation demonstration projects (Schnoor, 1997)

Plants

Performance

4.10. Soil washing (chemical leaching) Heavy metals can be removed from soils using various agents added to the soil (Fig. 5). This can be done in reactors or as heap leaching. These agents are: inorganic acids such as sulfuric and hydrochloric acids with pH less than 2, organic acids including acetic and citric acids (pH not less than 4), chelating agents such as ethylenediaminetetraacetic acid (EDTA) and nitrilotriacetate (NTA), and various combinations of the above (USEPA, 1991). The cleaned soil can then be returned to the original site. Soils with less than 10±20% clay and organic content (i.e. sandy soils) are most effectively remediated with these extractants. Both organics and metals are removed. However, modi®cations to the process, which is commercially used, have to be made for each type of soil (Hinsenveld et al., 1990). In general, soils with low contents of cyanide, ¯uoride and sul®de, CEC of 50±100 meq/kg and particle sizes of 0.25±2 mm, with contaminant solubility in water of greater than 1000 mg/l, can be most effectively cleaned by soil washing (Hazardous Waste Consultant, 1996). In the EPA VISITT 2.0, there are more than 20 soil washing vendors and 5 vendors of acid extraction. The feasibility of using biodegradable biosurfactants to remove heavy metals from an oilcontaminated soil was recently demonstrated (Fig. 6) by laboratory scale batch washes with surfactin, a rhamnolipid and a sophorolipid (Mulligan et al., 1999a). The ®rst two agents are produced by bacteria while the last is produced by a yeast. The soil contained 890 mg/kg of zinc, 420 mg/kg of copper with a 12.6% oil and grease content. A series of ®ve batch washes removed 70% of the copper with 0.1% surfactin/1% NaOH while 4% sophorolipid/0.7% HCl was able to remove 100% of the zinc. The results clearly indicated the feasibility of removing metals with the anionic biosurfactants tested even though the exchangeable metal fractions were very low.

204


Contaminated soil

Surfactant injection

Well

Groundwater level

Low permeability zone Contaminant plume Fig. 5. Diagram of soil ¯ushing process using injection of water or solution containing chemicals including acids, chelating agents or surfactants.

These biosurfactants were also able to remove metals from sediments (Mulligan et al., 1999c). Since these agents are biodegradable, can enhance hydrocarbon removal and can potentially be produced in situ, they have a great potential for soil washing and soil ¯ushing applications. 4.11. Treatment of sediments Since sediments contain large quantities of water, de-watering is frequently necessary after dredging to enable treatment. Methods include draining of the water in lagoons with or without coagulants and ¯occulants or using presses or centrifuges. Treatment methods are similar to those used for soil such as

5. Conclusions A summary of the various remediation techniques is shown in Table 3. Physical containment is the least expensive approach but this leaves the contaminants in place without treatment. Since metals are considered relatively immobile, methods for metal decontamination have focused on solid-phase processes such as solidi®cation/stabilization and vitri®cation. These

B 100 Soph

80

Surf

60

Rham

40

Control1

20

Control2

0 0

1

2

3

4

Number of washings

5

Total zinc removal (%)

Total copper removal (%)

A

hydrocyclone pretreatment, solidi®cation/stabilization and chemical extraction (acids). Few of these techniques in comparison to soil treatment have been used commercially.

100 Soph

80

Surf

60

Rham

40

Control1

20

Control2

0 0

1

2

3

4

5

Number of washings

Fig. 6. Cumulative copper (A) and zinc (B) removal after ®ve washes using different biosurfactants and controls. Experiments were performed as batch washes in 50 ml centrifuge tubes according to Mulligan et al. (1999a). Control 1, 1%; surf, 0.1% surfactin/1% NaOH; rham, 0.1% rhamnolipid/1% NaOH; control 2, 0.7% HCl; soph, 4% sophorolipid/0.7% HCl.


