Contents
Introduction
Leaking Underground Storage Tanks (LUSTs) continue to represent a significant threat to the Nation's drinking water quality. Recent EPA studies have determined that conventional pump-and-treat groundwater remediation systems, installed in the 1980s, have failed to achieve drinking quality standards (U.S. EPA 1989, 1992). Is this a surprise? Not really! These systems were designed to recover dissolved contaminants in groundwater. While they also addressed recovery of mobile light non-aqueous phase liquids (LNAPLs) floating on the water table, they did not address the recovery of residual LNAPL sorbed or retained by capillary tension in the unsaturated soils of the source zone. Failing to address the source of the contamination, the residual LNAPLs continued to partition soluble and semi-soluble contaminants to infiltrating rainwater as it percolated through the source zone. The contaminated water moved downward to the water table were it recontaminated groundwater. Hence, even when (after years of pumping and the expenditure of countless millions of dollars) drinking water standards were achieved and the pump-and-treat systems were shut off, groundwater contamination reappeared. Left unaddressed, residual LNAPLs will continue to act as a source of groundwater contamination for decades to come (U.S. EPA, 1992.)
Researchers are currently evaluating the applicability of enhanced oil recovery (EOR) technology to the remediation of LNAPL sites. EOR technologies under consideration include the use of hot water or steam flooding, cosolvents, surfactants, and polymers. Primary LNAPL recovery systems (e.g., drains and pumping wells used in conjunction with conventional pump-and-treat systems) may result in the removal of approximately 30% to 40% of the total LNAPL volume. EOR systems may remove up to an additional 50% of the LNAPL volume (Mercer et al., 1990.)
The following remediation system illustrates how EOR technology can be effectively integrated with (possibly preexisting on-site) conventional technology (pump-and-treat, soil vapor extraction (SVE) and air stripping) to address the remediation of residual LNAPL in the soils surrounding a former LUST excavation. Our focus will be on system components representing emerging technologies as details of system design for preexisting proven technologies can be readily found elsewhere. It is assumed that the following conditions prevail prior to system installation: (1) all mobile LNAPL has been recovered by conventional physical recovery techniques, such as single or double well recovery systems; (2) soils at the site are permeable (K>10-4cm/sec) with low organic carbon content (1% by weight) and are generally free of contamination except near the source area; (3) the major groundwater contaminants at the site are BTEX; (4) the water table is close to the surface (<18 ft); (5) bacteria exist in the aquifer that are able and willing to degrade the contaminants as the carbon source; and (6) nutrients and electron acceptors exist in or can be delivered to the aquifer in adequate quantities to support metabolism.
Funnel and Gate Barrier System
A funnel-and-gate barrier system, analogous to a sand filter turned on its side, will be constructed to contain and direct the contaminant plume through a permeable treatment corridor for in-situ bioremediation. Based on the results of laboratory compatibility studies, for superior sorption, reduced permeability, reduced susceptibility to permeation by petroleum products, and reduced susceptibility to chemical attack, barrier walls will be composed of composite Portland cement-bentonite-fly ash (CBF) slurry combined with an interior high density polyethylene (HDPE) synthetic liner (Vesperman et al., 1985, Spooner et al., 1985, and Stewart, 1978). These results will be verified prior to system installation. The composite CBF/HDPE barrier wall will be keyed into an underlying continuous very low permeabilty clay layer so as to erect a continuous impermeable chemical barrier around and beneath the source zone. Proper QA/QC procedures will be followed during and after construction to identify and prevent defects, windows, cracks, seam separation, tears, and punctures that may result in increased permeability. Any contaminants that may breach the barrier system will be released to the surroundings at a such a rate that dissolved contaminant concentrations will be naturally attenuated (by dilution, dispersion, chemical and biological degradation), thus ensuring that potential receptors are protected. Passive (water) conductivities, Kp , for CB slurry walls are in the range of 10-12 to 10-11m/s (Kosmatka et al.,1988.) Additives, such as polymers and fly ash can further reduce the hydraulic conductivities and also increase the chemical resistance (Vesperman, et al. , 1985.) Synthetic HDPE liners are characterized by very low passive conductivities, << 10-12 m/s.
