Mechanistic modeling of carbonated waterflooding

Abstract CO 2 enhanced oil recovery is usually affected by poor sweep efficiency due to unfavorable mobility contrast between the injected CO 2 and oil. To alleviate this problem, CO 2 is added to the injected brine and transported in the reservoir by flood water. Therefore, Carbonated Water Injection (CWI) takes advantage of both CO 2 and water flooding processes. Furthermore, geochemical reactions between the injected carbonated brine and rock can alter petrophysical properties of the reservoir and affect final oil recovery. While there are several CWI coreflood experiments reported in the literature, simulation studies for this process are scarce. Accurate modeling of CWI performance requires a simulator with the ability to capture true physics of the CWI process. In this study, a compositional reservoir simulator developed at The University of Texas at Austin, UTCOMP, coupled with a state-of-the-art geochemical package developed by United States Geological Survey, IPhreeqc, is used to model CWI process. We considered the impact of CO2 mass transfer between brine and hydrocarbon phases based on thermodynamic constrains at the reservoir condition. In order to validate our simulation approach, the results of our CWI simulations were compared with a recently published coreflood experiment. Moreover, we investigated the fluid-rock interactions in CWI. The results of the simulations, indicated that prior to water breakthrough the main drive mechanism is displacement. But as more carbonated water is injected, CO2 diffuses more into the trapped oil left behind, which results in oil swelling and subsequent oil viscosity reduction. Moreover, reaction of carbonate minerals such as calcite with carbonated brine results in dissolution of the main rock matrix which consequently creates wormholes similar to carbonates acidizing. In this study we propose a novel approach for accurate modeling of carbonated waterflooding process. The results of this study highlight the importance of geochemical reactions in modeling CWI process. Our approach has been validated based on history matching at the backdrop of a recently published coreflood experiment.