Residual Oil Saturation, Asphaltene Deposition and High-Viscosity Crude Oil Production During Prolonged CO2 Miscible Core Flood for Carbonate Reservoir

Abstract

Abstract To evaluate residual oil saturation by miscible flooding (Sorm), as a key parameter for evaluating CO2 miscible injection, a coreflooding experiment was conducted using a composite carbonate core through a prolonged testing period. It was indicated that compositional alteration by mixing of CO2 and oil continuously caused the complicated phenomena such as in-situ asphaltene deposition and gradual viscous oil production. In this paper, we show the observation of those critical phenomena during the prolonged coreflooding. Whereas 100% oil recovery is rarely obtained in actual miscible CO2 flooding, zero oil saturation is often predicted by use of the compositional simulator. Despite that CO2 injection pore volume (PV) is crucial to obtain Sorm, a popular assumption sets the CO2 injection PV ranging from 1.0 to 2.5 in many experiments. In this study, we increased CO2 injection PV up to 15. During tertiary mode CO2 flooding, compositions of effluent fluids were continuously analyzed. After the flood test, fluid saturations were measured in each plug core by the Dean-Stark method and asphaltene contents were measured with extracted oil. The tertiary CO2 flooding under miscible condition achieved high incremental recovery. The Sorm was obtained in core scale based on two different measurements: effluent analysis and Dean-Stark method described earlier. To validate the investigation, the Sorm was discussed from a viewpoint of pore throat size distribution (PTSD). In this analytical discussion, the predicted minimum pore diameter of CO2 penetrated pore space revealed consistent to the experimental Sorm. Three intriguing events were found during the coreflooding: (1) waxy crude production, (2) CO2 dry-out phenomena, and (3) in-situ asphaltene deposition. A compositional analysis of effluent fluids implies that the produced oil gradually became more paraffinic. After the mid-stage of coreflooding, steady decrease was observed for chloride and sodium ion concentrations in the effluent water. This decrease can be interpreted to be caused by dilution of formation brine as the result of CO2 dry-out phenomenon. The asphaltene contents remaining in plug cores showed an evidence of CO2-induced asphaltene precipitation. Due to the limited PV injection, usual coreflooding experiments cannot reproduce the findings described earlier. In this study, we clarified overall trends of those events through entire CO2 flooding. In this study, we demonstrated the importance of CO2 injection PV to obtain more reasonable residual oil saturation than those derived from usual experiments. Furthermore, relating behaviors were clearly observed as in-situ asphaltene deposition and waxy crude oil production by the prolonged CO2 miscible coreflooding. These observations should provide a useful insight for a robust operation during entire CO2 EOR application through further risk evaluation of not only asphaltene but also paraffin.