EPW Forum

hi dear administrators and users
im working on thermal properties on an monolayer structure i reach stability using phonopy software, my question is can i use the data provided from phonopy to use EPW in order to investigate the electron phonon coupling

Best

Dear ilias-serifi:

Which data from phonopy are you referring to?

I don't know much about phonopy and I am not sure whether I understand correctly your question, but the main inputs to EPW to investigate the electron phonon coupling are dynamical matrices and perturbing potentials which are imported from the DFPT calculations using the phonon code in QE.

Sincerely,

H. Lee

I'm not the original poster, but I'm also looking at this. QE's DFPT doesn't seem to scale well enough for large/complex systems for us, so I think I'm going to try making an interface for VASP+phonopy->EPW. I'd like your input hlee on this.

You mention we'd need the dynamical matrices (can be generated with phonopy) and an equivalent to the dvscf files generated by QE (I believe recent versions of VASP compute dV_SCF/du_{jl}, the change in the KS potential w.r.t. real space atomic displacements, in their linear response module). There is a VASP2WANNIER functionality, so in theory we can also generate the Wannier Functions for the electrons.

Could you give us advice on how the pieces should fit together? Or would it just be easier to re-implement the EPW functionality for the VASP case? My end goal is merely to generate the Wannier-space electron-phonon matrix elements. I don't need to interpolate to finer k/q-grids.

Dear adenchfi:

I don't know about the recent version of VASP; in other words, I don't know how the DFPT is implemented in the recent version of VASP.
Therefore, I would suggest you to contact the VASP developers.

I would just emphasize again that the main inputs to EPW to investigate the electron phonon coupling are dynamical matrices and perturbing potentials. Regarding how to proceed with dynamical matrices and perturbing potentials, you can check Prof. Feliciano's paper ( https://journals.aps.org/prb/abstract/1 ... .76.165108 ).

PS) I realized that the original message was posted in the wrong place; the only messages about the compilation should be posted here.

Sincerely,

H. Lee

Hi hlee,
Thank you. I will look into this more.