Learning Forum

[Eliana]

Hi, 

I'm trying to simulate a species transport through a selective membrane, for example methane-water vapor mixture, The geometry is a pipe, one velocity inlet for the mixture, one pressure outlet for the methane and I'm trying to simulate water vapor being adsorbed on the internal membrane surface and desorbed on the external membrane surface out of the pipe. How can I setup this right?

I'm using pressure based solver.

[RK]

Hello, 

With just this top level information, I would suggest using a pressure based solver with velocity inlet. 

[Eliana]

Hi Rahul, I've updated my description including your input.

Regards, Eliana

[DrAmine]

you require UDF to model the membrane like Boundary condition. Please add more details.

[Eliana]

Hi Amine,

I'm trying to set the source terms (mass, momentum, water) at the cell zone: fluid-interior pipe, fluid-membrane (porous zone), fluid-exterior pipe. Is this procedure correct? If not, please give me more details about UDF.

Thanks and regards,

Eliana

[DrAmine]

Yes the procedure is correct to connect both sides the interior boundary

[Eliana]

ok thanks, so this is working fine when I simulate just one pipe. When I increase the number of pipes, how can I simulate only water flow through the membranes? Results of the simulation show methane concentration different from zero outside the pipes. How to set the source terms properly? Moreover, I cannot see representative results about the pressure drop between the inlet and outlet sections, how can I solve that?

Thanks and regards,

Eliana

[Rob]

Please post some images. 

[Rob]

You need to post images in line, staff aren't permitted to open or download attachments. 

For the source terms you'll need to limit the transfer to water via a mix of the UDF commands and how you hook it into the species and mass fields in the cell source term. 

[Eliana]

Thank you for your answer, this is the geometry I'm considering.

[Eliana]

[DrAmine]

I do not think that is straightforward here to resolve the porous membrane. You need to mode it either as thin membrane (porous jump) or porous zone. Now the task will be to have some source terms which will act on both side of the thin membrane so that you transfer mass fraction from one side to another side and just block the passage of water vapor. I guess the implementation here is easier with porous jump.

 

If the filter cannot be modeled as thin membrane (which itself requires UDF skills)  then more work and more thoughts are required.