[jiguozhao]

Dear all:
I was trying to simulate gas going across membrane process. Nitrogen goes from inlet and pass across membrane, at last permeates out from outlet. I tried several boundary conditions, like pressure inlet and flow rate outlet, velocity inlet and pressure outlet. However in all the conditions, there is a error called overflow appearing. Meanwhile another two notice are inlet blocked gas went out, and outlet blocked gas went into. This is very wierd because we take it for granted that inlet only let gas go in rather than go out.
My guessing is the membrane is too dense with small permeability, so the feed can't go through membrane and gas accumulates too much, resulting in a tendency to go out through inlet. In the permeate side the gas is so rare that the gas outside want to goes into the system from outlet.
However after I increased the permeability of the membrane (porous media), all the problems disappeared, and that time I set velocity inlet and 0 pressure outlet. Actually the permeability is 5.25E-19 m^2, but I changed it as 5.25E-12 m^2 which is far from reality. Although I succeeded in getting results from the simulation, I didn't use the true value.
What I need is a simulation close to reality, but when I input the true value of permeability, the software refuses to work. How to solve this problem?
Thanks in advance!
Best regards

Ji Guozhao
email: g.ji@uq.edu.au

[bfadewole]

Originally Posted by jiguozhao

Dear all:
I was trying to simulate gas going across membrane process. Nitrogen goes from inlet and pass across membrane, at last permeates out from outlet. I tried several boundary conditions, like pressure inlet and flow rate outlet, velocity inlet and pressure outlet. However in all the conditions, there is a error called overflow appearing. Meanwhile another two notice are inlet blocked gas went out, and outlet blocked gas went into. This is very wierd because we take it for granted that inlet only let gas go in rather than go out.
My guessing is the membrane is too dense with small permeability, so the feed can't go through membrane and gas accumulates too much, resulting in a tendency to go out through inlet. In the permeate side the gas is so rare that the gas outside want to goes into the system from outlet.
However after I increased the permeability of the membrane (porous media), all the problems disappeared, and that time I set velocity inlet and 0 pressure outlet. Actually the permeability is 5.25E-19 m^2, but I changed it as 5.25E-12 m^2 which is far from reality. Although I succeeded in getting results from the simulation, I didn't use the true value.
What I need is a simulation close to reality, but when I input the true value of permeability, the software refuses to work. How to solve this problem?
Thanks in advance!
Best regards

Ji Guozhao
email: g.ji@uq.edu.au

First of all, what membrane simulation software are you using? Then secondly, I do not quite have a good understanding of your problem. Are you saying that the simulator does not work because it does not accept what you think should be the permeability rates? The units for your perm rates are not correct and therefore you might be missing something with inconsistent units.
Are you just passing only Nitrogen as a single component gas through the membrane or you want to permeate Nitrogen our of a multi-component gas system? You obviously would not just for no reason be flowing a single gas though a separation media, it makes no sense. I am not a CFD guy but I have some experience with membranes.
Thanks

[jiguozhao]

Originally Posted by bfadewole

First of all, what membrane simulation software are you using? Then secondly, I do not quite have a good understanding of your problem. Are you saying that the simulator does not work because it does not accept what you think should be the permeability rates? The units for your perm rates are not correct and therefore you might be missing something with inconsistent units.
Are you just passing only Nitrogen as a single component gas through the membrane or you want to permeate Nitrogen our of a multi-component gas system? You obviously would not just for no reason be flowing a single gas though a separation media, it makes no sense. I am not a CFD guy but I have some experience with membranes.
Thanks

Dear Francis:
Thank you for your reply very much!!
To answer your question, the software i used is cfx.
secondly, the permeability built in cfx is a parameter in Darcy's Law (unit is m^2) rather than the permeability in gas separation (mol m^-2 s^-1 Pa^-1).
actually i am just starting this area, so i would better go step by step. therefore i simulate single gas to compare the single gas permeace experiments. for single gas we closed the retentate outlet, and just left permeate outlet open.
I am really happy to meet you by internet, and feel grateful of your enthusiastis reply. can you tell me your email address? i would like to communicate further, because i think it must be helpful for me in the future.
Many thanks!!
cheers
Ji Guozhao
g.ji@uq.edu.au