Winning the race against time
A mid-western research facility was decommissioning some of their laboratory buildings when contamination beneath a floor in one building was discovered. The contamination threatened to delay the renovation of the building -- scheduled to be turned into much-needed offices within the next two months. Contaminant levels were high enough to pose a potential risk to building inhabitants. Furthermore, the facility was planning to release the buildings at some future date and wanted to do so without restriction.
The cleanup was driven by the renovation schedule. Of the array of remedies available to address the polycyclic aromatic hydrocarbons (PAHs), only soil excavation could have remediated the problem in the required two months. However, cost estimates for excavation ranged well over $1 million. What was needed was a remedy that could be applied during and after renovation -- one that would not pose a potential risk to workers or building inhabitants during and after renovation, yet would eliminate future liability associated with the contaminated soils. Bioremediation was selected.
A site profile
Building 1 had been used since the 1950s for chemical and metallurgical research. The wastewater system from laboratory sinks and floor drains was a network of contaminated sewer pipe under the concrete floor of the basement.
The facility is located in a state with a voluntary action program (VAP), allowing facility owners -- under the oversight of a state-certified professional -- to remediate their sites without waiting for approval from the state. Facility owners must screen soils and groundwater for contaminants, based on whether the site is in an industrial or residential setting. Those sites exceeding screening levels must take steps to reduce potential risk and exposure.
More than 980 feet of soil was exposed in trenches in the basement of Building 1 as the pipes were removed. The trenches varied in depth and width, but generally were 2.5 feet wide and 2 feet deep. The soil underlying the floor was sandy, moderately-permeable clay. Initial concentrations of 64 PAHs ranged between non-detect to over 20,000 micrograms per kilogram (mg/kg). Table 1 summarizes the initial sampling results for those PAHs above the screening criteria. It can be seen that benzo(a) pyrene was the most elevated above the screening criteria. Indeed, benzo(a) pyrene also proved to be the most troublesome to remediate.
The research facility faced the following issues regarding their decision to remediate the soils. They could:
- excavate the soils, thereby moving the liability and quickly completing cleanup;
- reconstruct the floor and maintain the liability into the future, decreasing the building's worth and limiting its future use;
- implement a remedial technology that could interfere with the renovation schedule because of its duration; or
- implement a remedy that could be initiated during renovation.
Implementing the remedial system
The final option consisting of in-situ bioremediation was selected. A treatability study showed that using enhanced indigenous microbes was likely capable of remediating the soils within a reasonable time. Due to time constraints, the remedial system was designed and installed prior to completing the treatability study. Negative results from the study would have meant time and money wasted. The positive results minimized the schedule and saved money.
A treatability study was performed to evaluate the viability of various bioremediation options and to support the final design. The treatability study involved environmental sampling, database research and laboratory testing. Soil and groundwater data were needed. Two temporary groundwater monitoring wells were installed, one outside the building to establish a control point, and one inside the building to monitor any impact the remediation might have on ground water quality. Three soil samples were obtained for chemical analysis: two taken at depths of five feet in each of the boreholes where the monitoring wells were installed and one taken in an area of highest PAH concentration at a depth of three feet below the trench bottom. The treatability study required that an additional four soil samples be collected at a depth of 0.5 to 1.0 feet below the trench bottoms to test how the microbes were reacting to various levels of contamination. The soil samples had a moisture content of approximately 16 percent,
and the grain size distribution was spread over a wide range of sizes, from gravel to clay. Although the curves would not allow a direct calculation of hydraulic conductivity, an estimate of 1x10^-5 centimeters per second (cm/sec) was made based on visual classification and a description of the soils. Laboratory testing for PAHs showed that contamination did not exceed approximately three feet below the trench floors. The target for remediation was limited to a depth of three feet below the trench bottoms where the PAH concentrations were no longer detectable.
Sampling demonstrated that ground water did not contain detectable concentrations of pesticides or PAHs. However, ground water quality was monitored throughout the remediation and the data supported an exemption from ground water discharge permitting by the state. Nothing was detected in soils outside of the building foundations at a depth of five feet (SB-3). Therefore, remediation was limited to inside the building's foundations.
