ABSTRACT Title of Dissertation: CHROMIUM OXIDATION AND REDUCTION BY HYDROGEN PEROXIDE IN DIVERSE SOILS AND SIMPLE AQUEOUS SYSTEMS Melanie Louise Rock, Doctor of Philosophy, 1999 Dissertation directed by: Professor George R. Helz Department of Chemistry Professor Bruce R. James Department of Natural Resource Sciences Hydrogen peroxide is being tested for in situ remediation of buried contaminants - either as a direct chemical oxidant in Fenton-type reactions or as a source of oxidizing equivalents in bioremediation. How it affects a common co- contaminant, Cr, is explored here in four chemically diverse high-Cr soils. Soils contaminated with high levels of soluble Cr(VI) from ore processing and soils containing high levels ofrecently reduced Cr(III) from electroplating waste showed marked increases in chromate after single applications of J-25 mM peroxide. Cr(VI) in the leachates exceeded the drinking water standard (2?M) by 1-3 orders of magnitude. Soluble Cr(III), in the form of dissolved organic complexes, contributed to the likelihood of Cr(III) oxidation. Anaerobic soil conditions at a tannery site prevented oxidation of Cr(III). Naturally occurring Cr in serpentine soil also resisted oxidation. Ambient soluble Cr(VI) in a contaminated aquifer disappeared from peroxide leachates below pH 5, then reappeared as peroxide levels declined. In solutions prepared under environmentally relevant conditions, aged 280 ?M Cr(III) treated with I 00 ?M H20 2 showed increases in Cr(VI) over weeks with maximum oxidation rates achieved in solutions prepared with 2: I and 4: t OH -:Cr. Although Cr(III) speciation differs in fresh and aged aqueous systems, a similar mechanism involving the pre-equilibrium step: Cr(OH)/ + OH- .,. Cr(OH)/ may account for Cr(III) oxidation in both systems. Under alkaline conditions, H20 2 enhanced the oxidative dissolution of CrnCOH)3n?. The formation of peroxochrornium compounds in the presence of H20 2 and Cr(VI) may account for the disappearance and reappearance of Cr(VI) in H20 2 treated soils; as does the possible formation and subsequent reoxidation of Cr1\(0H) 23n_2 + oligomers. Mobilization of hazardous Cr(VI) must be considered in plans to use H20 2 for remediation of chemically complex wastes. Once Cr(III) is oxidized to Cr(VI) by H20 2 it may persist long after applied H20 2 treatments have disappeared. Further, hexavalent Cr will behave as a catalyst toward H20 2 in soils, enhancing its oxidative capacity while helping to dissipate high levels of applied H20 2? CHROMIUM OXIDATION AND REDUCTION BY HYDROGEN PEROXIDE IN DIVERSE SOILS AND SIMPLE AQUEOUS SYSTEMS . by Melanie Louise Rock Dissertation submitted to the Faculty of the Graduate School of the University of Maryland at College Park in partial fulfilhnent of the requirements for the degree of Doctor of Philosophy 1999 Advisory Committee: Professor George R. Helz, Chair Professor Bruce R. James Professor Neil V. Blough Professor Alice C. Mignerey Professor Robert S. Pilato ?Copyright by Melanie Louise Rock 1999 DEDICATION To my family 11 ACKNOWLEDGMENTS My graduate studies have been made possible through the support of an NSF Doctoral Traineeship in Groundwater Chemistry and Hydrology administered by Maryland's Water Resources Research Center. I also owe special thanks to University of Maryland professors Dr. Bob Pilato, Dr. Neil Blough, and Dr. Phil Kearney for their open and ongoing willingness to provide scientific insight and advice. I have been fortunate in this endeavor to have had the guidance and encouragement of my advisors, Dr. George Helz and Dr. Bruce James. Each of them has given generously of their time and expertise, and graciously accommodated the unusual, interdisciplinary nature of this project. They have been truly outstanding mentors. iii TABLE OF CONTENTS List of Tables vi List of Figures Vil Chapter l. Background on Soil Chemical Behavior of Chromium and Peroxide 1 Introduction 2 Peroxide in Soils 3 Peroxide Soil Chemistry 3 Using H2O2 to Enhance Subsurface Bioremediation 8 Using H20 2 in Fenton Remediation 11 Chromium in Soils 16 Extent of Chromium Contamination in Soils 16 Chromium Soil Chemistry 19 Deposition of Chromium in Industrial Sites 24 Chromium Reduction and Oxidation in Soils 26 Chemistry of Chromium and Peroxide Interactions 29 Cr(VI)/H20 2 Interactions 31 Cr(III)/H20 2 Interactions 37 Current Inquiry 39 Chapter 2. Hydrogen Peroxide Effect on Chromium Chemistry in Four Diverse Chromium-Enriched Soils 41 Introduction 42 Materials and Methods Chromite Ore Processing Residue (COPR) 44 Electroplating Waste Site in Connecticut 47 Aberjona Superfund Site 50 Maryland Serpentine Barrens 51 Soil Sampling 53 Chemicals 54 Experiments 54 Analytical Methods 55 Results and Discussion 59 Chromite Ore Processing Residue (COPR) 59 Electroplating Waste Site in Connecticut 63 Aberjona Superfund Site 80 Maryland Serpentine Barrens 80 Conclusions 82 iv Chapter 3. Chromium Oxidation, Reduction and Complexation by Hydrogen Peroxide in Defined Aqueous Systems 84 Introduction 85 Materials and Methods 86 Chemicals 86 Preparation of Reactant Solutions 87 Experiments 89 Analytical Methods 89 Results 91 Cr(IIl)/H2O2 Interactions 91 Cr(Vl)/H2O2 Interactions 99 Effects of Adding Methanol or Fe(II) 103 Discussion 114 Cr(III)/H2O2 Interactions 114 Cr(VI)/H2O2 Interactions 127 Effects of Adding Methanol or Fe(II) 131 Summary 132 Chapter 4. Chromium-Peroxide Interactions and Implications for the Use of Hydrogen Peroxide to Remediate Biorefractory Organic Waste 133 Appendix 140 References 151 V LIST OF TABLES 1-1. Data from the 1192 sites on EPA's National Priorities List 18 1-2. Instances of Cr contamination in different media at the NPL sites 19 1-3. Geographic distribution of chromium contamination at Superfund sites 20 2-1. Soil properties 45 2-2. Summary of analytical methods 58 2-3. Data for Figure 2-2. Cr(VI) concentrations in COPR soil upon single applications of H2O2 ( at day zero) 61 2-4. Data for Figure 2-3. Changes in Cr(VI) concentrations in Connecticut wetland plating waste soil (0-14 cm horizon) upon single applications ofH2O2 (at day zero) 65 2-5. Data for Figure 2-4. Short term changes in Cr(VI) concentrations in Connecticut plating waste site (0-14 cm) after single applications ofH2O2 68 2-6. Disappearance of 24.0 mM H2O2 after a single application in three soils 77 2-7. Cr(VI) undetected in Aberjona and Serpentine sites 77 3-1. Data for Figures 3-5a and 3-5b. The reaction of I 00 ?M Cr(VI) with varying initial concentrations ofH2O2 106 3-2a. Data for Figures 3-6a and 3-6b. The reaction of I 00 ?M Cr(VI) with 4500 ?M H2O2 in 0.0lM NaNO3 109 3-2b. Data for Figures 3-6a and 3-6b. The reaction of 100 ?M Cr(VI) with 3000 ?M H2O2 in 0.0IM NaNO3 110 3-2c. Data for Figure 3-6b. The reaction of 100 ?M Cr(VI) with 1500 ?M H2O2 in 0.01M NaNO3 111 A-1. Peak identification data for COPR soil 141 A-2. Peak identification data for Connecticut plating waste soil (0-14 cm) 142 A-3. Peak identification data for Serpentine soil (53-75 cm) 143 VI LIST OF FIGURES 1-1. Stability diagram for aqueous Cr(III) and Cr(VI) species 23 1-2. Stability diagram for aqueous Cr(III) and Cr(VI) species with oxidation and reduction lines for H20 2 30 1-3. Structure of peroxochrornium complexes 35 2-la. X-ray diffraction spectra for COPR soil 46 2-lb. X-ray diffraction spectra for Connecticut plating waste soil, 0-14 cm 49 2-lc. X-ray diffraction spectra for Serpentine soil, 53-75 cm 52 2-2. Changes in Cr(VI) concentrations in COPR soil upon single applications ofH20 2 60 2-3. Changes in Cr (VI) concentrations in Connecticut wetland plating waste soil (0-14 cm horizon) upon single applications of H20 2 64 2-4. Short term changes in pH and Cr(VI) concentrations in Connecticut wetland plating waste soil (0-14 cm horizon) upon single applications ofH20 2 67 2-5. Effect of varied solution to soil ratios on Cr(VI) measurements in the 0-14 cm horizon of the Connecticut plating waste soil 71 2-6. Cr(III) oxidation by a single application of 12.0 mM H20 2 in the 14-40 cm horizon of Connecticut wetland plating waste soil 74 2-7. Cr(III) oxidation by H20 2 in 14-40 cm horizon of Connecticut wetland plating waste soil 75 2-8. Short term changes in Cr(VI) concentrations in glacial till underlying a Connecticut wetland plating waste soil after a single application of3.00 mM H20 2 78 2-9. Reaction of3000 ?M H20 2 and 100 ?M HCr04? in aqueous solution of 0.0lM NaN03 at initial pH 4.5 79 2-10. Changes in H20 2 concentrations in glacial till underlying a Connecticut wetland plating waste soil after a single application of 3.00 mM H20 2 81 vii 3-1. Oxidation of nine 280 ?M Cr(III) solutions, using 100 ?M H 0 2 2 88 3-2a. Effect of initial pH on the reaction of 100 ?M Cr III) and 3000 ?M H20 2, initial pH 3 93 3-2b. Effect of initial pH on the reaction of 100 ?M Cr (III) and 3000 ?M H20 2, initial pH 4.75 94 3-2c. Effect of initial pH on the reaction of 100 ?M Cr(III) and 3000 ?M H20 2, initial pH 10 95 3-3a. Oxidation of 100 ?M Cr(III) solutions, using varying concentrations of H20 2 96 3-3b. Disappearance of H20 2 when different initial concentrations are added to 100 ?M aqueous, hydrolyzed Cr(III) solutions 97 3-3c. Changes in pH when varying concentrations of H20 react with 2 100 ?M aqueous, hydrolyzed Cr(III) 98 3-4a. Effect of initial pH on Cr(VI) behavior in the reaction of 100 ?M Cr (VI) and 3000 ?M H20 2 100 3-4b. Effect of initial pH on H20 2 behavior in the reaction of 100 ?M Cr(VI) and 3000 ?M H20 2 101 3-4c. pH changes in the reaction of 100 ?M Cr(VI) and 3000 ?M H 0 2 2 102 3-5a Effect of initial H20 2 concentration on its reaction with 100 ?M Cr(VI) 104 3-5b. pH changes in the reaction of different initial H20 2 concentrations with 100 ?M Cr(VI) 105 3-6a. Behavior of 4500 ?Mand 3000 ?M H20 2 when added to 100 ?M Cr(VI) at initial pH 4.0 in 0.01 M NaN03 107 3-6b. Changes in Cr(VI) and pH after addition of different concentrations of H20 2 in 0.01 M NaN03 108 3-7a. Reaction of 100 ?Maged, hydro1)7.ed (2: 1 OH?/Cr) Cr(IIl) and 3000 ?M H20 2 in 0.01 M NaN03, with and without methanol 112 3-7b. Reaction of 100 ?M Cr(VI) and 3000 ?M H20 2 in 0.01 M NaN0 , 3 with and without methanol 113 viii D 3-8a. Reaction of 100 ?Maged, hydrolyzed (2: 1 OH"/Cr) Cr(III) and 3000 ?M H2O2 in 0.01 M NaNO3 with and without the addition of 1.0 ?M FeSO4 115 3-8b. Reaction of 100 ?M Cr(VI) and 3000 ?M H2O2 in 0.01 M NaNO3 with and without the addition of 1 ?M FeSO4 116 3-9. Speciation diagrams for Cr(IIQ in aqueous systems 119 3-10. Proposed pre-equilibrium step to the oxidation of Cr(III) by H2O2 121 3-11. K vs. sampling time for the reaction of 100 ?M Cr(VI) with different concentrations ofH2O2 130 4-1. Stability diagram for aged aqueous Cr(III) and Cr(VI) species with a reduction line for FeOOH/Fe2+ 136 A-1. Oxidation of Cr(III) in Connecticut plating waste soil ( 14-40 cm) varying H2O2 concentrations 144 A-2. Oxidation of Cr(III) in Connecticut plating waste soil (14-40 cm) varying OH"/Cr ratios 145 A-3. Oxidation of Cr(III) in two soils after a single application of 3.00mMH2O2 146 A-4. Oxidation of Cr(III) in Serpentine soil (53-75 cm) after a single application of 3.00 mM H2O2 147 A-5. Interaction of 100 ?M Cr(III) and 3.00 mM H2O2 after a) sparging 30 min with 0 2 and b) sparging 30 min with N2 148 A-6. Interaction of 100 ?M Cr(VI) and 3.00 mM H2O2 at pH 4.0 after a) sparging 30 with 0 2 and b) sparging 30 min with N2 149 A-7. Ratio of H2O2 used to Cr(VI) produced from l 00 ?M Cr(Ill) 150 ix Chapter 1 Background on Soil Chemical Behavior of Chromium and Peroxide INTRODUCTION The oxidative treatment ofbiorefractory organic contaminants in soil and groundwater using hydrogen peroxide has recently attracted strong interest within EPA, but such treatment may oxidize the co-contaminant, Cr{III), if present. Hydrogen peroxide can be applied to remediation sites either as a direct oxidant of organic contaminants or as part of the well-known Fenton's reagent, in which it reacts with catalytic amounts ofFe(II) to produce the powerfully oxidizing hydroxyl radical (OH"). It has also been used to enhance aerobic bioremediation through production and delivery ofO2 (Carberry, 1994). The aim ofthis research is to investigate the possible chenucal interaction between peroxide and different forms of chronuum that may co-contaminate a site where this treatment is applied. Chronuum in soil and groundwater poses an environmental hazard only when the metal is found in its most highly oxidized state, as anionic Cr(VI), in which form it is toxic, mobile and classified as a class A human carcinogen (Calder, 1988; Katz and Salem, 1994). In contrast, chronuum(III) is nontoxic, an essential human nutrient involved with glucose metabolism, mostly insoluble in soils and not readily absorbed by plants. IfH2O2 were to oxidize Cr(III) to Cr(VI), a relatively innocuous waste material would be transformed into a hazardous one. This introduction will briefly review the chemistry of some of the possible biotic and abiotic reactions ofH2O2 in groundwater, and describe its use to remediate contaminated soil, both as an enhancement to bioremediation, and as a direct oxidant. It will also treat the chemistry of Cr speciation as it affects its behavior in soil and 2 groundwater, and the scope and nature of Cr contamination, typified by its presence at Superfund sites. Background will also be given on the known chemistry ofCr/H20 2 interactions that could be relevant under conditions of soil remediation. Chapter 2 will discuss four soils high in Cr, but different in almost every other respect, and examine their varied response to H20 2? Chapter 3 will examine the chemistry of those results in light of experiments using defined aqueous systems. In conclusion, Chapter 4 will look at current H20 2 remediation trends and examine implications for Cr contamination from the use of H20 2 to remediate high levels of organic contaminants in soil. PEROXIDE IN SOILS Peroxide Soil Chemistry Hydrogen peroxide levels have been measured in groundwater at I 0?1 to I o?8 M (Holm et al., 1987) and at 104; Min groundwater exposed to sunlight (Cooper and Zika, 1983). It is thought to be a respiration byproduct of aerobic soil microorganisms, certain Aerococcus species of which have been isolated and characterized (Kontchou and Blondeau, 1990). Enzymes such as aerobic dehydrogenases, amine oxidases, lysine monooxygenase, and xanthine oxidase produce H20 2 during nonnal cellular metabolic processes (Pardieck et al., 1992). H20 2 is also produced by microbes as the product of the superoxide dismutase catalyzed disproportionation of the superoxide anion radical (Oi?") (Price et al, 1992): 202?? + 2H+ .. H20 2 + 0 2 (I.I The superoxide radical is present in soils as a natural byproduct of microbial 3 I respiration: it results from the reduction of molecular 0 2 to H20 via single electron transfers, along with H20 2 and Off: 0 2 + e--+ 0 2?- (1.2 02? - + 2H+ + e- -+ H20 2 (1.3 H20 2 + H+ + e- -+ H20 + Off (1.4 Off + H+ + e- -+ H20 (1.5 Soil microorganism defenses which protect against excess cellular quantities of these reactive intermediate species could be expected to play a role in the response of soil biota to remedial H20 2 treatments. Hydrogen peroxide is a powerful oxidant, used commonly as a disinfectant. In soil, with or without the aid of mineral or biological catalysts, it would be capable of oxidizing reduced species such as Fe2+, Mn2+, or H2S, as well as some forms of soil organic matter. It may, in turn, be expected to react as a reductant in soils toward oxidized species like MnOOH, or Mn02? The oxidation state of oxygen in H20 2 (-1) is between that of 0 2 and H20, and it can be oxidized to 0 2 or reduced to H20 2 by 2 electron transfer reactions. The following half reactions summarize the redox chemistry of H20 2 under acidic and basic conditions (Greenwood and Earnshaw, 1994): H20 2 + 2H+ + 2e-.,. 2H20 E'N 1.776 (1.6 0 2 + 2H+ + 2e- .,. H20 2 0.695 (1.7 Ho2- +H20 + 2e- .,. 30ff 0.878 (1.8 0 2 + H20 + 2e- .,. H0 ? + Off -0.076 2 (1.9 Its strength as an oxidant decreases with increasing pH, but its reductive strength 4 increases, allowing disproportionation to be energetically feasible across the entire pH range. The thermodynamic prediction for the disproportionation ofH2O2 at 298 Kand atmospheric Po, is that reaction 1.10 will go to the right at any H20 2 concentration over 10-19M: (1.10 The reaction, however, is kinetically slow in a pure system (Brown et al., 1970). For some time it was thought (Duke and Haas, 1961) that the thermal decomposition of H2O2 occurs without a catalyst as a second order process via attack by HO2? on H2O2 with a rate maximwn at pH 11.8, the PKa ofH2O2 (Cotton and Wilkinson, 1988). Subsequently, work reviewed by Brown et al. (1970) showed that the H2O2 decomposition was probably catalyzed by trace metals present in the experimental reagents, and that when reagents are carefully purified and trace metal contamination controlled, H2O2 solutions are essentially stable, even under alkaline conditions. The thermodynamics of equations 1.6 and 1. 7 imply that any redox couple at a potential (E') greater than 0.695 V (O/H2O) and less than 1. 77 V (H2O2 IH20) would be a catalyst for the H2O2 disproportionation reaction, and in soils, Mn(lll,IV) (hydr)oxides and Fe (II,111) (hydr)oxides are the most likely non-biological mediators. For example, Mn (II) is oxidized by H2O2 , while Mn (111,IV) oxides may be reduced (Pardieck et al., 1992): H2O2 (aq) + 2Mn 2+(aq) + 2H2O .. 2MnOOH + 4H+ AG= -50.9 kJ/mol (1.11 H2O2 (aq) + 2MnOOH(s) + 4H+ .. 2Mn2+(aq) + Oz(aq) + 4H2O (1.12 5 ~G = -24.8 kJ/mol (values for both reactions at pH 7) Birnessite, (o-Mn02) a non stoichiometric Mn mineral common in terrestrial and aquatic environments, was found to decompose H20 2 following a first order kinetic rate law (Elprince and Mohamed, 1992) and was found to be the most probable inorganic catalyst ofH20 2 decomposition in a dry alluvial soil from an Egyptian floodplain (El- Wakil, 1986). Hydrogen peroxide will also decompose in the presence of the Fe3+/Fe2+ redox couple (E?= 0. 771 V) via the well known Fenton mechanism (Fenton, 1894; Haber and Weiss, 1932), notable for its production of the highly reactive OH" and 0 2? - intermediate species. In the Fenton mechanism, a reduced metal, e.g. Fe(II) or Cr(V), reacts with H20 2 in a rate limiting step to produce OH" ( equation 1.13 below). Trace amounts of a reduced metal will cycle between oxidized and reduced states in a series of one electron transfer reactions to catalyze the destruction of H20 2 ( equation 1.10) in a manner consistent with the reactions steps below (Evans and Upton, 1985; Wardman and Candeias, 1996): (1.13 (1.14 Fe2+ + OH" .., Fe3+ + Off (1.15 (1.16 (1.17 2 HO2 ? .., 2 0 2? - + 2H+ (1.18 6 Superoxide, (HO2"/Ot PKi = 4.8 Bielski et al., 1985) produced by the oxidation of H2O2 by either Fe 3 + or OH" ( equations 1.14 and 1. I 7 above), serves as a reductant for the oxidized metal, completing the catalytic cycle. Hydroxyl radicals produced in the initial step may be subsequently scavenged by Fe2+ ( equation 1.15), or by H2O2 ( equation I .17), and in a simple system, the reactions proceed to produce H2O and 0 2 as H2O2 is decomposed. In a natural system, the reactive intermediates would be expected to interact with organic compounds. Hydroxyl radicals are second only to F2 in oxidation potential, reacting non-specifically with organic compounds with bimolecular rate constants of l 07 to 1010 L/mole sec (Dorfam and Adams, 1973). Organic radicals may be produced by hydrogen extraction (Baxendale, 1955): (1.19 which may subsequently undergo dimerization or hydroxylation. Although the Fenton pathway for the decomposition ofH2O2 in soil may not predominate under natural conditions, it is purposefully introduced in remediation strategies which apply Fe2+ along with H2O2 to augment the oxidative treatment capacity with OH" radicals. Even more significant for the disproportionation ofH2O2 under field conditions than the mineral composition of the soil, however, is the ubiquitous presence of catalase-positive microrganisms. Catalase is a heme-protein that protects the cell from the reactivity of H2O2 by catalyzing its disproportionation through a cycling ofFe(III) and Fe(V), producing 0 2 and water. The specific activity of catalase is extremely high, with a turnover number, or number of molecules of substrate decomposed per molecule 7 of enzyme per minute, of I9 x I 06 (Herbert anq Pinsent, 1948). Spain et al. (1989) showed a rapid loss of H2O2 (I 00 mg/L) applied to the surface of a sand suspension (t112 = 4 hr), while no decomposition ofH2O2 was observed in sterile batch reactors. Pardieck et al. (1992) also observed less loss of H2O2 in autoclaved soil suspensions than occurred in field condition samples, and a greater loss in a silt loam with an organic matter content of3.25% than in a sandy loam with 0.95% organic matter. These results both point to the disappearance of H2O2 as a result of microbial activity. In addition to catalase, H2O2 may be biotically activated as an oxidant via peroxidase enzymes, which are also heme-proteins present in soil microbes, and reduce H2O2 without producing 0 2. A metal center in the enzyme is oxidized by H O , 2 2 producing H2O, while an organic substrate donates electrons to reduce the metal center. Thus, a catalytically-active species might behave like catalase in the absence of an organic electron donor (producing H2O and 0 2) , and like peroxidase when an organic substrate is present, producing H2O and an oxidized organic species. In sum, the biotic interactions ofH2O2 appear to be of three types: formation via superoxide dismutase, dismutation via catalase, and reduction coupled to oxidation of organic matter via peroxidase. Using H2O2 to Enhance Subsurface Bioremediation In the last decade, augmenting soils with hydrogen peroxide has been considered a means of facilitating the oxidation of organic contaminants by providing a source of 0 2 which would enhance bioremediatio~ a subject reviewed by Pardieck et 8 al. ( 1992). A vast array of subsurface organic contaminants are completely degraded by soil microbes under aerobic conditions; some examples include benzene, toluene, xylenes and alkyl benzenes from gasoline or solvent spills, components of diesel or heating oil such as naphthalene and other polynuclear aromatic compounds, and synthetic organic compounds, such as chlorobenzene and methylene chloride. The degradation of those compounds with aromatic structures especially requires the presence of 0 2 since ring cleavage occurs via oxygenases that add oxygen atoms to the aromatic ring. Oxidation of compounds which can be degraded under anaerobic conditions tends to be more complete under aerobic conditions because aerobic respiration is more energetically favorable to the microorganisms than the use of oxidants such as nitrate, Mn(IV) or Fe(III) as electron acceptors. Therefore, the rate of in situ microbial degradation is :frequently limited by oxygen availability in the subsurface. Factors which limit the availability ofO2 include its relatively low solubility in water (9.2 mg/Lat 20? C), its slow rate of diffusion through soil solution, and the high biological oxygen demand of the microorganisms. Peroxide is an effective supplier of dissolved oxygen as it is 107 times more soluble in water than 0 2 (Henry's Law constants are 7 .1 x 104 and 1.3 x 10?3 M atm?1 for H2O2 and 0 2 , respectively, Seinfeld, 1986) and it tends to disproportionate readily in soil. The rapid decomposition ofH2O2 in soil could represent an advantage for its use to enhance bioremediation in that it would not persist in the environment at its high application level. It is also inexpensive and available, can be added at high