Vibrational entropy

[sjtuzhanglei]

Dear CASM developers,

Wondering if vibrational entropy is or being considered in the CASM package right now? If not, what could be the best way to incorporate it under the CASM framework?

Thanks,

[bpuchala]

CASM has some features that could be useful in the process, but currently doesn't include something like a phonopy-like tool or couple effects in monte carlo. The casm enum method ConfigEnumSiteDoFs can enumerate symmetrically unique displacements as a part of the process of incorporating vibrational entropy. Also, a CASM can generate a Hamiltonian including site displacements. CASM can be used to find equilibrium phases and then phonopy could be used to calculate vibrational free energy.

[sjtuzhanglei]

Hi Brian,

Just a catch-up with the thread: what if I want to add ZPE and vibrational part (a constant under the room temperature we are interested in) on top of the formation_energy generated by DFT and then do the ECI fitting?

I want to have 3 set of ECIs that fit to 3 cases:

1. 0K Energy vs. corr
2. 0K Energy + ZPE vs. corr
3. 0K Energy + ZPE + Fvib(298.15K) vs. corr

Thanks!

[bpuchala]

For fitting to different energies, see the comment with #345

To include multiple cluster expansions with different eci in your project use ccasm settings options. The options include things like creating an additional directory for a new eci set (using --new-eci), to set the eci (--set-eci) used by the current cluster expansion, create a new named cluster expansion that can be referred to in a Monte Carlo input file or elsewhere (--new-clex), and set the default cluster expansion (--set-default-clex).