Author's personal copy - University of Virginia OF THE TOTAL ENVIRONMENT 391 (2008) 13 25 ... kinetics (Geelhoed et al., 2002) ... Author's personal copy

This article was published in an Elsevier journal. The attached copyis furnished to the author for non-commercial research and

education use, including for instruction at the author’s institution,sharing with colleagues and providing to institution administration.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

http://www.elsevier.com/copyright

Author's personal copy

Mobilization of Cr(VI) from chromite ore processing residuethrough acid treatment

James M. Tinjuma,⁎, Craig H. Bensonb, Tuncer B. Edilb

aCH2M HILL, 1717 Arch Street, Suite 4400, Philadelphia, PA 19103, United StatesbDepartment of Civil and Environmental Engineering, University of Wisconsin–Madison, 1415 Engineering Drive, Madison,Wisconsin 53706, United States

A R T I C L E I N F O A B S T R A C T

Article history:Received 8 June 2006Received in revised form2 October 2007Accepted 9 October 2007Available online 11 December 2007

Batch leaching studies on chromite ore processing residue (COPR) were performed usingacids to investigate leaching of hexavalent chromium, Cr(VI), with respect to particle size,reaction time, and type of acid (HNO3 and H2SO4). Aqueous Cr(VI) is maximized atapproximately 0.04 mol Cr(VI) per kg of dry COPR at pH 7.6–8.1. Cr(VI) mobilized more slowlyfor larger particles, and the pH increased with time and increased more rapidly for smallerparticles, suggesting that rate limitations occur in the solid phase. With H2SO4, the pHstabilized at a higher value (8.8 for H2SO4 vs. 8.0 for HNO3) and more rapidly (16 h vs. 30 h),and the differences in pH for different particle sizes were smaller. The acid neutralizationcapacity (ANC) of COPR is very large (8 mol HNO3 per kg of dry COPR for a stable eluate pH of7.5). Changes to the elemental and mineralogical composition and distribution in COPRparticles after mixing with acid indicate that Cr(VI)-bearing solids dissolved. However,concentrations of Cr(VI) N2800 mg kg−1 (N50% of the pre-treatment concentration) were stillfound after mixing with acid, regardless of the particle size, reaction time, or type of acidused. The residual Cr(VI) appears to be partially associated with poorly-ordered Fe and Aloxyhydroxides that precipitated in the interstitial areas of COPR particles. Remediationstrategies that use HNO3 or H2SO4 to neutralize COPR or to maximize Cr(VI) in solution arelikely to require extensive amounts of acid, may not mobilize all of the Cr(VI), and mayrequire extended contact time, even under well-mixed conditions.

© 2007 Elsevier B.V. All rights reserved.

Keywords:Chromite oreProcessing residueChromiumAcid treatmentLeaching

1. Introduction

Historical processing of chromite ore involved mixing dry-milled ore with lime (CaO) and soda ash (Na2CO3), followed byroasting at high temperatures to promote oxidation of thetrivalent chromium, Cr(III), within the ore. Hexavalent chro-mium, Cr(VI), was then recovered as soluble sodium chromate(Na2CrO4). The waste from this process, referred to aschromite ore processing residue (COPR), is a granular materialwith 2 to 7% Cr by oxide weight (Raghu and Hsieh, 1989; Burkeet al., 1991; Tinjum, 2006). Remnant Cr exists in COPR because

of incomplete oxidation of the ore, incomplete leaching ofchromate, and generation of Cr-bearing minerals such asbrownmillerite during processing (Public Health Service, 1953;Hillier et al., 2003). Production of chromates in the US createdmillions of tonnes of COPR, most of which was disposed inHudson County, New Jersey (2.5 Mt) and the Baltimore area(2 Mt) in what are now commercial districts, transportationfacilities, harbors, and residential areas (MEMT, 1990; Burkeet al., 1991; Lioy et al., 1992). Cr(VI), a carcinogen and adermatological and pulmonary sensitizer (USEPA, 1998; Klaas-sen, 2001), is leached from COPR over long periods of time

S C I E N C E O F T H E T O T A L E N V I R O N M E N T 3 9 1 ( 2 0 0 8 ) 1 3 – 2 5

⁎ Corresponding author. Tel.: +1 215 563 4220; fax: +1 215 640 9275.E-mail address: [email protected] (J.M. Tinjum).

0048-9697/$ – see front matter © 2007 Elsevier B.V. All rights reserved.doi:10.1016/j.scitotenv.2007.10.041

ava i l ab l e a t www.sc i enced i rec t . com

www.e l sev i e r. com/ loca te / sc i to tenv


(Higgins et al., 1998; Geelhoed et al., 1999;Weng et al., 1994 and2002). Thus, remediation of sites contaminatedwith COPR is ofsignificant interest (Interfaith, 2003; Truini, 2005).

COPR particles contain heterogeneous assemblages ofunreacted chromite ore, high temperature minerals formedduring roasting (e.g., brownmillerite, periclase, portlandite,larnite), minerals formed by ambient weathering (e.g., brucite,calcite, hydrocalumite, hydrogarnet), and cementing agentssuch as Ca3Al2O6 (Hillier et al., 2003; Chrysochoouet al., 2005;Tinjum, 2006). COPR particles are cemented conglomerationsof smaller particles and have significant micro-porosity(Tinjum, 2006). COPR is highly alkaline (8.1–12.3 paste pH)and contains a large amount of residual Cr (2000–40,000mg kg−1),ofwhich asmuch as 35%may beCr(VI) (Weng et al., 1994; Higginset al., 1998; Meegoda et al., 1999; Thomas et al., 2001; Tinjum,2006). The Cr is distributed throughout the particles in trivalentand hexavalent forms rather than just a surface contaminant(Hillier et al., 2003; Tinjum, 2006). These characteristics result inpersistent leaching of Cr(VI) from COPR and make mitigationdifficult.

Efforts to plan remedial measures, remediate COPRimpacts, and evaluate in-place remedial efforts at COPR sitesin the mid-Atlantic region of the US are ongoing (Henry et al.,2007; Tinjum, 2006; Truini, 2005). Numerous reducing agentshave been proposed to reduce the Cr(VI) in COPR to lessmobileand less toxic Cr(III), including ferrous iron (Eary and Rai, 1988;Thornton, 1995; Chowdhury, 2003; Geelhoed et al., 2003;Wazne et al., 2005); ascorbic acid (James, 1999); nitrate, lacticacid, steel wool, or hardwood tree leaf litter (James, 1994);sulfide ions (Gancy and Wamser, 1976); organic matter(Higgins, 1996; Higgins et al., 1998); calcium polysulfide(Graham et al., 2006; Wazne et al., 2005; Tinjum, 2006); and acombination of FeSO4 and sodiumdithionite (Na2S2O4) (Su andLudwig, 2005). Many of these approaches are optimized at pHsconsiderably lower than the in situ pH of COPR. The rate of Fe(II) oxidation by oxygen (O2) increases rapidly with pHincreases (Geelhoed et al., 2003; Millero, 1985). The optimumpH range for reduction of Cr(VI) by organic matter is 6.5–9.5(Higgins et al., 1998) and, for use of ferrous compounds,pHb9.5 is optimal to minimize the oxidation of O2 by ferrouscompounds (James, 1994; Geelhoed et al., 2003). Cr(VI) is moremobile (more soluble and less likely to adsorb to solid phases)under neutral to slightly alkaline conditions (Palmer andWittbrodt, 1991) and thus more available for reduction-typereactions in the liquid phase. Lowering the pH can create anenvironment where conversion of Cr(VI) to Cr(III) readilyoccurs in the presence of reductants such as ferrous iron,the amount of Cr(VI) adsorbed onto metal oxyhydroxidesurfaces is reduced, and the solubility of resulting Cr(III)hydroxides is low (Palmer and Wittbrodt, 1991; Richard andBourg, 1991; Shupack, 1991; James, 1994; Geelhoed et al., 1999and 2002; Weng et al., 1994).

