Introduction

This article summarizes ENSR International’s (ENSR’s) experience remediating methyl tert butyl ether (MTBE) and other gasoline constituents in soil and groundwater at retail gas station sites.  ENSR has extensive experience in this arena in over 22 states and several foreign countries as a result of numerous longstanding contracts with oil companies and convenience store owners.

There is the impression that remediating MTBE requires specialized technologies and strategies that go beyond, and cost more than, those required for other gasoline constituents.  This is generally contrary to ENSR’s experience, where we routinely remediate MTBE and other gasoline constituents using long-proven conventional technologies.  We approach a gasoline release site with the same toolbox, whether or not MTBE is involved.

This article first briefly describes the technologies in this toolbox and then summarizes our experience applying these technologies at actual field sites. Cost information is summarized, and lessons learned are shared.  Finally, some thoughts as to the future of gas station remediation are presented.

Remedial Technologies

MTBE and other gasoline constituents are relatively mobile in the environment.  Therefore, prompt release detection and source control are essential.  MTBE is highly soluble, but has a low Henry's Law constant, allowing MTBE to dissolve readily in water and then remain in the aqueous phase.  It also has a high vapor pressure, volatilizing readily from gasoline to air.  It has a low tendency to adsorb (or "stick") to soil particles, therefore its movement in the subsurface is less retarded than that of many other gasoline constituents.  Like other gasoline constituents, MTBE has a specific gravity less than one and a vapor density greater than one; therefore, in the product (gasoline) phase it floats on water, and in the vapor phase it tends to sink, accumulating in basements, utility trenches, and the like.

These same physical properties that can make MTBE mobile in the subsurface also make it more easily recovered from the subsurface compared to other gasoline constituents.  Technologies such as groundwater extraction, soil vapor extraction (SVE), and thermal desorption work exceptionally well with MTBE. 

Site remediation is generally carried out in phases.  Minimizing total project cost entails optimizing the level of effort of each phase.  Receptor protection and source control are the first tasks at hand, followed by remediation of residual product and dissolved contamination.  Monitored natural attenuation may be used to complete the process.

Treating MTBE, once it is recovered from the subsurface, can be accomplished with a number of proven technologies.  MTBE's low tendency to adsorb and high Henry's Law constant make it more difficult to treat using granular activated carbon (GAC) adsorption or air stripping, although these technologies are still applicable under the right circumstances and widely used by ENSR at gas station sites.  Use of GAC is limited to situations with relatively low MTBE concentrations.  Air stripping will require higher air to water ratios than for other gasoline constituents.  Other methods, such as thermal or catalytic oxidation, are equally effective for MTBE and other gasoline constituents.

Bioremediation of MTBE is a widely accepted technology.  Like chlorinated solvents, MTBE was initially thought to be recalcitrant to biological treatment; however, research and field data continue to demonstrate that MTBE is biodegradable under a variety of conditions, both in situ and ex situ.  Bioremediation technology is evolving rapidly, and we anticipate that bioremediation will become more commonly used to treat gasoline releases containing MTBE.

ENSR’s Experience Remediating MTBE

ENSR’s project experience includes over 1,000 gas station sites, with scopes of work ranging from preliminary site assessment through full remediation and closure.  ENSR has remediated many sites where gasoline containing MTBE had been released. The majority of these sites have received official regulatory closure; nine are anticipated to be closed in the near future after several additional rounds of post-closure monitoring or other relatively minor activity take place.

Remediation typically involved some combination of source control, such as soil excavation and separate-phase product removal (31 sites), and remediation of residual product and dissolved contamination in the subsurface by groundwater pump and treat (28 sites),  SVE (24 sites) and/or air sparging (5 sites).  (It should be noted that more of the sites may have had source removal activities prior to ENSR’s starting work on the sites.)  Monitored natural attenuation was used as a final remediation step at 9 sites and was used alone to close 6 sites.

Excavated soil from 19 of 29 sites was recycled at an asphalt batching plant; the presence of clay and other fine-textured materials was often the reason soil from the other 10 sites was not sent to an asphalt batching plant.  Recovered water and vapor was treated most often by GAC adsorption (29 sites).  Air stripping was used to treat recovered water at 5 sites, in conjunction with GAC adsorption at 3 sites and alone at 2 sites.  Catalytic oxidation was used at 5 sites to treat recovered vapors (actually, it was thermal oxidation that was used at one of these sites).

