B11B-0346
Uranium Association With Framboidal Pyrite and Magnetite-like Phases in Alluvial Sediments
During the recent drilling campaign at the Integrated Field Challenge (IFC) Site located in Rifle, Colorado, a former U milling site, several sediment cores were obtained at different depths prior to acetate injection, within vadose and saturated zones of the subsurface. The cores represented a cross section of sediment conditions that ranged from typical aquifer sediment (minimally reduced) to highly naturally bioreduced sediment from the saturated zone. The objective of the study was to determine the role of the Fe-minerals on U attenuation under the site-specific conditions at Rifle. The sediment samples were characterized by XRD, Mossbauer spectroscopy, U-XANES, micro-XRF, and various electron microscopic techniques (SEM-EMPA, SEM-FIB, and TEM-SAED). The sediments were analyzed for aqueous Fe(II), Fe(III), and U, labile U by bicarbonate-carbonate extraction, acid volatile sulfide (AVS), and inorganic and organic carbon contents. The combined spectroscopic and analytical studies showed that: a) Fe-oxides, Fe-containing clays, and pyrite are the dominant Fe-containing minerals. Mössbauer spectroscopy identified the Fe-oxides minerals to be magnetite, hematite, and goethite, while the predominant clay mineral was clinochlore, based on the XRD results, b) U and Fe(II) contents were substantially higher in the naturally bioreduced sediment samples, probably due to microbial Fe, sulfate, and U reduction stimulated by the natural organic matter, and c) U was primarily associated, with framboidal and euhedral pyrites and magnetite-like phases, and occurred as both U(IV) and U(VI). The observation of U association with pyrite was rather unexpected since dissolved U amounts increased when Fe-sulfides of unknown nature were precipitated when the Rifle U-contaminated aquifer's microbial community transitions from iron to sulfate reduction conditions. We believe this is the first time that U was found to be associated with framboidal pyrite and magnetite-like phases at a DOE contaminated site. However, the role of the framboidal pyrite, which occurred in different sizes (up to 10-micron in diameter), microcrystal morphologies, and packing densities, and the role of magnetite-like phases in attenuation of U in natural aquifers, is not well understood. Further characterization and additional experimental work is needed to address these issues.
B11B-0347
U(VI) Uptake and Reduction by Fe(II)-Bearing Secondary Mineralization Products of the Bioreduction of Fe(III) Oxides by Dissimilatory Fe(III)-Reducing Bacteria
Biogenic Fe(II) phases (magnetite, green rust, siderite, vivianite, etc.) provide a reservoir of reducing capacity in the subsurface that may contribute to the reduction of contaminants such as U(VI). In this study we examined the potential for the uptake and reduction of U(VI) in the presence of biogenic green rust, (BioGR) biogenic magnetite (BioMAG), and biogenic siderite (BioSID) resulting from the reduction of Fe(III) oxides by Shewanella putrefaciens CN32. Suspensions of biogenic Fe(II) phases were pasteurized (70 °C for 1 h) to eliminate the potential for microbial reduction of U(VI) and washed repeatedly to remove any soluble reductants. The suspensions were then spiked with uranyl chloride solution. Within 48 h, the total solution-phase U(VI) concentrations decreased from 500 μM to 1.5 μM in the U-BioGR system, to 392 μM in the U-BioMAG system, and to 472 μM in the U-BioSID system, as determined by ICP- OES. Analysis of the samples by U LIII extended X-ray absorption fine structure spectroscopy (EXAFS) indicated that despite a stoichiometric excess of Fe(II), no more than 6% of U(VI) was reduced in the U- BioSID system, and no more than 22% of U(VI) was reduced in the U-BioMAG system. For comparison, in the U-BioGR system, no less than 80% of U(VI) was reduced to U(IV). Uptake of U(VI) by BioGR and BioMAG was accompanied by the formation of nanoparticulate uraninite. The U EXAFS data from the U-BioSID system was consistent with partial U(VI)/U(IV) substitution for Fe(II) in the surface layer of siderite particles and adsorption of U(IV). Our results clearly demonstrate that there are significant differences in reactivity among biogenic Fe(II) phases with respect to the uptake and reduction of U(VI).
B11B-0348
Stability of Uranium Incorporated into Fe(hydr)oxide Structure upon Oxidation and Under Fluctuating Redox Conditions
Owing to both ecosystem and human health consequences, understanding uranium's potential for mobility in environmental settings is important. In anaerobic soils and sediments, the more mobile, oxidized form of uranium (UO22+) may be reduced through biological or chemical pathways to U(IV), forming sparingly soluble solids. However, formation of ternary calcium-carbonato complexes of uranium may limit reduction. Additionally, authogenic UO2 is potentially susceptible to reoxidation in the presence of commonly occurring oxidants including molecular oxygen. Therefore, determining reaction pathways of uranium that lead to solids stable (i.e., having limited solubility) under both aerobic and anaerobic conditions is critical for limiting dissolved concentrations and migration of uranium. We examined uranium incorporation into transforming Fe(hydr)oxides and secondary fate under cyclic oxidizing and reducing conditions. In reducing environments containing ferrihydrite, 3 mM Fe(II), 4 mM Ca, and 3.8 mM HCO3-, aqueous uranium concentrations decrease from 0.16 mM to 0 mM in a period of 15 d. Increasing Fe(II) concentration to 10 mM results in aqueous uranium concentrations decreasing from 0.16 mM to 0 mM within 5 d. In identical systems without Fe(II), uranium decreases from 0.16 to 0.10 mM after 5 days of anoxic incubation. Changes in solid phase color during incubations suggest changes in iron mineralogy as well as the potential for formation of reduced uranium solid phases. Uranium EXAFS and XANES analysis reveal oxidized uranium both adsorbed on and incorporated in the structure of goethite and magnetite. Additionally, uranium reduction and concomitant UO2(s) precipitation occurs within ferrihydrite systems reacted with 10 mM Fe(II). By contrast, U(VI) is retained strictly by adsorption in systems without Fe(II). Our results thus reveal that the fate of uranium is intimately tied not only to anaerobic/anaerobic conditions, but also to geochemical conditions controlling uranium speciation and to the fate of iron.