The efficient reduction and stabilization of Cr(VI)-contami-natedmaterial often requires pH adjustment. For example, thepH at COPR sites may require pH adjustment from ≈12 to anear-neutral or slightly alkaline pH. pH plays a significant rolein desorption/adsorption, dissolution/precipitation, and redoxmechanisms in COPR. The solubilities of Cr-bearing mineralsare pH-dependent and many Cr-bearing minerals in COPR arestable at alkaline pHs (Geelhoed et al., 2002). The amount of Cr

(VI) in solution is also pH-dependent (Palmer and Wittbrodt,1991). Maximum aqueous Cr(VI) concentrations were reportedat pH≈8 for Glasgow-based COPR (Geelhoed et al., 2002).Maximizing the soluble Cr(VI) content may lead to moreefficient groundwater pump and treat systems. In addition,reduction of Cr(VI) is contingent on its geometric availability(Bartlett and James, 1988) and high concentrations of aqueousCr(VI) allow for efficient use of chemical reductants. Loweringthe pH below the point of zero point charge (pHZPC) of COPR,however, may cause aqueous Cr(VI) to adsorb to positivelycharged metal oxyhydroxides (Zachara et al., 1987). Cr(VI) inthe aqueous phase is rapidly reduced to Cr(III) in the presenceof reductants such as S2− and Fe2+ ions (Eary and Rai, 1988;Saleh et al., 1989), and Cr(III) can be removed to b10−6 M at pHfrom 6 to 11 via precipitation (Sass and Rai, 1987; Eary and Rai,1988; Weng et al., 1994). Accordingly, remediation of COPR bychemical reduction and precipitation methods often set atarget pH of 7–9 (Palmer and Wittbrodt, 1991; James, 1999;Chowdhury, 2003).

Because Cr is distributed throughout COPR particles(Tinjum, 2006), methods to treat COPR need to fully penetratethe particles to ensure complete treatment. Cr remaining inthe interiormay continue to leach due to very slow dissolutionkinetics (Geelhoed et al., 2002) and rate limitations imposed bytransport of Cr(VI) from the interior to the surface of particles(Higgins et al., 1998; Weng et al., 1994). Consequently, contacttime is expected to be an important factor influencing theeffectiveness of chemical treatments of COPR (Tinjum, 2006).

The objectives of this study were to determine the impactsof acid additions on COPR from the eastern coast of the US.The quantity of acid required to lower the pH of highly alkalineCOPR, expressed as acid neutralization capacity (ANC), wasinvestigated in addition to the eluate pH at which themaximum mass of Cr(VI) is dissolved/desorbed into theaqueous phase. Mixing time, particle size, and type of acidwere explored, and the effectiveness and efficiency at whichacid removes Cr(VI) from the solid phase were determined.The mineralogical and chemical composition and spatialdistribution of elements in acid-leached COPR were investi-gated to determine whatminerals were dissolved/precipitatedand the amount and spatial distribution of remnant Cr.Implications of these findings for remediation based on acidtreatment are discussed.

2. Materials and methods

2.1. Materials

2.1.1. COPRCOPR from a site on the mid-Atlantic coast of the US was usedin this study. COPR was obtained during a field investigationfrom a granular stratum of COPR located above the water tableand approximately 2 m below ground surface. The COPRclassifies as poorly-graded sand (SP) according to the UnifiedSoil Classification System (ASTM D 2487), with the largerparticles being conglomerates of smaller particles. The as-received water content (ASTM D 2216) ranged from 19.8 to21.2%, the specific gravity (ASTM D 854) ranged from 2.74 to2.99 (average=2.88), and the initial paste pH of the COPR

14 S C I E N C E O F T H E T O T A L E N V I R O N M E N T 3 9 1 ( 2 0 0 8 ) 1 3 – 2 5


(determined by ASTM D 4972) ranged between 11.2 and 12.2,(average=11.7).

The elemental composition of size-fractionated COPR usedin this study was determined by X-ray fluorescence (XRF) (seeSection 2.2.3 for method). Five elements (Ca, Fe, Al, Mg, andCr), expressed by oxide weight, contribute N90% of theinorganic mass, and their proportions vary minimally withparticle size (average of 35.3% CaO, 21.1% Fe2O3, 11.6% Al2O3,8.9% MgO, 4.1% Cr2O3). The high percentage of Ca in the COPRindicates that a “high lime” manufacturing process was used(Darrie, 2001). The solid-phase concentration of Cr(VI) in theCOPR, determined by alkaline digestion (USEPAMethod 3060A)followed by colorimetric analysis of the digest (USEPA Method7196A), ranges from 4900 to 7700 mg kg−1 (Table 1), whichrepresents about 25% of the total chromium component. ThisCr(VI) percentage is within the range (1% to 35%) reported byothers (Burke et al., 1991; Geelhoed et al., 2002; Meegoda et al.,1999). The fine-sized COPR (0.075 to 0.425 mm) has the least Cr(VI) (4,900 mg kg−1) and the coarse-sized COPR (2.0 to 4.8 mm)has the most Cr(VI) (7700 mg kg−1), with limited variabilityobserved in the replicates of each size fraction.

A qualitative Cr X-raymap of a 400 μm×500 μmregion of thecross-section of a COPR particle is presented in Fig. 1a. The darkregions (absence of characteristic X-rays) correspond to micro-poreswithin theparticle. Thebrightest regions in theCrmaparerelict chromite grains (confirmed by wavelength-dispersivespectrometry) that were not converted during roasting. Cr isdistributed at a low intensity throughout the particle cross-section, and thus is not only a surface contaminant. Elevated Crconcentrations occur in rims surrounding sub-particle features,

as shown in the Cr map. The 2000 μm×1900 μm X-ray mapspresented in Fig. 1b (Ca) and 1c (Fe) show that the mostabundant elements in COPR are also distributed throughout thecross-section of a typical COPR particle.

2.1.2. ReagentsTwo common acids (H2SO4 and HNO3) were used to study theimpacts of acid on the leaching of metals from COPR. H2SO4

was used to assess competition between sulfate (SO42−) and

chromate (CrO42−) for adsorption sites in COPR. SO4

2− and CrO42−

have equivalent charge, similar structure, and comparablethermochemical radii (Baron et al., 1996) and ion exchange isdue to a combination of competitive and electrostatic effects(Zachara et al., 1987). H2SO4 has also been proposed for thetreatment of COPR (Chowdhury, 2003; Higgins et al., 1998;Shupack, 1991; Wazne et al., 2005), partly because of thepotential to generate Fe(II) and H2SO4 on site by the controlledoxidation of iron pyrite at reasonable cost (Chowdhury, 2003).HNO3 was used because, unlike the SO4

2− from H2SO4, NO3− is

not as likely to form secondary precipitates with potentiallymobilized cations (Al, Ca, Cr, and Mg) from COPR. As anexample, the formation of gypsum (CaSO4·2H2O) is thermo-dynamically favorable with high aqueous concentrations ofSO4

2− and Ca2+ (Geelhoed et al., 2002; Tinjum, 2006). NO3− may

provide some competition with CrO42−, but not as extensive as

with SO42− because of the valence difference (NO3

−vs. CrO42−)

(Richard and Bourg, 1991; Zachara et al., 1988).

