Significance of Incorporating Water-Wet Regions for Modeling Low Salinity Waterflood in Oil-Wet Carbonates

Abstract

Abstract Recent studies have shown that low salinity waterflood can recover additional oil from oil-wet carbonate reservoirs by altering their wettability. Further studies have investigated key mechanisms for wettability alteration, such as, interaction of sulfate ions with organic acids, importance of divalent cations, role of oil acid/base content. The progress, however, has been limited in terms of modeling the wettability alteration process. A few recent studies have reported mechanistic models based on surface-complexation reactions. However, these models have their limitations when applied to oil-wet reservoirs because the injected brine is assumed to interact with the entire rock surface. In reality, the surface reactions occurs in the water-wet regions and the water-wet regions increase as a result of wettability alteration. In this study, a multiphase multicomponent finite-difference reservoir simulator was developed, incorporating the change in oil-wet/water-wet rock fractions, for modeling low salinity waterflood in carbonates. The wettability alteration process was modeled by incorporating key surface-complexation reactions. Single-phase and oil recovery coreflood simulations were performed in which modified brines, such as seawater and diluted seawater, were injected. The simulation results show that the extent of reactions of low salinity brines is much less when these reactions are assumed to take place only in the water-wet regions of oil-wet rocks. Single-phase simulation results agreed well with sulfate delay and dolomitization reactions observed experimentally. Oil recovery simulation results showed reasonable agreement with coreflood results of low salinity water injection reported in the literature. In these simulations, the water-wet rock fraction evolved as a result of wettability alteration on injecting diluted seawater. Simulating geochemical reactions only on the water-wet regions is critical for modeling low salinity waterfloods. In this study, we developed an improved model with this feature and found it to be effective for modeling low salinity waterfloods.