concentrations because of its high solubility in water. Too rapid a rate of dismutation, 9 however, could waste oxidant capacity and produce undissolved oxygen in an aquifer, thereby decreasing its permeability. Also, the cell membrane of a microorganism has little resistance to the transport ofH2O2 across it, and levels within a cell are toxic above 0.1 mM (Schumb et al., 1955), and would have a bactericidal effect. Bioremediation may have limited utility in treating organics that are biorefractory or toxic to microorganisms; resistance to biodegradation (along with high Kow and low volatility) will determine persistence among surface contaminants. Those with a high degree ofhalogenation, like pentachlorophenol (PCP), are slowly biodegraded even under aerobic conditions because of their existing high oxidation state. Biodegradation may also be inhibited at the high concentrations characteristic of spills (Pignatello and Baehr, 1994). If the initial oxidation steps for these compounds could be carried out chemically, the resulting partially oxidized products could become more easily degraded than the parent toxic compounds (Carberry, 1994). Initial, mainly empirical field studies reviewed by Pardieck et al. (1992) on the use of H2O2 to enhance bioremediation appeared promising. In one, an aquifer contaminated with over 270 organic compounds at the site of an old lumbennill showed increases in pollutant degradation after it was injected with 3 mM H2O2? Another showed marked increases in microbial concentrations at the site of an unleaded gasoline spill (although ambient dissolved oxygen levels did not rise), where recirculated groundwater was treated with inorganic nutrients and 15 mM H2O2? More current thinking, however, according to Pardieck (personal communication, 1997), is that the viability of using hydrogen peroxide to enhance IO bioremediation is problematic due to bioinhibitory effects, and will probably depend ultimately on controlling the toxicity ofH2O2 to microorganisms. As a direct oxidant ofrecalcitrant organics via Fenton interactions, however, the use of H2O2 in contaminated soil, combined with applications of reduced iron, has generated a good deal of scientific interest. Using H2O2 in Fenton Remediation A number of recent studies have addressed factors influencing the effectiveness of applying H2O2a s Fenton' s ~agent to oxidize biorefractory organic contaminants in soils. These include biotic interactions such as the degree ofH2O2 decomposition by catalase, formation rates of Off, effects of concentrations and speciation of iron, soil organic matter content and pH range. Zepp et al. (1992) studied the kinetics of OH' production by photolytically-generated Fe(II) and H2O2 over a pH range expected to be found in natural waters, and Watts et al. (1990, 1993, 1999), Tyre et al. (1991), Ravikumar and Gurol (1994), and Pignatello and Baehr (1994) investigated in situ H2O2 treatment using a variety of contaminants and soils. Results of several of these studies are summarized below. Zepp's study addresses the need for an understanding of the formation rates of hydroxyl radicals in natural waters via Fenton interactions. As a source ofFe(II), he used photochemically reduced complexes ofFe(III), a form ofFe{II) which may be involved in a "photo-Fenton reaction" oxidizing agrochemicals in surface waters. The photochemical approach to generating Fe(II) from Fe(III) was selected mainly to avoid 11 the effects of concentration gradients that would occur in the initial stages of the mixing process ifFe2+ were to be directly added to a reaction mixture containing hydrogen peroxide. Hexaaquo Fe2+ would not be expected to be found uncomplexed in soil, therefore the photoreduced complexes also represent a more realistic system. To determine that Off is indeed the reaction intermediary and to determine its rate of formation, Zepp used a kinetic approach with anisole and nitrobenzene as Off probes. Oxalate and citrate were chosen as ligands for the Fe(l11) complexes because they would photoreact efficiently without directly producing other transients that would oxidize the probe compounds, their reactions with Off were slow compared to those of the probe compounds, and they minimized the formation of Fe precipitates at the pH ranges (3-8) used in the study. Hydrogen peroxide was added to the Fe(l11)-ligand solutions, which were then continuously irradiated at 436 nm. A steady state kinetic method was then applied to obtain the rate of generation of the Off radical. The kinetic approach involves using a dilute probe compound under conditions of continuous irradiation in which the reactive transient (Off) will rapidly reach a steady state. Rates of Off formation in both the Fe(Ill) citrate and Fe(Ill) oxalate systems, and across the pH range of3 to 8 gave a one to one correspondence with the measured rate ofFe(II) formation, indicating that the Fe(II) complexes reacted quantitatively with H2O2 across the pH range to produce hydroxyl .radicals. One of the first studies to look at the use ofFenton's reagent in contaminated soils was done by Watts et al. (1990). Their purpose was to follow the degradation of 12 pentachlorophenol (PCP), and determine the optimum pH for soil treatment. Two fine- loamy soils and silica sand were spiked with PCP and mixed in batch systems with 6.5% H2O2 and 480 mg/L FeSO4? H2O2 consumption corresponded to PCP degradation. The reaction was carried out in silica sand at pH 3; such a low pH was required to prevent precipitation of the iron via hydrolysis. Even at that pH, the soluble iron concentrations decreased over the first three hours, and the decomposition rates of PCP and H2O2 also decreased after the first three hours, indicating the importance of soluble iron to the Fenton reaction. Although the rate of PCP degradation in sand was minimal without the addition of iron, PCP decomposition in the natural soil samples proceeded without amendment by iron, probably because of the natural presence of iron oxides which may have dissolved or served as catalysts at the mineral surfaces. Degradation rates did increase upon addition of iron as well. The soil with lower organic content (0.05% vs. 0.58%) showed an overall efficiency of PCP degradation (kpaJk8202 ) about four times greater than that of the soil with higher organic content, probably due to competition by organic matter as an Off scavenger, or perhaps due to the greater activity of catalase and subsequently greater decomposition rate of hydrogen peroxide in the soil containing higher levels of organic matter. A subsequent study by Watts et al. (1993) showed that amending silica sand with goethite (FeOOH), was more effective than adding a soluble Fe (II) salt. The loss of PCP with the goethite system was well descn"bed by a z.ero order expression: -dC/dt = k, where results gave [PCP]= -l.53t + 245 (R2 = 0.97) where the PCP concentration is in mg/Land time is in hours. It is expected that a Fenton system is z.ero order, 13 because the production of Off approaches steady state, which appears to be the case when goethite is used. This is not observed on addition of an Fe(II) salt because of the changing concentration of soluble Fe in the system due to hydrolysis. The data suggest that iron minerals in the presence ofH2O2 are able to catalyze Fenton-like reactions on mineral surfaces. Tyre( 1991) investigated the conditions affecting the relative efficiency of the Fenton treatment of four contaminants: PCP, trifluralin, hexadecane and dieldrin. Soil samples with a range of organic matter content were used, and all four contaminants were added to each soil sample. PCP and trifluralin were degraded faster than hexadecane and dieldrin, probably because the rate constants for OH" attack on dieldrin and hexadecane are slower than for PCP and trifluralin. Also, a higher percentage of trifluralin and PCP were present in the aqueous phase than dieldrin and hexadecane, indicating that preferential partitioning to organic matter by dieldrin and hexadecane may also have affected their oxidation rates. Efficiencies were determined by comparing rate constant (~nlamillan/k8202 ) ratios. Since H2O2 will probably be the primary cost of remediation, the most efficient conditions favor contaminant degradation with minimal H2O2 consumption. As a result, the high iron concentrations that were found to favor contaminant degradation, but also favored H2O2 decomposition, were not necessarily the most efficient. The efficiency ratios were highest in soils with low levels of organic matter, which also did not receive iron amendment. Existing levels of iron minerals in soil appear to contnbute to the efficiency, as well as effectiveness of the Fenton treatment. 14 A 1994 study by Pignatello and Baehr addresses the issue of remediation in a more neutral pH range. All the prior studies in soil were carried out at pH 3 or lower. In application, the need to acidify the soil would make the remediation technology impractical because of the high buffering capacity of soil, and the polluting effects of acidification. Pignatello proposes to circumvent the low pH requirement by using Fe(III) complexes to catalyze the hydrogen peroxide, producing reactive high valent ferryl species (L)Fe,v, instead of, or in addition to OH". ? Metolachlor and 2,4-D were selected as contaminants to give contrasting sorption behavior since metolachlor will sorb much more readily to organic matter than 2,4-D. Of the ligands tested, the best results for contaminant oxidation were obtained using Fe-nitrilotriacetate (NTA) or Fe- hydroxyethyleniminodiacetate (HEIDA) at 0.01 moVkg and H20 2 at greater than 0.5 moVkg. Interestingly, these Fe(III) complexes were much more effective than Fe(II) in combination with hydrogen peroxide. Simple addition of hexaaquo Fe2+ removed 61 % of2,4-D and only 7% of metolachlor, while Fe-NTA removed 99.3% of2,4-D and 87% of metolachlor. Ho ( 1995) published a design for an injection system and a pilot scale test which suggested successful injection ofH20 2 into inaccessible contaminated sandy soil, and found that H20 2 decomposition increased with injection pressure and injection depth. Although the Fenton reagent approach represents a promising combination of biotic and abiotic techniques for contaminant remediation, the reaction mechanisms in heterogeneous soil systems are far from being well understood. Complexation of 15 l ... . contaminants and scavenging of hydroxyl radicals by soil organic matter, the effects of contaminant adsorption and of the adsorption of Fe complexes, the impact of the Fenton treatment on soil microorganisms and their ability to further degrade byproducts of the Fenton degradation process, and the effects of different types of soil are all parts of the complex picture of this remediation strategy that remain to be pieced together. Since the evidence is convincing that Fenton type reactions in soil solution are oxidizing organic pollutants, the effect that the strong oxidants might have on co- contaminants, in particular on reduced chromiwn, becomes a compelling question. Given that both oxidized and reduced iron are common soil constituents, the presence of H20 2 in groundwater, whether natural or anthropogenic, might result in chromium's oxidation and mobilization. We will now turn to a discussion of chromium in the subsurface, its chemistry, and issues relating to its deposition, remediation, and possible oxidation in soils. CHROMIUM IN SOILS Extent of Chromium Contamination in Soils Millions of tons of industrial chromium are processed each year for use in ferrous and non ferrous alloys, pigments, electroplating, corrosion inhibitors, printing inks and refractories (Greenwood and Earnshaw, 1994). Widespread chromium disposal was practiced without discrimination near industrial sites up to the 1970s, resulting in significant pollution of soils and groundwater, and chromium is second only 16 to lead in its incidence of contamination at Superfund sites (U.S. EPA., www.epa.gov/superfund/sites). On December 11, 1980, Congress passed the Superfund Law (Comprehensive Environmental Response Compensation and Liability Act). Along with establishing federal authority to respond to releases of hazardous waste, it taxed the chemical and petroleum industries to set up a fund for the long term remedial treatment of the country's most polluted and hazardous sites. The National Priorities List (NPL) was created to designate those sites considered to be most dangerous to the environment and to public health. It is possible to access the EPA data on the current NPL sites through their website (www.epa.gov/superfund/sites) and search the database by geographical location and form of contamination found at a site. The following charts were compiled in this way to obtain a sense of the relative importance of chromium as a contaminant compared to other metals and organic contaminants, the prevalence of chromium contamination in our region, and the form of chromium contamination found in most sites. As of October 7, 1998, 1192 sites were on the final National Priorities List, 152 federally owned, 1040 privately owned. Of these, 510 were contaminated with chromium. As Table 1-1 shows, many of the sites have multiple contaminants, for example, 799 of the 1192 sites are also contaminated with VOC's, implying that at the very least, 11 7 of these sites are contaminated with both Cr and organic waste. Different kinds of contamination (soil, air, sediment, etc.) are present at most sites. Soil and groundwater contamination account for more than half the contamination media, 17 Table 1-1. Data from the 1192 sites on EPA's National Priorities List Selected Number of Instances of Instances of Contaminants sites on contamination for contamination the final NPL all media for soil and groundwater All 1192 60,405 38,855 metals 746 18,949 11,613 VOC's 799 18,696 13,089 PAH's 538 8,207 5,573 PCB's 302 1,628 865 pesticides 295 3,648 2,372 radioactive 46 781 539 waste nitroarornatics 43 232 179 dioxins/dibenzof 152 541 327 urans chromium 510 1,895 1,210 lead 561 2,477 1,463 arsenic 516 1,869 1,165 mercury 297 857 490 zmc 391 l,308 743 nickel 367 1,010 679 selenium 178 386 227 cobalt 141 350 219 18 and are especially important for metals. Of 1,895 instances of all types of Cr contamination, 1,210 were in soil and groundwater. The different media considered by EPA are shown in the following counts for chromium: Table 1-2. Instances of Cr contamination in different media at the NPL sites Groundwater 576 Debris 79 Soil 634 Surface Water 150 Air 32 Leachate 25 Sediment 145 Sludge 53 Solid Waste 81 Liquid Waste 63 Other 55 Residuals 2 All Media 1895 Geographic distribution of superfund chromium contamination reflects the industrial activity of the northeastern states and, again. the prevalence of soil and groundwater contamination, as seen in Table 1-3. Chromium Soil Chemistry The two prevailing oxidation states of chromium found in soil differ markedly in their chemical behavior: the Cr(VI) oxyanion is a soluble and toxic carcinogen which can cause both skin ulceration and lung cancer (Nriagu and Niebor, 1988), while Cr(III), in contrast, is mainly insoluble in soil, and a required trace element with a daily intake recommended by the NRC of 50-200 ?g (National Research Council, 1990). Therefore, to accurately evaluate the environmental threat posed by the exposure of chromium to H2O2 in contaminated sites, it is important to first examine chromium 19 - Table 1-3. Geographic distribution of chromiwn contamination at Superfund sites. Selected States Total nwnber of Instances of Cr Instances of Cr and Superfund Sites contamination in contamination in Territories all media soil and groundwater Maryland 16 23 17 Delaware 17 15 11 Virginia 25 33 19 New J~rsey 107 232 140 Pennsylvania 98 117 62 West Virginia 6 2 1 Connecticut 14 4 3 Massachusetts 30 64 41 New Hamp. 18 43 25 California 90 93 83 Florida 52 79 55 Texas 30 24 19 Puerto Rico 9 13 11 20 -. speciation and its effects on the metal 's geochemical behavior. The chromium deposition process and geochemistry of a particular site will in turn influence chromium speciation. The environmental behavior of chromium can be generally described by the processes of oxidation-reduction reactions, precipitation-dissolution reactions, and adsorption-desorption exchanges (Palmer and Wittbrodt, 1991 ). The speciation of chromium, however, is a prime factor in determining its precipitation or adsorption, because the two oxidation states behave differently in aqueous solution. Cr(VI) does not strongly adsorb to surfaces, and tends to form oxy- compounds, the tetrahedral HCr04? I CrO42 ? (p~ = 6.4) being the most common in groundwater (Bartlett, 1991 ). At concentrations greater than 0.01 M, and at low pH {<6) it will dimerize to form dichromate (Cr2O/-), although this form is extremely rare in contaminated sites. Its tendency to adsorb to surfaces will increase with decreasing pH, especially in the presence of variably charged oxide minerals. At sites with high Cr(VI) concentrations, its solubility may be controlled by sparingly soluble salts such as CaCr04 (James, 1994). As an oxyanion, chromate is strongly oxidizing at low pH, but will persist in soils at neutral or high pH (J~es, 1996a). Transformation ofCr(VI) to Cr(III) within soils is caused by reduction with ferrous iron in solution, ferrous iron minerals (e.g. biotite and green rusts), reduced sulfur compounds, or soil organic matter (Eary and Rai, 1988). Chromium mobility is therefore significantly reduced in soils in the presence ofFe(II) or organic matter. Cr(III) has 3 d-electrons in a high spin state in readily formed octahedral complexes, although it demonstrates kinetic inertness toward ligand exchange. As a 21 - result, the rate of substitution of waters of hydration is extremely slow (half times in the range of several hours) (Cotton and Wilkinson, 1988). In soil it will form complexes with organic ligands, which increases its solubility. James and Bartlett (1983a) observed that Cr(III) remained in solution in the presence of citric acid and diethylenetriaminepentaacetic acid (DTPA) at pH values greater than 5, where it becomes insoluble in water. Trivalent Cr may be expected to have some mobility and availability for redox interactions at very low pH, at very high pH, or in the presence of high levels of organic matter with which it may form complexes, especially in the process of its reduction from Cr(VI) (Wittbrodt and Palmer, 1996). The pe-pH stability diagram shown in Figure 1-1 depicts the various Cr species that may be present in groundwater and soils. At low pH in its reduced state, as hexaaquochromium (Ill}, chromium shows a strong tendency to adsorb to negatively charged clay surfaces (Cranston and Murray, 1978). As pH approaches 6, Cr(H2O} 3 6 + becomes hydrolyzed, with CrOH2+ more stable than Cr(OH)2+ in an aqueous system (Rai et al., 1987) until it precipitates as Cr(OH)3? The positive species may polymerize through oxo- and hydroxo-bridging, forming dimers, trimers, tetramers and higher weight oligomers in solution in a process similar to the aging process at the surface of a chromium precipitate (Stunzi and Marty, 1983). These multimers remain stable in solution due to the relative inertness of the Cr(III} inner coordination sphere. In the aging process, changes in the chemical structure and composition of chromium(III) hydroxide reduction products take place, which involve hydrolytic polymerization and a concurrent loss of coordinated water or of protons from 22 ? 30,------------------------ region of environmental relevance 20 --------------, I I I Cr3+ "CrOH2+" 0 Cr(OH)J(s) : I I I I I L-------------- __ J H20 -10 H2 0 2 4 6 8 10 12 14 pH Figure 1-1. Stability diagram for aqueous Cr(III) and Cr(VI) species. From Cr data compiled by Ball and Nordstrom, 1998. Activities of aqueous species = 10-4 M, activities of Cr(OH)JCs) and H2O(l) = 1. Po2 = 0.21 atm, PH2 = 10-4 atm 23 coordinated H2O or OH". Spiccia and Marty (1986) have described the fonnation ofan initially crystalline "active" Cr(OH)3?3H2O solid phase which forms on addition of base to aqueous solutions of Cr(H2O)/+. It does not contain any bridging hydroxide ligands; its octahedral units are linked through hydrogen bonds between the OH" and H2O ligands of adjacent Cr(III) centers. In a site where Cr(VI) is discharged and initially reduced under acidic conditions (e.g. discarded chromium plating baths), "active" chromium hydroxide might form. It is, however, thennally unstable, and with time "ages" and becomes an amorphous phase of unknown composition, with an accompanying loss ofreactivity (Spiccia and Marty, 1986). Bartlett (1991) has also noted that freshly precipitated Cr(III) will be oxidized by Mn(III, IV) (hydr)oxides in soil faster than aged materials or well-ordered minerals. Eventually, amorphous oxyhydroxides of Cr(III) in soil will slowly change to an even less reactive and more crystalline a.-Cr2O3 phase. The aged Cr(OH)3 solid form has an extremely low solubility product (~p = 6. 7 x 10"31 ) (DeFilipp~ 1994). When it co-precipitates with iron, as CrxFe1_x{OH)3, the chromium will be even less soluble and less subject to oxidation (Sass and Rai, 1987). Above pH 9 or 10, Cr(III) regains some solubility in its anionic form, as Cr(OH)4?? Deposition of Chromium in Industrial Sites The mineral crocoite (PbCr04) from Siberia was identified by L.N. Vauquelin in I 797, and chromium was isolated a year later by reduction with carbon (Katz and Salem, 1994). Its name derives from the myriad colors of its compounds, trace 24 quantities provide the characteristic color of emeralds and rubies, and today it is mined mainly as chromite (FeO?Cr2O3) in Russia, South Africa and the Phillippines. Chromite is reduced with coke or ferrosilicone in an electric arc furnace for the iron/chromium alloys that are added to stainless steel. For other industrial purposes, chromite is processed via aerial oxidation in molten alkali (NaiCO3 and CaCO3) to give NaiCrO ? 4 The chromate is then leached with water and may be reduced to Cr 20 3 by carbon, and further reduced with aluminum or silicon to obtain the pure chromium metal. Residues from ore processing sites contain high levels of insoluble Cr(III) which was resistant to processing, as well as high residual levels (50 mg/kg or more) of soluble Cr(VI) which was not completely removed in the leaching process. The pH of these sites is from 8- 12, reflecting the alkaline refining process. From 1900 to 1970 in Hudson County, New Jersey, over two million tons of the chromium ore processing residue (COPR) was disposed of and used as general or low land fill (Burke, et al., 1991 ). Chromium electroplating was in great demand during World War II, when dozens of small plating shops set up operations. The process uses chromic acid/sulfuric acid baths, and the washing, dripping and spent plating solutions were often discharged into adjacent wetlands and discharge ponds, creating levels of Cr in the soil that reached as high as 6% (see Table 2-1). The pH of these soils, in contrast to the COPR soils, tend to be low, from 4-5. Beginning in the 1830's, chromium was also used extensively at tannery sites. In the tanning process Cr(VI) was reduced to fonn stable Cr(III) complexes that protected the leather from deterioration. The soils at these sites tend to become depleted of oxygen and form highly reducing environments, due to the 25 quantities of animal organic matter added to them. Although chromium appears to be completely reduced in these sites, it has a surprising degree of mobility in the groundwater, probably due to the enhanced solubility of Cr(III) organic complexes (Davis et al., 1994). The three types of chromium waste sites described above differ, due to the site conditions and deposition process, in soil pH, soil oxidizing or reducing conditions, and chromium speciation as chromate or bichromate, organic or inorganic chromium(III). These conditions are important to consider both when evaluating possible remediation strategies, and when trying to predict the effect of adding remedial reagents like H Q ? 