2.2. Experimental methods

2.2.1. Batch testsTwo series of batch tests were conducted. One seriesevaluated the pH-dependent release of Cr(VI) and othermetalsand the COPR-specific acid neutralization capacity (ANC), withHNO3 used as the reacting acid. The second series of batchtests evaluated the release of Cr(VI) and rebound of pH withrespect to batch time (1 to 72 h), particle size (fine-, medium-,and coarse-sized particles), and type of acid used (HNO3 andH2SO4). Each batch test was conducted in duplicate and eachdata point is reported in the graphical results. COPR used inthe batch tests was air-dried for 48 h at a temperature of 20 °Cand homogenized by coning and sectioning. Air-dry speci-mens of COPR were used to allow for size fractionation and tonormalize themoisture content of the pre-test specimens. Air-drying does not significantly alter the pre-test mineralogy ofCOPR because hydration of brownmillerite, the majoritycrystalline phase in COPR, is minimized in the absence ofmoisture (Dermatas et al., 2007). Carbonated phase content ofCOPR such as calcium aluminum carbonate hydrates [Ca4Al2(OH)12(CO3)·5(H2O) and Ca8Al4(OH)14CO2·24(H2O)] and calcite(CaCO3) may be slightly impacted, but these minerals aretypically trace constituents of COPR (Tinjum, 2006; Chryso-choou et al., 2005; Chrysochoou and Dermatas, 2006). COPRcontains a substantial amount of alkaline buffer, thus air-drying is not expected to significantly alter the pH of thematerial.

For the first set of tests, sub-2.0-mm-sized COPR particles(40 g) were contacted with HNO3 at a liquid–solid ratio (LS) of10:1 in an end-over-end rotator for 48 h. Thirty-six pH endpoints were evaluated, each in duplicate. An acid titration

Table 1 – Specific gravity and Cr(VI) concentration ofCOPR (untreated and acid-treated)

Size fraction Untreateda HNO3-treatedb

H2SO4-treatedc

Specific gravityFine-sized COPR(0.075 to 0.425 mm)

2.82 3.12 2.98

Medium-sized COPR(0.425 to 2.0 mm)

2.90 3.04 2.95

Coarse-sized COPR(2.0 to 4.8 mm)

2.77 2.95 2.92

Averaged 2.83 3.04 2.95

Cr(VI) Concentration(mg kg−1)Fine-sized COPR(0.075 to 0.425 mm)

4900 3400 2800

Medium-sized COPR(0.425 to 2.0 mm)

5800 2900 3500

Coarse-sized COPR(2.0 to 4.8 mm)

7700 3300 4000

Averaged 6100 3200 3400

a Average of the replicate for each size fraction.b Average of two specimens for each size fraction (reaction times of1, 8, 16, and 64 h).c Average of two specimens for each size fraction (reaction times of1 and 16 h).d Average of specific gravity and Cr(VI) concentrations are themeans for the different size fractions; i.e., the mean of the replicateaverages of the three size fractions.

15S C I E N C E O F T H E T O T A L E N V I R O N M E N T 3 9 1 ( 2 0 0 8 ) 1 3 – 2 5


curve using HNO3 was also developed by sequential additions(100 to 250 μL) of 2N HNO3, followed by a 20-min reactionperiod with end-over-end mixing, a 5-min settling period, andmeasurement of the eluate pH, as described in Kosson et al.(2002). The 20-min titration curve was used to compare theamount of acid uptake in COPR for a quick, 20-min mixingtime versus a 48-h batch time. The second series of batch testsused dry-sieved COPR consisting of fine-sized (0.18 to0.30 mm), medium-sized (0.42 to 0.85 mm), and coarse-sized

(2.00 to 2.36mm) particles. Aliquots of air-dry COPR (10 g) werereacted with 100 mL of 0.5 N acid (HNO3 or H2SO4) (LS=10:1) bytumbling end-over-end for 1, 4, 8, 16, 32, and 64 or 72 h at30 rpm.

2.2.2. Chemical analysesEluates from the batch experiments were analyzed for pH andconductivity within 5 min of removal from the rotator with anAccumet®Model 50 pH/ion/conductivitymeter, filtered with a

Fig. 1 – Distribution of (a) Cr, (b) Ca, and (c) Fe in particles of COPR obtained with X-ray mapping. Lighter colors indicate higherX-ray intensity (i.e., higher quantities of element relative to areas with lower X-ray intensity). A different particle is shown in(b) and (c) than in (a).



0.45-μm filter, acidified to pHb2 with HCl, and refrigerated at4 °C. Cr(VI) concentrations were determined colorimetricallyby reaction of an aliquot of the eluate with 1,5-diphenylcarba-zide in acid solution (USEPA Method 7196A) and correlatedagainst an absorbance fit for a series of control solutions(prepared from analytical reagent grade potassium dichro-mate, Fisher Scientific) of Cr(VI). The method detection limitthe Cr(VI) analyses was 0.08 mg L−1. Total element concentra-tions were determined by inductively coupled plasma opticalemission spectrometry (ICP-OES) (USEPA Method 6010B) usinga Thermo-Jerrel Ash IRIS 1000, with daily calibrations anddetection limits (DLs) typically 0.33mg L−1 for Cr, 1.0mg L−1 forFe and Mg, and 2.0 mg L−1 for Al, Ca, and Na. Solids remainingafter the batch experiments were resuspended in 100 mL of DIwater and decanted three times to remove unbonded pre-cipitates from the surface of the particles before composi-tional and mineralogical analyses (see Section 2.2.3).

2.2.3. Composition of untreated and acid-treated COPRSolid-phase concentrations of untreated COPR were deter-mined by XRF (reported as apparent oxide percentages). Solid-phase concentrations of Cr(VI) and Cr(total) for untreated andacid-treated COPR were determined by digesting the solids inalkaline (USEPA Method 3060A) and acid (USEPA Method3050B) solutions, respectively, with subsequent analysis ofthe digest solutions by colorimetric techniques (USEPAMethod 7196A) and ICP-OES (USEPA Method 6010B), respec-tively. The initial and final compositions of COPR were used toassess the effectiveness of the acid treatments in mobilizingCr. For the XRF analysis, sample decompositionwas by lithiumtetraborate fusion in which COPR (1.0 g) was added to lithiumtetraborate flux (9.0 g), mixed well, and fused in a furnace at1100 °C.

Mineralogical composition by X-ray powder diffraction(XRPD) was used to determine if acid treatments caused(1) impacts to Cr(VI)-bearing minerals, (2) a reduction in thealkaline mineral content, and (3) formation of new minerals.Specimens for XRPD were prepared from air-dry COPR or acid-treated COPR ground with a mortar and pestle until passing aU.S. No. 200 sieve (75 μm). The use of hand-ground, air-driedspecimens are recommended for providing optimal XRPDresults for COPR because excess moisture can result inextensive hydration reactions during sample preparation,causing mischaracterization and significant underestimationof the brownmillerite content (Dermatas et al., 2007). Thepowder (≈2 g) was placed in an acrylic sample holder with aspecimen depth of 3 mm. Copper Kα radiation was used in adiffracted beam monochromator with beam energy of 40 keVand a current of 35 mA. Diffraction patterns were recorded bystep (0.01°) scanning at 1 s per step for diffraction angles (2θ)from 7.5° to 72.5°.

