Answers are grounded in this paper and cite their sources.
Diffusion of CO2 and hydrocarbons in shale nanopores controls mass transfer during CO2-enhanced oil recovery. A finite-duration grand canonical Monte Carlo (GCMC)-assisted preparation protocol followed by canonical-ensemble molecular dynamics is used to investigate molecular distributions and in-plane self-diffusion of CO2/n-decane mixtures in 4, 7, and 10 nm hydrophobic slit pores at 344 K. Varying the relative exchange-attempt frequencies produced a protocol-defined composition series whose molecular populations are then held fixed. In-plane diffusion is obtained from two-dimensional molecular center-of-mass (COM) mean-square displacements relative to the mass-weighted COM of the entire mobile fluid; three independent trajectories are analyzed per condition. Both species exhibited pronounced interfacial layering. The fraction of n-decane united-atom interaction sites in the predefined wall regions decreased in the order 4 nm > 7 nm > 10 nm, reflecting the greater proportional contribution of wall-dominated regions in narrower pores. Fluid-COM referencing reduced the long-time mean-square displacements and between-trajectory variability, particularly for n-decane. CO2 diffused faster than n-decane under all conditions, and across the prepared composition series, the diffusivities of both species generally increased with pore width and decreased with increasing n-decane mass fraction. At the highest n-decane fractions, initially wall-associated populations are less mobile than interior populations in the 7 and 10 nm pores, whereas regional effective diffusivities are comparable within statistical uncertainty in the 4 nm pore, consistent with overlapping influences of the opposing walls. Overall, wider pores and CO2-rich prepared states are associated with greater molecular mobility, whereas narrow pores and wall-associated environments restricted n-decane transport under the hydrophobic surface condition considered here.
No takes yet. Share an insight, caveat, or question.
Chen et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: