An investigation of parameters affecting oil recovery efficiency of carbon-dioxide flooding in cross-sectional reservoirs

Among the enhanced oil recovery (EOR) methods carbon dioxide is considered the most promising for its broad applicability, high recovery efficiency, and attractive economics. In laboratory tests conducted at pressures above the minimum miscibility pressure (MMP), continuous carbon dioxide flood can recover more than 95% of oil in place (OIP). In field applications, the cost of continuous injection and the low oil recovery efficiency made this process impractical. The low oil recovery efficiency is attributed to low vertical and areal sweep efficiency. The major causes of the low recovery efficiencies may be classified into three categories: (1) gravity segregation, (2) reservoir heterogeneity, and (3) unstable viscous fingering. Water alternate with gas (WAG) injection processes have been employed in field operations to improve the recovery efficiency and cut the cost of gas injection.; The purpose of this study is to investigate the effects of reservoir and process parameters on the oil recovery efficiency of carbon dioxide WAG processes in cross-sectional reservoirs. To accomplish this, a two-dimensional compositional numerical simulator was developed. The simulator was functional and verified in this study. The simulator was then used to generate simulation data for studying the effects of seven dimensionless parameters on the oil recovery efficiency: (1) reservoir length to height ratio, (2) sine of the reservoir dip angle, (3) vertical to horizontal permeability ratio, (4) gravity to viscosity ratio (GVR), (5) injection rate, (6) water to gas (WAG) injection ration, and (7) pore volumes injected.; Results of the investigation showed that oil recovery efficiency as a percentage of the oil in place (OIP) is affected to different degrees by the seven parameters.; Two correlations of the oil recovery efficiency versus the seven dimensionless parameters were established. The first was established for pore volumes injection ranging from 0 to 1.0 and the second from 0 to 0.7. The second correlation showed better agreement with the simulation results. The correlations will provide useful information in the design of the carbon dioxide WAG processes in cross-sectional reservoirs.