Scanning electron microscope (SEM) and elemental X-raymapping analyses were conducted to determine if acidtreatments caused the formation of precipitates in theinterstitial areas of COPR particles or a redistribution ofelements. COPR particles and acid-treated particles were air-dried, embedded in low-viscosity epoxy resin, polished ingraduated increments, and coated with a 20-nm-thick layer ofvacuum-evaporated carbon. A Cameca SX50/51 SEM was usedwith a beam energy of 15 keV and a 220218 current of 20 nA.

Qualitative X-ray mapping was conducted on a 4-μm gridusing a count time of 0.2 s/pixel.

3. Results and discussion

3.1. Mobilization of chromium from COPR reacted withacid

Aqueous concentrations of Cr with respect to final pH forCOPR reacted with HNO3 (48-h mixing time) are shown inFig. 2. Data from Geelhoed et al. (2002) are also shown. Forneutral to alkaline pH, the soluble Cr peaks (≈0.04 mmol g−1)within a narrow pH band of 7.5 to 8.1 [Cr(total)≈Cr(VI) in thispH range]. The maximum soluble Cr(VI) concentration repre-sents only one-third of the Cr(VI) mass in the solid phase,suggesting that the majority of Cr(VI) was not mobilizedduring the acid treatment. The peak in soluble Cr correspondswell to the solution pH for which aqueous concentrations of Crwere maximized (pH≈8.1) in batch experiments conducted byGeelhoed et al. (2002) using COPR fromGlasgow. The change inconcentration of soluble Cr with respect to pH is shown inFig. 2 and the overall trend for total soluble Cr closely followsthe leaching data from Geelhoed et al. (2002) and suggests thatthe two COPR sources have similarmineralogical composition.

The increase in aqueous Cr as the pH decreases from ≈11.5to 8 is attributed to dissolution of Cr-bearing minerals andanion substitution as the pH decreases (Geelhoed et al., 2002).In batch experiments (COPR mixed with HCl) conducted byGeelhoed et al. (2002), the elevated Cr concentration at pH 8coincides with a peak in carbonate concentration, andsuggests that aqueous carbonate also competes for adsorptionsites with CrO4

2− (Zachara et al., 1989).The decreasing concentration of aqueous Cr for pHb7.5 is

likely due to adsorption of chromates. For example, amor-phous metal oxyhydroxides are positively charged at low toneutral pH and CrO4

2− may be adsorbed by protonation of thesurface hydroxyl sites (Parfitt, 1978; Richard and Bourg, 1991).CrO4

2− is adsorbed to positively charged compounds when theeluate pHb the pH of zero point charge (pHZPC) for existingcompounds (Rai et al., 1988). Hillier et al. (2003), andChrysochoou et al. (2005) report that COPR contains upwardsof 40% amorphous material, which may provide adsorptionsites. Al and Fe are two of the primary elements in COPR andpoorly-ordered phaseswith these cationsmay include Fe(OH)3(pHZPC=8.5), γ-AlOOH (8.2), α-FeOOH (7.8), γ-Fe2O3 (6.7), andα-Al(OH)3 (5.0) (Stumm and Morgan, 1996). Weng et al. (1994,2002) reported a pHZPC of 6.8 for a COPR sample from theeastern seaboard of the US (i.e., where the COPR from thisstudy was obtained). This COPR-specific, composite pHZPC of6.8 is centered in the range of pHZPC (5.0–8.5) of potentialpoorly-ordered phases in COPR and suggests that thesecompounds may be responsible for the adsorption of CrO4

2−

from pH 5 to 8.The increase in adsorption of CrO4

2− with decreasing pH isattributed to the increasing positive charge on the adsorbingoxyhydroxide surfaces (Davis and Leckie, 1980; Zachara et al.,1987, 1989). The concentration of bichromate (HCrO4

−) alsoincreases as the pH decreases, and bichromate is the preferredsorbateonoxidesurfaces.Chromatewill alsoadsorb tocrystalline



iron and aluminum oxides at low pH (Zachara et al., 1989). Theadsorptionof chromate to thesurfacesofamorphous ferrihydrate(Fe2O3·H2O(am)), α-FeOOH, and Al2O3 increases rapidly once thesolution pH is lowered below 7.6 (Rai et al., 1989), which closelymatches the observed decrease in Cr concentrations (Fig. 2).Zachara et al. (1987) report that Fe2O3·H2O(am), which readilyforms a surface coating on particles, has a particularly high Cr(VI)adsorption capacity.

Aqueous concentrations of Al, Cr(total), Cr(VI), Fe, and Mnin the batch experiments with HNO3 are shown in Fig. 3 foreluate pHb6 (Ca and Mg concentrationsN2,000 mg L−1 in thisrange of pH). Concentrations for pHN6 are not shown toemphasize the dissolution of metal oxyhydroxides and otherminerals in COPR. Concentrations of Al and Fe in solutionincrease rapidly as the pH decreases b5. Dissolution of thesesolids releases Cr(VI) into solution, which may be subse-quently reduced to Cr(III) due to the Fe in solution (Geelhoedet al., 2002). Below pH of approximately 4, Cr(III)-bearing

solids, such as brownmillerite [Ca2(Al,Cr,Fe)O5] and chromite[(Mg2+,Fe2+)O·(Cr3+,Al3+,Fe3+)2O3], dissolve (Geelhoed et al.,2002), and the recalcitrant Cr(III) in COPR (≈75% of total Crmass) is solubilized.

3.2. Dependence of Cr(VI) release on particle size and typeof acid

The effect of particle size on the mobilization of Cr(VI) in thebatch experiments with HNO3 or H2SO4 is shown in Fig. 4. Forreactions with HNO3 and H2SO4, aqueous Cr(VI) concentra-tions are greater for incrementally smaller particles for thefirst 8 h of mixing time. For example, at 2 h, HNO3 treatmentresulted in average Cr(VI) concentrations of 220 mg L−1 for thefine-sized particles, 160 mg L−1 for the medium-sized parti-cles, and 80 mg L−1 for the coarse-sized particles. The samepattern emerged at 2 h of reaction time for COPR reacted withH2SO4, but relatively higher Cr(VI) concentrations wereobserved (280 mg L−1 for fine-sized, 190 mg L−1 for medium-sized, and 180 mg L−1 for coarse-sized particles). However, thedifference between the trend lines of medium-sized andcoarse-sized COPR reacted with H2SO4 is minimal (Fig. 4b).The slower dissolution/desorption of Cr(VI) from largerparticles suggests that the transport of Cr(VI) from the interiorof COPR plays a role in the leaching kinetics, which is similarto transport limitations proposed by Burke et al. (1991), Higginset al. (1998), Geelhoed et al. (1999), and Tinjum (2006). Thetransport limitationsmay also be impacted by the penetrationof the reagent, including H+ penetration.

Fig. 3 – Aqueous concentrations of Al, Cr(total), Cr(VI), Fe,and Mn in batch experiments with COPR mixed with HNO3

for LS=10 after 48-h reaction time.

Fig. 4 – Cr(VI) concentrations in the eluatewhen reactedwith(a) HNO3 or (b) H2SO4 for coarse-, medium-, and fine-sizedparticles.

Fig. 2 – Concentration of Cr in eluate from COPR reactedwithHCl (Geelhoed et al., 2002) and HNO3 (this study). Liquid tosolid ratio of 10:1 used in both data sets.



After 60+ h of reaction time, the aqueous Cr(VI) concentra-tions for all particle sizes are approximately the same for agiven acid (≈160 mg L−1 for HNO3, 285 mg L−1 for H2SO4). Thesimilarity of the aqueous Cr(VI) concentration at long mixingtimes is supportive of similar initial mineralogical composi-tion (Tinjum, 2006) and elemental composition regardless ofparticle size. The only particle size dependent compositionvariability was the initial solid concentration of Cr(VI), withlower Cr(VI) concentrations for smaller particles (Table 1).

