Principal Investigators: Mojdeh Delshad and Gary A. Pope
Chlorinated solvents known as dense non-aqueous phase liquids (DNAPLs) are one of the most serious groundwater contamination problems in the U.S. Surfactant enhanced aquifer remediation has emerged in recent years as the best available technology for remediating groundwater contaminated by chlorinated solvents. Accurate flow and transport modeling with the University of Texas chemical flooding simulator, UTCHEM, has played a critical role in the understanding of surfactant remediation and its efficient and reliable use to clean up superfund sites.
Mathematical models of multiphase flow in subsurface environments generally employ a local equilibrium assumption. However, field data frequently indicate that contaminant concentrations in groundwater are lower than their corresponding equilibrium values.
The important implication of this in field remediation design is that the predicted clean up time and volume of treatment fluids calculated from a non-equilibrium model can be larger than those calculated with an equilibrium model. Despite these significant findings, current practices for estimating clean up time ignore the effect of the surfactant on mass transfer rate coefficient. Since the efficiency of surfactant remediation depends on the dissolved contaminant concentrations in the flowing phase, a more fundamental understanding of mass transfer limitations under field conditions is needed and will result in a more efficient and effective remediation.
The emphasis of this project was on investigating for the first time DNAPL saturation end points approaching zero since this is both the ultimate objective of clean up and the condition where mass transfer limitations are most significant. A central question in surfactant remediation is just how low the best available surfactants can reduce the contaminant concentration in the soil, or in other words the all important end point. The issue of rate-limited mass transfer becomes significant as the DNAPL saturation approaches zero. We need a better understanding of this limitation under realistic conditions, and this will require both experiments designed for this purpose and enhancements of our flow and transport model UTCHEM.
In our experimental part of the project, the potential for nonequilibrium conditions was investigated for TCE as the model NAPL. First, batch solubilization experiments were conducted to assess the significance of chemical rate limitations. Surfactant flood experiments were performed in packed columns with residual TCE. Mass transfer rate coefficients were determined as a function of aqueous phase pore velocity, NAPL volumetric fraction, and surfactant concentration. A new correlation for predicting mass transfer rate coefficients as a function of system properties was developed. The mass transfer rate coefficients and correlation were found by fitting the column effluent concentration with the results of a transport simulator. The correlation developed here predicts near-linear dependence of mass transfer rates on NAPL volumetric fraction and pore velocity.
The new mass transfer correlation was implemented in UTCHEM and field-scale simulations were performed to learn more about the importance of rate limited mass transfer in a typical aquifer suitable for surfactant remediation and the coupling and role of soil heterogeneities on all relevant scales. SEAR operating strategies and conditions such as surfactant structure and concentration and injection/extraction rates were evaluated with the emphasis on how to achieve very low end points and the roles of mass transfer and heterogeneity.
Center for Petroleum and Geosystems Engineering
1 University Station C0304
The University of Texas at Austin
Austin, Texas 78712-0228
Phone: (512) 471-3246 FAX: (512) 471-9605