Hybrid functionals with ISMEAR=-2

Questions of type "How do I calculate xxx ?"
(Open discussion, limited support from the helpdesk.)

Moderators: Global Moderator, Moderator

Hybrid functionals with ISMEAR=-2

Postby IBRAHIM » Sun Dec 24, 2017 7:50 pm

Dear All,
I would like to fix the occupancy of bands using ISMEAR=-2 and HSE06 functional.

I followed the following steps:
1- Run normal DFT calculation with ICHARG=2, ISMEAR=0 and ALGO= Normal.
2- Run normal HSE06 calculations with ICHARG=0, ISMEAR=0 and ALGO= All.
3- To fix the occupancy I used ISMEAR=-2, LDIAG=F, FERWE tag, FERDO tag, ALGO= All with WAVECAR from the second step.

In the last step I got the following warning:
ALGO=A and IALGO=5X tend to fail with the tetrahedron method
(e.g. Bloechls method ISMEAR=-5 is not variational)
please switch to IMSEAR=0-n, except for DOS calculations
For DOS calculations use IALGO=53 after preconverging with ISMEAR>=0
I HOPE YOU KNOW, WHAT YOU ARE DOING

Therefore in step 3, I used IALGO=53, however the warring message reappeared. But the jobs are finished in all cases.

I have two questions:
1- What should I do to remove this warning message?
2- Are my steps correct or I have to run first non-self consistently calculation with ICHARG=12 as written in the online handbook for atoms?
3- How does LDIAG tag affect on the results? With ALGO=All the results (density of states) are quite different with LDIAG = T and LDIAG=F. With IALGO=53 there is no difference and the DOS are similar to that of ALGO=All and LDIAG = T.

Many thanks in advance,
Ibrahim
IBRAHIM
Newbie
Newbie
 
Posts: 9
Joined: Mon Jan 30, 2017 6:00 pm

Re: Hybrid functionals with ISMEAR=-2

Postby admin » Tue Jan 02, 2018 11:40 am

How many k-points do you use? The tetrahedron method requires at least 4 points.
admin
Administrator
Administrator
 
Posts: 2864
Joined: Tue Aug 03, 2004 8:18 am

Re: Hybrid functionals with ISMEAR=-2

Postby IBRAHIM » Tue Jan 02, 2018 3:24 pm

Dear Admin,

I tried ISMEAR=-2 with only gamma point and get the preferred occupancy but with the issues mentioned above.
However I did not get the error message concerning the tetrahedron method "VERY BAD NEWS! internal error in subroutine IBZKPT: Tetrahedron method fails for NKPT<4. NKPT = 1"

Also there are some examples in the forum discussed before such as:
viewtopic.php?f=4&t=358&hilit=ISMEAR+%3D+2+FERDO
https://cms.mpi.univie.ac.at/vasp-forum ... =4&t=10207

Best regards,
Ibrahim
IBRAHIM
Newbie
Newbie
 
Posts: 9
Joined: Mon Jan 30, 2017 6:00 pm

Re: Hybrid functionals with ISMEAR=-2

Postby IBRAHIM » Fri Jan 12, 2018 8:19 am

Dear All,

- Could I use ISMEAR=-2 and one k-point as written in the manual and the forum for single atoms with one k-point?

- If not, why did not I get the error message related to the tetrahedron method with insufficient k-points? Also how did the job finish with the preferred occupancy?

Many thanks in advance,
Ibrahim
IBRAHIM
Newbie
Newbie
 
Posts: 9
Joined: Mon Jan 30, 2017 6:00 pm

Re: Hybrid functionals with ISMEAR=-2

Postby admin » Tue Jan 23, 2018 9:32 am

The gamma-point calculation on a single atom is not precise,
but acceptable.
admin
Administrator
Administrator
 
Posts: 2864
Joined: Tue Aug 03, 2004 8:18 am

Re: Hybrid functionals with ISMEAR=-2

Postby IBRAHIM » Wed Jan 24, 2018 9:55 am

Dear Admin,
Thank you very much.

Therefore, I will continue with ISMEAR=-2 and one k-point for my systems.

Best regards,
Ibrahim
IBRAHIM
Newbie
Newbie
 
Posts: 9
Joined: Mon Jan 30, 2017 6:00 pm


Return to Physics questions

Who is online

Users browsing this forum: No registered users and 2 guests