Simulation modeling of the transport coefficients for rarefied gases and gas nanosuspensions

Simulation of transport coefficients is very important from a practical point of view. The only method for simulation of the transport coefficients of dense gases and liquids is the molecular dynamics method. However, this method is not applicable for a rarefied gas due to the need to use a great number of molecules. This paper proposes an alternative simulation method of the molecular modeling of rarefied gas transport coefficients. In this approach, the phase trajectories of considered systems are simulated stochastically. The actual values of the transport coefficients are obtained using the corresponding Green–Kubo relations by averaging over a large number of phase trajectories. To test the developed algorithm, a set of problems was solved. The binary diffusion coefficients for noble gases (Kr-Ar, Xe-Ar, Xe-Kr), the viscosity coefficients for monatomic and polyatomic gases (Ar, Kr, Ne, Xe, CH$_4$, CO, CO$_2$, O$_2$), and the diffusion coefficient for nanoparticles in rarefied gases were simulated and analyzed. It was shown that the algorithm accuracy of the order of 1–2% could be achieved when using a relatively small number of molecules. The dependence of the accuracy on the number of molecules, statistics (the number of phase trajectories), and calculation time were analyzed.