MI: Testimony Before the California Energy Commision

Letter from Malcom Pirnie, Inc. on Behalf of MI to California State Water Resources Control Board Regarding Ethanol Fate and Transport

Re: Comments on Ethanol Fate and Transport

Dear Mr. Giannopoulos,

I recently attended the public workshop in Sacramento regarding the fate and transport of ethanol in the environment. As part of the public comment process, we are providing written comments to you on issues that we recommend should be further addressed in the study being conducted by the Lawrence Livermore National Laboratory (LLNL).

Malcolm Pirnie, Inc has been retained by the Methanol Institute to prepare these comments on the LLNL analysis of potential impacts of the use of ethanol in gasoline on the fate and transport in groundwater of other constituents in gasoline, namely the aromatic compounds, benzene, toluene, ethylbenzene and the xylenes (BETX). We believe that these comments will help to clarify and expand the LLNL conclusions presented at the Sacramento public workshop regarding impacts on the BETX plumes due to the use of ethanol-blended gasoline.

Using a screening-level model, LLNL concluded that the use of ethanol-blended gasoline could extend BTEX plumes by approximately 25%. These results presented by LLNL are consistent with two other recently completed modeling efforts to assess this issue. As referenced in the LLNL presentation, Malcolm Pirnie completed a detailed analysis of ethanol fate and transport in the environment last year entitled, “Evaluation of the Fate and Transport of Ethanol in the Environment.” Included in that analysis was a preliminary modeling evaluation of the effect of 10% ethanol in gasoline (gasohol) on the fate of BETX plumes.

The Malcolm Pirnie report concluded that a primary disadvantage of adding ethanol to gasoline is the potential impact of ethanol biodegradation on the natural biodegradation of other gasoline constituents present in the groundwater. Ethanol is known to readily biodegrade under a variety of aerobic and anaerobic conditions. Under these conditions, ethanol is a preferred substrate and will be preferentially utilized in the presence of BTEX. However, as ethanol is aerobically biodegraded, oxygen and other electron acceptors, as well as nutrients will become depleted in the groundwater. As a result, BTEX plumes may be lengthened due to the delay in biodegradation in the presence of ethanol. Based on modeling results, the presence of ethanol is expected to increase BTEX plume lengths by approximately 27% under typical California groundwater conditions (ranging from 16% to 34% increase in BTEX plume lengths). The potential impact of increasing BTEX plume lengths is either a greater probability that drinking water well fields could be impacted by BETX or higher BTEX concentrations at wells that are already contaminated. Additional migration of the BETX plumes could also cause greater property damage due to plumes extending beyond the boundaries of the source property. These impacts would result in higher cleanup costs for BTEX plumes, if cleanup is warranted.

Finally, the University of Waterloo, Ontario, Canada has recently completed a modeling effort to evaluate the use of ethanol-blended gasoline on gasoline in groundwater. Preliminary results from this research have been presented at the National Ground Water Association meeting in Houston, Texas, 1999 at the Petroleum Hydrocarbon Conference. This study concludes that the use of ethanol in gasoline could extended BTEX plumes 24 to 33%. Thus, three independent assessments using different groundwater modeling approaches have reached similar conclusions regarding the impact of ethanol on BTEX plumes, namely, that BTEX plumes may be extended 24 to 33%.

A second scenario not evaluated in the Malcolm Pirnie report or the LLNL analysis, but of importance for fully understanding the impact on groundwater of the use of ethanol, is the expected BTEX plume elongation resulting from a release of pure ethanol onto an existing BTEX plume with some residual gasoline containing BETX in the soil or groundwater. The University of Waterloo has also presented preliminary results from this research at the National Ground Water Association meeting in Houston, Texas, 1999 at the Petroleum Hydrocarbon Conference. The Waterloo model evaluated the effects of increased benzene dissolution, rapid depletion of electron acceptors due to the biodegradation of ethanol, and more rapid ethanol biodegradation rates. The scenario involved a release of pure ethanol onto a 10-year old BTEX release. The results indicate that BTEX plumes could be elongated from 55% to 142%, relative to non-ethanol conditions. The highest BTEX elongation occurred in soil with low organic carbon content, which results in limited retardation of the benzene in the groundwater. Results from Waterloo suggest that BTEX plume elongation will increase as the contact time between the ethanol and BTEX increases, i.e., the longer ethanol remains in contact with the BTEX plume, the more the BTEX plume will elongate.

