ment 364 (2006) 1423www.elsevier.com/locate/scitotenv

Science of the Total Environ

Comparative evaluation of short-term leach tests for heavy metalrelease from mineral processing waste

Souhail R. Al-Abed a,, Philip L. Hageman b, G. Jegadeesan c,Narain Madhavan d, Derrick Allen a

a National Risk Management Research Laboratory, U.S. Environmental Protection Agency, 26W.Martin Luther King Dr. Cincinnati, OH 45268, USAb Crustal Imaging and Characterization Team (CICT), U. S. Geological Survey (USGS), Bldg. 20, Box 25046 MS 973 DFC, Denver,

CO 80225-0046, USAc Department of Civil and Environmental Engineering, University of Cincinnati, Cincinnati, OH 45221, USA

d GE India Technology Centre Pvt. Ltd., Plot #122, Export Promotion Industrial Park, Phase 2, Hoodi Village, Whitefield Road,Bangalore-560 066, India

Received 29 June 2005; received in revised form 31 October 2005; accepted 31 October 2005Available online 5 December 2005

Abstract

Evaluation of metal leaching using a single leach test such as the Toxicity Characteristic Leaching Procedure (TCLP) is oftenquestionable. The pH, redox potential (Eh), particle size and contact time are critical variables in controlling metal stability, notaccounted for in the TCLP. This paper compares the leaching behavior of metals in mineral processing waste via short-termextraction tests such as TCLP, Field Leach Test (FLT) used by USGS and deionized water extraction tests. Variation in the extractedamounts was attributed to the use of different particle sizes, extraction fluid and contact time. In the controlled pH experiments,maximum metal extraction was obtained at acidic pH for cationic heavy metals such as Cu, Pb and Zn, while desorption of Se fromthe waste resulted in high extract concentrations in the alkaline region. Precipitation of iron, caused by a pH increase, probablyresulted in co-precipitation and immobilization of Cu, Pb and Zn in the alkaline pH region. A sequential extraction procedure wasperformed on the original waste and the solid residue from the EhpH experiments to determine the chemical speciation anddistribution of the heavy metals. In the as-received waste, Cu existed predominantly in water soluble or sulfidic phases, with nobinding to carbonates or iron oxides. Similar characteristics were observed for Pb and Zn, while Se existed mostly associated withiron oxides or sulfides. Adsorption/co-precipitation of Cu, Se and Pb on precipitated iron hydroxides was observed in theexperimental solid residues, resulting in metal immobilization above pH 7.Published by Elsevier B.V.

Keywords: Leaching; TCLP; pH; Mineral processing waste; Metals

1. Introduction

Heavy metal contamination of natural environmentsdue to hazardous waste leaching is a global problem.

Corresponding author. Tel.: +1 513 569 7849; fax: +1 513 5697879.

E-mail address: al-abed.souhail@epa.gov (S.R. Al-Abed).

0048-9697/$ - see front matter. Published by Elsevier B.V.doi:10.1016/j.scitotenv.2005.10.021

The mobility of toxic metals released from wastes isassessed using a variety of risk assessment proceduresincluding the Toxicity Characteristic Leaching Proce-dure (TCLP), among others (van der Sloot, 1996; Barnaet al., 1997; USEPA, 1997). The TCLP test, widely usedby state and national agencies, was designed to simulateleaching of heavy metals and organics from industrialwastes to be co-disposed in municipal solid landfills.

mailto:al-abed.souhail@epa.govhttp://dx.doi.org/10.1016/j.scitotenv.2005.10.021

15S.R. Al-Abed et al. / Science of the Total Environment 364 (2006) 1423

Over several years, severe limitations of using TCLP inassessing contaminant leaching from industrial wastessuch as mine and mineral processing wastes (MPW)have come to light due to its specificity only to landfillconditions and other physicochemical parameters(USEPA, 1995, 1999). The main reasons contributingtowards the inaccurate prediction of heavy metal leach-ing from MPW included: (1) choice of correct extrac-tion fluid (Hooper et al., 1998); (2) pH or alkalinity ofthe wastes (Batchelor, 1998; van der Sloot, 1996); (3)particle size (Janusa et al., 1998; Karius and Hammer,2001) and; (4) redox potential (Meng et al., 2001). Dueto variation in metal leachability with these physicalconditions in short-term experiments, standardizationof the results and procedures is critical in the riskassessment of hazardous waste disposal.

Field experiments on mine wastes suggest that metalmobility depended on the pH and redox potential, asthey influence the speciation and association of themetals in the waste matrix (Jurjovec et al., 2002; Pag-nanelli et al., 2004). While metals are immobilized inthe waste due to secondary precipitation or adsorptionreactions, changes in pH or a pH induced redox poten-tial (Eh) decrease could enhance metal mobility.

