Evaluation of fuel ethanol releases in a pilot-scale aquifer tank: Source dynamics, NAPL migration and microbial community response

Ethanol is playing a key role in current discussions on energy, agriculture, taxes and the environment. This work addresses the potential environmental impacts and behavior of subsurface fuel-ethanol releases. A continuous-flow 8,150-L pilot-aquifer tank packed with sand was used to simulate two spill scenarios: (1) fuel-grade ethanol (E95, 95% v/v ethanol, 5% v/v hydrocarbon mixture as a denaturant) into uncontaminated soil, and (2) neat ethanol (100% v/v) release onto gasoline-contaminated soil. Measurement of ethanol and hydrocarbon concentrations in groundwater and capillary-fringe pore water from over 30-locations over 120+ days provided a quantitative evaluation of the extent of plume migration, longevity, and impacts to groundwater quality. Real-time quantitative PCR (RTQ-PCR) was also used to estimate temporal and spatial trends in concentrations of total bacteria (16s rDNA) and various genotypes that inhabit different electron-accepting zones at sites undergoing natural attenuation. Furthermore, the anaerobic catabolic gene bssA (coding for benzylsuccinate synthase), and the aerobic catabolic genes dmpN (coding for phenol hydroxylase) and todC1 (coding for toluene dioxygenase) were also quantified as biomarkers for BTEX biodegradation. Ethanol, which is buoyant and hygroscopic, quickly migrated upwards and spread laterally within the capillary-zone. Horizontal migration of ethanol occurred through a shallow thin layer with minimal vertical dispersion, and was consistently 10-times slower than the preceding bromide tracer. Dyes, one hydrophobic (Sudan-IV) and one hydrophilic (Fluorescein) provided evidence that the fuel hydrocarbons phase separated from the E95 mixture as ethanol was diluted by pore water and its cosolvent effect was diminished. The neat ethanol spill mobilized the pre-existing hydrocarbon NAPL down-gradient. Neither of the highly concentrated spills had a bactericidal impact on the microbial community, and cell growth coincided with ethanol availability. Bacteria concentrations increased by at least one-order of magnitude as did bacteria harboring todC1 and dmpN after each spill. However, bacteria harboring bssA were not detected, suggesting that longer acclimation time may be required to establish anaerobic hydrocarbon degraders. It appears that microbial impacts are mainly related to O 2 depletion, but rebound can be relatively fast, and fortuitous proliferation of aerobic BTEX degraders (growing on ethanol) is likely following the relatively rapid ethanol washout