[kage]

Dear All,

I'm trying to model the aggregation and separation of magnetic nanoparticles in a sessile ferrofluid droplet.

I'm wondering if this could be done using interMixingFoam, where the 3rd phase is the miscible phase of magnetic nanoparticles in water.

I've tried setting up interMixingFoam and was getting good results up to coupling the magnetostatics into the fluid flow.

Basing off from magneticFoam, I've copied pretty much most of the code into interMixingFoam solver.

This was what I did:
1) merging the createFields.H
- copying the psi, B, H and HdotGradH field, murf and Mrf field
- reading the constant/transportProperties for the magnets into the new createFields.H
2) merging the magneticFoam with interMixingFoam. Since field was created in the createFields.H, this were the only other lines that I moved over, changing simple.correctNonOrthogonal() to pimple.correctNonOrthogonal(), as follows
- // Solving for magnetic field
while (pimple.correctNonOrthogonal())
{
solve(fvm::laplacian(murf, psi) + fvc::div(murf*Mrf));
}
H = fvc::reconstruct(fvc::snGrad(psi)*mesh.magSf());
HdotGradH = H & fvc::grad(H);
B = constant::electromagnetic::mu0*fvc::reconstruct(mu rf*fvc::snGrad(psi)*mesh.magSf() + murf*Mrf);

I was able to specify the magnetic properties (relative permeability) of the three phases involved by adding additional functions in the incompressibleThreePhaseMixture header and C++ files.

I was also able to set up the magnet in the simulation and obtain the B, H and HdotGradH fields for every time step. But the fields were constant and did not change. It seems that the murf for the system was not updated. So I proceeded to update it to read off the new murf after the droplet have moved. There seems to be some response to the B and H field now.

I tried updating the Mrf using the formula M = chi*H, where chi is the magnetic susceptibility = relative permeability - 1.

This is where I was stuck. How do I multiply a surfaceScalarField chi to a volVectorField?

I know that there is still the component of the magnetic force that I need to add in the momentum equation the Kelvin force. I have yet done so, as I wanted to make sure that the fields are properly set up before doing so.

Regards,
Kwok Hoe