I am implementing a PNP model of a RFB considering 5 species initially in steady-state.
I am working with the math module in COMSOL in a 1D problem considering the cathode and membrane. The anode is only included in the boundary of the membrane where a redox reaction takes place, so there is a source term for one species only.
In this anodic boundary, I hoped not to set the concentrations since I don't know them, but to put a transparent BC so that the flux of species can leave the membrane.
When I tried to include the source term at the anodic boundary with a flux BC, COMSOL asks to set the flux for all species, but I don't want a zero flow for the other species. So I added a variables section for the anode boundary where I defined the source term for one species and implemented a flux BC for the 5 species where I just wrote the Nernst-Planck equation based on the concentration and potential variables.
I am not sure if this is the correct way to implement this condition(?), and I find the problem that for certain conditions of SOC and applied current I get a negative profile on the membrane. I think that the anodic BC is probably not correct and I am not restricting the problem correctly.
Does anyone know the correct way to implement a transparent boundary with reaction?