Job not running. need help in checking INCAR file.

Queries about input and output files, running specific calculations, etc.


Moderators: Global Moderator, Moderator

Post Reply
Message
Author
kalpna
Newbie
Newbie
Posts: 10
Joined: Mon Jun 03, 2013 10:16 am

Job not running. need help in checking INCAR file.

#1 Post by kalpna » Mon Jun 16, 2014 2:57 am

Dear sir,
I am trying to run a perovskite structure MAPbI3 in vasp. But the job is not running and coming out. The INCAR file that I am using is:
SYSTEM = MAPbI3

PREC = Normal
ENCUT = 400
IBRION = 2
NSW = 200
ISIF = 4
NELMIN = 2
EDIFF = 1.0e-06
EDIFFG = -0.02
VOSKOWN = 1
NBLOCK = 1
NELM = 60
ALGO = Normal (blocked Davidson)
ISPIN = 1
INIWAV = 1
ISTART = 0
ICHARG = 2
LWAVE = .FALSE.
LCHARG = .FALSE.
ADDGRID = .FALSE.
ISMEAR = 1
SIGMA = 0.2
LREAL = .FALSE.
KSPACING = 0.12
KGAMMA = .FALSE.
RWIGS = 0.73 0.75 0.77 0.32

This file is working for my other structures having C,H,N,O atoms.
For a test I replaced 1 of atom in my other structures with Pb. Now It is not not running.Similiar thing is happening with Sn atom.
Is there anything specific that I need to learn if using Pb or Sn atom.

Thanking you,
Last edited by kalpna on Mon Jun 16, 2014 2:57 am, edited 1 time in total.

sponza
Newbie
Newbie
Posts: 13
Joined: Wed Nov 20, 2013 5:01 pm
License Nr.: 5-311

Job not running. need help in checking INCAR file.

#2 Post by sponza » Mon Jun 16, 2014 8:28 am

Hi.

Can you tell at which point the calculation stops and what is (if any) the error message?

I don't know if it can help, but I had many problems in converging my ground state calculations and I finally meanged.
1) I observed that changing the ALGO falg can make some difference. Get a look on possible alternatives of ALGO=Normal and give a try.
2) I also observed a strong dependence of the convergence on the pseudopotential used, but I don't think it can be your problem since you are likely using the lighter pseudo.
3) If your system is a semiconductor you can try to split the calculation into two steps: a first step including only occupied states (attention: NBANDS is modified according to NPAR). Then you can make a second calculation including empty bands if needed (optics, band structure, dos) using ALGO=EXACT. In my case this procedure worked perfectly even if it's more time-consuming.

Good luck!
Last edited by sponza on Mon Jun 16, 2014 8:28 am, edited 1 time in total.
-- -- --

LORENZO SPONZA

kalpna
Newbie
Newbie
Posts: 10
Joined: Mon Jun 03, 2013 10:16 am

Job not running. need help in checking INCAR file.

#3 Post by kalpna » Tue Jun 17, 2014 4:34 am

Dear sponza,
Thanks for the response. The problem is solved. I have deleted that RWIGS line as it will be automatically included from the POTCAR file. Now all the jobs are running smothly.
Last edited by kalpna on Tue Jun 17, 2014 4:34 am, edited 1 time in total.

Post Reply