Phonon System Errors

[fballestermacia]

Hello,

During the past few weeks I have been testing the TB model for phonon systems using WannierTools and I have found some clear inconsistencies.

First, I have found that the code applies the Acoustic Sum Rule incorrectly. The ASR should be applied to the force constants matrix when generating the wannier90_hr.dat file, not to the eigenvalues. This can be solved easily by removing the lines:
if (index(Particle,'phonon')/=0) then eigv_mpi = eigv_mpi - MINVAL(eigv_mpi) endif
from any ek_*.f90 file.

After this, the code works wonders for a apolar materials. However, when considering polar materials - and the subsequent LO-TO splitting, the units for the splitting are completely wrong and, in fact, no long-range dipole-dipole interaction is added to the Hamiltonian. The units for the non-analytic correction can be easily solved (and I believe should be divided by the number of cells in the supercell, as mentioned in the paper cited in the code), but the long-range term needs a bit more thought. I am currently trying to work on a fix, but I am struggling to implement this long-range term. It should be done using an Ewald sum (see eq 73 of X. Gonze et al, PRB 50. 13035 (1994)), which I do not know if it's already implemented in the program somehow.

Is there anyone currently working on the phonon implementation of WT? How could I contact them?

Thank you.