Binary gas diffusion in microchannel

Shaw · August 11, 2014, 04:03

Hello everyone,
I'm new to CFX (and never used any CFD software before), but my case seems quite simple.
I'd like to mix two gases, methanol and oxygen in a microchannel (2 inlets and 1 outlet). I've searched lots of discussion related to my case and tried so many boundary condition settings, but still can't get a reasonable result.

Here are my BCs, hoping someone can tell me where is the error from.
Steady State
Laminar flow
Isothermal, fluid temp.= 400 [k]
No-slip wall
Interfase Len. Scale = 1e-20 [m]
Drag coefficient = 0
Oxygen - ideal gas
I set CH3OH as an additional variable(specific), kinematic diffusivity (for O2) = 2.78e-5 [m^2s^-1] (Should I set the same value for CH3OH itself? )

Inlet1
volume fraction of CH3OH=1, O2=0
mass flow rate = 4.99e-8[kg s^-1]
AV Value = 0 [kg s^-1]

Inlet2
volume fraction of CH3OH=0, O2=1
mass flow rate = 4.39e-8[kg s^-1]
AV Value = 0 [kg s^-1]

Outlet
Relative Pressure = 0 [Pa] AV Value = 0 [kg s^-1]

In the simulation result, mixing length is too long compared to other people's research, and the volume fraction cross the mixing channel is strange. Changing the kinematic diffusivity or interfase len. scale wasn't much help.

Sorry for such a long description, if I can provide further detail please let me know.
Thanks!

ghorrocks · August 11, 2014, 07:45

You made the first mistake right at the beginning - this is not a multiphase simulation as all the fluids are in the same phase (ie gas). Instead this is a multicomponent mixture simulation.

Shaw · August 11, 2014, 09:43

Thank you Glenn,

I know this is multicomponet flow, but I think that setting the "Interfase Len. Scale" almost zero, "not" switching on homogeneous model, and using the additional variables is the way.
I don't know how to do this kind of simulation correctly, could you give me some suggestions? I would really appreciate it.
(I've seen one of the CFX tutorials, it's about a static mixer. But since there are two streams of water at different temperatures, maybe it's not like my case?)

ghorrocks · August 11, 2014, 09:49

Why are you trying to kludge your simulation into a multiphase model when it is not multiphase? And the correct model (multicomponent mixture) is much easier to set up, will be more stable and does not require any fudges?

Your model is similar to the static mixer with different temperatures in that the flow is convecting a scalar quantity. But in your case the quantities are different gases with different material properties. So the analogy is limited.

I would recommend doing all the CFX tutorials on multicomponent flows. There are a few of them. And doing the other tutorials will not be wasted time either, you will learn from all of them. If you have never done CFD before then you have a lot to learn before you do a complex simulation like a multicomponent mixture.

August 11, 2014, 04:03	Binary gas diffusion in microchannel	#1
Shaw New Member Join Date: Aug 2014 Posts: 3 Rep Power: 12	Hello everyone, I'm new to CFX (and never used any CFD software before), but my case seems quite simple. I'd like to mix two gases, methanol and oxygen in a microchannel (2 inlets and 1 outlet). I've searched lots of discussion related to my case and tried so many boundary condition settings, but still can't get a reasonable result. Here are my BCs, hoping someone can tell me where is the error from. Steady State Laminar flow Isothermal, fluid temp.= 400 [k] No-slip wall Interfase Len. Scale = 1e-20 [m] Drag coefficient = 0 Oxygen - ideal gas I set CH3OH as an additional variable(specific), kinematic diffusivity (for O2) = 2.78e-5 [m^2s^-1] (Should I set the same value for CH3OH itself? ) Inlet1 volume fraction of CH3OH=1, O2=0 mass flow rate = 4.99e-8[kg s^-1] AV Value = 0 [kg s^-1] Inlet2 volume fraction of CH3OH=0, O2=1 mass flow rate = 4.39e-8[kg s^-1] AV Value = 0 [kg s^-1] Outlet Relative Pressure = 0 [Pa] AV Value = 0 [kg s^-1] In the simulation result, mixing length is too long compared to other people's research, and the volume fraction cross the mixing channel is strange. Changing the kinematic diffusivity or interfase len. scale wasn't much help. Sorry for such a long description, if I can provide further detail please let me know. Thanks!

August 11, 2014, 07:45		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,871 Rep Power: 144	You made the first mistake right at the beginning - this is not a multiphase simulation as all the fluids are in the same phase (ie gas). Instead this is a multicomponent mixture simulation. Shaw likes this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Binary gas diffusion in microchannel	Shaw	Main CFD Forum	0	August 11, 2014 03:57
Moving mesh	Niklas Wikstrom (Wikstrom)	OpenFOAM Running, Solving & CFD	122	June 15, 2014 07:20
error message	cuteapathy	CFX	14	March 20, 2012 07:45
Error to re-open fluent case file	J.Gimbun	FLUENT	0	April 27, 2006 09:42
Gas pressure question	Dan Moskal	Main CFD Forum	0	October 24, 2002 23:02

August 11, 2014, 09:43		#3
Shaw New Member Join Date: Aug 2014 Posts: 3 Rep Power: 12	Thank you Glenn, I know this is multicomponet flow, but I think that setting the "Interfase Len. Scale" almost zero, "not" switching on homogeneous model, and using the additional variables is the way. I don't know how to do this kind of simulation correctly, could you give me some suggestions? I would really appreciate it. (I've seen one of the CFX tutorials, it's about a static mixer. But since there are two streams of water at different temperatures, maybe it's not like my case?)

August 11, 2014, 09:49		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,871 Rep Power: 144	Why are you trying to kludge your simulation into a multiphase model when it is not multiphase? And the correct model (multicomponent mixture) is much easier to set up, will be more stable and does not require any fudges? Your model is similar to the static mixer with different temperatures in that the flow is convecting a scalar quantity. But in your case the quantities are different gases with different material properties. So the analogy is limited. I would recommend doing all the CFX tutorials on multicomponent flows. There are a few of them. And doing the other tutorials will not be wasted time either, you will learn from all of them. If you have never done CFD before then you have a lot to learn before you do a complex simulation like a multicomponent mixture.