Hi,
I wanted to run Nudged elastic band method using lammps (7th Aug, 2019 version), providing in the input initial and final images. While NEB does linear interpolation followed by minimization, I noticed in the first frame of the output trajectory files of each replica that the bonds were abnormally stretched after the linear interpolation scheme was applied, which shouldn’t have been.
Being skeptical, I generated my own images using linear interpolation, and visualized those images too, which looked fine. I ran NEB again, using my own generated linearly interpolated images, and I found one image structure (probably transition state) being distorted completely, which is not realistic (attached image of the same). When I look at the trajectory for this image, I see that the structure gets more and more distorted as NEB minimization moves forward, which should not happen.
Attached herewith is a screenshot of the distorted NEB replica structure and the script that I used to perform the NEB run. Any help in understanding what is wrong with my simulations would be greatly appreciated.
Thanks a lot,
Shaunak Badani
CCNSB, IIIT Hyderabad
neb.inp (1.03 KB)