Dimensional Analysis of CO2 Injection in Shale Oil Reservoirs

Abstract Carbon dioxide (CO2) injection effectively enhances oil recovery from shale reservoirs while potentially storing CO2. Several operational and reservoir parameters impact the oil recovery factor and CO2 storage efficiency. Rather than conducting extensive simulations through sensitivity analysis, dimensional analysis offers valuable insights into the dominant parameters affecting the CO2 HnP recovery factor (RF) and CO2 storage efficiency (EF). Accordingly, we conduct a dimensional analysis of CO2 injection in shale reservoirs by defining the key dimensionless parameters, including Fourier number (Fo), diffusivity ratio (RD), and drawdown pressure (PD). We then use a multiphase, multicomponent transport model that accounts for molecular diffusion, Knudsen diffusion, and viscous flow to simulate a single cycle of the CO2 injection process at different operational and reservoir conditions. Using the simulation results, we investigate the scaling relations between key dimensionless parameters (Fo, RD, and PD) and RF and EF. The results indicate that RF increases with Fo and reaches a plateau, achieving RFmax at an optimal Fourier number, Foopt, such as Foopt = 0.0025, 0.0025, and 0.025 at PD = 0.25, 0.5, and 0.75, respectively. Beyond Foopt, RF remains constant at RFmax because additional CO2 injection primarily enlarges a near-fracture CO2 bank that contributes little to further oil recovery. At high PD=0.75, the two-phase flow renders the injected volume ineffective below specific Fo. Furthermore, RD shows a positive correlation with RF, indicating that diffusion during the soak period has a favorable impact on oil recovery, particularly under two-phase conditions (PD=0.75). In contrast, RD has little influence on EF. The relationship between Fo and EF is similar to that of Fo and RF, where higher RF values correspond to greater remaining CO2. This relationship highlights the dual benefit of enhanced oil recovery and CO2 sequestration, which can be better understood and optimized through the presented dimensional analysis.