Ethanol Fate and Transport:  An Updated Review of its Impact on Subsurface Gasoline Contamination
Jim Davidson, Exponent, Boulder, CO

The Fate of Oxygenates and BTEX from Gasolines Containing MTBE, TBA and Ethanol: Initial findings from a Controlled Field Experiment, Borden, Canada
Jim Barker, University of Waterloo, Waterloo, ON, Canada

The Impact of Fuel Ethanol on Groundwater:  Source Behavior
Bill Rixey, University of Houston, Houston, TX

The Impact of Fuel Ethanol on Groundwater:  Microbial Response
Pedro Alvarez, Rice University, Houston, TX

Ethanol Fate and Transport:  An Updated Review of its Impact on Subsurface Gasoline Contamination

Brooke Redding, Exponent Inc., 4875 Pearl East Circle, Suite 201, Boulder, CO 80301, Tel: 303-544-2005, Fax: 303-544-2099, E-mail: bredding@exponent.com
James M. Davidson, Exponent Inc., 4875 Pearl East Circle, Suite 201, Boulder, CO 80301, Tel: 303-544-2009, Fax: 303-544-2099, E-mail: jdavidson@exponent.com

The potential environmental effects of blending ethanol with gasoline have been studied extensively in recent years, and a June 2006 study concludes that our knowledge base on the topic has improved substantially since 2001.  However, uncertainty remains in predicting the subsurface behavior of gasohol, particularly as it relates to formation of contaminant plumes in groundwater.  This presentation assesses the current state of knowledge and directions for future research.  Most studies agree that the issue of co-solvency inducing elevated levels of benzene in groundwater is real for neat (pure) ethanol spills, but not for releases of gasoline containing ethanol at 10% (E10) or lower.  Yet to be answered, however, is the question of whether the ethanol in blended gasoline will enhance the transport of benzene in the subsurface.  If so, a release of ethanol-blended gasoline might produce a larger subsurface benzene plume than a release of ethanol-free gasoline.  Uncertainty regarding ethanol transport to the groundwater clouds the issue, and the body of field data from real ethanol-gasoline spill sites is still inadequate to refute or validate the models and theories.  As such, until much more field data and supporting knowledge are gathered, uncertainty remains about the extent of benzene-plume lengthening at gasohol spill sites.

The Fate of Oxygenates and BTEX from Gasolines Containing MTBE, TBA and Ethanol: Initial Findings from a Controlled Field Experiment, Borden, Canada

Erika Williams, Dept. of Earth Sciences, University of Waterloo, Waterloo, Ontario, CANADA N2L 3G1, Phone: 519 888 4567, ext. 7090, Email: ewilliams@alumni.uwaterloo.ca
Marian Mocanu, Dept. of Earth Sciences, University of Waterloo, Waterloo, Ontario, CANADA N2L 3G1, Tel: 519 888 4567, ext. 7278, Email: mtmocanu@scimail.uwaterloo.ca
Jim Barker, Dept. of Earth Sciences, University of Waterloo, Waterloo, Ontario, CANADA N2L 3G1, Phone: 519 888 4567, ext. 2103, Email: jfbarker@sciborg.uwaterloo.ca
John Molson, Dept. of Earth Sciences, University of Waterloo, Waterloo, Ontario, CANADA N2L 3G1, Email: molson@uwaterloo.ca
José Luiz Gomes Zoby, Instituto de Geociências, University of São Paulo, São Paulo, SP 05508-080, Brazil, Email: jlgzoby@hotmail.com

