[QE-users] vdW nonlocal term

[Fabrizio Ferrari Ruffino]

Hello,
'VDW1' is a Van-der-Waals contribution only, in order to get a consistent dft you should also add the LDA/GGA terms, for example 'sla+pw+revx+vdw1'.
Regarding the energy graph, I would try a smaller convergence threshold (conv_thr) first.
Cheers,
Fabrizio

CNR-IOM
________________________________
From: users <users-bounces at lists.quantum-espresso.org> on behalf of Buckova, Nina <nina.buckova at tuwien.ac.at>
Sent: Tuesday, May 28, 2024 2:18 PM
To: users at lists.quantum-espresso.org <users at lists.quantum-espresso.org>
Subject: [QE-users] vdW nonlocal term


Dear all,


I would like to carry out some SCF calculations using the vdW-df-cx density functional with Quantum Espresso and analyze the contributions of the semi-local part of the functional (the "sla+pw+cx13" part) and the nonlocal vdW term ("vdW1") separately. However, while I can run SCF calculations using "input_dft = 'sla+pw+cx13'" without any problem, the calculation using the nonlocal term only ("input_dft = 'vdW1'") results in the following error:


 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
     Error in routine set_dft_from_name (1):
     VDW1: strange dft, please check
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

I also considered and tried calculating the nonlocal term by simply carrying out an SCF calculation using the full vdW-df-cx functional and then subtracting the XC energy obtained for the semi-local part of the functional from the XC energy of the full vdW-df-cx functional (full vdW functional XC energy - "sla+pw+cx13" only XC energy). However, this results in a nonlocal contribution that looks strange. As an illustration of this, I attach a plot showing the nonlocal term calculated by the approach described above. The system for which I calculated it consisted of two water molecules placed at various distances from each other (expressed by the scaling factor). I compared the nonlocal part obtained by Quantum Espresso to that obtained using VASP (it's the "Total vdW correction in eV" which is printed out automatically in the vasp.out file). While the VASP nonlocal correction part decreases smoothly with the decreasing distance between the water molecules, the nonlocal term from Quantum Espresso decreases less smoothly and much less. This suggests that it is probably not correct to calculate the nonlocal vdW term by the method described above.

[cid:cc045084-7643-42c2-9f99-f0216604dd7e]


Please, is it possible to somehow calculate the vdW nonlocal term contribution only using Quantum Espresso?

The version of quantum espresso that I'm using is v.7.0.

I send you an example of an expresso.in file that I'm using as well:


&CONTROL
   calculation      = 'scf'
   prefix           = 'water_rpbe'
   pseudo_dir       = '/home/nbuckova/sys_admin/qe_pseudo/oncv'
/
&SYSTEM
   ibrav            = 0
   ecutwfc          = 110
   occupations      = 'smearing'
   degauss          = 0.05
   smearing         = 'gaussian'
   nspin            = 1
   input_dft        = 'vdw-df-cx'
   ntyp             = 2
   nat              = 6
/
&ELECTRONS
   electron_maxstep = 300
   conv_thr         = 1e-06
   diagonalization  = 'cg'
/
&IONS
/
&CELL
/

ATOMIC_SPECIES
H 1.008 H_ONCV_PBE-1.2.upf
O 15.999 O_ONCV_PBE-1.2.upf

K_POINTS automatic
1 1 1  0 0 0

CELL_PARAMETERS angstrom
23.46658708700000 0.00000000000000 0.00000000000000
0.00000000000000 21.29507073200000 0.00000000000000
0.00000000000000 0.00000000000000 21.51152688400000

ATOMIC_POSITIONS angstrom
H 10.0000000000 11.2950707320 10.7329347260
H 11.2797142860 10.4904164460 10.7496424390
H 13.4665870870 10.0378329090 11.5115268840
H 13.4622297890 10.0000000000 10.0000000000
O 10.3199971700 10.3922346940 10.7543657020
O 13.0899654720 10.4762129750 10.7454025020



Thank you in advance for your help !


Best regards,


Nina Buckova, phD candidate, Vienna University of Technology

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20240528/cdc8fafa/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: pastedImage.png
Type: image/png
Size: 32456 bytes
Desc: pastedImage.png
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20240528/cdc8fafa/attachment.png>