Stanford News


CONTACT: Stanford University News Service (650) 723-2558

Project shows native bacteria can destroy widespread pollutant in groundwater

Native bacteria can be used to destroy one of the most tenacious and widespread contaminants that is poisoning the nation's groundwater, Stanford researchers have shown.

The pollutant is trichloroethylene (TCE). In the 1980s, Americans used 150 million tons of TCE annually to dryclean their clothes and to degrease aircraft and automobile engines. Now it is one of the most widespread and difficult to treat groundwater contaminants in the country.

The researchers ­ working at Edwards Air Force Base in Southern California ­ have completed the first full-scale test to determine if TCE can be removed from groundwater using a method called in situ bioremediation, which uses naturally occurring microorganisms to break down toxic chemicals.

"The test has shown that this method can successfully reduce the levels of TCE in groundwater by 96 to 98 percent," said Perry L. McCarty, the Silas H. Palmer Professor of Civil Engineering who heads the project, report in a at the American Chemical Society meeting in San Francisco this week. Working with him on the project are: Mark Goltz, associate professor at the Air Force Institute of Technology, graduate student Jason Allan, post-doctoral student Mark Dolan and Gary Hopkins, a science and engineering associate.

In addition to performing effectively, the bioremediation process lived up fully to the researchers' predictions, indicating that they understand the basic processes involved well enough to determine how the method will work in other areas contaminated with TCE, McCarty said.

The conventional approach to groundwater decontamination is called pump-and-treat. Groundwater is pumped to the surface, where air is blown through it to strip out volatile contaminants, which are then adsorbed onto activated charcoal filters. This approach wastes water, because the treated water is seldom reinjected into the aquifer. The contaminated charcoal also must be disposed. Energy costs can be high for pumping contaminated water from deep aquifers. The clean water that replaces the water pumped to the surface can become contaminated when it comes into contact with the tainted sands that remain.

The basic idea behind in situ bioremediation is simple: If you provide naturally occurring bacteria in groundwater with enough air and the right type of food, then the microorganisms may respond by digesting the toxic chemicals as well. Then, when the operation is shut down, the bacterial populations will quickly return to normal.

According to McCarty, the main advantages of bioremediation are water conservation and a substantial reduction in pumping costs. On the other side of the ledger, however, is the cost of the chemicals involved. In the case of TCE decontamination, the highest cost item in the process is hydrogen peroxide, which breaks down in the water to provide the oxygen that the bacteria need to attack the compound.

Although in situ bioremediation is simple in principle, it is considerably more difficult to put into practice. The Edwards project is the culmination of more than 10 years of basic laboratory and experimental field work, most of which was performed under the auspices of the US Environmental Protection Agency's Western Region Hazardous Substance Research Center, a consortium between Stanford and Oregon State universities. It reflects the efforts of 23 faculty and staff members from Stanford, Oregon State, Michigan State University, the University of Western Florida and the University of Minnesota, together with 30 graduate students.

In a 1985 pilot project conducted at Moffett Field in Mountain View, Calif., headed by professor of civil engineering Paul Roberts, Stanford researchers determined that feeding native bacteria with the right mixture of oxygen and toluene, a naturally occurring petroleum product, causes them to produce the compound toluene monoxygenase that initiates the breakdown of TCE into carbon dioxide, water and chloride.

Despite the success of the pilot study, there were questions that could only be answered by a full-scale test. There was concern, for example, that the bacterial growth might become thick enough around the injection wells that they would clog up the pores in the aquifer. Several experts also had suggested that the effectiveness of the bacteria at degrading the toluene might lessen over time.

So McCarty and his colleagues began looking for a site for a full-scale test. One of the biggest barriers they faced was the fact that they wanted to use toluene, which is itself a regulated compound. Unlike TCE, toluene is not carcinogenic, but drinking water standards specify a limit of one part per million. But at a fraction of that level ­ about 25 parts per billion ­ it gives water an unpleasant taste and smell.

"Using toluene is a little like adding chlorine to drinking water to kill bacteria. Chlorine is poisonous at a high enough concentration so you don't like to add it except when there is a significant benefit. We've tried a number of other compounds, but none of them is as effective as toluene," McCarty said.

The Edwards site was nearly ideal for the project. The groundwater was close to the surface. TCE is the sole contaminant, and its source was known. The base staff were interested and supportive, as were the local state and federal officials, McCarty said.

After the site was selected, the researchers gathered aquifer sands containing native bacteria and tested them in the laboratory to determine how the bacteria would respond to the oxygen/toluene diet. In 1993, Stanford master's students designed the purification system as a class project. The site contained two contaminated aquifers at different depths. The design uses two treatment wells. One well was designed to create a bacterial-rich zone in the lower aquifer and to pump water from the upper aquifer to the lower one. The second well was designed to create a bacterial-rich zone in the upper aquifer and to pump water from the lower aquifer to the upper one.

The wells were placed in a plume of TCE-contaminated groundwater. As the contaminated water reached the well site, it was sucked into the circular flow set up by the treatment wells, which pulled it through the two bacterial zones. Each pass through one of these bacterial zones removed between 83 percent and 86 percent of the TCE, according to the monitoring data. The net result was to reduce the concentration of TCE in the water passing through the area by 96 to 98 percent, from 1.2 parts per million down to 20 to 50 parts per billion.

Running the process for a year cleared up the last major operational questions, McCarty said. It confirmed the observation, made at the Moffett pilot project, that the high concentrations of hydrogen peroxide near the wells kept down bacterial growth and prevented clogging. Also, the bacterial action showed no evidence of degrading over time. The researchers had promised to keep toluene levels below 20 parts per billion at the experiment's boundaries: Actual levels averaged one part per billion.

"We've demonstrated the technology. Now we will work toward applying it at other TCE-contaminated sites and with other contaminants, like vinyl chloride," McCarty said.

The full-scale demonstration project was financially supported by the Armstrong Laboratory Environics Directorate at Tyndall Air Force Base and the Environmental Quality Division of the US Air Force.


By David F. Salisbury