205

Table 3 Summary of remedial technologies Technology Containment Physical Encapsulation Vitri®cation Ex situ treatment Physical separation Soil washing Pyrometallurgical

In situ Reactive barriers Soil ¯ushing Electrokinetic Phytoremediation

Description

Applicability

Costs ($US/ton)

Prevent movement by preventing ¯uid ¯ow Creation of an inert waste

Land®ll covers and slurry walls

10±90

Injection of solidifying chemicals Shallow metal-contaminated soil, low volatility metals

60±290

Application of electrical energy to vitrify contaminant Includes, froth ¯otation, gravity separation, screening, etc. Addition of surfactants and other additives to solubilize Elevated temperature extraction and processing for metal removal Creation of a permeable barrier Water ¯ushing to leach contaminants Application of electrical current Use of plants for metal extraction

processes can be performed in situ which reduces handling costs. Costs depend on presence of debris, excess moisture, contaminant dept, and soil homogeneity. They are bene®cial where the area of contamination is shallow but large. Long-term stability of the solidi®ed/stabilized matrix is the major unknown. Vitri®cation is expensive but applicable to mixed wastes where few technologies are available. Electrokinetics and in situ ¯ushing have been used at a few sites but results are promising. More ®eld demonstrations are needed for both of these technologies. Electrokinetics is particularly promising for contamination at moderate depths in clays but R&D is required to optimize pore ¯uids and electrode con®guration. In situ ¯ushing is most effective for homogeneous, permeable, sandy and silty soils. Site hydrology must be understood to avoid the movement of contaminants into undesirable areas. More developments are needed in the area of non-toxic additives for in situ ¯ushing. Economics also could be improved

400±870

For high metal concentrations

60±245

For water soluble contaminants

25±300

Highly-contaminated soils (5±20%)

200±1000

Sorption or degradation of contaminants in barrier For soluble contaminants

60±245

Applicable for saturated soils with low groundwater ¯ow Shallow soils and water

100±200 Little info Good (50,000±200,000/acre)

by recovery of metals. Phytoremediation and bioleaching are not as well developed but could be useful for areas of low contamination although longer treatment times may be necessary. Numerous issues still need to be resolved and more demonstrations are required. Some of the issues include enhancing the accumulation of metals by plants, developing methods to extract meals from plants and determining correlations between soil components and bioavailability. Treatment walls are low cost, passive treatment methods that have potential. More research is required to match reactive media with contaminants, model lifetime performance, optimize retention times and develop methods for regeneration of reactive media. The selection of each technology is site-speci®c. Overall, metal-contaminated groundwater is treated above-ground. Metalcontaminated soils are excavated and are treated ex situ by solidi®cation/stabilization. In situ techniques are under development. Our research has

206


indicated that biosurfactants, biologically produced surfactants, may be viable agents for the enhanced removal of metals from contaminated soils and sediments. References Acar, Y.B., Gale, R.J., 1995. Electrokinetic remediation: basics and technology status. J. Hazard. Mater. 40, 117±137. American Academy of Environmental Engineers (AAEE), 1993. Soil Washing/Soil Flushing, Innovative Site Remediation. Anderson, W.C. (Ed.), vol. 3, WASTEC. Agency for Toxic Substances and Disease Registry, 1995. Division of Toxicology, Atlanta. Baker, A.J.M., Reeves, R.D., McGrath, S.P., 1991. In situ decontamination of heavy metal polluted soils using crops of metalaccumulating plants: A feasibility study. In: Hinchey, R.E., Offenbach, R.F., (Eds.), In Situ Bioreclamation. ButterworthHeinemann, Boston, pp. 601±605. Blais, J.F., Tyagi, R.D., Auclair, J.C., 1992. Bioleaching of metals from sewage sludge by sulfur-oxidizing bacteria. J. Environ. Engng 118/5, 690±707. Bolenz, S., Omran, H., Gierschner, K., 1990. Treatments of water hyancinth tissue to obtain useful products. Biol. Wastes 22, 263±274. Bourg, A.C.M., 1995. Speciation of Heavy Metals and Implications for their Mobility, Heavy Metals. Springer, Berlin, pp. 19±32. Brown, K.S., 1995. The green clean: the emerging ®eld of phytoremediation takes root. Bioscience 45, 579±582. Cameron, R.E., 1992. Guide to Site and Soil Description for Hazardous Waste Site Characterization. Volume 1: Metals. Environmental Protection Agency EPA/600/4-91/029. Chlopecka, A., 1992. Forms of trace metals from inorganic sources in soils and amounts found in spring barley. Water, Air, Soil Pollut. 66. Conner, J.R., 1990. Chemical Fixation and Solidi®cation of Hazardous Wastes. Van Nostrand Reinhold, New York. Evanko, C.R., Dzombak, D.A., 1997. Remediation of metalscontaminated soils and groundwater. Technology Evaluation Report, TE-97-01. Gound-Water Remediation Technologies Analysis Center, Pittsburgh, PA. Fassett, D.W., 1980. Cadmium. In: Waldron, H.A. (Ed.), Metals in the Environment. Academic Press, New York, pp. 61±110. Gombert II, D., 1994. Treatability testing to evaluate what can work. Nucl. Technol. 108, 90±99. Hazardous Waste Consultant, 1996. Remediating Soil and Sediment Contaminated with Heavy Metals, Nov/Dec. Elsevier Science, Netherlands. Hinsenveld, M., Soczo, E.R., van de Leur, G.J., Versluijs, C.W., Groenedijk, E., 1990. Alternative physico-chemical and thermal cleaning technologies for contaminated soil. In: Arendt, F., Hinsenveld, M., van der Brink (Eds.), Proceedings of the International Conference on Contaminated Soils '90. Kluwer Academic, Dordrecht, pp. 873±881. Jasperse, B.H., Ryan, C.R., 1992. Stabilization and Fixation Using