2 2 Chromium Reduction and Oxidation in Soils Other than removal and sequestering, remediation strategies for chromium need to effect the reduction and immobilization ofCr(VI), and the resulting Cr(III) precipitates must not be subject to reoxidation, once the natural aquifer conditions are again obtained. Bioremediation may be a viable method for chromium (Palmer and Puls, 1994), and may proceed directly, as when chromium serves as the terminal electron acceptor for carbohydrate metabolism by a species such as Bacillus subti/is under reducing_c onditions (Melhorn et al., 1994). It may also be an indirect process, where Cr(VI) is reduced by sulfides produced by sulfate reducing bacteria (Suthersan, 1997). Certain organic materials may also be effective reductants of chromium (James, 1996a). One of the major factors affecting the rate and extent of this type of reduction 26 is the presence of mineral surfaces which catalyse the reaction. Deng and Stone ( 1996)_ used goethite and aluminum oxides to investigate the catalytic effect of those surfaces on the reduction ofCr(VI) by low molecular weight organic compounds e.g. glycolic acid, lactic acid, mandelic acid, tartaric acid and their esters. They found that none of the compounds investigated would reduce Cr(VI) (pH 4. 7, reductant/Cr(VI) ratio 10/1) in the absence of catalytic surfaces. The formation of Cr-surface complexes on the minerals (Fendorf, et al., 1997) alters the reactivity of Cr(VI) toward organic compounds, facilitating the formation ofCr(VI) esters with organic materials containing R-OH functional groups. Formation of a chromate ester has been shown to be the preliminary step in the transfer of electrons from a phenolic compound in the reduction of Cr(VI) (Elovitz and Fish, 1995). Both Fe(II) and zero valent iron are capable ofreducing and immobilizing Cr(VI) (Eary and Rai, 1988; James, 1994; Buerge and Hug, 1998, Se~ et al., 1999), and both may be applied in situ, for instance by injecting a reactive barrier with ferrous sulfate, or :filling one with iron filings. The two forms of iron will react with chromium to form different products, and have different effects on site geochemistry. Under alkaline conditions, Fe (II) will reduce chromate to Cr (III), which will hydrolyze and precipitate, or co-precipitate with Fe (Ill): 3Fe2+ + CrO/- + 8W .. cr3+ + 3Fe3+ + 4H20 (1.20 cr3+ + 3H20 ... Cr(OH)3 ! + 3H+ (1.21 3Fe3+ + 9H20 ... Fe(OH)3 ! + 9W ( 1.22 3Fe2++crO/- +8H20 ... 3Fe(OH)3+cr(OH}3 +4H+ (1.23 27 The overall reaction will produce acidity, and injection of ferrous sulfate will be likely to be done in acid media, further enhancing acidity. Large molar excesses of Fe(II) may be necessary to overcome competition for the reduced iron from oxygen. At pH above 6.5-7.0 dissolved oxygen could begin to compete with chromate: (1.24 James (I 994) found Fe(II) treatment more effective than leaf litter, steel wool and lactic acid in removing soluble and exchangeable Cr(VI) from a contaminated alkaline soil with 460 mg/kg total Cr(VI) at a pH about I 0. The presence of organic matter may enhance Fe(II) reduction of chromate. Buerge and Hug (1998) found that chromate reduction was enhanced by the addition ofFe(III) stabilizing ligands such as carboxylates and phenolates, which made the Fe(II) a stronger reductant. A full scale field application of zero valent iron remediation of chromate has been constructed at a plating waste site in Elizabeth City, New Jersey (Power et al., I 995). Elemental iron reduces chromate, and unlike reduction with Fe(II), the process generates alkalinity: (1.25 A much narrower range ofreactions is responsible for the natural oxidation of Cr(III) in soil; before Barlett and James (1979) demonstrated that fresh soils would oxidize up to 15% of added cr3+ by way of indigenous Mn (III,IV) (hdyr)oxides, the oxidation of Cr(III) was not thought to take place at all. More recent studies characterizing the process (Eary and Rai, 1987; Fendorf and l.asoski, 1991; Fendorf et al., 1993; Johnson and Xyla, 1991; Manceau and Charlet, 1992; Silvester et al., 1995) 28 have shown Cr(III) oxidation to be controlled by the oxidation state and morphology of the Mn, and by the transport of dissolved Cr to the oxide surface, either as the hexaaquo cation, Cr(H2O)/+, or as a soluble Cr(III) organic complex. These experiments tended to be run between pH 3-5, where Cr(III) solubility and oxidation rates were greatest. By contrast, a soluble oxidant such as H2O may oxidatively 2 dissolve Cr(III) compounds (Cr2O3, Cr(OH)3, Cr(III)-humates, FeCr O ) under the 2 4 higher pH conditions more commonly found in soils. In order to evaluate the potential threat of chromium at a contaminated site and design appropriate remediation strategies, the tendency for chromium to oxidize under prevailing site conditions needs to be understood. One approach has been to devise a numerical rating scheme to evaluate the need for remediation at a given site (James et al., 1997). The model rates the site based on the form of chromium: ( oxidized or reduced, soluble or insoluble), the soil pH, the presence of manganese oxides and the presence of soil organic matter or other soil components that could reduce Cr(VI). In this way, remediation strategies can be based on a more realistic appraisal of the chromium hazard than a simple measure of chromium concentrations would provide. CHEMISTRY OF CHROMIUM AND PEROXIDE INTERACTIONS H2O2 is thennodynamically capable of both oxidizing and reducing chromium across a broad pH range, as illustrated by the position of the H2O2 /H2O and O/l{ o2 2 reduction lines on a stability diagram for aqueous chromiwn species (Figure 1-2): 29 30.--------------------- ......... ... __H _20 2 ....... .......- ------... __ _ H20 ........ ... _ -- --------...-....-......-...-.. 20 -------- - -10-8 M H20 2 ....-.. ........ .-..... ------......... _.. ... ....... .............-...- ------- 10 Cr3+ CrOH2+ 0 Cr(OH)J(s) Cr(OH)4- -10 4 6 8 10 12 2 14 0 pH Figure 1-2. Stability diagram for aqueous Cr(III) and Cr(VI) species with oxidation and reduction lines for H 0 ? From Cr data compiled by Ball and Nordstrom, 1998? 2 2 H 0 data from Woods and Garrells, 1987. Activities of aqueous species= 10-4 M, ex2ce2p t where noted for H 0 2, activities ofCr(OHMs) and H20(l) = 1. Po2 = 0.21 atm. 2 30 3H20i(aq) + 2CrOH2+ = 2HCr04? + 6H+ (1.26 2HCr04? + 6H+ + 3H20i(aq) = 2CrOH2+ + 6H20 + 30i(g) (1.27 Two H20 2c oncentrations are shown, one at I 0-4 M and one at I o?8 M. At low concentrations, the lines for the oxidation and reduction ofH20 2 will tend to converge; as H20 2 activity increases, the lines on the diagram move apart ( 1 pe unit for each ten- fold increase in [H202 ]), indicating that at higher concentrations being considered for remediation, H20 2 would potentially behave both as a stronger oxidant and as a stronger reductant of chromium. At high H20 2 concentrations, since H20 2 can be both an oxidant and a reductant, it will be out of equilibrium with dissolved Cr, regardless of the Cr oxidation state. Kinetics, not thermodynamics, will control the oxidation state of Cr. As H20 2 diminishes, Cr(VI) becomes stable in the presence of H20 , first at 2 high pH, and with further H20 2diminution, at low pH. Figure 1-2 shows that above pH 8.5 Cr(VI) and 10-4 H20 2 could be stable, but below this pH Cr(VI) would oxidiz.e H20 2 to 0 2 ? Throughout the pH range, Cr(III) would reduce H20 2t o water, and this behavior will persist to low H20 2 concentrations. Cr{Vl)/112O2 Interactions The chemistry ofCr(VI) and H20 2 has been studied for decades, and is particularly complex in the 4-7 pH range relevant to soils. Oscillating behavior, hysteresis, reduction to Cr(V) and Cr(IV) intermediate species, and mono-, di-, tri- and tetra- peroxochromium species have all been reported. Several reviews discuss progress in characterizing the system in studies conducted over the past century 31 (Spitalsky, 1907, 1908; Baxendale, 1952; Brown et al., 1970; Dickman and Pope, 1994; House, 1997). Interest in unraveling its intricacies has historically stemmed from two applications: the synthesis of stable cationic organochromium(III) complexes (House, 1997), and the determination of intermediate species that could be responsible for the toxicity and mutagenicity of chromium (VI) in living cells (Aiyar et al., 199 I; Shi et al., 1999). The chemistry of many of these intermediate species has been approached from the addition of hydrogen peroxide to a solution of chromate (CrO/-) or bichromate (HCrO4-), and has been well reviewed by Brown et al., (1970), Dickman and Pope, (1994), and House (1997). A plethora of possibilities for these species have been ,,t reported, covering a range of chromium oxidations states (II-VI), degrees of substitution by the peroxo ligand, and protonated or deprotonated fonns. An understanding of the Cr(VI)/I-12O2 reaction and its intermediates is further complicated by the catalytic decomposition ofH2O2, which varies with pH and Cr/I-12O2 ratios. The system also varies with temperature and reactant concentrations, and intermediates are unstable and cannot be measured spectroscopically at moderate concentrations. A broad range of reaction conditions, including the use of various buffers and solvents, are reported in the literature and make data comparisons difficult. Regeneration of the Cr(VI) reactants has been reported by dissociation ofCr(VI) peroxo intermediates (Perez-Benito and Arias, 1997), as well as the disproportionation of Cr(V) or Cr(IV) peroxo intermediates (Buxton and Djouter, 1996), further complicating the reaction mechanisms. For example, in weakly acidic solution (pH 2.5-5.5) in an isothermal 32 stirred tank reactor, Beck et al. ( 1991) observed hysteresis and oscillation in the Cr(VI)/H2O2 interaction. As a result, characterization of the intermediate species has mainly been accomplished under more extreme reaction conditions (pH, reactant concentrations) than would be relevant in soil. Nevertheless, well characterized intermediate species of this fascinating system give us important clues as to its possible behavior under more environmentally relevant conditions. Under acidic conditions in the presence of alcohol, HCro - is reduced in a 4 reversible reaction first to a [CrvO(I-~2O) 2 5] + complex, and then to [Cr"(H O) ]2+, which 2 6 in turn, in the presence of 0 2 produces a Cr(III) superoxocomplex, (CrDOi(H O) ]2"? 2 5 (House, 1997). The superoxochromium(III) complex will decompose to produce HCrO4- and the Cr(III) dimer, [(H2O)4Cr(OH)2Cr(OH2) 4]4+, as well as compete with 0 2 to react reversibly with the Cr(II) complex from which it was formed. The reaction of the Cr(II) species with H2O2 is the reaction important for the synthesis of stable Cr(III) alkyl complexes. Cr(II) acts as a Fenton metal with peroxide to produce OH? radicals; these react rapidly with added organic substrates to form alkyl radicals, and they in turn, react with (Cr"(H2O)6]2+t o form the stable Cr(III) alkyl compounds. Without the reducing influence of the alcohol, the addition ofH2O2 to a strongly acidified solution of Cr(VI) results in the rapid formation of a blue ''perchromic acid" (Brown, et al., 1970). It quickly decomposes on standing in aqueous solution, evolving oxygen, partially decomposing excess peroxide, and leaving chromium reduced to the trivalent state. The blue perchromic acid can be stabilized by extraction into a non aqueous solvent such as pyridine. Funahashi et al. ( 1978) 33 propose two-phase kinetics for the overall reaction ( 1.28): rapid formation of the peroxo complex (1.29), followed by its reduction ( 1.30): (1.28 (1.29 (1.30 The equilibrium constant for the formation of the oxodiperoxochromium(VI) complex In a subsequent XAFS study of the blue complex, Inada and Funahashi ( 1997) confirm a pseudo pentagonal pyramidal geometry with an oxo group at the apex, and the two peroxo ligands and a coordinating water molecule making up a five pointed base (Figure 1-3). They found a shortening of the Cr-O (peroxo) bond length relative to that found when the complex was prepared with pyridine (which substitutes for the water ligand), helping to explain the instability or ease of reduction of the Cr(VI) center in aqueous solvent. Above pH 7, the Cr(VI)/H2O2 reaction results in the red-brown anion, tetraperoxochromate(V), [Cr(O2) 4]3-. This complex has a distorted dodecahedral arrangement around a central Cr atom (D2d symmetry), and slow catalytic decomposition of hydrogen peroxide proceeds in its presence in alkaline solutions. Studies done to understand chromium toxicology under near physiological pH conditions may give a better indication of what may be expected in the soil environment. Cr(VI), unlike Cr(III), is readily transported across cell membranes via 34 Figure 1-3. Structure of peroxochromium complexes as described in Dickman and Pope (1994). a) 0 I /OH2 o-Cr \/\~o 0 0 ....... b) violet chromium(VI)oxodiperoxo complex, [CrO(O2)i{OH)]" I/ OH] - o-Cr \/\~o 0 0 c) red-brown tetraperoxochromate(V), [C r( 0 2) 4]3" o_'_-_,f 3 /,.-o'o ] - Cr oQ \':;o 0 0 35 non specific anion pathways, and it is thought that reduction by cellular constituents is necessary for Cr(VI)-induced DNA damage because Cr(VI) does not react directly with isolated DNA (Jennette, 1979; De Flora et al., 1990; Cohen et al., 1993). Compounds found in cells such as ascorbate (Stearns and Wetterhahn, 1994; 1997), and glutathione (Shi and Dalal, 1989; Kortenkamp, 1990) have been shown to reduce Cr(VI) to Cr(V) and Cr(IV), and these complexes of incompletely reduced chromium have come to be considered potentially powerful carcinogens (Zhang and Lay, 1996; Chiu et al., 1998). A Fenton type generation of hydroxyl radicals is thought to proceed from the reaction of the Cr(V, IV) intermediates with H2O2 that may also be present in the cell as a by-product of oxygen cellular metabolism (Shi and Dalal, 1990; Aiyar, et al., 1991; Itoh et al., 1996). If these or similar reduced complexes formed in chromate contaminated soils and were subsequently treated with peroxide, a Fenton type generation of the strongly oxidizing OH? radical could significantly affect the oxidizing capacity of the soil. Zhang and Lay (1998) have recently identified three Cr(V) peroxo complexes using EPR spectroscopy in the pH 4-7 range that form in the presence of Cr(VI) and peroxide alone. By analogy with V(V) chemistry, they identify three degrees of substitution by the peroxo ligand: [Cr0(O2)(OH2)nr in relatively low H2O2 concentrations in low pH, [Cr0(O2)i(OH2)]" in weakly acidic (pH 4-7) and somewhat low H2O2 solutions, and [Cr(O 2 2)lOH)] ? at solutions slightly above neutral. The trend for higher levels of substitution with rising pH continues with the previously identified tetraperoxochromate(V), [Cr(O2)4]3- which was prevalent in alkaline solutions. These 36 species provide evidence for the potential of Fenton interactions in the Cr(VI)/H20 2 system even without the presence of cellular ( or soil) reductants, and all of them decompose H20 2 catalytically. However, the unstable, soluble, violet chromium(VI) oxodiperoxo complex [Cr0(02MOH)l also fonns in the pH 4-7 range, and also decomposes peroxide catalytically (Perez-Benito and Arias, 1997). Its fonnation is favored by the use of phosphate buffers. This complex is the deprotonated form of the blue complex noted earlier, and may decompose peroxide under mildly acidic conditions via the auto reaction of its protonated and deprotonated fonns, as with the decomposition of peroxide at alkaline pH, near its PI 100 400 ? 30 79 ? 0.3 48 ? I 2.0 ? 0.4* 636 ? 2? 4.9 5.2? 261* 2.5Y5/2* (32)tt Aberjona 0-20 1,300 <0.05 < 0.1 2.7 ? 0.8* 568 ? 2? 6.7 154* 4424* I0YR2/2* ? 100 Aberjona 20-40 4,700 <0.05 < 0.1 6.7 ? 0.2? 458 ? I? 5.7 202? 2462* I0YR2/2* ?200 Serpentine 53-75 2,500 <0.05 <0.1 <0.3* 631 ? 2? 6.5 4.2* 212? I0YR5/3** ? 150 t James, 1994. ? Typrin, 1998 (Eh values are corrected for an Ag/AgCI reference electrode; Fe(II) determined colorimetrically with 2,2'-dipyridyl). ?? Rabenhorst, 1982. ? tt Expressed as% of Total Cr(VI) NA - field measurement not available ,,. ...., ......., ................... -- ....... ~ - -- ----- - ... Figure 2-la. X-ray diffraction spectra for COPR soil. For peak identification tables see Appendix A. 7!50 ~ :, 0 0 !500 ~ ~ ~ i ?' 2!50 0 IIJ.-0579> Calcite ? Ca(COJ) J___ I -lo 30 -~ 1, 4b 00 70 2-Theta(") '-'11111..'-.'\r..'-."L.'1.~'\..'-' ,,..._ .....- .,u.,, ........ ._ ... contaminating the leachate. The residue therefore retained high levels of Ca salts, insoluble Cr(III) which had been resistant to processing, and residual levels of sparingly soluble Cr(VI) salts. As pH increases, chromate requires more strongly reducing conditions in the soil for its reduction to Cr(III), (as reflected in the stability diagram, Figure 1-2), and at the high pH of COPR soil, it has persisted for decades (Burke et al., 1991; Weng et al., 1994). Soluble chromate salts wick to the surface and will "bloom" as a bright yellow precipitate during periods of drying and evaporation. An organic, C- rich "meadow mat" underlying the COPR soil may act as a natural, reducing barrier for Cr(VI), and may explain the lack of chromate contamination of the Hackensack River flowing adjacent to the residue sites (James, 1996b ). Soil samples were previously sampled at a historic COPR waste disposal site in Kearney, New Jersey on the flood plain about 600 m from the Hackensack River. Depth to groundwater was 2-3 m Since they have been significantly disturbed by the disposal of industrial waste, the soils have not been mapped, although they may be described as "disused and mixed industrial land." Electroplating Waste Site in Connecticut National Chromium, Inc., a small electroplating facility located near Putnam, Connecticut, discharged wastewater from Cr plating directly into an adjacent wetland from the beginning of the operation in 1939 up to 1975 (Nikolaidis et al., 1994), resulting in high levels of chromium contamination. The chromium electroplating process uses chromic acid/sulfuric acid baths, and the washing, dripping and spent plating 47 solutions were discharged into sewage (where the chromium killed treatment plant bacteria) or into wetlands and discharge ponds. Soil horizon samples were taken from the peat-like surface to the white cla ' yey, glacial till in the wetland soil 50 m downslope from the facility. The uppermost horizon contained the highest total Cr levels of any soil in this study: green chromium(III) hydroxide coatings were evident on fallen branches and plant debris surrounding the site, and samples were measured with as much as 6% total chromium (Table 2-1). The soil is very poorly drained, and its pH, in contrast to the COPR soils, is low, from 4-5. Despite high levels of organic matter (200 g C/kg soil) there are ambient levels of Cr(VI) (60-90 ?M) in the uppennost horizon. The XRD spectra of soil taken from this horizon (Figure 2-1 b) shows quartz, and a Cr(III) rich chlorite that was identifi~d as a possible major phase using JADE. Data identifying the peaks corresponding to these two phases is shown in Appendix A. Another likely fonn of Cr(III) present at the site is an aged, amorphous Cr(OH)3 (s) formed by reduction ofCr(VI) in the plating waste. An XRD spectra would not identify such a non-crystalline phase. The soil underlying the organic rich surface horizons is classified as a Saco silt loam (Soil Survey ofWmdham County, 1981). Chromium behavior in the soil profile is complex, Cr(Vl) disappears in the middle horizons (where soluble Cr(III) can be found, perhaps complexed to organic matter), and it reappears in the glacial till. Mattuck (1994) reported Cr(Vl) levels of up to 950 ?Min groundwater sampled from the underlying aquifer from a well site about 10 m downslope from the electroplating facility. Since no chromate was found in the middle horizons of the soil profile, it is likely that the 48 Figure 2-1 b. X-ray diffraction spectra for Connecticut plating waste soil, 0-14 cm. For peak identification tables see Appendix A. 1000 750 i ::::, 0 (.) ~ 500 ;,:: ".C,' :S ~ \l:) 250 0 l[L 20 L JLO-11 l_i_. - , L. ~ ,...._J Ir I I r-- '40 50 ' 00 ' 2-Thela(") ,,.,..."''"-"'~''' ,,~..._ ......., , ~~"'- chromate in the glacial till was transported to the wetland site through fractured flow from the underlying aquifer. Aberjona Superfund Site The Aberjona watershed near Woburn, Massachusetts was the site of over 100 chrome tanning operations which operated from 1838 to 1988 and also produced glue and grease from carcass residues. Inorganic arsenical and lead-based insecticides were manufactured in the same locale from the 1860s until the 1920s (Davis et al., 1994). In the tanning process, Cr(VI) was reduced with organic acids over the surface of animal hides, forming stable Cr(III) complexes that preserved the leather. Chromium was mainly discharged into lagoons that were used to dispose of the sludge remnants from leather production (U.S. EPA. 1981). Because of the high levels of animal organic waste, natural biodegradation processes depleted the soils at these sites of oxygen and formed highly reducing environments. Along with organic waste materials, the disposal site contains an array of metal co-contaminants, including As and Pb. Two horizons of a riverbank soil classified as Freetown muck soil series were sampled at the Superfund site downstream from the tannery disposal lagoons. Conditions in the subjacent groundwater are conducive to sulfate reduction; the sulfide species produced could be expected to reduce any ambient Cr(VI). Although chromium appears exclusively as Cr(III) in these sites, it has a surprising degree of mobility in the groundwater (Davis et al., 1994 ), probably due to the enhanced solubility of Cr(III) 50 organic complexes (James and Bartlett, I 983a) formed with dissolved organic carbon provided by the decomposing hides. Maryland Serpentine Barrens Unlike the three other soils sampled, the samples taken from the Maryland serpentine barrens contain only naturally elevated levels of Cr. In the Piedmont of the eastern United States, a belt of serpentinite bodies extends from New Jersey to Alabama. These bodies are characterized by hydrated magnesian phyllosilicate minerals that are often also rich in chromiwn. In the early nineteenth century, prospector Isaac Tyson, Jr. associated the low fertility of serpentine soils with the presence of chromite (FeCr20 4) ore. He purchased land across Maryland, including an area northwest of Baltimore known as Soldier's Delight, which became a major source of industrial chromiwn in the I 840s. These soils have been previously studied extensively in an effort to determine the cause of their low fertility (Rabenhorst et al., 1982). Soil was sampled at the Soldier's Delight site, I 00 m from an old chromite mine, by horizon to a depth of 107 cm. The soil is classified as a Typic Hapludalf (fine silty, serpentinitic, mesic) and a detailed description of its properties and morphology is given in Rabenhorst et al. (1982). Those workers reported chromiwn levels as high as 5,850 mg/kg, principally as chromite (FeCr20 4), at a depth of 50-100 cm Samples used in this study were taken from the 53-75 cm horizon, and contained 2,500 mg/kg total chromium, with no detectable soluble chromiwn (Table 2-1 ). The XRD spectra of soil from this horizon (Figure 2-1 c) shows quartz and antigorite, a magnesian serpentine 51 Figure 2-lc. X-ray diffraction spectra for Serpentine soil, 53-75 cm. For peak identification tables see Appendix A. (ssSolcfier's DelighlRAW) Soldiers Delight 53-75 an 2000 1500 'iii' c :::, u0 ,-.;.:; "Ca' 1000 , V, N ~ 500 ow l!t,,wiw ,, ...,J,,,.wW . ?,,,,,.,v hi.,~"" ?WM.J~V1'.11..