International Journal of Heat and Mass Transfer, Vol.114, 47-61, 2017
Flow of a monatomic rarefied gas over a circular cylinder: Calculations based on the ab initio potential method
Two-dimensional flows of argon and helium over a circular cylinder are calculated by the Direct Simulation Monte Carlo (DSMC) method in the range of the free stream Mach number Ma(infinity) from 0.5 to 10 in nearly free molecular, transitional, and nearly continuum flows. In DSMC simulations, interparticle collisions are calculated with the ab initio (Al) potential method based on the interatomic interaction potentials established in quantum mechanical calculations. It is shown that the Al potential method enables computationally efficient simulations of multidimensional rarefied gas flows without introducing semi-empirical models of collision cross sections. The calculated values of the drag CD and heat flux C-Q coefficients of the cylinder for Ar and He at the same values of Ma(infinity), rarefaction parameter, and surface-to-free-stream temperature ratio are found to be different in less than 1%, ensuring small sensitivity of C-D and C-Q to the species of a monatomic gas. The simulation results obtained with the Al potential method are systematically compared with results obtained with the hard sphere (HS) molecular model. It is found that the choice of the HS diameter based on the condition of the identical viscosity of the real and HS gases at the free stream temperature ensures calculations of C-D and C-Q in sub- and supersonic flows at Mac(infinity) <= 2 with errors less than 3% and 6.5%, correspondingly. For hypersonic flows, this choice of the HS diameter is unsatisfactory and results in the errors up to 7% in C-D and 28% in C-Q. A semi-empirical rule that defines an optimum HS diameter in super- and hypersonic flows is suggested. With the use of this rule, the HS model is capable of predicting C-D and C-Q with errors less than 1% and 3%, correspondingly, and also provides a good accuracy in calculations of local flow parameters. (C) 2017 Elsevier Ltd. All rights reserved.