Search

The factors controlling rates of contaminant reduction by magnetite (Fe3O4) are poorly understood. Here, we measured the reduction rates of three ArNO2 compounds by magnetite particles ranging from highly oxidized (x = Fe2+/Fe3+ = 0.31) to fully stoichiometric (x = 0.50). Rates of ArNO2 reduction became almost 5 orders of magnitude faster as the particle stoichiometry increased from x = 0.31 to 0.50. To evaluate what was controlling the rate of ArNO2 reduction, we measured apparent 15N kinetic isotope effects ((15)N-AKIE) values for nitrobenzene and magnetite open-circuit potentials (E(OCP)). 15N-AKIE values were greater than unity for all magnetite stoichiometries investigated, indicating that mass transfer processes are not controlling the rate of ArNO2 reduction by magnetite. E(OCP) measurements showed that the E(OCP) for magnetite was linearly related to the stoichiometry, with more stoichiometric magnetite having a lower potential. Based on these results, we propose that conceptual models that incorporate both redox and Fe2+ diffusion processes, rather than those that rely solely on diffusion of Fe2+, are more appropriate for understanding contaminant reduction by magnetite. Our work indicates that particle stoichiometry should be considered when evaluating rates of contaminant reduction by magnetite.

Magnetite (Fe3O4) is a common biomineralization product of microbial iron respiration and is often found in subsurface anoxic environments, such as groundwater aquifers where aqueous Fe(II) is present We investigated the reaction between aqueous Fe(II) and magnetite using the isotopic selectivity of 57Fe Mössbauer spectroscopy and revisited the reduction of nitrobenzene by magnetite. Similar to our previous findings with Fe3+ oxides, we did not observe the formation of a stable sorbed Fe(II) species; instead, we observed oxidation of the Fe(II) to a partially oxidized magnetite phase. Oxidation of Fe(II) was accompanied by reduction of the octahedral Fe3+ atoms in the underlying magnetite to octahedral Fe2+ atoms. The lack of a stable, sorbed Fe(II) species on magnetite prompted us to reevaluate what is controlling the extent of Fe(II) uptake on magnetite, as well as contaminant reduction in the presence of magnetite and Fe(II). Uptake of Fe(II) by magnetite appears to be limited by the stoichiometry of the magnetite particles, rather than the surface area of the particles. More oxidized (or less stoichiometric) magnetite particles take up more Fe(II), with the formation of stoichiometric magnetite (Fe2+/Fe3+ = 0.5) limiting the extent of Fe(II) uptake. We also showthat stoichiometric magnetite, in the absence of aqueous Fe(II), can rapidly reduce nitrobenzene. Based on these results, we speculate that contaminant reduction that was previously attributed to Fe(II) sorbed on magnetite is due to a process similar to negative (n) doping of a solid, which increases the stoichiometry of the magnetite and alters the bulk redox properties of the particle to make reduction more favorable.

The reaction of aqueous Fe(II) with Fe(III) oxides is a complex process, comprising sorption, electron transfer, and in some cases, reductive dissolution and transformation to secondary minerals. To better understand the dynamics of these reactions, we measured the extent and rate of Fe isotope exchange between aqueous Fe(II) and goethite using a 57Fe isotope tracer approach. We observed near-complete exchange of Fe atoms between the aqueous phase and goethite nanorods over a 30-day time period. Despite direct isotopic evidence for extensive mixing between the aqueous and goethite Fe, no phase transformation was observed, nor did the size or shape of the goethite rods change appreciably. High-resolution transmission electron microscopy images, however, appear to indicate that some recrystallization of the goethite particles may have occurred. Near-complete exchange of Fe between aqueous Fe(II) and goethite, coupled with negligible change in the goethite mineralogy and morphology, suggests a mechanism of coupled growth (via sorption and electron transfer) and dissolution at separate crystallographic goethite sites. We propose that sorption and dissolution sites are linked via conduction through the bulk crystal, as was recently demonstrated for hematite. Extensive mixing between aqueous Fe(II) and goethite, a relatively stable iron oxide, has significant implications for heavy metal sequestration and release (e.g., arsenic and uranium), as well as reduction of soil and groundwater contaminants.

