[Pw_forum] Pseudopotentials for non-local optimized vDW-DF functionals

[Ary Junior]

Hi,

The file Modules/funct.f90 is well commented and the user is invited to
read it in the documentation. For example, if you chose input_dft =
'vdw-df', the internal indices will be forced to be

    iexch = 1 ! Exchange
    icorr = 4 ! Correlation
    igcx = 4 ! Gradient Correction on Exchange
    igcc = 0 ! Gradient Correction on Correlation
    inlc = 1 ! Van der Waals functionals (nonlocal term only)
    imeta=0 ! Meta-GGA functionals

When generating your PP with the ld1.x code, this same module is used and
if you set dft = 'PBE' you have

    iexch = 1 ! Exchange
    icorr = 4 ! Correlation
    igcx = 3 ! Gradient Correction on Exchange
    igcc = 4 ! Gradient Correction on Correlation
    inlc = 0 ! Van der Waals functionals (nonlocal term only)
    imeta = 0 ! Meta-GGA functionals

And if you set dft = 'revBPE' you have

    iexch = 1 ! Exchange
    icorr = 4 ! Correlation
    igcx = 4 ! Gradient Correction on Exchange
    igcc = 4 ! Gradient Correction on Correlation
    inlc = 0 ! Van der Waals functionals (nonlocal term only)
    imeta = 0 ! Meta-GGA functionals

So, notice that if you use a PP generated with dft = 'PBE' and set
input_dft = 'vdw-df' in your pw.x input file, the gradient correction on
exchange used in your solid-state calculation and the one used to generate
the PP are not the same. So, I think you must generate you PP with dft =
'revBPE'. The gradient correction on correlation will be the same of PBE,
but in the solid-state calculation it will be replaced by the non-local
correlation contribution as implemented in the file Modules/xc_vdW_DF.f90.
Try to follow the same logic for other values of input_dft. With respect to
the PPs, you can start from the pslibrary

http://www.quantum-espresso.org/pseudopotentials/pslibrary/

Ary Ferreira

FAPESP postdoctoral fellow
UFSCar - Brazil


On Mon, Jun 5, 2017 at 2:13 PM, Vikan Manmathan <vikanmanmathan1 at gmail.com>
wrote:

> Dear Developers and Experts,
>
> I am interested in studying the adsorption of aromatic molecules on the
> metal surfaces. From the literature, it is shown that the van der Waal's
> interactions play a significant role in such systems and the optimized
> versions of the vDW-DF functional best describe the structural and
> electronic features. More details can be found on these reports, (J. Chem.
> Phys. 140, 084704 (2014), J. Phys. Chem. C 2015, 119, 1886−1897 and J.
> Phys. Chem. Lett. 2016, 7, 2228−2233).
>
> My question is regarding the pseudopotentials.
> Is it a proper way to do the calculations by taking the PBE
> pseudopotential (available at the quantum Espresso website) and include the
> van der Waal's interaction using input_dft flag?
>
> Any help is highly appreciated.
>
> Thanks in advance.
>
>
> Regards,
>
>
>
> *NANDHA KUMAR V.*
> Research scholar,
> IISER, PUNE.
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://pwscf.org/mailman/listinfo/pw_forum
>



-- 
http://lattes.cnpq.br/8221674673413336
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20170606/4a08e4b3/attachment.html>