A Multiscale Study of Foam: The Phase Behavior, Transport, and Rheology of Foam in Porous Media
Hirasaki, George J.
Doctor of Philosophy
This dissertation provides an in-depth multiscale understanding of the foam flow in porous media for subsurface applications such as gas mobility control, aquifer remediation, CO2 sequestration, water production control etc. In enhanced oil recovery (EOR), gas flooding has superior crude oil displacement efficiency where the gas sweeps. However, the overall oil recovery is often not much better than that of water flooding because of viscous fingering, gravity override, and reservoir heterogeneity. Using surfactant to generate foam in situ the porous media offer promise to simultaneously address all the three issues mentioned. Yet, the successful design of foam projects requires insightful understanding of the phase/component transport and the rheology of foam flow in porous media. The first part of my research investigates the foam transport dependence on its constituent components: the effect of gas types and surfactant structures. An experimental investigation of the effect of gas type and surfactant structure is presented. The effects of gas solubility, the stability of lamellae, the surfactant Gibbs adsorption, and the gas diffusion rate across the lamellae were examined. Our experimental results showed that the steady-state foam strength is inversely correlated with the gas permeability across a liquid lamella, a parameter that characterizes the rate of mass transport. We also calculated the limiting capillary pressure for different foaming surfactants and found that foam stability is correlated with the Gibbs surface excess concentration. My research also advances the understanding of “smart” foam rheology that can improve the sweep efficiency of gas flooding in porous media. Laboratory research work was conducted to capture the effect of heterogeneity on foam using actual reservoir rocks of varied permeabilities. It is observed that foam is more stable in high permeability cores compared to low permeability cores. Such smart rheology was also visualized in a 3-layered heterogeneous micromodel at the pore-level. Foam was shown to respond to the porous media heterogeneity by separating into a relatively dry and wet regime in the high- and low-perm regions respectively. Due to the capillary continuity between the layers of different permeability, the phase separation induced a saturation step change between the layers which resulted in responsive flow resistance. In addition, understanding the foam-oil interaction is crucial to the success of foam EOR projects. In this thesis, foam strength dependence on oil was probed using nuclear magnetic resonance (NMR) imaging technique. Manganese (II) was doped into the surfactant solution to reduce the T_2 relaxation time of water in order to differentiate the NMR response from the oil. The measured apparent viscosity was mapped out with respect to different oil fractional flows and oil saturations. It is found out that the presence of oil can not only weaken the foam strength but also emulsify with the surfactant solution. Unlike the bulk foam column test with oil, the overall apparent viscosity is found to be dependent on both oil saturation (oil fractional flow) and oil composition. Moreover, improved numerical algorithm is developed to estimate foam model parameters based on laboratory-scale experiment for field-scale reservoir simulation. Both dry-out effect and shear-thinning rheology of foam were considered. The algorithm reduced the five-parameter estimation to a few simpler steps, such as linear regression and single-variable optimization, and successfully avoided the sensitivities of initial estimates and the non-uniqueness solution issues. Our improved algorithm was also compared with others reported in literature. The robustness of the algorithm was validated by varied foam systems. Last but not least, the idea of injecting the surfactant with the gas phase (WAG+S, water-alternating-gas-plus-surfactant-in-gas process) has been conceptualized for the next generation CO2 foam EOR. Some novel nonionic or switchable surfactants are CO2 soluble, thus making it possible to inject the surfactant with CO2 slugs. Since surfactant could be present in both the CO2 and aqueous phases, it is important to understand how the surfactant partitioning between the phases influences foam transport in porous media. Foam simulations were conducted in both 1-D and 2-D systems. We conclude that when surfactant has approximately equal affinity to both the CO2 and the water, the transport of surfactant is in line with the gas propagation and therefore the sweep efficiency is maximized.