Page 1 of 1

Applying phonon eigenvector displacements to POSCAR

Posted: Tue Sep 22, 2026 12:54 am
by dominic_varghese

Hi everyone,

I am performing phonon calculations using Phonopy on a 2x2x4 supercell using the finite displacement method. From the resulting qpoints.yaml files, I have extracted the phonon frequencies and their corresponding eigenvectors at specific q-points.

I would like to create a modulated POSCAR with a specific phonon mode displacement applied.

I have a few technical questions regarding the standard workflow for this:

Should the displacements listed in qpoints.yaml be applied to the primitive unit cell or mapped directly onto the 2x2x4 supercell?

Since the eigenvectors from Phonopy are given as complex numbers (amplitude and phase) per atom in Cartesian or fractional coordinates, are there standard transformation matrices required to map non-zone-center eigenvectors onto real space supercells?

Is there a standard, built-in Phonopy command (or Python API method) to automatically generate the modulated POSCAR directly from the q-point/eigenvector data, avoiding manual coordinate transformations?

Thank you for your time and assistance!

Thanks
Dominic


Re: Applying phonon eigenvector displacements to POSCAR

Posted: Tue Sep 22, 2026 1:06 pm
by christopher_sheldon1

Dear Dominic,

Thank you for your question. It is primarily concerned with Phonopy, which I'm afraid I can't help with, so I will move this to the "From Users to Users" section of the forum. I have a few clarifying questions:

If I understand correctly, you want to displace the supercell along a specific phonon mode?

If so, then if you use the primitive or supercell will depend on whether or not the specific phonon mode can be reduced to the primitive cell by symmetry (perhaps these tags are what you mean: https://vasp.at/wiki/LPHON_DISPERSION, and https://vasp.at/wiki/Computing_the_phon ... on_and_DOS).

For Phonopy tools, I will leave it to someone with more expertise.

Best wishes,

Chris