A modeling study and an in-situ field test have been performed at a Superfund site to estimate the intensity of remediation required to reach groundwater clean-up standards in a shallow aquifer affected by DNAPLs (dense non-aqueous phase liquids). A previous pilot recovery test conducted in 1989 (Connor, Newell, and Wilson, 1989) had demonstrated that the mobile fraction of the DNAPL fluids could be removed from the water-bearing stratum using enhanced oil recovery technologies. However, a significant percentage of the DNAPL mass would remain immobilized within the aquifer matrix, possibly acting as a continued source of organic constituent release to groundwater due to dissolution of soluble DNAPL components. To quantify the effect of DNAPL dissolution on the aquifer remediation program now being planned for the site, a modeling study and in-situ field test were designed to simulate the performance of the full-scale groundwater recovery system.To obtain a preliminary estimate of the DNAPL dissolution rate, an existing dissolution relationship (Borden and Kao, 1989) was applied using field data previously collected from the site. The model assumed equilibrium partitioning of soluble organics between water and DNAPL, and required estimates of the mass of DNAPL in the aquifer and the chemical composition of the DNAPL itself.Using the model results, the field test was designed to simulate the actual hydraulic conditions that might be expected during a full-scale pumping and waterflooding program for DNAPL-affected areas. One injection well and one production well, spaced 11 feet apart, were used to flush fresh water through the test area. During the 30-day test period, groundwater samples were collected from 4 observation wells and analyzed for organic indicator constituents. During the test, 14 – 43 pore volumes of fresh water were drawn through the test area, resulting in significant reductions in the concentrations of key organic constituents (i.e., > 85%).The test results were analyzed to determine the number of pore volumes required to achieve the specified groundwater recovery standards, ranging from 2.4 to 39 mg/1 for two of the key dissolved organics indicator parameters. Pore volumes, or the number of times the groundwater in a given volume of aquifer is replaced by fresh water, is an important indicator of the remediation intensity required at a site. A large number of pore volumes, for example, indicate that either a large number of recovery wells or an extended clean-up period will be required. At this site, the field test showed that approximately 50 to 100 pore volumes of fresh water may be required to achieve the groundwater recovery standards in DNAPL-affected areas. These empirical results are in good agreement with modeling predictions, which indicated that roughly a 50 pore volume flush would be required to adequately reduce dissolution products from the residual DNAPL. These data indicate the significance of the DNAPL dissolution phenomena to groundwater remediation design.Due to the effects of DNAPL dissolution, restoration of groundwater to drinking water standards (organic constituents in the low ppb range) may prove infeasible at many DNAPL-affected sites. A general management strategy, consisting long-term containment of organics in the DNAPL zone and restoration of groundwater affected by dissolved plumes in DNAPL-free areas, is proposed.