B11B-0349
Mesophilic, Circumneutral Anaerobic Iron Oxidation as a Remediation Mechanism for Radionuclides, Nitrate and Perchlorate
Iron oxidation is a novel anaerobic metabolism where microorganisms obtain reducing equivalents from the oxidization of Fe(II) and assimilate carbon from organic carbon compounds or CO2. Recent evidence indicates that in combination with the activity of dissimilatory Fe(III)-reducing bacteria, anaerobic microbial Fe(II) oxidation can also contribute to the global iron redox cycle. Studies have also proved that Fe(II)- oxidation is ubiquitous in diverse environments and produce a broad range of insoluble iron forms as end products. These biogenic Fe(III)-oxides and mixed valence Fe minerals have a very high adsorption capacity of heavy metals and radionuclides. Adsorption and immobilization by these biogenic Fe phases produced at circumneutral pH, is now considered a very effective mode of remediation of radionuclides like Uranium, especially under variable redox conditions. By coupling soluble and insoluble Fe(II) oxidation with nitrate and perchlorate as terminal electron acceptors in-situ, anaerobic Fe-oxidation can also be used for environmental cleanup of Fe through Fe-mineral precipitation, as well as nitrate and perchlorate through reduction. Coupling of Fe as the sole electron and energy source to the reduction of perchlorate or nitrate boosts the metabolism without building up biomass hence also taking care of biofouling. To understand the mechanisms by which microorganisms can grow at circumneutral pH by mesophilic, anaerobic iron oxidation and the ability of microorganisms to reduce nitrate and perchlorate coupled to iron oxidation recent work in our lab involved the physiological characterization of Dechlorospirillum strain VDY which was capable of anaerobic iron-oxidation with either nitrate or perchlorate serving as terminal electron acceptor. Under non-growth conditions, VDY oxidized 3mM Fe(II) coupled to nitrate reduction, and 2mM Fe(II) coupled to perchlorate reduction, in 24 hours. It contained a copy of the RuBisCO cbbM subunit gene which was differentially regulated. With perchlorate as the sole terminal electron acceptor, cbbM was expressed under autotrophic growth with hydrogen as the electron donor but not during heterotrophic growth on acetate, indicating a putative carbon-fixation pathway. Similarly, Ferrutens uranioxidens strain 2002 was also capable of autotrophic growth during nitrate-dependent iron oxidation, although the carbon fixation pathway has yet to be identified. Anoxic XPRD analysis of the biogenic end products of nitrate-dependent Fe(II) oxidation by Diaphorobacter sp. strain TPSY and strain 2002 indicated the gradual appearance of green rust (GR II) with cacoxenite and lepidocrocite from the precursor vivianite over 81 days. SEM and TEM showed the presence of hexagonal plate like crystals surrounding the bacterial cells whose morphology closely resembled GR II, indicating a very low redox potential and a weakly acidic to weakly basic pH. Mixotrophic growth incubations of strain TPSY with 1, 5 and 10 mM Fe(II) showed markedly different end products. The identity of the mineral phases and the reason behind this difference is currently under investigation.
B11B-0350
Influence of Iron Reducing Bacteria on Phosphate and Arsenate Release and Sequestration Onto Iron Oxides
This study investigates the potential for iron oxides to capture phosphate and arsenate. While iron oxides
clearly bind phosphate and arsenate, one should not assume that iron oxides would straightforwardly trap
these contaminants. Instead, iron oxides in shallow groundwater sediments undergo chemical transformations
that could lead to the release of sorbed contaminants. Most notably, scientists commonly observe that the
biologically driven reduction of ferric iron (Fe[III]) to ferrous iron (Fe[II]) releases phosphate and/or arsenate
under some conditions. Despite this observation, it is generally unclear whether iron reduction will lead to
greater or lesser mobility of phosphate and arsenic. This is because sorbed phosphate and arsenate may
dissolve into the aqueous phase under iron-reducing conditions, thereby becoming more mobile, while under
the same reducing conditions, phosphate and arsenate may precipitate with ferrous iron solids, thus
decreasing their mobility. Further complications arise when the system is exposed to cycles of oxic and anoxic
conditions. Under oxic conditions, ferrous iron will oxidize and form new iron oxides, while existing solid
ferrous oxides will transform to new ferric oxides. In shallow groundwater, flooding and drought lead to cycles
of oxic and anoxic conditions that could produce cyclic patterns in iron redox states and in the mobility of
arsenic and phosphate. To better predict the evolution of arsenate and phosphate in the field, this research
employs a simplified laboratory model. This study tracks the fate of arsenate and phosphate sorbed onto
goethite, a common sediment component, as geobacter sulfurreducens reduces iron from Fe[III] to Fe[II], and
after re-oxidation of reduced iron. By studying that particular phenomenon, this research sheds light on how
microbes and one soil constituent influence the mobility of arsenate and phosphate contamination. The
results of this study show that freshly precipitated ferrous iron solids can sequester arsenate and phosphate
as they remove these anions from the aqueous phase and incorporate them into the matrix of freshly
precipitated ferrous iron solids.
http://www.princeton.edu/~lmacdona/iron-oxides.html
B11B-0351
Composition and arsenic-attenuating capacity of biogenic iron (hydr)oxide flocs at the Lava Cap Mine Superfund Site, Nevada County, CA.
The Lava Cap Mine Site (LCMS) is on the National Priority List due to the elevated human health risk presented by the catastrophic release of several thousand cubic meters of arsenic (As) enriched tailings (average: 500 ppm As ) from the site. These tailings were released into a creek and lake (former tailings retention pond) in a low-density residential area where ground water is the primary source of drinking water. Although oxidation of iron (Fe) sulfides (pyrite and arsenopyrite) from tailings are the main sources of As and Fe, buffering by carbonate minerals prevents formation of acidic waters. Macroscopic accumulations of fluffy Fe (hydr)oxide are observed suspended in the water column or at the sediment-water interface in creeks, ponds, and seeps of the LCMS. Microscopic analysis indicates that the Fe (hydr)oxide is predominantly associated with the sheaths of bacteria identified as members of the genus Leptothrix, which are known to enzymatically oxidize Fe and manganese (Mn) under oligotrophic, near-neutral, sub oxic conditions. Both Fe- encrusted Leptothrix sheaths (which are largely devoid of cells) and free aggolmerations of Fe hydr(oxide) support morphologically distinct Eubacteria whose identity is currently under investigation. Dried biogenic Fe (hydr)oxide averages 4.4 % organic carbon, 20.2 % Fe, and 0.91% As (9100 ppm), making it attractive as a potential natural biosorbent for As and Fe. Water flow rate is a very important control on the amount of As retained in biogenic Fe (hydr)oxide flocs, based on monitoring of a natural passive bioreactor system. In addition, a pond with nearly stagnant water accumulated approximately one order of magnitude more As (dried) than a seep site with faster-running water, even though there was only a 5-fold difference in their median filtered (0.45 micron) arsenic concentrations. Most Probable Number estimates and analysis of PCR amplicons of Eubacterial DNA indicate that populations of Fe-, As-, and sulfate- reducing microorganisms are present within the Leptothrix-supported Fe (hydr)oxide flocs, but their abundances are low under the oxic conditions prevailing in surface creeks. These microbial populations could potentially re-mobilize As if prevailing conditions changed from sub- to anoxic, and if more organic carbon entered the system.