Logically, speciation analysis is of major importancein environmental research as it provides crucial evi-dence on the mineral phases and chemical form of targetmetals present in the waste matrix (Scheckel et al.,2003). As a consequence, sequential extraction proce-dures (SEP), using extraction fluids of increasingstrengths, have been used to characterize the physicaland chemical properties of the wastes. Although theresults from the SEP are often variable and should beaccompanied by in-depth investigation including theuse of EXAFS, they are still useful in providing infor-mation on contaminant partitioning within an environ-mental matrix.

Despite the emphasis on the need for a more integraland scientific approach towards the determination ofwaste leaching mechanisms, very few studies have cor-related the results from standardized batch extractiontests to characteristics of the waste. To aid in the under-standing of these mechanisms, the main objective of thispaper is to: (1) compare leaching of Cu, Pb, Zn and Sefrom MPW in different short-term batch extraction pro-cedures; (2) quantify release of the selected metals underdifferent pH and redox conditions and; (3) correlatemetal mobility to its speciation in the waste matrix.The mineral processing waste was selected due to itshigh metal content and to also evaluate its applicabilityin landfill disposal. In addition to the TCLP, short-termprocedures such as the Field Leach Test (FLT) used by

the United States Geological Survey (USGS), and deio-nized water (DI) extraction tests were performed. Here,short-term extraction test is operationally defined astest conducted within a 48-h period. This work aims toillustrate the effect of pH on the characteristics of waste,which subsequently controls metal mobility.

2. Materials and methods

2.1. Waste sampling

The mineral processing waste was obtained from anabandoned copper mine facility in Nevada, USA.Details of the sampling location have been reportedelsewhere (USEPA, 2004). After collection, the sampleswere homogenized, air dried to remove moisture andsieved. Particles with sizes less than 9.5 mm were usedfor the TCLP test, while those with sizes less than 2 mmwere used for the DI, FLT and controlled pH experi-ments. Ideally, the waste should be homogenized pref-erably to pass through 125 m sieve to mitigate particlesize effects. However, in order to compare metal leach-ing between the different extraction tests, further reduc-tion of the waste sample was avoided. Although carewas taken to perform sufficient homogenization of thesample (as observed by the low variances in the totalmetal concentration), a certain degree of variability andheterogeneity within the sample is bound to exist due tothe complex nature of the waste matrix.

2.2. Batch leaching tests

DI extraction test was performed to simulate scenar-ios where the waste is in contact with solutions having alow buffering capacity such as rainwater. The TCLP(Method 1311) was used to simulate worst-case scenar-io of co-disposal in municipal landfills, and the FLTwasperformed to simulate chemical reactions of geogenicmaterials with water. Based on the pH of MPW in DI(pH ~ 3.1), extraction fluid # (pH 4.9 0.05) was usedfor TCLP (USEPA, 1997. Sodium hydroxide (NaOH),nitric acid (HNO3) and glacial acetic acid (CH3COOH)were used to prepare the TCLP extraction fluid. For theDI and FLT (U.S. Geological Survey, 2005) experi-ments, deionized water (Millipore Systems, MA, 18M) was used as the extraction fluid. The liquid/solidratio for the above extraction tests was maintained at20:1. The duration and agitation type for the tests were:DI (48 h, rotary tumbler at 30 2 rpm); TCLP (18 h,rotary tumbler at 30 2 rpm) and; FLT (5 min, manualshaking). The extract was then filtered, acidified andstored at 4 C for total metal analysis.

16 S.R. Al-Abed et al. / Science of the Total Environment 364 (2006) 1423

2.3. Sequential extraction procedure

The procedure, previously reported in literature(Scheckel et al., 2003; Kim et al., 2003), was employedwith minor modifications (Table 1). Due to low organicmatter content, extraction of metals bound to organicmatter was not performed. 1 M MgCl2 solution wasused to determine the exchangeable (non-specificallyadsorbed) or water soluble phase (Step A), 1 M sodiumacetate (CH3COONa) (Step B) for extracting metalsbound to the carbonate phase and 0.04 M hydroxyl-amine hydrochloride (NH2OHHCl) in 25% acetic acid(Step C) for metals associated with reducible FeMnoxides. The metals associated as sulfides are usuallyextracted using a mixture of H2O2 and HNO3 withresidual silicates extracted in the acid digestion step(Kim et al., 2003). However, in the case of MPW theresidual metal content, assumed to exist predominantlyas sulfides, was determined by the acid digestion tech-nique (EPA Method 3051). Between each extractionstep, the samples were rinsed with 810 ml of deionizedwater and the supernatant was collected and analy