[rando_foamer]

EDIT - in true rubber duck style, I thought of a way to do this immediately after hitting post!

I plan to modify the mixed BC code to change the way the updateCoeffs operation works.

======

Hi all,

I'm hoping someone can offer me some advice on how to implement the following mixed boundary condition in code (it's based on a chemical reaction rate occurring at a surface):

(DT/delta)*(C_P - C_b) = k * (C_b)^a

Here, the left hand side is simply a diffusive flux term, where DT is a diffusion coefficient, 1/delta is the patch deltaCoeffs, C_P is the concentration field of the current cell, C_b is the concentration of the boundary, k is a rate constant, and a is some arbitrary real value (positive or negative).

I can easily work out the answer for a=1, as the problem reduces to:

C_b = ( (DT/delta) / (k + DT/delta) )*C_P

And so, in typical mixed fashion, we can set:

refValue=0.0;
refGradient=0.0;
valueFraction=1-( (DT/delta) / (k + DT/delta) );

I've compared my coded BC for a=1 against a separate finite difference solution in Mathematica and they agree.

For a != 1, I'm completely stumped! C_b will end up being some strange polynomial in C_P and I have no idea how I should implement that in my code.

Any advice would be greatly appreciated. I'm assuming no such BC already exists in OF (at least, I can't find one that seems to be equivalent).