Environ. Sci. Technol. 2004, 38, 6161-6165
Establishing and Elucidating Reduction as the Removal Mechanism of Cr(VI) by Reclaimed Limestone Residual RLR (Modified Steel Slag) CHARLES E. OCHOLA* AND HORACE K. MOO-YOUNG College of Engineering, Villanova University, Villanova, Pennsylvania 19085
The viability of utilizing Reclaimed Limestone Residual RLR (Modified Steel Slag) to remove hexavalent chromium Cr(VI) from the aqueous phase was investigated. A physical characterization of RLR showed that it is composed of various minerals some of which can reduce and others adsorb Cr(VI). Preliminary results showed that RLR significantly reduced the concentration of Cr(VI) from the aqueous phase. Adsorption competition tests with orthophosphate (HPO42-) and sulfate (SO42-) showed that Cr(VI) was still effectively reduced from solution regardless of the competing anions present. Kinetic tests based on the relationship d[Cr(VI)]/dt ) kcr[RLR]R[Cr(VI)]β showed that under initially neutral to basic conditions kcr ) 3.45 ( (0.25) × 10-4 mg0.4 L-0.4 h-1, R ) 0.9, and β ) -0.3, while under initially acidic conditions kcr ) 5.65 ( (1.055) × 10-11 mg-0.4 L0.4 h-1, R ) 2.2, and β ) -0.8. Stirred batch tests with RLR in deionized water showed significant drops in the redox potential (Eh), and in the presence of oxygen Eh values dropped to between 50 and 100 mV while in the absence of oxygen Eh values as low as -200 mV were observed. These results lead to the conclusion that redox mechanisms were responsible for the reduction of hexavalent chromium by RLR.
or alkaline solutions Cr(VI) is generally reduced to chromic hydroxide (Cr(OH)3), while in acidic solutions it is reduced to either Cr3+ or if the conditions are highly reducing Cr2+ (5). Most Cr(VI) remediation methods are therefore based on reduction. Reclaimed limestone residual (RLR) is a hard dense material produced as a byproduct of the steel making process, that contains significant amounts of iron as free iron or various iron compounds (6, 7). It is composed primarily of calcium silicates and alumino-ferrites and fused oxides of calcium, iron, magnesium, and manganese and might vary slightly in chemical content depending on the properties required of the steel being produced. The unslaked lime and magnesia present in RLR generally makes it mildly alkaline with a solution pH ranging from 8 to 10, with solution leachate in this study capable of exceeding pH 11. Objectives. The objectives of this paper are to show the dominant mechanism, responsible for the removal of hexavalent chromium in the aqueous phase by RLR, and to propose a pathway by which the process occurs. Various researchers (8-13) have shown that elemental iron and various iron compounds are capable of reducing Cr(VI) concentrations in the aqueous phase. Chemical characterization of the RLR in this study shows that, besides iron, there exists other minerals with the potential of adsorbing Cr(VI) and effectively reducing its concentration in solution. Bartlett and Kimble (14) proposed that orthophosphate (HPO42-) competes with chromate for adsorption sites. They observed that in the presence of HPO42- chromate adsorption was severely retarded, and there were also similarities between the adsorption and desorption behavior of chromate and phosphate in Spodsol horizons as pH was increased with CaCO3. Observations by Bartlett and James (15) led them to the conclusion that there may be similarities on the adsorption of chromate (CrO42-), sulfate (SO42-), and phosphate (HPO42-) on colloids with positively charged surfaces. Based on these observations, tests were conducted using solutions containing mixtures of CrO42- with SO42-, CrO42with HPO42-, and CrO42- with SO42-, and HPO42- batched with RLR in order to determine whether adsorption, defined here as the mechanisms of electrostatic attraction and complexation reactions, was the mechanism responsible for the reduction of hexavalent chromium or whether reduction was the dominant Cr(VI) removal process.