Soil Mixing, Proceedings of the ASCE Specialty Conference on Grouting, Soil Improvement and Geosynthetics. ASCE, Reston, VA. Jawarsky, J., 1978. Effects of Lead in the Environment I & II. National Research Council, Canada. Kabata-Pendias, A., 1992. Trace metals in soils in Poland Ð occurrence and behaviour. Trace Subst. Environ. Health 25, 53±70. Karavaiko, G.I., Rossi, G., Agates, A.D., Groudev, S.N., Avakyan, Z.A., 1988. Biogeotechnology of Metals: Manual. Center for International Projects GKNT, Moscow, Soviet Union. Lake, D.L., 1987. Chemical speciation of heavy metals in sewage sludge and related matrices. In: Lester, J.N. (Ed.), Heavy Metals in Wastewater and Sludge Treatment Processes. Volume I: Sources, Analysis and Legislation. CRC Press, Boca Raton, pp. 125±153. Lakanen, E., Ervio, R., 1971. Acta Agral. Fenn. 123, 233±238. Li, W., Peters, R.W., Brewster, M.D., Miller, G.A., 1995. Sequential extraction evaluation of heavy-metal-contaminated soil: how clean is clean? Proceedings of the Air and Waste Management Association, 88th Annual Meeting and Exhibition, San Antonio, Texas, 18±23 June. Lum, K.R., Edgar, D.C., 1983. Analyst 108, 918. Matthews, P.J., 1984. Control of metal application rates from sewage sludge utilization in agriculture. Crit. Rev. Environ. Control 14, 199±230. Mulligan, C.N., Yong, R.N., Gibbs, B.F., 1999a. On the use of biosurfactants for the removal of heavy metals from oilcontaminated soil. Environ. Prog. 18 (1), 50±54. Mulligan, C.N., Galvez-Cloutier, R., Renaud, N., 1999b. Biological leaching of copper mine residues by Aspergillus niger. Presented at AMERICANA 1999, Pan-American Environmental Trade Show and Conference, Montreal, Canada, 24±26 March 1999. Mulligan, C.N., Yong, R.N., Gibbs, B.F., 1999c. Removal of heavy metals from contaminated soil and sediments using the biosurfactant surfactin. J. Soil Contam. 8, 231±254. National Pollutant Release Inventory (NPRI), 1995. Environment Canada. NATO/CCMS, 1999. NATO/CCMS Pilot Study, Evaluation of Demonstrated and Emerging Technologies for the Treatment of Contaminated Land and Groundwater (Phase III). 1999 Annual Report, Number 235, EPA/542/R-99/007. Nriagu, J.O., Pacyna, J.M., 1988. Nature (London) 333, 134±139. Patterson, J.W., 1985. Industrial Wastewater Treatment Technology, 2nd ed. Butterworth, Boston. Petrozelli, G., Giudi, G., Lubrano, L., 1983. Proceedings of the International Conference on Heavy Metals in the Environment, 475. Pickering, W.F., 1986. Metal ion speciation Ð soils and sediments (a review). Ore Geol. Rev. 1, 83±146. Ramos, L., Hernandez, L.M., Gonzalez, M.J., 1994. Sequential fractionation of copper, lead, cadmium and zinc in soils from or near Donana National Park. J. Environ. Qual. 23, 50±57. Ravishankar, B.R., Auclair, J.-C., Tyagi, R.D., 1994. Partitioning of heavy metals in some Quebec municipal sludges. Water Pollut. Res. J. Can 29, 457±470. Robins, R.G., 1992. Arsenic chemistry in relation to the

