Molecular Dynamics Simulation of the Thermal Diffusion Effect in n-Alkane Binary Mixtures

FX Chen and H Nasrabadi, JOURNAL OF PHYSICAL CHEMISTRY B, 126, 10164-10171 (2022).

DOI: 10.1021/acs.jpcb.2c0512110164J

In oil and gas reservoirs, the thermal diffusion effect leads to compositional variations of hydrocarbon fluids in both horizontal and vertical directions. Compared with experimental methods, molecular dynamics (MD) simulations can cover a broader range of pressures and temperatures for the investigation of the thermal diffusion effect. However, previous MD simulations of the thermal diffusion effect for n-alkane binary mixtures have been limited to only nC5-nC10, nC6-nC10, and nC6-nC12 mixtures. In this work, for the first time, we perform a series of MD simulations on n-alkane binary mixtures, C1-C3, C1-nC4,nC7-nC12,nC7- nC16, and nC10-nCi (i = 5, 7, 8, 12, 14, 16), with different mole fractions and temperature and pressure conditions. The boundary-driven nonequilibrium molecular dynamics (BD-NEMD) with the enhanced heat exchange (eHEX) algorithm is used to generate the temperature gradient and measure the thermal diffusion effect. Additionally, a workflow for molecular simulations of thermal diffusion of n-alkane binary mixtures is proposed to ensure their repeatability and reliability. The errors for our MD simulation results are generally less than 10% compared with experimental data. Our results show that in the binary mixture, the heavy component tends to move to the cold region, while the lighter component tends to aggregate near the hot region, which is consistent with experimental observations.

Return to Publications page