PM6 convergence with periodic boundary conditions

[Jack Evans]

Dear team,

I have been investigating the application of PM6 within the CP2K program 
for the study of periodic structures.

Currently, I understand that periodic NDDO calculations are not a standard 
application and thus require care and to begin I have been testing the 
convergence of structures defined in the regtest section. 

In particular, I have noticed the simulation of Al2O3 
(SE/regtest-3-4/Al2O3.inp) does not converge despite increasing the number 
of SCF cycles to 2000 and using a variety of optimiser settings.

I have been testing this with cp2k/2.4.0 and cp2k/2.6.2 complied with intel 
compilers.

Thank you for your help in advance.

Cheers,

Jack
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20151019/fa589154/attachment.htm>