+,,J'~,~~~U~?""""' TD..1Dt0> Ou,ir17 . Si02 ~ -_J .., 20 L 311 ~ ~ slJ 2-Thela(") ~~~~~'-11A....,._,, ? ,,....._ ....a -..,' ?n."IIIL'-'l.. mineral. The absence of chromite identified in the spectra does not rule out its presence in the soil under levels of 5%. Total chromite would amount to only 0.5% in a sample of this horizon if calculated on the basis of measured Cr. Soil Sampling Laboratory experiments were conducted using seven soil horizons sampled from the above four sites. A similar sampling protocol was followed at each site: an undisturbed area about 1m 2 was cleared of leaf and plant covering, a pit was dug, soil horizons marked and identified, and samples taken from each horizon. Horizons were ,- kept intact as large blocks, sealed in plastic bags, transported to the laboratory in coolers ~ ," and stored in a refrigerator at 4? C. I I It has been shown that drying a soil may cause the breaking up of soil organic ! , ,, - ,, polymers into more easily oxidized fragments (Bartlett and James, 1980). If soluble 'J ~ Cr(VI) is present when such a soil is dried, upon remoistening it may be reduced by the .. ?',,,, ? :fragmented soil organic matter, altering original levels of Cr(VI) (Bartlett, 1991 ). It is therefore important to use samples that have been maintained in field moist conditions when investigating the oxidation or reduction of chromium in soils. Intact blocks of soil from individual horizons were prepared by passing them through a polyethylene sieve using gentle hand pressure to obtain a relatively homogenous :fraction. COPR soil samples were prepared using a 0.40 cm sieve; a 0.25 cm sieve was used on all other soils. 53 Chemicals Analytical grade K2Cr04 aqueous concentrate was obtained from J.T. Balcer and diluted to 19.23 rnM. This was used to prepare Cr(VI) standards and stock solutions. Cr(VI) solutions were titrated in 0.0lM NaNO3 using NaOH or HNO3 to obtain a desired pH. The laboratory preparation of aged, hydrolyzed solutions of Cr(III) is discussed in detail in Chapter 3. Reagent grade 30% H2O2 from Balcer was used without stabilizers to make standards, which were freshly prepared for each series of experiments. Concentrations ofH2O2 stock solutions were verified by titration with KMnO4 using sodium oxalate (NaC2O4 detennined gravimetrically) as a primary standard (Skoog et al., 1994). Catalase prepared from bovine liver was obtained from Sigma (1540 units/mg where one unit will decompose 1.0 ?mol of H2O2 per min). Unless otherwise noted, all other chemicals were reagent grade, obtained from Baker, and used without further purification. Experiments Batch experiments were conducted in triplicate in 50 mL polycarbonate centrifuge tubes. Suspensions containing solution/soil ratios of 10/1 by mass were prepared by placing 3.00 g of soil in each tube and equilibrating for 24 ? 1 hat 25 ?Con an orbital shaker (100 cycles/min, 30 minutes on, 30 minutes off) with 30.0 mL of0.0lM NaNO3? Reactants (H2O2 , Fe(II)) were added to the soil suspensions in small volumes (50 ?L- aliquots) to obtain the desired initial concentration in the soil suspension so as not to 54 significantly dilute the original concentrations of chromium. Destructive sampling was used to monitor [Cr(VI)], [H2O2] and pH of the soil suspensions over time. Samples remained on the orbital shaker until analysis took place, at which time they were centrifuged (12,000 rpm, RCF 14,862, 15 min, 25 ?C), and aliquots withdrawn from the supernatant liquid for determination ofCr(Vi) and H2O2? Analytical Methods Soil solution pH (solution/soil ratio 10/1) was measured with an Orion flat surface combination pH electrode (calibrated using pHydrion buffers at pH 4, 7 and 10) ? i inserted directly into supernatant solution in each centrifuge tube to a depth just above ~ ,? the soil plug. All spectral readings were done with a Shimadzu UV-1601 PC scanning ! I spectrophotometer. Powder X-ray diffraction analysis was performed on air dried, ! ,, , crushed soil samples (see XRD data in Appendix for source specifications). -I J Soluble Cr(VI) in the soil suspensions was measured by the diphenylcarbazide ,i', (DPC) colorimetric method (Bartlett and James, 1979) using 0.5 mL of the DPC reagent (add 0.38 g DPC to 100 mL 95% ethanol and add to 120 mL 85% H3PO4 in 280 mL distilled H2O) with a 4.5 mL-aliquot of the reaction supernatant. In cases where Cr(VI) concentrations fell above the linear standard curve (0.1-40.0 ?M) , 1: 10 or 1: 20 dilutions were made before withdrawing an aliquot to add to the DPC reagent. Diphenylcarbazide reacts with Cr(VI) in acidic solution to form a Cr(IIl)- diphenylcarbazone complex which absorbs as 540 nm (detection limit 0.1 ?M). At concentrations higher than 1o -s M, H2O2 causes a negative interference with the DPC 55 detennination of Cr(VI), because it will competitively reduce Cr(VI) under the low pH conditions of the test. Pettine et al. (1988) found that negative interference ofH2O2 can be avoided at concentrations less than 1o ~ M by increasing reagent concentrations. In these experiments, however, much higher concentrations ofH2O2 were used (up to 0.1 M), and the removal of peroxide by adding catalytic ( 1o -sM ) amounts of catalase to a sample and allowing it to stand 30 minutes prior to analysis was shown to be effective. In those determinations where catalase was used, Cr(VI) standards (0, 1, 5, 10, 20, 30, 40 ?M) were also prepared with 1o -s M catalase. The catalase raised absorbance readings slightly at 540 nm for all standards, but did not appear to affect Cr(VI). Results for standards using DPC and catalase corresponded well with results obtained using a direct optical method (@350 nm for HCr04? at pH 4). Total Cr(VI) in the soil samples was determined using a heated carbonate- '? hydroxide extraction method found to be the most effective of several tested by James et ? I' al. (1995). The measurement included soluble, adsorbed or occluded Cr(VI), and Cr(VI) ',i, bound in a solid phase within the soil matrix. DPC measurements of extracted Cr(VI) were compared to samples prepared with DPC reagent blanks (without DPC). This accounted for slight discoloration of the sample solutions, probably due to the dissolution of fulvic acid under the initially alkaline, and subsequently acidic conditions of the test. Total soluble Cr in the soil was determined by atomic absorption, and total Cr in the soil was determined by atomic absorption following a digestion procedure using 56 H2S04 -H2O2 -HF dissolution of oven dried (105 ?C) crushed (35-mesh) soil samples (Bowman, 1988). A slight modification of the 4-amino antipyrene horseradish peroxidase method ' reviewed by Frew et al. (1983), was used to determine H2O2 concentrations. H Q will 2 2 oxidatively couple with 4-amino antipyrene (AAP) and phenol, in the presence of horseradish peroxidase to produce a quinoneimine dye with a maximwn absorption at 505 nm. The linear range was 1-300 ?M H2O2 (Frew et al., 1983). The modified aqueous reagent was mixed in a 500 mL volwnetric flask as follows: 0.001 0 g horseradish peroxidase (type VI) from Sigma (about 2x10-s M), 0.50 g 4- aminoantipyrene, 1.17 g phenoi 5.0 mL 0. lM triphosphate buffer (pH 6.9), 100 ?L of 0.01M H2O2 (or 2 ?M) to give more stable readings. Reagent (2.0 mL) and aqueous sample (3.0 mL) were vortexed; absorbance readings at 505 nm reached a maximum in 30-120 seconds, after which they were no longer stable. Sample color faded at a rate that varied with H2O2 concentrations. Analytic uncertainties are graphically indicated by the presence of error bars associated with each data point. These are based on reproducibility and are calculated as ? 1 SD. In many cases they are too small to be visible. Correlation coefficients {r2) for standard curves were greater than 0.998 for Cr(VI) methods and greater than 0.995 for ~02 methods. A summary of analytical methods with their associated experimental uncertainties follows in Table 2-2. 57 Table 2-2. Summary of Analytical Methods Analytical Error Linear Range Determination Method HN0 H 0 HF 3% 1 ?M- 120 ?M Total Cr 1. 3, 2 2, digestion, flame AA detennination 2% 0.1 ?M-40 ?M Total Cr(VI) 2? base, carbonate extraction, DPC detemination 2% 0.1 ?M-40 ?M Soluble Cr(VI) in Diphenyl carbazide soil supernatant solutions 3? 1% 1 ?M- 150 ?M Soluble Cr(VI) in Direct absorbance aqueous systems, @350nm pH4-5 2% 1 ?M-300 ?M H2O2 in soils and 4-antiaminopyrene aqueous systems 4? 1. Bowman, 1988. 2. James et al., 1995. 3. Bartlett and James, 1979. 4. Frew et al., 1983. 58 RES UL TS AND DISCUSSION Chromite Ore Processing Residue (COPR) A significant increase in soluble Cr(VI) was shown to take place in samples of the COPR soil treated with single applications of24 mM H20 2 (Figure 2-2), raising 900 ?M ambient chromate levels about 30%. The higher levels of chromate were sustained in the alkaline soil for over a week, while 25% of the peroxide disappeared in the first hour after treatment, and was undetectable (under 0.1 ?M) in the soil within 24 hours after treatment. Lower concentrations of peroxide treatment produced smaller increases in Cr(VI) concentrations (about 10%, from 900 ?M to 1000 ?M) over control samples with no added peroxide, but not systematically, i.e., higher H20 2 levels didn't necessarily produce higher Cr(VI) concentrations. Variability in the data exceeds analytical error (2%, due mainly to the l Oo r 20- fold dilution necessary to measure mM levels of Cr(VI)). Each data point was obtained from three separate subsamples taken from a larger, surface horizon field sample that Was sieved ( 4.0 mm) and thoroughly mixed. Even after sieving, however, COPR soil contains unevenly distributed, small pellets extremely high in chromate. As a result, a wide range of soluble Cr(VI) (856-1020 ?M) was measured in untreated control samples (see Table 2-3). It is probably due to this heterogeneity of the soil mixture that any effects ofH 0 treatments below 24.0 mM were not discernible. 2 2 Although the chromite ore processing residue was subject to efficient leaching methods to remove soluble N3:iCr04, it has continued to release Cr(VI) for decades after its disposal (Burke et al., 1991 ). Sparingly soluble chromate compounds have been 59 1300 1200 ~ 1100 u :i 6 mMH20 2 ~ --! !~ ::1. 1?~; 1000 ::4 I iJ !,~,~~ :~ I~ 900 I I 2 4 8 10 :i,i 0 days Figure _2-2. Changes in Cr(VI) concentrations in COPR soil (10/1 solution/soil by mass) upon smgle applications ofH 0 ? For data see Table 2-3. Error bars are shown for 2 2 each data point as? I SD. Variability in data exceeded expected analytical error of2% due to the heterogeneity of the COPR soil, which contained unevenly distributed, high- chromate pellets. H 0 disappears within 1 day (see Table 2-3). 2 2 60 Table 2-3. Data for Figure 2-2. Cr(VI) concentrations in COPR soil upon single applications ofH O (at day zero). H O disappearance in soil shown for highest 2 2 2 2 application level. Samples measured at 10/1 solution/soil by mass, pH for all samples 8.8 ? 0.1. Cr (VI) (?,M) Day 0.1 1 2 9 H20 2 added (mM) None 896 911 937 960 926 925 978 989 953 856 975 1020 1040 1040 0.75 989 1000 959 1010 1070 1060 980 978 1020 1.50 932 920 1010 1030 983 1040 938 1010 1050 3.00 941 1050 959 1030 929 1020 6.00 921 1020 1020 1050 1090 947 1020 1060 926 998 1010 1030 960 954 975 12.00 914 971 998 990 1030 966 1010 905 923 1130 1350 1180 24.00 971 962 1130 1220 1160 962 11 IO 1130 1150 H 0, (mM) 2 1.5 hour 24 hour H 0 added 2 2 (mM) 9.89 Not detectable 24.00 10.70 9.61 61 shown to be present in the residue at concentrations of between O. 7 to 5%. Among these, CaCrO4 may predominate, and other Cr(VI) compounds found in the ore residue include calcium alwninochromate (3CaO?A12O3? CaCrO4), tribasic calcium chromate [CaJCCrO4) 2], and basic ferric chromate (FeOHCr04)(Gancy and Wamser, 1976). Soluble chromate in COPR soil may be controlled by these solid phase Cr(VI) salts. The relatively high ~P of calcium chromate (7.1 x 10-4) and high levels of calcium (up to 20% as calcium carbonate or.calcium oxides) in the soil from the original lime treatment of the chromite ore suggest CaCr04 as the most likely candidate for controlling Cr(VI) solubility in COPR soils (James, 1994). The 30% increase in soluble Cr(VI) observed by the addition of24.0 mM Ho 2 2 could be explained in two ways. Hydrogen peroxide either oxidized Cr(III), or dissolved Cr(VI) salts. Cr(VI) dissolution by H2O2 would suggest a complexation reaction between Cr(VI) and H2O ? The fonnation of the red-brown anion, 2 tetraperoxochrornate(V) ([Cr(O2)4J3") has been observed in the reaction between H2Q2 and Cr0 ? under alkaline conditions (Dickman and Pope, 1994), although its formation 4 from a sparingly soluble Cr(VI) salt has not been shown. Its 4:1 H2O2:Cr ratio would make its formation highly dependent on H2O2 levels. (Due to the use of catalase in the DPC determination of Cr(VI) under conditions of ambient H202 (> I o-s M), we would not expect to see any decrease in [Cr0 24 ?J due to its complexation with H20 2, because cataJase d e stro ys am b1' e1;1t H2 O 2 ? The catalase would either shift equilibrium conditions hack toward CrO/", or attack the peroxo ligands on the Cr-H20 2 complex directly.) If 62 Cr(VI) salts dissolved through complexation with H2O2, the complex would then release CrO/ into the soil solutions as H2O2 levels dropped. The swift disappearance of the H2O2 in the COPR soil within one day, while enhanced Cr(VI) levels persisted for over a week, also supports the explanation that H202 oxidized Cr(III) components of the COPR soil. Cr(III) in COPR soil has been shown to be resistant to other oxidants: James found that neither the Cr(III) present in COPR soils, nor soluble Cr(III) added to COPR will oxidize when exposed to Mn (III,IV) (hydr)oxides, despite the high pH and low organic matter conditions favorable to sustaining Cr(VI). This implies that peroxide may be uniquely capable of oxidizing chromium in COPR soil. Electroplating Waste Site in Connecticut Within a single soil profile at the Connecticut plating waste site, H2O2 treatments Produced markedly different results. Results from three horizons are reported: the peat- like uppermost (0-14 cm) horizon, with ambient levels of soluble Cr(VI) ranging between 60-90 ?M; the more reducing underlying (14-40 cm) horizon which contained 5-6 ?M soluble Cr(IlI) (Typrin, 1998) probably in the form of soluble organic complexes; and the white and clayey glacial till layer (> 100 cm), with ambient soluble Cr(VI) in the 40-60 ?M range (see Table 2-1). I ncreases m? so1 u bl e Cr(VI) were observed in the uppermost horizon after single applications of peroxide at various concentrations (Figure 2-3). Unlike results in the 63 300 250 - -~ 200 ::1. -~... u 150 6mM I io II ,~ I t OmM I ? I .. 100 I ~ I -, ,I: ~'i I j so-------------------_, I ,ii I I I I I ; 0 5 10 15 20 25 I lflc I ~ I ~ days 1l ,i,,i :I~, Figure 2-3. Changes in Cr (VI) concentrations (I 0/1 soln/soil by mass) in Connecticut wetland plating waste soil (0-14 cm horizon) upon single applications of H20 2 ( at day zero). For data see Table 2-4. Error bars are shown for each data point as ? 1 SD. 64 Table 2-4. Data for Figure 2-3. Changes in Cr(VI) concentrations in Connecticut wetland plating waste soil (0-14 cm horizon) upon single applications ofH O (at day 2 2 zero). Analytical error associated with Cr(VI) determination is ? 2 ?M. Initial [Cr(VI)] for all samples taken as 95 ? 6 ?M (10/1 soln/soil by mass). Solution pH 5.4 ? 0.1 for all samples after day 1. H20 2 not detected day 1-22. Cr (VI) (?M) Day 0.2 1 2 10 22 Peroxide added (mM) None 90.0 92.6 94.4 104 88.7 90.0 94.1 99.1 105 88.9 91.2 94.6 96.8 102 87.5 0.75 100 100 101 101 91.0 101 108 102 101 88.4 104 102 105 105 92.5 1.50 104 105 105 98.6 82.9 101 105 106 99.5 85.5 105 107 103 103 90.7 3.00 108 111 116 104 90.1 110 111 113 104 88.4 111 114 113 107 92.2 6.00 137 130 127 111 91.6 142 130 129 111 92.8 134 126 127 113 93.9 12.00 198 174 159 122 91.6 195 174 165 123 92.5 195 169 159 119 91.3 24.00 270 227 198 120 83.4 271 229 195 116 77.6 272 230 198 119 77.3 65 COPR soil, soluble Cr(VI) increases over ambient chromate levels varied with the concentration ofH20 2 applied, from an increase in Cr(VI) of 30% from a 750 ?M application ofH20 2, to an increase of250% in soluble Cr(VI) after a single application of24 mM H20 2? Increases in chromate levels reached a maximum 4 hours after peroxide applications. As with the COPR soil, two explanations for the increases in Cr(VI) in this horizon are possible: either H20 2 oxidized Cr(III) present in the soil, or it released existing Cr(VI) from the soil matrix. Cr(III) oxidation is suggested by pH changes observed after applying H Q ? 2 2 Increases in Cr(VI) were accompanied by decreases in soil pH (solution/soil ration J 0/J) from original values of 5.5 to as low as 4.6 for the highest peroxide application levels (Figure 2-4). The lowering of pH in these samples may correspond to chromium Oxidation ( equation 1.26). Although the one unit pH change does not account for total Increases in Cr(VI), this is an expected result of the buffer capacity of the soil, further evidenced by the observation that soil pH returned to original levels after a day, while chromate levels still remained high. Within two hours after application, about half of the Peroxide in samples spiked with 24 mM levels had disappeared, and peroxide was not detectable in any sample after one day. Enhanced chromate levels, on the other hand, Persisted Jong after the peroxide disappeared, declining over a two week period at rates that Varied with chromate concentrations, until they reached initial ambient levels of soJubJe Cr(VI). At the highest treatment levels, enhanced soluble Cr(VI) levels (270 ?M) swpassed Cr(VI) levels that would have been present if all forms ofCr(VI) in the horiz.on sample had been released into solution (Table 2-1). This provides further 66 300 pH4.6 ~-2 250 24mMH2o2 - pH5.0 ~ :::1. 200 _, ~3 f 12~H2o2 ' ua -. ' pH5.2 a 150 pH5.3 ~ ----fl. 6mMH ::; 20 2 pH 5.4 I pH5.4 100 a 0 mMH ' 20 2 ~ ~ ~ t-- I I I 50 4 66 9 12 ~ 0 2 ,~ boors Figure 2-4. Short tenn changes in pH and Cr(VI) concentrations (10/1 soln/soil by mass) in Connecticut wetland plating waste soil (0-14 cm horizon) upon single applications of H O ? Solution pH given above Cr(VI) data points. Error bars are shown for each d2ata2 point as ? 1 SD. Analytical uncertainty? 2 ?M for Cr(VI). 67 Table 2-5. Data for Figure 2-4. Short term changes in Cr(VI) concentrations in Connecticut plating waste site (0-14 cm) after single applications ofH20 2 ? Analytical uncertainty for Cr(VI) is ? 2 ?M. 4.6 11.3 Hours 1.3 2.5 H O Cr(VI) (?M) Cr(VI) (?M) Cr(VI) (?M) Cr(VI) (?M) 2 2 added (mM) 6.00 100 135 142 128 129 134 123 95.6 132 137 127 98.0 198 176 12.00 119 187 195 175 116 182 182 195 174 116 249 270 240 24.00 144 247 271 237 139 238 141 249 272 68 evidence that oxidation of Cr(III) by peroxide contributed to the increase in soluble Cr(VI). Cr(III) present in this soil (60 g/kg, Table 2-1) was the result of the reduction, over many years, of large quantities of Cr(VI) in an overland flow of the plating plant discharge to the wetland site. Cr(III) in the soil is therefore relatively newly-reduced as amorphous Cr(III) (hydr)oxides or Cr(IIl)-humates, as opposed to "older" Cr(III) fonns fo und .m so il e.g. Cr20 3 or FeCr20 4? A reducing :fraction of the high levels of organic matter in this horizon could account for the return of soluble chromate to levels observed prior to H20 2 treatment. Wittbrodt and Palmer (1996) observed the reduction ofCr(VI) by soil humic and fulvic acids across a pH range of2-7. Nakayasu, et al. (1999) found that gallic and tannic acids (polyphenols that originate in decaying leaves, likely precursors to humic substances) reduced Cr(VI) at pH 5 at even faster rates than humic and fulvic acids. In the first step of the reduction process, Cr(VI) fonns an organic chromate ester, bonding to an alcohol or an aldehyde functional group on a particulate organic substrate (.Klaning, 1977): (2.2 Although small equilibrium constants for this reaction (Klaning, I 958) caused Kieber and Helz (I 992) to discount its importance in natural waters, equilibrium between soil solution Cr(VI) and Cr(VI) organic esters forming in this peat-like horizon may be a factor in the return of chromate levels in all samples to the 60-90 ?M range. 69 If ambient soluble chromate levels in this horizon were being controlled by a solid phase source of Cr(VI), one would expect to see an equilibrium concentration of chromate approached as solution to soil ratios were raised above the somewhat arbitrary 10: 1 ratio chosen for lab experiments done in centrifuge tubes. Figure 2-5 shows chromate concentrations leveling at about 20 ?M as solution-to-soil ratios were raised from 10: 1 to 80: 1. A solid phase possibility for the control of soluble chromate in this horizon is the iron-chromate precipitate (KFelCr04)iOH\) identified by Baron et al. ( 1996) in an Oregon soil contaminated by chrome plating solutions. Formation of this chromate analog of the sulfate mineraljarosite is consistent with the common occurrence ofjarosite in acid sulfate soils (Wagner et al., 1982). In the case of the chromate mineral, Cr(VI), K, Fe(III) and low pH conditions could all be derived from the discarded plating solutions, although K+ could also be available as a soil exchangeable ? cation, and Fe(III) from native oxyhydroxides. Baron and Palmer (1996) determined a solubility constant for the chromate mineral, where log Ksp = -18.4 ? 0.6 at 25 ?C for the dissolution reaction: Ifit is assumed that [Fe(III)] is controlled by ferrihydrite (log ~P = 4.89, Woods and Garrels, 1987): (2.4 and if the protonation of CrO/- to form HCr04? (log Kt,= 6.4) is also taken into account, the overall reaction becomes: 70 100,-:------------- 75- .... ---?? ???????? ??????? ???????? ?????????????? .. ---- - ---- ?-? -- - 50- ? ? .... ?? ????????? .. ????-? ?? - - -- ... - .. - - ..... - A 25- ??????-? ?? ? ????-? ???? ????????? !J!,??????? . . .. .... ????--?-?-- .... - A 0-;;----::::----:::-----:-::------1.. 0 20 40 60 80 Solution-to-soil ratio Soln/soil ratio 10 20 40 80 Cr(VI) (?M) 93.4 40.0 29.4 20.1 95.7 42.0 28.2 20.4 Figure 2-5. Effect of varied solution to soil ratios on Cr(VI) measurements in the 0-14 cm horizon of the Connecticut plating waste soil. Analytical uncertainty for Cr(VI) is ? 2 ?M. 71 with a calculated K == 5.0 x 10?21 ? If [Cr(VI)] == 20 ?M (solution/soil ratio 80/1 ), and K+ is assumed to be controlled by the chromate jarosite and is therefore 0.5[Cr(VI)], and pH is 5, then an empirical K == 4.0 x 10?20 is obtained, within an order of magnitude of the calculated K. A similar calculation using the less soluble goethite (FeOOH, log ~P = -1.0) would predict 1 M Cr(VI) in solution, clearly far from measured levels. We may therefore consider the chromate jarosite to be a viable possibility as a solid phase controlling ambient Cr(VI) in this horizon, if a relatively unstable Fe phase is also present. High sulfate conditions in plating waste (from H2S04 used in plating solutions) also suggest that solid solutions could form between the ferric chromate salt and a sulfate lattice, and could cause some variation in expected Cr(VI) solubilities. The hypothesis of a solid phase controlling the solubility of Cr(VI) in this horizon also provides an alternate interpretation of the return of soluble Cr(VI) to ambient levels. Instead of being reduced by an active fraction of soil organic matter, the extra HCr0 ? 