B11B-0352
Evidence for Anaerobic Methane Oxidation Under Iron-Reducing Conditions in a Crude-Oil Contaminated Aquifer
Although anaerobic methane oxidation (AMO) under iron reducing conditions is energetically feasible, its existence is still an open question. At a crude oil spill site near the town of Bemidji, MN, methanogenic degradation of entrapped oil floating at the water table has been occurring for more than 20 years. In the anaerobic portion of the hydrocarbon plume there is evidence for AMO under iron-reducing conditions between 75 m and 120 m downgradient of the oil body. In this zone, dissolved methane concentrations decrease steadily from over 0.6 mmol/L to less than 0.06 mmol/L. Decreases in dissolved methane are accompanied by an increase in δ13C-CH4 indicating that methane attenuation occurs through microbially mediated oxidation. The dissolved methane decrease of ~0.5 mmol/L occurs where dissolved sulfate is below 0.06 mmol/L, dissolved oxygen is below 3 μmol/L, and nitrate is below 0.02 mmol/L. Together these electron acceptors can account for degradation of only 0.07 mmol/L of methane. Moreover, hydrocarbon contaminants contribute an additional ~1 mmol/L total organic carbon in this area. Active iron- reduction is indicated by dissolved iron concentrations exceeding 0.15 mmol/L. To investigate the sediment bioavailable iron and microbial populations, 2-m-long cores were collected at four locations spaced 15 m horizontally and sampled at 50-cm-depth intervals. Values of bioavailable Fe(III) averaged 8 mmol/kg (n=16), which is over eight times the amount required to degrade 0.5 mmol/L methane. Geobacter detected by qPCR averaged over 2,600/g, while sulfate reducing bacteria were detected in only four samples with 100/g being the highest abundance measured. Laboratory incubations were performed with eight sediment samples from these cores. For each location, 1 g of sediment was added to 10 mL anaerobic medium and amended with amorphous FeOOH, Fe(III)NTA, sulfate, or nitrate. Loss of methane occurred in 33% of the iron-amended treatments, 25% of the nitrate-amended treatments, and 8% of the sulfate-amended treatments. The greatest loss occurred in sediments from the downgradient edge of the anaerobic zone where amorphous FeOOH- and nitrate-amended incubations exhibited ~30% loss of methane in 118 days. Loss also occurred in two out of five Fe(III)NTA treatments from this location, but no loss occurred in the sulfate-amended treatments. Overall these data are suggestive of AMO under iron-reducing conditions in favorable locations at the Bemidji, MN, site.
B11B-0353
Degradation of Nitrobenzene Using Bio-Reduced Fe-Clays: Progress Towards the Development of an in-situ Groundwater Remediation Technology
Clay minerals are widely used in agricultural, industrial and environmental engineering applications due to their specific physical and chemical properties and their high abundance in soils in sediments. Currently however, Fe-bearing clays are not widely exploited in these applied fields. Fe-rich smectites, such as nontronite, can contain up to 20wt% of Fe2O3 as structural Fe(III) and if a suitable electron donor is available, this Fe(III) can be utilized by Fe-reducing bacteria as a terminal electron acceptor. When reduced, the overall reactivity of Fe-smectites changes, particularly where interactions with water and various organic compounds are involved. For instance, the presence of reduced Fe-smectites has been found to induce the degradation of certain organic contaminants found in groundwaters and the subsurface, e.g. chlorinated aliphatics and nitroaromatic compounds. The goal of this study is to develop an in-situ groundwater remediation technology that targets redox- sensitive organics, in the form of a permeable Bio Fe-clay barrier. To achieve this, the iron-reducing bacterium Shewanella algae BrY was first used to reduce structural FeIII in <2micron fractions of the Fe- rich smectite nontronite (NAu-2, 41.74wt% Fe2O3) and a Fe-bearing montmorrillonite (Speeton Clay, Yorkshire, UK, ~8wt% Fe2O3). S. algae BrY was able to reduce structural FeIII within these clays to maximum Fe(II)/Fe(II)+Fe(III) ratios 0.34 and 0.19 for the nontronite and Speeton Clay, respectively, in the presence and absence of the electron shuttle, AQDS (9, 10-anthraquinone-2, 6-disulfonic acid). These results are novel because the capability of S. algae BrY to reduce structural Fe(III) in smectite clays has not previously been tested. Nitrobenzene was selected as the test redox-sensitive organic compound as it is a common subsurface contaminant and is of global ecotoxicological concern. To test the capability of bio- reduced Fe-clays to transform nitrobenzene to aniline (the less toxic and more stable degradation product of nitrobenzene), nontronite suspensions with reduction levels of 20% and 30% were spiked with various concentrations of nitrobenzene and monitored for 5 days. Results showed that when reduced clay, S. algae BrY and AQDS were present, 100% of the nitrobenzene had been transformed to aniline within 24 hours. Further recent results suggest that bio-reduced nontronite alone is also capable of degrading nitrobenzene but at a slower rate than when AQDS is present. Future experiments will constrain absorption and degradation rates of nitrobenzene in contact with the reduced Fe-clays and the role(s) of the Fe-reducing bacteria. Results to date will be presented.
B11B-0354
Reductive Dechlorination of Carbon Tetrachloride by Soil With Ferrous and Bisulfide
Batch and column experiments were conducted to investigate the effect of concentration of reductants, contact time to activate reductive capacity, and pH on reductive dechlorination by soil with Fe(II) and HS- in this study. Carbon tetrachloride (CT) was used as a representative target organic compound. Sorption kinetic and isotherm tests were performed to investigate the influence of adsorption on the soil surface. Target compound in the soil suspension reached sorption equilibrium in 4 hours and the type of isotherm was well fitted by a linear type isotherm. In batch experiment, kinetic rate constants for the reductive dechlorination of CT increased with increasing the concentration of the reductants (Fe(II) and HS-). However, Fe(II) was a much more effective reductant, producing higher k values than those of HS-. The contact time of one day for the soil with HS- and that of four hours with Fe(II) showed the highest reaction rates. Additionally, the rate constants increased with the increase of pH in soil suspension with Fe(II) (5.2~8) and HS- (8.3~10.3), respectively. In column experiment, the soil column with Fe(II) showed larger bed volumes (13.76) to reach a column breakthrough than that with HS- indicating the treatment of Fe(II) is more effective for the reductive dechlorination of CT. To enhance reductive capacity of soil column under an acidic condition, CaO addition to the column treated with Fe(II) showed better results for the reductive dechlorination of CT than that of HS-. Fe(II) showed better CT dechlorination than HS- in batch and column reactors therefore, it can be used as an effective reducing agent for the treatment of soil contaminated with chlorinated organic compounds.
