[molpro-user] DFT functionals in molpro

[Gerald Knizia]

On Wed, 2014-06-25 at 15:01 -0400, Shepard, Christopher wrote:
> Hi, I am trying to optimize benzene dimer cation using the M062X DFT
> and including Grimme's D3 dispersion function. However I can not
> figure out how to input the density functional correctly or get it to
> read the density functional.

We currently do not have the most recent version of the D3 program
included in Molpro, and the version we have does not yet support M06-2X.
Note, however, that M06-2X does not really need it as badly as other
functionals, as such effects were included in its parameterization.

Regarding input of the functional and the D3 correction: The following
input works for me:

memory,200,m;

geometry={
   C    0.0000    0.0000    0.0000
   C   -0.5800   -1.2200    0.0000
   C    0.2400   -2.4200    0.0000
   O    1.4400   -2.3600    0.0000
   O   -0.4200   -3.5800    0.0000
   H    1.0666    0.0853    0.0000
   H   -0.6072    0.8810    0.0000
   H   -1.6468   -1.3032    0.0000
   H    0.0652   -4.4083    0.0000
}

basis=tzvpp

{df-rks,b3lyp; disp,func='b3-lyp'}

{df-rks,m06-2x}


Note the different spelling of b3lyp in KS and in the D3 program, and
the single quotes...  I'll see if I can fix this.

> 
> Also I was wondering if the DFT B3LYP  was available in molpro? I
> couldn't find it in the manual. 

The documentation could probably be improved on this... it's described
under "17.4.1 Alias density functionals". The default page lists the
individual functional building blocks (which you normally never use
directly); we should probably restructure this.


Best wishes,
Gerald