Impact of Capillary Number, Fluid Viscosity Ratio, and Fracture Closure on Two-Phase Flow Regimes in Geological Fractures: An Experimental Study

Accurately predicting fluid-fluid interface displacement in fractured reservoirs is paramount for optimizing subsurface operations, particularly in the context of enhanced oil recovery and geological carbon sequestration (GCS). However, a comprehensive understanding of two-phase flow behavior in fractures, including the impact of fracture closure, fluid viscosity ratio, and capillary number, is yet to be achieved. To address this challenge, we have developed an analog experimental setup to investigate the intricate relationship between fracture surface roughness and fluid-fluid interface displacement. Our experimental setup features a transparent fracture flow cell with self-affine rough-walled surfaces that are matched to each other above a chosen length scale (denoted below as the correlation length) and a precisely controlled mean aperture. Realistic synthetic fracture geometries were generated numerically. They are characterized by their Hurst exponent, fracture closure, and correlation length. High-speed imaging captures the dynamic spatial distribution of fluid phases within the fracture plane during drainage processes in a given fracture geometry. The mean aperture can be varied between experiments for a given geometry of the fracture walls. We investigate a comprehensive range of capillary numbers, spanning both viscous and capillary-dominated regimes, vary viscosity ratios, and characterize the resulting displacement regimes. Our results reveal a profound impact of fracture closure and correlation length on trapping efficacy, particularly in the capillary-dominated regime. These findings can be interpreted in terms of residual trapping of CO2 during GCS in fractured reservoirs.