Using the MCY3 functional in CP2K

[Leopold Talirz]

Hi Samuel,

thanks a lot for pointing this out, I should have thought of this.

>From looking at the XC_FUNCTIONAL section in these regtests, I have some 
remaining questions


   - Do I understand correctly that the &LYP_ADIABATIC together with the 
   &BECKE88 represent the first term in eq. (18) of the paper 
   http://scitation.aip.org/content/aip/journal/jcp/126/19/10.1063/1.2741248 
   (the W_{\lambda_p}^{BLYP}) ?
   - Is it possible that in the regtest 
   tests/QS/regtest-hybrid-3/Li-hybrid-MCY3.inp and Ne-hybrid-MCY3.inp the 
   scaling parameters c_2 and c_3 have been confused? According to the paper, 
   the scaling c2 for the long-range B88 exchange should be -(1+a)=-1.05601, 
   while c_3 for Hartree-Fock should be 1-a=0.94399.
   - Some indications on the meaning of the two HF sections would still be 
   greatly appreciated.


Best,
Leopold



On Wednesday, March 25, 2015 at 8:02:39 AM UTC+1, Samuel Andermatt wrote:
>
> I do not know the solution, but what could help is to have a look at the 
> files in the test section of CP2K. There are two files that use the MCY3 
> functional:
> tests/QS/regtest-hybrid-3/Li-hybrid-MCY3.inp
> tests/QS/regtest-hybrid-3/Ne-hybrid-MCY3.inp
>
> If you didn't compile CP2K yourself you can also find the files under: 
> http://sourceforge.net/p/cp2k/code/HEAD/tree/trunk/cp2k/tests/QS/regtest-hybrid-3/
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20150325/a1c2763b/attachment.htm>