Evaluating the Pressure-Dependent Equivalent Permeability Evolutions for Shale Matrix: Experiments and Modeling

The ability to model and predict matrix permeability changes during reservoir depletion is critical for accurate analysis of long-term production performance in unconventional gas reservoirs (UGRs), including shale gas and coalbed methane reservoirs. Yet, flow quantification in the nanoporous matrix is still challenging due to the complex pore structure and morphology. To understand the pressure-dependent matrix permeability evolution, this study conducted laboratory permeability measurements using pulverized samples. Equivalent permeability was estimated from the pressure decay profile for the Devonian shale sample. A novel experimental system, a differential volumetric unit, has been established and applied to capture the accurate transient gas flows for the shale sample. The measured permeability of shale exhibited overall decreasing trends with pressure depletion. Due to the presence of slip flow and Knudsen diffusion, low-pore-pressure data did not follow the same decline trend fitted by high-pore-pressure data as observed in the shale sample. This study also utilized methane as the invasion fluid to examine the adsorption effect on matrix permeability, whose value could be up to 40% lower than permeability without correction for adsorption because of the condensation of adsorbed phase at pore surface occupying available pore space. Since these tight rock matrixes are composed of micro- and nanopores, matrix permeability is primarily related to pore structure (e.g., the pore size distribution, porosity and tortuosity). Low-pressure N2 adsorption was conducted to characterize the complex pore structure of the Marcellus shale sample. A multimechanic model was proposed to predict the pressure-dependent matrix permeability based on pore structure information and investigate the effect of gas adsorption on apparent permeability. This model has successfully linked the realistic, complex pore structure with the pressure-dependent matrix permeability of shale and coal. The proposed model could be coupled into the commercially available simulator to forecast long-term production profiles for UGRs wells.