Abstract
We have employed a recently developed nonequilibrium molecular dynamics algorithm to compute the thermal diffusion factor, thermal conductivity, and heat of transfer of the artificially binary Lennard-Jones mixtures composed of particles with equal mass, but different potential parameters. The thermal diffusion factor has been shown to be quite sensitive to variations in the particle size and energy well depth, and can be uses as a convenient tool for testing of accuracy of the potential energy function. We show that the thermal diffusion factor strongly increases with increasing the size ratio σ11/σ22 or the energy well-depth ratio ε11/ε22, when the size or energy well-depth ratio is about less than one, and decreases with increasing the size or the energy well-depth ratio, when the size or energy well-depth ratio is about larger than one. We also observed that the thermal conductivity is a strong monotonic decreasing function of the size and energy well-depth ratios.
