FIELD EXPERIMENTS SHOWING STRONG MATRIX DIFFUSION EFFECTS ON PCE AND TCE DNAPLS IN FRACTURED CLAY

Field experiments involving chlorinated solvent DNAPLs were conducted in a surficial clay aquitard with natural fractures to examine the influence of fracture network and matrix diffusion on contaminant mass distribution. Five months after each infiltration, the separate contaminated zones were excavated in lifts to observe fractures and take numerous samples at 1 cm spacing along transects orthogonal to fractures containing visible DNAPL. Concentration profiles of contaminant mass in the low-permeability matrix blocks between fractures were used to quantify rates of DNAPL phase transfer by molecular diffusion causing complete dissolution of the TCE DNAPL in many fractures in a 4.5 month period, with remaining DNAPL phase present in disconnected filaments along larger, first-order fractures. A diffusion profile obtained near the TCE infiltration point was used to obtain a best-fit combined diffusion coefficient (De/R) and retardation factor (R) given a known time for DNAPL arrival and assuming a constant aqueous solubility concentration at the boundary. The best-fit porous medium diffusion coefficient (De) and retardation factor for the measured profile match calculated values using the modified Wilke-Chang correlation equation and independently measured porous medium parameters. The rate of matrix diffusion phase transfer in this clay deposit is 5 times faster for TCE than PCE and explains observations made during the excavations, where PCE DNAPL was found frequently in fractures within the 25 m3 contaminated volume but TCE DNAPL was only found in few larger fractures within the 13 m3 volume of contamination. Additional data suggest that a different rate of diffusion mass transfer to the matrix is the most important factor for the distinction in contaminant source zone sizes and shapes. Despite the greater persistence of the PCE DNAPL after 5 months, the storage capacity for dissolved and sorbed mass in the matrix of clays are sufficiently large to allow complete dissolution from fractures in a few years, which is short compared to the age of most contaminated sites. These experiments show complete dissolution of DNAPL phase for common chlorinated solvents is likely at many such sites if the mass retention capacity of the matrix is not exceeded due to numerous repetitive releases to the fractures.