Maximize EOR Potential by Optimizing Miscible CO2 Injection Parameters in Carbonate Reservoirs

Abstract

Abstract CO2 injection to improve oil recovery has been successfully implemented in many oil reservoirs worldwide for almost half century. Conventional CO2 injection strategies, such as continuous CO2 flooding, water-alternating-gas (WAG) or carbonated water injection (CWI), were usually studied for their technical viability and cost effectiveness before deployment. This paper aimed to optimize the parametric design of CO2 injection schemes in a carbonate reservoir, and evaluated the impacts of WAG slug size, WAG sequence, and the initial gas saturation (Sgi) on the incremental oil recovery factors. This paper presents the results from experimental and numerical studies on the different forms of CO2 floods. We compared the secondary and tertiary recovery modes, and studied the effects of initial gas saturation on EOR performance, through a series of core flood experiments under reservoir conditions and detailed compositional simulation in a commercial simulator. The EOR core flood experiments were performed in a 2-ft long composite of carbonate cores, which were restored to reservoir wettability at reservoir conditions. The experiment indicates that carbonated water injection (CWI) in tertiary mode renders additional oil recovery ~ 20% OOIP above the waterflood (42% OOIP). In the cases of zero initial gas saturation or the reservoir pressure is above the bubble point, all three forms of CO2 injection (i.e. CO2 continuous, CO2-brine WAG and Brine-CO2 WAG) achieved > 93% OOIP recovery factor in the laboratory coreflood studies. If the reservoir was injected with produced/injector gas under immiscible conditions to support the reservoir pressure, the miscible gas EOR using CO2 WAG dropped by 10-15% OOIP compared with the corresponding cases of zero initial gas saturation. Therefore, it is recommended to implement miscible CO2 WAG flood after the initial gas saturation is minimized. Simulation studies with EOS model were carried out to simulate all of these EOR corefloods, taking into account of the effects of relative permeability hysteresis, wettability and Sorm (residual oil saturation) on oil recovery. The simulation results agree well with the laboratory results. In Summary, the presence of initial gas saturation in carbonate reservoirs reduces CO2 EOR performance under the miscible conditions. Without the presence of Sgi, the recovery factors from all the forms of CO2 EOR appear to be very similar in laboratory scale study. CWI was proved to be a good alternative to CO2 WAG, only when a significant amount of CWI is injected into the reservoir. These key findings have been validated by core flood experiments and compositional simulation studies with tuned EOS models.