ABINIT, Van der Waals input variables:

Used when vdw_xc>0, to read previously calculated vdW-DF variables.
Supported values:

• 0 => do not read vdW-DF variables
• 1 => read vdW-DF variables

Sets a threshold for the energy gradient that, when reached, will cause the vdW-DF interactions to be calculated.
Adjust it to a big value (e.g. 1e12) to enable it all along the SCF calculation. Too small values, as well as negative values, will result in the vdW-DF energy contributions never being calculated.

The absolute value of vdw_nfrag is the number of vdW interacting fragments in the unit cell. As wannierization takes place in reciprocal space, the MLWF center positions could be translated by some lattice vector from the cell where atoms are placed. If vdw_nfrag >= 1 then MLWFs are translated to the original unit cell, otherwise the program will keep the positions obtained by Wannier90. The later is usually correct if some atoms are located at the corners or at limiting faces of the unit cell.

Set of dimensionless positive numbers which define the maximum multiples of the primitive translations (rprimd) in the supercell construction. Each component of vdw_supercell indicates the maximum number of cells along both positive or negative directions of the corresponding primitive vector i.e. the components of rprimd. In the case of layered systems for which vdW interactions occur between layers made of tightly bound atoms, the evaluation of vdW corrections coming from MLWFs in the same layer (fragment) must be avoided. Both a negative or null value for one component of vdw_supercell will indicate that the corresponding direction is normal to the layers.

The DFT-D methods (S. Grimme approach) dispersion potentials, vdw_xc==5 or 6 or 7, include a pair potential. The number of pairs of atoms contributing to the potential is necessarily limited. To be included in the potential a pair of atom must have contribution to the energy larger than vdw_tol.

Control the computation of the 3-body correction inside DFT-D3 dispersion correction (Grimme approach) to the total energy:
-If vdw_tol_3bt<0, no 3-body correction.
-If vdw_tol_3bt>0, the 3-body term is included with a tolerance = vdw_tol_3bt

DFT-D3 as proposed by S. Grimme adds two contributions to the total energy in order to take into account of the dispersion:

• A pair-wise potential for which the tolerance is controlled by vdw_tol


• A 3-body term which is obtained by summing over all triplets of atoms. Each individual contribution depends of the distances and angles between the three atoms. As it is impossible to sum over all the triplets in a periodic system, one has to define a stopping criterium which is here that an additional contribution to the energy must be higher than vdw_tol_3bt

This array defines the interacting fragments by assigning to each atom an integer index from 1 to vdw_nfrag . The ordering of vdw_typfrag is the same as typat or xcart. Internally each MLWF is assigned to a given fragment by computing the distance to the atoms. MLWFs belong to the same fragment as their nearest atom. The resulting set of MLWFs in each interacting fragment can be found in the output file in xyz format for easy visualization.

Selects a van-der-Waals density functional to apply the corresponding correction to the exchange-correlation energy. If set to zero, no correction will be applied.
Possible values are:

• 0: no correction.
• 1: apply vdW-DF1 (DRSLL) from Dion et al.
  doi:10.1103/PhysRevLett.92.246401
• 2: apply vdw-DF2 (LMKLL) from Lee et al.
  arXiv:1003.5255v1
• 5: apply vdw-DFT-D2 as proposed by S. Grimme (adding a semi-empirical dispersion potential)
  Available only for ground-state calculations and response functions; see vdw_tol variable to control convergency
  J. Comp. Chem. 27, 1787 (2006)
• 6: apply vdw-DFT-D3 as proposed by S. Grimme (refined version of DFT-D2)
  Available only for ground-state calculations and response functions; see vdw_tol variable to control convergency and vdw_tol_3bt variable to include 3-body corrections
  J. Chem. Phys. 132, 154104 (2010)
• 7: apply vdw-DFT-D3(BJ) as proposed by Grimme (based on Becke-Jonhson method J. Chem. Phys. 2004-2006)
  Available only for ground-state calculations and response functions; see vdw_tol variable to control convergency
  J. Comput. Chem. 32, 1456 (2011)
• 10: evaluate the vdW correlation energy from maximally localized Wannier functions, as proposed by P. L. Silvestrelli, also known as vdW-WF1 method.
  doi:10.1103/PhysRevLett.100.053002. For details on this implementation please check: doi:10.1016/j.cpc.2011.11.003
  The improvements introduced by Andrinopoulos et al. in J. Chem. Phys. 135, 154105 (2011) namely the amalgamation procedure, splitting of p-like MLWFs into
  two s-like Wannier functions and fractional occupation of MLWFs are performed automatically.
  
• 11: evaluate the vdW correlation energy from maximally localized Wannier functions, as proposed by A. Ambrosetti and P. L. Silvestrelli, also known as vdW-WF2 method.
  doi:10.1103/PhysRevB.85.073101
• 14: apply DFT/vdW-QHO-WF method as proposed by Silvestrelli, which combines the quantum harmonic oscillator-model with localized Wannier functions.
  J. Chem. Phys. 139, 054106 (2013)
  For periodic systems a supercell approach has to be used since vdw_supercell is not enabled in this case.