I was just curious if anyone had any particular thoughts on FF features related to
this exercise. It is nominally anodic etching of a metal with a normalized/fictional
Butler-Volmer equation for now. The initial equillibrium ( zero transport ) ion
densities are found with the SNES non[linear solver and written to a file to be
used in a seprate run. The initial and time stepping runs may need different
number of cores etc.
The etching was solved with the sparsesolver and time stepped.
The etch progress is encoded in a phase variable, 1 in the solid and 0 in the liqud.
In the images below, the top is the liquid and the bottom red is the metal
being etched.
At the interface, the diffusivity changes drastically and the amount of
reactive ion that diffuses into the metal determines the etch amount 
The one bump becomes a point before finally polishing out a bit.
" any similiaritties between these simulations and real physical systems is
purely coincidental " for now…
