Strain Rate Dependence of Heat Transfer as Applied to Planar Poiseuille Flow
Principal Investigator
Denis J. Evans
Research School of Chemistry
If a fluid is forced to flow between two stationary flat plates the fluid will develop a velocity gradient between the plates. Conventional hydrodynamics predicts the velocity profile will be quartic in the separation distance. As the fluid flows viscous heating will also occur. This project aims to study viscous heating in some detail.
Co-Investigators
Billy D. Todd
Karl Travis
Peter Daivis
Research School of Chemistry
Projects
r61 - VPP, PC
What are the results to date and the future of the work?
This project was initiated to investigate the coupling of the heat flux vector to the strain rate in a simulation of planar Poiseuille flow. The original aims of the project have now been achieved, and a coupling of the heat flux to the gradient of the square of the strain rate tensor has been established. The work has been accepted for publication in Physical Review E, ('Temperature Profile for Poiseuille flow', 'Departure from Navier-Stokes hydrodynamics in confined fluids'). Another paper is to appear in Physica A ('Poiseuille Flow of Molecular Fluids') while a fourth is to appear in Journal of Chemical Physics ('A study of viscosity inhomogeneity in porous media').
The next phase of this project involves a closer examination of the coupling coefficient x, which couples the strain rate to the heat flux. In particular, we need to investigate the dependence of this coefficient on the size of the simulation cell, as well as examine the behaviour of x as a function of thermodynamic state point. These simulations are extremely time consuming to perform, and will require literally thousands of hours of supercomputer time.
What computational techniques are used?
The basic technique employed is nonequilibrium molecular dynamics. We simulate atomic and molecular liquids sandwiched between molecular walls. We also plan to calculate the coupling coefficient from the so-called sinusoidal tranverse force algorithm using the TTCF formalism. This should improve efficiency enormously for weak fields.