Adsorption,sequestration, and bioaccessibility of As(V) in soils

The influence of various soil physical and chemical properties (Fe and Mn oxides, pH, cation exchange capacity, total inorganic and organic carbon, and particle size) on As(V) adsorption, sequestration, and relative bioaccessibility (as a surrogate for oral bioavailability) was investigated in a wide range of well-characterized soils over a 6-month period. Arsenic(V) bioaccessibility was measured using a streamlined version of a physiologically based extracton test (PBET), designed to replicate the solubility-limiting conditions in a child's digestive tract. The soil's dithionite-citrate-bicarbonate (DCB) extractable Fe oxide content was the most important land only statistically significant) soil property controlling the initial degree of adsorption. Sequestration, as measured by the reduction in bioaccessibility over time, occurred to a significant extent in 17 of 36 (47.2%) soils over the first 3 months. In contrast, only 4 of 36 (11.1%) soils exhibited a significant reduction in bioaccessibility from 3 to 6 months. Soil pH was the most important (and only statistically significant) soil property affecting the decrease in bioaccessibility upon aging for 6 months. Soils with pH < 6 generally sequestered As(V) more strongly over time, whereas those with pH > 6 generally did not. The Fe oxide content and pH were the most important soil properties governing the steady-state bioaccessibility of As(V) in soil. Two multivariable linear regression models of steady-state As(V) bioaccessibility were developed using soil properties as independent variables. Generally, soils having higher Fe oxide content and lower soil pH exhibited lower bioaccessibility. These models were able to account for approximately 75-80% of the variability in steady-state bioaccessibility and independently predict bioaccessibility in five soils within a root-mean-square error (RMSE) of 8.2-10.9%. One of these models was also able to predict within an RMSE of 9.5% the in vivo bioavailability of As in nine contaminated soils previously used in swine dosing trials. These results indicate the bioaccessibility, and thus, potentially the bioavailability of otherwise soluble As(V) added to soils (i.e., the worst-case bioavailability scenario) is significantly reduced in some soils over time, particularly those with lower pH and higher Fe oxide content. These results also provide a means of estimating As(V) bioaccessibility and bioavailability on the basis of soil properties.     

Using nitrogen and carbon dioxide molecules to probe arsenic(V) bioaccessibility in soils

Highly specialized personnel and high cost are typically required for in vivo risk assessment of arsenic (As) exposure to humans in As-contaminated soils. Arsenic bioaccessibility in soils, as determined with the aid of in vitro tests, is quite variable, and its magnitude depends upon unidentified soil properties. Use of soil chemical properties is a common practice for construction of As(V) sorption and bioaccessibility models with relative success. We propose a novel As(V) bioaccessibility model, which was tested on 17 soils. The model includes only two parameters characterizing surface properties of soils that are readily determined from N2- and CO2-based specific surface areas (SSAs), and total organic carbon (OC) content. We found that N2 and CO2 molecules act as As(V) "surrogates", probing easily accessible and relatively difficult to access soil porosity, respectively. Three interrelated linear models were constructed using two terms (CO2/N2-based SSAs and OC) that were significant (p <0.001) in explaining 51 and 95% of the variability observed in As(V) sorption and bioaccessibility, respectively. The proposed models successfully predicted bioaccessible As concentrations for 4 out of the 5 soils that were not included in the bioaccessibility models, reaching RMSE values of < or =10%.     

In vivo validation of the unified BARGE method to assess the bioaccessibility of arsenic, antimony, cadmium, and lead in soils