Treatment technology selection for recovered groundwater and vapors is site-specific and is based on flow rate, initial VOC concentrations, required final VOC concentrations, and other factors. For recovered water containing more than approximately 10,000 micrograms-per-liter (ug/l) of total VOCs, ENSR would consider employing air stripping followed by catalytic oxidation of the stripped vapor-phase VOCs.  VOC concentrations this elevated contribute enough BTUs to the catalytic oxidation process to preclude the need for large amounts of fuel, making this technology economical.  In the 1,000 to 10,000 ug/l range of total VOCs, air stripping followed by GAC adsorption is often the best approach.  For VOC concentrations less than about 1,000 ug/l, GAC adsorption and monitored natural attenuation are appropriate.  For vapor-phase treatment, catalytic oxidation is often favored for total VOC concentrations greater than 75 parts per million (ppm) and GAC adsorption is often applied for lower concentrations.  In cases where MTBE is the major constituent being treated, all of the above numbers can be reduced by roughly 25 percent.

ENSR often changes technologies for the treatment of residuals as concentrations decrease with time.  It is very common to start with catalytic oxidation and switch to GAC adsorption once concentrations become low enough to make GAC economical.  At all five sites where catalytic oxidation was used for vapor treatment, we later switched to GAC adsorption.

ENSR has recently been using more biologically based technologies such as in situ oxygen addition by means of solid- or liquid-phase oxygen-containing compounds (5 sites).  This has worked very well at some sites, but not at others.  Our sense is that high concentrations of non-target compounds (e.g., iron or total organic carbon), or incomplete source control, or low soil permeability account for cases where oxygen amendment has not been successful.

Remediation Costs

Remediation cost is highly site-specific and largely a function of the period of time the release has gone unremediated, the size of the release, hydrogeologic conditions, and the presence of nearby sensitive receptors and pathways to them.  At some sites, MTBE drives the remediation due to high concentrations or mobility; at others, benzene drives the remediation due to benzene's higher toxicity (and thus more stringent required cleanup concentrations). 

Generally, releases that have impacted soil but not groundwater cost on the order of $100,000 to remediate (on average).  Releases that have impacted both soil and groundwater may cost on the order of $250,000 to remediate. The most costly gas station remediation to our knowledge is upwards of $4 million (and incidentally, MTBE is not a primary constituent at this site).  Obviously, there will be a wide range of variability.  The impact of the presence of MTBE on remediation cost is likewise site-specific.  In many cases, there is little or no impact.  At sites where the MTBE plume is larger than the BTEX plume, remediation costs can be higher than they otherwise might have been had MTBE not been a factor.  ENSR has found that such sites are often in areas of relatively shallow bedrock or where soil permeability is low and the silt content is high.

Future Trends

There are many other technologies ENSR considers for remediating gasoline releases containing MTBE; however, based on the positive results with physical treatment technologies such as groundwater pump and treat, SVE, and air sparging, we will continue to utilize these tried and true methods.  ENSR has had success using in situ oxygen amendment, and anticipates using this and other biological methods more extensively in the future. In addition, ENSR is increasingly using monitored natural attenuation as a final polishing step.

ENSR recently started using bioaugmentation at three sites with MTBE.  It is too soon to tell whether this approach is working at two of the sites.  At the third site, a gas station in Connecticut, BTEX and MTBE concentrations had remained fairly stable for 7 years.  ENSR then began injections of cultured aerobic microorganisms in late 2000 and saw BTEX concentrations fall from 10,000 to 12,000 ug/l to less than 100 ug/l, and MTBE concentrations fall from 1,000 to 2,000 ug/l to 500 ug/l, over a two-month period. 

The main take-home lesson from the totality of ENSR’s gas station remediation experience is the importance of release prevention, early detection, and prompt source identification and control in minimizing total costs.  We believe that gas station owners will increasingly realize it is in their interests to focus more on these activities so that less remediation effort and cost are required.  Historically, releases have been detected long after their occurrence; by then, plumes have grown, often extending beyond the property boundary. In conjunction with the smaller plumes of gasoline constituents that result from early detection and prompt control, in situ bioremediation may come to be more widely used at gas station sites.

Conclusion

ENSR has been remediating releases of gasoline containing MTBE on a routine basis in a wide variety of geological settings.  We have found that the same technologies that are effective for treating other gasoline constituents are effective for treating MTBE. Addressing MTBE does not need to be overly complicated.  It just takes an understanding of specific site conditions, appropriate selection and sequencing of remedial technologies, good engineering, adequate monitoring, and flexible operation and maintenance.  The real key to minimizing costs is release prevention, early detection, and effective source control.

Top