B11B-0355
Development Of A New Redox-Active Porous Material For Groundwater Remediation
Laboratory experiments have shown that reducing iron in smectites promotes the degradation of various redox sensitive organics, including nitroaromatics and chlorinated compounds. Fe-bearing smectites have however never been used in the design of permeable reactive barriers (PRBs) for groundwater remediation. One basic requirement when designing PRBs is to keep their permeability equal to or higher than that of the surrounding aquifer materials to avoid affecting groundwater flow. Smectite clays are very low permeability materials and, when physically mixed with permeable materials, such as sand, clay particles can migrate and clog up pores, resulting in a progressive loss in permeability. In this study, we are developing a novel Fe-bearing clay-material suitable for permeable water treatment systems, including PRBs. Fe-smectite particles are tightly attached to the surface of sand grains using polyvinyl alcohol (PVA). To identify optimum procedures, we are studying the relationships between the size and texture of the sand grains, the clay/PVA and clay/sand ratio, the quality and extent of clay coverage, the stability of the clay-coated sand to changes in pH and redox conditions, and its hydraulic properties before and after iron reduction. The best clay coatings have been obtained using the most angular sands with rough surfaces and medium grain sizes (0.3-0.6mm). An optimum coating of 61.5 mg clay/g sand was obtained using the nontronite Nau- 2. The clay-coated sand is stable when pH is below 7 (no detachment of the clay particles). For pH higher than 7, a maximum of 14% of the clay-coating is detaching when the sample is not disturbed, and 28% if shaken. XRD analyses of the clay-coated sand also show that the coated smectite retains its swelling properties (d-spacing at 17.1Å after ethylene glycol treatment). The clay-coated sand is also stable to changes in redox conditions, with less than 15% detachment after 4h of treatment with sodium dithionite at 25°C. The coated clay is reducible, with a maximum reduction level of 83% (Fe(II)/Fe total). The hydraulic properties of the clay-coated sand are suitable for use in permeable systems. The effect of the reduced clay-coated sand on the degradation of redox-sensitive organics will be tested using nitrobenzene.
B11B-0356
Integration of Genome-Scale Metabolic Nodels of Iron-Reducing Bacteria With Subsurface Flow and Geochemical Reactive Transport Models
Several field and laboratory experiments have demonstrated that the growth and activity of iron-reducing bacteria can be stimulated in many subsurface environments by amendment of groundwater with a soluble electron donor. Under strong iron-reducing conditions, these organisms mediate reactions that can impact a wide range of subsurface contaminants including chlorinated hydrocarbons, metals, and radionuclides. Therefore there is strong interest in in-situ bioremediation as a potential technology for cleanup of contaminated aquifers. To evaluate and design bioremediation systems, as well as to evaluate the viability of monitored natural attenuation as an alternative, quantitative models of biogeochemically reactive transport are needed. To date, most such models represent microbial activity in terms of kinetic rate (e.g., Monod- type) formulations. Such models do not account for fundamental changes in microbial functionality (such as utilization of alternative respiratory pathways) that occur as the result of spatial and temporal variations in the geochemical environment experienced by microorganisms. Constraint-based genome-scale in silico models of microbial metabolism present an alternative to simplified rate formulations that provide flexibility to account for changes in microbial function in response to local geochemical conditions. We have developed and applied a methodology for coupling a constraint-based in silico model of Geobacter sulfurreducens with a conventional model of groundwater flow, transport, and geochemical reaction. Two uses of the in silico model are tested: 1) incorporation of modified microbial growth yield coefficients based on the in silico model, and 2) variation of reaction rates in a reactive transport model based on in silico modeling of a range of local geochemical conditions. Preliminary results from this integrated model will be presented.
B11B-0357
Spatial Modeling of Iron Transformations Within Artificial Soil Aggregates
Structured soils exhibit significant variations in transport characteristics at the aggregate scale. Preferential flow occurs through macropores while predominantly diffusive exchange takes place in intra-aggregate micropores. Such environments characterized by mass transfer limitations are conducive to the formation of small-scale chemical gradients and promote strong spatial variation in processes controlling the fate of redox-sensitive elements such as Fe. In this study, we present a reactive transport model used to spatially resolve iron bioreductive processes occurring within a spherical aggregate at the interface between advective and diffusive domains. The model is derived from current conceptual models of iron(hydr)oxide (HFO) transformations and constrained by literature and experimental data. Data were obtained from flow-through experiments on artificial soil aggregates inoculated with Shewanella putrefaciens strain CN32, and include the temporal evolution of the bulk solution composition, as well as spatial information on the final solid phase distribution within aggregates. With all iron initially in the form of ferrihydrite, spatially heterogeneous formation of goethite/lepidocrocite, magnetite and siderite was observed during the course of the experiments. These transformations were reproduced by the model, which ascribes a central role to divalent iron as a driver of HFO transformations and master variable in the rate laws of the considered reaction network. The predicted dissolved iron breakthrough curves also match the experimental ones closely. Thus, the computed chemical concentration fields help identify factors governing the observed trends in the solid phase distribution patterns inside the aggregate. Building on a mechanistic description of transformation reactions, fluid flow and solute transport, the model was able to describe the observations and hence illustrates the importance of small-scale gradients and dynamics of bioreductive processes for assessing bulk iron cycling. As HFOs are ubiquitous in soils, such process-level understanding of aggregate-scale iron dynamics has broad implications for the prediction of the subsurface fate of nutrients and contaminants that interact strongly with HFO surfaces.