Based on the results of the treatability study, indigenous microbes present in non-contaminated soils could be made capable of remediating the contaminants. However, they were not thriving and could not be expected to degrade the contaminants in the foreseeable future without enhancement. The target duration for the cleanup was two years.
The tactical goal of this remedial action was to expose contaminated soils to treatment by transmitting the inoculum to an area larger than that believed to be contaminated. Contaminated wastewater was thought to have leaked from the drain lines in the trench. The remedial action sought to provide a larger hydraulic head than that which existed during the release scenario, thereby increasing soil penetration.
PAHs were in soils in the trench bottoms and at similar depths (approximately four feet below the floor) outside the trench walls -- below which concentrations fell under generic screening levels. The basic concept for the ecological restoration of the finite resources of soil and water is firmly grounded in proven natural laws. When faced with environmental stress, organisms respond by migrating, adapting or dying off. Site-specific microbes were enhanced in a balanced matrix of bacteria, fungi, algae and protozoa in the laboratory. The process was conformed to fit the environment. One application in the contaminated areas re-balanced and maintained the system, while targeting specific contaminants.
The contaminants in the soils beneath the basement floor in Building 1 consisted of PAHs. The soils were exposed in trenches that had been excavated to remove buried floor drains. The plan was to use these pre-existing trenches as part of the remedial design to minimize costs and compress the cleanup schedule.
As a first step in the remediation, the trenches were partially backfilled with clean gravel, designed to have a much higher hydraulic conductivity than the surrounding soils and to accept a discharge of over 200 gallons per minute. A four-inch, perforated horizontal piping was placed on the gravel bed along the upper one-third of all of the trenches. Figure 1 shows the current configuration of the treatment system under the floor of Building 1. This drawing shows groundwater monitoring wells (MW-1 and MW-2), injection ports (labeled "IP"), and soil sampling ports (labeled "SP") in approximately 950 linear feet of pipe. Each pipe length has at least one injection port, capped at either end. The injection ports are four-inch, schedule 40 riser pipe, connected with a T-fitting to the slotted pipe, capped with four-inch, rubber-ringed well caps and protected by flush-mounted manholes.
The soil sampling ports consist of two-foot square steel vaults with bolting covers. The vaults are generally two to four feet long, with the open bottom resting on soil and the top flush with the floor slab. They can be re-entered periodically to collect soil samples to monitor the progress of the cleanup. A new drain line for the renovated building is located approximately 1.0 to 0.5 feet above the bottom of the trenches. The soil hydraulic conductivity (1x10^-5 cm/sec) was estimated from grain size analysis, although the lack of uniformity in size distribution makes this estimate approximate. The hydraulic conductivity of the gravel is expected to be on the order of one cm/sec, estimated from literature and field values. Therefore, the trench was designed to fill before much liquid percolates into the surrounding soils.
The inoculum could flow vertically downward and slightly laterally, due to a temporary mound in the trenches, until the soils were penetrated at approximately 40 percent saturation to a depth approximately three feet below the trench bottom. The volume of liquid delivered was approximately the volume of the trench pore space (estimated to be 35 percent of the total volume based on the anticipated packing of the gravel). Whereas the release scenario had impacted soils beneath the trench and one foot laterally, the remedial action impacted the soils beneath the trenches to a depth of three feet and laterally out to a distance of approximately two feet on either side of the trenches.
The inoculum distributed the microbes throughout the soils. This was accomplished by first injecting inoculum and then injecting water to help drive it to the desired depth and distance. Once the microbes began to reproduce, this further distributed them throughout the targeted soils under anaerobic conditions. Ten thousand gallons of inoculum and 10,000 gallons of nutrient-enhanced, dechlorinated water were injected into the bioremediation system on November 9, 1996. The injection took approximately eight hours. It was conducted at night to reduce potential complaints of odors.