Introduction
Experimental Section
Cr(VI) is toxic to living cells and is believed to be a carcinogen. Unfortunately as an anion Cr(VI) is quite soluble over a very large pH range and readily penetrates cell membranes. For this reason the U.S. Environmental Protection Agency (1) set a maximum drinking water standard of 50 µg L-1. Cr(VI) is strongly oxidizing and exists only in oxo species, such as CrO3 (chromic acid), CrO2Cl2 (chromyl chloride), and the chromate ion CrO42- (2). Cr(III) is generally nontoxic and is believed to be essential in glucose metabolism in mammals (3). The National Academy of Science actually estimates a safe adequate intake of 0.05 to 0.20 mg/day (4). There exist various mechanisms that can reduce the amount of Cr(VI) in the aqueous phase, and these include sorption, precipitation, transformation, and or various combinations of these processes. Under reducing conditions (low Eh values) or in the presence of oxidizable compounds, it is possible to reduce hexavalent chromium, and in neutral
RLR. The RLR utilized in this study was obtained from the Waylite Corporation located in Bethlehem, Pennsylvania. It was characterized as fines and had the consistency of coarsegrained sand. The results of the major physical properties tested showed that the RLR had a water content of 0.4%, an organic content of 1.5%, a porosity of 30%, and a specific gravity Gs of 3.46. Grain size analysis showed that the RLR analyzed was well graded. Results of X-ray diffraction analysis showed that the RLR is approximately 92% crystalline and was composed of several compounds including but not exclusive to those listed in Table 1. Analysis for elemental content in the RLR was achieved by a scanning electron microscope (SEM) using energy-dispersive X-ray (EDX) spectroscopy. Before analysis, the RLR was finely ground and sieved to create a proper sample for (EDX). An average of the results in weight percent from three areas scanned excluding oxygen and carbon is presented in Table 2. Chemicals. Chemical reagents utilized in this study were reagent grade or better and included potassium dichromate (K2Cr2O7) from Aldrich chemical company, potassium di-
* Corresponding author phone: (610)519-7440; fax: (610)519-4941; e-mail:
[email protected]. 10.1021/es049670l CCC: $27.50 Published on Web 10/19/2004
2004 American Chemical Society
VOL. 38, NO. 22, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6161
TABLE 1. Mineralogical Composition of RLR mineral periclase larnite ingersonite calcium magnesium silicate calcium magnesium silicate bredigite ghelnite galaxite fayalite manganese calcium aluminum oxide manganese oxide manganosite calcite lime
chemical formula MgO Ca2SiO4 Ca2SiO4 Ca5MgSi3O12
general type oxide olivine olivine
Ca7Mg(SiO4)4 Ca4Mg2(SiO4)8 Ca2Al2SiO7 MnAlO Fe2SiO4 Mn Ca2Al2O4
oxide olivine native element oxide
MnO2 MnO CaCO3 CaO
oxide oxide carbonate oxide
TABLE 2. Elemental Composition of RLR
a
element
av wt percenta (%)
magnesium aluminum silicon sulfur calcium manganese iron
6.2 3.9 17.9 0.7 58.6 2.0 11.3
The average weight percents do not include oxygen or carbon.
hydrogen phosphate (KH2PO4), sodium sulfate (Na2SO4), and hydrochloric acid (HCl) from Fisher scientific. All solutions were prepared with deionized water (electrical resistivity at 25 °C ) 18MΩ cm) obtained from a Labconco Water Pro. Shaken Batch Testing. During preliminary testing, K2Cr2O7 was dissolved in deionized water to prepare hexavalent chromium solutions in concentrations of 0.5, 5.0, 9.1, 22.6, 44.9, and 89.5 mg/L. Ten and 20 g of RLR were placed in 120 mL plastic bottles, and 100 mL of the various hexavalent chromium solutions was poured into the bottles and tightly capped. All the various RLR and Cr(VI) mix ratios were prepared in triplicate. The samples were placed in a Lars Lande rotary extractor and rotated at 30 ( 2 rpm at an ambient temperature of 24 degrees centigrade. At the end of a 24-h batching period, the bottle contents were filtered through a 0.45 µm filter membrane, and the final concentration of hexavalent chromium was determined by the diphenylcarbohydrazide method in an Agilent 8453 spectrophotometer. Based on the various minerals inherent in RLR, the effects of adsorption defined here as the mechanisms of electrostatic attraction and complexation reactions were considered. Magnesium oxide (MgO) was of particular concern because it has a very high point of zero charge (PZC) of 12.4 (16, 17). Such a high PZC could result in the sorption of the chromate anion. To establish the effects of electrostatic considerations, subsequent testing involved the preparation of 50 mg/L Cr(VI) mixed with various concentrations of orthophosphate (HPO42-) prepared from KH2PO4 and sulfate (SO42-) prepared from Na2SO4 to determine the effects of anion competition. The pH was adjusted to approximately 7.3 to ensure that orthophosphate was speciated as HPO42-. Twenty grams of RLR was placed in 120 mL glass bottles, and 100 mL of the various chromate compositions were poured in the bottles, sealed with Teflon lined caps, and batched for 48 h. All testing was done in triplicate. 6162
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 22, 2004
FIGURE 1. Cr(VI) reduction vs initial concentration. Averaged data replicates shown with error bars; where no error bars are seen, they are smaller than the size of the data symbol.
