Near-wall effects in rarefied gas micro-flows: some modern hydrodynamic approaches

Methods for simulating the critical near-wall region in hydrodynamic models of gas micro-flows are discussed. Two important non-equilibrium flow features – velocity slip at solid walls, and the Knudsen layer (which extends one or two molecular mean free paths into the gas from a surface) – are inves...

Full description

Saved in:
Bibliographic Details
Published inThe International journal of heat and fluid flow Vol. 28; no. 1; pp. 37 - 43
Main Authors O’Hare, Lynne, Lockerby, Duncan A., Reese, Jason M., Emerson, David R.
Format Journal Article Conference Proceeding
LanguageEnglish
Published New York, NY Elsevier Inc 01.02.2007
Elsevier Science
Subjects
Applied fluid mechanics
Exact sciences and technology
Fluid dynamics
Fluidics
Free molecular flows
Fundamental areas of phenomenology (including applications)
Gas microsystems
Knudsen layer
Microfluidics
Physics
Rarefied gas dynamics
Slip flows
Velocity slip
Wall-function
Gas microsystems
Knudsen layer
Velocity slip
Microfluidics
Rarefied gas dynamics
Wall-function
Slip flow
Rarefied gas flow
Digital simulation
Modelling
Wall effects
Microstructure
Online AccessGet full text

Cover

More Information
Summary:Methods for simulating the critical near-wall region in hydrodynamic models of gas micro-flows are discussed. Two important non-equilibrium flow features – velocity slip at solid walls, and the Knudsen layer (which extends one or two molecular mean free paths into the gas from a surface) – are investigated using different modelling approaches. In addition to a discussion of Maxwell’s slip boundary condition, a newly implemented ‘wall-function’ model that has been developed to improve hydrodynamic simulations of the Knudsen layer is described. Phenomenological methods are compared to physical modelling and it is shown that, while both simulation types have merit, and both can quantitatively improve results in most cases, there are drawbacks associated with each approach. Phenomenological techniques, for example, may not be sufficiently general, whilst issues with applicability and stability are known to exist in some physical models. It is concluded that, at present, neither approach is unambiguously preferable to the other, and that both physical and phenomenological modelling should be the subject of future work.
ISSN:0142-727X
1879-2278
DOI:10.1016/j.ijheatfluidflow.2006.04.012