[CP2K-user] Classical geometry optimation: infinite loop

[Bas van Beek]

Dear all,

Recently I attempted a number of geometry optimization of nanocrystals with 
(CHARMM-based) forcefields, this, however, proved to be surprisingly 
problematic.
While using the LBFGS or CG optimizers CP2K seems, after a while, to get 
stuck in some sort of infinite loop while calculating the energy.
Tweaking the LBFGS- or CG-specific settings, unfortunately, has not 
remedied the issue.
The only thing that has worked so far is switching to BFGS, and, while more 
robust, this is optimizer is easily an order of magnitude slower (*i.e.* it 
is prohibitedly expensive).

Is this a known issue and, more importantly, are there any recommended 
solutions?
Calculations were done on both Mac (see attached files) and Unix with CP2K 
6.1.

Regards,
Bas van Beek
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20191120/ad1009c9/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: QD_opt.002.in
Type: application/octet-stream
Size: 13582 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20191120/ad1009c9/attachment.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: QD_opt.prm
Type: application/octet-stream
Size: 1849 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20191120/ad1009c9/attachment-0001.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: QD_opt.psf
Type: application/octet-stream
Size: 79308 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20191120/ad1009c9/attachment-0002.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: QD_opt.002.out
Type: application/octet-stream
Size: 507356 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20191120/ad1009c9/attachment-0003.obj>