Dear Prof. Graeme and all developers and users,
Finally I have managed to link the vtst tools to vasp5.4.1. As the title, is my CI-NEB running properly? I have been looking in this forum but found no definite solution. I run with 100 steps and it has not converged. Is my step not enough or something wrong with the input files/systems. I have checked the geometry in each file and they seem to be OK. Any suggestion are welcome. Please find attached all file in my calculation.
Is my CI-NEB going to right direction
Moderator: moderators
Is my CI-NEB going to right direction
- Attachments
-
- NEB.tar.gz
- (24.96 MiB) Downloaded 16963 times
Re: Is my CI-NEB going to right direction
This is not too bad but you should make a couple of changes. First, turn off symmetry (ISYM=0) in both your endpoint and NEB calculations. You can see the the energy drops significantly between the endpoint 00 and the first image 01; similarly between 09 and 08. Try relaxing the geometry of the 01/CONTCAR and see if you get a lower energy initial state -- one that is not restricted by symmetry.
Then, start a new band and use IOPT=3 until the forces drop below 1 eV/Ang. This should take just a few tens of iterations. If the path and barrier look reasonable, you can switch to any other optimizer for more rapid convergence to the minimum energy path.
Then, start a new band and use IOPT=3 until the forces drop below 1 eV/Ang. This should take just a few tens of iterations. If the path and barrier look reasonable, you can switch to any other optimizer for more rapid convergence to the minimum energy path.