The relative bioavailability of arsenic, antimony, cadmium, and lead for the ingestion pathway was measured in 16 soils contaminated by either smelting or mining activities using a juvenile swine model. The soils contained 18 to 25,000 mg kg(-1) As, 18 to 60,000 mg kg(-1) Sb, 20 to 184 mg kg(-1) Cd, and 1460 to 40,214 mg kg(-1) Pb. The bioavailability in the soils was measured in kidney, liver, bone, and urine relative to soluble salts of the four elements. The variety of soil types, the total concentrations of the elements, and the range of bioavailabilities found were considered to be suitable for calibrating the in vitro Unified BARGE bioaccessibility method. The bioaccessibility test has been developed by the BioAccessibility Research Group of Europe (BARGE) and is known as the Unified BARGE Method (UBM). The study looked at four end points from the in vivo measurements and two compartments in the in vitro study ("stomach" and "stomach and intestine"). Using benchmark criteria for assessing the "fitness for purpose" of the UBM bioaccessibility data to act as an analogue for bioavailability in risk assessment, the study shows that the UBM met criteria on repeatability (median relative standard deviation value <10%) and the regression statistics (slope 0.8 to 1.2 and r-square > 0.6) for As, Cd, and Pb. The data suggest a small bias in the UBM relative bioaccessibility of As and Pb compared to the relative bioavailability measurements of 3% and 5% respectively. Sb did not meet the criteria due to the small range of bioaccessibility values found in the samples.     

In vitro gastrointestinal bioavailability of arsenic in soils collected near CCA-treated utility poles

Because of the potentially high arsenic concentrations found in soils immediately adjacent to chromated copper arsenate (CCA)-treated wood structures and utility poles, CCA-contaminated soil ingestion may be a significant exposure route to arsenic for children. Therefore, a strong need exists to provide accurate data on oral relative bioavailability (RBA) of arsenic (in vivo or in vitro) in field-collected CCA-contaminated soils. The objectives of this study were (1) to assess arsenic bioaccessibility in contaminated soils collected near in-service CCA-treated utility poles, (2) to determine the influence of soil properties and arsenic fractionation on arsenic bioaccessibility, and (3) to estimate an average daily arsenic intake from incidental soil ingestion. Arsenic bioaccessibility (in vitro gastrointestinal (IVG) method) was determined on surface soil samples collected immediately adjacent to 12 CCA-treated utility poles after 18 months of service. Bioaccessible arsenic was also determined in 3 certified reference materials. Total arsenic concentrations in soils (<300 microm) varied from 37 +/- 2 to 251 +/- 12 mg/kg, irrespective of soil organic matter contentwith the major soil-bound arsenic species being As(V). Arsenic bioaccessibility ranged between 25.0 +/- 2.7 and 66.3 +/- 2.3% (mean value 40.7 +/- 14.9%). The mean value was in agreement with the in vivo arsenic RBA reported by Casteel et al. (2003) in soil near CCA-treated utility poles. Bioaccessible arsenic was positively correlated with total organic carbon content (r2 = 0.36, p < 0.05) and with water-soluble arsenic (2 = 0.51, p < 0.01), and was negatively correlated with clay content (r2 = 0.43, p < 0.05). Using conservative exposure parameters, the mean daily arsenic intake from incidental ingestion of contaminated soil near CCA-treated utility poles was 0.18 +/- 0.09 microg As kg(-1) d(-1). This arsenic intake appeared negligible compared to the daily intake of inorganic arsenic from water and food ingestion for children.     

Comparison of five in vitro digestion models to study the bioaccessibility of soil contaminants

Soil ingestion can be a major exposure route for humans to many immobile soil contaminants. Exposure to soil contaminants can be overestimated if oral bioavailability is not taken into account. Several in vitro digestion models simulating the human gastrointestinal tract have been developed to assess mobilization of contaminants from soil during digestion, i.e., bioaccessibility. Bioaccessibility is a crucial step in controlling the oral bioavailability for soil contaminants. To what extent in vitro determination of bioaccessibility is method dependent has, until now, not been studied. This paper describes a multi-laboratory comparison and evaluation of five in vitro digestion models. Their experimental design and the results of a round robin evaluation of three soils, each contaminated with arsenic, cadmium, and lead, are presented and discussed. A wide range of bioaccessibility values were found for the three soils: for As 6-95%, 1-19%, and 10-59%; for Cd 7-92%, 5-92%, and 6-99%; and for Pb 4-91%, 1-56%, and 3-90%. Bioaccessibility in many cases is less than 50%, indicating that a reduction of bioavailability can have implications for health risk assessment. Although the experimental designs of the different digestion systems are distinct, the main differences in test results of bioaccessibility can be explained on the basis of the applied gastric pH. High values are typically observed for a simple gastric method, which measures bioaccessibility in the gastric compartment at low pHs of 1.5. Other methods that also apply a low gastric pH, and include intestinal conditions, produce lower bioaccessibility values. The lowest bioaccessibility values are observed for a gastrointestinal method which employs a high gastric pH of 4.0.     