The rate and extent of hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX) transformation was measured in the presence of carbonate and sulfate green rust suspended in solutions containing common groundwater anions. Formaldehyde (HCHO), nitrous oxide gas (N2O(g)), and ammonium (NH4+) were the major end products, accounting for about 70% of the carbon mass balance and about half of the nitrogen mass balance. Results from experiments with both 14C-RDX and LC-MS analysis indicate that the remaining carbon products are soluble and most likely small (< 50 Da). The transient appearance of 1,3-dinitro-5-nitroso-1,3,5-triazacyclohexane (MNX), 1,3-dinitroso-5-nitro-1,3,5-triazacyclohexane (DNX), and 1,3,5-trinitroso-1,3,5-triazacyclohexane (TNX) indicate that some nitro-group reduction occurred. The kinetics of RDX transformation was rapid with a half-life of less than an hour in a pH 7.0 KBr solution. Little difference in rates of RDX transformation or product distribution was observed between carbonate and sulfate green rust, and an apparent reaction order of 1.0 was measured with respect to Fe(II) in both green rusts. Phosphate anions completely inhibited RDX reduction, and carbonate and sulfate anions resulted in slower kinetics, and in some cases, an initial lag period, compared to bromide and chloride. Our results suggest that green rusts may contribute to abiotic natural attenuation of RDX in Fe-rich subsurface environments, but that it will be important to consider groundwater composition when assessing rates of attenuation.

We collected Mössbauer spectra of 57Fe(II) interacting with 56hematite (alpha-Fe2O3) over a range of Fe(II) concentrations and pH values to explore whether a sorbed Fe(II) species would form. Several models of Fe(II) sorption (e.g., surface complexation models) assume that stable, sorbed Fe(II) species form on ligand binding sites of Fe(III) oxides and other minerals. Model predictions of changes in both speciation and concentration of sorbed Fe(II) species are often invoked to explain Fe(II) sorption patterns, as well as rates of contaminant reduction and microbial respiration of Fe(III) oxides. Here we demonstrate that, at low Fe(II) concentrations, sorbed Fe(II) species are transient and quickly undergo interfacial electron transfer with structural Fe(III) in hematite. At higher Fe(II) concentrations, however, we observe the formation of a stable, sorbed Fe(II) phase on hematite that we believe to be the first spectroscopic confirmation for a sorbed Fe(II) phase forming on an iron oxide. Low-temperature Mössbauer spectra suggest that the sorbed Fe(II) phase contains varying degrees of Fe(II)-Fe(II) interaction and likely contains a mixture of adsorbed Fe(II) species and surface precipitated Fe(OH)2(s). The transition from Fe(II)-Fe(III) interfacial electron transfer to formation of a stable, sorbed Fe(II) phase coincides with the macroscopically observed change in isotherm slope, as well as the estimated surface site saturation suggesting that the finite capacity for interfacial electron transfer is influenced by surface properties. The spectroscopic demonstration of two distinctly different sorption endpoints, that is an Fe(III) coating formed from electron transfer or a stable, sorbed Fe(II) phase, challenges us to reconsider our traditional interpretations and modeling of Fe(II) sorption behavior (as well as, we would argue, of any other redox active sorbate-sorbent couple).