The following equation was used to conservatively estimate the time to breakthrough for petroleum contaminants (Boscardin, 1993):
t = tb + [neL/Kci] [(((Kc/Kb) -1)1/b (b=2) + ((Kc/Kb) - b)) / (b-1)] where
tb = [neL/Kbi] [(a+(Kb/Kp))/(a+1)] Breakthrough time, discharge of 1 pore volume through wall
ne = effective porosity
L = wall thickness
Kp , Kb , Kc = passive, breakthrough and contaminated conductivities
i = hydraulic head gradient across wall
a, b = pre- and post-breakthrough exponents
A combined total fluid/vapor recovery well (possibly preexisting) constructed at exit of the funnel-gate barrier system will be screened across both the unsaturated and the saturated zones and will recover contaminated fluids (including any unused dissolved nutrients added to affect bioremediation) and any vapors that may be generated by the treatment system. The extraction well will consist of slotted schedule 40 PVC pipe surrounded by permeable sand pack. Wells will be screened to accommodate seasonal fluctuations in the water table level. An impermeable barrier will be constructed at the surface around the well head to prevent direct inflow of air from the surface to the well. Prior to system installation, a pump test will be conducted in the well to determine the pumping rate that can be sustained in the aquifer, an important consideration because it limits the amount and rate of water that can be circulated in the system during the bioremediation process. Due to the proximity of the barrier walls, relatively little water is anticipated to be required to be pumped from the well to achieve a significant hydraulic gradient from the source area in the bioremediation treatment corridor. Dewatering of the aquifer and increasing the hydraulic gradient in the treatment corridor relative to the surrounding water table will improve the performance of the barrier wall system, residual LNAPL stripping in the source zone and vapor recovery from the vadose zone. An explosion-proof high-volume vacuum diaphragm pump equipped with an air/water separator and associated piping will convey the extracted fluids and gases to the clarifier. Extracted vapors will be sorbed by passing through a GAC prior to discharge to the atmosphere.
Fluids pumped from the recovery well enter a (possible preexisting) clarifier tank equipped with baffles. The tank will be used to settle suspended solids and to recover any mobile LNAPL (if any). Accumulated sediment will be removed as needed. The clarified fluid will be conveyed via associated piping to the air stripping tower.
Clarified fluids enter the top of the (possibly preexisting) air stripping tower. Air supplied by an air compressor is forced upwards in the column against the flow of water. Volatile organic compounds (VOCs) transferred from the water to the air in accordance with Henry's Law will be sorbed by passing through a GAC prior to being exhausted to the atmosphere. The tower will be packed with materials designed to maximize the available surface area to facilitate transfer. The system will be sized according to the rate of flow of water and the rate of contaminant transfer required. Oxygen (the necessary electron receptor for aerobic biodegradation) is added to the water by contact with the moving air and the oxygen saturated water (with dissolved oxygen concentrations ª 8 mg/L) then exits the tower and is conveyed via associated piping to the infiltration gallery.
After water falls through the air stripping tower, nutrients required to enhance microbial growth will be flash mixed with supplied hot water in a tank and released into the water as needed at a steady rate. Microbes not only require carbon sources for digestion, but also require other nutrients for bioremediation. Common macronutrients are carbon, hydrogen, oxygen, nitrogen, phosphorus, calcium, magnesium, iron, sulfur, and iodine. Common micronutrients may include toxic metals such as nickel or chromium in very trace amounts. Microbes are 50% carbon, 12% nitrogen, and 2% to 3% phosphorus by dry mass (Bedient et al., 1994.) Since the contaminants will serve as the carbon and energy source and oxygen will be added elsewhere in the treatment process, nitrogen and phosphorus are likely to be limiting and will need to be added in the largest quantities. Soil sampling in conjunction with lab testing will be conducted prior to system installation to determine which limiting nutrients will be necessary to add for effective bioremediation. The addition of hot water will increase the rate of biodegradation in two ways: (1) by increasing the supply of carbon source by reducing the density and viscosity of residual LNAPLS hence increasing its mobility and bioavailability (2) by increasing the metabolic rate of the bacteria (U.S. EPA, 1992). After nutrient addition, water will be conveyed by gravity flow to the infiltration gallery.
After nutrient addition, non-sudsing biosurfactants (biodegradable surfactants) will be flash mixed with supplied hot water in a tank and released into the water as needed at a steady rate. The addition of biosurfactants should increase the effective solubility of the LNAPL components (by up to 1 to 2 orders of magnitude) making it available as a carbon source for bioremediation. Soil sampling in conjunction with lab testing will be conducted prior to system installation to determine the effectiveness and optimal concentration of biosurfactant to add for effective LNAPL dissolution. The addition of hot water will increase the effectiveness of the biosurfactant by reducing the density and viscosity of residual LNAPLS hence increasing its mobility and availability (U.S. EPA, 1992). After biosurfactant addition, water will be conveyed by gravity flow to the infiltration gallery.
