Firefly and PC GAMESS-related discussion club

Learn how to ask questions correctly


Alex Granovsky


it is always a good idea to test your scan using semiempirical
calculations. Sometimes, performing successful scan is like the
art - you need to adjust various parameters etc...

I'm attaching the scan that seems to work.
Note, the only undocumented parameter in the input file is
$statpt regtol(2)=10000 - this sets internal threshold used
to reset approximate hessian to the very large value and thus
effectively disables this feature - it seems it causes some
problems in your particular case. You can try to rerun this
scan with RHF or DFT.

As to C2h scan, your mistake is the numbering of atoms - it is not
the same as for C1 symmetry. Thus, your scan variables are a little
bit random and do not preserve symmetry; and so, the transformation
to Cartesians cannot converge.


On Mon Sep 6 '10 5:05pm, Jonas Baltrusaitis wrote
>Thank you Alex. Seems like there are still issues with this particular molecule.

>I tried nosymmetry scan, molecule implodes into itself after several steps.

>C2h scan is much worse, it fails convergence from internals to cartesians at the very first step

>I would appreciate any suggestions


This message contains the 7508 kb attachment
[ art_of_scan.rar ] The art of scan

[ Previous ] [ Next ] [ Index ]           Mon Sep 6 '10 10:28pm
[ Reply ] [ Edit ] [ Delete ]           This message read 771 times