Experimental evaluation of in situ chemical displacement of trichloroethylene (TCE) from aquifer media using ethanol

Restoration of aquifers contaminated with denser-than-water non-aqueous phase liquids (DNAPLs) such as trichloroethene (TCE) presents a significant remedial challenge. The problems associated with conventional pump-and-treat remedial approaches have served as the impetus to develop technologies such as in-situ chemical displacement using alcohols to accelerate DNAPL remediation. The ability of ethanol to displace TCE from coarse and fine sands was therefore investigated. Ethanol was chosen because it is mutually soluble with TCE and water and can therefore significantly enhance the aqueous solubility of TCE. Moreover, the ethanol reduces the interfacial tension between the DNAPL and water thereby lowering the capillary pressures required for mobilization.;While these phenomena are purportedly beneficial to enhanced recovery of TCE, the complex interplay of capillary forces arising from interfacial tension reduction and differences in the bulk fluid properties (density, viscosity) can lead to flow instabilities which may result in only partial recovery of TCE, its complete bypassing, or even result in downward mobilization of TCE, making it even more inaccessible for recovery. Many previous studies have addressed the efficiency of alcohol flooding, but the flow direction with respect to gravity, flow geometry and dimensions of the experimental apparats were often intentionally chosen to avoid any potential flow instabilities. Yet under typical field conditions, the governing flow equations suggest that instabilities arising from both gravity and viscosity are likely to occur.;Horizontal flow experiments were conducted in homogeneous and layered sandy soils using a vertically-oriented, two-dimensional apparatus with a flow geometry and velocities that simulated typical aquifer conditions. Six ethanol displacements of water were conducted in uncontaminated porous media to establish a base-line performance for ethanol flooding. The observed inclination angles of the propagating ethanol interface in homogeneous sands compared well with the inclination angles predicted by the Gravity Number. For flow velocities less than 5 m/day, the time required for the propagating ethanol interface to reach its stable configuration compared well with predictions based on a model of gravity segregation of miscible liquids in a no-flow domain.;After the ethanol was completely flushed out of each soil sample, a small quantity (5-7 ml) of pure TCE was injected into the center of the sample and ethanol displacement was repeated under the same flow conditions. Once the ethanol front contacted the region contaminated with TCE, rapid downward mobilization of TCE occurred at the leading edge of the ethanol front, indicating that interfacial tension reduction dominates NAPL dissolution under these conditions. In addition, TCE was observed to migrate downward, and upgradient, along the sloping ethanol-water interface.;Moreover, heterogeneities promoted a variety of bypassing and liquid mixing phenomena that resulted in multiple instabilities, which include, but are not limited to: physical displacement of TCE at the edges of discrete fine-grained lenses due to localized confined flow conditions; upward migration of ethanol through finer-grained soils due to buoyancy; rapid vertically-downward mobilization of TCE; upgradient mobilization of TCE around heterogeneities; migration of TCE into fine-grained soils at locations both upgradient and downgradient of the initial emplacement of TCE; and enhanced TCE dissolution.;Overall, the experimental results indicate significant complications could be encountered when applying (pure) cosolvent displacement for in-situ remediation of DNAPLs. Therefore, this methodology is likely to be applicable only in very limited situations in which excellent control over the flow field can be exercised.