From owner-chemistry -x- at -x- ccl.net Fri Jan 22 11:09:01 2016
From: "Christoph Bannwarth cbannwar*|*uni-bonn.de" <owner-chemistry-,-server.ccl.net>
To: CCL
Subject: CCL: HF-D dispersion software
Message-Id: <-51975-160122110627-31395-dV8PwI867S3IVPIMw2bkYw-,-server.ccl.net>
X-Original-From: Christoph Bannwarth <cbannwar .. uni-bonn.de>
Content-Transfer-Encoding: 7bit
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Date: Fri, 22 Jan 2016 17:06:06 +0100
MIME-Version: 1.0


Sent to CCL by: Christoph Bannwarth [cbannwar^uni-bonn.de]
Dear Marco,

as Diego has already pointed out, the D3 method is parametrized for 
Hartree-Fock.

In ORCA: Run ORCA with the following options:
! HF D3BJ
The BJ refers to the Becke-Johnson-damped variant of D3 (which is the 
default).

In TURBOMOLE:
Choose Hartree-Fock (i.e., do not set $dft) and specifiy in the control 
file.:
$disp3 -bj

In both programs, the dispersion energy will be added to the total 
energy (you can however find its individual contribution in the outputs 
as well).


For the plain dispersion contribution, you can also use the standalone 
dftd3 code from
Grimme's homepage:
http://www.thch.uni-bonn.de/tc/index.php?section=downloads&subsection=getd3
Just run it as
dftd3 <geometry.xyz> -func hf -bj


The HF-3c method which Diego also mentioned is a low-cost method.
It combines Hartree-Fock in a minimal basis with D3 and two geometric 
corrections for
basis set incompleteness and superposition errors.
(see: http://onlinelibrary.wiley.com/doi/10.1002/jcc.23317/abstract )
If you are interested in using this, the method is available in ORCA as 
well ( ! HF-3c ).

I hope, this helps!

Best regards,
Christoph