Page 1 of 1

With LCLIMB = TRUE LNEBCELL=TRUE

Posted: Mon Jun 16, 2014 5:54 am
by Jiang12
Dear Prof. Graeme and others,

With LCLIMB = TRUE LNEBCELL=TRUE, the result can not converge. My cell of calculation is composed with 12 atoms.

But with LCLIMB = FALSE LNEBCELL=TRUE, the result converges. But I want to obtain the energy barrier, so I need to set LCLIMB = TRUE, what should I do to make the result converged.

My INCAR file is as follows:
PREC = Accurate
ENCUT = 800
IBRION = 3
ISIF = 3
PSTRESS = 90
POTIM = 0
NSW = 100
ALGO = Normal I(blocked Davidson)
NELM = 100
NELMIN = 2
EDIFF = 1.0e-08
EDIFFG = -0.01
ISPIN = 1
INIWAV = 1
ISTART = 0
ICHARG = 2
LWAVE = .FALSE.
LCHARG = .FALSE.
ADDGRID = .FALSE.
ISMEAR = 1
SIGMA = 0.05
LREAL = .FALSE.
IMAGES = 1
SPRING = -5
LCLIMB = TRUE
IOPT = 3
ICHAIN = 0
LNEBCELL= TRUE
ISYM = 0


Thank you very much.

Re: With LCLIMB = TRUE LNEBCELL=TRUE

Posted: Mon Jun 16, 2014 3:01 pm
by graeme
I would be interested to see what's going on here. If you post the two calculations, we'll take a look.

Re: With LCLIMB = TRUE LNEBCELL=TRUE

Posted: Tue Jun 17, 2014 1:51 am
by Jiang12
[quote="graeme"]I would be interested to see what's going on here. If you post the two calculations, we'll take a look.[/quote]

Re: With LCLIMB = TRUE LNEBCELL=TRUE

Posted: Tue Jun 17, 2014 1:57 am
by graeme
A tar.gz file of the files would be best; there is no need to include the CHG* WAV*.

Re: With LCLIMB = TRUE LNEBCELL=TRUE

Posted: Tue Jun 17, 2014 2:44 am
by Jiang12
[quote="graeme"]A tar.gz file of the files would be best; there is no need to include the CHG* WAV*.[/quote]
Dear Prof. Graeme, how should I do in detail to send you the tar.gz file?

Thank you

Re: With LCLIMB = TRUE LNEBCELL=TRUE

Posted: Tue Jun 17, 2014 2:45 am
by graeme
You can attach it here, email it (if it's not too large) or make it available by a web link or dropbox.