(432b) Three-Dimensional Pore Networks in Miocene Stevens Sandstone, California: Implications for CO2 Injection and Trapping

The Miocene Stevens sandstone in San Joaquin basin of California is a promising target for CO₂ geological sequestration. Among the four trapping mechanisms that have been identified so far, structural trapping which relies on the impermeable caprock, and residual trapping which relies on the capillary effect of non-wetting fluid are the most important during the injection stage, which occurs up to several decades. Significant heterogeneity in reservoir properties has been noticed in the Stevens sandstone due to the complicated geology in Miocene sedimentation in San Joaquin basin. To investigate the impact of heterogenous and complicated pore structures on fluid flow and trapping, core samples from Stevens sandstone were scanned by high-resolution X-ray micro-computed tomography (micro-CT), a non-destructive imaging method.

A three-dimensional characterization of the pore networks of these samples was conducted. Three rock phases were identified, namely mineral grains, pores, and cement. The cement phase includes cement, matrix, and microporosity that can hardly be resolved by micro-CT scanning. The three-dimensional pore systems of samples were segmented, and porosity values were measured. Absolute permeability simulation was then conducted using Lattice Boltzmann method. An idealized pore network model (PNM) was extracted from each sample, in which pores and pore throats are simplified as spheres and cylinders respectively. A two-phase fluid-flow simulation with CO₂ and water was then conducted at reservoir temperature and pressure conditions. The pore system model was validated through comparison with lab measured porosity, particle sizes, and permeability. While the relative permeability is consistent with published works on a number of sandstone reservoirs.

The tested samples demonstrated diverse reservoir properties, with porosity ranges from 15.3% to 19.3%, and permeability ranges from 0.55 md to 271 md. Several wettability scenarios were simulated, including strong water-wet, intermediate water-wet, mixed-wet with 25% CO₂-wet, and mixed-wet with 50% CO₂-wet. Initial and residual CO₂ saturations were simulated for each scenario.

Results indicate that microporosity is widely distributed in the Stevens sandstone, which can contribute up to 30% of the total porosity, yet it hardly contribute any to the permeability of a sample. Initial CO₂ saturation is largely controlled by the irreducible water saturation which is decided by the pore and pore throat sizes of pore system. In addition, Wettability exerts major influence on the residual trapping saturation. Although positively correlated with initial CO₂ saturation during primary drainage, a mixed-wet system provides the highest residual trapping capacity. Lastly, for heterogeneous sandstone reservoirs, injected CO₂ will likely form layered flow that first saturates the high-permeability beds.