Chlorinated solvents are often disposed of in such a manner that they form pools on subsurface clay layers. There they slowly migrate into the clay layers, accumulating therein over time. Due to the low permeability of these layers, it is assumed that the migration occurs by dissolution and diffusion. However, field evidence suggests that more solvent may be stored in such layers than can be accounted for through this mechanism. The overall objective of this research was to examine the validity of diffusion as the dominant mechanism of transport and to examine the possibility that cracks in low permeability clay layers contribute to the accumulation of contamination therein and the long remediation times associated with these sites.

Since there are few reported measurements of the diffusion coefficient in clayey soils for contaminants of interest, steady-state measurements using the time-lag method were made in silt and silt-clay mixtures (75% silica silt and 25% Na-montmorillonite). Changes in clay structure for smectitic clays were measured using x-ray diffraction and through macroscopic observations of crack apertures. The mechanism of cracking was examined using x-ray diffraction, Fourier transform infrared (FTIR), and sorption measurements. The impact of cracking and dissolution rates on down gradient concentrations and remediation time frames was evaluated using numerical simulation.

The diffusion measurements showed that the diffusion coefficient for trichloroethene in a silt-clay mixture was at least two to four fold smaller than estimates used in field studies. Calculations based on the measurements obtained in this research suggest that there is an even greater discrepancy between the amount of mass storage in low permeability layers and that which can be attributed to diffusion. To account for this enhanced transport, it was postulated that direct contact between the waste and these layers altered the structure of the clay, and consequently the transport properties. Measurements using x-ray diffraction showed that contact with chlorinated field wastes decreased the basal spacing of water-saturated smectites from 19 Å to 15 Å, accompanied by cracks with apertures as large as 1 mm, within weeks. Calculations showed that even minimal cracking could easily account for the enhanced mass storage observed in the field.

To investigate the mechanism of basal spacing decrease, a set of screening experiments was performed, which identified a nonionic surfactant, an anionic surfactant, and a chlorinated solvent, as the minimum waste components necessary for cracking to occur. Sorption measurements showed enhanced synergistic sorption of the surfactants in the presence of the chlorinated solvent, while FTIR spectroscopy suggested a displacement of water from the interlayer space. Based on all the accumulated evidence, it was hypothesized that the nonionic surfactant sorbs on the margins of the interlayer space, displacing some of the interlayer water. The anionic surfactant sorbs via an interaction with the nonionic surfactant and enhances the dehydration of the interlayer space through the solvation of water in micellar aggregates. Thus, the reduction in basal spacing is predominantly through a process of syneresis.

The numerical simulations showed that the presence of cracks in a clay layer has a significant effect on down gradient concentrations, with the storage of the contaminant as a dense nonaqueous phase liquid (DNAPL) in the cracks and the rate of dissolution being key. In particular, if DNAPL is present in the cracks, it is the process of DNAPL dissolution which drives the contaminant farther into the clay matrix that impacts the down gradient concentration temporal history. Simulations comparing the influence of constant and variable dissolution rates showed that when the dissolution rate depends on the DNAPL saturation, the mass of DNAPL that persists in the clay is greater, extending the remediation time by decades.

This research focused on how significant amounts of low permeability storage may occur. The results suggest that contact between DNAPLs and clay minerals in the low permeability layers alters the structure of the clay, resulting in the formation of cracks. Calculations performed as part of this research indicated that even small cracks (naturally occurring or forming as a result of contact with DNAPLs) can significantly increase the amount of storage in these low permeability layers. Numerical simulations showed that extended remediation times may be associated with DNAPL residing in cracks, the dissolution of which serves to drive the contaminant farther into the low permeability matrix. These results indicate that conventional remediation, even if effective in removing contaminant from high permeability areas, will fail to decrease contaminant concentrations below the maximum contaminant level (MCL) for decades, suggesting lengthy and costly cleanups for contaminated sites where the remedial objective is meeting the MCL.