Determination of the solid-phase arsenic speciation in sediments hosting high-arsenic groundwaters, utilized for drinking and irrigation in Bengal, SE Asia and elsewhere is important in order to understand the biogeochemistry of arsenic. Despite this, there is a relative paucity of speciation data for solid-phase arsenic in such systems, due to preservation difficulties, low arsenic concentrations in the sediments, multiple coordination environments and sample heterogeneity. In this study, X-ray absorption near edge structure spectroscopy was used in conjunction with linear least-squares fitting of model compounds to determine the oxidation state of arsenic in sediments from West Bengal and Cambodia. Whatever the collection and storage method used, substantial oxidation of arsenic was commonly observed over periods of weeks to several months. Sands were particularly susceptible to changes in arsenic oxidation state during storage. Analysis within two or three weeks of collection is therefore recommended, whilst on-site storage under a nitrogen atmosphere immediately after collection is particularly recommended for the preservation of sandy samples. Both muds and sands from West Bengal and Cambodia were dominated by arsenite (As(III)) with <35±10% arsenate (As(V)). Complete oxidation to arsenate was never observed suggesting that a significant proportion of the sedimentary arsenic is inaccessible within crystalline phases. Centrifuging under anaerobic conditions enabled more detailed information about a variety of arsenic coordination environments to be determined.

Possible relationships between groundwater arsenic concentration and alluvial sediment characteristics in a ∼19 km2 area in West Bengal have been investigated using a combination of hydrogeochemical, lithogeochemical and geophysical techniques. Arsenic hotspots, typically associated with elevated groundwater Fe and Mn, were found to be correlated to some extent with old river channels (abandoned meanders, oxbow lakes), where sandy aquifers included intercalated fine-grained overbank deposits, rich in As, Fe, Mn and Corg. Otherwise no demonstrably significant overall differences in any of lithology, grain-size distribution, mineral composition or Fe, Mn and organic C content of the sediments were found between two representative sites with contrastingly low (<50 μg 1—1) and high (>200 μg 1—1) As groundwater contents.

Our results are consistent with microbially mediated redox reactions controlled by the presence of natural organic matter within the aquifer and the occurrence of As-bearing redox traps, primarily formed by Fe and Mn oxides/hydroxides, being the most important factors which control the release of As into shallow groundwaters at the study site.

Arsenic mobilization and Fe(III) reduction in acetate-amended sediments collected from a range of depths from an aquifer with elevated groundwater arsenic concentrations in West Bengal were monitored over a 1 month period. Significant arsenic release was noted in sediment collected from 24 m and 45 m depth, with some Fe(III) reduction also observed in the 24 m sample. The structure of the microbial communities present in the sediments prior to incubation showed marked differences down the sediment column. Profiling of the microbial community in the 24 m and 45 m samples revealed a relatively complex make-up, with Acinetobacter species comprising the bulk of the 24 m sedimentary bacterial population, but no previously characterized As(V)-reducers were detected in either sample.

Previous studies from our laboratory have suggested a role for indigenous metal-reducing bacteria in the reduction of sediment-bound As(V), and have also shown that a stable enrichment culture of Fe(III)-reducing bacteria was able to mobilize arsenic (as As(III)) from sediments collected from West Bengal (Islam et al., 2004). To identify the Fe(III)-reducing bacteria that may play a role in the reduction of As(V) and mobilization of As(III), we made a detailed molecular analysis of this enrichment culture. It was dominated by a close relative of Geothrix fermentans, but the type strain of this organism was unable to conserve energy for growth via the dissimilatory reduction of As(V), or reduce As(V) present in a defined medium containing fumarate as the electron acceptor. Furthermore, when the cells were grown using soluble Fe(III)-citrate as an electron acceptor in the presence of As(V), bacterial Fe(III) reduction resulted in the precipitation of the Fe(II)-bearing mineral vivianite in 2 weeks. This was accompanied by the efficient removal of As from solution. These results demonstrate that Geothrix fermentans, in common with other key Fe(III)-reducing bacteria such as Geobacter sulfurreducens, does not reduce As(V) enzymatically, but can capture arsenic in Fe(II) minerals formed during respiration using Fe(III) as an electron acceptor. Thus, the reduction of arsenic-bearing Fe(III) oxide minerals is not sufficient to mobilize arsenic, but may result in the formation of Fe(II) biominerals that could potentially act as sinks for arsenic in sediments. Additional mechanisms, including dissimilatory As(V) reduction by other specialist anaerobic bacteria, are implicated in the mobilization of arsenic from sediments.