There are three potential sources of mobilized Cr(VI) in thealkaline pH range (8–12): (1) dissolution of Cr(VI)-bearingminerals such as hydrotalcite and hydrogarnet, (2) anionexchange from the interlayers of Cr(VI)-bearing minerals suchas hydrocalumite, and (3) CrO4

2− exchange from the externalsurfaces of other compounds such as poorly-ordered metaloxyhydroxides (Dermatas et al., 2007; Geelhoed et al., 2002;Hillier et al., 2003; Zachara et al., 1987). Model calculations by

Geelhoed et al. (2002) suggest that hydrocalumite and hydro-garnet, Cr(VI)-bearing minerals found in COPR (Hillier et al.,2003; Tinjum, 2006), are dissolved at equilibrium conditionsfor pHb11. However, Geelhoed et al. (2002) found thatequilibrium had not been reached at 26 d of batch timebecause of slow dissolution kinetics, suggesting that double-layered hydroxide minerals in the COPR of this study may nothave completely dissolved in the 48-h of mixing time.However, SO4

2− from H2SO4 may preferentially exchange,relative to CrO4

2−, in the structure of these double-layeredhydroxide minerals and suggests that anion exchange con-tributes to the mobilization of Cr(VI) at alkaline pH (8–12). Thehigher equilibrium concentration of aqueous Cr(VI) in COPRmixed with H2SO4 is likely attributed to the increased ionexchange of SO4

2− for CrO42−, relative to NO3

− (Fig. 4). NO3 is not aslikely to exchange for CrO4

2− in double-layered hydroxidesbecause of the valence difference (electrostatic binding

Fig. 5 – Temporal consumption of acid (HNO3) by COPR for 20-min pH titration intervals and individual 48-h batch times.

Fig. 6 – pH rebound for COPR reacted with (a) nitric acid and (b) sulfuric acid.



mechanism), although specific chemical bindingmechanismsmay also be responsible for Cr(VI) adsorption (Stollenwerk andGrove, 1985).

Depression of CrO42− binding in the presence of competing

anions is reported by Zachara et al. (1989) and Rai et al.(1989), and SO4

2− from COPR treated with H2SO4 may competewith CrO4

2− for exchange sites in double-layered hydroxides(minerals with hydroxide charge deficit balanced by inter-layer anions such as CrO4

2− and SO42−; Pinnavaia, 1993) or for

adsorption sites on the surface of amorphous metalcoxy-hydroxides (Rai et al., 1989). Desorption of CrO4

2− from metaloxyhydroxides occurs in the presence of SO4

2−, especiallywhen the concentration of SO4

2− is increased (Davis and Kent,1990), and Richard and Bourg (1991) report a 33% reduction inadsorbed chromate to Fe2O3·H2O(am) with addition of 210 mMof sulfate. However, the pH of COPR mixed with acids was

alkaline after long mixing times (8.0 with HNO3 and 8.8 withH2SO4, Fig. 6), and few poorly-ordered, simple Al- or Fe-basedcompounds are positively charged at pHN8 (pHZPC for Fe(OH)3=8.5, γ-AlOOH=8.2) (Fig. 2). The likely sources ofmobilized Cr(VI) are thus from the dissolution of Cr(VI)-bearing minerals and anion exchange in double-layeredhydroxides.

3.3. Acid consumption by COPR

Because pH is an important factor impacting the solubility of Cr-bearing compounds and Cr(VI) adsorption/desorption, the quan-tity of acid required to neutralize COPR and the mixing timerequired for acid to fully penetrate COPR particles were investi-gated.AsshowninFig. 5, thequantityofacid required to lower thepH to a near-neutral pH of 7.5 (ANC7.5) is very large (8 mol HNO3

Fig. 7 – Metal concentrations in eluate from COPR reacted with HNO3 H2SO4. Concentrations for HNO3 tests are shown in (a) Aland (c) Ca. Concentrations for H2SO4 tests are shown in (b) Al and (d) Ca.



per kg dry COPR). The ANC7.5 is less than that reported byGeelhoed et al. (2002) for COPR from Glasgow reacted with HCl(13molH+kg−1) andcomparable to thatofsteel slag (8molH+kg−1)(Yan et al., 1998). Like COPR, steel slag contains a large amount ofhydroxide-based minerals such as Ca(OH)2 and Mg(OH)2 (Craw-ford andBurn, 1969). Because of the veryhighANCofCOPR, directtreatmentmethods are necessary if the pH is to be reduced in thenear term. Dissolution of alkaline minerals by groundwater oracidic rainwater infiltration is unlikely to significantly affect the insitupHofCOPRexceptoververy longperiodsof time (SreeramandRamasami, 2001; Geelhoed et al., 2002; Tinjum, 2006).

The acid consumption in COPR depends on mixing time,particle size, and acid type (Fig. 6). Underwell-mixed conditions(10:1 LS ratio, end-over-endmixing), pHof the eluate reboundedwith time after a single addition of acid. As an example, forcoarse-sized COPR mixed with HNO3 (Fig. 6a), the pH started at3.8 (1h), increased to 6.3 (8h), and finally stabilized at 8.0 (≈32h).The pH rebound indicates that acid penetration and consump-tion in COPR is rate-limited, most likely due to transportmechanisms within the solid-phase. The pH–time dependencyfor a similar, highly alkaline material (cement-stabilized, metalcontaminated soil) leached with acid was reported to be rate-limited by three mechanisms: H+ penetration into the solid,subsequent reactions with alkaline minerals such as portlan-dite, and diffusion of OH− out of the solid (Islam et al., 2004).Similar rate-limiting steps are likely in COPR.

Rate-limited acid consumption in COPR suggests thatminimum mixing times may be required to achieve constantpHandstabilizationofpHmaydependonthe largest particles inCOPR (Fig. 6). For example, for 2 h of mixing with HNO3 (Fig. 6a),the average eluate pH was ≈7 for the fine-sized particles, ≈6 forthe medium-sized particles, and ≈4 for the coarse-sized

particles (i.e., the pH in the eluate rebounded quicker for smallerparticles). The final eluate pH stabilized at 8.0 for all particlessizes after approximately 30 h of reaction time with HNO3,whereas the pH stabilized at 8.8 after approximately 16 h ofreaction timewith H2SO4. The H2SO4 reactions weremore rapidand complete as indicated by stabilization at higher pH (8.8 forH2SO4 vs. 8.0 for HNO3), less time required for pH stabilization(16 h vs. 30 h), and less differences in pH between particle sizes.For acid treatments in which the mixing time is desired to beminimized with less rate limitations based on particle size, theuse ofH2SO4maybe advantageous. In addition,H2SO4mobilizesmore Cr(VI) from the solid phase (Fig. 4).

Acid titration curves based on 20 min reaction times arerecommended to determine the buffering capacity of wastematerials (Kosson et al., 2002). A 20-min acid titration curve forCOPR is shown in Fig. 5 along with a curve for 48-hmixing times(per Fig. 6, a minimum mixing time of 32 h was required tostabilize the pH of COPR mixed with HNO3). Penetration andconsumptionof acidbyCOPRduring the20-minacid titration testis not complete. For example, reducing the eluate pH of COPR to7.5 (ANC7.5) requires approximately 9 timesmore HNO3 for a 48-hmixing time versus a 20-min mixing time. Use of quick, acidtitration tests to determine buffering tendencies inCOPRmaynotrealistically portray the true, long-term ANC of COPR. The time-dependency of the pH response may also explain why pHrebound has occurred in pilot studies where COPR is treatedwith acid (Chowdhury, 2003).