In vivo-in vitro and XANES spectroscopy assessments of lead bioavailability in contaminated periurban soils

Lead (Pb) bioaccessibility was assessed using 2 in vitro methods in 12 Pb-contaminated soils and compared to relative Pb bioavailability using an in vivo mouse model. In vitro Pb bioaccessibility, determined using the intestinal phase of the Solubility Bioaccessibility Research Consortium (SBRC) assay, strongly correlated with in vivo relative Pb bioavailability (R(2) = 0.88) following adjustment of Pb dissolution in the intestinal phase with the solubility of Pb acetate at pH 6.5 (i.e., relative Pb bioaccessibility). A strong correlation (R(2) = 0.78) was also observed for the relative bioaccessibility leaching procedure (RBALP), although the method overpredicted in vivo relative Pb bioavailability for soils where values were <40%. Statistical analysis of fit results from X-ray absorption near-edge structure (XANES) data for selected soils (n = 3) showed that Pb was strongly associated with Fe oxyhydroxide minerals or the soil organic fraction prior to in vitro analysis. XANES analysis of Pb speciation during the in vitro procedure demonstrated that Pb associated with Fe minerals and the organic fraction was predominantly solubilized in the gastric phase. However, during the intestinal phase of the in vitro procedure, Pb was strongly associated with formation of ferrihydrite which precipitated due to the pH (6.5) of the SBRC intestinal phase. Soils where Fe dissolution was limited had markedly higher concentrations of Pb in solution and hence exhibited greater relative bioavailability in the mouse model. This data suggests that coexistence of Fe in the intestinal phase plays an important role in reducing Pb bioaccessibility and relative bioavailability.     

Assessment of DDT relative bioavailability and bioaccessibility in historically contaminated soils using an in vivo mouse model and fed and unfed batch in vitro assays

In this study, DDTr (DDTr = DDT + DDD + DDE) relative bioavailability in historically contaminated soils (n = 7) was assessed using an in vivo mouse model. Soils or reference materials were administered to mice daily over a 7 day exposure period with bioavailability determined using DDTr accumulation in adipose, kidney, or liver tissues. Depending on the target tissue used for its calculation, some variability in DDTr relative bioavailability was observed; however, it did not exceed 25% (range 2-25%). When DDTr bioaccessibility was determined using organic physiologically based extraction test (Org-PBET), unified BARGE method (UBM), and fed organic estimation human simulation test (FOREhST) in vitro assays, bioaccessibility was less than 4% irrespective of the assay utilized and the concentration of DDTr in the contaminated soil. Pearson correlations demonstrate a poor relationship between DDTr relative bioavailability and DDTr bioaccessibility (0.47, 0.38, and 0.28, respectively), illustrating the limitations of the static in vitro methods for predicting the dynamic processes of the mammalian digestive system for this hydrophobic organic contaminant.     

In vitro model improves the prediction of soil arsenic bioavailability: worst-case scenario

There is a strong interest in developing an in vitro arsenic (As) model that satisfactorily estimates the variability in in vivo relative oral bioavailability (RBA) measurements. Several in vitro tests have been developed, but none is universally accepted due to their limited success in predicting soil As RBA. A suite of amorphous and crystalline solid As phases were chosen, utilizing a worst-case scenario (WCS) that simulated fasting children's gastric solution chemistry. The objectives of this study were to (i) determine the effects of residence time, pH, and solid-to-solution ratio on As bioaccessibility and speciation in the in vitro gastric test; (ii) provide the fundamental basis for an optimized in vitro model constrained by the WCS; and (iii) validate the optimized in vitro test with the in vivo RBA obtained with BALB/c mice. The gastric pH was the only significant (p < 0.05) factor influencing solid As bioaccessibility. Bioaccessible As retained the oxidation state after its release from the solid into the gastric solution. The optimized in vitro model adequately predicted RBA values for a suite of solid As phases typically encountered in soils, with the exception of aluminum-based solids. This study is an excellent starting point for developing an in vitro test applicable to different As-contaminated soils.     

