Measurement and Theory of Gas Trapping in Porous Media During Steady-State Foam Flow

Abstract

Abstract Aqueous foams are useful to reduce gas mobility during steam or CO2 flooding and for aquifer remediation. Trapped or stationary gas contributes significantly to the extent of gas mobility reduction. There is no consensus regarding the functional dependence of trapped gas fraction, or conversely the flowing gas fraction, but it is believed to vary with pressure gradient, pore geometry, foam-bubble size (i.e., texture), surfactant type, and concentration. This work uses a gas-tracer technique in combination with X-ray CT scanning to measure the fraction of flowing gas and the in-situ aqueous phase saturation during steady-state foam flow in Berea sandstone. Statistical network concepts are used to interpret the mechanisms that govern the mobile gas fraction. The experiments show that as gas velocity increases, the flowing gas fraction of the foam increases. Similarly, as injected gas-liquid ratio increases, the fraction of gas flowing increases. Hence, the absolute velocities of gas and aqueous surfactant solution are fundamental to foamed-gas mobility reduction. Effluent foam-bubble sizes were measured in a visualization cell installed at the core outlet. Bubbles range in size from about 60–120 μm in diameter. The smaller the effluent bubble size, the less the fraction of mobile gas. Lastly, experiments show that as surfactant concentration increased, the flowing gas fraction decreased. While it is likely that an empirical correlation could be deduced from the experimental results, such an approach does not engender a mechanistic understanding of the role of the various parameters. Instead, statistical network theory is employed incorporating the parameters identified as important in the experimental program: pressure gradient and bubble size. A closed form expression is developed for the fraction of mobile gas. It is predicted that the fraction of gas flowing is a weak function of pressure gradient, foam-bubble size, and the permeability of the porous medium. Moreover, the theory shows excellent ability to reproduce the newly obtained experimental data