Hybrid functional and ISMEAR=-2 does not work in VASP 6.4

Problems running VASP: crashes, internal errors, "wrong" results.


Moderators: Global Moderator, Moderator

Post Reply
Message
Author
joshua_claes
Newbie
Newbie
Posts: 1
Joined: Mon Oct 04, 2021 2:58 pm

Hybrid functional and ISMEAR=-2 does not work in VASP 6.4

#1 Post by joshua_claes » Mon Mar 03, 2025 1:30 pm

Hi,

I recently performed some calculations to relax a defect in hBN. After noticing that the system wasn’t converging, I checked the output and found that the forces and total energy were oscillating. This issue seems to occur only when using the HSE06 functional with the ISMEAR=-2 tag.

In the attachment, I’ve included a minimal working example with a small unit cell, showing the forces during relaxation:

HSE_no_extra_input.png

I tried various parameter combinations to resolve this issue, but the only solution was reverting to VASP 5.4.4. I’ve also observed this behavior in other systems, such as the silicon vacancy center in diamond, where switching to an older VASP version also fixed the problem.

This might be related to previous discussions about this parameter combination (listed below), though I am using a newer version of VASP.
- https://www.vasp.at/forum/viewtopic.php ... rwe#p23031
- https://www.vasp.at/forum/viewtopic.php?f=4&t=18725

Is the combination of hybid functionals and ISMEAR=-2 still broken in VASP? Or am I missing something?

You do not have the required permissions to view the files attached to this post.

ferenc_karsai
Global Moderator
Global Moderator
Posts: 517
Joined: Mon Nov 04, 2019 12:44 pm

Re: Hybrid functional and ISMEAR=-2 does not work in VASP 6.4

#2 Post by ferenc_karsai » Fri Mar 07, 2025 7:50 am

Thank you for reporting the problem. I was able to reproduce it. I even see it without ISMEAR=-2. I will discuss with my colleagues if they have a solution for you at the moment.


Post Reply