[adhikary]

Dear Scholars,

I am working in a Gaussian 09 in Linux operating system. I am using Santos-5. I am trying to find the TS-structure using QST2 protocol. The reactant and Product structures were fully optimized. The order of the atoms in reactant and product structures were same. After submitting the job almost immediately I received an error message as follows:

Old curvilinear step not converged, using linear step:
 SCX= 3.31D+01 DXMaxT= 2.16-318 SCLim= 0.00D+00 Fact= 0.00D+00
 RedCar/ORedCr failed for GTrans.
 Error termination via Lnk1e in /home/adhikary/g09/l101.exe

For better understanding, the part of input and output is given here as an attachment. I am expecting your kind suggestions and help.

With kind regards,

Keshab K. Adhikary

[wildfyr]

I think our computational expertise is actually a bit low here, for something this advanced I would suggest posting on https://www.reddit.com/r/Chempros/

[Corribus]

Also, as a general rule, I'd avoid displaying your email address as you do at the end of your post. It's a great way to attract a lot of spam.

[wildfyr]

Good catch Corribus. I have manually changed his forum signature to not include his email. Only cached versions of this site should have it now.

[adhikary]

@wildfyr,

I think our computational expertise is actually a bit low here, for something this advanced I would suggest posting on https://www.reddit.com/r/Chempros/

Thank you for concern and comments. I think, a 'Computational Chemistry' as a sub branch should be added in this forum. Please think on it. Of course, I shall gratefully accept your proposal to repost the problem in 'reddit.com'. Thanks a lot.

Keshab K. Adhikary