B11B-0358
Effect of Electron Capacitance on Geobacter Growth and Metal Reduction in Subsurface
Background: Microbial reduction has been established as a promising bioremediation strategy to reduce and immobilize hexavalent uranium [U (VI)] as precipitated U (IV). This method however relies on the availability of Fe (III) oxides prevalent in the subsurface and their concurrent Fe (III) reduction. Unfortunately, the biogeochemical study on the physiology of simultaneous reduction of multiple metals is still poorly understood. A detailed model is therefore required to clarify the pathways leading to U (VI) and Fe (III) reduction in Geobacter species. Results: We propose a novel kinetic model that physically distinguishes Geobacter species into neutral and electron-charged states based on the recent experimental evidence of temporary electron sinks in Geobacter. This physical separation allows prediction of the environmentally relevant physiological status of Geobacter species in subsurface. The simulation clearly indicates that the decrease in neutral suspended cells and the increase in electron-charged cells are due to the Fe (III) limitation in the subsurface. Furthermore, this model illustrates a capacitance-dependent electron load-unload cycle that can be used to identify mechanisms responsible for the efficient U (VI) reduction and the correlation between U (VI) and Fe (III)-reducing activity. It shows that the electron load at cytochromes is not only responsible for providing maintenance and motility energy for Geobacter growth, but also responsible for facilitating the U (VI) removal. Global sensitivity analysis was used to validate the beneficial effects of electron capacitance and determine the level of importance and interactions of physicochemical and biogeochemical processes. In addition to identify the key biogeochemical processes responsible for U(VI) removal, the sensitivity analysis pinpoints several physicochemical processes that have significant impact on the U(VI) removal, such as the release of attached Geobacter from Fe (III) surface sites and Fe (II) precipitation and complexation. Furthermore, the sensitivity analysis has been shown to provide a guide for future data-collection efforts for this two-state modeling of in situ bioremediation process. Application: As compared with current modeling approaches in which biomass is often assumed to maintain the same metabolic and physiological state over all conditions, the structured two-state model accounts for important aspects of the dynamic electron capacitance of subsurface Geobacter, thereby facilitating further applications in the optimal bioremediation design strategy. The immediate application will address issues associated with the coupling of this structured two-state kinetic model and a genome-scale metabolism model with geochemical and hydrological models in an effort to describe the bioremediation process under spatially and temporally different conditions.
B11B-0359
Acid-Tolerant Sulfate-Reducing Bacteria Play a Major Role in Iron Cycling in Acidic Iron Rich Sediments
Climate change drives drying and acidification of many rivers and lakes. Abundant sedimentary iron in these systems oxidizes chemically and biologically to form iron-ox(yhydrox)ide crusts and "hardpans". Given generally high sulfate concentrations, the mobilization and cycling of iron in these environments can be strongly influenced by bacterial sulfate reduction. Sulfate-reducing bacteria (SRB) induce reductive dissolution of oxidized iron phases by producing the reductant bisulfide as a metabolic product. These environmentally ubiquitous microbes also recycle much of the fixed carbon in sediment-hosted microbial mat communities. With prevalent drying, the buffering capacity for protons liberated from iron oxidation is exceeded, and the activity of sulfate-reducers is restricted to those species capable of tolerating low pH (and generally highly saline, i.e. sulfate-rich) conditions. These species will sustain the recycling of iron from more crystalline phases to more bioavailable species, as well as act as the only source of bisulfide for photosynthesizing microbial communities. The phylogeny and physiology of acid-tolerant SRB is therefore important to Fe, S and C cycling in iron-rich sedimentary environments, particularly those on a geochemical trajectory towards acidification. Previous studies have shown that these SRB species tend to be highly novel. We studied two distinct environments along a geochemical continuum towards acidification. In both settings, iron redox transformations exert a major, if not controlling, influence on reduction potential. An acidified, iron- rich tidal marsh receiving acid-mine drainage (San Francisco Bay, CA, USA) contained abundant textural evidence for reductive dissolution of Fe(III) in sediments with pH values varying from 2.4 – 3.8. From these sediments, full-length novel dsrAB gene sequences from acid-tolerant SRB were recovered, and sulfur isotope profiles reflected biological fractionation of sulfur under even the most acidic conditions. The dsrAB genes are related to other novel SRB lineages derived from acidic environments in previous reports, suggesting that these species have adapted to the acidity rather than colonized more circumneutral microenvironments. In an acidic hypersaline lake system in NW Victoria (Australia), previous studies suggested that pore water bisulfide derived from anoxic groundwater transported from distal locations. However, isolated potholes of oxic Fe(III)-rich springwater exhibited nearly a two-fold increase in conductivity and pH increase from 4.5 to 8.0 over time periods on the order of days; and biogeochemical and mineralogical observations were consistent with the presence of active acid- and halo-tolerant SRB. Furthermore, stratified active microbial mat communities, with zones of black FeS formation localized several millimeters below the sediment-air interface, were identified in cross-section from lakeshore sediments near groundwater discharge springs. Culture-independent and culture-based work to characterize the SRB population is ongoing at this site. We infer, from previous sulfur isotope tracer experiments at the lake, that overall sulfate reduction rates may be slow, but are nonetheless proceeding and contributing to the recycling of oxidized iron to a significant degree given the abundance of sulfate evidenced by widespread gypsum precipitation. We conclude from the two study-sites described above that acid-tolerant SRB species play an important role in the linked S, Fe and C cycles in acidifying, iron-rich environments, and their phylogenetic and physiological diversity should be further investigated.
B11B-0360
Fractionation of Fe Isotopes During the Reductive Dissolution of Fe Oxide Minerals by Dissolved Sulphide
Previous discussions regarding Fe isotope systematics during Fe oxide reduction in sediments have often focused on the potential for identifying microbial processes such as dissimilatory Fe reduction. Indeed, during the reduction of Fe(III) oxides to dissolved Fe(II), mediated by Schewanella Alga, a fractionation of - 1.3‰ has been reported. Such fractionations, when combined with experimentally-derived fractionations observed during diagenetic processes such as FeS formation, have been used to investigate the antiquity of microbial Fe reduction. However, previous studies investigating microbial Fe reduction in ancient sediments have been limited by a lack of knowledge concerning isotope fractionations during key abiotic reactions. The reductive dissolution of Fe oxides mediated by dissolved sulphide is a major process during anoxic sediment diagenesis. Due to the change in redox state from Fe(III) to Fe(II), a significant Fe isotope fractionation might be expected. Therefore, the aim of this study is to ascertain whether Fe isotopes fractionate during this process, and whether different Fe oxides fractionate to different extents. To investigate these possibilities, a range of controlled sulphidation experiments were performed using a variety of synthetic Fe oxide minerals (ferrihydrite, lepidocrocite, hematite). In this presentation we will discuss the results of these experiments, which indicate a significant Fe isotope fractionation which should be considered when assessing Fe isotope systematics in modern and ancient sediments.
