In this 10 year study, Brazilian gasoline (100 L, containing 24% ethanol by volume) was released to a sandy aquifer to evaluate the natural attenuation of benzene, toluene, ethylbenzene, and total xylenes (BTEX) in the presence of ethanol. Groundwater concentrations of BTEX, ethanol, and degradation products (e.g., acetate and methane) were measured over the entire plume using an array of monitoring well clusters, to quantify changes in plume mass and region of influence. Ethanol biodegradation coincided with the development of methanogenic conditions while acetate (a common anaerobic metabolite) accumulated. The benzene plume expanded beyond the 30 m long monitored area and began to recede after 2.7 years, when ethanol had disappeared. Theoretical calculations suggest that the transient accumulation of acetate (up to 166 mg L(-1)) may have hindered the thermodynamic feasibility of benzene degradation under methanogenic conditions. Yet, benzene removal proceeded relatively fast compared to literature values (and faster than the alkylbenzenes present at this site) after acetate concentrations had decreased below inhibitory levels. Thus, site investigations of ethanol blend releases should consider monitoring acetate concentrations. Overall, this study shows that inhibitory effects of ethanol and acetate are relatively short-lived, and demonstrates that monitored natural attenuation can be a viable option to deal with ethanol blend releases.

The potential groundwater impacts of biodiesel releases have received limited attention despite the increasing probability of such events. In this work, microcosms were prepared with unacclimated sediment and groundwater from the Ressacada Experimental Site (Florianopolis, Santa Catarina, Brazil) and spiked with 54.8 mg/L of pure soybean or castor oil biodiesel (B100). Oxygen was purged from the microcosms to mimic commonly anoxic and hypoxic conditions at fuel‐impacted sites; low background concentrations of nitrate (1.2 to 2.5 mg/L) and sulfate (2.2 to 3.0 mg/L) were present. Biodegradation was assessed by the removal of fatty acid methyl esters and hydrocarbon components relative to sterile controls. Approximately 80% of soybean biodiesel was biotransformed in 41 d, compared to only 40% of castor oil biodiesel removed in 90 d. The higher persistence of castor biodiesel was attributed to its higher viscosity and lower bioavailability. Additional microcosms were prepared similarly to assess the impact of biodiesel on hydrocarbon degradation. These microcosms were spiked with benzene (2.9 mg/L) and toluene (0.8 mg/L) with or without soybean biodiesel (54.8 mg/L). The biodiesel had an inhibitory effect, increasing the time required to remove toluene from 25 to 34 d. Similarly, 45% of benzene was removed in the presence of biodiesel within 34 d, compared to 90% in the absence of biodiesel. Overall, we postulate that the relatively high viscosity of biodiesel is conducive to limited migration potential and a smaller but longer lasting inhibitory region of influence, compared to that exerted by more soluble, more mobile, and readily degradable biofuels such as ethanol. However, controlled release studies are needed to test this hypothesis and characterize the complex dynamics of such releases.

The assessment of biodegradation activity in contaminated aquifers is critical to demonstrate the performance of bioremediation and natural attenuation and to parameterize models of contaminant plume dynamics. Real time quantitative PCR (qPCR) was used to target the catabolic bssA gene (coding for benzylsuccinate synthase) and a 16S rDNA phylogenetic gene (for total Bacteria) as potential biomarkers to infer on anaerobic toluene degradation rates. A significant correlation (P = 0.0003) was found over a wide range of initial toluene concentrations (1-100 mg/l) between toluene degradation rates and bssA concentrations in anaerobic microcosms prepared with aquifer material from a hydrocarbon contaminated site. In contrast, the correlation between toluene degradation activity and total Bacteria concentrations was not significant (P = 0.1125). This suggests that qPCR targeting of functional genes might offer a simple approach to estimate in situ biodegradation activity, which would enhance site investigation and modeling of natural attenuation at hydrocarbon-contaminated sites.

A pilot‐scale aquifer system (8 m3 continuous‐flow tank packed with fine grain sand) was used to evaluate groundwater quality impacts from a continuous release of 10% v:v ethanol solution in water mixed with benzene and toluene (50 mg/L each). The geochemical footprint (methane [CH4], volatile fatty acids [VFAs], pH, oxidation reduction potential [ORP], dissolved oxygen [DO], and temperature) was monitored more than 11 months. A rapid depletion of DO (from 5.3 to less than 0.1 mg/L) and a decrease of ORP (from 110 to –310 mV) were observed within 25 d of the release. The high‐biochemical oxygen demand exerted by ethanol resulted in strongly anaerobic conditions, indicated by the accumulation of CH4 (up to 17.9 mg/L) and VFAs (up to 226 mg/L acetic acid and 280 mg/L n‐butyric acid). Measurements at the sand surface (40 cm from the water table) using a portable combustible gas detector did not detect CH4. However, accumulation of VFAs (particularly n‐butyric acid) during the summer exceeded the secondary maximum contaminant level value for odor (odor levels extrapolated from aqueous concentrations), which represents a previously unreported aesthetic impact. Temperature variations (3.9 to 30.0 °C) significantly affected microbial activities, and a strong correlation was observed between groundwater temperature and CH4/VFAs generation (p less than 0.05). Overall, these results suggest that seasonal variation of odor generation and CH4 concentration should be considered at sites contaminated with fuel ethanol blends.