DFT-D3 parameters for PBEsol

[Tobias Kraemer]

Hi all,

I am doing a few test calculations with the PBEsol functional, and it appears that D3 parameters are not implemented for this functional (I am using version 2.7).

I guess it might be an option to use the parameters from PBE, but I am not convinced this is a good idea. On a similar note, I am wondering about the specification

of the potentials in this case as well, since I don't find any for PBEsol. I have used the specific potentials for PBE, assuming that this might be a valid approximation.

The presence of Rhodium in my system seems to limit the choice of functional to PBE (which I am currently using), PADE and BLYP. So, more generally I am asking 

the question if there is a way to increase the selection. How dangerous is it to mix potentials and functionals. Perhaps using PBE with PBEsol is an option, the combination of BLYP with B3LYP has been suggested somewhere on this forum if I am not mistaken. 

This is something I have been wondering about for a while, since I would like to explore a few more functionals, but I seem to be somewhat limited in my choice. Closely interlinked is the question for the D3 parametrization. 

You suggestions are highly welcomed. 

Thanks 

Tobias  

[hut...@chem.uzh.ch]

Hi

you can find the latest complete list of parametrized functionals
for D3 on the home page of Stefan Grimme:

http://www.thch.uni-bonn.de/tc/downloads/DFT-D3/functionals.html

there are also the values for PBEsol.

It has become the standard (although not optimal) case to use PBE
pseudopotentials for all GGA functionals (and even hybrids).

regards

Juerg
--------------------------------------------------------------
Juerg Hutter                         Phone : ++41 44 635 4491
Institut für Chemie C                FAX   : ++41 44 635 6838
Universität Zürich                   E-mail: hut...@chem.uzh.ch
Winterthurerstrasse 190
CH-8057 Zürich, Switzerland
---------------------------------------------------------------

-----cp...@googlegroups.com wrote: -----To: cp2k <cp...@googlegroups.com>
From: Tobias Kraemer <161brun...@gmail.com>
Sent by: cp...@googlegroups.com
Date: 10/16/2015 11:31AM
Subject: [CP2K:7063] DFT-D3 parameters for PBEsol

Hi all,


I am doing a few test calculations with the PBEsol functional, and it appears that D3 parameters are not implemented for this functional (I am using version 2.7).I guess it might be an option to use the parameters from PBE, but I am not convinced this is a good idea. On a similar note, I am wondering about the specificationof the potentials in this case as well, since I don't find any for PBEsol. I have used the specific potentials for PBE, assuming that this might be a valid approximation.

The presence of Rhodium in my system seems to limit the choice of functional to PBE (which I am currently using), PADE and BLYP. So, more generally I am asking the question if there is a way to increase the selection. How dangerous is it to mix potentials and functionals. Perhaps using PBE with PBEsol is an option, the combination of BLYP with B3LYP has been suggested somewhere on this forum if I am not mistaken. 
This is something I have been wondering about for a while, since I would like to explore a few more functionals, but I seem to be somewhat limited in my choice. Closely interlinked is the question for the D3 parametrization. 
You suggestions are highly welcomed. 

Thanks 

Tobias  

--

You received this message because you are subscribed to the Google Groups "cp2k" group.

To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+uns...@googlegroups.com.

To post to this group, send email to cp...@googlegroups.com.

Visit this group at http://groups.google.com/group/cp2k.

For more options, visit https://groups.google.com/d/optout.



[attachment "input.inp" removed by Jürg Hutter/at/UZH]

[Tobias Kraemer]

Hello Juerg. 

as far as the dftd3.dat file is concerned, this one is left unchanged I assume? The parameters can then easily be set with the D3_SCALING keyword.

Also thanks for the clarification with regards to the choice of basis/potential. This certainly is helpful to know.

Tobias

Tobi

[S Ling]

Yes, the dftd3.dat file is the same, and you can set the D3_SCALING parameters manually in the input. Please use the correct scaling parameters either corresponding to zero damping or to BJ-damping. Please note these parameters listed on the D3 webpage are for calculations using very large basis sets in which case the contribution from basis set superposition errors (BSSEs) can be neglected. If you look at the Supporting Information of the original paper on the D3 method, the authors have provided additional scaling parameters for calculations using moderate basis sets which are more likely to be affected by BSSE. You may try both set of scaling parameters and see which gives better agreement in comparison with experiment for your particular system.

SL

--

[Tobias Kraemer]

Dear SL,

thanks for the clarification, this confirms what I thought. I am fairly familiar with the D3 method as far as more conventional quantum chemistry packages are concerned, so I wanted to make sure that I am also doing the right thing in CP2K.  

Your comment on the basis set dependence is interesting, something that I must admit I wasn't aware of. I will look into this. Thanks for pointing this out, very helpful.

Tobi