I asked this question to Marios Zacharias via email and he encouraged me to post it here:
Did we use equation12 in https://arxiv.org/abs/2212.10633 to get the coordinates in Zg-relax.in, using the monomorphous harmonic calculation's IFC? So effectively I already did the first ZG displacement before I did the band structures right?
A-SDM - ZG-relax.in
Moderator: stiwari
Re: A-SDM - ZG-relax.in
To generate the coordinates in ZG-relax.in, indeed, we use that equation. However, the ZG coordinates in ZG-relax.in are simply used to initiate the relaxation and nothing else. The resulting polymorphous structure (in ZG-relax.out) does not incorporate any temperature dependence.
Then, the polymorphous structure is used to compute the starting dynamically stable phonons (iteration 0); from there you can start introducing thermal ZG displacements iteratively. So if you want to proceed to the next step (iteration 1) you need to choose the target temperature and use the interatomic force constants in "poly_iter_00.fc" to obtain a new ZG configuration.
Then, the polymorphous structure is used to compute the starting dynamically stable phonons (iteration 0); from there you can start introducing thermal ZG displacements iteratively. So if you want to proceed to the next step (iteration 1) you need to choose the target temperature and use the interatomic force constants in "poly_iter_00.fc" to obtain a new ZG configuration.