The results from these evaluations of ethanol’s impact on BETX plumes pose two questions. First, are these predicted BTEX plume extensions significant? Second, are there limitations with these modeling efforts that may underestimate the actual impacts under field conditions? The following presents a list of three factors that have not been addressed in any of the modeling analyses, which could result in BTEX plumes extending beyond what has been predicted in these three modeling studies.

Due to the complexity of modeling real systems, all of the models have ignored subsurface heterogeneities, which may prove to be their most significant limitation. Subsurface heterogeneities can result in preferential groundwater pathways, where the impact of ethanol on BTEX compounds is unknown. Similarly, the fate of BTEX compounds in fractured bedrock is likely to change when exposed to ethanol. Under both of these conditions, i.e., preferential pathways and fractured media, groundwater velocities are high and the fraction of organic carbon is low. Under these conditions, the Waterloo results suggest that BTEX plumes could be extended up to 142 % of the plume length without ethanol compared to the 30-50% predicted by models that ignore preferential pathways.
A second factor that could result in further elongation of BTEX plumes occurs when multiple discrete releases of gasohol occur over several years. Under these conditions, ethanol will be released in pulses to the subsurface over a period of several years and thus, will remain in contact with the BTEX over several years. As the Waterloo results suggest, a long contact time between ethanol and BTEX could lead to increased BTEX plume elongation.
Due to the lack of field data to verify modeling, dissolution kinetics of gasoline in groundwater have been estimated from controlled lab, modeling, and limited field experiments. Based on an evaluation of the properties of ethanol, ethanol dissolution should occur very quickly under ideal mixing conditions; however, if dissolution occurs slowly, ethanol will remain in contact with BTEX longer and BTEX plumes may experience greater elongation. A similar issue has been addressed with respect to MTBE. One would expect MTBE to dissolve from the source area rapidly, resulting in detached MTBE plumes. However, in the field, MTBE plumes remain attached to the source for extended periods. Recently, Dr. Bill Rixey at the University of Houston estimated that over 100 pore volumes could be required for complete MTBE dissolution in a heterogeneous source area with minimal groundwater/NAPL interfacial contact. Ethanol is expected to behave similarly to MTBE in heterogeneous environments, and thus, ethanol dissolution may occur over a much longer period than theoretically predicted.

Each of these factors could result in BTEX plumes that are extended beyond their non-ethanol maximum length. The relevant question is whether these elongated BTEX plumes are more likely to impact drinking water wells. We suggest that this issue should be addressed quantitatively by the LLNL study.

In conclusion, there are significant unknowns regarding the real impact of ethanol on BTEX plumes; however, it is generally acknowledged that the use of ethanol in gasoline will extend BTEX plumes. Three independent modeling assessments have consistently concluded that the use of ethanol-blended gasoline will extend BTEX plumes 24 to 33% on average, relative to gasoline without ethanol under presumed homogeneous aquifer conditions. In addition, modeling results at the University of Waterloo suggest that a pure release of ethanol on an existing BTEX plume could extend the BTEX plume up to 142%, relative to non-ethanol conditions. As noted, however, it is likely that release and subsurface factors exist where BTEX plumes could be extended even further. Thus, the use of ethanol in gasoline and the increased transport of pure ethanol are expected to increase the probability of detecting benzene in drinking water wells as well as exacerbating property impact issues. We therefore recommend that LLNL carefully consider the potential impacts of these findings on costs of soil and groundwater cleanup in California, and on the potential for impacts of drinking water wells. The effects of heterogeneous aquifer conditions on BETX plume lengths in the presence of ethanol should be further evaluated. Finally, the potential impacts on groundwater quality of denaturants that must be added to ethanol should be considered to provide a more comprehensive assessment of the relative merits of ethanol in gasoline compared to other options.

If you have any questions or would like to discuss these comments further, please contact either me or my associate, Andrew Stocking. We can be reached at (510) 451-8900.

Sincerely,

MALCOLM PIRNIE, INC

Michael C. Kavanaugh, Ph.D., P.E. Andrew Stocking, P.E.
Vice President Project Engineer

Top | Home | Reformulated Gasolines | Search | E-Mail