NMR, Orbital shielding and Fermicontact with hybrid functionals

[arobendo mondal]

Hello all,

I am doing some test for NMR (orbital shielding and fermicontact 
(non-relativistic)) for molecules, with a large unitcell keeping 
periodicity. these are organic radicals where the spin state is doublet.

I am using, 

UNRESTRICTED_KOHN_SHAM T
MULTIPLICITY 2

setting on. 

I have compared the results with Gaussian-09. results using BLYP functional 
are in well agreement where as when I tried to use hybrid functional (PBE, 
B3LYP, BHandHLYP) the results are okay for hydrogen but for other atoms its 
deviating just randomly. 


I was trying to find reasons about the deviations, but not able to 
understand anything. In some of the older posts I saw there was problem 
with exact_exchange part but is this the reason why hybrid doesn't preform 
well for orbital shielding.


Best wishes
Arobendo  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20150513/18b1de6c/attachment.htm>