Pore-scale Numerical Study of Spontaneous Imbibition in 3D Granular Porous Media under Different Boundary Conditions

Spontaneous imbibition is a fundamental recovery mechanism in tight oil reservoirs with natural or artificial fractures. Core-scale laboratory/numerical studies have extensively demonstrated that spontaneous imbibition is significantly influenced by boundary conditions and rock structures. However, due to the limitations of experimental conditions and the complexity of porous media, it is challenging for these studies to focus systematically on the effects of individual geometric parameters on spontaneous imbibition, especially at the pore scale. To overcome these limitations and further understand the pore-scale flow mechanism of spontaneous imbibition under different boundary conditions, we developed a numerical simulation based on the lattice Boltzmann method to directly simulate the process of spontaneous imbibition in 3D synthetic grain packs. The porous media model in this study is obtained by simulating the same number of solid grains packing in a square container with the discrete element method. All models have the same porosity and grain size distribution, but the grain shape and pack mode are significantly different. Then these models are applied in spontaneous imbibition simulation under three boundary conditions, including one faces open (OFO), two faces open (TFO) and all face open (AFO) boundary conditions. Through a total of 30 sets of simulation results, we analyze systematically the influence of grain shape and heterogeneity on the oil-water displacement dynamics, remaining oil distribution patterns and oil recovery rate in the process of spontaneous imbibition under different boundary conditions. Results show that, no matter what boundary conditions, grain shapes have a more influence on the pore-scale dynamics of oil-water displacement but less impact on oil recovery rate during spontaneous imbibition. However, the heterogeneity, characterized by the uniformity of grain distribution, has different impacts on both flow dynamics and patterns and amounts of oil trapped, depending on the boundary conditions. The results obtained can improve the most fundamental understanding of spontaneous imbibition for enhancing oil recovery.