[mech-engineer]

Hi every all
I'm simulating Eulerian two-phase flow (dispersed solid-continuous fluid) in a microchannel heat sink. Because of presence of nano-particles in modeling, I should consider Brownian motion at simulation, but CFX hasn't any option about it. Please help me in this matter and tell me how apply Brownian motion at Eulerian two-phase flow. Also, tell me Brownian motion in the which group of equations (continuity,momentum and energy) and which phase (dispersed solid or continuous fluid) can add?

[ghorrocks]

CFX does not include Brownian motion as the multiphase models assume the flow is mixed at scales far larger than the atomic scale. Brownian motion is only apparaent at atomic scales. Sounds like you are using the multiphase model in a flow regime it is not intended for.

Can you model the Brownian motion with diffusion?

[mech-engineer]

I want to add Brownian motion as a source term to continuity equation of dispersed solid particles only,because Brownian motion is a diffusion mechanism.Is it possible to do this?

[ghorrocks]

As I said in my previous post, can you use diffusion to simulate Brownian motion?

If you are modelling nanoparticles I assume there is no relative slip between the particles and the fluid, but there is a diffusive process due to Brownian motion. This sounds better modelled as an additional variable with a convection/diffusion equation, where the diffusion is matched to the Brownian motion diffusion.

[mech-engineer]

Because of two-phase modeling, I should consider relative velocity and temp. between phases.Multiphase method is Eulerian-Eulerian and according to modeling guide of CFX 12.1,I want add Brownian motion as a source term to continuity equation of particle phase only.From your opinion, is it possible to do this?

[ghorrocks]

From your opinion, is it possible to do this?

Yes, but it would be tricky to implement as you would have to do it in a fashion which is conservative. This is certainly the hard way of doing it.

I do not think you have understood the points I have made previously.
1) You cannot use a multiphase model for an atomic scale process.
2) An atomic scale process will have no velocity slip or thermal differences between the "phases".
3) If you model this as an additional variable then you can use a trivial diffusion term to model the Brownian motion.

Is there some reason why an additional variable with diffusion is not a more appropriate way of modelling it?

[mech-engineer]

I don't know,but since the apply of Brownian motion is one of the most important parts of the project, the only way that comes to mind is considering of Brownian motion as a source term and adding it to continuity equation of particle phase only.Is it true?

[ghorrocks]

This conversation is getting very tedious. Please do some reading about multiphase models and Brownian motion as it does not appear that you understand what I am saying.