Predicting arsenic relative bioavailability in contaminated soils using meta analysis and relative bioavailability-bioaccessibility regression models

A number of in vitro assays are available for the determination of arsenic (As) bioaccessibility and prediction of As relative bioavailability (RBA) to quantify exposure for site-specific risk assessment. These data are usually considered in isolation; however, meta analysis may provide predictive capabilities for source-specific As bioaccessibility and RBA. The objectives of this study were to predict As RBA using previously published in vivo/in vitro correlations and to assess the influence of As sources on As RBA independent of geographical location. Data representing 351 soils (classified based on As source) and 514 independent bioaccessibility values were retrieved from the literature for comparison. Arsenic RBA was predicted using published in vivo/in vitro regression models, and 90th and 95th percentiles were determined for each As source classification and in vitro methodology. Differences in predicted mean As RBA were observed among soils contaminated from different As sources and within source materials when various in vitro methodologies were utilized. However, when in vitro data were standardized by transforming SBRC intestinal, IVG, and PBET data to SBRC gastric phase values (through linear regression models), predicted As RBA values for As sources followed the order CCA posts ≥ herbicide/pesticide > mining/smelting > gossan soils with 95th percentiles for predicted As RBA of 78.0, 78.4, 67.0, and 23.7%, respectively.     

Bioaccessibility of arsenic(V) bound to ferrihydrite using a simulated gastrointestinal system

The risk posed from incidental ingestion to humans of arsenic-contaminated soil may depend on sorption of arsenate (As(V)) to oxide surfaces in soil. Arsenate sorbed to ferrihydrite, a model soil mineral, was used to simulate possible effects on ingestion of soil contaminated with As-(V) sorbed to Fe oxide surfaces. Arsenate sorbed to ferrihydrite was placed in a simulated gastrointestinal tract (in vitro) to ascertain the bioaccessibility of As(V) and changes in As(V) surface speciation caused by the gastrointestinal system. The speciation of As was determined using extended X-ray absorption fine structure (EXAFS) analysis and X-ray absorption near-edge spectroscopy (XANES). The As(V) adsorption maximum was found to be 93 mmol kg(-1). The bioaccessible As(V) ranged from 0 to 5%, and surface speciation was determined to be binuclear bidentate with no changes in speciation observed post in vitro. Arsenate concentration in the intestine was not constant and varied from 0.001 to 0.53 mM for the 177 mmol kg(-1) As(V) treated sample. These results suggest that the bioaccessibility of As(V) is related to the As(V) concentration, the As(V) adsorption maximum, and that multiple measurements of dissolved As(V) in the intestinal phase may be needed to calculate the bioaccessibility of As(V) adsorbed to ferrihydrite.     

Kinetics of arsenate adsorption-desorption in soils

Adsorption-desorption of arsenic is the primary factor that impacts the bioavailability and mobility of arsenic in soils. To examine the characteristics of arsenate [As(V)] adsorption-desorption, kinetic batch experiments were carried out on three soils having different properties, followed by arsenic release using successive dilutions. Adsorption of As(V) was highly nonlinear, with a Freundlich reaction order N much less than 1 for Olivier loam, Sharkey clay, and Windsor sand. Adsorption of arsenate by all soils was strongly kinetic, where the rate of As(V) retention was rapid initially and was followed by gradual or somewhat slow retention behavior with increasing reaction time. Freundlich distribution coefficients and Langmuir adsorption maxima exhibited continued increase with reaction time for all soils. Desorption of As(V) was hysteretic in nature and is an indication of lack of equilibrium retention and/or irreversible or slowly reversible processes. A sequential extraction procedure provided evidence that a significant amount of As(V) was irreversibly adsorbed on all soils. A multireaction model (MRM) with nonlinear equilibrium and kinetic sorption successfully described the adsorption kinetics of As(V) for Olivier loam and Windsor sand. The model was also capable of predicting As(V) desorption kinetics for both soils. However, for Sharkey clay, which exhibited strongest affinity for arsenic, an additional irreversible reaction phase was required to predict As(V) desorption or release with time.     

