Simulating Multi-Component Systems to Study Turbulence

Abstract

In the inaugural editorial of this journal, it was suggested that nonelastic collisions of molecules may modify the traditional continuum model of a turbulent fluid. Non-elastic collisions of molecules violate the historic assumption that molecular properties do not matter for as long as viscosity and density are used as macroscopic parameters, contributing to the dimensionless Reynolds number. In fact, inelastic collisions presume the existence of quantum states of molecules. An energetic gas consists of molecules in various quantum states. The gas is therefore composed of different components, each component belongs to an excited state, and the gas may be considered multi-component. An accurate physical description will consist of different species of what used to be considered a single specie of unexcited molecules, as in the Navier-Stokes equation. Note that traditional numerical simulations only assume a single specie. The purpose of this guest editorial is to suggest that if computational fluid dynamics admits the existence of several interacting species, very rich results, including turbulence, will be found.