Transition state convergence unstoppable

vitalys

Hello,
I had run a transition state calculation for the reaction between two organic molecules, but the calculation can't finish, since it looks like there is a repetition and no significant change occured. I killed the calculation and run it again with different keywords like opt =calcall, tight, int=ultrafine, but unfortunately every time I got the same issue.
I read intermediates geometries of the output file, and show the total energy graph of the optimization process (See attached image), it shows a zigzag and it seems that the optimization will never complete.

I have no idea how to fix this issue, since i tried all the keywords explained by Pr. Sobereva for situations like this.
I need advice.

vitalys

wangzhe 发表于 2023-10-31 10:51
If the structure looks like your target TS state, you can use the structure of the last optimization ...

Thank you very much dear wangzhe, that's exactly what i did several times, but unfortunately it keeps turning without stopping.


I aslo attached the convergence criteria for all steps, and it seems that when Maximum Force and RMS Force converged (get "YES"), the Maximum Displacement and RMS Displacement does not at all, and again Maximum Force and RMS Force return unconverged, and this happen again and again without reaching an end.

Convergence.txt (43.69 KB, 下载次数 Times of downloads: 1)

2023-10-31 16:36 上传 Uploaded

点击下载
Click to download

wangzhe

If the structure looks like your target TS state, you can use the structure of the last optimization step and submit to TS calculation again.

sobereva

Please carefully check optimization trajectory, if the last structure doesn't look like the expected TS structure, usually you need to adjust initial guess structure and redo the calculation. Simply using calcall is not helpful in this case.