Flow-through columns packed with "aged" zero-valent iron (ZVI) between layers of soil and sand were constructed to mimic a one-dimensional permeable reactive iron barrier (PRB). The columns were continuously fed RDX (hexahydro-1,3,5-trinitro-1,3,5-triazine, ca. 18 mg l(-1)) for over one year. Two columns were bioaugmented with dissimilatory iron reducing bacteria (DIRB) Shewanella algae BrY or Geobacter metallireducens GS-15 to investigate their potential to enhance the reactivity of aged iron by reductive dissolution of passivating iron oxides or via production of biogenic reactive minerals. A third column was not bioaugmented to evaluate colonization by indigenous soil microorganisms. [14C]-RDX was completely removed in all columns at the start of the iron layer, and concentration profiles showed rapid and sustainable RDX removal over one year; however, a phylogenetic profile conducted after one year using DGGE analysis of recovered DNA did not detect S. algae BrY or G. metallireducens in their respective columns. Bacterial DNA was recovered from within the ZVI. Several unidentified 14C-labeled byproducts were present in the effluent of all columns. Dissolved 14C removal and the detection of dissolved inorganic 14C in these columns (but not in the sterile control) suggest microbial-mediated mineralization of RDX and sorption/precipitation of degradation products. Enhanced RDX mineralization in bioaugmented columns was temporary relative to the indigenously colonized column. However, shorter acclimation periods associated with bioaugmented PRBs may be desirable for rapid RDX mineralization, thereby preventing breakthrough of potentially undesirable byproducts. Overall, these results show that high RDX removal efficiency by ZVI-PRBs is achievable and sustainable and that the efficacy and start-up of ZVI-PRBs might be enhanced by bioaugmentation.

Microbial respiration of Fe(III) oxides has been shown to produce reduced Fe phases that are capable of transforming a variety of oxidized contaminants. Little data, however, are available on how these Fe phases evolve over time and how this evolution may affect their ability to reduce contaminants. Here,the evolution and reactivity of biologically reduced ferrihydrite were monitored over a period of 14 months. Solids were collected from a culture of Geobacter metallireducens (GS-15) thatwas incubated with ferrihydrite (as the electron acceptor) for 0, 7, 10, 20, 75, and 400 days. Mineralogical composition and surface area of the biologically reduced solids were characterized using Mössbauer spectroscopy, X-ray diffraction, and BET with N2 adsorption. By day 10, ferrihydrite began to transform, and a nanoparticle magnetite/maghemite phase, as well as two ferrous phases, was observed. One of the ferrous phases was identified as siderite, whereas the other could not be positively identified. Likely candidates, however, include Fe(OH)2(s) or an adsorbed Fe(II) species. Over the next few months, ferrihydrite was completely reduced and evolved into a mixture containing about 70% magnetite/maghemite, 19% siderite, and 11% of the second Fe(II) phase. The effect of incubation time on the reactivity of the biologically reduced solids was evaluated by measuring the kinetics of hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX) transformation. The only products observed were the three reduced nitroso products. Rate coefficients (k) for RDX transformation were dramatically influenced by incubation time with half-lives of about 1 month observed in the presence of solids incubated for 10 and 20 days, 3 months with solids incubated for 75 days, and negligible removal with solids incubated for 400 days. The loss of reactivity was not directly correlated to any one mineralogical variable but may be due to particle size or surface chemistry changes in the reactive Fe phase or to cell die-off and the accumulation of cell lysis products after consumption of the electron acceptor. The dramatic effect of incubation time on the rate of RDX removal highlights a potential limitation of studying complex systems, as we have here, in batch reactors and suggests that incubation time is an important variable to consider when measuring and comparing rates of contaminant reduction.

Perchlorate was reduced by a mixed bacterial culture over a pH range of 7.0-8.9. Similar rates of perchlorate reduction were observed between pH 7.0 and 8.5, whereas significantly slower reduction occurred at pH 8.9. Addition of iron metal, Fe(0), to the mixed bacterial culture resulted in slower rates of perchlorate reduction. Negligible perchlorate reduction was observed under abiotic conditions with Fe(0) alone in a reduced anaerobic medium. The inhibition of perchlorate reduction observed in the presence of Fe(0) is in contrast to previous studies that have shown faster rates of contaminant reduction when bacteria and Fe(0) were combined compared to bacteria alone. The addition of Fe(0) resulted in a rise in pH, as well as precipitation of Fe minerals that appeared to encapsulate the bacterial cells. In experiments where pH was kept constant, the addition of Fe(0) still resulted in slower rates of perchlorate reduction suggesting that encapsulation of bacteria by Fe precipitates contributed to the inhibition of the bacterial activity independent of the effect of pH on bacteria. These results provide the first evidence linking accumulation of iron precipitates at the cell surface to inhibition of environmental contaminant degradation. Fe(0) was not a suitable amendment to stimulate perchlorate-degrading bacteria and the bacterial inhibition caused by precipitation of reduced Fe species may be important in other combined anaerobic bacterial-Fe(0) systems. Furthermore, the inhibition of bacterial activity by iron precipitation may have significant implications for the design of in situ bioremediation technologies for treatment of perchlorate plumes.

