Consultation on the post processing of phonon calculation (IBRION=6)

Question on input files/tags, interpreting output, etc.

Please check whether the answer to your question is given in the VASP online manual or has been discussed in this forum previously!

Moderators: Global Moderator, Moderator

Post Reply
Message
Author
xiancai
Newbie
Newbie
Posts: 4
Joined: Sun Nov 17, 2019 9:42 am

Consultation on the post processing of phonon calculation (IBRION=6)

#1 Post by xiancai » Wed Jul 27, 2022 7:08 am

Dear VASP developers and users,

I encounter a problem on the post processing of phonon calculation. I have finished a finite difference based phonon frequency calculation by using VASP (i.e., IBRION=6 and NFREE=2. NOT by using phonopy!), obtaining the frequencies at the Gamma point as well as a series of output files such as OUTCAR and DYNMAT. Now I want to change the masses of several atoms, and obtain the corresponding phonon frequencies. I am wondering if there are any programs or software could derive the phonon frequencies based on the output files (such as OUTCAR or DYNMAT) directly, without re-performing the phonon calculation with different atomic masses?

Sincerely
Hua Tian

henrique_miranda
Global Moderator
Global Moderator
Posts: 414
Joined: Mon Nov 04, 2019 12:41 pm
Contact:

Re: Consultation on the post processing of phonon calculation (IBRION=6)

#2 Post by henrique_miranda » Wed Jul 27, 2022 9:09 am

Yes, this should be possible.

If you are familiar with phonopy you can use it to read in the hessian matrix produced by VASP using this procedure:
http://phonopy.github.io/phonopy/vasp-dfpt.html

Note that this page seems to indicate that this is only possible when using the DFPT driver in VASP ( IBRION=7 or 8 ) but from my testing, it also works with (IBRION=5 or 6) since VASP produces the same entries in the vasprun.xml file.

xiancai
Newbie
Newbie
Posts: 4
Joined: Sun Nov 17, 2019 9:42 am

Re: Consultation on the post processing of phonon calculation (IBRION=6)

#3 Post by xiancai » Wed Jul 27, 2022 1:15 pm

Dear Dr. Miranda,

Thank you very much for your guidance! Indeed, I am familiar with phonopy, and this procedure works very well.

Sincerely,
Tian Hua
henrique_miranda wrote: Wed Jul 27, 2022 9:09 am Yes, this should be possible.

If you are familiar with phonopy you can use it to read in the hessian matrix produced by VASP using this procedure:
http://phonopy.github.io/phonopy/vasp-dfpt.html

Note that this page seems to indicate that this is only possible when using the DFPT driver in VASP ( IBRION=7 or 8 ) but from my testing, it also works with (IBRION=5 or 6) since VASP produces the same entries in the vasprun.xml file.

Post Reply