Intermediate - Scale Investigation of Nonaqueous - Phase Liquid Architecture on Partitioning Tracer Test Performance

725 Chlorinated solvents in the form of DNAPLs exist at a large number of hazardous waste sites. The distribution of DNAPLs is controlled by fi ngering (Held and Illangasekare, 1995), preferential channeling, and heterogeneity of the subsurface formation (Kueper and Frind, 1991). All of these factors increase the complexity of NAPL movement and subsequent entrapment (Schwille, 1988; Kueper et al., 1989; Illangasekare et al., 1995; Oostrom et al., 1999). Due to the complexity of entrapment architecture, the tasks of delineating and characterizing source zones have become a challenge when selecting and implementing effective remediation schemes. Traditional soil coring methods present diffi culties in resolving the spatial distribution of NAPL from discrete samples (Dai et al., 2001). Considering the heterogeneity of the subsurface, it is highly unlikely that the interpolation of core data will reproduce the distribution of contaminants. The importance and prevalence of the DNAPL problems at many waste sites have driven the need to investigate new techniques for source zone characterization. The partitioning interwell tracer technique (PITT) was modifi ed by Jin et al. (1995) for detecting and characterizing the distribution of NAPL contaminants in subsurface environments from a technique originally developed by petroleum engineers in enhanced oil recovery (Allison et al., 1991). The PITT has been used at both the laboratory and fi eld scales to detect and characterize NAPL source zones in the subsurface (Jin et al., 1995; Nelson and Brusseau, 1996; Annable et al., 1998; Cain et al., 2000; Meinardus et al., 2002). The method involves the injection of a set of reactive and nonreactive tracers that travel through the source zone located between injection and extraction wells. The transport of the reactive tracers is delayed with respect to the nonreactive tracers due to partitioning into and out of the DNAPL phase. If equilibrium conditions are present, the retardation depends only on the DNAPL saturation and the partition coeffi cient of the tracer (Kp). The entrapped DNAPL saturation can then be calculated using the following expression (Jin et al., 1995):