Hi all,
I am currently performing GSSNEB calculations in VASP 6.3.1. Upon achieving convergence, I have observed the presence of two local minima in energy along the migration path (please refer to the attached figure). Is this a reasonable result, or should we contemplate dividing the entire path into three segments and conducting separate GSSNEB calculations for each? Thank you very much.
I attach my INCAR file here.
SYSTEM = ABX3
LWAVE=.FALSE.
LCHARG=.FALSE.
ISTART=0
ICHARG=2
ALGO=Normal
PREC=Normal
LASPH=.TRUE.
LREAL= Auto
NPAR = 10
ISPIN=1
ISYM=0
ENCUT=400
ISMEAR=0
SIGMA=0.05
NELM=1000
NSW=200
POTIM=0
ISIF=3
IBRION=3
EDIFF=1E-5
EDIFFG=-0.05
ICHAIN = 0
SPRING = -5
LNEBCELL= .TRUE.
LCLIMB = .TRUE.
IMAGES = 25
IOPT=3
TIMESETP=0.01
SYMPREC = 1E-6
GSSNEB
Moderator: moderators
Re: GSSNEB
This looks like a very nice path to me. If you want to improve your estimate of the minima and the 2 lower-energy saddles, then you can break up the path. But if you just want to know the three steps along the path and the overall barrier - it looks like you have that converged nicely.