Biologically mediated abiotic degradation (BMAD) of chlorinated solvents using Fe-bearing clay minerals

Abstract

Chlorinated solvents, such as tetrachloroethene (PCE), trichloroethene (TCE), and trihalomethanes (THMs), are extremely persistent and widespread groundwater contaminants. Due to their ecotoxicological and physiochemical properties, chlorinated solvent contamination poses a major threat to the quality of safe potable groundwater. Bioremediation is often employed for sites requiring a low-cost long-term remediation scheme to transform these pollutants to lesser chlorinated products. Whilst many studies have shown successful removal of PCE and TCE to a certain extent, bioremediation often stalls at cis-DCE and vinyl chloride (VC), which are much more toxic than either PCE or TCE and requires specific bacterial assemblages to remove. Alternatively, the abiotic degradation pathway, for example in the presence of minerals, favours the conversion of PCE and TCE to benign C2 gases. Fe-bearing clay minerals are seen as a promising choice of mineral for abiotic degradation because of their resistance to dissolution and ubiquity in the subsurface, which make them a sustainable source of reduction. Furthermore, interactions between aqueous Fe(II) and aqueous S(-II), which is produced via microbial activity, and structural Fe(III) bound inside clay minerals leads to the formation of an electron-doped clay mineral which can be used to reduce contaminants. However, electron transfer from aqueous Fe(II) or S(-II) also leads to the formation of solid oxidation product(s), with their exact nature and identity causing some controversy, with recent work suggesting the formation of nanocrystalline mixed valent Fe phases such as green rust, which, in turn, could contribute to contaminant transformation as potential reactive mineral intermediates (RMIs). Here we evaluated the reactivity of Fe-bearing clay minerals toward the reductive degradation of PCE, TCE, and THMs. We compared the reactivity with dithionite-reduced clay minerals and for high and low Fe content clay minerals (nontronite, NAu-1; montmorillonite, SWy-2), to modulate the clay mineral Fe(II) concentration, and varied the concentrations of aqueous Fe(II) or aqueous S(-II) to investigate the effect the concentration of each aqueous species has on the identity and potential reactivity of any RMIs that may form. We also used a separate reactive probe, hexachloroethane (HCA), which follows the same reductive -elimination reaction mechanism as the abiotic degradation pathway PCE and TCE follows, to further investigate any differences in the Fe-bearing RMIs reactivity towards degrading recalcitrant contaminants. We complemented our reactivity studies with techniques to identify any RMIs and used a combination of 57Fe-Mössbauer spectroscopy and X-ray diffraction (XRD). Transformation of PCE and TCE was not observed when NAu-1 (20 wt% Fe) or dithionitereduced SWy-2 (2.5 wt% Fe) was present, suggesting that clay mineral structural Fe(II) alone was not capable of reducing PCE and TCE. Similarly, we found no detection of any PCE or TCE transformation products from any of our experimental conditions when Fe-bearing clay minerals were reduced using aqueous S(-II), and also found no detectable transformation products from our THMs degradation study. Interestingly, PCE and TCE transformation products (acetylene, ethene, and ethane) were detected only in reactors containing SWy-2 amended with 20 mM aqueous Fe(II). We found from our experiments using HCA that the reactivity of HCA is dependent on both the clay mineral Fe content, as well as the initial aqueous Fe(II) concentration used. We believe that the rate of HCA removal is affected by the precipitate that is formed, with the reactivity of the precipitate dependent on the Fe stoichiometry and the mineral phase that has formed. Mössbauer spectroscopy was used to identify the Fe stoichiometry of both the structural clay mineral Fe as well as any Fe-bearing precipitates that formed. In the case of Fe(II)-reduced clay minerals, Mössbauer shows that both precipitates that form in the presence of 20 mM Fe(II) have similar amounts of Fe(II) present (NAu-1: 80.9%; SWy-2: 82.9%), the ratio of Fe(II) bound in the precipitate relative to the clay mineral Fe total is far greater when SWy-2 is present compared with NAu-1 ( 4.3 vs 0.5 for 20 mM amended SWy-2 and NAu-1 respectively). XRD analysis has shown that the precipitate formed for 20 mM amended SWy-2 bears some resemblance to a 1:1 trioctahedral Fe-silicate, yet hyperfine parameters from Mössbauer spectroscopy show Fe(III) coordination for a 2:1 trioctahedral mineral. The reverse appears to be true in the case of S(-II)-reduced clay minerals, where on visual inspection precipitates only appeared to form when NAu-1 was present, but due to the limitations of Mössbauer spectroscopy and XRD on identifying RMIs, it was difficult to confirm the identity of these Fe-S precipitates. Overall, while we have evaluated that Fe-bearing clay minerals may not play a pivotal role in the reductive degradation of chlorinated solvents, we have highlighted the potential importance of RMIs in the subsurface as contributors to abiotic natural attenuation, with the composition and identity of RMIs that form due to interfacial electron transfer being strongly controlled by aqueous Fe(II) or S(-II) concentrations and clay mineral Fe content. The formation of these RMIs from interfacial electron transfer can also provide further insight into Fe-bearing mineral formation throughout geological history such as the formation of low-Fe(III) greenalite.