Hitting the target concentrations
Soil and ground water sampling was conducted monthly in each soil port and monitoring well for four months, then quarterly for 18 months until the soils reached concentrations below the soil screening criteria. The samples were analyzed for semi-volatile organic compounds by gas chromatography/mass spectrometry (GC/MS) by a state-certified laboratory (Figure 2). Although specific constituents cannot be identified from this figure (other than benzo(a) pyrene), the general distribution and magnitude of the contaminants can be seen. These results are discussed in the following section. Several peaks appear in the data in Figure 2, but the general trend is downward. A linear regression analysis, performed on these data, shows a downward trend in all of the data.
The degradation of products in the unsaturated "smear zone" is a multi-step process with each step requiring a separate microbial ecosystem that drives that step down to the next level. This continues until the contaminant has been reduced to non-toxic end products, such as carbon dioxide, dioxide and water. The intermediary products in the degradation process can produce what appears as random peaks, as shown in Figure 2. In most cases, these peaks represent the movement of microbes and the contaminant as it degrades. The life cycles of the microbes also vary and fluctuate, with different microorganisms peaking at different times. This also can cause fluctuations in the analytical results. Thus, the peaks and valleys are typical of indigenous microbial activity.
The remedy was successful in reducing all PAHs below the target concentrations. The concentrations of benzo(a) pyrene in SP-6 were the most difficult to remediate because one of the latest sampling events uncovered what appeared to be chunks of asphalt. The microbes were degrading the asphalt and the degradation products from that activity were being seen on the graph.
Conclusion
This study illustrates several important concepts about the use of indigenous microbes for bioremediation. The most fundamental point is that natural systems can be fine-tuned to successfully remediate contaminants without upsetting the natural balance of matter and energy flow. Using indigenous microbes and the concept of negative feedback reduces those factors that inhibit the degradation of a given chemical mix, reinforcing the use of balance to degrade contaminants by exploiting fairly passive techniques. In systems that have been stressed during remediation (e.g., pump and treat, highly oxidized bioremediation, and other aggressive extraction processes), once the stress has been removed, the system will try to readjust and contaminants often reappear from the smear zone or beyond the area of treatment.
Major factors of remedial design
. Energy is stored in the chemical bonds of compounds found in the soil and are released and restored during the degradation process. Only heat is lost, and energy is changed into useful abiotic components (such as iron, sulfur, potassium, magnesium, etc.);
Matter. A basic ecosystem contains producers that make food, consumers that use food and decomposers -- decomposing dead producers and consumers into basic elements taken up by producers to start the cycle again. Sunlight provides the solar energy that changes form as it passes through the system. All matter cycles and all energy flows through the system, making it self sustaining;
Carrying Capacity. An ecosystem maintains its carrying capacity (maximum sustainable populations) by introducing or removing limiting factors to prevent over- or under-population. The remedial process starts with a balanced carrying capacity and nature maintains it, making the cleanup durable and long-lasting;
Actions and Reactions. When one understands the biological and/or chemical processes that created the problem, find the opposite process to control it (such as changing the valiance state to make the contaminant unavailable in the environment); and
Stress. Introducing pollution creates extreme environmental stress. Organisms impacted by the greatest contamination have no time to migrate or adapt, so they die off. As one moves away from the center of contamination to areas of lower concentrations, these same organisms become capable of responding to stress genetically by mutating or "adapting" so that the contamination is no longer toxic to them. The indigenous microbe remedy calls for taking microbes from the areas of highest contamination and creating the optimum conditions in the laboratory for their adaptation. When they are re-injected at the site they continue to reproduce and pass on the enhanced selected genes to their progeny and biodegradation can continue until the clean-up standards are achieved.
Table 1. Summary of initial soil screening results
|
ConstituentState cleanup levels (m g/kg) |
|
|
|
|
|
|
Benzo(b) fluoranthene |
900-19,000 |
101 |
68 |
7 |
17,000 |
|
Benzo(a) pyrene |
390-14,000 |
101 |
78 |
59 |
1,700 |
|
Benzo (a) anthracene |
920-20,000 |
101 |
62 |
6 |
17,000 |
|
Dibenzo(a,h) anthracene |
390-2,100 |
101 |
14 |
3 |
1,700 |
This article appears in the January 2001 issue of Environmental Protection, Vol. 12, No. 1, page 46.
This article originally appeared in the 01/01/2001 issue of Environmental Protection.