4 generated by oxidation ofCr(III) may be gradually precipitated by reaction with an excess Fe phase. Since the abundant quantities ofCr(III) oxyhydroxides in this soil have been fonned as the result of the reduction of Cr(VI) in plating waste, it is possible that ?cr(VI) became sequestered within the Cr(III) precipitation matrix during the reduction process. Such "matrix chromate" could be released in any process affecting dissolution of the solid, such as oxidation of surrounding Cr(III). Attempts to reproduce such a phenomenon in the laboratory yielded some evidence for its occurrence: adsorbed chromate that was exchangeable from the surface of a chromium hydroxide precipitate 72 by equilibration in phosphate solution (method of James et al., 1995) was three to four times higher in a system that was prepared with chromate added after the hydroxide precipitation of a Cr(III) salt than in a similar system where precipitation took place in the presence of the chromate. Attempts to measure any Cr(VI) that may have been sequestered in the solid phase of the latter system were unsuccessful due to the oxidation of Cr(III) during the base extraction process (James et al., 1995) designed to release solid phase Cr(VI). Whatever may be controlling ambient levels of soluble Cr(VI) in the uppermost Connecticut horizon, it is clear that chromium in this soil does oxidize in response to treatment with peroxide. Chromium (III) in the adjacent underlying horizon, which has no ambient soluble chromate, was also oxidized by peroxide, but to a lesser extent. Chromate levels reached 5-6 ?M (Figure 2-6), which corresponded to initial levels of soluble Cr(III) in this horizon. If Cr(VI) were being released from a solid phase by H2O2, higher Cr(VI) levels should have been seen in this horizon after H2O2 treatment, since it contained total Cr(VI) levels comparable to the surface horizon (Table 2-1 ). Figure 2-7 shows no significant enhancement of Cr(III) oxidation in this system by the addition of catalytic amounts of Fe(II), and also shows the effect of spiking this soil with I 00 ?M aged, hydrolyzed, Cr(III). An increase in resulting chromate is observed, but it only accounts for about 4-5% of the added Cr(III), indicating that chromium reduction is favored by soil processes in this horizon, which is also observed in the rapid disappearance of chromate after it forms. 73 6 I. u ~ .::1. 2 1 2 3 4 5 days Cr(VI) (?M) days a YI ?2 ?3 0.0 0.0 0.0 0.0 1.0 5.2 5.8 5.2 2.0 0.5 0.5 0.2 5.0 0.0 0.0 0.0 Fi~e 2-6. Cr(IIl) oxidation by a single application of 12.0 mM H20 2 in the 14-40 cm ho?Zon of Connecticut wetland plating waste soil. Error bars are shown for each data P0 mt as? 1 SD. Analytical uncertainty for Cr(VI) is? 0.2 ?M. 74 20,---------=-:;;,-------, JS days day a b Yl Y2 Y3 YI Y2 Y3 0 0.0 0.0 0.0 0.0 0.0 I 12.6 12.6 9.7 9.9 6.2 2 19.6 19.0 0.5 1.0 0.5 4 26.0 26.6 5 1.0 0.7 0.5 day C d YI Y2 Y3 YJ Y2 Y3 0 0.0 0.0 0.0 0.0 0.0 0.0 1 5.9 6.0 5.8 5.2 5.8 5.2 2 0.9 0.6 0.5 0.5 0.5 0.2 5 0.0 0.1 0.1 0.0 0.0 0.0 Figure 2-7. Cr(III) oxidation by H20 2 in 14-40 cm horizon of Connecticut wetJand P~ting waste soil (10/1 solution/soil by mass for all soil samples). a) aqueous control us~g 100 ?Maged, hydrolyzed Cr(II1)(2 OH/I Cr(III)) and 1.00 mM H20 2 b) soil spiked with 100 ?Maged, hydrolyud Cr(III) (2 Off/1 Cr(Ill)) and 12.0 mM H20 2 c) soi] spiked with I.00 ?M Fe(Il) and 12.0 mM H202 d) soil only with 12.0 mM H202. Error bars are shown for each data point as ? 1 SD. Analytical uncertainty for Cr(VJ) is ? 0.2 ?M. 75 Not surprisingly, peroxide took twice as long to disappear in the 14-40 horizon than in the 0-14 horizon, which contained much higher levels of organic matter (Table 2- 6). Reduced chromium may be ''tanning" the organic matter in this horizon, stabilizing it, and rendering it less bioavailable. In their study of the products of Cr(VI) reduction by gallic and tannic acids, Nakayasu et al. (1999) observe the polymerization of the polyphenols during Cr(VI) reduction, and complexation ofCr(III) with the polymerized compounds. This effect may prevent the high organic matter soil from becoming completely anaero hie, perhaps explaining why chromium oxidation occurs to any extent in this poorly drained horizon. Application of 3 mM peroxide to the glacial till horizon produced an entirely different effect (Figure 2-8): an initial rise in pH and a loss of ambient chromate levels, followed by a return of chromate to its initial concentration. A sample spiked with 50 ?M HCr04- showed a similar pattern. Figure 2-9 is taken from data presented in Chapter 3 (Figures 3-5a, 3~5b) from experiments in an aqueous Cr(VI)IH20 2 system with an original pH of 4.5, and shows similar behavior, albeit over a longer time period. The Cr(VI) disappearance and reappearance in this horizon could be explained by Cr(VI) reduction and Cr(III) reoxidation, or by the fonnation of a soluble peroxochromium (VI) complex, which reverts to HCr0 ? as H20 2 levels in the aqueous system drop. 4 The low levels of soil organic matter in this horizon allowed for a direct determination of Cr(VI) by absorbance at 350 nm, rather than via the determination of the DPC derivative. The behavior of a Cr(VI) peroxo complex which could not be observed in the other soils due to the addition of catalase to avoid H20 2 interference in 76 Table 2?6. Disappearance of24.0 mM H202 after a single application in three soils. Soil 1.5 hr 24 hr 48 hr COPR 9.9? 0.3 <0.005 for <0.005 for 10.7 all samples all samples 9.6 Connecticut o. 14 cm 11.5 ? 0.3 2.5 ? 0.3 <0.005 for 11.8 2.8 all samples 11. 7 2.5 Connecticut 1440 cm 18.2?0.3 <0.005 for <0.005 for 18.6 all samples all samples 18.6 Table 2. 7. Cr(VI) (?M) was not detected over the course of one week in two soils treated with single applications of 0.1, 0.05 and 0.025 M H202. Data was the same for all three treatment levels. Cr(VI) ?M lday 2day 4day 7day Aberjona < 0.1 <0.1 <0.1 <0.1 ~S erpentine <0.1 <0.1 <0.1 <0.1 77 5.25 --a 5.00 ::: g. hrs pH 0.0 4.90 4.92 4.93 0.3 5.03 5.03 5.02 0.7 5.16 5.10 5.08 4.75 1.7 5.02 5.01 4.99 3.5 5.02 5.01 5.02 100 2 J 4 5 0 ' 90 hours - 80 ~ .:_:,t 70 s: ua.. 50 -0-b 40 ~c 30 20 4 6 8 10 0 2 boors hours C hours b YI Y2 Y3 ' YI Y2 Y3 0.0 47 47 49 0.0 83 86 84 0.5 40 41 42 0.3 74 79 75 1.1 39 40 41 0.7 72 74 74 2.8 42 44 47 1.7 81 83 79 7.3 53 54 65 61 3.5 76 74 79 24.5 58 59 82 82 50.0 51 52 53 22.5 81 - ~igure 2~8. Short term changes in Cr(VI) eo~rations in g_lacial till underlying a 2 2 onnechcut wetland plating waste soil after a smgle application of3.00 mM H 0 a) pH in soil amended with 50 ?M HCrO. b) Cr(VI) changes in amended soil c) Cr(VI) changes in unamended soil. Analytical uncertainty for Cr(VI) is ? 2 ?M. 78 a) 3000 100 80 ~ 2250 ----- Cr VI (..",' ) c 0 ~ ,-. =0 1500 ~ _; 750 ......-H2O2 0 0 1 2 3 4 5 6 days ! b) 'J ,. 5.5 -~? ~ ~ "-? _,,., =Q. i ~ Ji 4.5 ~ ~ 4.0 5 6 0 1 2 3 4 days Figure 2-9. Reaction of 3000 ?M H 0 2 and 100 ?M HCr04- in aqueous solution of 2 0.0l M NaN0 at initial pH 4.5. a) Changes in H20 2 and HCr04? concentrations 3 b) changes in pH. Data in Figures 3-4a, 3-4b, 3-4c. 79 the DPC test, may possibly be evident here. A violet, diperoxochrornium(VI) complex has been shown to form at pH 4-7 (Funashi, 1978; Perez-Benito and Arias, 1997): (2.6 Determination of its I(,. continues to be problematic due to the complicated behavior of H2O2 in its presence (Zhang and Lay, 1998). Initial concentrations of peroxide (3 mM) in the glacial till took about two days to d~appear in the soil (Figure 2-10). Any peroxochrornium complex formed in the soil would presumably undergo degradation via surface reactions more quickly than in an aqueous system. Aberjona Superfund Site Oxidation of chromium was not observed in the highly reducing environment of the Massachusetts tannery waste site soil, at peroxide application levels of up to O. l M (Table 2-7). Bartlett and James (1979) did observe the oxidation of soluble Cr(III) added to a high Mn sewage sludge and tannery waste, and observed its subsequent reduction over a period of two months. Maryland Serpentine Barrens Samples high in chromium (2,500 ? 150 mg/kg soil) did not produce soluble chromate on treatment with up to 0. l M H2O2 (Table 2-7). Cr(III) has been shown to appear in serpentine soils as chromite (FeCr20 4) (Rabenhorst et al., 1982), and it has been observed that co-precipitation with iron will reduce chromium reactivity and 80 -~-~----~- -----,;;::. _._ --~ 3000 - 2250 ---itr- soil amended with 50 ?M HCro --~ 4 ::t -o- unamended soil ~ 0 1500 ~ 750 0 0 JO 20 30 40 50 60 hours I ,,I, hours a hours t, b ,.,, YI Y2 Y3 YI Y2 Y3 ~; 0.0 3000 3000 3000 0.0 3000 3000 3000 0.3 2822 2846 0.5 2742 2779 2925 0.7 2745 2546 2642 I.J 2647 2660 2774 1.7 2400 2424 2374 2.8 2485 2469 2475 3.5 2214 2177 2217 7.3 1947 1942 1770 22.5 796 727 756 24.5 727 727 626 48.0 85 107 95 50.0 122 108 75 72.0 25 27 23 121.0 0 0 0 ' Figure 2-1 0. Changes in H O concentrations (?M) in glacial till underlying a Connecticut wetland platin~ ~aste soil after a single application of3.00 mM H20 2? Analytical uncertainty in H O detemrination is 2%. Error bars are shown for each data , 2 2 Pomt as? l SD. 81 ------------------'----- solubility (Sass and Rai, 1987). Samples spiked with I 00 ?M soluble Cr(III) showed only a I% oxidation of chromium when treated with 3 mM peroxide, but the Cr(VI) persisted at I ?M for days showing no decreasing trend (Appendix Figure A-3, A-4). CONCLUSIONS Soils with elevated chromium from four different sites, and even soils from horizons within the sam~ profile responded differently to treatment with H20 ? Soils 2 with high ambient levels of soluble Cr(VI), such as ore processing residues, and high levels of recently reduced Cr (III), such as electroplating waste sites, showed marked ! mcreases in chromate after single applications ofm M peroxide over a 4- IO pH range. j .I Soluble Cr(III), in the form of dissolved organic complexes, as found in the 14-40 cm ,I:' ,?~,. horizon of the plating waste site aJso contributes to the likelihood of Cr(III) oxidation by ,,,, ,,,, peroxide. Anaerobic soil conditions found in the Aberjona tannery site, however, may ,. ~ prevent the oxidation of soluble Cr(III). Chromium (III) present in soil as chromite, as ,1, ?? :~, in the Serpentine Barrens site, a1so appeared to be resistant to peroxide oxidation. The disappearance of soluble Cr(VI) after treatment with peroxide in soils above PH 4 could be due to the formation of soluble Cr(VI) peroxo complexes and should not be assumed to be caused by chromate reduction. Once H20 2 levels dissipate in a soil, soluble Cr(VI) which disappeared upon initial H20 2 treatment could reappear. It should be kept in mind that these experiments were conducted with one application of Peroxide, and most remedial peroxide treatments call for continuous delivery ofp eroxide over many days, and at much higher concentrations than those used in these experiments. 82 The extent and persistence of chromiwn oxidation under such conditions would be much greater, as would be the possibility of forming reactive peroxochromiurn intermediate species. ,, :': ,, ,;,, ,."., ,, ,, ,;,; ,, ,J,I ,, ,1,? :, 83 Chapter 3 Chromium Oxidation, Reduction and Complexation by ,, Hydrogen Peroxide in ' Defined Aqueous Systems jl I! ,/,I ,, ,,. ,",, .,i.,i ,I,f ;, 84 INTRODUCTION Understanding the interaction of chromium and hydrogen peroxide has become relevant to environmental science due to clean-up strategies currently being tested by EPA which use high levels ofH2O2 (U.S. EPA, 1998) to oxidatively remediate hiorefractory organic contaminants in soils. In the context of soil remediation, not only are we compelled to look at the possibility of the oxidation of chromium in soils by hydrogen peroxide to its soluble and toxic hexavalent fonn, we should also consider the fonnation of intermediate species which may be generated in the soil in the presence of peroxide and chromium, and which may be the very species responsible for the toxicity and mutagenicity ofCr(VI) (Kawanishi, et al., 1986; Aiyar et al., 1991; Shi et al., I I I 1999). ,, fl ll The kinetics of Cr(III) oxidation by H2O2 under highly alkaline conditions (pH ,,,, ,. 12) has been reported by Baloga and Earley (1961), and under very low [Cr(III)J (1.9 ,,,. ,,,1.,,? ?M) conditions in artificial seawater (Pettine and Millero, l 990; Pettine et al., l 99 l ). ,,, ,,,,, ,, Shi et al. (1993, 1998) used buffered solutions (pH 3.0, 7.2, and 10.0) to study free radical production in the Cr(III)/H2O2 system, but did not monitor Cr(VI). In this study, Cr(IIl)/H O interactions were examined using aged, aqueous Cr(III) systems in 2 2 order to observe whether Cr(III) would be oxidized by relatively low concentrations of "202 under conditions relevant to soils. Chromium(Vl)/H o interactions were examined to determine conditions under 2 2 Which Cr(VI) could be reduced by H O2 and for evidence of the possible formation of 2 peroxochromium complexes under pH conditions and reactant concentrations that 85 w .,...;::-cecer could be present in the context of soil remediation. If high levels ofH2O2 were added to a contaminated soil containing Cr(III) and Cr(VI) in an alkaline environment ( e.g. COPR soil), the tetraperoxochromate(V), [Cr(O2) 4]3" species could form (Dickman and Pope, 1994). Under more neutral conditions, peroxide has been shown to oxidize Cr(III) to Cr(VI) in a contaminated plating waste site (Chapter 2), and high H O levels 2 2 could produce peroxochromium intermediates, In particular, an intermediate such as the soluble violet chromium(VI) oxodiperoxo complex [CrO(O2)i{OH)]", which forms in the pH 4-7 range (Perez-Benito and Arias, 1997), may be capable of forming under high H2O2 and Cr(VI) conditions in this type of contaminated waste site. Fonnation of chromiumperoxo intermediates in the presence of high levels ofH2O2 could exacerbate ,I, the threat already posed by Cr(VI) in contaminated soils. I: II I: In keeping with our inquiry into the possible oxidation of chromium by peroxide ,", in soils, our aim has been to investigate chromium/peroxide behavior under ,, ," ",, , environmentally relevant conditions, and to any extent possible, find clues that could I"I "" :"?-, help establish intermediate species forming under those conditions. Experiments were conducted without buffers, using moderate reactant levels, and were followed over days to observe long-term outcomes. MATERIALS AND METHODS Chemicals Reagent grade Cr(NO1k9H2O ' NaNO3, NaOH, HN01, FeS04, KMn04, lI2C204 and 30% H2O2 were obtained from J. T. Baker and used without further 86 ???era:~. Purification. Analytical grade aqueous K2CrO4 concentrate was also obtained from Baker, diluted to 19.23 mM and used to prepare Cr(VI) standards and stock. solutions. Preparation of Reactant Solutions Hexaaquochromium(III) will dimerize and undergo further polymeriz.ation via oxo and hydroxo bridging as solution pH is raised to levels commonly found in soils (pH 4-5). Since our inquiry relates to the response ofCr(III) in the environment to peroxide, these experiments were conducted with operationally defined "aged" and "hydrolyzed" solutions of Cr(III). This allowed us to assume consistency in our reactant solutions at a mid range pH (3.8-5.5). Cr(III) stock solutions (500 mL) were I prepared in 0.01 M NaNO3 using Off/Cr ratios of 0, 0.5, 1.0, 1.5, 2.0, 2.5, 2.75, 3.0 ,. /1 It H and 4.0. Hydroxide was added as 0.005 M NaOH dropwise at a rate of2 mL/minjust 1", below the liquid surface, while stirring, in order to avoid high local OH? concentrations. ,,", ,",, Formation of a solid phase was observed only in the 3: 1 and 4: I Off:Cr solutions. "" i, I",, I Blue-green, floe-like particles appeared suspended in these solutions, and settling 1, occurred slowly in the absence of stirring. The solutions were then equilibrated with gentle swirling on an orbital shaker (50 cycles/min) for at least one week, after which time, pH values remained stable for several months (see Day Op H data, Figure 3-1 ). Cr(VI) solutions were titrated in 0.01M NaNO3 using NaOH or HNO3 to obtain a desired pH. Reagent grade 30% H2O2 was used without stabiliz.ers to make standards, which were freshly prepared for each series of experiments. Concentrations ofH202 stock solutions were verified by titration with KMn04 using sodium oxalate 87 so,------------- 40 ~ -:::t 30 ->... 0 20 10 0 2 3 4 0 1 after equilibration 9 and before oxidation' I I = I: 6 1! 0.i 1",,, 3 ,, ;,,; 0 4 2 3 ii 0 1 If If Olf/Cr(III) ratio ,,:'' , pH Cr(VI) ?M 115 days Day0 Day 17 4days 6days 17 days ~- OH../Cr 2days 25.1 ? 3.8 3.1 14.1 0.0 5.6 9.9 25.3 28.7 4.1 3.2 15.4 0.5 6.3 11.0 29.4 33.1 42 3.3 15.1 19.9 1.0 9.4 32.5 35.1 4.3 3.4 24.7 1.5 13.4 20.4 33.I 312 4.4 3.4 27.l 2.0 14.4 21.9 22.5 24.8 4.7 3.8 9.7 15.0 17.6 4.1 2.5 52 7.2 6.6 6.3 5.0 4.0 7.1 5.5 4.5 2.8 3.0 3.3 IO.O 6.9 3.0 1.8 41.0 46.8 42.3 4.0 25.0 39.5 - ~ Figure 3-1. Oxidation ofoine 280 ?MC~) solutions. using 100 ?M H,O,. Solutions ~e prepared and equilibrated with off/ Cr(III) ratios from 0/1 to 4/ I. Extent of :XIdation is shown after 2 days, 6 days, and 17 daY:- AJi:'1Ylical uncertain!Y f~r Cr(VI) is O. t ?M Solution pH values are shown after equiblnallOD but befure oXKlatlOD and after 17 days of oxidation . 88 -========~==-=--- --- --- --~ _,Jj/,~ (N"2C204, detennined gravimetricaily) in 0.1 M H2SO4 as a primary standard (Skoog and West, 1994). In order to avoid possible effects that have been noted from the interaction of buffers with Cr/H2O2 intermediate species (Perez-Benito and Arias, 1997; Beck et al., l991), and in order to observe pH changes in the systems we investigated, solutions Were prepared without the addition of buffers. Experiments Batch experiments were conducted in duplicate in 50 mL polycarbonate centrifuge tubes. Reactants (H2O2 , Fe(II)) were added to the chromiwn stock solutions in small volumes to obtain the desired initial concentration in solution without I l1 ,l 'I diluting the original concentrations of chromiwn. Samples were immediately vortexed I; I I and remained on a bench-top orbital shaker ( 100 cycles/min, 30 minutes on, 30 minutes ',. ? 0 ' ft) at 25 ?C until anaJysis took place. Destructive sampling was used to monitor I' .I I, ,I ? ,? [Cr(VI)], (H2O ] and pH of the reaction mixtures over time. 2 Analytical Methods Solution pH was measured with an Orion flat surface combination pH electrode inserted just below the solution surface in each centrifuge tube. All spectral readings Were done using a Shimadzu UV-160 J PC spectrophotometer. 5 Where peroxide concentrations were low ( < 10? M) the diphenylcarbazide (DPC) colorimetric method (Bartlett and James, 1979) was used to determine soluble 89 Cr(VI) using 0.5 mL of the DPC reagent (add 0.38 g DPC to 100 ~ 95% ethanol and add to 120 mL 85% H3PO4 in 280 mL distilled H2O) with a 4.5 mL-aliquot of the rea cti?o n supernatant. In cases where Cr(VI) concentrations fell above the linear standard curve (0.1-40.0 ?M), l: IO or l :20 dilutions were made before withdrawing an aliquot to add to the DPC reagent. Diphenylcarbazide reacts with Cr(VI) in acidic solution to form a Cr(III)-diphenylcarbazone complex which absorbs at 540 run (detection limit 0.1 ?M). The test is specific for Cr(VI): addition of the DPC reagent to hexaaquo Cr3+ produces no absorbance at 540 nm. Under conditions of [H O J 2 2 greater than 10?5 M, soluble concentrations ofHCrO4? or CrO/- were determined by direct absorbance measurements oft he reaction mixtures at 350 run ( E = l 600 cm?1 1 M? ) and at 372 nm (e = 4800 cm?' M~1) respectively. Although this method is not as sensitive (detection limit 1 ?M for HCr04? for al cm pathlength) as the diphenylcarbazide (DPC) colorimetric method, it avoids the negative interference of , ;' peroxide with the DPC determination of Cr(VI), which is significant above I 0?5 M , ,,: concentrations of H o (Pettine et al, 1988). In regions of overlap, where [H2O2 ] was 2 2 less than I o-5 M, the DPC and direct optical methods agreed within 0.5%, and correlation coefficients (r2) of standard curves for both methods were greater than 0.9998. Error bars are shown graphically for each measured data point, and represent the range between duplicate sample values. A slight modification of the 4-amino antipyrene horseradish peroxidase method reviewed by Frew et al, 0983) was used to determine hydrogen peroxide. Hydrogen J)eroxide will oxidatively couple with 4-aminoantipyrene (AAP) and phenol. in the 90 presence of horseradish peroxidase to produce a quinoneimine dye with a maximum absorbance at 505 nm. The linear range was 5-300 ?M peroxide (Frew et al., 1983). The modified reagent was mixed as follows: 0.0010 g horseradish peroxidase (type.VI) from Sigma (about 2xl o-s M), 0.50 g 4-aminoantipyrene, 1.17 g phenol, 5.0 mL 0. lM triphosphate buffer(pH 6.9), and 100 ?L of0.0lM hydrogen peroxide (or 2 ?M) added to give more stable readings. Reagent (2.0 rnL) and aqueous sample (3.0 mL) Were vortexed; absorbance readings at 505 nm were taken when they reached a maximum (30-120 seconds after mixing), after which time they began to decrease at a rate that depended on the concentration of H20 2 in the sample. RESULTS Cr(IIl)/112O2 Interactions Figure 3-1 shows the response of aged 280 ?M Cr(III) solutions (prepared , I i with OH?Jcr ratios from zero to four) to treatment with 100 ?M H2O2 ? Oxidation of ,,,,' Cr(III) to HCr0 ? (CrO/- at the 4:1 Off/Cr ratio) occurred across the range ofOH?Jcr 4 ~tios. Cr(VI) continued to appear slowly over days. Overall pH values decreased, as shown in Figure 3-1. An increase in Off/Cr ratios resulted in increased Cr(III) oxidation up to a lllaximum at the 2:1 ratio, then decreased Cr(lll) oxidation up to the 3:1 region where floccuJation became visible and oxidation ceased. The greatest oxidation occurred at OIJ-/iC r -- 4 , w h ere, at t he end o f 1 15 days , 71 % of stoichiometric yield of chromate CXJ>ected from 100 ?M H20 2 was obtained. 91 -------- -- ---~ ,...,_...., Initial solution pH strongly affected the production of Cr(VI) and the destruction of H20 2 in the Cr(III)/H20 2 system (Figures 3-2 a-c). Excess peroxide (JOOO ?M) was applied to three different 100 ?M preparations of aged Cr(III): a) Cr(III) titrated with HN03 to pH 3; b) Cr(III) prepared with a 2:1 OH?/Cr ratio at pH 4-75; and c) Cr(III) slowly titrated to pH IO with NaOH. Oxidation ofCr(III) at pH 3 Was not observed, nor was a change in pH measured (Figure 3-2a). H20 decreased 2 slightly after several days. The 2:1 OH?:Cr(III) solution showed a steady increase in Cr(VI) over 15 days (Figure 3-2b) along with catalytic decreases in H20 2, with 46 times as much peroxide used as chromate produced. At an initial pH of 10 (Figure 3. 2c ), initial rates of chromium oxidation were higher than at pH 4. 7, and rates of H 0 2 2 destruction were lower, but still catalytic in nature with about 20 times as much H20 2 destroyed as chromate produced in two weeks. At the end of four weeks, about 92 % of the chromium was oxidiz.ed. When 100 ?M 2: 1 Off/Cr(III) solutions were treated with peroxide levels i I ,?I from 500 ?M to 4500 ?M (Figure 3-3 a-c) a linear relationship (r2 = 0.999) between the amount of Cr(VI) produced in 7 days and the initial H20 2 concentration was seen for the lower (500, 750, 1000 and 1500 ?M) H20 2 levels. Above 1500 ?M initial H202, the pattern became erratic, with less Cr(VI) produced in seven days when using 3000 ?M H o initially, and about the same amount ofCr(VI) produced when using 2 2 4500 ?M H o as when using 1500 ?M H20 2 ? Peroxide disappeared catalytically 2 2 When initial levels were above 1000 ?M (Figure 3-3b), and at seven days, measurable .Peroxide levels in the 3000 and 4500 ?M initial H20 2 solutions have reached 92 - - -- ----------.,,.,...