In situ alcohol flushing of a DNAPL source zone at a dry cleaner site

A pilot-scale field test of in-situ alcohol flushing for enhanced solubilization and extraction of a dense nonaqueous phase liquid (DNAPL) source zone was conducted at a former dry cleaner site located in Jacksonville, Florida. This study was conducted to evaluate the feasibility of in-situ flushing for remediation of DNAPL sites in Florida. Groundwater at this site was contaminated with tetrachloroethylene (PCE) that had migrated below the water table, located at 3 m below ground surface (bgs), and collected at high saturations in thin, discontinuous layers in the 7.9 m to 9.4 m bgs depth interval. An oblong source zone (7.3 m x 2.7 m) was delineated using direct-push technologies and further characterized using soil coring and partitioning tracer techniques. Tracer tests and in-situ alcohol flushing were conducted using three injection wells that approximately bisected the source zone and six recovery wells located on the outer perimeter of the source zone. Over-extraction through the recovery wells ensured hydraulic containment within the test zone. A partitioning tracer test conducted before the alcohol flood provided an estimate of about 68 L of PCE within the zone swept by the wells. The test zone was flushed with 34 kL (equivalent to 2 pore volumes) of a 95% ethanol/5% water mixture over a period of 3 days. Packers were used in the injection wells to focus the flushing solution delivery to regions of the swept zone that showed larger initial NAPL saturations. Alcohol flushing removed approximately 43 L of PCE from the test zone (62% removal effectiveness). These results were in agreement with soil core data that indicated approximately 65% removal and a postflushing partitioning tracer test that indicated approximately 26 L of PCE remaining (63% removal). Postflushing groundwater concentrations of PCE were an average of 92% lower than preflushing values at 21 of 35 multilevel sampling locations within the test zone, but the combined effects of residual ethanol and incomplete flushing resulted in elevated postflushing PCE concentrations at the other 14 locations, Alcohol flushing successfully removed a substantial volume of DNAPL; however, evidence indicated that continued alcohol flushing would have resulted in a greater NAPL removal effectiveness.