SCF does not converge for ZnO slab

Problems running VASP, runtime errors, technical questions.
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
donniedarko
Newbie
Newbie
Posts: 1
Joined: Tue Aug 28, 2018 9:31 am

SCF does not converge for ZnO slab

#1 Post by donniedarko » Thu May 16, 2019 7:29 am

Hi,

I have a problem converging a in my eyes fairly simple slab structure. The structure is a ZnO slab:

Zn O
1.0000000000000000
3.2909990940830092 0.0000000000000000 0.0000000000000000
-1.6454995470415046 2.8500894957131062 0.0000000000000000
0.0000000000000000 -0.0000000000000000 37.1257455150137687
4 4
Direct
0.3333330000000032 0.6666669999999968 0.4556748981770778
0.6666669999999968 0.3333330000000032 0.3842463267485064
0.3333330000000032 0.6666669999999968 0.5985320410342206
0.6666669999999968 0.3333330000000032 0.5271034696056492
0.3333330000000032 0.6666669999999968 0.4014679589657794
0.6666669999999968 0.3333330000000032 0.4728965303943508
0.3333330000000032 0.6666669999999968 0.5443251018229223
0.6666669999999968 0.3333330000000032 0.6157536732514937

and the INCAR reads:

EDIFF = 1.00e-07
EDIFFG = -5.00e-03
PREC = Accurate
GGA = PE
ISMEAR = -5
NELM = 100
IBRION = 2
NSW = 0
LREAL = Auto

However, the scf-cycle does not converge. As suggested in https://www.vasp.at/vasp-workshop/slide ... ectron.pdf I tried using ICHARG = 12 which lead to convergence. Yet when using ICHARG = 2 and playing with the mixing parameters the calculation again did not converge. I also tried ALGO = Fast, and different combinations of AMIN, AMIX, and BMIX. The employed KPOINT mesh is 5x5x1.

If you have any suggestions on how to converge this simulation I would be very grateful!

Daniel

admin
Administrator
Administrator
Posts: 2921
Joined: Tue Aug 03, 2004 8:18 am
License Nr.: 458

Re: SCF does not converge for ZnO slab

#2 Post by admin » Mon May 20, 2019 8:37 am

Predefined spin state (NUPDOWN=0) will help

Post Reply