Problems with a direct simulation Monte Carlo method applied to the shock structure in a binary gas mixture

An attempt is made to explain why, in applying Bird's (1968, 1976) method to the shock structure in a binary mixture of inert noble gases, under certain conditions the temperatures of the components do not relax to the common equilibrium value downstream of the shock wave. This phenomenon is observed especially when the number of molecules per cell of the simulation procedure is small, and may be due to a distorted statistical balance in connection with the collision calculation procedure. An examination of the numerical simulation method and a comparison with experimental results shows that, on the basis of the shock thickness and the component separation in the shock, the number of molecules per cell has a certain influence on the calculated shock profiles in the shock Mach number range 2-6 and for binary Ar-He mixtures from 3 percent up to 97 percent Ar.