Bioaccessibility of uranium in soil samples from port hope, ontario, Canada

Adequate assessment of human health risk of uranium contamination at hazardous waste sites, which is an important step in determining the cleanup strategy, is based on bioavailability data. Bioavailability of uranium from contaminated soil has not been properly determined yet. Bioaccessibility is an in vitro conservative estimate of bioavailability and is thus frequently used for site-specific risk assessment. Bioaccessibility of uranium was measured in 33 soil samples from the Port Hope area in Ontario, Canada, by the physiologically based extraction test (PBET). Higher bioaccessibility values in the gastric plus intestinal phase, 48.4% ± 16.8%, than in the gastric phase, 20.8% ± 11.7%, are very probably the result of more efficient extraction of uranium from soil by intestinal fluid rich in carbonate ions. The observed variability of measured bioaccessibility values is discussed in light of the results of scanning electron microscope examination of the soil samples. Uranium bioaccessibility values in both gastric (acidic) and gastric plus intestinal (neutral) phases are higher in soil samples with smaller uranium-bearing particles and lower in samples where the uranium-bearing particles are larger. We postulate that the most important reason for variability of measured bioaccessibility values in Port Hope soil samples may be the difference in particle size of uranium-bearing particles.     

Coupling speciation and isotope dilution techniques to study arsenic mobilization in the environment

We have developed an approach to isolate mechanisms controlling mobility and speciation of As in soil-water systems. The approach uses a combination of isotopic exchange and chromatographic/mass spectrometric As speciation techniques. We used this approach to identify mechanisms responsible for changes in the concentration of soluble As in two contaminated soils (Eaglehawk and Tavistock) subjected to different redox conditions and microbial activity. A high proportion of the total As in both soils was present in a nonlabile form. Incubation of the soils under anaerobic conditions led to changes in the concentration of soluble As in each soil but did not change the As speciation or the proportion of total As in labile forms in the soils. Hence, a decrease in soluble As in the Eaglehawk soil was the result of an Eh-induced pH decrease, enhancing the solid-phase sorption of As(V). An increase in soluble As in the Tavistock soil was due to an Eh-induced pH increase, decreasing solid-phase sorption of As(V). Incubation of the soils under aerobic conditions with microbial activity stimulated by addition of glucose resulted in no change in the solution concentration or speciation of As in the Eaglehawk soil, but led to a large increase in the concentration of soluble As in the Tavistock soil. This increase was due to conversion of exchangeable forms of As(V) into less strongly sorbed As(III) species. Incubation under anaerobic conditions in the presence of glucose resulted in a large increase in the concentration of soluble As in both soils; however, different mechanisms were found to be responsible for the increase in each soil. In the Eaglehawk soil higher concentrations of As were again due to conversion of exchangeable forms of As(V) into less strongly sorbed As(III) species. In contrast in the Tavistock soil, the increased As in solution was the result of release of As(V) from the large reservoir of nonlabile soil As.     

Contrasting effects of dissimilatory iron (III) and arsenic (V) reduction on arsenic retention and transport

Reduction of arsenate As(V) and As-bearing Fe (hydr)- oxides have been proposed as dominant pathways of As release within soils and aquifers. Here we examine As elution from columns loaded with ferrihydrite-coated sand presorbed with As(V) or As(III) at circumneutral pH upon Fe and/or As reduction; biotic stimulated reduction is then compared to abiotic elution. Columns were inoculated with Shewanella putrefaciens strain CN-32 or Sulfurospirillum barnesii strain SES-3, organisms capable of As (V) and Fe (III) reduction, or Bacillus benzoevorans strain HT-1, an organism capable of As(V) but not Fe(III) reduction. On the basis of equal surface coverages, As(III) elution from abiotic columns exceeded As(V) elution by a factor of 2; thus, As(III) is more readily released from ferrihydrite under the imposed reaction conditions. Biologically mediated Asreduction induced by B. benzoevorans enhances the release of total As relative to As (V) under abiotic conditions. However, under Fe reducing conditions invoked by either S. barnesii or S. putrefaciens, approximately three times more As (V or III) was retained within column solids relative to the abiotic experiments, despite appreciable decreases in surface area due to biotransformation of solid phases. Enhanced As sequestration upon ferrihydrite reduction is consistent with adsorption or incorporation of As into biotransformed solids. Our observations indicate that As retention and release from Fe (hydr)oxide(s) is controlled by complex pathways of Fe biotransformation and that reductive dissolution of As-bearing ferrihydrite can promote As sequestration rather than desorption under conditions examined here.     

