Phonopy_VESTA

Export Eigenvectors from Phonopy format to VESTA

This Code can Extract the Eigenvectors of Phonon Dispersion from Phonopy and plot it in VESTA Files for visualisation.

If you use this code, please cite this article: A.P. Roy, Jayakrishnan SS , A. Khandelwal, M.K. Chattopadhyay, V. Sathe, S.K. Mishra, R. Mittal, P.D. Babu, M. Duc Le, D. Bansal, “Evidence of strong orbital-selective spin-orbital-phonon coupling in CrVO4”. Physical Review Letters Vol. 132, 026701, 2024, DOI: https://doi.org/10.1103/PhysRevLett.132.026701

Requirements :

Phonopy

VESTA

Python

Check if the below libraries are present by trying to import in python

numpy

sys

os

shutil

Steps :

1)

Save the Following as band.conf (Different for your material)

DIM = 2 2 2

FORCE_CONSTANTS = WRITE

BAND = 0 0 0 0.5 0.5 0.5

BAND_POINTS = 2

BAND_LABELS = $\Gamma$ R

EIGENVECTORS = .TRUE.

THz --> cm-1

FREQUENCY_CONVERSION_FACTOR = 521.47083116

Please look at the Phonopy website for other required files for your interface (e.g, VASP, QE). Note : Here I have used BAND_POINTS = 2 to find the eigenvectors only at the High Symmetry Points, you can use more.

2) Run

$ phonopy -c POSCAR -p band.conf -s

Note : This lines are written for VASP interface, but simillar things are for other interfaces supported by phonopy e.g, for QE

$ phonopy --qe -c prim_cell.in -p band.conf -s

3) You should get band.yaml as an output with the following format

phonon:

`- q-position: [ 0.0000000, 0.0000000, 0.0000000 ]

distance: 0.0000000

band:

` - # 1

frequency:   -0.0241026373

eigenvector:

- # atom 1

  - [ -0.00000504564764,  0.00000000000000 ]

  - [ -0.31749032741153,  0.00000000000000 ]

  - [  0.15556314465407,  0.00000000000000 ]

- # atom 2

  - [ -0.00000504564764,  0.00000000000000 ]

  - [ -0.31749032741153,  0.00000000000000 ]

  - [  0.15556314465407,  0.00000000000000 ]

4) Run

$ python3 extract_vectors_phonopy.py band.yaml POSCAR.vesta

Here POSCAR.vesta is the VESTA file of the primitive positions of atoms used in the initial SCF calculation, not the supercell from Phonopy. To make it, for VASP you can directly open POSCAR file in VESTA and save as POSCAR.vesta Easiest way in QE is to export your input file to .xsf format by XCRYSDEN and open it in VESTA and futher save as POSCAR.vesta Or if you used ibrav=0 in QE, then simply copy the CELL_PARAMETERS Card in Angstron units and ATOMIC_POSITIONS in Direct or Cartesian Format as in VASP POSCAR file, and save it as POSCAR.vasp, open it in VESTA, and save as POSCAR.vesta. Please look in the internet on how to create such files for other supported interfaces, since I only have experience in QE and VASP.

Note : Typing band.yaml and POSCAR.vesta is completely optional, since the code, by default takes those two filenames as input, in case you have the same files saved in other names but in same format, you can put those names in the command line but should be in the same order. (See the code for more info)

5)

Finally all the VESTA files are ready for visualization and will be saved in VESTA_FILES folder with each q-point and inside them, the VESTA files saved for each band with the name in format of band_index_(frequency_in_cm-1).vesta. You can change the name of the files to any units by changing the Tag FREQUENCY_CONVERSION_FACTOR = 521.47083116 in band.conf, which converts from THz to cm-1. But keep in mind, as some bands are degenerate, the format of the files to be written should follow the trend as written in the code, other wise some degenerate bands with same frequency will get overwritten and you wont get the number of VESTA files you are expecting for your material i.e., 3 x natoms.

I will keep updating the code, please comment if you face any problem.

References

https://github.com/Stanford-MCTG/VASP-plot-modes