At Knudsen numbers >> 1 the continuum assumption of CFD breaks down as the flow exhibits particle type behaviour. Does anyone know what Knudsen numbers are typical in MEMS? Have people adapted CFD methods to be MEMS applicable? Have they had to?

Any pointers would be welcome.

Fred.

Fred,

I think the answer to your first question is that it varies. It all depends on the size of the channel and the state of the fluid being studied. If you are approaching the non-continuum regime, various people have done work extending continuum equations to this regime by modifying the boundary conditions (i.e. temperature jump and velocity slip at the wall as a function of an "accomodation coefficient").

The answer to the second question is yes. People are doing CFD for MEMS devices. I know that CFD Research Corporation is doing quite a bit of work in this area and has several sofware packages specifically designed to work with MEMS devices. See their web page at www.cfdrc.com.

Mark,

Great answer, thanks. Do you know how exactly fluid thermal variation effects a tailored CFD approach to high Knudsen number simulations?

Cheers Fred.