~ 4 :~c Jr--------------- 2 0 J 6 9 12 15 3000 days 100 2250 .....,_H202 80 f r., ~ ., 0 1500 -~ ? a1= 750 ----cr(VI) 20 0 0 3 6 9 12 15 days , I ,,, ? days Cr(Vl) peroxide pH YI Y2 YI Y2 YJ Y2 0.0 0.0 0.0 3000 3000 3.07 3.07 J.O 1.2 0.6 2770 2820 3.06 3.05 2.0 0.0 0.0 2770 2800 3.05 3.06 4.0 0.0 0.0 2830 2800 3.09 3.08 7.0 0.0 0.0 2700 2680 3.06 3.05 28.0 0.0 0.0 2350 2260 3.05 3.06 !igUre 3-2a Effect of initial pH on the reaction of 100 ?M Cr ill) and 3000 ?M Ho 2 2 queo~ Cr(III) prepared with initial pH 3 by titrating with HN03 in 0.01 M NaNQ ? ? ~lyticaJ uncertainty for H 0 2 detennination is 2%, and experimental error for Cr(VI) is 0.5 ?M. 2 93 J - - - ---.---.,._... 5.0 4.5 4.0 3.5 0 J 6 9 12 15 days 100 _,._H20 2 2250 0 f ~Cr(VI) Q ~ 6 ISOO ~ ? -? :e 7SO 20 0 0 3 6 9 . 12 15 I i days I?' I I days Cr(VI) peroxide pH YI Y2 YI Y2 YI Y2 0.0 0.0 0.0 3000 3000 4.73 4.73 1.0 14.9 14.9 2590 2560 4.44 4.44 2.0 21.8 21.8 21 JO 2090 4.39 4.39 4.0 30.J 28.9 1410 1420 4.35 4.36 7.0 40.0 40.7 826 829 4.18 4.17 15.0 58.8 59.4 243 246 4.05 4.05 Figure 3-2b. Effect ofi nitial pH on the ion of 100 ?M Cr (III) and 3000 ?M H o Aqueous, hydro1yz.ed Cr(III) prepared in 0.0 I M_N~ O~ using Off ~r(1II). 22 11 Initial ? ~ll 4? 7s. Analytical uncertainty for H202 determmat10n JS 2%, and expenmental error or Cr(VI) is ? 0.5 ?M. 94 J -- -- ---------.... IO 9 :c C. 8 7 6 0 J 6 9 12 15 days 100 -cr(VI) 2250 0 i' .0. ~ 0 1500 -H20z -s ~ -i:: i: 750 20 I 0 I' I 0 3 6 9 12 15 days days Cr(VI) peroxide pH YJ ?2 YI ?2 YJ ?2 0.0 0.0 0.0 3000 3000 9.67 9.67 1.0 24.0 26.0 2860 2750 7.00 7.0J 2.0 49.0 51.0 2650 2460 6.59 6.62 4.0 69.0 68.2 2490 2680 6.44 6.46 7.0 75.9 74.0 2060 2080 6.43 6.47 28.0 93.4 91.7 467 461 6.23 6.21 Figure 3-2c. Effect of initial pH on the reaction of 100 ?M Cr(III) and 3000 ?M H o Aqueo~ Cr(III) prepared with initial pH 10 by titrating with NaOH in 0.01 M NaNO ~. Analyt1cal Wlcertainty for H2O2 determination is 2%, and experimental error for Cr(VI) is ::t:O.S?M. 95 J - ---------- 70 -0-4500 ?M H20 2 60 -D-3000 ?M H20i -A- 1500 ?M H20 2 -<>- 500 ?M H20i 20 10 O~----,---,---r----.----r---,--------. 0 1 2 3 4 s 6 7 days Cr(VI) (?M) with different initial H2O2 (?M) levels days 500 750 1000 YI Y2 YI Y2 YI Y2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.0 8.2 8.2 10.7 IO.I 12.6 12.6 2.0 12.6 13.3 16.4 16.4 19.6 19.0 4.0 20.2 20.2 24.I 23.4 26.0 26.6 7.0 26.6 27.9 32.3 31.7 36.J 35.5 14.0 44.2 45.3 47.4 47.4 50.6 52.5 days 1500 3000 4500 ?1 Y2 YI ?2 YI ?2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.0 17.6 16.3 14.9 14.9 163 16.3 2.0 28.9 29.6 21.8 21.8 23.9 233 4.0 383 383 JO.I 28.9 383 37.0 7.0 45.7 45.7 40.0 40.7 48,8 48.2 15.0 58.8 59.4 28.0 64.2 64.8 69.9 69.2 ~igure 3-3a Oxidation of I 00 ?M Cr(III) solutions, using varying concentrations of 20 2 ? Aqueous, hydrolyzed Cr(III) solutions were prepared in 0.01 M NaNO3 with 2 OI-r- /1 Cr(ID). Experimental error for Cr(VI) is ? 0.5 ?M. 96 I ---------- 0 ;0--~1--~2----.:3--:4---.-s---,- -~,--- 6 s days days 500 750 1000 Y1 Y2 YI Y2 Y1 Y2 0 500 500 750 750 1000 1000 I 473 476 680 688 868 873 2 441 443 626 615 773 781 4 430 433 568 575 690 700 7 369 353 457 459 526 534 I 14 I 194 195 236 239 263 247 I days 1500 3000 4500 YI Y2 Y1 Y2 YI Y2 0 1500 1500 3000 3000 4500 4500 1 1350 1290 2590 2560 3520 3640 2 1080 1080 21 IO 2090 2390 2480 4 807 796 1410 1420 1350 1390 7 509 526 826 829 732 746 15 243 246 28 56 56 52 52 - i~gtire 3-3b. Disappearance ofH20 2 when different initial concentrations are added to ~?~ aqueous, hydro1yud Cr(III) solutions (2 Off/I Cr(III)) in 0.01 M NaN03? Yhcal uncertainty for H20 2 determination is 2%. 97 I s.o =Q, 4 .S 4.0~-r--..--,-----..--.---.---. 0 2 3 4 S 6 7 2 3 4 5 6 7 days days pH data days a) b) c) YI Y2 YI Y2 YI Y2 0 4.72 4.72 4.72 4.72 4.72 4.72 1 4.61 4.64 4.53 4.53 4.43 4.45 2 4.46 4.48 4.38 4.39 4.33 4.34 4 4.34 4.38 4.30 4.30 4.29 4.27 7 4.25 4.26 4.21 4.20 4.17 4.16 14 4.13 4.14 4.07 4.09 4.04 4.02 days d) e) f) YI Y2 Yl Y2 YI Y2 0 4.69 4.69 4.73 4.73 4.69 4.69 1 4.33 4.38 4.44 4.44 4.48 4.51 2 4.25 4.24 4.39 4.39 4.40 4.41 4 4.19 4.20 4.35 4.36 4.31 4.33 7 4.12 4.13 4.18 4.17 4.17 4.18 15 4.05 4.05 28 3.97 3.98 3.97 3.98 Figure 3-3c. Changes in pH when varying concentrations ofH 2O2 react with 100 ?M .J~ u eous, hydrolyzed Cr(III) ( 2 Off/1 Cr(III)) in 0.01 M NaNO3 a) 500 ?M H O7 2so 2 M H20 2 c) 1000 ?M H20 2 d) 1500 ?M H 20 2 e) 3000 ?M H 20 2 f) 4500 ?M 20 2. 98 I comparable levels (700-800 ?M). Figure 3-3c shows similar behavior in pH changes in these solutions. The lower initial H20 2 levels (up to 1500 ?M) show pH decreasing as the amount ofCr(VI) produced increases. However, at the higher H20 2 levels, pH did not increase as much, nor was there an appreciable difference in its behavior between )OOO ?Mand 4500 ?M initial H20 2 ? Cr(Vl)IH202 Interactions Figures 3-4 a-c show the effect of adding 3000 ?M peroxide to 100 ?M Cr(VI) at different pH's. When the initial pH was 3, Cr(VI) was completely reduced and Ho 2 2 was catalytically destroyed while reduction occurred, after which time it maintained a stable level in solution (Figures 3-4a, 3-4b). The pH of the solution increased (Figure 3-4c) , accounting for a change of 400 ? 20 ?M H+, corresponding stoichiometrically to the reduction ofHCrO/ to Cr(H20)/+ (e~uation 1.28). A different pattern emerged between initial pH's 4 and 5. At pH 4, Cr(VI) initially disappeared, reached a minimum value in 2 hours, and began to gradually increase over days, and approached its original concentration. The same pattern was 0 bserved at pH 4.5 and 5, only with less of an initial reduction in Cr(VI). Above pH 5, Cr(VI) levels did not change when H 0 was applied. The pH of these solutions 2 2 increased as Cr(VI) decreased, and decreased as Cr(VI) recovered (Figure 3--4c). Interestingly, the behavior of H o was also comparable in all three solutions (Figure 3- 2 2 4b), disappearing at a similar rate in the pH 5 solution, where very little of the intern..~Ai"'t e speci.e s was 1c.o nm?n g, as it did in the pH 4 solution, where over half the A??.ft;;U ... 99 ----~,~~-- - ~--- 100 75 i .:._:t -lr-pH3.0 .S._' 50 -0-pH4.0 ur.., -9-pH4.5 25 -0--pHS.O 1 2 3 4 5 6 7 days Cr(VI) (?M) days pH3.0 pH4.0 pH4.5 pH5.0 YI Y2 YJ Y2 YI Y2 YI Y2 0.0 97.0 100.0 100.0 100.0 98.5 97.9 IO0.0 IO0.0 0.2 55.6 56.2 1.0 3.1 3.1 61.7 62.9 83.J 82.5 93.2 94.5 2.0 2.1 1.5 71.6 71.6 86.9 86.9 97.0 96.4 3.0 0.0 0.0 90.2 90.2 4.0 75.8 75.8 99.6 99.6 6.0 0.0 0.0 93J 93J 7.0 79.4 79.4 100.0 IO0.0 15.0 87.2 87.2 99.3 100.6 27.0 0.0 0.0 99.0 97.J Fi&ure 3-4a Effect of initial pH on Cr(VI) behavior in the reaction of 100 ?M Cr (VI) ; 3000 ?M H 0 ? Cr(VI) solutions prepared in 0.01 M NaN03 and titrated with 2 2 OJ or NaOH to set initial pH values. Experimental error for Cr(VI) is? 0.5 ?M. 100 I _...., __ ___ _ -A-pH3.0 2500 --O-pH4.0 i 2000 --9--- pH 4 .5 ~ 0 -0-pHS.O 1500 :? 1000 500 0-::---.---.----r----.---~--.-----. 0 2 3 4 5 6 7 days H202 (?M) - days pH 3.0 pH4.0 . pH 4.5 pH 5.0 Yl Y2 Yl Y2 YI Y2 YI Y2 0.0 3000 3000 3000 3000 3000 3000 3000 3000 0.2 2560 2520 1.0 1300 1240 1720 1700 1400 l4l0 1650 1590 2.0 1220 JJ80 JJ60 1180 1050 1010 1070 1090 3.0 JJ50 Jl90 796 810 4.0 724 729 752 731 6.0 J145 1235 507 496 7.0 456 456 465 472 15.0 166 164 192 197 27.0 1023 1069 86 77 figure 3-4b. Effect ofi nitial pH on H202 behavior in the reaction of 100 ?M Cr(VI) and Nooo ?M H20 2 ? Cr(VI) solutions prepared in 0.~1 M NaN03 and titr~t~ ~h HNO3 or aOH to set initial pH values. Analytical uncertamty for H 20 2 detenrunat1on is 2%. 101 I 5.5 H5.0 pH4.5 = pH4.0 C. 4.0 J.5 pH3.0 J.O 0 2 3 4 5 6 7 days pH days pH3.0 pH4.0 pH4.5 pH5.0 I YJ ?2 YI ?2 ?1 ?2 YJ ?2 0.0 3.02 3.02 4.06 4.06 4.47 4.47 4.90 4.90 0.2 4.58 4.57 1.0 3.24 3.24 4.68 4.69 5.05 5.14 5.31 5.34 2.0 3.24 3.26 4.58 4.58 4.89 4.94 5.25 5.27 3.0 3.27 3.27 4.84 4.87 4.0 4.43 4.49 5.14 5.16 6.0 3.25 3.24 4.74 4.78 7.0 4.35 4.34 5.10 5.08 15.0 4.27 4.28 5.04 4.99 27.0 3.24 3.24 4.55 4.59 - Fi~ 3-4c. pH changes in the reaction of 100 ?M Cr(VI) and 3000 ?M H2O2 ? Cr(VI) SOiutions prepared in 0.0 J M NaN03 and titrated with HN03 or NaOH to set initial pH Values. Analytical uncertainty for H2Q2d etermination is 2%, and experimental error for Cr(VI) is? 0.5 ?M. 102 I --~-i? i,,,., .. ,~UNIV. UI" MD COI IFAI=" DAov Cr(VI) disappeared after one day. The quantity of intermediate species being formed does not appear to affect the catalytic dismutation of peroxide. Different initial levels ofp eroxide applied to 100 ?M Cr(VI) at pH 4 produced the same pattern of initially disappearing, and then recovering Cr(VI) over days, with higher initial H20 2 levels resulting in a greater initial decrease in Cr(VI) (Figure 3-Sa). Long tenn pH changes (Figure 3-Sb) inversely reflected Cr(VI) changes, rising as Cr(VI) decreased and falling as Cr(VI) recovered. Peroxide exhibited complicated, oscillatory behavior over the first minutes of these experiments (Figure 3-6a). Initial H202 oscillations dampened as Cr(VI) concentrations reached their minimum values, and pH reached maximum values, after about 2 hours (Figure 3-6b). Effects of Adding Methanol or Fe(II) Adding 9.0 mM methanol to a Cr(III)/H20 2 system (100 ?M 2:1 OH?:Cr{III) and 3000 ?M H Q ) did not affect the initial rate ofCr(III) oxidation, but did inhibit 2 2 Cr(III) oxidation over time (Figure 3-7a). The extent and rate of H20 2 destruction over days was not altered by the addition of methanol. When added to the Cr(VI)fH 0 system (100 ?M Cr(VI), pH 4, 3000 ?M H20 2 ) (Figure 3-7b), 9.0 mM 2 2 methanol prevented the recovery ofCr(VI) after its initial disappearance. Control experiments using 9_0 mM methanol with either H20 2 or HCrO/ showed no effect on lI202 or HCrO/ concentrations (data shown on Figure 3-7b). Replacing 9.o mM methanol with 3_0 mM methanol produced the same results as using the higher coilCentration of methanol (data shown on ~igureS 3-7 a-b). 103 U~I\I , OF Mn r.n1 1 11:rue ,.. , ... ,~ 100 --<>- 750 ?M H2O2 ~.._., u 60 _,.... I 500 ?M H2O2 --o-- 3000 ?M H20 2 -v-- 4500 ?M H2O2 40 6 9 12 15 0 3 days 4500 3750 =022 50 9 12 15 3 6 ctays Figure 3-Sa Effect of initial H O concentration on its reaction with 100 ?M Cr(VI). 2 2 Cr(VI) solutions prepared at pH 4 in 0.01 M NaNOJ? 104 11 ? __ ,}UNIV. OF MD r::ni I a:,-,.a: ",....,., 5.0 4.5 =C. -:<>- 750 mM H20 2 4.0 -A- 1500 mM H20 2 -0- 3000 mM H20 2 J/1- 4500 mM H20 2 3.5 4 5 6 7 3 0 1 2 days Figure 3-Sb. pH changes in the reaction of different initial H,O, concentrations with I 00 ?M Cr(VI). Cr(Vl) ,olutions prepared at pH 4 in 0.01 M NaNO,. 105 ".,. , _UNIV. OF MD ~nl I IC'r..lC' ,.,.,...,. !~~le 3- 1. Data for Figures 3-Sa and 3-Sb. The reaction of 100 ?M Cr(VI) with varying lnthal concentrations of H 0 ? Analytical uncertainty for H20 2 determination is 2%, and 2 2 experimental error for Cr(Vl) is ? 0.5 ?M. All concentrations are given as ?M. pH days Cr(VI) peroxide YI Y2 YI Y2 YI Y2 a) 750 750 3.97 3.97 0.0 100.0 100.0 4.02 4.02 0.2 95.l 95.4 533 520 4.13 4.14 1.0 89.4 90.0 4.07 90.0 88.8 422 416 4.09 2.0 347 4.18 4.15 355 4.0 88.I 88.8 182 4.14 4.14 ]68 7.0 91.9 91.3 77 4.14 4.15 67 14.0 92.4 93 .6 peroxide pH days Cr(Vl) YI Y2 YI Y2 YI Y2 1500 ]500 3.97 3.97 b) 0.0 100.0 100.0 4.14 4.15 0.2 83.0 82.7 893 4.32 4.29 947 1.0 78.0 78.0 694 4.26 4.25 2.0 80.5 80.5 683 511 518 4.29 4.28 4.0 83.I 82.4 314 319 4.21 4.23 7.0 85.6 85.0 123 4.19 4.16 14.0 91.l 90.5 120 peroxide pH days Cr(VI) Y2 Yl Y2 YI YI Y2 3000 3000 4.06 4.06 c) 0.0 100.0 100.0 2560 2520 0.1 4.58 4.59 0.2 55.6 56.2 1700 4.68 4.69 1.0 61.7 62.9 1720 1180 4.58 4.58 2.0 71.6 71.6 1160 729 4.43 4.49 724 4,0 75.8 ' 75.8 456 4.35 4.34 7.0 79.4 79.4 456 164 4.27 4.28 166 15.0 87.2 87.2 pH peroxide days Cr(Vl) Y2 YI Y2 YI YI Y2 4.03 4.03 d) 4500 4500 0.0 97.2 97.9 0.1 42.9 4.92 4.90 0.2 49.2 49.0 1950 4.78 4.80 1960 1.0 59.8 59.8 4.66 1300 1260 4.65 68.6 2.0 68.6 959 4.53 4.53 75.0 74.4 953 4.41 3.0 539 528 4.44 6.0 80.0 80.0 4.24 63 55 4.23 93.3 27.0 91.4 106 -~-.. ... ' ? t:,, IV. OF MD C(U I i=n.i:: D Iii "" --------~ _,._ H202 control 4500,t,_..-----'~ treated as sample 4000 ?0 3500 ~ :t 3000 2500 2000.------r----..------..-- - - 75 1 -00. -----125 50 0 25 minutes ~~~ 3-6a. Behavior of 4500 ?M and 3000 ?M ll,O, when added to I 00 ?M Cr(VI) at nutiaI pH 4.0 in 0.0 I M NaNO,. Expernnental uncertainty fur ll,O, is ? 40 ?M. 107 ~--?? ???? ~U~l1V'. OF MO Ct"H I Fn~ aanv ~~~~~-~~ 100 ? ~o 1500 ?MH20 2 & Oc, co co 0 0 0 B~o 75 ? 0~ ? 3000 ?MH20 2 ? ~_, ? <> <> ~ <> <> 0 u:r.. ? 50 ? :g ' Do ?? ?? ? ? ? ? ::1. 4500 ?MH20 2 25 0 I I r I 200 300 0 100 minutes 5.0 D D D D DD ? ? D ? 4500 ?MH20 2 DD 0 ? 0 4.5 0 C> ? d> 0 = 0 3000 ?MH202 a? 0 0 c:i. i 0 0 0~ ID t9 ~o oo 1500 ?M H202 4.0 I 3.5 I 300 I r . 200 100 0 minutes Figure 3--lib. Changes in Cr(VI) and pH after addition of difrerelll eoncentrations of H20 2 in 0.01 M NaNO3? 108 t i ... ..... . Y~ IV. OF "1Q C01 I 1=n.1: S:,Aftl,I Table 3-2a. Data for Figures 3-6a and 3-6b. The reaction of l 00 ?M Cr(VI) with 4500 ?M H20 in 0.0IM NaN0 ? Control data is for peroxide alone. All concentrations 3 (?M). A2n alytical uncertainty for H O detennination is 2%, and experimental error for 2 2 Cr(VI) is ? 0.5 ?M. Cr VI pH minutes Peroxide Control minutes ]00.0 4.00 0.0 4500 4509 0.0 1.5 4290 0.5 89.3 4.06 3.8 4280 2.0 81.7 4.07 7.0 4160 2.8 79.8 9.8 4210 3.5 4.09 4.12 12.0 4120 4.0 ]4.4 4310 4.6 74.7 16.8 4400 5.5 4.15 19.4 4280 7.0 4.19 22.0 4090 8.0 69.0 4.24 24.0 . 4000 4570 9.5 27.0 4080 10.0 65.8 4.29 32.0 4220 12.5 35.0 4360 13.5 62.0 4490 4.40 39.4 3980 20.5 45.0 4070 22.0 54.3 4.55 47.8 3930 26.5 50.5 4060 29.0 51.1 4.54 55.5 4020 30.8 3980 4520 59.5 35.0 4.59 62.5 4090 36.0 49.2 65.8 4120 37.0 4.60 71.5 4000 40.0 48.6 75.0 4070 41.5 4.64 81.0 3880 44.5 4.65 84.0 3920 4480 45.3 47.3 87.0 3830 48.0 46.0 92.5 3880 4.71 51.0 4520 99.0 3840 60.0 44.l 4.75 101.5 3850 4.80 68.0 42.9 103.8 3850 4.85 80.0 42.9 123.0 3830 42.2 4.87 94.0 4.91 120.0 42.9 4.88 111.0 43.5 4.91 135.0 42.9 4.95 154.0 42.9 4.91 177.0 42.9 195.0 43.5 4.90 4.92 256.0 45.4 330.0 49.2 4.93 109 ~--J?????? -~?~ . OF MD COllf:(U: DAPI' Table 3-2b. Data for Figures 3-6a and 3-6b. The reaction of 100 ?M Cr(VI) with 3000 ?M H 0 in 0.0 IM NaN0 ? Analytical uncertainty for H20 2 determination is 2%, and 2 2 3 experimental error for Cr(VI) is? 0.5 ?M. All concentrations (?M) . minutes Peroxide Minutes Cr VI pH 0.0 3000 0.0 100.0 4.00 1.8 2920 1.5 91.2 4.oo ? 3.4 3020 2.8 90.0 4.01 5.5 2850 4.1 88.7 7.5 2890 4.5 4.02 86.l 4.03 9.0 3070 6.0 2970 8.0 84.9 I 0.5 8.5 4.05 l 1.8 2910 2810 10.5 83.6 16.4 1 I.0 4.07 20.5 2970 4.08 22.2 3080 13.0 14.0 80.4 27.5 2900 20.6 76.0 33.5 2790 21.0 4.15 38.5 2790 26.8 74.0 48.0 2720 27.0 4.18 50.5 2800 30.0 72.l 4.20 53.0 2740 47.0 66.4 55.5 2790 49.0 4.26 72.0 2640 50.0 65.8 74.0 2720 65.0 4.32 91.0 2630 66.0 63.9 2680 85.0 4.36 93.0 2570 59.4 4.37 120.0 88.0 2570 59.4 4.39 122.0 103.0 157.0 2490 121.0 59.4 4.42 2560 122.0 58.8 4.42 160.0 182.0 2520 4.45 150.0 57.5 53.8 4.52 185.0 2560 203.0 290.0 56.2 4.58 llO ---~, . ' ? ? . ~.'!'IV~ OF MD COi I F~S: DAOV Table 3-2c. Data for Figure 3-6b. The reaction of I 00 ?M Cr(VI) with 1500 ?M H20 2 in 0.0IM NaNO ? Cr(VI) concentrations given in ?M (? 0.5). 3 Minutes Cr VI pH 0.0 100.0 4.00 0.8 96.9 4.00 2.0 96.3 4.00 3.5 95.7 8.0 95.7 4.00 9.0 4.01 10.0 96.3 4.01 17.5 95.1 4.01 37.5 91.9 4.05 46.0 90.6 4.05 47.0 90.0 76.0 88.l 4.06 78.0 86.8 4.06 97.0 87.4 4.07 99.0 87.4 4.08 138.0 85.5 4.12 190.0 83.6 4.13 260.0 83.6 4.14 111 11 1 11 111.u?~~. . OF "lO COUFr.J: DAC:Ut' 3000 100 2500 --.a-- YAth 9 ? 0 mM metha nol ~ 2000 75 ::t 0 N ..:!. =0 15 00 s 0 ,-:: 1000 ~ 500 O~--~----.----r-----.._L 0 1 2 3 4 days ?MCr(VI) Day Cr (HI )/perox Cr(III)/pero/meth L Cr(lll)/pero/meth H YI Y2 YI Y2 YI Y2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.0 19.2 18.6 15.4 14.8 14.8 14.8 2.0 26.7 26.7 18.5 19.1 19.1 19.1 I 4.3 1 42.8 43.5 I 21.4 22.0 j 21.4 21.4 j I Day Cr (III)/perox Cr (IIl/pero/meth L Cr (IIJ/pero/meth H YI Y2 YI Y2 YI Y2 0.0 3000 3000 3000 3000 3000 3000 1.0 2620 2600 2380 2420 2480 2426 2.0 1870 1910 1920 1940 1910 4.3 961 1000 1090 1060 1030 1044 pH Day 1 1 Cr (III)/perox Cr (IJI/pero/meth L Cr (lllpero~meth H ' YJ ?2 YI ?2 YI Y2 0.0 4.46 4.46 4.46 4.46 4.46 4.46 1.0 4.33 4.32 437 4.37 439 439 2.0 4.27 4.28 4.27 4.27 431 430 4.3 4.09 4.09 4.05 4.07 I 4.09 4.09 ~igur~ 3-7a Reaction of l 00 ?M aged, hydrolyzed (~: 1 OH"/Cr) Cr(Ill) and 3000 ?M 202 m 0.01 M NaNO . Data shown for reactions wrth 2 levels of methanol: meth H = d9. O ~ methanoi meth L = 3.0 mM methanol Analyti~ uncertainty for H20 2 etermination is 2%, and experimental error for Cr(VI) 18 ? 0.5 ?M. 112 .ll:H~Jl+ueYNIVU OF MD COUFA~ D.&t:u, 3000 100 2500 -4'- ?with 9.0 mM methanol 75 ~ 2000 ;:j_ N 0 1500 =N .... ..... ... .... 1000 ? -~---.7. - H20z -..... ---..... 25 500 ---.. __ --- O~-----.-----,-----r----,_J.ii 0 2 3 4 days - ?MCr(VI) _ Day Cr(VT)/peroxide Cr(VJ)/perox/meth L Cr(VI)/perox/meth H - Cr (Vl)/meth H - YI Y2 Yl Y2 YI Y2 o.o Yl Y2-100.0 100.0 l00.0 100.0 100.0 100.0 LO 100.0 100--:0 69.3 68.7 59.8 60.4 60.4 60.4 100.4 2.0 100.4 75.4 75.4 59.6 60.2 61.5 61.5 99.4 ..__ 4.3 100.0 83.2 83.2 55.9 55.9 56.5 56.5 100.8 l01.-4 Day Cr (VI)/perox Cr(VI)/perox/meth L Cr(Vl)/peroxlmeth H Yl Y2 Yl Y2 Yl Y2 0.0 3000 3000 3000 3000 3000 3000 1.0 1570 1570 1510 1490 1550 1540 2.0 1040 1040 917 925 945 960 4.3 623 563 333 341 346 357 Day Perox control Perox/meth H Yl Y2 YI Y2 0.0 3000 3000 3000 3000 1.0 2970 3000 2930 2960 2.0 2930 3060 2980 3000 4.3 3050 2990 2920 3040 pH ~ Day Cr (Vl)IH202 I Cr (VI)/H202/meth _L Cr (V0/H202 metb H Cr (Vl)/meth H YI Y2 YI Y2 YI Y2 YI Y2 0.0 4.02 4.02 4.02 4.02 4.02 4.02 4.02 4.02 LO 4.52 4.55 4.42 4.42 4.46 4.47 4.07 4.07 2.0 4.40 4.40 4.26 4.24 4.27 4.28 4.07 4.07 - 4.3 4.29 4.28 4.09 4.11 4.06 4.06 4.01 4.01 - Figure 3-7b. Reaction of J 00 ?M Cr(VI) and 3000 ?M H2O2 in 0.01 M NaNO3? Data shoWn for reactions with 2 levels ofm ethanol: meth H = 9.0 mM methanoi meth L ~ 3.0 ~ methanol. Data also shown for controls with peroxide alone, 9.0 mM methanol Witb Cr(VI) and 9.0 mM methanol with H202 ? Analytical uncertainty for H2O2 determination is 2%, and experimental error for Cr(VI) is? 0.5 ?M. I 13 The addition of small (1 ?M) amounts ofFe(II) to the same Cr(IIl)fH2o2 and Cr(VI)IH202 systems resulted in significantly enhanced Cr(III) oxidation, and Jess Cr(VI) reduction/complex fonnation coupled with quicker Cr(VI) recovery (Figures 3_ 8 a-b) . Not surprisingly, H2O2 dismutation rates increase in both systems. DISCUSSION Cr(III)IH202 Interactions The oxidation of aged solutions of Cr(III) shown across a range of OH"/Cr ratios (Figure 3-1) can be attributed to the addition ofH2O2. Evidence that eliminates 0 2a s a potential oxidant in the system is provided by experiments that showed sparging Cr(Ill) (2: I OH?/cr ratio) reactant solutions for 30 min with N2 before adding H2O2 had no significant effect on oxidation rates (see Appendix Figure A-5). Sparging Cr(III) solutions with 0 2 showed a similarly slight effect when compared to the same experiments done without sparging (Figure 3-2b). One problem inherent in studying the Cr(III)/H2O2 system is in characterizing the initial reactant solutions ofCr(III). Ligand displacement reactions of hexacoordinate Cr(III) complexes are slow, with half times in the range of several hours (Cotton and Wilkinson, l988). In aqueous systems, the resulting kinetic inertness of oxygen in the first Cr(III) coordination shell gives rise to two Cr(III) chenu?s tn ?e s, a fast one an d a s1 o w on e. The fast chemistry involves protonation, deprot ? bo d" m? response to the system pH. whereas the slow onat10n and hydrogen n mg chemi~ ... , . . f covalent OH bridges between Cr(III) atoms, giving ---"J mvolves the fonnat1on o 114 11? ....' .!!'V? OF MD COLI f:AS: DA Dt,, 100 ?--<>-- with 1 ?M Fe (II) 80 i 2000 -::t .n., ..-.. 0 0 s $ -1= 1000 -~ 20 0 0 2 4 6 8 days Cr(VI) (?M), H2O2 (?M) and pH for samples without Fe(II) days Cr(VI) peroxide pH Yl Y2 YJ Y2 YI Y2 0.0 0.0 0.0 3000 3000 4.73 4.73 1.0 14.9 14.9 2594 2560 4.44 4.44 2.0 21.8 21.8 2109 2091 4.39 4.39 4.0 30.1 28.9 1411 1416 4.35 4.36 7.0 40.0 40.7 826 829 4.18 4.17 15.0 58.8 59.4 243 246 4.05 4.05 Cr(Vl) (?M), H2O2 (?M) and pH for samples with 1? M Fe(II) days CrVI peroxide . pH YJ Y2 YI ?2 Yl Y2 0.0 0.0 0.0 3000 3000 4.57 4.57 1.0 27.5 28.1 1997 2102 4.22 4.23 2.0 34.7 35.3 1462 1405 4.21 4.19 5.0 46.9 46.9 793 739 4.14 4.1 3 Figure 3-8a Reaction of 1o o ?M aged, hydrolyzed (2: 1 Off/Cr) Cr(III) and 3000 ?M 8 202 in 0.01 M NaNO with and without the addition of 1.0 ?M FeSO4, Analytical uncertainty for H202 d:termination is 2%, and experimental error for Cr(VI) is ? 0.5 ?M. 115 100 3000 \ .. ??... _..()-______________________ _- I ?? -0 i ?o------- ........._ 80 \.\, /Cr(VI) .n, i 2000 --o-- with 1 ?M Fe (II) 0 - ~ :::1.. \ --i: ~ =0 ', 0 \ -1000 b. \ 20 ???? ?.? - H202 ???o-------------?------o O;-----r-----.--~---.----.L 6 8 0 2 4 days Cr(VI) (?M), H O (?M) and pH for samples without Fe(II) 2 2 peroxide pH I days CrVI Y2 YI Y2 Y2 YI YI 3000 3000 4.06 4.06 0.0 100.0 100.0 1715 1702 4.68 4.69 1.0 61.7 62.9 1180 4.58 4.58 2.0 71.6 71.6 1157 729 4.43 4.49 4.0 75.8 75.8 724 456 456 4.35 4.34 7.0 79.4 79.4 166 164 4.27 4.28 87.2 872 15.0 Cr(VI) (?M), H 0, (?M) and pH for samples with I ?M Fe(Il) 2 pH . ~xide days CrVI Y2 YI Y2 YI YI Y2 3000 4.06 4.06 3000 100.0 100.0 817 4.24 4.24 0.0 86.7 85.4 753 363 4.18 4.18 1.0 91.4 436 4.13 2.0 89.5 4.11 94.0 45 112 5.0 96.5 F~igu re 3-Sb. Reaction of JOO ?M Cr(VI) and 3000 ?M H,0 2 in 0.01 M NaNO, with without the addition of I ?M FeSO,. ,\nlllytical uncertain!Y fur H,O, determination 18 2%, and experimental error for cr(VI) is? 0.5 ?M. 116 ,. 111 .1o ..U N1~. OF MD COLI f:'RF D.tDV rise to oligomers and polymers. Coordinative inertness of ligands surrounding Cr(III) allow a variety of oligomers to persist for weeks to months in solution (Stunzi and Marty, 1983). The "fast" and "slow" Cr(III) chemistries are interdependent: lability of Cr(H20)/+ has been shown to increase upon deprotonation. Xu et al. (1985) measured Water exchange rates for Cr0H(H20) 2 5 + that were 75 times faster than those for Cr(H20)/+ (kex = 2.4 x 10-6 s?1 at 298.15 K). Rate increases of 50-200 fold in the dimeru.ation of Cr(H2O)/+ were observed for each additional deprotonation step CRotzinger et al., 1986). Deprotonation ( or increasing Off/Cr ratios in solution species) therefore corresponds to increased rates of condensation or oligomer formation. Cr(III) oligomers from dimer to hexamer begin to form within minutes in aqueous solutions (Stunzi and Marty, 1983). The particular mixture of oligomers Present in a Cr(III) solution will depend on temperature, time, pH, Cr(III) concentrations, concentrations of added Off, extent of stirring, and whether solutions Were formed from the deprotonation ofCr(H2O)/+ or the protonation ofCr(OH)4? (Spiccia and Marty, 1986; Spiccia et al., 1987; Spiccia et al., 1988). Stunzi and Marty developed a technique using acidification and ion exchange separation ofCr(III) solutions (as well as amorphous Cr(IIl) solids) to identify the fully protonated forms of Cr(III) oligomers: the blue"purple monomer cr3+( aq), the greenish"blue dimer Cr(OH) Cr4+(aq), the green trimer Cr/OH)/+(aq), and the olive tetramer 2 Cr (OH) 6+(aq) I neach case, H o molecules complete the presumed octahedral, six" 4 6 ? 