How do I know NEB is done and ready to plot the barrier?

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


Moderators: Global Moderator, Moderator

Post Reply
Message
Author
ashokneb

How do I know NEB is done and ready to plot the barrier?

#1 Post by ashokneb » Fri Feb 01, 2008 4:01 am

Hello All,

How do we know that a NEB run is done?
Do I check it in the OUTCAR file that all the forces are less than or equal to 0.03 ev/Angst.
when can we say that a NEB run is fully converged and ready to plot the barrier using
images Vs Eo(sigma->0).

I did a NEB run with 3 images and then copied CONTCAR to POSCAR for respective images and re-ran the NEB. I repeated the above procedure for three times. Every time I checked the OUTCAR file forces for each image where some of the forces reach below 0.03 ev/Angst and some do not.
Is this the right method to do??

how can I make all the forces less than 0.03 ?
Is this criterion show that the NEB is fully converged and ready to plot?
Is there any criterion that should be met.

Here is my INCAR file

SYSTEM = Si bulk
ENCUT = 150.0 eV,
PREC = High
IMAGES = 3
SPRING = -5
NPAR = 12
EDIFFG = -0.005
EDIFF = 0.0001
POTIM = 0.05
IBRION = 3
LREAL = Auto
NSW = 500
LCHARG = .FALSE.
LVTOT = .FALSE.


Please reply.

Thank you,
Ashok
Last edited by ashokneb on Fri Feb 01, 2008 4:01 am, edited 1 time in total.

lcyin
Newbie
Newbie
Posts: 31
Joined: Fri Mar 18, 2005 7:07 am
License Nr.: 375

How do I know NEB is done and ready to plot the barrier?

#2 Post by lcyin » Wed Mar 05, 2008 7:04 am

From your INCAR setting, that the NEB run should be ended after the forces on each atom are less than 0.005 eV/Angst, pls. check it.
Last edited by lcyin on Wed Mar 05, 2008 7:04 am, edited 1 time in total.

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

How do I know NEB is done and ready to plot the barrier?

#3 Post by admin » Wed Mar 05, 2008 3:42 pm

please note that NEB only allows for relaxation along the hypertangents perpendicular to the reaction path, therefore the
force components along the reaction path will remain quite large sometimes (otherwise, the "images" would instantaneously relax into their adjacent minimum.) For an NEB run, I would rather choose the deltaE criterium (EDIFFG>0) to check convergence.
Last edited by admin on Wed Mar 05, 2008 3:42 pm, edited 1 time in total.

christine1201

How do I know NEB is done and ready to plot the barrier?

#4 Post by christine1201 » Mon Mar 10, 2008 9:02 pm

you didnt turn LCLIMB = .TRUE. on.
I think CNEB will be better.
Last edited by christine1201 on Mon Mar 10, 2008 9:02 pm, edited 1 time in total.

Post Reply