[Brian_]

Hi,

My analysis currently only involves the energy equation. I have a 1 mm plastic sphere filled with deuterium-tritium, DT, (heavy hydrogen). The sphere is surrounded by helium. Outside the helium there is an adjustable temperature BC. When the DT is cooled below 19.2 K a layer of ice sublimes out on the inside of the plastic shell. The average thickness and uniformity of thickness is of importance. The material is radioactive (heat source) so the DT is hottest in the center of the sphere (i.e. remains a gas).

I have material properties (density, conductivity, and heat generation due to beta decay that are a function of temperture). At 19.2 K the material properties change from a gas to a solid. In order to conserve mass (solid density >> than gas density) I wrote a UDF the sums the system total mass and the UDF adjusts the BC outside the helium so that mass is conserved (i.e. the mass of gas + mass of ice=the original mass of gas). My analysis is axisymmetric steady-state.

This process works reasonably well and is "fast" (> 5 Mdof takes ~8-24 hours to converge on a Dell "workstation" laptop). But I was wondering if mass can be conserved automatically if I couple the energy equation with the fluid equations and use the Eulerian multi-phase flow. I have been reluctant to do this due to my lack of knowledge of multi-phase flows and assuming my run time would increase significantly.

Regards,

Brian