2 coordinate first coordination sphere of each Cr center. Stirring, time, local OH" 117 concentrations and Cr(III) concentrations all increase the extent of higher oligomer formation (Spiccia and Marty, 1986). The tendency of oligomers to coagulate mcreases with increasing pH, and they flocculate as OH"/Cr approaches 3:1, forming an amorphous solid. Above pH 13, a deep green solution forms, preswnably an extensively polymerized "Cr(OH)4 - ", considering the increased Jability of its highly deprotonated form (Spiccia et al., 1988). Characterizing Cr(III) solutions involves distinguishing between "fresh" and "aged" systems, where fast or slow Cr(III) chemistry respectively prevails. In fresh solutions of aqueous Cr(III) the monomer cr3+(a q) successively deprotonates as OH-/Cr ratios increase to form monomers CrOH2+, Cr(OH)2+ , and Cr(OH)/. An ' 'active" Cr(OH)ls) precipitates in fresh systems, and, unlike the amorphous Cr(OH)ls), it does not contain bridging hydroxide ligands. Its units are linked through hydrogen bonds between Off and H20 ligands of adjacent Cr(III) centers, and it Produces only the monomer cr3+(a q) upon acidification (Spiccia and Marty, 1986). Its solubility is significantly higher than that of amorphous Cr(OH)ls). Values of Jog K determined for its dissolution: (3.1 are 8.0 for the "active" precipitate (von Meyenburg et al., 1973) vs. 5. 78 for the amorphous solid (Rai et al., 1987). The active precipitate will revert in time and at ambient temperatures to a more amorphous phase (Spiccia and Marty, 1986), F1. gure h3- 9 s o ws spec.l.a t1'o n diagrams constructed from equilibrium data representing: a) a "fresh" monomeric system (acid dissociation constants taken from 118 a) 1.2 l.O Cr 3+ Cr(OH)2 + - CrOH 2+ Q 0.8 "-' 0.6 u... t$ 0.4 0.2 0.0 -0.2 3 4 5 6. 7 pH .- 10-s ~ "-' a-10u -6 "-' - 10-1 10-s 10-9 3 4 5 6 7 8 pH Figure 3-9. Speciation diagrams for Cr(ITI) in aqueous systems a) representing a "fresh" :r Inonomer ~stem. At ~Cr(ITI) = 10-6 M, ~!111:1tion with respect to active Cr(OH)J(s) not reached m this pH range (based on equilibnu.m data for soluble Cr(III) species !om Stunzi and Marty (1983) and solubility data fo; active Cr(O~J(s) fro~ von }; eyenburg et al., 1973) b) representing an "aged' system or oligomer IIUXture. At Cr(UI) = 10-4 M saturation with respect to amorphous polymeric Cr(OH)JO is reached at PH 4.8 (based on equilibrium data from Rai et al., 1987) ? 119 R0 tz m? ger et al., 1986, and based on Stunzi and Marty (1983) data); and b) an "aged" system containing a mixture of aqueous Cr(III) oligomers in equilibrium with amorphous Cr(III) solids. The "aged" system is based on data provided by Rai et al. (l 987) that has become the basis for much of the Cr thermodynamic data reported in the literature ( e.g. Ball and Nordstrom, l 998). The equilibrium model of Rai and co- workers identified CrOH2+, Cr(OH)/, and Cr(OH)4? as the Cr(III) species which account for Cr(III) solubilities across the pH range from 4-14. Their data indicated that the only significant Cr(III) species found in solution between pH 3-6 was CrOH2+, based on a 2: l slope that resulted from plotting log [Cr(III)J vs. pH in that pH range. They therefore concluded that multimers ( e.g. Cr(OH)2Cr 4+, CrlOH)/+) were not Present in the aqueous phase. Spiccia (I 988) predicted that experimental conditions used by Rai would produce oligorners, and applied his separation technique for 0 ligomers on Cr(III) systems as prepared in Rai et al. ( l 987). He found the aqueous Phase to consist ahnost completely (>98%) of rnultinuclear species, and made the point that Rai's conclusions were based on the charge, not the nuclearity of the Cr(III) species being measured. Since Spiccia's technique for determining oligorners used an acidification step before separation, all the oligorners became protonated. They might have been present in a depr o t ? d /:'. as Cr n (OH) 3n-22 + in Rai's solutions. A structure for onate 10ITl1 Cr (OH) 2+ h as he line ar onfiguration depicted in Figure 3-IO could account for n Jn-2 SUC t C the 2+ har aili"n ? Rai"' system, as well as charges assigned to the multinuclear c ge prev: g m s COmnn els 4+ C (OH) ,-+- Cr (OH)66 +) by Spiccia and co-workers, -o/Vun ( e.g. Cri(OH)2 , r3 4 , 4 120 - OH" Figure 3-10. Proposed pre-equilibrium step to the oxidation of Cr(l11) by H2O2 : a) in a ''fresh" system: Cr(OH)/ + OH- .. Cr(OH)/ b) in an "aged system": Crn(OH)3n_/+ + OH- .. Crn(OH)30_, + Each octahedron represents the inner coordination sphere surrounding a single Cr(ill) atom Matrix points not occupied by an OH ligand are occupied by H2O. 121 Where no n bn ? dg m? g OH groups would be protonated under the acidic conditions in Whi ch they are separated. The 2+ charge on a poJynuclear Cr.(OH)3n_/+ could be expected to be distributed on opposite ends of the molecule, as indicated in Figure 3- JO. The aged Cr(III) solutions used in trus study, prepared while stirring with dropwise addition of dilute NaOH, and equilibrated for at least a week, undoubtably contained a mixture of oligomers, and are best described using the speciation data for an"a ged" system (Figure 3-9). The pH of solutions with Off/Cr ratios up to 2. 75 remained below pH 5.2. Assuming a 2+ charge for the solution species, the low pH at these OH-/Cr ratios also supports the presence of oligomers, because solution pH Would have been much higher if the Cr0H2+ monomer were the predominant species. Some deprotonation appears to be necessary for the oxidation of Cr(III) by fI202, as none occurred at pH 3 (Figure 3-2a) where the cr3+(aq) monomer is the Prevalent species. The rate of oxidation by I 00 ?M H2O2 (Figure 3-1) reached a lllaximum at the 2: 1 Off/Cr ratio, above that ratio Crn(OH)Jn ?( aq) oligomers possibly began to form and flocculate up to the 3: I Off/Cr ratio, impeding oxidation by limiting access to deprotonated OH groups as they l,ecame buried in floe. At OH-/Cr ratios over 3, further deprotonation may have caused increasing rates ofCr(III) oxidation by fI20 2? Oxidation of octahedral Cr(III) to tetrahedral Cr(VI) requires a change in coordination number :from 6 to 4, a change that may be facilitated by the increased ~bili ty that accompanies deprotonat1?o n. Pettine and Millero (1990) determined the rate constant for the oxidation of 122 Cr(III) by H20 2 to be: log k = 8.13-2.17r r for the rate law: -d[Cr(ITI)]/dt = k [Cr(III)][H20 2 ][Off] (3.2 Where the rate is in M/min and Tis in ?C. They used dilute (1.9 ?M) Cr(III) solutions buffered to 7.4-8.5. In this pH range they found a linear relationship between k?213 and aging time of Cr(III) reactant solutions. Their rate constant was determined by extrapolating the line back to zero time, thus correcting for aging effects that slowed the oxidation reaction. We can therefore conclude theirs approximates a "fresh" system as described in Figure 3-9. Using the speciation diagram for the ''fresh" system and the solubility constant for the "active" Cr(OH)ls) (equation 3.1), it is predicted that saturation would not be reached in the fresh system at pH 7.5 at Cr concentrations Used in Pettine and Millero 's experiments. It follows that the predominant species in their system is the monomer Cr(OH)2 +, and their rate law becomes: (3.3 The case may then be made that Cr(OH)3? is the active species in Pettine and Millero's experiments. ? Their rate law can be interpreted as involving a pre-equilibrium step: Cr(OH)2 + (aq) +OH"~ Cr(OH)3 ? (aq) (3.4 Where [Cr(OH) +][OH?] = [Cr(OH) ?]/K. K is calculated from equilibrium constants 2 3 8 2 from Rotzinger et al. ( 1986) and Rai et al., (1987) to be 10 ? , and substitution into Pettine and Millero 's rate law produces: (3.5 Where klK = 10o.39 within reason for a bimolecular mechanism. ' ! __, ,, that a similar mechanism applies to the oxidation I t may be further hypothesu.cu 123 of Cr(lIT) by H2O2 in the aged systems used in the present study. Figure 3-2b shows the oxidation of 100 ?M Cr(III) (at 2:1 Off/Cr) by 3000 ?M H20 ? If average 2 concentrations from day Oto day 1 of the oxidation are applied to Pettine and MiHero 's rate law ( equation 3 .2), a rate is calculated (1.03 x 1o -s M/min) that corresponds exactly to the measured initial rate of oxidation. As we are taking Crn(OH) _/+ to be 30 the predominant species in our 2: 1 Off/Cr system, one interpretation of the correspondence of our ~easured rate to Pettine and Millero's rate law is that oligomers in our system become activated toward oxidation by deprotonation by an OH" at one eod of the molecule, behaving near an active site like CrOH2 +. A diagrammatic comparison of the deprotonation of aged and fresh species is made in Figure 3-10. I P<>stuiate that once a terminal Cr(III) center is deprotonated, it becomes more labile and subject to oxidatio.n by H 0 ? Pettine and Millero 's rate Jaw for our aged system 2 2 then becomes: -d[Cr(III)]/dt = k (1/n) [CrnCOH) 23n-2 +]fH202 ][OH"] (3.6 -Measured and calculated rates diverge somewhat due to the factor 1/n where 1< n<6, and to corrections to thennodynamic data from different sources obtained at various ionic strengths, but still fall within an order of magnitude ofo ne another. In this work , un?n g 'd a t' measured pH values decreased across the range d oXI 100, of OH?/cr ratios (Figure 3-1 ), consistent with the oxidation of Cr(III) by H202 : (3.7 (3.8 3H2O2 + 2CrOH2+ .. 2HCr0/ + m+ (3.9 3H2O + Cri(OH) 4+ .. 2HCr04- + 6W 2 2 124 3H2O2 + Crz(OH)/+ .,. 2HCrO/ + 4H+ + 2H2O (3.10 3H2O2 + 2Cr(OH)2+ .,. 2HCr04- + 4H+ + 2H2O (3. I 1 3H2O2 + 2Cr(OH)4- .,. 2Cr0/" + 2H+ + 6H2O (3.12 Production ofH+r elative to production ofCr(VI) at 17 days exceeds stoichiome t n?c rati?o s. For example at OH"', Cr = O, 25 tu? nes as much H+I? S produced as Cr(VI), compared to the 4:1 ratio expected from equation (3.7), at OH"/Cr == 2, 11 times as much H+i s produced as Cr(VI), compared to the 2: 1 ratios expected from equations (3.10) or (3.11 ). Another process in this system that could generate extra H+i s the further P<>l)'Jneriz.ation of Cr(III) species during oxidation. For example, if, as a result of the Oxidation of the terminal Cr(III) center in an oligomer, a Crn_i(OH)Jn-4 +s pecies was Produced, it could bond with another multimer (e.g.): (3.13 to Produce H+. An additional W for each Cr(VI) produced still does not account for total H+ generated during oxidation. It should be kept in mind that other reactions may complicate these systems as the oxidation ofCr(III) proceeds, including reduction of Cr(VI) by H2o , or the fonnation ofp eroxochromium complexes and their subsequent 2 lllteraction with H2O2 . As OH? levels increase beyond the develop~nt of solid floe at OH?/Cr ratio 3: l, Cr(Ill) becomes soluble once again as ''Cr(OH) R.ai et al. (1987) obtained 4 ?". log K == -18.3 for: (3.14 Cr(OH)ls) + H O .,. Cr(OH)/ (aq) + W 2 125 This would suggest that the concentration of soluble Cr(III) in the initial 4: 1 solution (pH -10) was in the region of 1o -s M, and that the oxidation of Cr(III) by peroxide ( equation 3 .12) enhanced chromium solubility under alkaline conditions. Figures 3-2 a-c show how H2O2 behavior varied markedly with pH in the presence ofCr(III). Peroxide standards prepared at pH 3, 4.5, and 10 in 0.01 M NaN0 3 retained consistent absorbance readings in the course of experiments and showed no catalytic disappearance ofH2O2 in the absence of Cr. Beck et al. (1991) noted no catalytic destruction ofhigh levels ofH2O2 (3M) added to Cr(III) nitrate in "weakly acidic solution," and those results correspond to results in Figure 3-2a, at pH 3, where no catalytic destruction of H2O2 is observed. This suggests that the presence of Cr(VI) plays a critical role in the catalytic destruction ofH2O2 ? The observation that the rate of chromium oxidation in the OH"/Cr 4: 1 system (Figure 3-2c) appears to level off at about one week, while peroxide levels are still high, could be an indication that an intennediate species, such as the tetraperoxochromium(V) complex could be forming initially, contnbuting to the dismutation of the peroxide, and slowly decomposing back to chromate over time. At the end of four weeks, about 92% of the Cr was present in the system as Cr(VI). Similarly, intermediate species such as the violet diperoxochromium(VI) or the chromium(V) peroxo species detected by Zhang and Lay (1998) (see Chapter 1) could be forming in the midrange pH solution, once chromium begins to be oxidized. The fonnation of persistent peroxochromiwn complexes may also explain the changing 126 pattern in the behavior of [Cr(VI)] and pH when levels of H2O2 above 1500 ?Mare applied to Cr(III) (Figure 3-3a). Cr(Vl)/H2O2 Interactions As with the Cr(III)/H2O2 system, the Cr(Vl)/H2O2 interaction depended strongly on pH, and its behavior changed significantly across a relatively narrow pH range. In three systems from pH 4-5 (Figure 3-4 a-c), application ofH2O2 to Cr(VI) initially caused Cr(VI) to disappear, reach a minimum level within hours, then reappear over days. Cr(VI) is either being reduced and reoxidized, or forming a peroxochromium(VI) complex that reverts to HCr04? as H2O2 levels decline. At minimum Cr(VI) levels, the ratio of H+ used to Cr(VI) consumed is 1.5-1.6 for all three solutions, in contrast to the 4: I ratio observed at pH 3 which corresponded to complete reduction of chromium to hexaaquochromium{III): 2HCr04? + 3H20 2 + 8H+.,. 2Cr3+ + 302 + 8H20 (3.15 A closer fit is the reduction to Cro(OH)3n}+: nHCr04? + (3n/2)H20 2 + (n+2)W .,. Crn(OH)3n-2 i+ + (n+2)H2O + (3n/2)O2 (3.16 Formation of polymers as Cr(VI) is reduced consumes W per atom of Cr in the ratio (n+2)/n which has a maximum value of3 for a ?monomer and a minimum value at of 1 for a large polymer. Measured pH changes correspond to n z 4 in this scheme. Cr(III) oligomers could be subsequently reoxidized to Cr(VI), as in equation (3.10). The pe- pH diagram in Chapter I (Figure 1-2) further illustrates how, as H2O2 diminishes and pH increases to arowid pH 5 during reduction of Cr(VI) to Cr(III), Cr(VI) will become 127 stable with respect to H2O2 , while Cr(III) could continue to be oxidized by H2O2 even at low H2O2 concentrations. Using pH changes to consider the formation ofperoxochromium complexes, we observe that the violet peroxochromium complex that has been shown to form in this pH range (Witt and Hayes, 1982; Dickman and Pope, 1994), does not require protons to form: (3.16 Subsequent reduction to the Cr(V) species postulated by Zhang and Lay (1998) would also not require protons: (3.17 However, if this complex lost_one of its peroxo ligands to form the monoperoxochromium(V) species Zhang and Lay suggested would form at low H 20/Cr levels, it would account for two protons: [CrvO{02) 2OH2l + H2O + 2W-= [CrvO(O2)OH2r + H2O2 (3.18 If about half of the hydroperoxyl radical formed in (3.17) were deprotonated: (3.19 (pI ? 2.()952 ? ?? 104 11.2 calcite 1.9799 4.0 0 2 4 47.125 . 0.164 -- - 26 -~:-660 ?1.9853 '75 8.1 Quartz 1.9269 6.4 0 2 4 47.125 -0.281 --?- ? !7 ~-960 . ? 1.9333 44 4. 7 Calcite 1.9269 6.4 O 1 8 47.526 -0.109 - - ~ .~ -7.406 1.9116 18.5 1 6 48.519 . 0.078 - --? 1.9161 122 13.2 Calcite . - ??? ?- -- --- . --~-i.634 ? 1.9075 10 1.6 calcite 1,8747 19,4 30 _4 a?:,w 1 . 1.8Tl'6. . .. ? 139 1s :o Caicite . - - - -- 31 . -~ .'.~4-.. 1.8602? . . 90 9f 32 _49.181 ? Tas1c;- -? -si i.2 : ~_:_405 ~ .!432 ? ? n 1:s1ao 13.0 8.3 o 2 5,4.873 -- ?..o:oos' --?- -- . 50.121 1.8185 - 246 26.5 Quartz 2 3s 53?_13a? ? -:; r-- -?- - . . - 1 o 3 55.323 . 0>110? .. --- 3376 54.941???:/a_: . ?-54~~ ~:~ Quartz ?. 11.667519?27' 42.0 2 1 ?o? ? sr.234 -0.006 38 . 154 1.6639- 32 ?is? "Quartz _.. f ? ? 1 ---c s9.958 ---- 0.051 ? ? ----55 ? ? ? -? 6()83 1.0 --- ------ --- 57.239 - 1:6081 . ?? ?103 1-1.?1 ?auiirtz 5 -9-?0? 2 39 .90T-?1.5429 ? ? 151 rn::r ?auartz? .. ?::~~~~iff ?59 5o. 51?- ?1.5324 - . ?24 2.?.6. - . ... . .. . .. 15249 5.1 3 60 .724 . ??,[5239? . . 42 4.5 caicite . . -? _. .. - 42 -~6 0 7- ?? ?1 --.5239 -? -? 42 4.5 : ..... . -.. ..? __ .. --? ,;-:4728 ???T9 43 62.843 - ?T.4TTS ?8 1 a.7 calcite t41 Table A-2. XRD data from Connecticut plating waste soil. [ssCONN(0-14).RAW] Connecticut aoil, 0-14 cm I Peak ID Report SCAN: 10.0/70.0/0.02/1.2(sec), Cu(40kV,30mA), l(max)=967, 07/22/99 19:30 PEAK: 13-pts/Parabolic Filter, Threshold=3.0, Cutoff=0.1%, BG=3/1.0, Peak-Top=Summit NOTE: Intensity= Counts, 2T(0)=0.0("), Wavelength to Computed-Spacing = 1.54056A(Cu/K-alpha1) # 2-Theta d(A) Intensity 1% Phase ID d(A) 1% h k I 2-Theta Delta 1 10.453 8.4557 54 5. 7 2 12.295 1.1921 75 8.0 ?ch1orite ia 7.1659 100.0 0 6 ..2. ..1.2..3.4.2. .. _0._045_ _ _ 3 12.550 7.0475 -??? 94 10.0? ... 4 --18.167 4.8792 - . 43 4.6 .. ??- 5 18.295 4.8452- 70 7.4 -- . . - .. - - .. ?-?? . -- . ------1 6 18.564 4.7756 ??? 28 3.0 Chlorite la - ?- 4.7773 74.0 0 0 3 18.558 -0.007 7 19.879 4.46:i7. - 40 4.2 Chlorite la ... 4.5130 34:9 -1 1 1 19.655 -0.224 8 20 .779 4.27{2 ?? -291 30.9 Quartz 4 .2550 16.0 1 0 -0 --20.859 0.080 9 20.942 4.2383 158 16.8 Chlorite la -- 4.2407 10.7 1 -1 1 20.930 ?0.012 110 23.622 3. 7632 -- 62 6.6 - . . .. --- --- - -- 11 25.556 3.4827 - -60 6.4 Chloritela - ???- 3.5031 15.2 -1 f - 3 ? ?25.405 -0.151 12 26.658 3.3411 943 100.0 ..Q uartz 3.3435 foo.o --., ?? 0 1 '26.639 -0-.0-19_ __ _ 13 21.493 3.241 6- ?- ? ss 6.2 ?? ??? ?? ??-? ? ?? - - - ---- '14~21.915 3.1935 . - 72 7.6 ? ?- ? -??- - --- 15 28.060 . . 3~1773 .. ?102 10.8 Chiorite la .... - 3.1456 4.6 1 -1 3 28.349 -0.289 - - -? 16 ?- 29.692 ? 3.0063. ? .. 56 5.9 c"t,lorite la ?2.9832 5.0 -1 1 4 29.927 o:fa5 --- -- 17 30.891 2.892j ???so 5.3 ? --- ?? - 18 31.043 2.8785 ------46 4.9 Ciilorite la . - 2.8664 10.1 o o ?5? ??J1.1n?- 0.134 119 34.893 2.5692 .. ??? 49 5.2 Chlorite la 2.5901 11.9 1 -3 1 . 34.603 ?0.290_ __ '20 . 35.087 2:5555 .. 5?7 6.0 Chlorite la . - 2.5572 1.9 -1 1 -5- 35.062_ .. -0.024 21 35~234? Z:5451 56 5.9 Chlorite la 2.5489 3.4 -1 3 2 35.1ao ?- =-a:054 - ?- 22 35~700 2.5129 32 3.4 23 . 36:699 2.4468 58 6.2 Chlorite la . - 2.4476 1.6 1 .3 - 2 36.686 -0.013 24 38.958 2.3fOO 38 4.0 Chlorite la ..... 2.3059 0.3 -2 .i . ?1 - 39.029 0.071 25 39.478 2.2807 1?54 16.3 Quartz ? ?? ? ? -2.2815 8.0 1 0.. 2 39.464--.Q~_-=-01-c4----1 ...... .. ------l 26 39.676 2.2698 70 7.4 Chlorite la 2.2674 8.9 1 -3 3 39.719 0.044 2.7 ?-40241 . 2:2392 36 3.8 28 ... 40.305 2.2358 34 3.6 Quartz 2.2361 4.0 1 1 1 . 40.299 - : 0]06 29 ? 4-2.460 ? 2. -fi ii 89 9.4 Quartz 2.1277 6.0 2 0 0 42.449 -0.011 30 45.602 1.9876 56 5.9 Chlorite la 1.9880 0.9 2 -2 3 45.593 - ~Ci: 009 -- - - --? '31 45.779 1.9804 79 8.4 Quartz 1.9799 4.0 2 0 .. 1? 45.792 0.013 -- 32 48.763 1.8659 ?ss 5.9 dilorite la 1.8720 0.5 0 .z" ? 7 48.596 -0.168 33 50.159 1.a112 ?? 24s 26.3 Quartz 1.8180 13.0 1 1 2 . 50.f:38 .Q.021 - -- ~ --------? . - 34 50.658 1.8005 34 3.6 Chlorite la - . - 1.7993 2.3 0 -4 5 50.696 0.037 ?- '35?~52.778 1.7331 ? ? 150 15.9 Chlorite la? ?- . - ?1.1'318 -3 r -?,r ?5 2.iffg o.041 36 53.261 1.7185 ? - 49 5.2 Chlorite ia - ? ?? ? 1.7250 1 3 6 5f642 . -0.219 - - 37 54.834 - f.illa ? 80 8.5 Clilorite la ? ? ? - ? 1.6739 27 .4 -1 -3 . 7 ?54)95? ??:.0.039 ?-?? -- - 38 55.004 1.6681 69 7.3 Chlorite la 1.6704 15.2 0 ~2 ? -8 . 54.921 - -0.083 .. ?- - -? . ...... - ???-?------ 39 55.004 T.6681 -? 69 7.3 '4o - 59.981 ? 'f.5410 ?fas 13.4 auaitz ?? - ?? 1"].415 9.o 2 1 . 1 . 59.958 .. -0.023- - - > ? -? ?-? -?H - - ?-- --? 41 60.356 1.5323 - 28 3.0 42 - 65.850- - 1.4172 ? - 24 2.5 au'ariz' -??-- .. 1.4184 . :u, 3 "o o 65.784 -0.066 43 - 66.493 1.4050 19 2.0 Chlorite la ?-? '{4049 1.7 -1 -3 9 aa.498 --,ro~o=5- - --1 142 Table A-3. XRD data from Serpentine soil. [ssSoldler's Dellght.RAW] Soldiers Delight 53.75 cm I Peak ID Report SCAN: 10.0/70.0/0.02/1 .2(sec), Cu(40kV,30mA), l(max)=2139. 07/26/99 15:43 PEAK: 17-pts/Parabolic Filter, Threshold=3.0, Cutoff=0.1%, BG=3/1 .0, Peak-Top=Summit NOTE: Intensity = Counts, 2T(0)=0.0(?), Wavelength to Compute d-Spacing = 1.54056A(Cu/K-alpha1) # 2-Theta d(A) Intensity I% Phase ID d(A) I% h k I 2-Theta Delta 1 12.181 7.2598 2129 100.0 Antigorite 6M 7.3000 100.0 O O 1 12.114 -0.067 2 12.498 7-:-5767 48222.6 - Antigorite 6M ifgsoo 6.0 -2 o 1 12.727 - 0-_2=2'"9"'--~ ...... -?- ---- - ------1 3 17.804 4.9777 100 4.7 - -,-- -----??-????. -? -? - - -------- 4 19.713 4.4999 34 1.6 ---~--------. - - . ?? ?? ??- - --- - -----------< 5 19.899 4 .4582- ? 42 2.0 ------- --?? ?- - --? - --?- - - ---- ------------< 6 19.998 4 .4363 54 2.5 7 20.859 4.2550 452 21 .2 Quartz - - 4.2557 21.0 1 OO ? 20.856 -0.003 ,-.--+- --------- ~ --?? ?--- -?? - -- - . -- - . - -? ?- .... - ? 8 21 .341 4.1600 111 5.2 ------?-- ----- -?- ?----?- --------------- 9 21.478 4.1339 83 3.9 .. To -4.0787' . ----?--?-- -?????--- -?--? - ?? --??-- ? ??- ? -----?------21 .772 43 2.0 11 22.015 4.0341-- 57 2.7 Antlgorite 6 M 4.0100 . 2.0 ? ??9 0 1 22.150 0.1 34 12 24.698 ?3.6017 1129 53.0 Antigorite 6M 3.6300 75.0 -0 0 2 24.502 -0.196 13 25.254 3.523f 80 3.8 ? Antigorite 6M 3.5100 6.0 -11 0 . 1 25.354 0.100 14 --26.100 -3.3360? --2035 95.6 ?auariz -3.3439 ? 100.0 1 0 .. .; ? 26?1i36-- -0.064 15 . 27.583 3.2312 96 4.5 ---?- .. 16 ? 21.72!f -3.2145 - 101 4.7 - ? -- .... ---? ... ?- 17 27.979 3.1 863 - 179 - 8.4 - ..... . - . ??---? -? - 18~ 28.607 ?- 3.1178- - 25 - 1.2 -- ... . -----?-?? ? - -- - ?? - ?-?---- ---< 19 - 29j,04 ? J:oTso ? ? 42 2.0 ??- - - -- -? . - -- - ?- --- - -< '20 - 2s:aiJ fii944 ? ?- 63 3.o ' ?-?-? - --------l '21 ?3-i":'331 2:5s21 ?. . 43- ? 2.0 . --- .. ??? ?? --? - ----- 22 . -:f:fffg '" ?- -? - --- ------ -2.7026 39 1.8 23 ?-? 33.409 - 2.6799- - ?37 1.7 - .... ?- -----------1 124 ? ? 34.7 0 i - 2.5825 - -? eo - 2.8 Antigorit8 ?s M -? ??? 2.5900 -1.0 -7 3 1 - 34.604 -0.104 Ts ?3 -4.981 2:5629 ? 67 3.1 Antigorite 6M 2.5700 2.0 7 3 1 3(882 -0.100 Ts 35:s"f1 2.5252 f,41 ?6.9 AiitiQiinte 6M 2.5200 18.0 13 2 1 ?J s.sg1?- o.ois? -- 27 ?35_54?2? 2~4-569 165 ?? -7~8 aua?rtz?? . -? 2.4570 6.7 1 1 ""6 -? 36.541 -0.001 28 - . - - ? -36:-931 - 2.4319 73 3.4 - 29 37.243 2.4123 f34 6.3? Antigorlte 6M 2.4200 10.0 0 0 3 "J?i.120 -0.123 30 :t92 a? 2 ?31.so3 -:o.ie9_ __ _ - 37 2.3785 36 1. 7 Antigorite 6M 2.3900 3.0 -14 31 - 37.815 2~3n1 --,.t 22-- 321--39_519 2.2785 ?? ? 134. .. 6.3 Quartz - - 2.2817 6.6 0 1 2 -:39.461 . -0.058 33 - 040.398 2.2309 ? ?105- 4.9 Q-uarti" -? ?? 2.2368 2:9 1 1 1 . 40.286 .. -0.112 34 ~ 41.600--2.1691 ???? 41 1.9 Antigorite6M - 2.1670? ?- S] 7 4 -,,-? 41 .643 0.043 ---- 35 ?-;ff980- 2T?504 46 2.2 Antlgorite6M .. . . ?2.1soci 5.0 16 0 2 -41:e88 - ? ? o:008 ?? - .. ?- ?- 35 42.575 2.121i ?"769?? 7.9 Antigorite 6M 2.1260 1.0 ~9 1 3 - 4i4B5 ? -0.091 37~ 45.520 - 1.9911 50?- -i.3 - ???????- - -~ --- - -? --- - ?-?-?? - -- --- ?- 38 45.801 1.9795 - 101?? --;a Quartz 1.9800 2.7 0 ? ? 2 ? f 45.789 -0.012 39 so.200 1.8158 221 10.4 AntigortteifM -- ? ?1.8150??-? s.o :1 0"" 4 50.225-- 0.025 ___ 40 51.296 1.TT96 - ??29 1.4 Antigorite 6M - ?1.7810 -4.ci -1 ?1 ? 4 - 51.252 -0.043 -- 41 - 53.027 1.7255 - 58 2.7 -- . - ?- ?- ? -- - --? --- - ?- 42 ...- 53_301 1.1113 ???--39 ? 1.a ??- ?------------- -- ??-??-?-- - - ???? ?-- ? ---~-----=-==-:~_-_-_-_-_: 43 -- 53.562 1.7095 ? - ? 33 1.6? - ?? ?------ - ?? .... . .. ?- -???-??--? ? 143 15 _,.._ 12000 mMHi02 ,-.. > ___._. 10 3000 mMH20.,-._,, 2 u --- 750 ?M H20 2 ~ :::t 5 o~~~~~~~;;;;;;;;;;;;;~~L----. 0 1 2 3 4 5 6 days Figure A-1. Oxidation of Cr(III) in Connecticut plating waste soil (14-40 cm) after single applications of different concentrations of H2O2 ? Soil amended with 100 ?M Cr(III) prepared with 2: l OR/Cr ratio. 144 3 -- 4:1 Olf/Cr ---- 3:1 Olf/Cr 2:1 Olf/Cr oL---,-----:===:=:==::;~~--- o 1 2 3 4 5 6 days Figure A-2. Oxidation of Cr(III) in Connecticut plating waste soil (14-40 cm) after a single application of 3.00 mM H20 2? Soil amended with 100 ?M Cr(III) prepared with different Off/Cr ratios. 145 3 -1t- Connecticut (14-40 cm) ---o-- Connecticut with 1 ?M Fe(II) -.- Serpentine (53-75 cm) ---ts-- Serpentine with 1 ?M Fe(II) --------------- --------- --------------- 0 0 1 2 3 4 5 6 days Figure A-3. Oxidation of Cr(III) in two soils after a single application of3.00 mM H2O2? Soils amended with 100 ?M Cr(III) prepared with 2: 1 Off/ Cr ratio. Applications ofH2O2 were made with and without 1.0 ?M FeSO4? 146 -0- 2:1 OH/Cr 1.5 -0- 4:1 OH/Cr -I!.- 3:1 OH/Cr ,~_., 1.0 uJ. ~ ::t 0.5 0.0 0 1 2 3 4 5 6 days Figure A-4. Oxidation of Cr(III) in Serpentine soil (53-75 cm) after a single application of 3.00 mM H20 2? Soil amended with 100 ?M Cr(lll) prepared with different oH-/Cr ratios. 