3.4. Characterization of COPR after acid treatment

The average specific gravity (s.g.) of solids from the batch testswith HNO3 and H2SO4 are summarized in Table 1. No consistent

Fig. 8 – Comparison of XRPD patterns for COPR reacted with nitric acid for 1, 8, 16, and 64 h, including an untreated controlspecimen.



trends in s.g. with respect to timewere observed for either of theacid-treated batch series. The average s.g. increases from 2.83 to3.04 for HNO3, and to 2.95 for H2SO4. The increase in average s.g.is evidence that select minerals in COPR with lower s.g. (e.g.,portlandite=2.23, brucite=2.39, and hydrogarnet=2.63) areselectively dissolving in deference to minerals with higher s.g.(e.g., calcium chromate=3.14, brownmillerite=3.76, chro-mite=4.79). The relatively smaller average s.g. of COPR reactedwithH2SO4 (2.95 vs. 3.04 forHNO3) ispotentiallydue to formationof gypsum (s.g.=2.3). Ca andAl leached fromCOPR are shown inFig. 7. Ca was dissolved from COPR mixed with H2SO4 atconcentration N190 mg L−1 regardless of mixing time or particle

size (Fig. 7a), and gypsumwas observed in column tests onCOPRpermeated with a FeSO4\H2SO4 solution(Tinjum, 2006).

Average solid-phase concentrations of Cr(VI) after HNO3 orH2SO4 treatments are summarized in Table 1. Similar toobservations of s.g., no consistent trends in solid-phase Cr(VI)concentrations were observed for either of the acid-treatedbatch series. A large amount of Cr(VI) (typically N50% of theinitial concentration) remainsafter acid treatment, regardlessofthe type of acid used. Residual Cr(VI) N50% of the initialconcentration supports results from the batch experiments inwhichb33%ofCr(VI)wasmobilized.Basedonresults reported inthe literature, the residual Cr(VI)may reside in the crystal lattice

Fig. 9 – X-ray maps (1200 μm×900 μm) for Al, Ca, Cr, and Fe (in rows from top to bottom). The columns represent, from left toright: COPR reacted with HNO3 for 2 h, COPR reacted with HNO3 for 32 h, and COPR reacted with H2SO4 for 32 h. Lighter colorsindicate higher X-ray intensity (and higher concentration of element), with the scale ranging from 0 (black) to 10 (white).



ofundissolvedCr(VI)-bearingminerals suchasCAChydrates, asunexchanged anions in double-layered hydroxides such ashydrocalumite and hydrotalcite, and may also be adsorbed topositively charged amorphous solids such as ferrihydrates andAl oxyhydroxides that were precipitated in the pores of COPRparticlesduring treatment (Dermatas etal., 2007;Geelhoedet al.,2002; Hillier et al., 2003; Zachara et al., 1987). More Cr(VI) wasmobilized in COPR treated with H2SO4; consequently, theaverage solid-phase concentration of HNO3-treated COPR wasslightly greater (3400 vs. 3200 mg kg−1).

XRPD patterns of solids from COPR mixed with HNO3 fordifferent times along with the pattern for untreated COPR areshown in Fig. 8. Cr(VI)-bearing minerals identified in theuntreated COPR (calcium aluminate chromate hydrate, hydro-garnet, and hydrotalcite) generally have diminished peakintensities in comparison to acid-treated COPR, indicating thatthese Cr(VI)-bearing minerals partially dissolved during acidtreatment and contributed to the amount of Cr(VI) in solution.At lower pH (b8), Cr(VI) was probably adsorbed to amorphousprecipitates such as Al2O3(am) and Fe2O3·H2O(am) and in poorly-ordered oxyhydroxides such as Al(OH)3. The broad ‘hump’ thatrises above the background for diffraction angles between 10°and 16° (not observed for untreated COPR) is characteristic ofthese precipitates, which have been shown to form whenCOPR is reacted with acid (Geelhoed et al., 2002). Trends in theconcentration of aqueous Al in batch tests conducted withHNO3 (Fig. 7c) and H2SO4 (Fig. 7d) suggest that dissolved Alrapidly precipitates. For mixing times b4 h, and only forcoarse-sized particles, aqueous Al is at elevated concentra-tions (AlN50 mg L−1) and then drops to low concentrationsafter 4 h (≈20 mg L−1 for HNO3, b5 mg L−1 for H2SO4). Thedecrease in dissolved Al for medium-sized particles treatedwith HNO3 is also perceptible (Fig. 7c) by comparing 1-hmixingtimes (32 mg L−1) with 2-h mixing times (18 mg L−1). Elevatedconcentrations of dissolved Al for smallmixing times followedby stable, low concentrations of dissolved Al after smallincreases in mixing time indicate that Al compounds arerapidly precipitated in acid-treated COPR. Precipitated Aloxyhydroxides are positively charged at pHbpHZPC and serveas sorption sites for CrO2− (Rai et al., 1988).

Elemental X-ray mapping scans are shown in Fig. 9 forcoarse-sized particles reacted with HNO3 for 2 and 32 h andwith H2SO4 for 32 h. These maps show the distribution of Al,Ca, Cr, and Fe within COPR particles after acid treatment andwhere precipitates form within the particles. Enhanced Alintensities are present on the periphery of sub-particles (seeFig. 9) treated with HNO3 for 32 h, indicating that Al hasprecipitated within the micro-pores in the particle matrix, assuggested by the drop in Al concentration in the eluate shownin Fig. 7c. Precipitation of Al is not observed in particles mixedwith HNO3 for 2 h and is consistent with the high Al concen-trations in the eluate at 2 h (Fig. 7c).

Ca and Fe also are found in the areas where Al precipitated,particularly in the interstitial areas. A similar effect occurswhen acids react with cement-stabilized wastes. Acid releasescations such as Ca from the rapid dissolution of alkalineoxyhydroxides at a reaction boundary, resulting in a highconcentration of the cation that partially reprecipitates inlocalized areas of higher pH within the solid (Cussler, 1984).These precipitates (especially Al and Fe) probably contain

metal oxyhydroxides that arepositively charged for pHbpHZPC,and act as sorption sites for CrO4

2−. Cr is present in theseinterstitial areaswith Al and Fe precipitates (Fig. 9), supportingthe assertion that Cr(VI) is adsorbed to positively chargedmetal oxyhydroxides precipitated in the particles.

The absence of Cr in the interstitial areas of the particletreated with H2SO4 for 32 h indicates that the acid removed Crfrom the interstitial areas in this particle. However, Cr is stillpresent in the interior of the sub-particles, as Cr(III) and Cr(VI)(see Table 1), indicating that H2SO4-treated COPR did notefficiently mobilize all of the Cr(VI) from the solid-phase.

4. Summary and conclusions

The experimental findings indicate the following: (1) mobili-zation of Cr is highly sensitive to pH, (2) single applications ofacid treatmentsmay not efficientlymobilize Cr(VI) fromCOPR,(3) the volume of acid required to neutralize COPR is large, and(4) pH and metal mobilization from COPR exposed to acidictreatments vary temporally. These factors will play a role inimplementation of acid treatments of COPR in field scenarios.

