Internal error in vhdf5.F at line: 121 (phonon dispersion)

[reynaldo.putra]

Hello everyone,

I ran several phonon dispersion calculations for plotting, and I got the error as mentioned in the title.

I am following the steps as mentioned in the phonon tutorial here: https://vasp.at/tutorials/latest/phonon/part2/, but I am using the DFPT method instead.

I got this error as I follow step 6 (phonon dispersion plotting), all the files are attached in this post.

3_phondisp.zip

Initially I thought it was because I ran a DFPT calculation using VASP 6.3.1, so I switched to 6.4.2. However, even after doing the DFTP force calculation using 6.4.2, I still get this error when I try to plot the dispersion.

Also, if possible I'd like to check if the QPOINTS are already properly chosen. Even though I got this error, the plotting using Py4VASP still somehow succeeded, but the band structure does not look right... Could have I chosen the wrong high-symmetry points?

newplot_LTON2.png

Any help would be appreciated.

Thank you for your time!
Reynaldo

[merzuk.kaltak]

Dear Reynaldo,

thank you for sending a bug report.
The reason you see this message is due to the missing KPOINTS file in the working directory.
If you add a proper KPOINTS file, the "error" message goes away. This issue will be fixed in the next vasp release.

Also, you have to remove the "1" after the Kx Ky Kz coordinates in QPOINTS to make py4vasp understand the input, that is change the QPOINTS to

Code: Select all

Special k-points for band structure
50  ! intersections 
line-mode
reciprocal
0.0000000000     0.0000000000     0.0000000000   GAMMA
0.5000000000     0.0000000000     0.0000000000   X

0.0000000000     0.5000000000     0.0000000000   Y 
0.0000000000     0.0000000000     0.0000000000   GAMMA
...

Nevertheless, these changes will not fix your issue, since vasp still writes the phonon dispersion successfully to file.
There are two possible reasons for this.

• structure is not in equilibrium

• super cell is too small

Try setting EDIFFG to a negative number (e.g. -0.001) during relaxation to tell vasp not to stop relaxation before all forces are smaller than 0.001 eV/A.
If this doesn't help, I suggest to use a larger super cell for the computation of the force constants.