Grimme dispersion corrections for Hessian with .VIBANA?

Find answers or ask questions regarding Dalton calculations.
Please upload an output file showing the problem, if applicable.
(It is not necessary to upload input files, they can be found in the output file.)

Post Reply
ringholm
Posts: 6
Joined: 24 Feb 2014, 15:44
First name(s): Magnus
Last name(s): Ringholm
Affiliation: CTCC - UiT
Country: Norway

Grimme dispersion corrections for Hessian with .VIBANA?

Post by ringholm » 25 Apr 2018, 11:50

Dear all,

I want to do a vibrational analysis with .VIBANA using Grimme's dispersion corrections like e.g. in the following input:

Code: Select all

**DALTON INPUT
.RUN PROPERTIES
.PARALLEL
**WAVE FUNCTION
.DFT
 PBE0
*DFT INPUT
.DFTD3
**PROPERTIES
.VIBANA
*VIBANA
.HESPUN
**END OF INPUT
I notice that the manual says that at least for .DFTD3 and .DFD3BJ, which I think are the most relevant for me to use, that analytic gradient contributions are available (but I assume no analytic Hessian contributions). Since .VIBANA entails evaluating the Hessian, I wonder:

How are the dispersion corrections taken into account when I calculate the Hessian with .VIBANA? Would I get a (finite difference or otherwise) full dispersion correction to the Hessian or something else? That is, is for instance the gradient or the energy contribution differentiated numerically? Would there be any neglect of dispersion contributions to the Hessian?

ringholm
Posts: 6
Joined: 24 Feb 2014, 15:44
First name(s): Magnus
Last name(s): Ringholm
Affiliation: CTCC - UiT
Country: Norway

Re: Grimme dispersion corrections for Hessian with .VIBANA?

Post by ringholm » 26 Apr 2018, 10:28

Update: I have now run tests with a medium-size system (thymine + three proximate water molecules) with and without the D3 dispersion corrections, and it doesn't look like the vibrational frequencies or those elements of the Hessian that I looked at have changed between those runs above what I assume to be machine precision. So I guess this means that the corrections are not included in the vibrational analysis.

ringholm
Posts: 6
Joined: 24 Feb 2014, 15:44
First name(s): Magnus
Last name(s): Ringholm
Affiliation: CTCC - UiT
Country: Norway

Re: Grimme dispersion corrections for Hessian with .VIBANA?

Post by ringholm » 26 Apr 2018, 16:05

Another update: I was using an older version of Dalton for the last post and Jógvan Magnus Olsen told me that there may be changes in a new version. There are still no D3 corrections to the Hessian in the most recent version, but it is now caught by the program if you try to calculate that.

The conclusion here could well be that I implement this myself.

Post Reply

Who is online

Users browsing this forum: No registered users and 1 guest