C.N. Mulligan et al. / Engineering Geology 60 (2001) 193±207 disposal and stability of metallurgical wastes. Arsenic and Mercury-Workshop on Removal, Recovery, Treatment and Disposal, Alexandria, Virginia, August 17±20, pp. 4±7, EPA/600/R-92/105. Of®ce of Research and Development, Washington, D.C. Rodsand, T., Acar, Y.B., 1995. Electrokinetic extraction of lead from spiked Norwegian marine clay. Geoenvironment 2000 2, 1518±1534. Rumer, R.R., Ryan, M.E., 1995. Barrier Containment Technologies for Environmental Remediation Applications. Wiley, New York. Schnoor, J.L., 1997. Phytoremediation. Technology Evaluation Report TE-98-01, GWRTAC, Pittsburg, PA. Shuman, L.M., 1985. Fractionation method for soil microelements. Soil Sci. 140, 11±22. Smith, L.A., Means, J.L., Chen, A., Alleman, B., Chapma, C.C., Tixier Jr., J.S., Brauning, S.E., Gavaskar, A.R., Royer, M.D., 1995. Remedial Option for Metals-Contaminated Sites. Lewis, Boca Raton, FL. Tessier, A., Campbell, P.G.C., Bisson, M., 1979. Sequential extraction procedure for the speciation of particulate trace metals. Anal. Chem. 51, 844±851. Thompson-Eagle, E.T., Frankenberger, Jr., W.T., 1990. Proteinmediated selenium biomethylation in evaporation pond water. Environ. Toxicol. Chem. 9, 1453±1560. Tichy, R., Grotenhuis, J.T.C., Rulkens, W.H., 1992. Bioleaching of zinc-contaminated soil with thiobacilli. Proc. Int. Conf. Eurosol. 92, September 1992, Masstricht, The Netherlands. Urlings, L.G.C.M., 1990. In situ cadmium removal Ð full-scale

207

remedial action of contaminated soil. International Symposium on Hazardous Waste Treatment: Treatment of Contaminated Soil, Air & Waste Association and U.S. EPA Risk Education Laboratory, Cincinatti, Ohio, February 5±8 1990. USEPA, 1987. Teatability Studies Under CERCLA: An Overview. OSWER Directive 9380.3-02FS. USEPA, 1991. Innovative Treatment Technologies. Semi-annual status report (third edition) EPA/540/2-91/001, U.S. EPA Of®ce of Solid Waste and Emergency Response, Washington, DC. USEPA, 1994. Selection of Control Technologies for Remediation of Soil Contaminated with Arsenic, Cadmium, Chromium, Lead or Mercury. Revised Draft Engineering Bulletin, January 31. USEPA, 1996. Superfund Fact Sheet. United Chrome Products Inc., Corvallis, Oregon, June 10. USEPA, 1996. Engineering Bulletin: Technology Alternatives for the Remediation of Soils Contaminated with Arsenic, Cadmium, Chromium, Mercury and Lead. U.S. Environmental Protection Agency, Of®ce of Emergency and Remedial Response, Cincinnati, OH. Vidac, R.D., Pohland, F.G., 1996. Treatment Walls. Technology Evaluation Report TE-96-01, Groundwater Remediation Technologies Analysis Center, Pittsburgh, PA. Wheeler, P., 1995. Leach repellent. Ground Engng 28, 20±22. World Resources Institute, 1992. World Resources 1992/93. Oxford University Press, New York. Yong, R.N., Galvez-Cloutier, R., Phadingchewit, Y., 1993. Selective extraction analysis of heavy-metal retention in soil. Can. J. Geotech. 30, 834±847.