Using the isotope specificity of 57Fe Mössbauer spectroscopy, we report spectroscopic observations of Fe(II) reacted with oxide surfaces under conditions typical of natural environments (i.e., wet, anoxic, circumneutral pH, and about 1% Fe(II)). Mössbauer spectra of Fe(II) adsorbed to rutile (TiO2) and aluminum oxide (Al2O3) show only Fe(II) species, whereas spectra of Fe(II) reacted with goethite (alpha-FeOOH), hematite (alpha-Fe2O3), and ferrihydrite (Fe5HO8) demonstrate electron transfer between the adsorbed Fe(II) and the underlying iron(III) oxide. Electron-transfer induces growth of an Fe(III) layer on the oxide surface that is similar to the bulk oxide. The resulting oxide is capable of reducing nitrobenzene (as expected based on previous studies), but interestingly, the oxide is only reactive when aqueous Fe(II) is present. This finding suggests a novel pathway for the biogeochemical cycling of Fe and also raises important questions regarding the mechanism of contaminant reduction by Fe(II) in the presence of oxide surfaces.

RDX (hexahydro-1,3,5-trinitro-1,3,5-triazine), a nitramine explosive, is often found as a subsurface contaminant at military installations. Though biological transformations of RDX are often reported, abiotic studies in a defined medium are uncommon. The work reported here was initiated to investigate the transformation of RDX by ferrous iron (Fe(II)) associated with a mineral surface. RDX is transformed by Fe(II) in aqueous suspensions of magnetite (Fe3O4). Negligible transformation of RDX occurred when it was exposed to Fe(II) or magnetite alone. The sequential nitroso reduction products (MNX, DNX, and TNX) were observed as intermediates. NH4+, N2O, and HCHO were stable products of the transformation. Experiments with radiolabeled RDX indicate that 90% of the carbon end products remained in solution and that negligible mineralization occurred. Rates of RDX transformation measured for a range of initial Fe(II) concentrations and solution pH values indicate that greater amounts of adsorbed Fe(II) result in faster transformation rates. As pH increases, more Fe(II) adsorbs and k(obs) increases. The degradation of RDX by Fe(II)-magnetite suspensions indicates a possible remedial option that could be employed in natural and engineered environments where iron oxides are abundant and ferrous iron is present.

The reactions of eight model contaminants with nine types of granular Fe(0) were studied in batch experiments using consistent experimental conditions. The model contaminants (herein referred to as "reductates" because they were reduced by the iron metal) included cations (Cu2+), anions (CrO4(2-), NO3(-), and 5,5',7,7'-indigotetrasulfonate), and neutral species (2-chloroacetophenone, 2,4,6-trinitrotoluene, carbon tetrachloride, and trichloroethene). The diversity of this range of reductates offers a uniquely broad perspective on the reactivity of Fe(0). Rate constants for disappearance of the reductates vary over as much as four orders of magnitude for particular reductates (due to differences in the nine types of iron) but differences among the reductates were even larger, ranging over almost seven orders of magnitude. Various ways of summarizing the data all suggest that relative reactivities with Fe(0) vary in the order Cu2+, 5,5',7,7'-indigotetrasulfonate > 2-chloroacetophenone, 2,4,6-trinitrotoluene > carbon tetrachloride, CrO4(2-) > trichloroethene > NO3(-). Although the reductate has the largest effect on disappearance kinetics, more subtle differences in reactivity due to the type of Fe(0) suggests that removal of CrO2(2-) and NO3(-) (the inorganic anions) involves adsorption to oxides on the Fe(0), whereas the disappearance kinetics of all other types of reductants is favored by reduction on comparatively oxide-free metal. Correlation analysis of the disappearance rate constants using descriptors of the reductates calculated by molecular modeling (energies of the lowest unoccupied molecular orbitals, LUMO, highest occupied molecular orbitals, HOMO, and HOMO-LUMO gaps) showed that reactivities generally decrease with increasing E(LUMO) and increasing E(GAP) (and, therefore, increasing chemical hardness eta).