Enhanced Dissolution Infiltration Gallery
After nutrient and biosurfactant addition, the clean, oxygen-saturated, heated water will be discharged into the high permeability pea gravel backfill of the former LUST excavation. The chemically and (to a lesser extent thermally) enhanced water, infiltrating under the influence of a high hydraulic gradient will dissolve the residual LNAPL in the source zone and transport it at a controlled rate into the bioremediation corridor. Soil sampling in conjunction with lab testing and field infiltration tests will be conducted prior to system installation to determine the ability of the infiltration system to dissolve residual LNAPL from the source zone and to deliver the additional nutrients and electron acceptors (oxygen) to the bioremediation corridor at the proper rate and concentration to effect bioremediation.
Infiltration galleries are regulated under the Safe Drinking Water Act (SDWA) of 1974 Underground Injection Control (UIC) program (40 CFR Part 146). The EPA considers a drinking-water source to be endangered if underground injection results in the placement of any contaminant into an underground source of drinking water and such contaminant results (or could result) in a violation of the national primary drinking water standards. The Infiltration gallery is a Class V injection well. A UIC permit will be obtained from the appropriate state regulatory authority prior to system installation. A substantial amount of site specific geologic and construction data may be required to show containment, recovery and impact of the injectate.
Bioremediation Treatment Corridor
Chemically and (to a lesser extent thermally) enhanced water containing dissolved contaminants (carbon and energy source), oxygen (electron acceptor) at saturation, i.e., 8 mg/L and growth nutrients will migrate at a controlled rate under the influence of an induced hydraulic gradient through the bioremediation treatment corridor. Indigenous microorganisms attached to the aquifer matrix substrate will then biodegrade the contaminants to CO2 and H2O. A feasibility study will be conducted prior to system design. Field sampling of soil coupled with laboratory microcosm and electrolytic respirometer studies etc. will be conducted to determine microbial nutrient, oxygen, and environmental requirements for optimal metabolic growth, and biodegradation rates necessary for system design (Flathman et al., 1984). These requirements will be monitored in response to changing conditions as necessary throughout the remediation process. Field sampling of soil and groundwater coupled with laboratory studies will also be conducted to determine chemical reactions associated with nutrient addition and sorption of organics onto the soil matrix which might affect the remediation process.
In addition to the above requirements, the performance of the system and proper distribution of nutrients will be monitored by periodically measuring the organic, inorganic, and bacterial levels. Carbon dioxide levels will also be measured as an indicator of microbial activity in the formation.
A key element in this system is contaminant monitoring. Both groundwater monitoring and vapor extraction wells will be installed outside the perimeter of the slurry walls. The wells will contain sensitive fiber optic detectors to warn against contaminant leaks. The total organic vapor detectors, or "sniffers," have detection limits of one part per million. The sensors will be linked to the existing alarm system in the service station that will notify pertinent personnel about possible leaks. With large-scale jobs, it may be cost effective to hook the detection system up to a telephone or even a satellite network that may be surveyed by personal computers at an office. Labor costs of sampling may be heavily reduced.
Once contaminant concentrations inside the treatment corridor are below drinking water quality standards, the system will be operated as necessary to flush biosurfactant out of the aquifer (if any) to ensure that the aquifer will return to its natural condition. The system will then be shut off, disassembled and removed from the site (as necessary).
Following completion of the treatment, a very low permeability bentonite cap will be constructed over the former LUST basin source zone/infiltration gallery to prevent infiltration of rainwater. The cap will then be paved with asphalt. The prior application of enhanced dissolution under the influence of a high hydraulic gradient should assure that any residual LNAPL will be immobilized under the reduced flow conditions. Any infiltration percolating through the remaining residual LNAPL will reintroduce contaminants to the aquifer in such low concentrations as to be readily degraded to below drinking water quality standards by the natural aquifer.
Conclusion
As an extension to an existing pump-and-treat system, our system may offer distinct cost advantages for residual LNAPL remediation of petroleum contamination. The system is not without its limitations. Advantages and disadvantages are summarized below:
Due to the difficulties associated with LNAPL remediation, the
heterogeneous nature of field hydrogeology, and the costs
associated with cleaning up LNAPL sites, LNAPL remediation system
design is, and will continue to be, site specific. No single
system is likely to emerge as superior to other methods.
