Dispersion correction for semi-empirical methods (PM6)

[Jan-Niclas Luy]

Dear developers,

I would like to test the applicability of semi-empirical methods for fast 
structure screening. I see that PM6 is implemented in CP2K.
In addition, some form of dispersion correction is often necessary for 
larger systems. Is it possible to use the D3 method by Grimme in 
conjunction with semi-empirical methods like PM6?

Considering that S. Grimme has already determined the parameters I would 
assume that supplying them to the input of CP2K would not be too difficult.

see for reference the SI of https://doi.org/10.1002/chem.201200497

Is there a way to do this similar to DFTB or with the VDW_POTENTIAL key?

Best regards
Jan-Niclas Luy
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20180529/2847589d/attachment.htm>