Using Microorganisms to Clean-up Bedrock Contamination

UNH's Bedrock Bioremediation Center
The U.S. Environmental Protection Agency (USEPA) has stated that annually, billions of dollars from federal, state and local governments and private industry will be spent over the next several decades to remediate hazardous waste and petroleum contaminated sites. At most of these sites, both the subsurface soil and groundwater have been affected. Of the 217,000 sites requiring clean-up in the United States, 70-80% need remediation of groundwater. While there are several technologies for remediating groundwater flowing through soil, there has been a much more limited effort made to clean-up contaminated bedrock sites, even though bedrock aquifers are used as drinking water supplies by a significant number of people. In New Hampshire, for example, 60% of the drinking water supplies rely upon groundwater and of these 97% get their water from bedrock aquifers. Bedrock aquifers serve about 25% of the state's population, and are heavily relied upon as domestic and small public drinking water sources throughout much of the United States.

Remediation of bedrock contamination is often deferred because it is very difficult to address. For example, in New Hampshire, there are approximately 400 hazardous waste sites, 3,000 petroleum contamination sites, and 197 unlined solid waste landfills, some of which are impacting, or have the potential to impact, bedrock aquifer. Of the 18 Federal Superfund sites located in New Hampshire, 15 have bedrock contamination. Again this pattern is typical in other areas of the country and is particularly true for chlorinated solvents, such as trichloroethylene (TCE), which are more dense than water and continue to sink when they enter the saturated zone as non-aqueous phase liquids (NAPLs).

Cleaning-up Bedrock Aquifers Using In Situ Bioremediation
Remediation of contaminated bedrock aquifers has been largely neglected because of the difficulties in characterizing the extent of contamination at these sites and the lack of technologies able to successfully clean-up the problems. Frequently, the contamination is deep within the rock, especially in the zone where there are few fractures (competent bedrock) and the hydraulics of the fractured media are complex. In addition, drilling is often prohibitively expensive and excavation is virtually impossible. One possible method for remediating these sites is in situ bioremediation, which has generally been recognized as offering a low-cost and long-term solution to clean-up of organically contaminated soils and groundwater. Bioremediation uses naturally occurring microorganisms to degrade hazardous substances into less toxic or non-toxic, inert compounds. The microbes ingest and degrade organic substances, often using them as their food and energy source.

Until recently, no one knew that microbes were living in bedrock aquifers. Their existence has been confirmed by scientists studying the potential for subsurface disposal of wastes. While data have been collected in several countries, the most detailed studies are from the U.S, Sweden and Canada. These studies indicate that microorganisms exist in uncontaminated bedrock aquifers. They live along the fractures that permeate the rock, and in the water that flows through the fractures. Many of the microorganisms are similar to types found in other environments such as soils and surface water.

Most bedrock aquifers have little or no oxygen present because this gas cannot diffuse down into the rock in large quantities. Thus, it is not surprising that the microorganisms found in these aquifers are capable of living without oxygen (anaerobes). Furthermore, experiments with these microorganisms indicate that they are active and capable of degrading many different types of organics. However, under normal uncontaminated conditions, these microorganisms divide very slowly, perhaps only once every 10 to 100 years, because there is little food present to support their growth. The key issue in using these microorganisms to degrade organic contaminants in the bedrock aquifer is how to speed the process so it occurs within a reasonable time frame.

One of the biggest problems in using in situ microbial populations to remediate contaminated bedrock aquifers is the difficulty in predicting the groundwater flow pattern through the fractures. In addition, because substances such as nutrients must often be added to enhance the rate of biodegradation, fracture patterns and hydraulics must be understood to ensure that the injected stimulants reach the microbes.

Bedrock typically contains a network of fractures. The pattern of fractures is usually very heterogeneous. Some fractures will have higher hydraulic conductivity and these may have more active microbial populations capable of potentially higher rates of biodegradation. However, in highly conductive fractures, the groundwater may have such a short contact time with the microorganisms relative to the time required for biodegradation, that the contaminant may not be degraded completely.

The mineral composition of the bedrock along the fractures may also affect the conditions experienced by the microorganisms. Fractures may contain clay-like substances, iron oxide coatings, and carbonate deposits to which contaminants can adhere. These and other minerals may affect the groundwater chemistry in the fractures. The minerals may also supply the microbes with substances they need to degrade the contaminants.

