From:	Eric Breynaert <eric.breynaert=at=agr.kuleuven.ac.be>
Date:	Wed, 08 Sep 2004 11:00:14 +0200
Subject:	CCL: Good initial guess for transition metal calculations

Dear ccl'ers,

What would be a good initial guess if one is doing geometry optimisations
on a large range of similar transition metal structures?

I'm doing ub3lyp/lanl2dz calculations on Technetium compounds. These
calculations often need a very large computational effort. Improving that
initial guess would probably reduce the computation time. In some
literature it says to read the inital guess from a HF calculation with the
same basis set. But HF seems to be a really bad method for TM-systems.
Would it be possible to reuse the wavefunction for an optimised structure
as starting point for a similar structure? Or would it be possible to
optimise for example a Tc(OH)4 molecule or a [TcO]2+ core, and use that
wavefunction as an inital guess ?

thanks,

Eric





--
Eric Breynaert
Labo voor Colloodchemie
Kasteelpark Arenberg 23
B-3001 Leuven

Tel: +3216321457
Fax: +3216321998






Raw Message Text