MODELING KINETIC INTERPHASE MASS TRANSFER OF CONTAMINANTS DURING AIR SPARGING WITH A DUAL-CONTINUUM APPROACH

 FALTA, Ron, faltar@clemson.edu, Department of Geological Sciences, Clemson University, Clemson, SC 29634-0976

The unstable nature of air sparging gas flow below the water table causes injected gas to flow locally through small channels, which are separated by a distance of one or several millimeters.  Although it is possible to model the average air sparging darcy flux using a conventional multiphase flow approach, this type of model does not resolve local mass transfer effects which arise due to the millimeter scale gas channels.  For this reason, compositional multiphase flow simulators which assume local chemical equilibrium between the phases may overestimate the rate of contaminant mass transfer to the gas phase during air sparging.
    An alternative method is proposed for modeling the local mass transfer during air sparging.  The method is based on the existing dual media formulation which is commonly used for modeling flow and transport processes in fractured media.  Instead of considering fractures and matrix blocks, the method is applied to porous media to simulate the local gas channels which form during air sparging.  This allows resolution of the local diffusive mass transfer between the flowing gas phase, and nearby stagnant water filled zones.  Because each media is modeled with a single node separated by some local average distance, the diffusive mass transfer is mathematically analogous to a first order interphase mass transfer reaction.
    The new model is applied to laboratory column experiments in which dissolved trichloroethylene (TCE) is removed by air sparging.  The new numerical formulation provides a good match with these data, it is shown that the simulation results are very sensitive to the nature of the local mass transfer limitations.
    The numerical model is then applied to a hypothetical field air sparging case involving a dissolved plume of TCE, with the same material parameters as the laboratory experiments.  In this instance, the effluent gas concentration from the SVE recovery well is found to be relatively insensitive to the details of the local mass transfer.  Instead, the long term behavior of the system depends more on the large scale diffusive mass transfer of the contaminant in this case.