Discussion Forum

Hello,

In a very simple example I´m trying to do I use Navier-Stokes and Maxwell-Stefan equations for the momentum and mass transport in a pipe. In the MS equations I have two components, O2 and H2O, and one of the boundary conditions is an oxygen flux through one of the pipe boundaries. When solving, COMSOL automatically considers that the same oxygen mass flux that is "disappearing" is because water is being produced, i.e., the total mass is conserved in the subdomain and therefore the same amount of water "appears" in this boundary. However, this is not my case of study and I would like to specify the amount of water that is being produced (higher or lower than the oxygen flux that I have specified, or even no water flux). This could be the case of a membrane selective to oxygen (the oxygen disappears but not the water and the total mass is not conserved) or a reactive boundary where oxygen reacts with protons coming from elsewhere and then producing water (a fuel cell cathode). In this case, the total mass would increase as the chemical reaction occurs and thus the total mass would not be conserved either. How could I specify this?
By the way, I used the conservative form of the MS equations because total mass was supposed not to be constant. However, I see a rather bad stability in the solution (strange mass fraction profiles) when coupling with Navier-Stokes equations for the momentum balance. Can this be improved?

Thank you very much in advance.