The development and fate of groundwater plumes derived from three emplaced oxygenate-gasoline residuals in the Borden Research Aquifer are being monitored with multilevel monitoring fences. One fuel contains 90% gasoline with 9.8% MTBE and 0.2% TBA (GMT); another contains 90% gasoline and 10% ethanol (E10); the third contains 95% ethanol and 5% gasoline (E95). Ethanol was found 15 m downgradient of the sources and after 250 days at concentrations of 300 mg/L (E10) and 3000 mg/L (E95). While ethanol from E10 appears to have undergone considerable biodegradation, ethanol from E95 underwent relatively less. Surprisingly, MTBE and TBA mass fluxes declined precipitously and very little MTBE or TBA mass flux was found downgradient. However, previous experience at Borden and the lack of stable C isotopes fractionation suggests this apparent decline in MTBE mass flux may represent uncertainties in monitoring short slugs (<100 days) with only a few mass flux measurements. Considerable bioattenuation of BTEX, trimethylbenzenes and naphthalene is noted and is consistent with past Borden experience. Ethanol appears to enhance the mass flux of benzene and toluene, but not trimethylbenzenes or naphthalene, as compared to the GMT case.  

The Impact of Fuel Ethanol on Groundwater: Source Behavior

Brent P. Stafford, Dept. Civil & Environmental Engineering, University of Houston, Houston, TX 77204-4003, Phone: 713-743-4267, Fax: 713-743-4260
Natalie L. Cápiro,  Dept. Civil & Environmental Engineering, Rice University, Houston, TX 77251, Phone: 713-348-3798, Fax: 713-348-5239, E-mail: ncapiro@rice.edu
William G. Rixey, Dept. Civil & Environmental Engineering, University of Houston, Houston, TX 77204-4003, Phone: 713-743-4279, Fax: 713-743-4260, E-mail: wrixey@uh.edu 
Pedro J.J. Alvarez, Dept. Civil & Environmental Engineering, Rice University, Houston, TX 77251, Phone: 713-348-5903, Fax: 713-348-2411, E-mail: alvarez@rice.edu

In addition to concerns over the impact of gasohol on groundwater quality, there is concern over the potential impact of spills of fuel grade ethanol (E95 and E85).  Two important spill scenarios are fuel grade ethanol spilled into the subsurface onto existing NAPL contamination and fuel grade ethanol spilled into the subsurface without existing contamination.  This research assesses the extent of NAPL mobilization, phase separation and groundwater impacts for these two spill scenarios in both 2D bench-scale and pilot-scale (8-m³ tank) systems.  Complementary experiments (presented in a companion paper) are investigating the impact of these same two spill scenarios on microbial processes.  Results from the pilot-scale E95 release (supported by 2D bench-scale visualization studies) indicate that significant migration of ethanol, BTX and other hydrocarbons occurred within the capillary fringe downstream of the source, and NAPL was redistributed in the capillary fringe.  Low concentrations of hydrocarbons and ethanol were observed below the water table at several positions downgradient of the injection point.  Higher concentrations of ethanol were found at the water table ([EtOH] _max = 0.5% v/v) and at the tank outlet ([EtOH] _max =1% v/v), suggesting that for the flow rate and media used in this study, downgradient migration of ethanol occurred above the water table.  Ethanol was undetected in the outlet after 30 days.  Analysis of vadose zone soil cores taken on day 60 yielded BTX and other hydrocarbons in soils downgradient of the injection point.  For the E100 release onto pre-existing NAPL, ethanol concentrations were greatest in the capillary fringe ([EtOH] _max=20% v/v), with lower levels measured in the outlet ([EtOH] _max=0.6%) and in groundwater samples ([EtOH] _max=0.2% v/v).  Ethanol was undetected in the outlet after ~90 days.  The corresponding 2D bench-scale visualization experiments showed that ethanol flushed pre-existing residual NAPL emplaced above the water table, then redistributed the NAPL downgradient from the source following phase separation from the ethanol as it migrated through the capillary zone.  These results suggest that ethanol spills (at least for the specific spill volumes considered in this research) can be expected to migrate primarily in the capillary zone and that physico-chemical effects will have a significant impact (in conjunction with microbial processes) on the observed concentrations of ethanol and hydrocarbons in groundwater at various points downstream of fuel ethanol spills.  In addition, these results have implications for the longevity of the ethanol source as well as the distribution of NAPL at ethanol spill sites.    