B11B-0361
Onset of Turonian oceanic red beds in Central Italy: geochemical constrains for paleoceanographic change
Oceanic red beds were widespread during the Late Cretaceous in association with major climate perturbations. Despite their relevance to the debate concerning the effects of climate change on ocean redox, the detailed geochemistry and the mechanism of formation of Cretaceous oceanic red beds remain poorly understood. Here we present a detailed examination of Fe partitioning, major and trace element data across the pre-OAE2 to the red beds segments of Contessa Quarry section (bottom of Scaglia Bianca to the top of Scaglia Rossa Formation), central Italy. Our geochemical data indicate that the limestones from the Contessa Quarry section show a distinct distribution pattern along the carbonate dilution line, indicating that they represent simple background sedimentation (average shale, AS) diluted by carbonates. Very low Al2O3 concentrations only 0.5 ~ 2.15 wt% indicate that terrigenous detrital input in the Contessa Quarry section is relatively low. Many of the common conservative lithogenic elements (i.e. Ti, K, Mg, Rb, Zr) trend close to AS when plotted against Al2O3, pointing to homogenous source area material. Furthermore, terrigenous detrital input and source area have almost no changes throughout the transition from the Scaglia Bianica to the Scaglia Rossa. Geochemical data shows that the red limestones deposited in a more oxic condition near the sediment-water interface, comparing to the white limestones in the Contessa Quarry section, which is strongly supported by: 1) extremely low TOC in the Scaglia Rossa only 0.01~0.07 wt%; 2) higher value of ferric oxide (hematite) and lower carbonate- associated iron in the red beds ; 3) lower concentrations of redox-sensitive elements such as U, V, Cr, Co, Ni; 4) strongly negative Ce anomaly. 5) lower biogenetic Ba thus lower productivity. Altogether, these data show that an increase in dissolved oxygen in bottom waters or the decrease of the productivity in the surface water (or the combination of the two) lead to the deposition of oceanic red beds during this time. The related paleoceanographic conditions resulted from these two factors will be presented in the presentation.
B11B-0362
Application Of Bacterial Iron Reduction For The Removal Of Iron Impurities From Industrial Silica Sand And Kaolin
Biogeochemical evidence exists to support the potential importance of crystalline or amorphous Fe minerals
as electron acceptor for Fe reducing bacteria in soils and subsurface sediments. This microbial metabolic
activity can be exploited as alternative method in different industrial applications. For instance, the removal of
ferric iron impurities from minerals for the glass and paper industries currently rely on physical and chemical
treatments having substantial economical and environmental disadvantages. The ability to remove iron by
other means, such as bacterial iron reduction, may reduce costs, allow lower grade material to be mined, and
improve the efficiency of mineral processing.
Kaolin clay and silica sand are used in a wide range of industrial applications, particularly in paper, ceramics
and glass manufacturing. Depending on the geological conditions of deposition, they are often associated
with iron (hydr)oxides that are either adsorbed to the mineral surfaces or admixed as separate iron bearing
minerals.
In this study, we have examined the Fe(III) removal efficiency from kaolin and silica sand by a series of iron-
reducing bacteria from the Shewanella species (S. alga BrY, S. oneidensis MR-1, S.
putrefaciens CN32 and S. putrefaciens ATCC 8071) in the presence of anthraquinone 2,6 disulfonate
(AQDS). We have also investigated the effectiveness of a natural organic matter, extracted with the silica
sand, as a substitute to AQDS for enhancing Fe(III) reduction kinetics. The microbial reduction of Fe(III) was
achieved using batch cultures under non-growth conditions. The rate and the extent of Fe(III) reduction was
monitored as a function of the initial Fe(III) content, Shewanella species and temperature. The bacterially-
treated minerals were analyzed by transmission electron microscopy (TEM) and X-ray diffraction (XRD) to
observe any textural and mineralogical transformation. The whiteness and ISO brightness of the kaolin was
also measured by spectrophotometry for quality testing.
All Shewanella species were able to couple the oxidation of lactate to the reduction of Fe(III) associated
with the kaolins and silica sands. However, there are differences among species with respect to the rate and
extent of iron leaching. S. putrefaciens ATCC 8071 is the most effective, with a 10% increase in kaolin
whiteness and 4% increase in ISO brightness in less than 5 days.
http://www.ceg.ncl.ac.uk/
B11B-0363
In situ characterization of green rust in the presence of arsenate and phosphate in simulated oxidized and reduced environments.
Nano- to micron-scale particles of mixed-valent iron hydroxide, specifically green rust (GR [FeII6- x(OH)y FeIIIx(OH)12-y]x+[Anionx- + H2O]x-), have been identified and studied as corrosion products of steel, and recently rediscovered in hydromorphic soils and sediments. Green rusts are intermediate phases produced by biotic and abiotic reductive dissolution of ferric oxyhydroxides, or by oxidation of dissolved ferrous iron. Adsorbed oxyanions can stabilize GR phases and inhibit the formation of thermodynamically favored iron phases such as magnetite or lepidocrocite in subsurface environments. This study used synchrotron XRD to characterize iron (hydr)oxide minerals precipitated from solution and subsequent aging products under different environmental conditions of pH and Eh. Here we show the in situ abiotic development of green rust and its stabilization by the addition of adsorbed oxyanions or alternatively, subsequent rapid transformation to magnetite or lepidocrocite in the absence of added anions. A closed batch reactor with an in-line capillary was used to expose the reaction products to continuous synchrotron radiation. Laue patterns were collected at time intervals of 3-5 minutes and used to detect the formation of crystalline iron (hydr)oxide minerals that precipitate as a function time and chemical perturbations to the system, i.e. changing the pH, redox potential, ratio of Fe2+ to OH- , and addition of an oxyanion, arsenate or phosphate. The reactions were monitored by observing the development of diagnostic green rust XRD d-spacing peak at 10.9 Å (300), the 3.29 Å (210) d- spacing for lepidocrocite, and the 2.53 Å (100) d-spacing for magnetite, with continuous in-line measurement of pH and ORP. We found that green rust was stabilized by the adsorption of arsenate and phosphate. In the presence of arsenate or phosphate at pH =7, green rust transformed to lepidocrocite after several hours when anoxic controls were removed. When pH and Eh were constant, GR did not transform to magnetite or lepidocrocite. However, without arsenate or phosphate, the transformation occurred in less than 1 hour. At pH > 7.5, green rust transformed to magnetite within 1 hour. In the presence of phosphate or arsenate, the transformation was retarded and took up to 8 hours. The rates of transformation and meta- stability of iron (hydr)oxide phases in natural redoximorphic sediments play important roles in the cycling of contaminants and nutrients.
