Stuck NEB calculation
Moderator: moderators
Stuck NEB calculation
Hello,
One of my NEB calculations is stuck at and not running beyond 16 electronic steps within the first ionic cycle itself. It is at this state for the past 60 minutes. What can possibly be wrong?
One of my NEB calculations is stuck at and not running beyond 16 electronic steps within the first ionic cycle itself. It is at this state for the past 60 minutes. What can possibly be wrong?
Re: Stuck NEB calculation
If you post a .tar.gz of the calculation, I can take a look at it.
Re: Stuck NEB calculation
Dear Prof. Henkelman,
thank you for your quick reply. Here is the file for the calculation. It is to study the migration of vacancy from one site to another.
thank you for your quick reply. Here is the file for the calculation. It is to study the migration of vacancy from one site to another.
- Attachments
-
- lisitemigration_NEB.zip
- (297.03 KiB) Downloaded 10900 times
Re: Stuck NEB calculation
I don't understand the structure of the files that you sent. I'm expecting a main directory with your INCAR, POTCAR, KPOINTS, and then a set of directories: 00, 01, ... NN, with POSCAR files and OUTCAR files. Then I can see the results of your calculation and what might be wrong. The li4toli3_NEB subdirectory appears to be empty and there are no OUTCAR files in what you sent.
Re: Stuck NEB calculation
Sorry for the inconvenience, i had only sent the input files. Here is main directory. Thank you.
- Attachments
-
- li_migration_neb.zip
- (1.51 MiB) Downloaded 10971 times
Re: Stuck NEB calculation
If you look at image 04, you will see that the migrating Li atom is sitting right on top of another atom. This calculation will never converge.
You need a more sensible initial band. One option is to choose a sensible intermediate (04) structure and make two bands that you can put together. Another option is to use the IDPP method - you can do this using our nebmake.py script.
You need a more sensible initial band. One option is to choose a sensible intermediate (04) structure and make two bands that you can put together. Another option is to use the IDPP method - you can do this using our nebmake.py script.
Re: Stuck NEB calculation
Thank you for your reply.
Do i use the nebmake.py in the same way as the nebmake.pl script? i.e have the initial images generated using nebmake.py and then run the calculation as i would a normal vasp calulation? Also how different is IDPP from linear interpolation?
Do i use the nebmake.py in the same way as the nebmake.pl script? i.e have the initial images generated using nebmake.py and then run the calculation as i would a normal vasp calulation? Also how different is IDPP from linear interpolation?
Re: Stuck NEB calculation
The two scripts work the same way. For IDPP, see: https://wiki.fysik.dtu.dk/ase/tutorials/neb/idpp.html
Re: Stuck NEB calculation
Thank you Prof. Henkelman
Re: Stuck NEB calculation
Dear Prof. Henkelman, i had a followup question.
I tried using the nebmake.py script but am running into the following error: File "/home/anandak4/bin/nebmake.py", line 3, in <module>
from aselite import read_vasp, write_vasp, NEB
ImportError: cannot import name NEB. I already installed ase.
I tried using the nebmake.py script but am running into the following error: File "/home/anandak4/bin/nebmake.py", line 3, in <module>
from aselite import read_vasp, write_vasp, NEB
ImportError: cannot import name NEB. I already installed ase.
Re: Stuck NEB calculation
You have to download all of the scripts and add it to your path. You might also need to add the directory to your PYTHONPATH ; that is what you need for scripts to be able to import python modules, such as aselite. By the way, you can also do this through ase.