Stirred Batch Testing. The preceding method did not facilitate the testing of parameters such as pH, the oxidationreduction potential (ORP), or dissolved oxygen concentration with time, and therefore an additional batch testing procedure was utilized. This procedure involved the utilization of a reaction vessel where the Cr(VI) solution and RLR was mixed with a blade driven by a motor. Probes inserted into the system enabled the measurement of Eh, pH, and DO (dissolved oxygen). Nitrogen gas could also be purged through the system to eliminate the dissolved oxygen. Stirred batch testing was conducted on 25, 50, and 100 g of RLR placed in 1 L of deionized water and monitored for Eh and pH changes over time. The first set of tests were conducted in open air, and in the second set of testing nitrogen was purged through the solution for the duration of the test to eliminate any dissolved oxygen. Kinetic Testing. Kinetic batch testing was utilized to determine the reaction rate of RLR with Cr(VI). To minimize the effects of oxygen, N2 was purged through all Cr(VI) solutions for at least 1 h before batching. From the initial kinetic testing it was noticed that reduction rates by RLR of initially neutral Cr(VI) were very similar to reduction rates of basic Cr(VI). There was however a significant difference in these reaction rates compared to those of initially acidic solutions. Based on these observations, it was determined that reaction rate constant for the initially neutral to basic Cr(VI) could be determined simultaneously, and the reaction rate constant for the initially acidic Cr(VI) had to be determined separately. The test procedure involved the preparation of Cr(VI) solutions of varying concentrations and pH, followed by batch reactions in triplicate of 100 mL of these solutions in 10 and 20 g of RLR. The concentration of Cr(VI) in solution was then analyzed at specific times resulting in a concentration profile over time.
Results and Discussion Batch Tests. The averaged results of the preliminary tests are shown in Figure 1. For the lower concentration of 0.5 mg/L the reduction of Cr(VI) was almost 100% in both the 10- and 20-gram samples. As the concentration increased the reduction was less for both the 10- and 20-gram samples. It is of interest to note that at a concentration of 9.1 mg/L, the rate of change of reduction with increased Cr(VI) concentration dropped significantly. For example in the 20gram RLR sample mixed with 9.1 mg/L Cr(VI) the reduction was approximately 80%, and for the increased concentration of 22.6 mg/L the reduction was only slightly different at approximately 76%. The same phenomenon was also apparent in the 10-gram samples. These results showed that RLR reduced the concentration of Cr(VI) in solution. However
TABLE 3. Competition Batch Test Resultsa
competing anions
final Cr(VI) concn (mg/L)
reduction (%)
none 100 mg/L HPO42100 mg/L SO42200 mg/L HPO42200 mg/L SO42200 mg/L HPO42- + 200 mg/L SO42500 mg/L HPO42- + 500 mg/L SO42-
3.82 1.78 4.03 2.01 3.42 2.41 3.76
92.2 96.3 91.8 95.9 93.0 95.1 92.4
a All solution mixtures contained an initial Cr(VI) concentration of 49.2 mg/L.
FIGURE 3. ORP vs time comparison for 25, 50, and 100 g RLR batched in 1 L DI water (nitrogen).
