Formatted Title
Documentation of Kilometer-Scale Transport of DNAPL Resulting in a Remote Secondary Source and Persistent High-Concentration Groundwater
Background/Objectives
Large releases of dense nonaqueous phase liquid (DNAPL), travelling long distances driven by gravity, have the potential to support remote areas of persistent high concentration groundwater contamination. The prevailing view, however, generally presumes that dissolved phase transport and matrix back diffusion are the principal reasons for persistent contamination in plume areas far from the original source release. Unfortunately, standard characterization technologies do not provide the resolution to directly identify small, discrete occurrences of DNAPL in the subsurface to assess the role of DNAPL transport in the persistence of contamination in the mid and distal plume areas. At the U.S. Department of Energy’s (DOE) Savannah River Site (SRS), 1.7 million kilograms of chlorinated solvent DNAPL were released to the subsurface via a settling basin. Despite the installation of hundreds of monitoring wells, DNAPL was directly detected in only one well adjacent to the settling basin. Notably, high aqueous concentrations indicative of potential DNAPL were measured in monitoring wells along a tortuous path consistent with density-driven travel along the structure contour of fine grain strata below the water table. The mapped path extended a kilometer west of the basin and the orientation was not aligned with groundwater flow. During a field test to assess the viability of chemical oxidant injections for remediating primary and secondary sources of DNAPL, we also collected data to assess the presence of remote DNAPL. Following oxidant injection, the multiple lines of DNAPL evidence included a novel-opportunistic DNAPL tracer (mercury) as well as a chlorine balance.
Approach/Activities
We conducted an oxidant injection field test at SRS in a small portion of the extensive groundwater plume approximately 650 m from the source release area and orthogonal to prevailing groundwater flow. Permanganate solution followed by persulfate solution were injected through wells screened in the Lost Lake Aquifer (> 50 m bgs). Groundwater from monitoring wells downgradient of the injection wells were analyzed for a comprehensive suite of chemical parameters in several campaigns prior to and after injection. Elemental mercury was co-disposed with and transported by the DNAPL. Following the application of oxidant, mercury from the affected DNAPL was converted to an easily measured ionic form, providing a sensitive DNAPL indicator. Another key DNAPL indicator was generation of excess chloride beyond what could be accounted for by a combination of background and stoichiometric conversion of aqueous phase contaminant concentrations.
Results/Lessons Learned
Following oxidant deployment, we consistently found stoichiometric excesses in chloride concentrations along with statistically significant increases in aqueous mercury concentrations. The data confirm that DNAPL traveled far from the source release area but was not directly detected by other characterization methods. The determination of kilometer-scale DNAPL migration supports a conceptual model that encourages consideration of alternatives to matrix diffusion. This finding also may inform remediation strategies to reduce cleanup time since DNAPL has >1,000 times the contamination potential of aqueous-phase contaminants. The destruction of potent DNAPL secondary sources will significantly reduce the persistence and magnitude in impacted plume areas. Identification of a remote DNAPL rather than large-scale matrix diffusion as a secondary source scenario may allow converting a technically impracticable site to one that might be effectively remediated.