References
Bedient, Philip, B., Hanadi S. Rifai, and Charles J Newell, 1994. Groundwater Contamination. Prentice Hall PTR, Englewood Cliffs, New Jersey
Boscardin, M.D., Ostendorf, D.W., 1993. "Barrier Walls to Contain Contaminated Soils", in Principals and Practices for Petroleum Contaminated Soils, Calabrese, E.J., and Kostecki, P.T., eds., Lewis Publishers, Chelsea, MI, 433-451.
Evans, J.C., 1986, "Slurry Trench Cutoff Walls for Waste Containment", International Symposium on Environmental Geotechnology, Vol. 1, Envo Publishing Company, 303-311.
Flathman, P.E., Quince, J.R., and Bottomley, L.S., 1984. "Biological Treatment of Ethylene Glycol-Contaminated Groundwater at Naval Engineering Center in Lakehurst, New Jersey," in Proc. 4th Natl. Symp. On Aquifer Restoration and Groundwater Monitoring, Nielsen, D.M., and Curl, M., eds., NWWA, Houston, Texas.
Hinchee, R.E., Downey, D.C., and Coleman, E.J., 1987. "Enhanced Bioremediation, Soil Venting, and Groundwater Extraction; A Cost-Effectiveness and Feasibility Comparison," in Proc. Conf. On Petroleum Hydrocarbons and Organic Chemicals in Groundwater, NWWA, Houston, Texas, pp. 147-563.
Kosmatka, S.H., and Panarese, W.C., 1988. Design and Control of Concrete Mixtures, 13th Ed., Portland Cement Association, Skokie, Illinois.
Lee, M.D., Thomas, J.M., Borden, R.C., Bedient, P.B., Ward, C.H., and Wilson, J.T., 1988. "Biorestoration of Aquifers Contaminated with Organic Compounds," CRC Critical Rev. Environ. Control, 18(1):29-89.
Lee, M.D., and Ward, C.H., 1985. "Biological Methods for the Restoration of Contaminated Aquifers," Environ. Toxicol. Chem. 4:743.
Mercer, J.W., and Cohen, R.M., 1990, A Review of Immiscible Fluids in the Subsurface, J. Contaminant Hydrology, 6: 107-163.
Nash, J.H. and Traver, R.P., 1993. "Field Studies of Insitu Soil Washing," in Principals and Practices for Petroleum Contaminated Soils, Calabrese, E.J., and Kostecki, P.T., eds., Lewis Publishers, Chelsea, MI, 403-409.
Office of Technology Assessment, 1984. "Protecting the Nation's Groundwater from Contamination," Report.
Raymond, R.I., Jamison, V.W., and Hudson, J.O., 1976. "Beneficial Stimulation of Bacterial Activity in Groundwaters Containing Petroleum Products," AIChE Symp. Ser. 73, p. 390.
Spooner, P., Wetzel, R., Spooner, C., Furman, C., Tokarshi, E., Hunt, G., Hodge, V., and Robinson, T., 1985. "Slurry Trench Construction for Pollution Migration Control," Noyes, Publications, Park Ridge, New Jersey.
Stewart, W.S., 1978. "State of the Art Study of Impoundment Techniques," USEPA Report No. EPA-600/2-78-196, Cincinnati, Ohio.
U.S. Environmental Protection Agency, 1989, "Evaluation of Groundwater Extraction Remedies", Vol. 1, Summary Report, EPA/540/2-89/054, Washington, DC.
U.S. Environmental Protection Agency, 1990. "The National Water Quality Inventory - 1988 Report to Congress," Washington, D.C.
U.S. Environmental Protection Agency, 1992, "Evaluation of Groundwater Extraction Remedies: Phase II," Summary Report, Office of Emergency and Remedial Response, Publication 9355.4-04, Washington, DC.
Vesperman, K.D., Edil, T.B., and Berthouex, P.M., 1985.
"Permeability of Fly-Ash and Fly-Ash-Sand Mixtures," Hydraulic
Barriers in Soil and Rock, ASTM STP 874, 289-298.
Links
Send comments or suggestions to:
Student Authors: Eric Anderson,
Joe Wallen, Micheal Kislek, Andy Carter
Faculty Advisor: Naraine Persaud, npers@vt.edu
Copyright © 1998 Daniel Gallagher, dang@vt.edu
Last Modified: June 7, 1998
