Restricted CO2/CH4 diffusion in nanopores: A quantitative framework to characterize nanoconfinement effect of shale organic pore

•The self-diffusion coefficients of CH4/CO2 in shale organic nanopores were investigated.•A workflow combined Monte Carlo simulation with molecular dynamics method was proposed.•A novel nondimensional self-diffusion coefficient D* was defined to build up concise relation with Knudsen number (Kn).•Two-segmented correlation formula between D* and Kn was proposed.•The causation of binary mixture weakening fluid diffusion was interpretated by fluid-pore surface interaction. The utilization of carbon dioxide (CO2) to enhance depleted unconventional resources can reduce green-house gas emission and increase the recovery ratio. Self-diffusion coefficient of nanoconfined CO2/CH4 is a crucial factor in CO2-enhanced shale gas recovery and CO2 geological sequestration. A workflow was proposed which included the reconstruction of amorphous shale organic nanopores, detection of effective pore size, and integration of grand canonical Monte Carlo (GCMC) method with molecular dynamics (MD) simulations to calculate self-diffusion coefficient. Challenges such as complex pore surface and extreme reservoir condition were considered, and the self-diffusion coefficients of nanoconfined fluid were obtained. It was found that the self-diffusion coefficient of nanoconfined CO2/CH4 is much smaller than that of bulk CO2/CH4, as a concequence of nanoconfinement effect. The effects of temperature, pressure and pore size were considered, and a nondimensional self-diffusion coefficient was proposed to derive a concise relation with Knudsen number. Additionally, a two-segment correlation formula was proposed to describe the concise relation. Owing to the strong CO2-pore surface interaction, CO2 was adsorbed on the pore surface and CH4 was restricted in a smaller pore space. Therefore, this fluid-pore surface interaction reduced the self-diffusion coefficient of binary mixture CO2/CH4. The workflow, nondimensional self-diffusion coefficient, and correlation formula derived in this study can be used for extensive applications, such as catalyst, fuel cell and sewage treatment.