Summary: Cyclic gas injection effectively enhances oil recovery for low-permeability oil reservoirs. Numerous theoretical, mathematical, and laboratory investigations have attempted to unlock underlying recovery mechanisms and optimal design for the cyclic gas injection in these reservoirs. While these investigations have shed light on various aspects of the process, different descriptions of key recovery mechanisms and optimal design parameters can be found in the literature. Many of these published studies consider conventional approaches and concepts, such as assuming advection-dominated fluid flow and mixing between injected gas and oil within the matrix, to simulate the process or conduct experiments. Under different reservoir and operational conditions, to be reviewed, one or a combination of mechanisms can be responsible for improving oil recovery. This review aims to critically examine the published modeling and experimental studies regarding the recovery mechanisms of gas cyclic injection and the conditions under which the process can enhance oil recovery. The review will identify lessons learned and areas in need of further research.

We study the stability of dissolution-driven convection in the presence of a capillary transition zone and hydrodynamic dispersion in a saturated anisotropic porous medium, where the solute concentration is assumed to decay via a first-order chemical reaction. While the reaction enhances stability by consuming the solute, porous media anisotropy, hydrodynamic dispersion, and capillary transition zone destabilize the diffusive boundary layer that is unstably formed in a gravitational field. We perform linear stability analysis, based on the quasi-steady-state approximation, to assess critical times, critical wavenumbers, and neutral stability curves as a function of anisotropy ratio, dispersivity ratio, dispersion strength, material parameter, Bond number, Damköhler number, and Rayleigh number. The results show that the diffusive boundary layer becomes unstable in anisotropic porous media where both the capillary transition zone and dispersion are considered, even if the geochemical reaction is significantly large. Using direct numerical simulations, based on the finite difference method, we study the nonlinear dynamics of the system by examining dissolution flux, interaction of convective fingers, and flow topology. The results of nonlinear simulations confirm the predictions from the linear stability analysis and reveal that the fingering pattern is significantly influenced by combined effects of reaction, anisotropy, dispersion, and capillarity. Finally, we draw conclusions on implications of our results on carbon dioxide sequestration in deep saline aquifers.

Cyclic solvent injection, known as solvent huff-n-puff, is one of the promising techniques for enhancing oil recovery from shale reservoirs. This study investigates the huff-n-puff performance in ultratight shale reservoirs by conducting large-scale numerical simulations for a wide range of reservoir fluid types (retrograde condensate, volatile oil, and black oil) and different injection gases (CO₂, C₂H₆, and C₃H₈). A dual-porosity compositional model is utilized to comprehensively evaluate the impact of multicomponent diffusion, adsorption, and hysteresis on the production performance of each reservoir fluid and the retention capacity of the injection gases. The results show that the huff-n-puff process improves oil recovery by 4–6% when injected with 10% PV of gas. Huff-n-puff efficiency increases with decreasing gas-oil ratio (GOR). C₂H₆ provides the highest recovery for the black oil and volatile oil systems, and CO₂ provides the highest recovery for retrograde condensate fluid type. Diffusion and adsorption are essential mechanisms to be considered when modeling gas injection in shale reservoirs. However, the relative permeability hysteresis effect is not significant. Diffusion impact increases with GOR, while adsorption impact decreases with increasing GOR. Oil density reduction caused by diffusion is observed more during the soaking period considering that the diffusion of the injected gas caused a low prediction error, while adsorption for the injected gas showed a noticeable error.

This paper investigates the impact of hydrodynamic dispersion on the stability of free convection in a saturated horizontal porous layer subject to a transient vertical concentration gradient and a steady horizontal background flow. A linear stability analysis (LSA) was conducted using the quasi-steady-state approximation to obtain neutral stability curves, critical times, and the corresponding wavenumbers as a function of dispersivity ratio (α) and longitudinal dispersion strength (β). The LSA results showed that the dispersive boundary layer becomes less unstable as longitudinal and transverse dispersivity increase. In addition, for the isotropic dispersive system with α = 1, the critical time and its corresponding wavenumber follow τ_c = 167.6/(1 − β) and κ_c = 0.0696 (1 − β), respectively. The nonlinear dynamics of the system were studied by examining the interaction of convective fingers, dissolution flux, and the time-dependent Sherwood number. Finally, the results were applied to 24 deep saline aquifers in the Alberta Basin.

Summary: A thorough understanding of fluid transport in ultratight shale reservoirs is crucial for designing and optimizing cyclic solvent injection processes, known as huff ’n’ puff (HnP). We develop a two-phase multicomponent numerical model to investigate hydrocarbon and solvent transport and species mixing during HnP. Unlike the conventional modeling approaches that rely on bulk fluid (advective) transport frameworks, the proposed model considers species transport within nanopores. The chemical potential gradient is considered the driving force for the movement of nonideal fluid mixtures. A binary friction concept is adopted that considers friction between different fluid molecules and between fluid molecules and pore walls. After validating the developed model against analytical solutions and experimental data, the model examines solvent HnP enhanced oil recovery (EOR) mechanisms by considering four-component oil and Eagle Ford crude oil systems. The impacts of injection pressure, primary production duration, soaking time, and solvent type on the oil recovery are examined. The results reveal that the formation of a solvent-oil mixing zone during the huff period and oil swelling and vaporization of oil components during the puff period are key mechanisms for enhancing oil recovery. Furthermore, the incremental recovery factor (RF) increases with injection pressure, even when the injection pressure exceeds the minimum miscibility pressure (MMP), implying that MMP may not play a critical role in the design of HnP in ultratight reservoirs. The results suggest that injecting solvents after a sufficient primary production period is more effective, allowing reservoir pressure depletion. Injecting the solvent without enough primary production may result in significant production of the injected solvent. The results show that the solvent-oil mixing zone expands, and the solvent recycling ratio decreases as soaking time increases. However, short soaking periods with higher HnP cycles are recommended for improving oil recovery at a given time frame. Finally, CO2 HnP outperforms CH4 or N2 HnP due to the higher ability of CO2 to extract a larger amount of intermediate and heavy components into the vapor phase, which has higher transmissibilities as compared with the liquid phase.