Modeling of Immiscible, Two-Phase Flows in a Natural Rock Fracture

One potential method of geologically sequestering carbon dioxide (CO₂) is to inject the gas into brine-filled, subsurface formations. Within these low-permeability rocks, fractures exist that can act as natural fluid conduits. Understanding how a less viscous fluid moves when injected into an initially saturated rock fracture is important for the prediction of CO₂ transport within fractured rocks. Our study examined experimentally and numerically the motion of immiscible fluids as they were transported through models of a fracture in Berea sandstone. The natural fracture geometry was initially scanned using micro-computerized tomography (CT) at a fine volume-pixel (voxel) resolution by Karpyn et al. [1]. This CT scanned fracture was converted into a numerical mesh for two-phase flow calculations using the finite-volume solver FLUENT^® and the volume-of-fluid method. Additionally, a translucent experimental model was constructed using stereolithography. The numerical model was shown to agree well with experiments for the case of a constant rate injection of air into the initially water-saturated fracture. The invading air moved intermittently, quickly invading large-aperture regions of the fracture. Relative permeability curves were developed to describe the fluid motion. These permeability curves can be used in reservoir-scale discrete fracture models for predictions of fluid motion within fractured geological formations. The numerical model was then changed to better mimic the subsurface conditions at which CO₂ will move into brine saturated fractures. The different fluid properties of the modeled subsurface fluids were shown to increase the amount of volume the less-viscous invading gas would occupy while traversing the fracture.