To evaluate the effect of potential interactions between methanogenic bacteria and iron sulfide minerals during transformation of 1,1,1-trichloroethane (1,1,1-TCA), we measured the kinetics of 1,1,1-TCA transformation by mackinawite (FeS(1 - x), but abbreviated as FeS) and a methanogenic consortium enriched on lactate (termed LEC). Results from batch kinetic experiments show that 1,1,1-TCA transformation by FeS and resting LEC can be described by second-order rate expressions, with rates depending on 1,1,1-TCA concentration (M), FeS surface area concentration (m2 L(-1)), and LEC concentration (as measured by mg L(-1) volatile suspended solids (VSS)). In reactors containing FeS alone, 1,1-dichloroethane (1,1-DCA) and 2-butyne were identified as products, but only accounted for 6% of the 1,1,1-TCA transformed. In reactors containing LEC alone, the only identified product was 1,1-DCA, which accounted for 46 +/- 8% of the 1,1,1-TCA transformed. Supernatant from LEC-alone reactors also transformed 1,1,1-TCA, suggesting that 1,1,1-TCA may be transformed by some non-cell component (such as a extracellular compound excreted by the organisms) that either reacts directly with 1,1,1-TCA or with the abiotic media to form a reactive species. Comparison of 1,1,1-TCA transformation rates from experiments with combinations of FeS (varying surface area concentrations) and LEC (varying VSS concentrations) to those with just FeS alone or LEC alone suggests some synergism occurs between the two reactive species. Observed enhancements took the form of faster 1,1,1-TCA transformation and faster 1,1-DCA appearance but less production of 1,1-DCA per unit of 1,1,1-TCA transformed. These observations suggest that the faster 1,1,1-TCA transformation in the combined systems (compared to the FeS-alone and LEC-alone experiments) is due to increased reactivity of both FeS and LEC, possibly due to production of soluble microbial products that make the FeS more reactive or less inhibition of LEC by 1,1,1-TCA due to FeS transformation of 1,1,1-TCA.

The kinetics of nitrate, nitrite, and Cr(VI) reduction by three types of iron metal (Fe0) were studied in batch reactors for a range of Fe0 surface area concentrations and solution pH values (5.5-9.0). At pH 7.0, there was only a modest difference (2-4x) in first-order rate coefficients (k(obs)) for each contaminant among the three Fe0 types investigated (Fisher, Peerless, and Connelly). The k(obs) values at pH 7.0 for both nitrite and Cr(VI) reduction were first-order with respect to Fe0 surface area concentration, and average surface area normalized rate coefficients (kSA) of 9.0 x 10(-3) and 2.2 x 10(-1) L m(-2) h(-1) were determined for nitrite and Cr(VI), respectively. Unlike nitrite and Cr(VI), Fe0 surface area concentration had little effect on rates of nitrate reduction (with the exception of Connelly Fe0, which reduced nitrate at slower rates at higher Fe0 surface areas). The rates of nitrate, nitrite, and Cr(VI) reduction by Fisher Fe0 decreased with increasing pH with apparent reaction orders of 0.49 +/- 0.04 for nitrate, 0.61 +/- 0.02 for nitrite, and 0.72 +/- 0.07 for Cr(VI). Buffer type had minimal effects on reduction rates, indicating that pH was primarily responsible for the differences in rate. At high pH values, Cr(VI) reduction ceased after a short time period, and negligible nitrite reduction was observed over 48 h.

