The Composition Dependence of Self and Transport Diffusivities from Molecular Dynamics Simulations

Using molecular dynamics (MD) simulations, we determine the composition dependence of the self-diffusivity and transport diffusivity of a methane/ethane mixture at high pressure. The transport diffusivity is generated from the self-diffusivities using the Darken equation. We perform a careful analysis of the molecular dynamics simulations and show that it is possible to reproduce the results in the microcanonical, canonical, and isobaric–isothermal ensembles. We demonstrate that in order to capture the sensitive dependence of the diffusivities on composition, it is necessary to run simulations with larger systems and for longer durations than is typical. We report the trends in the diffusivities as a function of composition, temperature, pressure, and density. We modify an existing empirical correlation, which when combined with a corresponding states chart, is capable of quantitatively reproducing the simulated diffusivity dependence on composition, temperature, pressure, and density. Finally, we quantify the effect that the choice of equation of state (EOS) used to evaluate the thermodynamic factor in the Darken equation has on the transport diffusivity.