Leaching of Perfluoroalkyl Acids during Unsaturated Zone Flushing at a Field Site Impacted with Aqueous Film Forming Foam.

While several studies have focused on perfluoroalkyl acid (PFAA) leaching from soils, field studies evaluating the relationship between PFAA mass removal and porewater concentrations as the PFAA source becomes depleted are lacking. Herein, in situ water flushing was performed at a site historically impacted with AFFF to accelerate the leaching of PFAAs from unsaturated soils in a highly characterized field test cell. Porous cup suction lysimeters were used to assess the changes in PFAA porewater concentrations as a function of PFAA mass removal from the unsaturated soils, where flushing was intermittently paused to determine ambient PFAA porewater concentrations. Results showed that the fractional decreases in PFAA porewater concentrations during flushing exceeded the fractional decrease in PFAA mass removal from the soil. PFOS porewater concentrations decrease by 76% (with negligible rebound) compared to only a 7.4% decrease in overall PFOS mass removed from the unsaturated zone. Overall, the results observed herein suggest that, when considering soil impacts to groundwater, less stringent soil cleanup criteria than those that consider an equivalent relationship between mass removal and mass discharge may be appropriate. In addition, remedial approaches that remove only a modest fraction of the PFAA soil mass may be protective of underlying groundwater, particularly for perfluorinated sulfonates with at least six carbons.

A field study to assess the role of air-water interfacial sorption on PFAS leaching in an AFFF source area.

Field-deployed lysimeters were used to measure the concentrations of poly- and perfluoroalkyl substances (PFASs) in soil porewater at a site historically impacted with aqueous film forming foam (AFFF). Samples collected over a 49-day period showed that perfluorooctane sulfonate (PFOS) and perfluorohexane sulfonate (PFHxS) were the PFASs with the highest concentrations in porewater, with concentrations of approximately 10,000 and 25,000 ng L-1, respectively. The corresponding average mass flux to underlying groundwater observed for PFOS and PFHxS was 28,000 ± 11,000 and 92,000 ± 32,000 ng m-2 d-1, respectively. Employing the use of batch desorption isotherms (soil:water slurries) to determine desorption Kd values resulted in an overestimation of PFAS porewater concentrations by a factor for 1.4 to 4. However, using the desorption Kd values from the batch desorption isotherms in combination with a PFAS mass balance that incorporated PFAS sorption at the air-water interface resulted in improved predictions of the PFAS porewater concentrations. This improvement was most notable for PFOS, where inclusion of air-water interfacial sorption resulted in a 58% reduction in the predicted PFOS porewater concentration and predicted PFOS porewater concentrations that were identical (within the 95% confidence interval) to the lysimeter measured PFOS porewater concentration. Overall these results highlight the potentially important role of air-water interfacial sorption on PFAS migration in AFFF-impacted unsaturated soils in an in situ field setting.

Field demonstration of N-Nitrosodimethylamine (NDMA) treatment in groundwater using propane biosparging.

N-Nitrosodimethylamine (NDMA) is found in groundwater and drinking water from industrial, agricultural, water treatment, and military/aerospace sources, and it must often be treated to part-per-trillion (ng/L) concentrations. The most effective remedial technology for NDMA in groundwater is pump-and-treat with ultraviolet irradiation (UV), but this approach is expensive because it requires ex situ infrastructure and high energy input. The objective of this project was to evaluate an in situ biological treatment approach for NDMA. Previous laboratory studies have revealed that propane-oxidizing bacteria are capable of biodegrading NDMA from µg/L to low ng/L concentrations (Fournier et al., 2009; Webster et al., 2013). During this field study, air and propane gas were sparged into an NDMA-contaminated aquifer for more than 1 year. Groundwater samples were collected throughout the study from a series of monitoring wells within, downgradient, and sidegradient of the zone of influence of the biosparge system. Over the course of the study, NDMA concentrations declined by 99.7% to >99.9% in the four monitoring wells within the zone of influence of the biosparge system, reaching low ng/L concentrations whereas the control well declined by only 14%. Pseudo first-order degradation rate constants for NDMA in system monitoring wells ranged from ∼0.019 day -1 to 0.037 day -1 equating to half-lives ranging from 19 to 36 days. Native propanotrophs increased by more than one order of magnitude in the propane-impacted wells but not in the control well. The field data show for the first time that propane biosparging can be an effective in situ approach to reduce the concentrations of NDMA in a groundwater to ng/L concentrations.

Passive in situ biobarrier for treatment of comingled nitramine explosives and perchlorate in groundwater on an active range.

Hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX), octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine (HMX), and perchlorate (ClO4-) are common, and often co-mingled, contaminants at military ranges worldwide. This project investigated the feasibility of using a passive emulsified oil biobarrier plus a slow release pH buffering reagent to remediate RDX, HMX, and ClO4- in a low pH aquifer at an active range. A 33 m biobarrier was emplaced perpendicular to the contaminant plumes, and dissolved explosives, perchlorate, and other relevant parameters were monitored. The pH increased and the DO and ORP decreased after emulsified oil injection, leading to >90% reductions in perchlorate, RDX, and HMX compared to upgradient groundwater. Some nitroso breakdown products were observed immediately downstream of the barrier, but generally decreased to below detection limits farther downgradient. First-order rate constants of approximately 0.1/d were obtained for all three contaminants. Dissolved metals (including As) also increased in the wells immediately adjacent to the barrier, but attenuated as the plume re-aerated in downgradient areas. Biobarrier installation and sampling were performed during scheduled range downtime and had no impacts to ongoing range activities. The field trial suggests that an emulsified oil biobarrier with pH buffering can be a viable alternative to remove explosives and perchlorate from shallow groundwater on active ranges.

