An assessment of the effects of field-scale formation heterogeneity on surfactant-enhanced aquifer remediation.

Abstract

Surfactant-enhanced aquifer remediation (SEAR) is currently being studied as a method for dramatically reducing the time required for cleanup of entrapped nonaqueous phase liquids (NAPLs) from the subsurface. While much has been learned about NAPL entrapment and solubilization behavior at the laboratory scale, extension of experimental results to field applications is likely to be strongly influenced by field-scale heterogeneity. In this study, numerical simulation techniques are used to explore the entrapment and subsequent surfactant-enhanced solubilization of a spill of tetrachloroethylene in a heterogeneous saturated sandy aquifer. A mathematical modeling approach is developed to describe surfactant-enhanced solubilization at the column scale, and the approach is incorporated into a one-dimensional numerical simulator. Model predictions are calibrated to experimental studies of surfactant flushing of dodecane in sand columns. A realistic depiction of aquifer heterogeneity is generated, using the turning bands method to create multiple realizations of random, spatially correlated permeability fields. Utilizing the generated fields, a multiphase flow simulator is employed to explore the effect of heterogeneity on the initial spill configuration. The influence of heterogeneity on subsequent cleanup of the entrapped NAPL is examined through application of a two-dimensional, multicomponent, surfactant-enhanced solubilization simulator which is an extension of the one-dimensional simulator described above. A matrix of model simulations is conducted and analyzed using a single-factor statistical modeling approach to identify the input parameters and processes most critical to the determination of NAPL spill configuration and SEAR performance. The analysis suggests that the most significant factors influencing NAPL distributions will be reliable estimates of the mean and variance of the formation permeability, an estimate of the vertical correlation scale of the formation, and an accurate representation of the relationship between the capillary pressure-saturation function and the permeability. The permeability variance and correlation structure are also shown to have a strong influence on organic source removal and long-term tailing of organic concentrations. The insight gained from model simulations is then used to develop simplified, vertically averaged methods for modeling of SEAR without explicit characterization of the permeability field.