[Phigo90]

Hi everyone. At the present time, I am trying to simulate an evaporation phenomenon with liquid nitrogen in OpenFoam. Therefore I've analyzed the InterCondensatingEvaporatinFoam, which should be well adapted to simulate the problem. In the "alphaEqn.H" the conservation equation for the phase fraction alpha is built.
The Code is the following:

        fvScalarMatrix alpha1Eqn
        (
            fv::EulerDdtScheme<scalar>(mesh).fvmDdt(alpha1)
          + fv::gaussConvectionScheme<scalar>
            (
                mesh,
                phi,
                upwind<scalar>(mesh, phi)
            ).fvmDiv(phi, alpha1)
          - fvm::Sp(divU, alpha1)
         ==
            fvm::Sp(vDotvmcAlphal, alpha1)
          + vDotcAlphal
        );
        alpha1Eqn.solve();

In mathematical notation, that should be
d/dt(alpha)+div(v*alpha)-div(v)*alpha=C'cond*(1-alpha)+C'vap*alpha.
I understand that the source term is tretead in an implicit way (because of Sp()).


In my view, it does not make sense to use a linear dependency of alpha for the source term. This linear dependency results in a large transient area from alpha=0 to alpha=1. Furthermore,in general, I do not understand why the mass stream of the evaporated mass should be a function on the value of alpha.

To cut a long story short, I want to change the linear dependecy on alpha of the source term to a non-linear one.
My first approach is that I want to replace alpha at the RHS. of the equation by the expression: (1-exp(-a/dd))/(1-exp(-1/dd)), where dd is a scalar predefined variable. This function should approximate the heaviside step function near alpha=0.

Does anyone know how it is possible to use a function that dependes on alpha instead of just alpha?