Bedrock Bioremediation Center
The Bedrock Bioremediation Center (BBC), situated within the Environmental Research Group (ERG) at the University of New Hampshire (UNH), specializes in multi-disciplinary research on bioremediation of organically-contaminated aquifers. The Center is comprised of a consortium of faculty from ERG and the University's Departments of Microbiology, Earth Sciences and Natural Resources. Support for the BBC has been championed by Congressman John E. Sununu who believes that "no groundwater issue is more challenging than dealing with pollutants in deep bedrock formations". The BBC has received $1.6 million of funding from the USEPA's Robert S. Kerr Laboratory in Ada, Oklahoma to develop, test and evaluate innovative technologies for enhanced bioremediation of bedrock aquifers contaminated with organic chemicals.

The overall, long-term objectives of the BBC are to conduct research in order to:

1. More efficiently and economically characterize the direction of groundwater flow and fracture patterns in contaminated bedrock aquifers. Without an understanding of the fracture patterns, flow paths and contaminant distributions, it is very difficult to develop strategies for implementing bioremediation in situ.
2. Improve and develop new field technologies to characterize and control hydraulic flow conditions in the contaminant zone. Without controlling these parameters, the ability to enhance biodegradation in bedrock is limited.
3. Develop laboratory and field methods to estimate and accelerate in situ rates of bioremediation of organic contaminants in bedrock aquifers. These rates must be known to predict whether bioremediation will degrade the contaminants before they reach the nearest drinking water wells.
4. Develop and apply innovative microbiological and molecular biology techniques to identify and characterize the specific components of the microbial communities in the fractures of competent (deep) bedrock responsible for contaminant biodegradation, and enhance and assess the effectiveness of in situ bioremediation strategies. The BBC will transfer information regarding these technologies and the methods it gains during its research to federal (including military), state, and local regulatory agencies, and environmental consultants.

In order to be able to achieve its goals, the BBC is creating a field test site where new in situ technologies can be evaluated. One problem with evaluating the effectiveness of methods developed for use in bedrock is that it is difficult to track a parcel of water as it moves through an aquifer. To overcome this issue, the BBC's field facility will consist of sets of paired boreholes, each located in a region of the aquifer with different levels of contamination. Each set of paired boreholes will be connected by at least one common fracture (Figure 1). With this method, it will be possible to monitor the conditions as water moves through the fracture from the upgradient to the downgradient borehole. This monitoring will help determine the effectiveness of a new technology in reducing the concentration or toxicity of a contaminant. The paired boreholes will be located in an uncontaminated area of the aquifer (a control zone) and in zones of high and low contamination. This will enable testing of new methods under a variety of conditions to determine the full range of their effectiveness.

Each set of paired boreholes will be spaced no greater than 30 feet apart, which while close in terms of surface water flow, can represent a large distance and time in competent bedrock where the groundwater may move through the fractures at a rate of less than one foot per day. The creation of a set of paired boreholes begins with drilling a "preliminary" borehole in each zone. Once the preliminary wells are completed, a full suite of geophysical tests will be carried out in each borehole including:

• Electrical resistivity measurements that monitor water quality (which is often distinct in each fracture).
• Self-potential measurements that locate changes in the specific mineral content of the rock.
• Natural gamma radiation monitoring which detects the presence of radioactive minerals deposited on the fracture surfaces.
• Temperature logs that reveal zones of water coming from different fractures.
• Videologs and optical and acoustic televiewer logs that provide actual pictures of the fractures in the borehole walls.
• Borehole flowmeter measurements that monitor how well the fractures transmit water.
• Omni-directional borehole radar that pinpoints the location of specific fractures.
• Directional borehole radar that determines the orientation and location of the fractures for distances within a 30 foot radius of the borehole.

The geophysical data, especially information obtained from the directional borehole radar, will be used to locate the downgradient "test" boreholes that will be the second well of each pair. When drilling a test borehole, the borehole radar's antenna will be set in the "preliminary" well to obtain echoes off the advancing drill string in the "test" borehole, in order to directly monitor the approach of the drill to the target fracture that will link the two boreholes.