The Impact of Fuel Ethanol on Groundwater: Microbial Response

Natalie L. Cápiro, Marcio L. B. Da Silva, and Pedro J.J. Alvarez, Rice University, Houston, TX Brent P. Stafford and William G. Rixey, University of Houston, TX

The use of ethanol as a gasoline additive is rapidly increasing due to the phase out of methyl tertiary-buytl ether (MTBE), which despite helping to meet Clean Air Act requirements, has caused widespread groundwater contamination problems. Ethanol is also being widely used to meet renewable fuel requirements.  This increases the probability of groundwater contamination by ethanol-amended gasoline, which requires an improved understanding of how ethanol affects microbial communities associated with the biodegradation of hydrocarbons in the subsurface.

In this study, a 8-m³ pilot-scale aquifer tank packed with sandy soil was used to simulate two spill scenarios that are likely to occur in the field:  1) fuel-grade ethanol (E95, ethanol 95% v/v containing gasoline 5% v/v as a denaturant) into soil containing no existing contamination, and 2) neat ethanol (100% v/v) release onto gasoline-contaminated soil. Real-time quantitative polymerase chain reaction (RQT-PCR) and most probable number PCR were used to estimate the concentration of Bacteria (16s rRNA), Archaea (e.g., methanogens), nitrate-reducers (nir), sulfate-reducers (dsr), and iron-reducers (Geobacter). Additionally, bacteria-harboring benzylsuccinate synthase (bssA), phenol hydroxylase (phe) or toluene dioxygenase (tod) genes were also quantified as biomarkers for potential BTEX biodegradation, since these genes are ubiquitous in bacteria capable of degrading aromatic hydrocarbons under anaerobic and aerobic conditions, respectively.

The fuel-grade ethanol spill at the water table interface was unique, the high reaeration rate resulting from the shallow vadose zone (0.5 m) and high inflow rate (2 m day^-1) maintained aerobic conditions throughout experiment, inhibiting the growth of any anaerobic species. Bacteria concentration in the soil prior to E95 injection was 1.9 ± 1.3 ´10³ cells g-soil ^-1. Within 6 months after the E95 pulse release (75-L), the Bacteria concentration increased significantly (p≤ 0.05) to 6.8 ± 5.1 × 10⁴ cells g-soil^-1, while the population of bacteria harboring tod genes also increased (0.13±0.05 to 1.8 ±1.8 × 10³ cells g-soil^-1).  Even though bacteria-harboring tod increased by one order of magnitude, ethanol did not select for specific BTEX degraders (decreased tod/Bac ratio). These preliminary results suggest that ethanol stimulated the growth of total Bacteria and thus the fortuitous growth of putative aerobic BTEX degraders (tod), suggesting that higher BTEX degradation rates could occur in the presence of ethanol provided that oxygen is not limiting.

The spill of neat ethanol onto residual BTEX mimicked potential spills that could occur at bulk (blending) terminals. In this experiment, higher ambient temperatures (summer) associated with lower influent D.O. saturation and higher microbial activity, coupled with a slower flow rate (0.7 m d^-1), resulted in fast depletion of oxygen from the system. The high oxygen demand exerted by ethanol consumption lead to strongly anaerobic conditions (i.e., methanogenic) 20 days after ethanol addition. After ethanol had been washed out (50%), the system returned to aerobic conditions (rebound time of 60 days), which favored the growth of BTEX degraders (tod) by over two orders of magnitude.  However, the short acclimation time limited the growth of other anaerobic communities, which was corroborated by the absence of their corresponding geochemical footprints. Microbial growth coincided with ethanol migration and availability, which was restricted to a relatively thin layer at the capillary fringe and water table interface, and toxic effects from ethanol exceeding 10,000 mg/L were not detected. These results suggest that releases of highly concentrated ethanol into the subsurface may not be as detrimental to microbial communities as previously believed because of short rebound times and minimal genotype dilution.

Top