Effects of Pressure and Pore Structure on CO2 Accessibility and Residual Oil Mobilization After Waterflooding

To clarify how pressure and pore structure regulate CO2 spatial accessibility, residual oil mobilization, and morphology redistribution after waterflooding, three quasi-two-dimensional microfluidic pore networks were investigated at nominal outlet back pressures of 10, 15, and 20 MPa. The apparent minimum miscibility pressure (MMP) of the simulated oil/CO2 system was approximately 20 MPa, as determined from slim-tube recovery tests. End-point CO2 occupancy after approximately 3 PV of CO2 injection, incremental oil recovery, and four residual oil morphologies were quantified. Two complementary descriptive indices were introduced: the occupancy-relative recovery index (ORRI), which relates incremental recovery to end-point CO2 occupancy, and the residual oil dispersion ratio (RDR), which characterizes the morphology balance between sheet/cluster-dominated and film/isolated-dominated residual oil. Higher nominal outlet back pressure was associated with lower sheet-like and clustered residual oil saturation. The cross-structure average sheet-like saturation decreased from 15.77% at 10 MPa to 3.63% at 20 MPa, whereas isolated oil saturation increased from 2.71% to 4.98%. Although pooled RDR and incremental recovery were positively correlated across the nine condition-specific observations (r = 0.847), the partial correlation decreased to r = 0.196 after controlling for pressure, indicating that the pooled relationship was largely associated with their common pressure dependence. At 20 MPa, Chips 1 and 3 exhibited nearly identical end-point CO2 occupancies of 40.04% and 39.98%, respectively, but different observed incremental recoveries of 65.02% and 69.85%. These observations indicate that CO2 spatial occupancy, oil recovery, and residual oil morphology provide complementary descriptions of displacement performance. The main contribution of this study is the integrated pore scale evaluation of CO2 spatial accessibility, incremental recovery, and residual oil morphology within a common experimental framework. Because each chip–pressure CO2 flooding condition was represented by one experimental run, chip-to-chip differences are interpreted as condition-specific observations rather than statistically resolved structural effects.