Questions about convex hull constructions with ternary compounds

icet
1

Hi,

I have a few questions about the convex hull construction with ternary compounds. As I can see from ICET tutorial, the ConvexHull function is mostly used with binaries, where there’s only one independent concentration. However, my current system is a Ti-C-N-Vac ternary system, where [[Ti], [C, N, X]] Ti atoms takes one sulattice and C, N, Vacancies take another sublattice. I wonder if the ConvexHull function work well for such a system?

If so, which concentration should I use? (sublattice concentration or the element total concentration?) For the Ti-C-N system, the sublattice concentration of Ti is always 1.0, so I think using sublattice concentration doesn’t work here. I’m a bit confused about this part right now. Would you please show me how I should set the concentration correctly in this function?

Thank you so much. Look forward to your reply.

2

Im not sure about this, maybe can check out the code and discussion here An example using ASE's PhaseDiagram for convex hull energies (#615) · Issues · materials-modeling / icet · GitLab
and possibly Michael would know more about this.