The BBC, along with USEPA Region I, the New Hampshire Department of Environmental Services (NHDES) and the New Hampshire/Vermont District of the U.S. Geological Survey (USGS), evaluated many potential field sites for this research. In order to rank these sites, six criteria were established that would be desirable for the BBC's research:

• Bedrock aquifer readily accessible (e.g., depth to bedrock not excessive);
• Good understanding of the site's hydrogeological and physical characteristics;
• Good understanding of the specific contaminants present and their distribution and concentrations in situ;
• Existing network of monitoring wells;
• Site readily accessible for several years without fear that remediation/litigation will impact the BBC's work;
• Good understanding of the type of bedrock present (competent vs. highly fractured or weathered bedrock), fracture characterization, sorption and permeability.

Ultimately, it was decided that Site 32 in Zone 3 at the Pease International Tradeport (formerly Pease Air Force Base) in Portsmouth, New Hampshire was best suited for the BBC's field site (Figure 2). The BBC is working at Site 32 under the auspices of the Federal Facilities Agreement (FFA) between the U.S. Air Force (USAF), Pease Development Authority (PDA), USEPA and NHDES. Site 32 is located near the center of Pease within the area known as the Industrial Shop/Parking Apron. Building 113, located at Site 32, was used for equipment maintenance, which included degreasing operations. A 1,200 gallon concrete tank buried beneath the northeastern corner of the building served as a reservoir for degreasing wastes, primarily TCE. The tank contained an overflow discharge pipe that released contaminants associated with the degreasing wastes directly into the soil (overburden). The TCE migrated downward into the groundwater. A contaminant plume containing primarily TCE and its degradation products cis- and trans-1,2-dichloroethylene (cis- and trans-1,2-DCE) and vinyl chloride (VC) now extends approximately 0.3 miles beyond the identified source area. The plume (Figure 2) has migrated with the groundwater into the weathered and competent bedrock. The USAF has implemented an extensive groundwater monitoring program that analyzes samples from wells located throughout the contaminant plume. The apparent loss of TCE over time and the presence of DCE and VC in discrete zones within the plume suggest that biodegradation is occurring naturally at Site 32.

The first year of the BBC project, starting in August 1999, commenced with a review of the existing data in USAF, USEPA and NHDES reports on Site 32 and adjacent sites. The reviews were done to gain as much insight as possible about the subsurface geology, hydrogeology, and contaminant distribution in the bedrock. Two other preliminary tasks included the development of a Quality Assurance Project Plan (QAPP), to insure that the project sampling and analysis would provide valid data, and a Health and Safety Plan (HASP. In addition, geophysical logs were made by the USGS of some of the existing wells at Site 32 owned by the USAF. These logs helped to better our understanding of the bedrock fracture pattern at the site. All of the existing bedrock wells at the site were videologged. Fracture zones in the existing bedrock wells were identified and subsequently sampled, using a packer system, for TCE, DCE and VC. The system consisted of two inflatable packers placed immediately above and below the fracture of interest. A pneumatic sampling pump, located between the inflated packers and adjacent to the fracture of interest, was used to evacuate the borehole interval and draw water at a slow rate from the isolated fracture zone. The data gathered from the first-year preliminary activities were used to select the locations of the "preliminary" boreholes for each of the sets of paired boreholes the BBC would drill.

BBC Bedrock Drilling
Logistics for the actual bedrock drilling and coring at the BBC site presented a series of challenges. The first challenge involved drilling fluids. Normally, drilling muds are used when drilling or coring bedrock. However, for the BBC project, use of muds would create unacceptable interferences in the microbiological analyses. Drilling muds have been found to be highly contaminated with microorganisms. Because the goal of the project was to obtain naturally occurring bedrock microorganisms, the drilling team took great pains to eliminate the possibility that other microbes, such as those in the drilling muds or overburden soils, would be present in the cores. This meant that the selection of the drilling fluid was limited to groundwater from an uncontaminated zone of the bedrock aquifer. It was thought that microorganisms in the uncontaminated drilling water would be distinct from those in the contaminant zone bedrock because they were not normally exposed to the contaminants and thus would not be producing substances (enzymes) to degrade them. For the same reason, it was not possible to re-circulate the drilling water, as re-circulation would bring microorganisms from other parts of the borehole in contact with the new core. The constant re-circulation of water within the borehole would cross-contaminate the borehole, and make the detection of distinct differences in the natural microbial populations and their biodegradation capabilities very difficult. Furthering the drilling challenge was the fact that the BBC researchers wanted to recover bedrock cores from depths of at least 100 feet into the competent rock. One of the goals was to extract microbes from the surface of each water-bearing fracture encountered. To do this, the surface area, and consequently the core, had to be as large as possible. Therefore, consecutive 4-inch diameter cores, 5-feet in length, were recovered from each borehole.