There is a very narrow band of pH (7.6–8.1) in which aqueousCr(VI) concentrations are maximized (0.04 mol/kg) when COPR istreatedwith acid. This suggests that treatment of COPRwith acidto maximize the amount of Cr(VI) in solution can be expected tobedifficult inheterogeneous in situenvironments. In addition, themaximumequilibriumconcentration of Cr(VI) fromCOPR treatedwith HNO3 (pH≈8) represents only ≈33% of the available Cr(VI) inthe solid phase, which suggests that multiple sequences of acidadditions (and porewater flushes) may be required to remove Cr(VI) from the solid phase adequately.

Although acid treatments dissolved many Cr(VI)-bearingcompounds from COPR, concentrations of Cr(VI) N2800 mgkg−1 exist in COPR mixed with solutions of HNO3 and H2SO4.Residual Cr(VI) likely remains in undissolved Cr(VI)-bearingminerals because of relatively slow dissolution kinetics(Geelhoed et al., 2002). Residual Cr(VI) may also reside onpoorly-ordered or amorphous precipitates, especially atslightly acidic to neutral pH (5–7.5). If pH rebound occurs inthe field, as observed in the batch experiments, Cr(VI) that wasadsorbed to Fe and Al oxyhydroxide precipitates may beremobilized as the sorptive potential of these surfaceschanges with pH. Fe and Al oxyhydroxide precipitates wereobserved in XRPD patterns and X-ray scans of COPR treatedwith acid, and they likely adsorbed chromates. These pH-dependent Cr(VI) adsorption/desorption tendencies may limitthe amount of Cr(VI) rapidly removed from with acid treat-ments and extend the time of Cr(VI) leaching. To mobilize ahighmajority of Cr(VI) from the solid phase, a very low (b3) pHwill need to be sustained.

The ANC of COPR is very large (nearly 8 mol of HNO3 per kgsolid COPR to reduce the starting eluate pH of 11.3 to a near-neutral pH of 7.5). To put the quantities of acid required toneutralize COPR in perspective, approximately 1 L of 8N HNO3

would be required to neutralize 1 kg of dry COPR in an ex situbatch operation. For an in situ injection approach using 2NHNO3, about 10 pore volumes of flow would be required underideal conditions. Greater efficiency could be achieved withH2SO4. The H2SO4 reactions were more complete based on



stabilization of the pH (8.8 for H2SO4 vs. 8.0 for HNO3; i.e., lessH+ in solution), less time required for pH stabilization (16 h vs.30 h), and less differences in pH for different particle sizes. Inaddition, the amount of Cr(VI) from COPR in solution nearlydoubled when the reacting acid was H2SO4.

Once acidified pore fluid is flushed from COPR treated withacid, a rebound in pH may subsequently cause a release ofadsorbed Cr(VI) from the surface of amorphous oxyhydrox-ides. Under well-mixed conditions (10:1 LS ratio, end-over-endmixing), the pH of the eluate continued to rebound after singleadditions of an acid (N16 h). This observation has importantimplications for field applications of acids to COPR; acidsmustbe introduced andmixedwith COPR for a long period of time toachieve a stabilized pH in the mixing fluid, which may limitthe feasibility of ex situ or in situ mixing applications.

Acknowledgements

Funding for this research was provided by the Federal HighwayAdministration, the Graduate School at the University ofWisconsin–Madison, and theWisconsin Industrial and EconomicDevelopment Research Program. John Fournelle (EugeneCameron Electron Microprobe Laboratory) and Huifang Xu (S.W.Bailey X-Ray Diffraction Laboratory) of the University of Wiscon-sin–Madison assisted with the experimental setup for themineralogical analyses; their assistance is gratefully acknowl-edged. RMT Inc. of Madison, Wisconsin, conducted the ICPanalyses.

R E F E R E N C E S

Barlett RJ, JamesBR.Mobilityandbioavailability of chromiuminsoils.In: Nriagu JO, Nieboer E, editors. Chromium in the natural andhuman environments, 20. John Wiley & Sons; 1988. p. 267–303.

Baron D, Palmer CD, Stanley JT. Identification of twoiron-chromate precipitates in a Cr(VI)-contaminated soil.Environ Sci Technol 1996;30(3):1810–3.

Burke T, Fagliano J, Goldoft M, Hazen RE, Tglewicz R, McKee T.Chromite ore processing residue in Hudson County, NewJersey. Environ Health Perspect 1991;92:131–7.

Chowdhury AK. Method for stabilizing chromium-contaminatedmaterial. US Patent 6,607,474. Aug 19, 2003.

Chrysochoou M, Dermatas D. Application of the Rietveld methodto assess chromium (VI) speciation in chromite ore processingresidue. J Hazard Mater 2006;141:370–7.

Chrysochoou M, Moon DH, Wazne M, Christodoulatos C, Meng X,KaourisM,Morris J, FrenchC, Sass BM.Mineralogical analysis ofchromite ore processing residue by X- ray powder diffraction.In: AllemanBC, KelleyME, editors. 8th International In- situ andOn-site Bioremediation Symposium, 6–9 June. Baltimore,Maryland: Battelle Press; 2005.

Crawford CB, Burn KN. Building damage from expansive steel slagbackfill. J Soil Mech Found Div 1969;95(SM6):1325–34.

Cussler EL. Diffusion — Mass transfer in fluid systems. 1st ed.Cambridge: Cambridge University Press; 1984.

Darrie RG. Commercial extraction technology and process wastedisposal in the manufacture of chromium chemicals from ore.Environ Geochem Health 2001;23:187–93.

Davis JA, Leckie JO. Surface ionization and complexation in theoxide/water interface 3. Adsorption of anions. J Colloid InterfaceSci 1980;74:32–43.

Davis JA, Kent DB. Surface complexation modeling in aqueousgeochemistry. In: Hochella MF, White AF, editors.Mineral–water interface chemistry, vol. 23. Review inMineralogy; 1990. p. 177–260.

DermatasD, ChrysochoouM, Pardali S, GrubbDG. Influence of X-raydiffraction sample preparation on quantitative mineralogy:implications for chromate waste treatment. J Environ Qual2007;36:487–97.

Eary LL, Rai D. Chromate removal from aqueous wastes byreduction with ferrous iron. Environ Sci Technol 1988;22(8):676–83.

Gancy AB, Wamser CA. Treatment of chromium ore residues tosuppress pollution. US Patent 3,937,785. Feb 10, 1976.

Geelhoed JS, Meeussen JCL, Lumsdon DG, Roe MJ, Thomas RP,Farmer JG, Paterson E. Processes determining the behaviour ofchromium in chromite ore processing residue used as landfill.Land Contam Reclam 1999;7(4):271–9.

Geelhoed JS, Meeussen JC, Hillier S, Lumsdon DG, Thomas RP,Farmer JG, Paterson E. Identification and geochemicalmodelingof processes controlling leaching of Cr(VI) and other majorelements from chromite ore processing residue. GeochimCosmochim Acta 2002;66(22):3927–42.

Geelhoed JS, Meeussen JCL, Roe MJ, Hillier S, Thomas RP, FarmerJG, Paterson E. Chromium remediation or release? Effect of iron(II) sulfate addition on chromium(VI) leaching from columns ofchromite ore processing residue. Environ Sci Technol 2003;37(14):3206–13.

GrahamMC, Farmer JG, Anderson P, Paterson E, Hillier S, LumsdonDG, Bewley RJF. Calcium polysulfide remediation of hexavalentchromiumcontamination fromchromiteore processing residue.Sci Total Environ 2006;364(1–3):32–44.

Henry KS, Petura JC, Brooks S, Dentico S, Kessel SA, Harris M.Preventing surface deposition of chromium with asphalt capsat chromite ore processing residue sites: a case study. CanGeotech J 2007;44:814–39.

