[Pw_forum] Implementation of Van der Waals Interactions

[Giuseppe Mattioli]

Dear Manuel
Quantum ESPRESSO uses several approaches to simulate, in an ab initio as well as in a semiempirical fashion, the dispersion interactions. However, the 
information on this subject is still far from being well organized...:-)

As you already know, the semiempirical DFT-D2 approach is invoked by the following keywords, which I copy from the PW manual

   +--------------------------------------------------------------------
   Variable:       london

   Type:           LOGICAL
   Default:        .FALSE.
   Description:    if .TRUE. compute semi-empirical dispersion term (DFT-D).
                   See S. Grimme, J. Comp. Chem. 27, 1787 (2006), and
                   V. Barone et al., J. Comp. Chem. 30, 934 (2009).
   +--------------------------------------------------------------------

   +--------------------------------------------------------------------
   Variable:       london_s6

   Type:           REAL
   Default:        0.75
   Description:    global scaling parameter for DFT-D. Default is good for PBE.
   +--------------------------------------------------------------------

   +--------------------------------------------------------------------
   Variable:       london_rcut

   Type:           REAL
   Default:        200
   Description:    cutoff radius (a.u.) for dispersion interactions

As you see, the london_s6=0.75 default is considered good for PBE. Without further information about your input file this seems to be the only source 
of discrepancy between Persson's results and yours.

Regarding the (more or less) ab initio approaches, the stable 5.0.2 version (but also 4.3.2, if I remember well) contains the long-range VDW-DF 
correlation functional in different flavours. They are invoked by the keyword

input_dft='XXXX'

A complete list of functionals can be found in this file:

YOURQE/Modules/funct.f90

from where the following values of the input_dft variable are extracted

  !              "vdw-df"= "sla+pw+rpb+vdw1"   = vdW-DF
  !              "vdw-df2"="sla+pw+rw86+vdw2"  = vdW-DF2
  !              "vdw-df-c09"="sla+pw+c09x+vdw1"
  !              "vdw-df2-c09"="sla+pw+c09x+vdw2"

Flick through the corresponding documentation to check which kind of pseudopotential is required/suggested for the above functionals.
More choice is allowed in the unstable (but completely working...) svn version. The Thatchenko-Scheffler method is now implemented among the cp.x 
options, while the XDM method (for PAW pseudopotentials only) is a new feature of pw.x.

HTH

Giuseppe


On Tuesday 19 November 2013 20:44:38 Manuel Otero wrote:
> I am a new user of Quantum-Espresso (QE) and I am trying to study the
> adsorption energy of lithium ions in graphite. I want to reproduce some
> calculations from other people (Persson), to ensure that I am using the
> right parameters.
> 
> In this system the Van der Waals (vdw) interactions play an important role
> and I want to know more about vdw in QE.
> In her work Persson (Nano Lett., 2012, 12 (9), pp 4624–4628) says that
> there are two approacches to include vdw interactions:
> - "DFT-D2 approach, which adds a semiempirical pairwise force field"
> - "vdW-DF approach, which adds a nonlocal correlation functional that
> approximately account for dispersion interactions"
> which approach use by QE?
> 
> I am using espresso-4.3.2. In my calculations I use:
> 
> &SYSTEM
> ...
>   london = .true.,
>   london_rcut =200,
>   london_s6 =0.75,
> 
> and
> 
> ATOMIC_SPECIES
>  C  12.0107    C.pw91-van_ak.UPF
> Li   6.939    Li.pw91-n-van.UPF
> 
> I cant reproduce Persson calculations, but I dont know where is the
> problem.
> 
> It is OK to use this parametres with this pseudopotentials?
> I have tryed changing the parametres but my calculations do not improve.
> 
> QE enables to change other parameters?
> 
> Thanks in advance for any help.
> 
> Manuel


********************************************************
- Article premier - Les hommes naissent et demeurent
libres et ègaux en droits. Les distinctions sociales
ne peuvent être fondèes que sur l'utilitè commune
- Article 2 - Le but de toute association politique
est la conservation des droits naturels et 
imprescriptibles de l'homme. Ces droits sont la libertè,
la propriètè, la sùretè et la rèsistance à l'oppression.
********************************************************

   Giuseppe Mattioli                            
   CNR - ISTITUTO DI STRUTTURA DELLA MATERIA   
   v. Salaria Km 29,300 - C.P. 10                
   I 00015 - Monterotondo Stazione (RM)          
   Tel + 39 06 90672836 - Fax +39 06 90672316    
   E-mail: <giuseppe.mattioli at ism.cnr.it>
   ResearcherID: F-6308-2012