147 a) . 3000 Sparge with 0 2 100 - 2500 80 ~ (i - 2000 ... ::1. 0 -:;; N 0 1500 ,:: =N 0 1000 -::: 500 0 0 2 4 6 8 10 12 14 days b) 3000 Sparge with N2 100 - 2500 80 (i -~ 2000 "'I ::1. 0 -:;; 0 1500 ~ N 0 ::t ::: 1000 - 500 o...-----r-----r-~----.---,.----.----~~ 0 2 4 6 8 10 12 14 days Figure A-5. Interaction of 100 ?M Cr(III) (prepared with 2:1 Off/Cr ratio) and 3.00 mM H2O2 after a) sparging 30 min with 0 2 and b) sparging 30 min with N2? 148 a) Sparge with 0 3000 2 100 -Cr(VI) - 2500 80 (.",) -~ 2000 ::::l 0 0 ~ 1500 N 0:: 1000 500 20 0 2 4 6 8 10 12 14 days b) Sparge with N 3000 2 100 2500 -CrVI - (~ 2000 .",) -::::l 0 ~ 0 1500 i-:: =N 0 _; 1000 500 20 o--~----- ---~---~~ 0 2 4 6 8 10 12 14 days Figure A-6. Interaction of 100 ?M Cr(VI) and 3.00 mM H20 2 at pH 4.0 after a) sparging 30 with 0 2 and b) sparging 30 min with N2? 149 100 -0-1/30 "C ~ (j = -1120 "C 0 75 I,,, Q., -+--1/10 I,,, ~ --116.7 ~ 50 =~ -115 ~ "C 'i< e 25 --113.3 ~ ~ ~ : : 0 0 2 ...,. 4 5 6 7 days Figure A-7. Ratio ofH2O2 used to Cr(VI) produced from 100 ?M Cr(III) (prepared with 2: 1 OH-/Cr ratio) for different initial concentrations of H2O2? Different initial H 2O2 concentrations given as Cr(IIl)/H2O2 ratios. 150 REFERENCES Aiyar, J. , Berkovits, H.J., Floyd, R.A. and Wetterhahn, K.E. (1991) Reaction of chromium (VI) with glutathione or with hydrogen peroxide: Identification of reactive intermediates and their role in chromium(VI)-induced DNA damage. Environ. Health Perspect. 92: 53-62. Ball, J. W. and Nordstrom, D.K. (1998) Critical evaluation and selection of standard state thermodynamic properties for chromium metal and its aqueous ions, hydrolysis species, oxides and hydroxides. J. Chem. Eng. Data 43: 895-918. Baloga, M.R. and Earley, J.E. (1961) The kinetics of the oxidation ofCr(III) to Cr(VI) by hydrogen peroxide. J. Am. Chem. Soc. 83: 4906-4909. Baron, D. and Palmer, C.D. (1996) Solubility of KFeiCr04h(OH)6 at 4 to35? C. Geochim. Cosmochim. Acta 60, 3815-3824. Baron, D., Palmer, C.D. and Stanley, J.T. (1996) Identification of two iron-chromate precipitates in a Cr(VI)-contaminated soil. Environ. Sci. Technol. 30, 964-968. Bartlett, R.J. ( 1991) Chromium cycling in soils and water: Links, gaps, and metho~s. Environ. Health Perspect. 92: 17-24. Bartlett, R.J. and James, B.R ( 1979) Behavior of chromium in soils: III. Oxidation. J. Environ. Qual. 8: 31-35. Bartlett, R.J. and James, B.R. (1980) Studying air-dried, stored soil samples - some pitfalls. Soil Sci. Soc. Am. J. 44: 721-724. Baxendale, J.H. ( 1952) Decomposition of hydrogen peroxide by catalysts in homogenous aqueous solution. Adv. Catal. IV: 31-87. Baxendale, J.H. (1955) Oxidation of Benzene by Hydrogen Peroxide, University of Notre Dame Press, Notre Dame, IN, p34. Beck, M.T., Nagy, I.P. and Szekely, G. (1991) Oscillatory kinetics and mechanism of the chromate catalyz.ed decomposition of hydrogen peroxide. Int. J. Chem. Kinet. 23: 881-896. Bielski, B.H.J., Cabelli, D.E. and Arudi, R.L. (1985) Reactivity ofHO2"/Ot radicals in aqueous solution. J. Phys. Chem. Ref. Data 14: 1041-1100. Bowman, RA. ( 1988) A rapid method to determine total phosphorus in soils. Soil Sci. Soc. Am. J. 52: 1301-1304. 151 Brown, S.B., Jones, P. and Suggett, A. (1970) Recent developments in the redox chemistry of peroxides. In: Progress in Inorganic Chemistry Vol. 13, pp. 159- 203. Buerge, I.J. and Hug, S.J. ( 1998) Influence of organic ligands on chromium (VI) reduction by iron (II). Environ. Sci. Technol. 32: 2092-2099. Burke, T., Fagliano, J., Goldoft, M., Hazen, R.E., lglewicz, R. and McKee, T. (1991) Chromite ore processing residue in Hudson County, New Jersey. Environ. Health Perspect. 92: 131-137. Buxton, G.V. and Djouder, F. (1996) Disproportionation of Cr V generated by the radiation-induced reduction of Cr VI in aqueous solution containing fonnate: A pulse radiolysis study. J. Chem. Soc., Faraday Trans. 92: 4173-4176. Buxton, G.V., Djouder, F., Lynch. D.A. and Malone, T.N. (1997) Oxidation of Cr III to Cr VI initiated by OH? and SO4?? in acidic aqueous solution. J. Chem. Soc., Faraday Trans. 93: 4265-4268. Calder, L.M. ( 1988) Chromium contamination of groundwater. In: Chromium in the Natural and Human Environments (Nriagu, J.O. and Niebor, E., Eds.), pp. 215- 229, John Wiley, New York. Carberry, J.B. (1994) Enhancement ofbioremediation by partial preoxidation. In Remediation of Waste Contaminated Soils (Wise, D. and Trantolo, D., Eds.), pp. 543-596. Chernikov, S. (1998) Chromate Reduction in Soils and Simple Systems by Humic Substances and a Variety of Organic Materials, Ph.D. Dissertation, University of Maryland, Dept ofNatural Resource Science, College Park. Chiu, A., Chiu, N., Shi, X.L., Beaubier, J. and Dalal, N.S. (1998) Activation of a procarcinogen by reduction: Cr6+-> Cr5+-> Cr4+-> Cr3+ a case study by electron spin resonance (ESR/PMR). Environmental Carcinogenesis & Ecotoxicology Reviews-Part C of Journal of Environmental Science and Health 16, 135-148. Cohen, M.D., Kargacin, B., Klein, C.B. and Costa, M. (1993) Mechanism of chromium carcinogenicity and toxicity. Crit. Rev. Toxicol. 23: 255-281. Cooper, W.J. and Zika, R.G. (1983) Photochemical formation of hydrogen peroxide in surfuce and ground waters exposed to sunlight. Science 220: 711-712. Cotton, F.A. and Wilkinson, G. (1988) Advanced Inorganic Chemistry, 5th Edition, John Wiley & Sons, New York. 152 cl"anst on, RE. and M naturaJ Waters.:, ~~ (1978) The detennination ofc hromium species in Cre~ ? ? Acta 99: 275-282. ~~D . Che .m (1is9t8ry3 )o GfP eenre ral. and th eo~eti.c al aspects of the peroxide group. In The York. OJades (Patai, S., Ed.), pp. 1-84, John Wiley& Sons, New DaVis , A K.e an'd chIrnopmtoiunm, J ~ Ht ? ~d N~. cholson, A. (1994) Growidwatertransport ofa rsenic Geochem. 9? 569 a5histoncal tannery, Woburn, Massachusetts, U.S.A. Applied I) . ? - 82. aVJs A ' re?m aenddi aOtilose n? R?L ? ( l 995) The geochemistry ofc hromium migration and DeFJ nm tbe subsurfuce. Ground Water 33: 759-768. ora, S., Bagnasc chromium o, M., Serra, D. and .lanacchi, P. (1990) Genotoxicity of Den compounds: A review. Mutation Res. 238: 99-172. g B. and St cotnpar/;ne, AT._ (1996) Surface catalyzed chromiwn(VI) reduction: Reactivity Environ ~~ ofd ifferent organic reactants and different oxide surmces. Die . c1. Technol. 60: 99-106. kma1n1l,oMly.bHd ? andPo pe, M. T. (1994) Peroxo and superoxo complexes ofc hromium, ,n rfinan, enum, and tungsten. Chem. Rev. 94: 569-584. 0 BUr1;:;1? and Adams, G.E. (1973) Reactivity of the hydroxyl radical. National DuJc u ofS tandards Rep. NSRDS-NBS-46 NBS, Washington, D.C. e, Ph.yR.~ a nd lla as, T. W. (1961) The homogeneous base-catalyzed decomposition of Eary ogen peroxide. J. Phys. Chem. 65: 304-306. 'L.bEy. and R. ai, D. (1987) Kinetics ofc hromium (111) oxidati.o n to chrom1.w n (VI) llory reaction with manganese dioxide. Environ. Sci. Technol. 21: 1187-1193. , LW.Ei~ ha nd Rai?, D. (1988) Chromate removal from aqueous wastes by reduction EJ~\V: . ferrous iron. Environ. Sci. Tecbnol: 22: 972-977. ~!M. (1 :~6) Studies on Some Chemical Pr~perties ofS ome ~gypt~ Soils: lJni~ ~sttmn ofH ydrogen Peroxide by Torrifluvents, Ph.D. dissertation, r:- ers1ty ofTanta., Egypt cfuvnz., ? ? p~?8? and Fish, W. (1995) Redox interactions ofCr(VI) and substitutes r:- enols: products and mechanism Environ. Sci. Technol. 29: 1933-1943. clpri.n ce A ._... ' ?1v1 ? and Mohamed, W.H. (1 992) Catalytic decompo_s ?m ?o n ki.n ea ?c ~ of . ~~eous hydrogen peroxide and solid .magnesium peroxide by bimess1te. Soil . Soc. Am. J. 56: 17 84-1788. 153 Evans, D.F. and Upton, M.J. (1985) Studies on singlet oxygen in aqueous solution, Part 4. The 'spontaneous' and catalyzed decomposition of hydrogen peroxide. J. Chem. Soc. Dalton Trans. 2525-2529. Fendorf, S.E. and Zasoski, R.J. (1991) Chromium oxidation by d-MnO2 - Characterization. Environ. Sci. Technol. 26: 79-85. Fendorf, S.E., Zasoski, R.J. and Burau, R.G. (1993) Competing metal ion influences on chromium (Ill) oxidation by birnessite. Soil Sci. Soc. Am. J. 57: 1508-1515. Fendorf, S., Eick, M.J., Gross~ P. and Sparks, D.L. (1997) Arsenate and chromate retention mechanisms on goethite. 1. Surface structure. Environ. Sci. Technol. 31: 315-320. Fenton, H.J.H. (1894) Oxidation of tartaric acid in the presence of iron. J. Chem. Soc. 65: 899-910. Frew, J.E., Jones, P. and Scholes, G. (1983) Spectrophotometric determination of hydrogen peroxide and organic hydroperoxides at low concentrations in aqueous solutions. Anal. Chim. Acta 155: 139-150. Funahas~ S., Uchida, F. and Tanaka, M. (1978) Reactions of hydrogen peroxide with metal complexes. 3. Thermodynamic and kinetic studies on the formation, dissociation, and decomposition of peroxochromium (VI) complexes in acid media. Inorg. Chem. 17: 2784-2789. Ganey, A.B. and Wamser, G.A. (1976) Suppression of water pollution caused by solid wastes containing chromium compounds. U.S. Patent No. 3,981,965 . Greenwood, N .N. and Earnshaw, A. (1994) Chemistry of the Elements, Pergamon Press, New York. Haber, F. and Weiss, J. (1934) The catalytic decomposition of hydrogen peroxide by iron salts. Proc. R. Soc. London Al 47: 332-351. Herbert, D. and Pinsent, J. (1948) Crystalline bacterial catalase. Biochem. J. 43: 193- 202. Ho, C.L., Shebl, M.A. and Watts, R.J. (1995) Development of an injection system for in situ catalyz.ed peroxide remediation of contaminated soil. Haz. Waste & Haz. Materials 12: 15-25. Holm, T.R., George, G.K. and Barcelona, M.J. (1987) Fluorometric determination of hydrogen peroxide in groundwater. Anal. Chem. 59: 582-586. House, D.A. (1997) Advances in Inorganic Chemistry, Vol. 44 Vol. 44, pp. 341-373. 154 lnada., Y. and Funahashi, S. (1997) Laboratory stopped-flow XAFS apparatus. Structure determination of the short-lived peroxochromium intermediate formed during reduction process of chromate(VI) ion by hydrogen peroxide. Zeitschrift Fur Naturforschung Section B-a Journal of Chemical Sciences 52, 711-718. . Itoh, M., Nakamura, M., Suzuki, T., Kawai, K., Horitsu, H. and Takamiz.awa, K. ( 1996) Mechanism of chromium (VI) toxicity in Escherichia coli: Is hydrogen peroxide essential in Cr(VI) toxicity? J. Biochem (Tokyo) 117: 780-786. JADE: software for XRD spectra identification. (1999) Materials Data, Inc., Livermore CA. James, B.R. ( 1994) Hexavalent chromium solubility and reduction in alkaline soils enriched with chromite ore processing residue. J. Environ. Qual. 23: 227-233. James, B.R. (1996a) The challenge ofremediating chromium contaminated soil. Environ. Sci. Technol. 30: 248A-251A. James, B.R. (1996b) Chromium Oxidation and Reduction Chemistry in Soils: Relevance to Chromate Contamination of Groundwater of the Northeastern United States, Research proposal submitted to the Maryland Water Resources Research Center. James, B.R. and Bartlett, R.J. (1983a) Behavior of chromium in soils. V. Fate of organically complexed Cr(ill) added to soil. J. Environ. Qual. 12: 169-172. James, B.R. and Bartlett, R.J. (1983b) Behavior of chromium in soils. VII. Adsorption and reduction ofhexavalent fonns. J. Environ. Qual 12: 177-181. James. B.R., Petura, J.C., Vitale, R.J. and Mussoline, G.R. (1995) Hexavalent chromium extraction from soils: a comparison of five methods. Environ. Sci. Technol. 29: 2377-2381. James. B.R., Petura, J.C., Vitale, R.J. and Mussoline, G.R. (1997) Oxidation-reduction chemistry of chromium: Relevance to the regulation and remediation of chromate-contaminated soils. J. Soil Contamination 6(6): 569-580. Jennette, K. ( 1979) Chromate metabolism in liver microsomes. Biol. Trace Elem. Res. 1: 55-62. Johnso?' C.A. and Xyla, A.G. (1991) The oxidation of chromium (III) to chromium (VI) on the surface of manganite (g - MnOOH). Geochim. Cosmochim. Acta 55: 2861-2866. 155 Katz, S.A. and Salem, H. (1994) The Biological and Environmental Chemistry of ChromiUI11, VCH Publishers, New York. Kawanishi, S., Inoue, S. and Sano, S. (1986) Mechanism of DNA cleavage induced by sodium chromate{Vl) in the presence of hydrogen peroxide. J. Biol. Chem. 261: 5952-5958. Kieber, R.J. and Helz, G.R. ( 1992) Indirect photoreduction of aqueous chromium (VI). Environ. Sci. Technol. 26: 307-313. Klaning, U .K. ( 1958) Photo induced oxidations of alcohols by acid chromate. Acta Chem. Scan. 12: 576-577. Klaning, U.K. (1977) Photoinduced oxidation ofpropan-2-ol by acid chromate. J. Chem. Soc., Faraday Trans. 73: 434-455. Kontchou. C. Y. and Blondeau. R. ( 1990) Isolation and characterization of hydrogen peroxide producing Aerococcus sp. :from soil samples. FEMS Microbiol. Lett. 68: 323-328. Kortenkamp, A., Oetken, G. and Beyersmann, D. (1990) The DNA cleavage induced by chromium (V) complex and by chromate and glutathione? is mediated by activated oxygen species. Mutat. Res. 232: 155-161. Manceau, A. and Charlet, L. (1992) X-ray absorption spectroscopic study of the sorption of Cr (III) at the oxide-water interface - Molecular mechanism of Cr (III) oxidation on Mn oxides. J. Colloid Interface Sci. 148: 425-442. Mattuck, R. (1994) The Role of Wetlands in the Immobiliz.ation of Chromium Contamination, Master's Thesis, University of Connecticut-Storrs (Tech. Rep. ERI-94.05). Melhorn, R.J., Buchanan, B.B. and Leighton, T. (1994) Bacterial chromate reduction and product characterization. In: Emerging Technology for Bioremediation of Metals (Mean, J.L. and Hinchee, R.E., Eds.), Lewis Publishers, Boca Raton, Florida. Nakayasu, K., Fukushima, M., Sasaki, K., Tanaka, S. and Nakamura, H. ( 1999) Comparative studies of the reduction behavior ofchromium(VI) by humic substances and their precursors. Environ. ToxicoL Chem. 18: 1085-1090. National Research Council (1990) Recommended Dietary Allowances, 10th Ed.,Committee on Dietary Allowances, Food and Nutrition Board, National Academy of Sciences, Washington, D.C. 156 NikoJaid? IS, N.P., Robbins J. ~d Suib, S.L. j ~A., Sc~erer,_M ._, M~Aninch, B., Binkhorst, G., Asikainen, glac1of1uvial Jc 4) Vencal distnbut1on and partitioning ofc hromium in a N . aqu er. Ground Wat. Monit. Remed. llagu, J and N? J? .W iley iebo, Se n.e r, E. ? (1988) ci.-s ui omn . un m. the Natural and Human Environments Y. m Adv ances . Enw. N m onmental Science and Technology, V 20,, Palmer C ' .D and w? chro~um Jttbrndt, P.R. (1991) Processes affecting the remediation of Pai- contamina ted si?t es. Enw. on. Health Perspect. 92: 25-40 . .uoer C ' . and Puls R groundw ' ? (1 994) Natural attenuation ofhexavalent chromium in Solid wa::er and soils. USEPA, Office ofR esearch and Development, Office of Pard? e and Emergency Response. EPA/540/S-941505. Jeck, D L Ox.id~~? Bou~er, E.J. and Stone, A. (1992) Hydrogen peroxide use to increase reView lapacJty for in situ bioremediation ofc ontaminated soils and aquifers: A Pe ? ? ournaJ of Contaminant Hydrology 9: 221-242. rez-Be . pe:~:?F. an~ Arias, C. (1997) A kinetic study oft he chromium(VI) hydrogen Physfoat~::ct~on. Role of the diperoxochromate(VI) intennediates. Journal of Pett? 1l1JStry A.. 101, 4726-4733. llle 11 , ivi., LaN . intern oc~, T., Liberatori, A. and Loreti, L. (1988) Hydrogen peroxide llleth:;ence m th~ detennination ofc hromium(Vl) by the diphenylcarbazide Pett? . Anal Chim. Acta 209: 315-319. ? -M . ~oJ~ ::.! Minero, F.J. (1990) Chromium speciation in seawater: the probable Pett? Ycirogen peroxide. Limnol. Oceanogr. 35: 730-736. llle, ?., Miller F . . . seaWat O .J. and La Noce, T. (1991) Chromium (III) mteractJons m l>j er through its oxidation kinetics. Mar. Chem 34: 29-46, ~teUo, JJ " degr. ? ~d Baehr, K. (1994) Ferric complexes as catalysts for "Fenton p adation of2 ,4-D and metolachlor in soil. J. Environ. Qual. 23: 365-370? 0 we11,c Ro. ::??. Puls, R. W., Hightower, S.K. and Sabatini, D.A. (1995) Coup~ed- iron En ? sion and chromate reduction: mechanisms for subsurface remediation. l>rj VU-on. S . CJ. TechnoJ. 29: 1913-1922. Ce, D., Wo ?~ . r..,1old p J and M t R.F 'd . th C (1992) Hydrogen peroxi em e lllarine ? ' ? ? an oura. ? ? ? nd 5 Anal Chem. 11 ? 379_ ;nvironment: Cycling and methods ofa nalysis. Tre ? ? 38 ~ . benborst J i1s fo ' .E., Foss, J.E. and Fanning, D.S. (1982) Genesis ofM aryland so llned from setpentinite. Soil Sci. Soc. Am. J. 46: 607.616? 157 Rai, D., Sass, B. and Moore, D.A. ( 1987) Chromiwn (Ill) hydrolysis constants and solubility of chromium (III) hydroxide. Inorg. Chem. 26: 345-349. Ravikumar J X and G I, M ( 1994) Chemical oxidation of chlorinated organics by h , ? ? uro ? f d E viron Sci Technol. 28: 394-400. Y d rogen peroxide in the presence o san ? n ? ? Rotzinger F.P St?? . H d Marty W. (l 986) Early stages of the hydrolysis of ' . ?, WlZI, ? an ' ? ? t ? fth c hro nuum(III) in aqueous solut1. 0n. 3. Ki?n eti?c s 0 f dnnenz.a 100 o e deprotonated aqua ion. Inorg. Chem. 25: 489-495. Sass, B.M. and Rai, D. (1987) Solubility of amorphous chromiwn(III)-iron(III) hydroxide solid solutions. Inorg. Chem. 26: 2228-2232. Schumb, W.C., Satterfield, C.N. and Wentworth, R.L. Hydrogen Peroxide, Reinhold, New York p 759. Seaman, J .C., Bertsch, P.M. and Schwallie, L. (I 999) In situ Cr(Y1) reduction ~bin coarse-textured, oxide-coated soil and aquifer systems usmg Fe(II) solutions. Environ. Sci. Technol. 33, 938-944. ? Seinfeld, J.H. (1986) Atmospheric Chemistry and Physics of Air Pollution, Wtley- Interscience, NY p 739. Shi, X.L., Chi~ A., Chen, C.T., Halliwell, B., Castranova, V. and Vallyathan, V. (1999) Reduction of chromium(VI) and its relationship to carcinogenesis. Journal of Toxicology and Environmental Health-Part B-Critical Reviews 2, 87- 104. Shi, X.L. and Dalal, N .S. (1989) Chromiwn (V) and hydroxyl radical formation during the glutathione reductase-catalyzed reduction ofc hromium (VI). Biochem. Biophys. Res. Corrnnun. 163: 627-634. Shi, X .L. and Dalal, N.S. {1990) Evidence for a Fenton-type mechanism for the generation of OH? radicals in the reduction ofCr(VI) in cellular media. Arch. Biochem. Biophys. 281: 90-95. Shi, X .L. and Dalal, N.S. ( 1993) Generation of free radicals from hydrogen peroxide and lipid hydroperoxides in the presence of Cr(III). Arch. Biochem. Biophys. 302: 294-299. Shi, X.L., Leonard, S.S., Liu, K.J., Zang, L.Y., Gannett, P.M., Rojanasakul, Y., Castranova, V. and Vallyathan, V. (1998) Cr(III)-mediated hydroxyl radical generation via Haber-Weiss cycle. J. Inorg. Biochem 69, 263-268. Silvester, ~-, ~harlet, L. and Mancea~ A (l 995) Mechanism ofc hromium (Ill) oxtdat1on by Na-buserite. J. Phys. Chem. 99: 16662-16669. 158 11111 Ill Skoog, D .A., West, D.M. and Holler, F.J. (1994) Analytical Chemistry, 6th Ed., Pp. 305-327, Saunders College Pub., Philadelphia. Soil Survey of Windham County, C. (1981) USDA, Soil Conservation Service, U.S. Govt. Print. Off Spain, J.C., Milligan, J.D., Downey, D.C. and Slaughter, J.K. (1989) Excessive bacterial decomposition of H2O2 during enhanced biodegradation. Ground Water 27: 163-167. Spiccia, L . ( 1988) Solubility of chromium(III) hydroxides. Inorg. Chem. 27: 432-434. Spiccia, L. and Marty, W. (1986) The fate of"active" chromium hydroxide, Cr(OH)3 x3H20 , in aqueous suspension. Study of the chemical changes involved in its aging. Inorg. Chem. 25: 266-271. Spiccia, L., Marty, W. and Giovanoli, R. (1988) Hydrolytic trimer of chromium(III). Synthesis through chromite cleavage and use in the preparation of the "active" trimer hydroxide. Inorg. Chem. 27: 2660-2666. Spiccia, L., Stoeckli-Evans, H., Marty, W. and Giovanoli, R. (1987) A new "active" chromium(III) hydroxide: Cri{?-OH)i{OH)lOH2) 4-2H2O. lnorg. Chem. 26: 474-482. Spitalsky, E . (I 907) Catalysis by means of chromic acid and its salts I. Z. Anorg. Alig. Chem 53: 184. Spitalsky, E . ( 1908) Catalysis by means of chromic acid and its salts II. Z. Anorg. Alig. Chem 56: 72. Steams, D .M. and Wetterhahn, K..E. (1994) Reaction of chromium (VI) with ascorbate produces chromium (V), chromium (IV) and carbon based radicals. Chem. Res. Toxicol. 7: 219-230. Steams, D.M. and Wetterhahn, K..E. (1997) Intermediates produced in the reaction of chromium (VI) with dehydroascorbate cause single-strand breaks in plasmid DNA. Chem Res. Toxicol. 10: 271-278. Stunzi, H. and Marty, W. {1983) Early stages of the hydrolysis of chromium (Ill) in aqueous solution I. Characterization of a tetrameric species. Inorg. Chem. 22: 2145-2150. Suther~ S.S. (1997) Remediation Engineering Design Concepts, CRC Lewis Publishers, New York pp. 215-236. 159 11111111111111111111111111111111111111111111111111111111111 111 11 Tyre, B.W., Watts, R.J. and Miller, G.C. (1991) Treatment of four biorefractory contaminants in soils using catalyzed hydrogen peroxide. J. Environ. Qual. 20: 832-838. Tyrpin, L.R. (1998) Using a Thermodynamically Based Approach to Assess the Redox Status and Predict the Valence State of Chromium in Simple, Aqueous Systems and Chromium Enriched Soils, Master's Thesis, University of Maryland, College Park, MD. U.S. EPA (1981) Interim report on the groundwater quality of east and north Woburn, Massachusetts. USEPA Contract No. 68-01-6056, TDD FI-8010-04B, Washington, D.C. U.S. EPA (1998) In situ remediation technology: in situ chemical oxidation. USEPA, Office of Research and Development, Office of Solid Waste and Emergency Response, Technology Innovation Office. EPA/542/S-98/008. ? Vitale, R.J., Mussoline, G.R., Rinehimer, K.A., Petura, J.C. and James, B.R. (1997) Extraction of sparingly soluble chromate from soils: Evaluation of methods and E(h)-pH effects. Environ. Sci. Technol. 31, 390-394. von Meyenburg, U., Siroky, O. and Schwarzenbach, G. (1973) Die deprontonierung von metallaquoionen II: Aquochrom(IIl)-ion. Zur struktur activer cbromhydroxide. Helv. Chim. Acta 56: 1099. Wagner, D.P., Fanning, D.S., Foss, J.E., Patterson, M.S. and Snow, P.A. (1982) In Acid Sulfate Weathering (Kittrick, J.A., Ed.), pp. 109-125, Special Publication No. 10; Soil Science Society of America, Madison, WI. Wardman, P. and Candeias, L.P. (1996) Fenton chemistry, an introduction. Radiat. Res. 145: 523-531. Watts, RJ., Udell. M.D., Kong, S.H. and Leung, S.W. (1999) Fenton-like soil remediation catalyz.ed by naturally occurring iron minerals. Environmental Engineering Science 16, 93-103. Watts, RJ., Udell. M.D. and Monsen, RM. (1993) Use of iron minerals in optimizing the peroxide treatment of contaminated soils. Water Environ. Res. 65: 839- 844. Watts, RJ., Udell, M.D., Rauch, P.A. and Leung, S. (1990) Treatment of pentachlorophenol-contaminated soils using Fenton's reagent. Haz. Waste & Haz. Materials 7: 335-345. 160 Weng, C.H., Huang, C.P., Allen, H.E., Cheng, A.H.D. and Sanders, P.F. (1994) Chromiwn leaching behavior in soil derived from chromite ore processing waste. Sci. Total Environ. 154, 71-86. Witt, S.N. and Hayes, D.H. (1982) Kinetics and mechanism of the formation of violet peroxychromate in aqueous solution. Inorg. Chem. 21: 4014-4016. Wittbrodt, P.R. and Palmer, C.D. (1996) Reduction ofCr(VI) by soil humic acids. Eur. J. Soil Sci. 47: 151-162. Woods, T.L. and Garrels, R.M. (1987) Thennodynamic Values at Low Temperature for Natural Inorganic Materials: An Uncritical Summary, Oxford University Press, Oxford. Xu, F.G., Krouse, R. and Swaddle, T. W. (1985) Conjugate base pathway for water exchange on aqueous chromiwn(III): variable-pressure and -temperature kinetic study. Inorg. Chem. 24: 267-270. Zepp, R.G., Faust, B.C. and Hoigne, J. (1992) Hydroxyl radical formation in aqueous reaction (pH3-8) of iron(II) with hydrogen peroxide: The phot-Fenton reaction. Environ. Sci. Technol. 26: 313-319. Zhang, L.B. and Lay, P.A. (1996) EPR spectroscopic studies of the reactions ofCr(VI) with L-ascorbic acid, L-dehydroascorbic acid, and 5,6-O-isopropylidene-L- ascorbic acid in water. Implications for chromium (VI) genotoxicity. J. Am. Chem Soc. 118, 12624-12637. Zhang, L.B. and Lay, P.A. (1998) EPR spectroscopic studies on the formation of chromiwn (V) peroxo complexes in the reaction of Cr(VI) with hydrogen peroxide. Inorg. Chem 37, 1729-1733. 161