The coring equipment was designed and built by Christensen Products in Salt Lake City, Utah. The 5_ x 4 inch core barrel with lexan liner was used with a hard formation diamond- impregnated coring bit. The cores were collected in lexan liners, which had been sterilized in diluted bleach. Using the liners facilitated handling the cores, that were occasionally in several pieces, and minimized the chance of microbial contamination. The core barrel contained a special piston that was initially placed in the end of the inner barrel. Check valves on the top of the inner core barrel assembly and in the piston kept fluids from the drill stem out of the inner core barrel.

The original standard bit manufactured by Christensen was 4 inches in diameter. The rock cores produced by this bit, especially those from more fractured material, jammed inside the lexan liners and caused the liners to deform (Figure 3) making core recovery difficult. Christensen-DESI (Drilling Equipment Supply, Inc.) worked with NH/VT USGS drilling personnel to overcome this problem by making the inner diameter of the diamond bit slightly smaller (Figure 4), and increasing the inner diameter of the liner retaining ring.

All drilling equipment and instrumentation used to drill the bedrock boreholes had to be cleaned with pressurized steam before use to remove any organic contaminants. Then, the equipment was soaked for 30 minutes in diluted bleach (the time required to kill bacterial spores) and rinsed with uncontaminated bedrock drilling water. All pumps and hose lines also had to be decontaminated by soaking or pumping the bleach solution through them for 30 minutes.

In order to minimize microorganisms from the overburden and weathered bedrock from contaminating the rock cores, these zones had to be isolated from the competent bedrock. This was done using a telescoping casing. Initially, a 10-inch casing was advanced as the contractor drilled through the overburden (primarily sand, marine clay/silt, and glacial till) using a 4 _- inch diameter hollow stem auger with an outer diameter of 7 3/4-inch. Once the 10-inch casing came in contact with the top of the bedrock, the overburden material remaining in the borehole after augering was flushed out of the casing. Then, a 12-inch socket was drilled into the weathered bedrock. The entire hole was then filled with a cement grout. Most bentonite is highly contaminated with microorganisms, so it could not be used as a grout for this project. An 8-inch casing was installed through the grout to the bottom of the bedrock socket. The bottom of the 8-inch casing was sealed using a heavy plastic bag or sheeting to prevent grout from entering it. The casing was filled with water to sink it through the grout. The 10-inch casing was withdrawn as the grout was pumped into the hole. In this way, the grout was forced up into the annular space between the 8-inch casing and the outside of the borehole. The casing completion sealed off the overburden and the grout displaced or encapsulated contaminants that had been in the borehole fluids.

Once hardened, the grout inside the 8-inch casing was drilled out using a rotary wash technique with a tri-cone roller bit, leaving a 6-inch hole extending down to the weathered bedrock. The weathered bedrock was cored with the 5_ x 4-inch core barrel with the Christensen diamond impregnated bit. Continuous cores were taken down to competent bedrock (usually <15 feet at Site 32). Once it was established that competent bedrock had been reached, the borehole was reamed with an 8-inch diameter bit using rotary wash techniques.

A 6-inch casing was installed extending from the ground surface down through the overburden and weathered bedrock zones and at least 12 inches into the competent bedrock. Before installing this casing, the entire borehole was filled with cement grout. A plastic cap was placed on the end of the 6-inch casing prior to inserting it in the borehole, to keep the grout from flowing into it as it was lowered. The 6-inch casing was filled with drilling water to prevent it from floating. The grouted casing isolated the competent bedrock from the weathered bedrock. A roller bit was used to drill through the plastic cap on the end of the 6-inch casing. Once the plug had been drilled out, the core barrel was used to extract cores of the competent bedrock.

New Hampshire Boring of Londonderry, New Hampshire was the subcontractor to the BBC responsible for drilling all boreholes.