FIGURE 2. ORP vs time comparison for 25, 50, and 100 g RLR batched in 1 L DI water (air). Averaged data replicates shown with error bars; where no error bars are seen, they are smaller than the size of the data symbol. these results did not give an indication as to what mechanism was responsible for this reduction. The results of batch testing with competing anions are shown in Table 3. In all cases regardless of any competing anions present in solution, RLR was still able to reduce the concentration of hexavalent chromium in solution by over 90%. In all cases the final pH was approximately 11. Cr(VI) hydrolyzes extensively in water, and in basic media the only significant species is CrO42-. McBride (3) has shown that CrO42- adsorbs less strongly than Cr3+, and consequently its mobility is higher. These facts and the observations by Bartlett and Kimble (14) and Bartlett and James (15) on competitive adsorption between chromate, orthophosphate, and sulfate previously discussed are compelling evidence that sorption mechanisms by RLR were not responsible for the reduction of Cr(VI) in its chromate form. The results of the subsequent batch tests that enabled the measurements of the redox potential of the system are shown in Figures 3 and 4. Upon addition of RLR there was a rapid drop in redox potential (Eh) irrespective of the quantity of RLR in solution. In the case where oxygen was allowed into the system, the redox potential dropped to between 50 mV and 100 mV and stayed there for approximately 12 h after which it slowly rose to approximately 200 mV. In the case where nitrogen was purged through the system to remove any dissolved oxygen (Figure 3), the redox potential quickly dropped to -200 mV for all the three quantities of RLR. The 100-gram RLR solution held the lower redox potential much longer than either the 25- or 50-gram RLR solutions. These results showed that RLR does indeed cause reducing conditions in an aqueous solution. Chromate Reduction Kinetics Analysis. Batch techniques were utilized in determining the rate of chromium reduction by RLR. Figure 4 is a plot of the reaction time for 20-gram
FIGURE 4. Cr(VI) vs time by 20 g RLR at various initial pH values. Averaged data replicates shown with error bars; where no error bars are seen, they are smaller than the size of the data symbol. RLR batched in 100 mL Cr(VI) solutions. The objective was to vary the initial pH values of the Cr(VI) solutions, while keeping the initial Cr(VI) concentrations constant. Due to pH adjustments by addition of either HCl or NaOH, the concentrations varied slightly. At initial pH values of 1.88, 7.26, and 11.43, the initial Cr(VI) concentrations were 55.06 mg/L, 49.20 mg/L, and 58.49 mg/L, respectively. Although the final concentrations for both the 10- and 20-gram batches were different, the trends in both cases were very similar. In both cases the reaction rates for the acidic Cr(VI) was slower than the reaction rates for the Cr(VI) systems with initial neutral to basic pH values, although the final concentrations after a couple of days were approximately the same, and all systems were ultimately basic. For reactions occurring in a liquid system at constant volume, reaction rate is expressed as the number of reactant species changed into product species per unit of time and per unit of volume of the reactant species (18). Rates can be expressed as either a decrease in reactant concentration per unit time or an increase in product concentration per unit time. In this study, the reduction of Cr(VI) concentration was dependent on the quantity of RLR and concentration of Cr(VI). From Sparks (18), the Cr(VI) reduction rate can then be expressed as
d[Cr(VI)] ) kcr[RLR]R[Cr(VI)]β dt
(1)
where kcr is the rate constant, and R and β are the orders of the reaction with respect to RLR and Cr(VI), respectively. Rate constants can be determined in several ways (19, 20), and these include ascertaining initial rates, using integrated rate equations directly and graphing the data, and VOL. 38, NO. 22, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6163
employing nonlinear least-squares techniques (18). In this research the initial rate method was utilized, and this involved plotting the concentration of the reactant (Cr(VI)) for a very short initial period of the reaction during which the change of concentration was so small that the instantaneous rate is hardly affected (19). Alternatively determining the initial linear slope of a Cr(VI) concentration vs time plot will give the same results. Experimental procedures involved varying the initial concentrations of each reactant, while holding the other reactants constant (18). The experiment called for the nonvaried reactant to be constant over the time of measurement, such as being provided in excess to the varied reactant. However because the rate was measured over a very short time period of the reaction, the nonvaried reactant did not need to be necessarily in excess for this approach (10). In this study it was observed that RLR reacted faster in basic waters than in acidic waters although the overall capacity was the same in both cases. Under initially neutral to basic conditions it was determined that the rate constant kcr ) 3.45 ( (0.25) × 10-4 mg0.4 L-0.4 h-1, the order of dependence on RLR is 0.9, and the order of dependence on Cr(VI) is -0.3, while under initially acidic conditions the rate constant kcr ) 5.65 ( (1.055) × 10-11 mg-0.4 L0.4 h-1, the order of dependence on RLR is 2.2, and the order of dependence on Cr(VI) is -0.8. Stumm and Morgan (21) proposed a reaction scheme for the dissolution of minerals where the reaction can be shown schematically by the two following sequences: fast
