On Wed, 25 Apr 2001, Dr R J Deeth wrote:

> Dear All,
>
> Further to Luigi's message, we too have noticed that the internal
> coordinate optimisation in ADF2000 is nowhere near as reliable as in
> 2.3. We find in particular that torsion terms involving say H atoms
> attached to aromatic rings become unstable and oscillate through +-10
> degrees so that the energies and gradients are low but the angle
> convergence is not met.
>
 
Exactly what happens to me. By the way, H atoms of methyl groups do the
same here. The dirty workaround I found, it is to set STEP ANG=1.0 in the
GEOMETRY section. It slows convergence if you are far from the minimum,
but, at least you have more chances to converge...

When I asked Stan about this issue, he kindly replied that they have it in
the TODO list. Maybe, if this topic becomes of general interest A'dam
could give it a somewhat higher priority (to be honest, they could be
already working hard on it...)

Regards,

Luigi

------------------------------------------------------------------------------
| Dr. Luigi Cavallo Department of Chemistry |
| Ph: ++39-81-674448 University of Naples |
| Fax: ++39-81-674442 Complesso Monte S.Angelo |
| Email cavallo_at_chemistry.unina.it Via Cintia, I-80126 Napoli, ITALY |
------------------------------------------------------------------------------
Received on 2001-04-26 17:29:55