School of Engineering and Materials Science
Research Student Awards
PhD Thesis: Direct Simulation Monte Carlo of non-equilibrium rarefied flows
Author: ALVES, N
Supervisor(s): John Stark
Detailed studies of rarefied gases are of importance to space travel industry due to the requirement to understand the flow physics around the spacecraft during its re-entry to Earth. The low-density nature of these gases is such that continuum laws become invalid so the flow physics is underpinned by microscopic events.
The Direct Simulation Monte Carlo (DSMC) is a simulation technique suitable to solve engineering problems under rarefied conditions. This technique adopts a discrete particle approach to model the molecular processes prescribed by kinetic theory, and it follows a statistical mechanics approach to determine the macroscopic effects. DSMC is a computationally efficient method and it has been validated against atomic near-equilibrium experimental data. However, the lack of non-equilibrium laboratory data has precluded the validation of the method when strong flow gradients exist
The objective of this research is to assess the performance of DSMC in non-equilibrium conditions for the case of Nitrogen gas. This assessment focused on three areas: collision mechanics, transport processes and high-temperature processes.
A parametric study was undertaken on the binary collision mechanics and it was found that the DSMC simulated macroscopic properties are insensitive to the selected scattering law. Moreover, this law has no impact on the physical outcome of a collision, and the physics in conventional molecular models is governed by the adopted viscosity-temperature relationship. An investigation followed on DSMC transport processes in strong non-equilibrium flows and these were compared with a solution of Boltzmann equation. It was observed that when strong flow gradients are present momentum transfer is well compared, but significant errors are observed in energy transfers.
Finally, a study on DSMC calculated transport coefficients in high-temperature near-equilibrium flows was performed, and this was compared to available theoretical data. It was found that as temperature increases, chemical reactions processes appear not to be well represented by DSMC.
It is concluded the molecular collision physics represented by a viscosity-temperature law followed a separate model for thermo-chemistry leads to inconsistencies in the prediction of more complex flows. Moreover, the binary collision mechanics may not be appropriate for representation of recombination processes.