permutation not found (2)
P.J.Knowles at bham.ac.uk
Thu May 17 17:05:57 BST 2001
For a geometry optimization for which analytical gradients are not
available, you basically have to decide at the outset
what symmetry you wish to use, and to then ensure that this symmetry
is never broken. If you use the full set of cartesian coordinates in
the optimization, then you can be sure that when finite displacements
are taken, some of the points will have lower symmetry, and the chaos
that you observed will ensue. However, if you set up the geometry with
a Z matrix appropriately parameterized with molpro variables such that
the symmetry is always conserved, then you should be in good
shape. The finite displacements are always taken on the variables that
define the geometry.
Daniel Boese wrote at 01:19 on 16 May 2001:
> > As additional information, this seems only to happen if I try to optimise a
> > molecule with a higher order correlation method enforcing symmetry. Is that
> > perhaps the molecule breaks symmetry????
Prof. Peter J. Knowles
Email P.J.Knowles at bham.ac.uk Phone +44-121-414-7472 Fax +44-121-414-7471
School of Chemistry, Univ. of Birmingham, Edgbaston, Birmingham, B15 2TT, UK.
More information about the Molpro-user