Removal of arsenic from contaminated soils by microbial reduction of arsenate and quinone

We investigated bioremediation of As-contaminated soils by reductive dissolution of As using a dissimilatory As(V)-reducing bacterium (DARB), Bacillus selenatarsenatis SF-1. We also examined the effect of anthraquinone-2,6-disulfonate (AQDS), an extracellular electron-shuttling quinone, on the As extraction. When B. selenatarsenatis was incubated with As(V)-laden Al precipitates, no acceleration of As dissolution was observed in the presence of AQDS, even though the microbial reduction of AQDS occurred actively. In contrast, AQDS addition significantly enhanced the reductive dissolution of As and Fe in analogous experiments with As(V)-laden Fe(III) precipitates, whereas As dissolution did not occur in the absence of the As(V) reducer. These results indicate the dissolution of As was accelerated by indirect reduction of solid-phase Fe(III) following microbial AQDS reduction, although As(V) reduction is vital for As extraction. B. selenatarsenatis was able to extract As from two types of industrially contaminated soils through reduction of solid-phase As(V) and Fe(III). The copresence of AQDS with B. selenatarsenatis improved the removal efficiency of As from the contaminated soils, concomitantly releasing Fe(II), suggesting that simultaneous use of DARB and electron-shuttling compounds can be an effective strategy for remediation of As-contaminated soils.     

Bioaccessibility of metal cations in soil is linearly related to its water exchange rate constant

Site-specific risk assessments often incorporate the concepts of bioaccessibility (i.e., contaminant fraction released into gastrointestinal fluids) or bioavailability (i.e., contaminant fraction absorbed into systemic circulation) into the calculation of ingestion exposure. We evaluated total and bioaccessible metal concentrations for 19 soil samples under simulated stomach and duodenal conditions using an in vitro gastrointestinal model. We demonstrated that the median bioaccessibility of 23 metals ranged between <1 and 41% under simulated stomach conditions and < 1 and 63% under simulated duodenal conditions. Notably, these large differences in metal bioaccessibility were independent of equilibrium solubility and stability constants. Instead, the relationship (stomach phase R = 0.927; duodenum phase R = 0.891) between bioaccessibility and water exchange rates of metal cations (k(H?O)) indicated that desorption kinetics may influence if not control metal bioaccessibility.     

A model for the effect of rhizodeposition on the fate of phenanthrene in aged contaminated soil

Microcosm data were used to develop a deterministic model to describe how rhizodeposition affects the fate of phenanthrene in aged contaminated soil. Microbial mineralization and soil sequestration of 14C-phenanthrene were compared in microcosms amended weekly with phenolic-rich mulberry root extracts versus unamended controls. Mineralization was higher in the amended soils simulating the rhizosphere (57.7 +/- 0.9%) than in controls simulating bulk (unplanted) soils (53.2 +/- 0.7%) after 201 days (p < 0.05). Humin was the main soil sink for the residual 14C-label. Whereas the total 14C-label associated with humin remained constant in biologically active soils (at about 30%), it increased up to 80% after 201 days in sterile controls. The initial phenanthrene extraction with n-butanol (commonly used to assess bioavailability) slightly underestimated the fraction thatwas mineralized (assessed by 14CO2 recovery). Changes in the unextractable fraction (determined by combustion in a biological oxidizer) suggested the presence of two soil sequestration domains: (1) irreversibly bound residue, and (2) an intermediate transition phase that is unextractable by solvents at a given point in time but could become bioavailable due to physicochemical or biological transformations of the binding matrix. The fate of phenanthrene was accurately modeled by considering the transfer of the 14C label between different soil compartments as first-order kinetic processes. Model simulations suggested that the system was approaching a stable end-point after 201 days of simulated rhizoremediation, and corroborated that microorganisms have a significant impact on the fate of phenanthrene in soil.     

