Pore-scale fluid dynamics in heterogeneous porous media: Insights from numerical simulation and nuclear magnetic resonance

Shale oil reservoirs contain complex pore networks where fluid distribution and displacement are governed by intricate capillary and wettability effects. This study combines pore-scale computational fluid dynamics (CFD) simulations and laboratory-scale experiments to investigate fluid flow and oil recovery behavior in shale formations. Using the Volume of Fluid (VOF) method, we simulate two-dimensional displacement in porous media with varying heterogeneity and wettability. Results show that uniform pore structures exhibit a broader swept region, whereas non-uniform structures promote preferential flow along high-permeability pathways, leading to lower recovery. Experimental validation was performed using high-temperature, high-pressure static imbibition tests on shale core samples from four lithofacies regions, supported by two-dimensional nuclear magnetic resonance (2D NMR) and interfacial tension (IFT) measurements. CO₂ and surfactant-enhanced fluids demonstrated superior oil mobilization in organic matter pores, while water-based fluids primarily recovered oil from clay-bound micropores. An energy-based analysis revealed that capillary-driven imbibition accounts for over 70% of total recovery, surpassing elastic expansion and gas dissolution contributions. IFT measurements confirmed the effectiveness of low-IFT fluids in altering wettability and improving pore accessibility. Together, the integrated CFD-experimental approach offers mechanistic insights into pore-scale flow dynamics and recovery processes in heterogeneous shale systems, providing guidance for the design of targeted enhanced oil recovery (EOR) strategies.