Application of Biochar to Stabilize Mercury in Riverbank Sediments and Floodplain Soils from South River, VA under Conditions Relevant to Riverine Environments

Abstract

Disposal of mercury (Hg) containing products related to industrial activities has led to large-scale watershed contamination across the globe, posing long-term risks to human health due to its persistent properties. Hg in terrestrial systems can re-enter aquatic systems directly through soil erosion and sediment resuspension, and indirectly through reductive dissolution of manganese (Mn) and iron (Fe) oxides, desorption from clays and other minerals, and breakdown of soil organic matter. Hg is transformed into methylmercury (MeHg), a well-known neurotoxin that accumulates through the food chain, mainly by microbially driven processes under anoxic conditions. Remediation of Hg in riverine environments is challenging due to dynamic redox oscillations caused by flooding and drainage which influence Hg mobility and bioavailability. Mercury sulfate (HgSO4) was used by a textile plant in Waynesboro, VA between 1920-1950s, and as a result of inadvertent discharge elevated Hg concentrations have been observed in the South River watershed since 1970, long after cessation of HgSO4 use. Biochars have been proposed for use in reactive capping mats or as soil amendments for in situ Hg stabilization. Studies evaluating the effectiveness of biochar for stabilizing Hg focus on the effectiveness under fully-saturated conditions, but how treatment systems respond to more environmentally relevant conditions, such as drying and rewetting, is less studied. This dissertation evaluates selected biochars for Hg stabilization in river bank sediments and floodplain soils collected along the South River using laboratory-scale experiments under conditions relevant to riverine environments, including flooding and drainage, fully-saturated anoxic, and drying and rewetting conditions. Five biochars selected for study were: hardwood biochar (OAK), sulfurized-hardwood biochar (MOAK), and biochar prepared from ethanol refinery by-products, including distillers` grains (DIS), anaerobic digestate (DIG), and a mixture of digestate and distillers` grains (75G25S). OAK was evaluated for potential application as a reactive capping mat as well as a soil amendment, and the other biochars were evaluated as soil amendments. To evaluate OAK as a reactive capping mat which intercepts flow paths under flooding and drainage conditions, the treatment system consisted of two sets of modified humidity-cell experiments operated for 100 weekly cycles. The weekly cycles started with dry air, water-saturated air, and were followed by an aqueous leach at day 7 of each week. Each set contained a source column and a treatment column. Source columns contained river bank sediment and floodplain soil collected from different locations along the South River. Treatment columns contained 50% v/v OAK and non-reactive quartz sand. South River water (SRW) was used as input solution for the source column containing river bank sediment, and acidic rain water (ARW) was used as an input solution for the source column containing floodplain soil. Leachates collected from the source columns were used as input solutions for the treatment columns. More than 80% Hg was retained in the treatment columns with limited formation of MeHg in both aqueous and solid phases. Results of micro X-ray fluorescence mapping (µ-XRF) indicate Hg retained on OAK co-occurred with Si, S, Fe and Cu within the biochar porous structure. Results of S K-edge X-ray absorption near edge structure (XANES) analyses indicate lower fractions of sulfoxide for OAK in treatment systems than in untreated OAK. These synchrotron-based analyses indicate that Hg accumulation in OAK when used as a reactive capping mat under flooding and drainage conditions can be attributed to retention of the Hg as particulates in the biochar porous structure as well as formation of complexes with O-containing functionalities on the biochar. OAK and MOAK were evaluated as soil amendments in a floodplain soil under anoxic saturated conditions using laboratory-scale microcosm experiments followed by drying and rewetting of the biochar-amended systems. Floodplain soil was mixed with or without biochars and equilibrated with SRW under anoxic conditions for up to 200 d, and solid materials collected at selected reaction intervals were dried under oxic conditions for 90 d and rewet under anoxic conditions for an additional 90 d. Limited Hg removal was observed in OAK-amended systems. Addition of MOAK enhanced Hg removal under anoxic conditions without promoting MeHg production. After drying and rewetting, Hg in OAK-amended systems was remobilized, likely due to association with dissolved organic carbon (DOC), while Hg in MOAK-amended systems remained at low concentrations. Increases in solid MeHg concentrations coupled with increases in aqueous Mn, Fe, SO42- and HS- concentrations in MOAK-amended systems were observed. 16s RNA pyrosequencing analysis suggests shifts in Hg methylating community composition toward sulfur-reducing bacteria (SRB). Drying and rewetting alters the structure of the microbial community, therefore generating conditions favourable for MeHg production in MOAK-amended systems. DIG, DIS, and 75G25S were evaluated as alternative soil amendments in floodplain soil following the same experimental protocol used to evaluate OAK and MOAK as soil amendments. Addition of digestate-based biochar (DIG and 75G25S) led to greater Hg uptake and lower MeHg concentrations than the addition of DIS. After drying and rewetting, increases in DOC concentrations were observed in digestate-biochar amended systems, while concentrations of Hg remained at low concentrations, suggesting Hg in digestate-based biochars was less likely affected by the release of DOC. Concentrations of MeHg in these biochar-amended systems remained at low concentrations, and solid MeHg content was 50% lower after drying and rewetting than under initial anoxic conditions. These experiments suggest that digestate-based biochars can potentially be used as soil amendment in fully-saturated anoxic environments with limited impacts from drying and rewetting on the system performance. MOAK and DIG were further evaluated as soil amendments under periodic wetting and drying conditions due to their greater control of Hg under anoxic conditions. The periodic wetting and drying conditions were mimicked using a modified humidity-cell experiment. Each wetting and drying cycle contained a wetting, leaching and drainage period, and a total of ten cycles was conducted. In wetting periods, SRW was added and the system was allowed to stagnate for 14 d, and then drained by gravity during leaching periods, after which solid materials were dried before the next wetting period. An early period of elevated leachate concentrations of THg, MeHg, DOC, and Mn, was observed in the soil control and biochar-amended systems. Limited Hg removal (up to 57%) was observed in the biochar-amended systems at steady state. Minimal MeHg (<0.6 ng L-1) was observed in soil control and DIG-amended systems, while MeHg concentrations were up to 158 ng L-1 in the MOAK-amended system during the early flush. THg and MeHg concentrations in the early flush were positively correlated with DOC and Mn concentrations. Concentrations of SO42- increased with decreases in pH and alkalinity in the MOAK-amended system. Initial release of elevated (SO42-) was observed in the DIG-amended system. S K-edge XANES spectra indicate polysulfur is the predominant S form in the biochar-amended systems. Results of 16s rRNA pyrosequencing suggest the microbial community in the MOAK-amended system shifted toward sulfur oxidizers, while the microbial community in the DIG-amended system was similar to the soil control. The greater abundances of sulfur oxidizers in the MOAK-amended system suggest MOAK is more available for microbial organisms to promote microbially-driven oxidation under periodic wetting and drying conditions. Results of this study suggest that Hg removal under conditions relevant to riverine environment depends on application methods, biochar properties and biogeochemical conditions. Biochars may be used as reactive material embedded in geotextiles for river bank stabilization and as soil amendments. Dynamic oscillations representative of riverine environments indirectly influences the effectiveness of Hg removal and may result in unintended consequences. Careful characterization of biochar properties and local environments are recommended prior to implementing large-scale biochar applications.