The drilling fluid requirement for 4-inch diameter rock cores in the granitic rock at Site 32 was approximately 3,000 gallons per day (gpd) of water from an uncontaminated area of the bedrock aquifer. Every three days, approximately 9,000 gallons of water was pumped from a bedrock supply well, located 2.5 miles from the site, into a 10,000-gallon pool water truck. The water was transported to Site 32 and transferred into three large plastic storage tanks. The drilling water also had to have low oxygen content, so as not to change the anaerobic conditions in the bedrock and shock the indigenous microbes. The water was therefore pumped on demand from the large tanks into two 500-gallon tanks where it was purged of oxygen by sparging with nitrogen gas.

The spent drilling water exited the borehole, and was channeled through a baffled tank where the heavy rock particles settled out. The water was then pumped into a large storage tank where further settling occured. Each day, the water was transferred from this storage tank into a tank truck. The truck transported the spent fluid to an existing wastewater treatment facility, where it was gradually fed into the existing waste stream. The USAF owns the treatment facility and treated the spent drilling water as part of its commitment to the BBC project. All borehole cuttings also had to be containerized, tested and disposed properly, based on their level of contamination.

Another challenge when retrieving the cores was to preserve, as best as possible, the in situ conditions of the deep bedrock. This required special handling procedures to minimize microbial contamination and prevent contact with oxygen. The top of the core was sealed off from drilling fluids in the core rod, and ultimately air at the surface, by a piston inside the liner. The piston was pushed to the top of the liner as the core moved into the barrel. The challenge of sealing the fluids in the core, and preventing air from contacting the bottom of the core was solved by using a lexan tank or "fishbowl". As the core was brought to the surface, it was pulled into a "fishbowl" that was attached to the top of the well casing (Figure 5). The fishbowl was a lexan glove-box chamber that was filled with drilling water. It was designed to maintain a column of water approximately 8 feet above the top of the borehole. The lower portion of the fishbowl consisted of a square chamber that had two long rubber gloves protruding into it. As the core was drawn up into the chamber, the person using the gloves clamped a sterile rubber cap on the bottom of the core barrel while it was still submerged. The cap also contained a metal piston that was later used to extrude the core from the barrel. This procedure helped to retain the pore fluids in the core and prevented contact with oxygen. Once the bottom endcap was secured, the core was pulled out of the top of the fishbowl and placed upright in a stabilizing ring. The top of the core barrel was quickly removed and a sterile rubber cap was fastened on the end. The core was not exposed to oxygen during this maneuver because the sterile piston inside the core barrel liner prevented air or microbial contamination from entering. The sealed barrel containing the core was placed in a "coffin" (Figure 6), the lid was closed, and the box was flooded with nitrogen to maintain oxygen-free conditions. The coffin was brought to the BBC field trailer where it was placed on a hydraulic table. The end of the coffin that contained the top of the core was slid into the entry port of a large glovebox chamber that contained no oxygen. The cap on the bottom of the core which contained the piston was then loosened and a 4-ton hydraulic extruding device was used to push the piston against the lexan core liner and the core itself. The liner and rock core were extruded into a sterilized PVC trough in the first oxygen-free glovebox chamber (Figure 7). Once the extrusion was complete, the port allowing access to the chamber was closed, and the coffin was removed. The lexan liner was split longitudinally using a dremel tool and the pieces of rock were transferred into a second, adjacent oxygen-free glovebox chamber for microbial processing.

Assessment of Microbial Contamination
The next phase of this project is to assess the amount of microbial contamination introduced into the bedrock cores during the drilling process. The first experiment will use a chemical tracer (lithium bromide). This will be introduced into the drilling water at a known concentration (~50 ppm above background) and will assess the overall level of drilling-associated contamination. Porewater recovered from the cores and distilled water rinsate from the fracture surfaces will be analyzed for bromide to estimate the extent of drilling water penetration into the core and its fractures.

Biological tracers will be used as a second independent means of identifying the source and extent of microbial contamination during drilling. Two different bacteria will be used, Chromobacterium violaceum and an ice nucleation active (INA) Pseudomonas syringae, as independent tracers to assess the source and extent of drilling-associated microbial contamination. Chromobacterium is easy to detect because of its distinctive violet color. INA bacteria are unique and easy to identify because they freeze water very quickly.