Arsenic and fluoride in groundwater from Quaternary loess deposits in Argentina pose major health concerns. Common sources for arsenic and fluoride have been suggested but the processes of mobilization are disputed, and distributions in groundwater are largely unresolved at a sample density >1/50 km2. At Los Pereyras in Tucumán Province, northern Argentina, we have evaluated distributions and hydrochemical associations of arsenic and fluoride with a sample density of 0.75 per km2 over an area of 75 km2, to a depth of 230 m. Groundwater in the loess is oxic and alkaline. Fluoride is restricted to the upper 20 m of the Quaternary loess, where it reaches 8.3 mg/1. Arsenic has a vertical layering consistent with that of fluoride, ranging from 20 to 760 μg/1 in the upper 20 m and 58—163 μg/l below this. There are two sources of arsenic, one unrelated to the fluoride source. Positive correlations between arsenic and fluoride with pH, but not with alkalinity, support desorption from iron oxyhydroxides as the likely mechanism of release to groundwater for arsenic and fluoride, rather than the weathering of silicate minerals. Stratigraphic and/or palaeohydrological controls may explain the observed depth distributions within the loess aquifer.

Degradation of contaminant hydrocarbons in groundwater by microbially mediated oxidation, linked to the reduction of electron acceptors, is fundamental to the strategy of ‘monitored natural attenuation’ (MNA) for oxidizable hydrocarbons, which is increasingly being adopted at polluted aquifer sites throughout Europe and North America. Commonly, oxygen is depleted and following the reduction of nitrate, solid-phase Fe oxides become the dominant electron acceptors. Arsenic, associated with Fe and Mn oxides in soils and sediments, may therefore be mobilized to groundwater and pose an additional threat to environmental receptors. In a pilot study of three aquifers in England, we have examined the extent to which arsenic is released to groundwater under Fe(III)-reducing conditions imposed by contaminant hydrocarbons. Results show that arsenic is locally mobilized in the Chalk to <10 μg/1, in Quaternary gravels to 70 μg/1 and in the Triassic sandstones to 160 μg/1. At the Chalk and Quaternary gravels sites arsenic mobilization is demonstrably linked to reduction of Fe- and Mn-oxides. This is not so at the Triassic sandstone site, where release of arsenic is related to elevated bicarbonate alkalinity. Redox-driven arsenic mobilization at other Triassic sandstone locations is possible. Further work is required on the solid-phase sources of arsenic in the aquifers, and to relate the hydrochemical observations to groundwater hydraulic conditions.

Arsenic-rich scales are widely associated with geothermal fields and constitute a potential hazard to human health. Such arsenic has hitherto been reported to be almost exclusively hosted by sulphide or oxide phases or occurring as surface species. We report here, however, the occurrence of an arsenic-rich (1500 to 4000 mg kg—1 As) smectite from geothermal precipitates from a geothermal field in northwestern Japan and present evidence that the arsenic is predominantly hosted within this silicate mineral.

Consistently ∼80% of the total arsenic determined in these geothermal precipitates was found by selective chemical extractions to be associated with an operationally defined clay mineral fraction, with lesser proportions being associated with operationally defined amorphous silica, Fe oxide and sulphide fractions. Analysis by XRD, ATR IR and XRF showed the clay fraction to be dominated by Mg-rich trioctahedral smectite.

Arsenic K-edge XAS spectra of the smectite suggested the dominance of As(III)-O coordinated species with significant contributions from As(V)-O coordinated species. Both XPS and a magnesium chloride chemical extraction indicated minimal adsorption of arsenic on smectite surfaces suggesting that the arsenic was predominantly either dissolved within the smectite or occurred within mineral occlusions. No such occlusions greater than 1 μm in size were observed in the As-rich smectites.

The potential occurrence of arsenic-bearing clays should be considered when determining remediation strategies for geothermal environments or evaluating risks associated with the industrial usage of geothermal precipitates.