Recent studies of contaminant reduction by zero-valent iron metal (Fe0) have highlighted the role of iron oxides at the metal-water interface and the effect that sorption has at the oxide-water interface on contaminant reduction kinetics. The results suggest that a variety of organic surface-active substances might enhance or inhibit contaminant degradation, depending on the degree to which they promote solubilization, sorption. and/or reaction. Of particular interest is the effect of natural organic matter (NOM), because of its ubiquitous presence in natural waters and amphiphilic properties; anthropogenic surfactants, because of their use in groundwater remediation; and certain quinones, which represent the redox-active functional groups associated with NOM. In this study, no well-defined effects of these substances were found on the reduction of nitro benzene by Fe0, but the reduction of carbon tetrachloride and trichloroethylene (TCE) was inhibited by NOM. Results with carbon tetrachloride showed that the inhibitory effect of humic acids was greater than fulvic acids, but that several quinonoid NOM model compounds (juglone, lawsone. and anthraquinone disulfonate) increased the rate of reduction by Fe0. Isotherms for adsorption of TCE and NOM onto Fe0 showed evidence of competition for surface sites.

The kinetics of Cr(VI) reduction to Cr(III) by carbonate green rust were studied for a range of reactant concentrations and pH values. Carbonate green rust, [FeII4FeIII2(OH)12][4H2O x CO3], was synthesized by induced hydrolysis (i.e., coprecipitation) of an Fe(ll)/Fe(III) solution held at a constant pH of 8. An average specific surface area of 47 +/- 7 m2 g(-1) was measured for five separate batches of freeze-dried green rust precipitate. Heterogeneous reduction by Fe(II) associated with the carbonate green rust appears to be the dominant pathway controlling Cr(VI) loss from solution. The apparent stoichiometry of the reaction between ferrous iron associated with green rust ([Fe(II)GR]) and Cr(VI) was slightly higherthan the expected 3:1 ratio, possibly due to the presence of other oxidants, such as oxygen, protons, or interlayer carbonate ions. The rate of Cr(VI) reduction was proportional to the green rust surface area concentration, and psuedo-first-order rate coefficients (kobs) ranging from 1.2 x 10(-3) to 11.2 x 10(-3) s(-1) were determined. The effect of pH was small with a 5-fold decrease in rate with increasing pH (from 5.0 to 9.0). At low Cr(VI) concentrations (<200 microM), the rate of reaction was first order with respect to Cr(VI) concentration, whereas, at high Cr(VI) concentrations, rates appearto deviate from first-order kinetics and approach a constant value. Estimated amounts of surface Fe(II) and total Fe(II) suggest that the deviation from first-order kinetics observed at higher Cr(VI) concentrations and the 50-fold decrease in rate observed upon three sequential exposures to Cr(VI) is due to exhaustion of available Fe(II).

To evaluate the importance of external mass transport on the overall rates of contaminant reduction by iron metal (Fe0), we have compared measured rates of surface reaction for nitrobenzene (ArNO2) to estimated rates of external mass transport in a permeable reactive barrier (PRB). The rate of surface reaction was measured at a polished Fe0 rotating disk electrode (RDE) in an electrochemical cell, and the rate of mass transport was estimated from a correlation for mass transport in packed-bed reactors. The kinetics of ArNO2 reduction were studied in pH 8.4 borate buffer at a potential below which an oxide film would form. The cathodic current measured in this system was dependent on the electrode rotation rate, and the measured first-order heterogeneous rate coefficient for surface reaction [krxn = (1.7 +/- 0.2) x 10(-3) cm s-1] was about 10 times faster than the first-order mass transport rate coefficient (kmt approximately 2 x 10(-4) cm s-1) estimated for PRBs. The similarity between rates of surface reaction and mass transport suggest that it may be important to consider mass transport processes in the design of PRBs for contaminants such as nitroaromatics that are highly reactive with Fe0.

Permeable reactive barriers (PRBs) are receiving a great deal of attention as an innovative, cost-effective technology for in situ clean up of groundwater contamination. A wide variety of materials are being proposed for use in PRBs, including zero-valent metals (e.g., iron metal), humic materials, oxides, surfactant-modified zeolites (SMZs), and oxygen- and nitrate-releasing compounds. PRB materials remove dissolved groundwater contaminants by immobilization within the barrier or transformation to less harmful products. The primary removal processes include: (1) sorption and precipitation, (2) chemical reaction, and (3) biologically mediated reactions. This article presents an overview of the mechanisms and factors controlling these individual processes and discusses the implications for the feasibility and long-term effectiveness of PRB technologies.