surface sites + reactants (H+, OH-, or ligands) 98 surface species slow
surface species 9 8 Me(aq) detachment of Me Although each sequence may consist of a series of smaller reaction steps, the rate law of surface controlled dissolution is based on the idea that the attachments of reactants to the surface sites is fast, and the subsequent detachment of the metal species from the surface of the crystalline lattice into the solution is slow and thus rate limiting (21). This theory may account for the sudden drop in Cr(VI) reduction shown in Figure 1 especially if a majority of the reactive surface sites is initially utilized. Another hypothesis for the control of dissolution is known as the armoring precipitate hypothesis where diffusion of the species through a reprecipitated layer limits the rate of hydrolysis (22). Other researchers, Schott and Berner (23, 24), and Chou and Wollast (25) have also attributed the decrease in mineral dissolution rates over time to the formation of a diffusion inhibiting leached layer or secondary precipitates on the mineral surface. The dissolution rates of iron containing silicates have been determined by various researchers including Siever and Woodford (26), Eggleton and Boland (27), Berner and Schott (28), and Schott and Berner (23). McBride (3) presented a silicate mineral dissolution rate in the lab of 10-7-10-8 mol/cm2 s although he also states that in natural systems the rate is much slower. Wogelius and Walther (29) described a dissolution rate expression for fayalite as 1.1 × 10-10[H]0.69 + 3.2 × 10-14 mol/ cm2 s, and Westrich et al. (30) presented a fayalite dissolution rate of 10-10-10-11 mol/cm2 s. Various researchers (31-33) have observed a U-shaped curve for dissolution rate and pH for silicate minerals with the minimum dissolution rates generally observed near the PZC (34, 21). The dissolution rates of silicate and oxide minerals are pH dependent, reflecting the incorporation of H+, H2O, or OH- into rate controlling surface complexes (35). Given the high pH of RLR solutions, OH- is believed to be strongly associated with any rate controlling surface complexes. Buerge and Hug (36) postulated that HCrO4- and CrO42might have different reactivities. This suggestion coupled 6164
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 22, 2004
with observations that the reduction of acidified Cr(VI) occurred at a different rate than that of neutral to basic Cr(VI), necessitated the investigation of chromate reduction kinetics by RLR. Most dissolution rate expressions for silicates and oxides, only consider proton promoted dissolution mechanisms since there is very little if any literature for dissolution rate expressions at near-neutral and basic pH conditions (35). Tests conducted to establish the reaction rate constants of initially acidified chromium and initially neutral to basic chromium revealed that initially neutral to basic chromium i.e., HCrO4- had a rate constant kcr ) 3.45 (( 0.25) × 10-4 mg0.4 L-0.4 h-1 which is greater than the initially acidified chromium i.e., CrO42- rate constant kcr ) 5.645 (( 1.055) × 10-11 mg-0.4 L0.4 h-1. Although the rate constants differ, the capacity of RLR to reduce chromium is the same in both cases. Eary and Rai (8) conducted experiments that investigated the reduction of chromate by ferrous iron. One of the objectives of their studies was to determine how pH affected Cr(VI) reduction by Fe(II). Their experiments involved batching Cr(VI) and Fe(II) at pH values ranging from 3.5 to 12. Their results showed that under acidic conditions it was possible to predict the stoichiometry of the reaction as 3.0 mol of aqueous Fe(II) oxidized to 1.0 mol of Cr(VI) reduced. However under alkaline conditions the reaction was nonstoichiometric. Westheimer (37) based on kinetic studies characterized the rate-determining step for the reduction of Cr(VI) by Fe(II) as an inner sphere reaction. Furthermore Cotton and Wilkinson (38) stated that inner sphere reactions could generally be expected to increase in rate as pH increased because hydrolysis of the reactants promote the transfer of electrons through bridging hydroxyl ions. Eary and Rai (8) thus conclude that “On the basis of the characterization of the reaction between Cr(VI) and Fe(II) as being an inner sphere reaction, it is unlikely that an increase in pH, which results in increased hydrolysis of Fe(II) ions, would cause a decrease in reaction rate. In fact, the opposite effect is more likely”. Experimental observation in this study showed that this was the case with RLR and Cr(VI), and therefore this could be one possible explanation for the greater reaction rate of initially neutral to basic Cr(VI) as opposed to the acidic Cr(VI) via the following proposed scheme which is based on the premise that the reduction process in RLR is a one-electron-transfer process similar to instances often found in soils (39): [RLR] + [OH ] + [Cr6+] T [RLR-(OH)-Cr]5+ oxidant reductant inner-sphere complex
Step 1: A bridged complex is formed. e- transfer
[RLR-(OH)-Cr]5+ 98 [RLR-(OH)-Cr]5+ inner-sphere complex inner-sphere complex Step 2: An electron is transferred via OH-. + 5+ [RLR-(OH)-Cr]5+ f [RLR ] + [Cr ] + [OH ] inner-sphere complex
Step 3: The bridged complex splits up. The preceding scheme is repeated two more times with subsequent reduced chromium complexes until a stable trivalent chromium ion is produced. This ability of OH- to promote electron transfer has also been used by Sedlak and Chan (40) to explain the increase in reaction rates at elevated pH values, and they state that the electron-donating ligands stabilize Fe(III) and make Fe(II) a more reductive reactant (41).
