From:	Timm Lankau <lankau' at \`blubber.chemie.uni-hamburg.de>
Date:	Tue, 21 Jan 1997 12:57:44 +0100 (NFT)
Subject:	Summary: broken symmetry in G94

Dear Netters,

thank you very much for your answers. At the end of this mail is an
extract from the most helpful answers.
Stephan Irle's suggestion seems to the useful but also to be the most
tedeous. He suggest to use another element and to redefine it later with
the 'massage' command and an external basis set.
Once again thank you.

Best wishes Timm

==========================================================================

Timm Lankau                            phone  (+)40 4123 3686
Institut fuer Physikalische Chemie     fax    (+)40 4123 3452
Universitaet Hamburg                   e-mail lankau <-at-> chemie.uni-hamburg.de
Bundesstr. 45
20146 Hamburg
Germany

==========================================================================


Summary:
========

Dear Timm,

If you have the cartesian coordinates of your computation with the first
symmetry, you might just want to change the last digit of the
respective coordinates one number, to bring it in accordance with the
required symmetry.  Since G94 outputs give many digits in the cartesisian
coordinates, this changes (let's say for an arbitrary number from
1.2853967 to 1.2853968) changes the symmetry recognition by G94 (see
reference manual for the limits that is keeps by default for this
recognition), but will not affect the energy itself significantly (often
less than a microhartree!).
I hope this helps,
Han

***************************************************************************
**  Dr. Han Zuilhof          **  E-mail: ZUILHOF -AatT- chem.columbia.edu
**
**  Department of Chemistry  **                                          **
**  MC 3156                  **                                          **
**  Columbia University      **                                          **
**  New York, NY, 10027      **  Fax:   (212) 932-1289                   **
**  USA                      **  Voice: (212) 854-2179                   **
***************************************************************************

Viele Gruesse von Wien nach Hamburg!

Meine pragmatische Loesung fuer das Problem ist die Folgende:
Gaussian einfach ein anderes chemisches Element fuer das 2.
Atom in der Z-Matrix vorsetzen, und dieses dann entsprechend
mittels listengesteuertem Basissatz-Input und dem Keyword
'massage' fuer die Kernladung so zu modifizieren, dass de
facto zwei identische Atome vorliegen. Fuer N2 hat der Trick
auf diese Weise funktioniert.

Input:
# rhf/gen gfinput

n2 full symmetry

0 1
n
n 1 r1

r1=1.0

  1 2 0
 S    3 1.00
   .2427660000D+03   .5986570000D-01
   .3648510000D+02   .3529550000D+00
   .7814490000D+01   .7065130000D+00
 SP   2 1.00
   .5425220000D+01  -.4133010000D+00   .2379720000D+00
   .1149150000D+01   .1224420000D+01   .8589530000D+00
 SP   1 1.00
   .2832050000D+00   .1000000000D+01   .1000000000D+01
 ****

--Link1--
# rhf/gen gfinput massage

n2 broken symmetry (2nd nitrogen atom massaged)

0 1
n
o 1 r1

r1=1.0

  1 2 0
 S    3 1.00
   .2427660000D+03   .5986570000D-01
   .3648510000D+02   .3529550000D+00
   .7814490000D+01   .7065130000D+00
 SP   2 1.00
   .5425220000D+01  -.4133010000D+00   .2379720000D+00
   .1149150000D+01   .1224420000D+01   .8589530000D+00
 SP   1 1.00
   .2832050000D+00   .1000000000D+01   .1000000000D+01
 ****

2 0 7.0

-Stephan-Irle--stephan %-% at %-%
itc.univie.ac.at---------------------------------------
 http://www.itc.univie.ac.at/~stephan/
 voice:+43/1/40480-679


Similar Messages
05/14/1997:  Problem with an MP2 calculation using G94
12/01/1995:  RE:DZP' basis set
12/18/1996:  RE: GAMESS BASIS SET INPUT
09/12/1995:  CCL:Using valence double zeta basis in MOLPRO94?


Raw Message Text