B11B-0364
Oxidation of Structural Fe(II) in Biotite by Lithotrophic Fe(II)-oxidizing microorganisms
The potential for microbial involvement in the oxidation of Fe(II)-bearing phyllosilicates is an understudied aspect of soil/sediment Fe biogeochemistry. An important property of structural Fe in Fe-bearing smectites is their ability to undergo multiple redox cycles without being mobilized. An obvious choice of mineral substrate for enumeration/isolation of Fe(II)-oxidizing microorganisms would be reduced smectite. But reduced smectite is readily oxidized by air. That is why biotite was chosen as a substrate for this study. In contrast to smectite, biotite is more stable in the presence of air, but incapable of redox cycling. Once Fe(II) is oxidized, biotite is weathered to expendable 2:1 phyllosilicates or kaolinite. First, we evaluated the ability of a neutral-pH lithoautotrophic nitrate-reducing enrichment culture (MPI culture), recovered by Straub et al (Appl. Environ. Microbiol., 1996, 62:1458-1460) from a freshwater ditch, to oxidize two different specimens of biotite. The culture was capable of multiple transfers in anaerobic nitrate-containing biotite suspensions. The growth of MPI culture resulted in decrease of 0.5 N HCl-extractable Fe(II) content and simultaneous nitrate reduction. Cell yields were comparable to those observed for other neutral-pH lithoautotrophic Fe(II)-oxidizing bacteria. High resolution TEM examination revealed structural and chemical changes at the edges of oxidized biotite and formation of reddish amorphous precipitates dominated by Si and Fe. To further evaluate efficiency of biotite for recovery of oxygen- and nitrate-dependent Fe(II) oxidizing cultures microbial enumeration study was performed using subsoil from a site near Madison, WI. The soil is rich in Fe-bearing smectite and shows evidence of redoximorphic features. The enumeration of Fe(II) oxidizing organisms from this sediment showed 10-fold higher efficiency of biotite over soluble Fe(II) for recovery of Fe(II)-oxidizers. Isolation and identification of both aerobic and nitrate-utilizing Fe(II)-oxidizing cultures is under way. This study demonstrates that biotite can be effectively used to recover and study microorganisms involved in the oxidative side of iron redox cycle in phyllosilicates. Our findings also indicate that microbial redox metabolism has the potential to vastly accelerate the oxidative weathering of otherwise relatively stable Fe(II)-bearing phyllosilicates.
B11B-0365
The role of FeS(aq) molecular clusters in microbial redox cycling and iron mineralization.
Iron sulfide molecular clusters, FeS(aq), are a group of polynuclear Fe-S complexes varying in size between a few and a few hundred molecules that occur in many environments and are critical parts of cycling between soluble iron and iron sulfide minerals. These clusters react uniquely with voltammetric Au-amalgam electrodes, and the signal for these molecules has now been observed in many terrestrial and marine aquatic settings. FeS(aq) clusters form when aqueous sulfide and iron(II) interact, but the source of those ions can come from abiotic or microbial sulfate and iron reduction or from the abiotic non-oxidative dissolution of iron sulfide minerals. Formation of iron sulfide minerals, principally mackinawite as the first solid nanocrystalline phase in many settings, is necessarily preceeded by formation and evolution of these molecular clusters as mineralization proceeds, and the clusters have been suggested to additionally be part of the pyritization process (Rickard and Luther, 1997; Luther and Rickard, 2005). In several systems, we have also observed FeS(aq) clusters to be the link between Fe-S mineral dissolution and oxidation of iron and sulfide, with important implications for changes to the overall oxidation pathway. Microorganisms can clearly be involved in the formation of FeS(aq) through iron and sulfate reduction, but it is not clear to date if organisms can utilize these clusters either as metabolic components or as anabolic 'building blocks' for enzyme production. Cycling of iron in the Fe-S system linked to FeS(aq) would clearly be a critical part of understanding iron isotope dynamics preserved in iron sulfide minerals. We will review ongoing work towards understanding the role of FeS(aq) in iron cycling and isotope fractionation as well as the measurement and characterization of this key class of iron complexes using environmental voltammetry.
B11B-0366
Aqueous Iron-Sulfide Clusters in Variably Saturated Soil Systems: Implications for Iron Cycling and Fluid Flow
Iron and sulfur cycling is an important control on contaminant fate and transport, the availability of micronutrients and the physics of water flow. This study explores the effects of soil structure (i.e. layers, lenses, macropores, or fractures) on linked biogeochemical and hydrological processes involving Fe and S cycling in the vadose zone using packed soil columns. Three laboratory soil columns were constructed: a homogenized medium-grained sand, homogenized organic-rich loam, and a sand-over-loam layered column. Both upward and downward infiltration of water was evaluated during experiments to simulate rising water table and rainfall events respectively. Water samples extracted by lysimeter were analyzed for reduced species (including total sulfide, Fe(II), and FeSaq) voltammetrically using a mercury drop electrode. In addition to other reduced species, aqueous FeS clusters (FeSaq) were observed in two of the columns, with the greatest concentrations of FeSaq occurring in close proximity to the soil interface in the layered column. To our knowledge, this is the first documentation of aqueous FeS clusters in partially saturated sediments. The aqueous nature of FeSaq allows it to be transported instead of precipitating and suggests that current conceptual models of iron-sulfur cycling may need to be adapted to account for an aqueous phase. The presence of iron-rich soil aggregates near the soil interface may indicate that FeS clusters played a critical role in the formation of soil aggregates that subsequently caused up to an order of magnitude decrease in hydraulic conductivity.
B11B-0367
Fluxes of Oxidized and Reduced Iron Through a Northern Hardwood Forest Spodosol
Iron (Fe) is abundant among trace elements in forest ecosystems and is important in the development and function of soils. In this study we use measurements from the Hubbard Brook Experimental Forest in the White Mountains of New Hampshire, USA. To better understand the biogeochemical behavior of Fe and its role in the development of Spodosol soils (podsolization), we have constructed a series of mass balance equations to determine fluxes of reduced (ferrous, Fe(II)) and oxidized (ferric, Fe(III)) iron draining through the soil profile. Additionally, we measured Fe in throughfall and leaf litterfall as well as stream water to better assess inputs to and output from the soil. Soil solution fluxes of Fe were highest from the organic (Oa) horizon and decreased with depth in the mineral (Bh and Bs) horizons, consistent with podsolization theories predicting immobilization of Fe in mineral soil. The fluxes of Fe(II), Fe(III), and dissolved organic carbon (DOC) show similar patterns to each other, also consistent with hypotheses of organically-complexed Fe translocated to the spodic horizon, where co-precipitation of Fe and C occur. The portion of total Fe as Fe(II) ranges approximately 10-60% in soil solutions, seemingly high for soils that are typically considered well- drained, oxidizing environments. Analysis of total Fe and Fe(II) in leaf litter extracts from the three most abundant hardwood species show leachate to be a major source of reduced Fe to solutions draining the forest floor as approximately 50% of this Fe is Fe(II). The dissolved Fe draining the forest floor is either complexed by organic compounds during litter decomposition or is in leached directly from leaves in a complexed form. Our results indicate these organic complexes stabilize Fe(II) in solution when oxidizing conditions should promote considerably higher Fe(III)-to-Fe(II) ratios. Qualitative measurements of dissolved oxygen concentration in the soil solution range from nearly depleted to completely saturated, while the Fe concentrations remain more constant. This study indicates primarily a vegetation-derived source of dissolved Fe(II) to soils from the forest floor rather than direct reduction of soil Fe. However, the translocation of Fe(II) could have implications for redox chemistry in deeper mineral soils.