Acknowledgments The authors wish to thank PITA the Pennsylvania Infrastructure Technology Alliance for its financial support for
this study and the Waylite Corporation Bethlehem, Pennsylvania for providing the raw material for RLR.
Literature Cited (1) U.S. EPA. Chromium. Quality Criteria for Water; Washington, DC, 1976. (2) Nieboer, E.; Jusys, A. A. Biologic chemistry of chromium. In Chromium in the Natural and Human Environments; Nriagu, J. O., Nieboer, E., Eds.; John Wiley & Sons: New York, 1988; pp 21-79. (3) McBride, M. B. Environmental Chemistry of Soils; Oxford University Press: New York, 1994. (4) Tate, C. H.; Arnold, K. F. Health and aesthetic aspects of water quality. In Water Quality and Treatment: A Handbook of Community Water Supplies; Pontius, F. W., Tech Ed.; McGrawHill Inc.: New York, 1990; pp 63-154. (5) Faust, S. D.; Aly, O. M. Chemistry of Natural Waters; Butterworth Publishers: Stoneham, Mass, 1981. (6) Kalyoncu, R. S. Slag-Iron and Steel: U.S. Geological Survey Mineral Commodity Summaries, 2000. (7) Emery, J. J. Extending Aggregate Resources; ASTM Special Technical Publication 774; American Society for Testing and Materials: 1982; pp 95-118. (8) Eary, L. E.; Rai, D. Chromium removal from aqueous wastes by reduction with ferrous iron. Environ. Sci. Technol. 1988, 22, 972-977. (9) Eary, L. E.; Rai, D. Kinetics of chromate reduction by ferrous ions derived from hematite and biotite at 25 °C. Am. J. Sci. 1989, 289, 180-213. (10) Fendorf, S. E.; Li, G. Kinetics of chromate reduction by ferrous iron. Environ. Sci. Technol. 1996, 30, 1614-1617. (11) Loyaux-Lawniczak, S.; Lecomte, P.; Ehrhardt, J. Behavior of hexavalent chromium in a polluted groundwater: Redox processes and immobilization in soils. Environ. Sci. Technol. 2001, 35, 1350-1357. (12) White, A. F.; Yee, A. Aqueous oxidation-reduction kinetics associated with coupled electron-cation transfer from ironcontaining silicates at 25 °C. Geochim. Cosmochim. Acta 1985, 49, 1263-1275. (13) White, A. F. Heterogeneous electrochemical reactions associated with oxidation of ferrous oxide and silicate surfaces. Rev. Mineral. 1990, 23, 467-509. (14) Bartlett, R. J.; Kimble, J. M. Behavior of chromium in soils. II. Hexavalent forms. J. Environ. Qual. 1976, 5, 383-386. (15) Bartlett, R. J.; James, B. R. Mobility and bioavailability of chromium in soils. In Chromium in the Natural and Human Environments; Nriagu, J. O., Nieboer, E., Eds.; John Wiley & Sons: New York, 1988; pp 267-315. (16) Parks, G. A. The isoelectric points of solid oxides, solid hydroxides and aqueous hydroxo complex systems. Chem. Rev. 1965, 65, 177-198. (17) Yoon, R. H.; Salaman, T.; Donnay, G. Predicting points of zero charge of oxides and hydroxides. J. Colloid. Interface Sci. 1979, 70, 483-493. (18) Sparks, D. Kinetics of Soil Chemical Processes; Academic Press: San Diego, Ca, 1989. (19) Bunnett, J. F. Techniques of chemistry Vol 6. In Investigations of Rates and Mechanisms of Reactions, 4th ed.; Bernasconi, C. F., Ed; Wiley: New York, 1986; pp 251-372. (20) Skopp, J. Analysis of time-dependence process in soils. J. Environ. Qual. 1986, 15, 205-213. (21) Stumm, W.; Morgan, J. J. Aquatic Chemistry: Chemical Equilibria and Rates in Natural Waters, 3rd ed.; John Wiley & Sons: New York, 1996.
