Arsenic contaminated groundwater: exploration of the role of organic carbon in mobilization processes and evaluation of mechanisms of arsenic sequestration by in situ treatment systems

Thesis (Ph.D.)--University of Washington, 2016-12Arsenic contamination of groundwater is a global concern, with environmental presence attributable to both geogenic and anthropogenic factors. Presence in the environment threatens both public and environmental health. In this study, we explored both the mobilization and sequestration of arsenic in subsurface environments, both in in situ treatment systems that sought to sequester the arsenic by intentional interference with the biochemistry and in natural systems where anthropogenic activities hold the potential to inadvertently exacerbate existing public health problems caused by arsenic contamination. Our investigation into these arsenic stabilization mechanisms involved fieldwork and laboratory-based experiments and analysis methods to study three different groundwater systems. The fieldwork collected sediment and water samples from these different locations. Laboratory batch incubation experiments were conducted for two locations, to elucidate arsenic mobilization processes, with further laboratory characterizations of these experimental materials and those collected from the third site, including solid-phase speciation of arsenic and iron in sediment, molecular characterization of microbially-altered organic carbon, and isotopic characterization of carbon pools. Our work conducted on in situ treatment systems analyzed two different treatments that utilized induced sulfate reduction (ISR) in permeable reactive barriers (PRB) as a means to form sulfide and iron minerals that incorporate or sorb arsenic, removing it from groundwater. Our work found that PRBs that are formed using zerovalent iron (ZVI) in conjunction with the sulfate and carbon materials are more effective at arsenic removal, likely due to the formation of more stable minerals. Our data suggest that arsenic stabilization in non-ZVI PRBs are primarily controlled by arsenic-sulfide associations, while ZVI PRBs saw greater formation of iron sulfides and iron oxyhydroxides. Our investigations into mechanisms responsible for subsurface mobilization of arsenic into groundwater were primarily driven by laboratory incubation experiments of aquifer materials collected from arsenic-contaminated locations. Materials collected from an aquifer in Bangladesh were incubated and sampled over one-half year, with the goal of better understanding the labile organic carbon available to fuel reductive dissolution processes that drive arsenic mobilization. Ultrahigh resolution mass spectrometry revealed a large, heterogeneous pool of bioavailable organic carbon in the sediment porewater. Despite the substantial microbial degradation of this carbon pool, most of the compounds were those types traditionally considered recalcitrant or less energetically-favorable. Our results lend weight to an ever-growing shift in understanding of carbon bioavailability. Additionally, both this sediment porewater and the aquifer recharge waters contained many organosulfur compounds, suggesting that the presence of such compounds may potentially be useful as a fingerprint for organic matter derived from organic-rich, anaerobic subsurface environments. The existence of this pool of organic carbon in subsurface sediments that holds the potential to fuel microbial reactions that impact groundwater quality, led to our final investigation of a process that may occur in situ and mobilize such a pool: microbial priming. Laboratory incubations of aquifer sediment from Cambodia were conducted to show the occurrence of microbial priming and its potential to release mobilize arsenic in the subsurface. The results did not show explicit evidence of priming, but did show an increase in the onset of microbial activity in treatments amended with labile organic carbon. Moreover, even in treatments amended with high concentrations of labile carbon, notable arsenic mobilization was not observed. More energetically-favorable electron acceptors, iron and manganese, were reduced and dissolved into solution, but at later timepoints in the experiment began to decrease in dissolved concentrations, indicating mineral precipitation