Adsorption and diffusion of shale oil on kerogen in CO2 flooding: molecular dynamics and NMR studies of temperature and multi-component effects

The adsorption and diffusion behavior of shale oil within kerogen nanopores represents a core scientific issue for improving the efficiency of CO 2 -enhanced oil recovery in shale reservoirs. Temperature regulates competitive adsorption between light and heavy components in distinct manners, yet the underlying mechanism remains unclear at molecular and pore scales. Molecular dynamics (MD) simulations are performed to explore adsorption and diffusion characteristics of single-component alkanes (C 7 H 16 , C 12 H 26 , C 18 H 38 ) and multi-component mixtures on kerogen surfaces within a temperature range of 303–403 K. Nuclear magnetic resonance (NMR) tests are carried out to verify pore-scale oil displacement responses under the same temperature gradient. Simulation results indicate that long-chain alkanes form highly ordered adsorbed packing structures at low temperatures. In multi-component systems, size-complementary filling increases the mass density by 20–40% relative to single components, whereas intermolecular entanglement reduces the diffusion coefficient by approximately 80%. Experimental results from NMR tests confirm that oil recovery presents a consistent positive trend with the diffusion coefficient obtained from MD simulations. A quantitative relationship is established among temperature, alkane chain length, and displacement efficiency. Clear differences in temperature sensitivity are identified between light oil and heavy oil systems. A temperature-differentiated CO 2 flooding strategy is accordingly proposed: 383–403 K for light oil reservoirs and 363 K for heavy oil reservoirs. This work bridges molecular‑scale dynamics and pore‑scale experiments, providing quantitative insights into the nonlinear coupling between chain length and temperature and offering an experimentally grounded temperature strategy for CO 2 -enhanced oil recovery in shale reservoirs.