In the first of the bacterial tracer experiments, a sterile plastic bag containing the Chromobacterium tracer will be taped to the drill bit just prior to drilling through the plug at the bottom of the 6-inch casing and into the competent bedrock. The first three cores of the competent bedrock will be processed in the glovebox and the fracture faces and porewater associated with each one will be analyzed for the Chromobacterium tracer. Spot tests may also be made on cores recovered from deeper zones of the competent bedrock during subsequent days of drilling. The presence of Chromobacterium will indicate the potential extent of microbial contamination in the cores originating from the overburden and weathered bedrock.

The INA Pseudomonas syringae tracer will be used in another borehole to quantify the extent of core contamination resulting from the drilling water. A concentrated, freeze-dried and killed preparation of commercially-available Pseudomonas syringae (Snomax Technologies, Rochester, NY) will be introduced into the drilling water at the beginning of the first day of coring in competent bedrock. It will be dispersed throughout the drilling water in the 500-gallon holding tanks using the nitrogen gas bubbled into the bottom of the tanks. The fracture faces and porewater of the weathered and competent bedrock cores obtained during the drilling process will be analyzed for the INA tracer. The drilling water will only contain the INA bacteria on the first day of preliminary coring of the borehole. On subsequent days, drilling water will not contain the bacterial tracer. The presence of INA bacteria in the cores will indicate the extent that the drilling water penetrates into the fractures.

In a separate experiment in the laboratory, bedrock fracture faces will be exposed to solutions of Chromobacterium and the INA bacteria for different periods of time to determine their sticking efficiency on the cores. This will help in understanding how readily these bacteria might become associated with the core surfaces during drilling.

A third independent approach will also be used to characterize the microbial populations and assess the extent of microbial contamination of bedrock fracture faces from the overburden and drilling water. Community-level physiological profiles (CLPP) will be determined for microorganisms in the overburden, the drilling water, the porewater associated with the bedrock cores, and the fracture faces of the bedrock cores. In CLPP, the microorganisms are placed in tiny plastic cups, each of which contains a different food or nutrient source they might be able to use. If the microbes degrade a specific food or nutrient source, the cup will change color. The amount of color change is measured and indicates the microbes' affinity for that food/nutrient source. Data for all food sources for each sample are analyzed and help distinguish between microbial communities from the groundwater, overburden, drilling water, and fracture faces. The CLPP information will not only allow for the detection of drilling-associated contamination, but will also provide insight into the capability of the microbial communities at Site 32 to use certain contaminants as food sources. In a similar technique, community-level phylogenetic profiles (CLGP) will be used to compare the DNA (genetic material) from the microbial communities living in the groundwater, overburden, drilling water and fracture faces to detect drilling contamination.

Drilling Parameter Recorder
Drilling parameter recorders (DPR) consist of computerized systems which monitor a series of transducers installed on conventional drilling rigs to collect data on all aspects of drilling, including advance rate, downthrust and pull-up pressures, rod torque, rotation rate, mud/water pressure and flow, depth and time. Some DPR systems are also equipped to record the percussion wave reflected by the rock or soil in the drill string and the inclination of the drill mast. The data for the different parameters are acquired automatically by simply pushing a button on a rugged drill rig-mounted recorder operated by the driller. The data is displayed in real time on a strip chart or thermal paper and is also stored on an electronic medium for further analysis. These systems ensure that all information is registered from the ground surface to the final depth of the boring, and also that the depths stated on the boring logs are accurate.

The results of the DPR logs will be compared with those obtained during standard geophysical testing, acoustic televiewer monitoring, videologging and the information gained from direct examination of the bedrock cores. These comparisons will help determine how these methods could be used most effectively, separately or in combination, to gain the most information possible about lithology and fracture patterns in a bedrock aquifer. From this information, it will be possible to develop relationships between the energy needed to core a soil or rock layer and conventional geotechnical parameters such as standard penetration resistance, cone tip resistance, modulus of deformation, logging density, rock quality designation and seismic velocities. In addition, information concerning the drilling parameters can significantly enhance the quality of the data obtained from the boreholes and lead to improvements to drilling equipment and techniques. Such advances can increase efficiency, productivity and control of the drilling process.

by Nancy E. Kinner, Kimberly S. Newman, Larry K. Brannaka, and Jean