In situ bioremediation of 1,2-dibromoethane (EDB) in groundwater to part-per-trillion concentrations using cometabolism.

1,2-Dibromoethane (ethylene dibromide; EDB) is a probable human carcinogen that was historically added to leaded gasoline as a scavenger to prevent the build-up of lead oxide deposits in engines. Studies indicate that EDB is present at thousands of past fuel spill sites above its stringent EPA Maximum Contaminant Level (MCL) of 0.05 µg/L. There are currently no proven in situ options to enhance EDB degradation in groundwater to meet this requirement. Based on successful laboratory studies showing that ethane can be used as a primary substrate to stimulate the aerobic, cometabolic biodegradation of EDB to <0.015 µg/L (Hatzinger et al., 2015), a groundwater recirculation system was installed at the FS-12 EDB plume on Joint Base Cape Cod (JBCC), MA to facilitate in situ treatment. Groundwater was taken from an existing extraction well, amended with ethane, oxygen, and inorganic nutrients and then recharged into the aquifer upgradient of the extraction well creating an in situ reactive zone. The concentrations of EDB, ethane, oxygen, and anions in groundwater were measured with time in a series of nested monitoring wells installed between the extraction and injection well. EDB concentrations in the six monitoring wells that were hydraulically well-connected to the pumping system declined from ~ 0.3 µg/L (the average concentration in the recirculation cell after 3 months of operation without amendment addition) to <0.02 µg/L during the 4-month amendment period, meeting both the federal MCL and the more stringent Massachusetts MCL (0.02 µg/L). The data indicate that cometabolic treatment is a promising in situ technology for EDB, and that low regulatory levels can be achieved with this biological approach.

Evidence of rock matrix back-diffusion and abiotic dechlorination using a field testing approach.

An in situ field demonstration was performed in fractured rock impacted with trichloroethene (TCE) and cis-1,2-dichloroethene (DCE) to assess the impacts of contaminant rebound after removing dissolved contaminants within hydraulically conductive fractures. Using a bedrock well pair spaced 2.4m apart, TCE and DCE were first flushed with water to create a decrease in dissolved contaminant concentrations. While hydraulically isolating the well pair from upgradient contaminant impacts, contaminant rebound then was observed between the well pair over 151days. The magnitude, but not trend, of TCE rebound was reasonably described by a matrix back-diffusion screening model that employed an effective diffusion coefficient and first-order abiotic TCE dechlorination rate constant that was based on bench-scale testing. Furthermore, a shift in the TCE:DCE ratio and carbon isotopic enrichment was observed during the rebound, suggesting that both biotic and abiotic dechlorination were occurring within the rock matrix. The isotopic data and back-diffusion model together served as a convincing argument that matrix back-diffusion was the mechanism responsible for the observed contaminant rebound. Results of this field demonstration highlight the importance and applicability of rock matrix parameters determined at the bench-scale, and suggest that carbon isotopic enrichment can be used as a line of evidence for abiotic dechlorination within rock matrices.

Abiotic dechlorination in rock matrices impacted by long-term exposure to TCE.

Field and laboratory tests were performed to evaluate the abiotic reaction of trichloroethene (TCE) in sedimentary rock matrices. Hydraulically conductive fractures, and the rock directly adjacent to the hydraulically conductive fractures, within a historically contaminated TCE bedrock aquifer were used as the basis for this study. These results were compared to previous work using rock that had not been exposed to TCE (Schaefer et al., 2013) to assess the impact of long-term TCE exposure on the abiotic dechlorination reaction, as the longevity of these reactions after long-term exposure to TCE was hitherto unknown. Results showed that potential abiotic TCE degradation products, including ethane, ethene, and acetylene, were present in the conductive fractures. Using minimally disturbed slices of rock core at and near the fracture faces, laboratory testing on the rocks confirmed that abiotic dechlorination reactions between the rock matrix and TCE were occurring. Abiotic daughter products measured in the laboratory under controlled conditions were consistent with those measured in the conductive fractures, except that propane also was observed as a daughter product. TCE degradation measured in the laboratory was well described by a first order rate constant through the 118-d study. Observed bulk first-order TCE degradation rate constants within the rock matrix were 1.3×10(-8) s(-1). These results clearly show that abiotic dechlorination of TCE is occurring within the rock matrix, despite decades of exposure to TCE. Furthermore, these observed rates of TCE dechlorination are expected to have a substantial impact on TCE migration and uptake/release from rock matrices.

Coupled diffusion and abiotic reaction of trichlorethene in minimally disturbed rock matrices.

Laboratory experiments were performed using minimally disturbed sedimentary rocks to measure the coupled diffusion and abiotic reaction of trichloroethene (TCE) through rock core samples. Results showed that, for all rock types studied, TCE dechlorination occurred, as evidenced by generation of acetylene, ethene, and/or ethane daughter products. First-order bulk reaction rate constants for TCE degradation ranged from 8.3 × 10(-10) to 4.2 × 10(-8) s(-1). Observed reaction rate constants showed a general correlation to the available ferrous iron content of the rock, which was determined by evaluating the spatial distribution of ferrous iron relative to that of the rock porosity. For some rock types, exposure to TCE resulted in a decrease in the effective diffusivity. Scanning electron microscopy (SEM) indicated that the decrease in the effective diffusivity was due to a decrease in the porosity that occurred after exposure to TCE. Overall, these coupled diffusion and reaction results suggest that diffusion of TCE into rock matrices as well as the rate and extent of back-diffusion may be substantially mitigated in rocks that contain ferrous iron or other naturally occurring reactive metals, thereby lessening the impacts of matrix diffusion on sustaining dissolved contaminant plumes in bedrock aquifers.