CCL:G: Constrained optimization scan in Gaussian03

Try to visualize the path of the scan using the animation feature in molekel (I am not sure if Molden might be useful too). Maybe then you might observe if there is something wrong with the geometry at the final step and rethink your input
Good luck!

"Daniel Norberg daniel.norberg~!" <> escribió:
Sent to CCL by: "Daniel Norberg" [daniel.norberg(0)]
Hello everyone,

I have a question regarding constrained optimization scans in Gaussian03 using the opt=modredundant keyword for requesting the scan.

The scan I am performing have two angular constraints. The scan parameter is a distance which I decrease by an increment of -0.1 Angstrom/optimization step. All other geometrical degrees of freedom are optimized in every optimization step. The calculation proceeds fine for about 10 optimization steps (i.e. ca 1.0 Angstrom), each optimization step ending with the message "Optimization completed". However at a particular point the scan stops with the following error message:

New curvilinear step not converged.
Error imposing constraints
Error termination via Lnk1e in /usr/g03/l103.exe at Tue Nov 29 02:01:27 2005.
Job cpu time: 0 days 6 hours 15 minutes 52.8 seconds.
File lengths (MBytes): RWF= 21 Int= 0 D2E= 0 Chk= 8 Scr= 1

Based on the second line in the error message above it seems that it has to do with the angular constraints which I have imposed to the optimizations.

Can anyone help me to interpret this error, and perhaps suggest what I could try to do in order to cicumvent this problem.

Thanks in advance!

Daniel, PhD-student

Q. Joaquín Barroso Flores

Instituto de Química UNAM
Correo Alterno:

Do You Yahoo!? La mejor conexión a Internet y 2GB extra a tu correo por $100 al mes.