B11B-0368
Effects of Carbon Addition on Iron and Phosphorus in a Highly Weathered Tropical Soil
In the highly weathered iron (Fe)-rich soils of wet tropical forests, Fe may play a key role in controlling ecosystem processes because of its interactions with carbon (C) and phosphorus (P). The high NPP typical of tropical forests contributes significantly to the global C cycle. In Fe-rich tropical soils, NPP is thought to be limited by P. The periodic reducing conditions that occur in upland tropical soils may be associated with pulses of increased P availability because of the release of Fe-bound P during iron reduction. While little is known about the factors controlling Fe reduction in soils, it is likely that C availability plays a role. Typically, only simple C sources like acetate or glucose have been used to examine this limitation. However, the source of much of the C in nature is the complex mixture of organic compounds leached from leaves and litter. To investigate the linkages between Fe, C, and P, we compared the effects adding either acetate (200 mg C/L) or leaf leachate in low (50-100 mg C/L) or high (150-200 mg C/L) concentrations to incubated soils from a tropical rain forest in Puerto Rico under ambient atmospheric conditions. We measured pools of iron and phosphorus as well as pH at four time points over a month. Both Fe(II) and pH exhibited significant treatment effects, but not until the last sampling date. At this time, the Fe(II) concentration could explain 49% of the variability in soil pH. The pH was significantly higher in the acetate treatments than both the leaf leachate treatments. While Fe(II) concentration was significantly higher in the acetate treatment than the control and low leaf leachate treatment, there was no difference compared to the high leaf leachate treatment After one month microbial biomass P had increased significantly while the NaOH extractable organic P had decreased significantly. These changes suggest the rapid microbial uptake of P liberated from Fe. In conclusion, microbes appear to utilize more complex C in leaf leachate at a similar rate as acetate to promote Fe reduction. The simultaneous immobilization of P by microbes during the incubation suggests that periodic reducing conditions in the field may be associated with enhanced microbial activity and carbon cycling in these highly productive ecosystems.
B11B-0369
Long-Term Exposure of Tropical Soils to Pressure Treated Lumber, Barro Colorado Island, Panama: Impacts on Soil Metal Mobility and Microbial Community Structure
Pressure treated lumber (CCA) has been used in a variety of structures for over seven decades, but recent concerns have been raised about leaching of metals such as chromium (Cr) and arsenic (As) into proximal soils and water supplies. Pressure treated lumber abundance and its continued use necessitate a thorough understanding of metal release and sequestration in the subsurface. To date, no long-term, in situ study on the migration of CCA compounds from lumber has been performed. Barro Colorado Island, Panama is the site of several previous CCA studies and provides an opportunity to investigate the long-term (>70 years) effects of pressure treated lumber in oxisols, where high rainfall and warm temperatures may represent an end-member condition for the leaching and mobility of these metals. Soil samples from CCA and control sites were measured for Cr, As, Cu, Zn, and Fe abundances, microbial biomass and community structure via phospholipid fatty acid analysis, along with basic soil properties. CCA lumber samples were also characterized for their metal abundance. Lumber treated with zinc meta-arsenite displayed advanced decay with elevated As, Cu, and Zn concentrations observed in the adjacent soil. Increased soil organic matter and microbial biomass correlate to decreases in Fe and Fe-associated metals compared to the control. High As concentrations persist to <1 m of the source. Lumber treated with potassium dichromate contained high chromium concentrations and displayed little decay, however, soil concentrations of Cr, Fe, and Cu were generally less than control soils. Over these same intervals, soil organic matter and microbial biomass increased, particularly the fraction of metal reducing bacteria (MRB). We hypothesize that organic carbon loading from lumber stimulates MRB, leading to mobilization of Fe and Fe-associated metals from these oxide-rich soils. Principal component analysis of PLFA data confirms a distinction between controls and samples with elevated metal abundance at each site. This study provides fundamental insight into the long-term persistence of CCA compounds in Fe-rich soils and could serve in practical applications related to CCA contamination.
B11B-0370
Evaluation of clay-TiO2 nanocomposite efficiency on the photocatalytic elimination of a model hydrophobic air pollutant.
Clay-supported TiO2 photocatalysts can potentially improve the performance of air treatment technologies due to enhanced adsorption and reactivity of volatile organic compounds (VOCs). In this study, a bench-top photocatalytic flow reactor was used to evaluate hectorite-TiO2 and kaolinite-TiO2, two novel composite materials synthesized in our laboratory. Toluene, a model hydrophobic VOC and a common indoor air pollutant, was introduced in the air stream at realistic concentrations and reacted under UVA (max = 365 nm) or UVC (max = 254 nm) irradiation. The UVC lamp presented secondary emission at 185 nm, leading to the formation of ozone and other short-lived reactive species. Performance of clay-TiO2 composites was compared with that of pure TiO2 (Degussa P25) and with UV irradiation in the absence of photocatalyst under identical conditions. Films of clay-TiO2 composites and of P25 were prepared by a dip- coating method on the surface of Raschig rings placed inside the flow reactor. An upstream toluene concentration of 150 ppbv was generated by diluting a constant vapor flow with dry air, or with humid air at 10, 33, and 66 % relative humidity (RH). Toluene concentrations downstream were determined by collecting Tenax-TA ® sorbent tubes and subsequent thermal desorption/gas chromatography/mass spectrometry (TD/GC/MS) analysis. The fraction of toluene removed, %R, and the reaction rate, Tr, were calculated for each experimental condition from the concentration changes registered with and without UV irradiation. Use of UVC light (UV/TiO2/O3) led to overall higher reactivity, which can be partially attributed to contribution of gas phase reactions by O3 and other short-lived species. When the reaction rate was normalized by the light irradiance, Tr/Iλ, the UV/TiO2 reaction under UVA irradiation was more efficient for samples with a higher content of TiO2 (P25 and Hecto-TiO2), but not for Kao-TiO2. Considering the effect of relative humidity, in all cases reaction rates peaked at 10% RH, with values 10-50% higher than those measured under dry air. However, a net inhibition was observed as RH increased, indicating that water molecules competed effectively with toluene for reactive surface sites and limited the overall photocatalytic conversion. As compared with P25, inhibition by co-adsorbed water was less significant for Kao-TiO2 samples, but was more dramatic in the case of Hecto-TiO2 due to the high water uptake capacity of the hectorite clay.