Higgins TE, Halloran AR, Dobbins ME, Pittignano AJ. In situreduction of hexavalent chromium in alkaline soils enrichedwith chromite ore processing residue. J Air Waste ManageAssoc 1998;48(11):1100–6.

Higgins TE. Process for the in situ bioremediation of Cr(VI)-bearingsolids. US Patent 5,562,588. Oct 8, 1996.

Hillier S, RoeMJ, Geelhoed JS, Fraser AR, Farmer JG, Paterson E. Roleof quantitative mineralogical analysis in the investigation ofsites contaminated by chromite ore processing residue. SciTotal Environ 2003;308:195–210.

Interfaith Community Organization v Honeywell International,Inc. Civil Action No. 95-2097 (DMC), US District Court, NewJersey May 21, 2003.

Islam MZ, Catalan LJJ, Yanful EK. Effect of remineralization onheavy-metal leaching from cement-stabilized/solidified waste.Environ Sci Technol 2004;38(5):1561–8.

James BR. Hexavalent chromium solubility and reduction inalkaline soils enriched with chromite ore processing residue.J Environ Qual 1994;23:227–33.

James BR. Method to reduce hexavalent chromium in soils,sediments, industrial waste and other contaminated materialsusing ascorbic acid. US Patent 5,951,457, Sept 14 1999.

Klaassen CD, editor. Casarett and Doull toxicology: The basicscience of poisons. 6th ed. NY, NY: McGraw–Hill; 2001.

KossonDS,vanderSlootHA,SanchezF,GarrabrantsAC.An integratedframework for evaluating leaching in waste management andutilization of secondary materials. Environ Eng Sci2002;19(3):159–204.

Lioy PJ, Freeman NCG, Wainman T, Stern AH, Boesch R, Howell T,Shupack SI. Microenvironmental analysis of residentialexposure to chromium-laden wastes in and around newJersey homes. Risk Anal 1992;12(2):287–99.

Meegoda JN, Kamolpornwijit W, Vaccari DA, Ezeldin AS, Noval BA,Mueller RT, Santora S. Remediation of chromium-contaminated



soils: Bench-Scale investigation. Pract Period Hazard, Toxic,Radioact Waste Manag 1999;3(3):124–31.

Millero FJ. The effect of ionic interactions on the oxidation ofmetals in natural waters. Geochim Comochim Acta1985;49:547–55.

Montclair EnvironmentalManagementTeam (MEMT).Managementplan for chromium- contaminated soil. Int J Environ Stud1990;35:263–75.

Palmer CD, Wittbrodt PR. Processes affecting the remediation ofchromium- contaminated sites. Environ Health Perspect1991;92:25–40.

Parfitt RL. Anion adsorption by soils and soil materials. Adv Agron1978;30:1–50.

Pinnavaia TJ. Clay catalysts: opportunities for use in improvingenvironmental quality. In: Churchman GJ, Fitzpatrick RW,EggletonRA,editors. ClaysControlling theEnvironment, Proc 10th

International Clay Conference, Adelaide, Australia. Melbourne,Australia: CSIRO; 1993. p. 3–7.

Public Health Service. Health of Workers in the ChromiumProducing Industry. Public Health Service Publication No. 192.Washington, DC: US Government Printing Office; 1953.

Raghu D, Hsieh H-N. Origin, properties and disposal problems ofchromium ore residue. Intern Environ Stud 1989;34:227–35.

Rai D, Zachara JM, Eary LE, Ainsworth CC, Amonette JE, Cowan CE,Szelmeczka RW, Schmidt RL, Girvin DC, Smith SC. Chromiumreactions in geologic materials. Electric Power Res Inst (EPRI).Palo Alto: CA; 1988. Rept EA-5741.

Rai D, Eary LE, Zachara JM. Environmental chemistry of chromium.Sci Total Environ 1989;86:15–23.

Richard Jr LR, Bourg ACM. Aqueous chemistry of chromium: areview. Water Res 1991;25:807–16.

Saleh FY, Parkerton TF, Lewis RV, Huang JH, Dickson KL. Kineticsof chromium transformations in the environment. Sci TotalEnviron 1989;86:25–41.

Sass BM, Rai D. Solubility of amorphous chromium(III)-iron(III)hydroxide solid solution. Inorg Chem 1987;26(14):2228–32.

Shupack SI. The chemistry of chromium and some resultinganalytical problems. Environ Health Perspect 1991;92:7–11.

Sreeram KJ, Ramasami T. Speciation and recovery of chromiumfrom chromite ore processing residues. J Environ Monit2001;3:526–30.

StollenwerkKG, GroveDB. Adsorption and desorption of hexavalentchromium in an alluvial aquifer near Telluride, Colorado.J Environ Qual 1985;14(1):150–5.

Stumm W, Morgan JJ. Aquatic chemistry, chemical equilibria andrates in natural waters. 3rd ed. New York, NY: John Wiley andSons, Inc.; 1996.

Su C, Ludwig RD. Treatment of hexavalent chromium in chromiteore processing solid waste using a mixed reductant solution offerrous sulfate and sodium dithionite. Environ Sci Technol2005;39(16):6208–16.

Thomas RP, Hillier SJ, Roe MJ, Geelhoed JS, GrahamMC, Paterson E,Farmer JG. Analytical in characterization of solid- andsolution-phase chromium species at COPR-contaminatedsites. Environ Geochem Health 2001;23:195–9.

Thornton RF. Removal of chromium from solution usingmechanically agitated iron particles. US Patent 5,380,441. Jan10, 1995.

Tinjum JM. Mineralogical properties of chromium ore processingresidue and chemical remediation strategies. PhD Dissertation;U Wisconsin–Madison 2006; 257 p.

Truini J. N.J. urges chromium cleanup. Waste News 2005;11(1):3.US Environmental Protection Agency. Toxicological Review of

Hexavalent Chromium. In Support of Summary Information onthe Integrated Risk Assessment System; 1998.

Wazne M, Papazoglou P, Papastavrou M, Meng X, Dermatas D,Christodoulatos C, et al. Remediation of chromite oreprocessing residue by chemical reductants. In: Alleman BC,Kelley ME, editors. 8th International In-situ and On-siteBioremediation Symposium, 6–9 June. Baltimore, Maryland:Battelle Press; 2005.

Weng CH, Huang CP, Allen HE, Cheng AH-D, Sanders PF.Chromium leaching behavior in soil derived from chromite oreprocessing waste. Sci Total Environ 1994;154(1):71–86.

Weng C-H, Huang CP, Sanders PF. Transport of Cr(VI) in soilscontaminated with chromite ore processing residue (COPR).Pract Period Hazard, Toxic, Radioact Waste 2002;6(1):6–13.

Yan J, Bäverman C, Moreno L, Neretnieks I. Evaluation of thetime-dependent neutralising behaviours of MSWI bottom ashand steel slag. Sci Total Environ 1998;216:41–54.

Zachara JM, GirvinDC, Schmidt RL, ReschCT. Chromate adsorptionon amorphous iron oxyhydroxide in presence of majorgroundwater ions. Environ Sci Technol 1987;21:589–94.

Zachara JM, Cowan CE, Schmidt RL, Ainsworth CC. Chromateadsorption on kaolinite. Clays Clay Miner 1988;36(4):317–26.

Zachara JM, Ainsworth CC, Cowan CE, Resch CT. Adsorption ofchromate by subsurface soil horizons. Soil Sci Soc Am J1989;53:418–28.


Documents

Author's personal copy - University of Virginia OF THE TOTAL ENVIRONMENT 391 (2008) 13 25 ... kinetics (Geelhoed et al., 2002) ... Author's personal copy