CAPILLARY PRESSURE-SATURATION CURVES FOR AIR AND DNAPL ENTRY INTO A WATER-SATURATED FRACTURED SHALE SAPROLITE

Dense non-aqueous phase liquids (DNAPL’s) are important contaminants at many hazardous waste disposal sites. While DNAPL behavior has been widely studied in homogeneous aquifer materials, relatively little information is available on DNAPL behavior in heterogeneous porous media such as fractured saprolite. We measured capillary pressure-saturation curves for air and Fluorinert^TM (a non-toxic DNAPL surrogate) intrusion into an 18-cm long by 10-cm diameter undisturbed column of water-saturated fractured shale saprolite. The experiments provided insight into the ability of DNAPL's to penetrate saprolite, as well as information on the pore structure of the material. Initial entry of air and Fluorinert into the saprolite column occurred at relatively low capillary pressures, equivalent to 13 and 7.5 cm of water head, respectively. These entry pressures correspond to maximum fracture apertures of 121 and 142 mm, respectively. As capillary pressures increased, the volume invaded increased gradually rather than in one or two discrete steps. This indicates there is not a distinct division in size between pores in the fine-grained matrix and fractures or other macropores, which is consistent with microscopic examination of pore structure in thin-sections of the saprolite. A simple method was developed to estimate the elevation of the moving immiscible interface within the column from the applied non-wetting phase pressure and initial matrix entry pressure. This moving boundary was used to calculate capillary - pressure saturation curves that are accurate at the initial entry pressure, and then steadily approach the effluent elevation curve at residual saturation. Although an approximation, these curves better represent the capillary pressure of a tall column much better than any single constant elevation. Predictions of Fluorinert behavior from the air-water data, using a scaling factor based on the fluid interfacial tensions, overestimated the initial fracture entry pressure by 26%, and underestimated the matrix entry pressure by 26%. The magnitude of the disagreement between predicted and measured values was relatively small compared to the variation in pore sizes observed in thin sections. This suggests that air intrusion experiments can be used to predict the entry of DNAPL’s and other immiscible fluids into water-saturated saprolite.