(22) Stumm, W. Aquatic Surface Chemistry: Chemical Processes at the Particle-Water Interface, 2nd ed.; John Wiley & Sons: New York, 1987. (23) Schott, J.; Berner, R. A. X-ray photoelectron studies of the mechanism of iron silicate dissolution during weathering. Geochim. Cosmochim. Acta 1983, 47, 2233-2240. (24) Schott, J.; Berner, R. A. Dissolution mechanism of pyroxenes and olivines during weathering. In The Chemical of Weathering; Drever, J. I., Ed.; Reidel: Dordrecht, The Netherlands, 1985; Vol. 149, pp 35-53. (25) Chou, L.; Wollast, R. Study of the weathering of albite at room temperature and pressure with a fluidized bed reactor. Geochim. Cosmochim. Acta 1984, 48, 2205-2217. (26) Siever, R.; Woodford, N. Dissolution kinetics and the weathering of mafic minerals. Geochim. Cosmochim. Acta 1979, 43, 717724. (27) Eggleton, R. A.; Boland, J. N. Weathering of enstatite to talc through a sequence of transitional phases. Clays Clay Miner. 1982, 30, 11-20. (28) Berner, R. A.; Schott, J. Mechanism of pyroxene and amphibole weathering. 11 observations of soil grains. Am. J. Sci. 1982, 282, 1214-1231. (29) Wogelius, R. A.; Walther, J. V. Olivine dissolution kinetics at near surface conditions. Chem. Geol. 1992, 97, 101-112. (30) Westrich, H. R.; Cygan, R. T.; Casey, W. H.; Zemitis, C.; Arnold, G. W. The dissolution kinetics of mixed-cation orthosilicate minerals. Am. J. Sci. 1993, 293, 869-893. (31) Blum, A. E.; Lasaga, A. C. Role of surface speciation in the lowtemperature dissolution of minerals. Nature 1988, 331, 431433. (32) Brady, P. V.; Walter, J. V. Controls on silicate dissolution rates in neutral and basic pH solutions at 25 °C. Geochim. Cosmochim. Acta 1989, 53, 2823-2830. (33) Oelkers, E. H. Physical and chemical properties of rocks and fluids for chemical mass transport calculations. Rev. Mineral. 1996, 34, 131-191. (34) Wieland, E.; Wehrli, B.; Stumm, W. The coordination chemistry of weathering: III. A generalization on the dissolution rates of minerals. Geochim. Cosmochim. Acta 1988, 52, 1969-1981. (35) Parsons, M. B.; Bird, D. K.; Einaudi, M. T.; Alpers, C. N. Geochemical and mineralogical controls on trace element release from the Penn Mine base-metal slag dump, California. Appl. Geochem. 2001, 16, 1567-1593. (36) Buerge, I. J.; Hug, S. J. Kinetics and pH dependence of Chromium(VI) reduction by iron(II). Environ. Sci. Technol. 1997, 31, 1426-1432. (37) Westheimer, F. H. The mechanisms of Chromic acid oxidations. Chem. Rev. 1949, 45, 419-451. (38) Cotton, F. A.; Wilkinson, G. Advanced Inorganic Chemistry; John Wiley & Sons: New York, 1980. (39) Yen, T. F. Environmental Chemistry Chemical Principles for Environmental Process; Prentice Hall PTR: Upper Saddle River, NJ, 1999. (40) Sedlak, D. L.; Chan, P. G. Reduction of hexavalent chromium by ferrous iron. Geochim. Cosmochim. Acta 1997, 60, 21852192. (41) Wehrli, B. Redox reactions of metal ions at mineral surfaces. In Aquatic Chemical Kinetics: Reaction Rates of Processes in Natural Waters; Stumm, W., Ed.; Wiley: 1990; pp 311-336.
Received for review March 2, 2004. Revised manuscript received September 5, 2004. Accepted September 6, 2004. ES049670L
VOL. 38, NO. 22, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6165