XAS evidence of As(V) association with iron oxyhydroxides in a contaminated soil at a former arsenical pesticide processing plant

The molecular-level speciation of arsenic has been determined in a soil profile in the Massif Central near Auzon, France that was impacted by As-based pesticides by combining conventional techniques (XRD, selective chemical extractions) with X-ray absorption spectroscopy (XAS). The arsenic concentration is very high at the top (>7000 mg kg(-1)) and decreases rapidly downward to a few hundreds of milligrams per kilogram. A thin layer of schultenite (PbHAsO4), a lead arsenate commonly used as an insecticide until the middle of the 20th century, was found at 10 cm depth. Despite the occurrence of this As-bearing mineral, oxalate extraction indicated that more than 65% of the arsenic was released upon dissolution of amorphous iron oxides, suggesting a major association of arsenic with these phases within the soil profile. Since oxalate extraction cannot unambiguously distinguish among the various chemical forms of arsenic, these results were confirmed by a direct in situ determination of arsenic speciation using XAS analysis. XANES data indicate that arsenic occurs mainly as As(V) along the soil profile except for the topsoil sample where a minor amount (7%) of As(III) was detected. EXAFS spectra of soil samples were fit by linear combinations of model compounds spectra and by a shell-by-shell method. These procedures clearly confirmed that As(V) is mainly (at least 80 wt %) associated with amorphous Fe(III) oxides as coprecipitates within the soil profile. If any, the proportion of schultenite, which was evidenced by XRD in a separate thin white layer, does not account for more than 10 wt % of arsenic in soil samples. This study emphasizes the importance of iron oxides in restricting arsenic dispersal within soils following dissolution of primary As-bearing solids manufactured for use as pesticides and released into the soils.     

Gastrointestinal microbes increase arsenic bioaccessibility of ingested mine tailings using the simulator of the human intestinal microbial ecosystem

It is widely accepted that the use of total metal concentrations in soil overestimates metal risk from human ingestion of contaminated soils. In vitro simulators have been used to estimate the fraction of arsenic present in soil that is bioaccessible in the human digestive track. These approaches assume that the bioaccessible fraction remains constant across soil total metal concentrations and that intestinal microbiota do not contribute to arsenic release. Here, we evaluate both of these assumptions in two size fractions (bulk and <38 microm) of arsenic-rich mine tailings from the Goldenville, Lower Seal Harbour, and Montague Gold Districts, Nova Scotia. These samples were evaluated using an in vitro gastrointestinal model, the Simulator of the Human Intestinal Ecosystem (SHIME). Arsenic bioaccessibility, which ranged between 2 and 20% in the small intestine and 4 and 70% in the colon, was inversely related to total arsenic concentration in the mine tailings. Additionally, arsenic bioaccessibility was greater in the bulk fraction than in the <38 microm fraction in the small intestine and colon while colon microbes increased the bioaccessibility of arsenic in mine tailings. These results suggest that the practice of using a constant percent arsenic bioaccessibility across all metal concentrations in risk assessment should be revisited.     

Passive samplers provide a better prediction of PAH bioaccumulation in earthworms and plant roots than exhaustive, mild solvent, and cyclodextrin extractions

A number of extraction methods have been developed to assess polycyclic aromatic hydrocarbon (PAH) bioavailability in soils. As these methods are rarely tested in a comparative manner, against different test organisms, and using field-contaminated soils, it is unclear which method gives the most accurate measure of the actual soil ecosystem exposure. In this study, PAH bioavailability was assessed in ten field-contaminated soils by using exhaustive acetone/hexane extractions, mild solvent (butanol) extractions, cyclodextrin extractions, and two passive sampling methods; solid phase micro extraction (SPME) and polyoxymethylene solid phase extraction (POM-SPE). Results were compared to actual PAH bioaccumulation in earthworms (Eisenia fetida) and rye grass (Lolium multiflorum) roots. Exhaustive, mild solvent and cyclodextrin extractions consistently overpredicted biotic concentrations by a factor of 10-10?000 and therefore seem inappropriate for predicting PAH bioaccumulation in field contaminated soils. In contrast, passive samplers generally predicted PAH concentrations in earthworms within a factor of 10, although correlations between predicted and measured concentrations were considerably scattered. The same applied to the plant data, where passive samplers also tended to underpredict root concentrations. These results indicate the potential of passive samplers to predict PAH bioaccumulation, yet call for comparative studies between passive samplers and further research on plant bioavailability.