[Bruno Machado]

Hi everyone,

My PhD project involves the simulation of a fuel cell. I built a model using UDF and it works ok. The reason it is ok is that my species concentration gradient of species is very small within the "flow channel/gas diffusion layer/catalyst layer".

I am considering a (compressible) mass-flow-inlet boundary condition where I define the mass flow rate, temperature, mass fraction of species. In the outlet, a pressure-outlet boundary is defined and backflow temperature and species are defined as well. Is this the best boundaries for this case?

Does anyone worked with UDF and reactions? Anyone had the same problem?
Anything that could help me fix this issue are more than welcome.

Thanks all.

[maverick123]

@ Bruno : As a single fuel cell channels c/s is very very small, laminar flow considerations would be better.

[Bruno Machado]

Originally Posted by maverick123

@ Bruno : As a single fuel cell channels c/s is very very small, laminar flow considerations would be better.

Dear Maverick, thank you for your response.

I am considering a laminar flow in the flow channel. Even when considering a single fuel cell channel, the inlet is adjusted to change accordingly to the stoichiometric ratio for anode and cathode. In literature there are many studies where, even for a single channel, the concentration difference between inlet and outlet changed considerably( let's say about half for a stoichiometric number 2).

What I see now is that my electronic and ionic potential are very close to each other, so the overpotential is very small (it is bigger near the interface of the catalyst with the membrane and goes near 0 through the catalyst layer). Any comments on that? I am simulation an Anion Exchange Membrane Fuel Cell and my equations for electronic and ionic potential potential are:

-grad(diff_ele.grad(ele_pot)) = S_ele
S_ele = - j_a (Butler Volmer) if anode catalyst layer
S_ele = j_c (Butler Volmer) if cathode catalyst layer

-grad(diff_ion.grad(ion_pot)) = S_ion
S_ion = j_a (Butler Volmer) if anode catalyst layer
S_ion = - j_c (Butler Volmer) if cathode catalyst layer

Anything comes to your attention?
Any help or comments are welcome.

Regards

[maverick123]

Hi Bruno,

You are quite right that there will considerable difference between H2 concentration on Anode side (I checked for Proton Exchange Membrane Fuel cell) . Are you using Fluent Fuel Cell Module ? You said, you are simulating Anion Exchange Membrane Fuel Cell. Is it alkaline AEMFC ?

Best Regards,

[Bruno Machado]

Originally Posted by maverick123

Hi Bruno,

You are quite right that there will considerable difference between H2 concentration on Anode side (I checked for Proton Exchange Membrane Fuel cell) . Are you using Fluent Fuel Cell Module ? You said, you are simulating Anion Exchange Membrane Fuel Cell. Is it alkaline AEMFC ?

Best Regards,

Hi Maverik123,

I am not using the Fuel Cell module, I built an entire code to simulate an alkaline AEMFC.

[maverick123]

Hi Bruno,

Apart from the two potential equations, what are the other equations?
You may need other Scalar Transport Equations as well. (Honestly speaking I don't have any experience with alkaline AEMFC.)

In case on PEM FC , one need Scalar Transport Equation for Electric Potential, Protonic Potential, Water Content and water Saturation.

Best Regards,

[Bruno Machado]

Originally Posted by maverick123

Hi Bruno,

Apart from the two potential equations, what are the other equations?
You may need other Scalar Transport Equations as well. (Honestly speaking I don't have any experience with alkaline AEMFC.)

In case on PEM FC , one need Scalar Transport Equation for Electric Potential, Protonic Potential, Water Content and water Saturation.

Best Regards,

Thanks for your answer.

Yes, I do have 4 scalar equations for electronic and ionic potential, water content and water saturation (liquid water). Both water phases equations seems to be working well, with values as expected. I only mentioned before the electronic and ionic because, apparently, the problem is that both values are closer to each other, and I am expecting a bigger gap (bigger overpotential -> bigger reaction rate -> bigger species gradient).