Page 1 of 1

The influence of images used for CINEB

Posted: Mon Jul 12, 2021 3:58 am
by YZW825651869
Dear all,
I'm trying to use CINEB to search the transition state.Here is my problem:
1、Now I have used 4 images to run CINEB for a round,and I think image 03 is close to the transition state I'm looking for.
However ,it costs me a lot of time ,so I want to know if I only use image 03 to run (1 image for CINEB),will this image climb to the transition state?

2、If I can use 1 image to run the CINEB, should I reset my EDIFF , EDIFFG , IOPT or other parameters?
The parameters i use now:

SYSTEM = ts_butene partial frequency

Electronic minimisation
GGA = PE ! density functional
PREC = Normal ! precision tag
EDIFF = 1E-6 ! criterion to end SCF loop
ENCUT = 400.0 ! energy cutoff
NELMIN = 5 ! minimum number of SCF-iterations
NSW = 20 ! Number of ionic steps
NELM = 50 ! Number of electronic iterations
EDIFFG = -0.02 ! Global break cond. for the elec. Ionic-loop
ISIF = 0

DOS related values
ISMEAR = 0 ! Gaussian smearing for big cells
SPRING = -5

Single point energy calculation
IBRION =3 ! calculate frequencies
POTIM = 0 ! step width
LVDW = .TRUE.
IMAGES = 4
ICHAIN=0
LCILMB=.TRUE.

Do not write huge files - restart options
LCHARG = .TRUE. ! do not write charges
LWAVE = .TRUE. ! do not write wavefunction
ISTART = 1 ! do not read previous SCF guess
IOPT=1

Re: The influence of images used for CINEB

Posted: Tue Jul 13, 2021 1:40 am
by graeme
If you can attach a .tar.gz file of your calculation, I can help to debug the problem.

Re: The influence of images used for CINEB

Posted: Thu Jul 15, 2021 5:25 am
by YZW825651869
graeme wrote:
> If you can attach a .tar.gz file of your calculation, I can help to debug
> the problem.


Sorry I forgot attaching the file,thanks for attention!