Reservoir Fluid Characterization for Miscible Enhanced Oil Recovery

Abstract CO2 injection is an effective method for enhanced oil recovery (EOR). The Pennsylvanian Morrowan sequence of the panhandle of Texas through Southeast Colorado in the United States currently presents an opportunity for CO2 EOR and carbon sequestration. The Farnsworth Unit (FWU) of Ochiltree County, Texas operated by Chaparral Energy L.L.C. is the site of a CO2-EOR project using anthropogenic CO2. The Southwest Regional Partnership on Carbon Sequestration is working closely with Chaparral Energy to study the effectiveness of large-scale CO2 injection into the FWU. This paper presents the reservoir fluid model for the FWU. It includes a comprehensive PVT analysis of fluid that has been tuned to an equation of state (EOS) with 9 components. After tuning, the experimental and simulated data had an excellent match. A series of slim tube simulated experiments were carried out to determine the minimum miscibility pressure (MMP). The C7+ fractions were split into two pseudo-components with appropriate critical properties without affecting the actual fluid properties. This aided in optimizing the predictions even while using fewer pseudo components to characterize the fluid. Simulation of the MMP using the EOS tuned fluid was conducted and the predicted MMP was close to the experimental value with an error less than five percent. A sensitivity analysis was also conducted with varying impurities in the CO2 injection fluid to analyze the effect of recycle gas on the MMP. The approach used in this study demonstrated an improved strategy in lumping and splitting of compositions to ensure accuracy in reservoir fluid characterization. Introduction Fluid characterization plays an important role in the compositional simulation of enhanced oil recovery processes (EOR) like miscible gas flooding. Black oil models do not efficiently deal with the complexities of phase behavior in the case of compositional modelling. This makes the use of compositional fluid characterization essential.