Ga-N dimer

Total energy, geometry optimization, DFT+U, spin....

Moderator: bguster

Locked
zav_16
Posts: 1
Joined: Mon Mar 11, 2019 3:55 pm

Ga-N dimer

Post by zav_16 » Fri Mar 15, 2019 9:09 am

Hello,

I try to compute cohesive energy (Ecoh) and equilibrium distance (req) of Ga-N dimer. From previous computations I have results of N-N and Ga-Ga dimers which are consistent with references. But I still have wrong results for Ga-N dimer (Ecoh = 0.8eV ref. 2.4eV, req = 1.75A ref. 1.9A). The best results (according to references) I have got in run with spin polarized option (nsppol 2 - script attached) - Ecoh = 2.3eV, req = 1.75A.

Cohesive energy is computed with total energies of single atoms Ga/N in the box with the same dimensions.
All calculations were done with norm-conserving pseudopotentials from abinit webpage (https://www.abinit.org/psp-tables - Ga.psp8, N.psp8).

Thank you for any advise

Code: Select all

# Ga-N
# Structural optimalization of Ga-N dimer bonding distance


      ndtset   3

       acell   32 32 32                     # dimensions of the empty box
        ecut   50                           # maximal kinetic energy cut-off [Ha]

# First dataset: dimer Ga-N

      natom1   2                            # there are two atoms in the unit cell
     occopt1   3
     ionmov1   2                            # can move freely
     tsmear1   0.001

       xred1   0.45 0.0 0.0                 # The starting values of the
               0.55 0.0 0.0                 # atomic coordinates
      nband1   14

       typat1  1 2                          # these are  Ga N

# Second dataset: single atom of Ga

      natom2   1                            # there is one atom in the unit cell
     occopt2   2                            # Allow occupation numbers to be set by hand
      nband2   7 6                          # Number of bands for spin up and spin down   -----> in valence 4 spin up a 1 spin down
        occ2   1 1 1 1 1 1 1                # Occupation numbers for spin up state and spin down state.  ---> 1st row spin up,  2nd row spin down
               1 1 1 1 1 1 0                # dimensions (nband(1)+nband(2)+...+nband(nkpt))*nsppol

       xred2   0.5 0.0 0.0                  # The starting values
       nsym2   1                            # not spherical symmetry
     spinat2   0.0 0.0 1.0                  # Initialisation of spin

      typat2   1                            # This is Ga

# Third dataset: single atom of N

      natom3   1                            # there is one atom in the unit cell
     occopt3   2                            # Allow occupation numbers to be set by hand
      nband3   4 1                          # Number of bands for spin up and spin down   -----> in valence 4 spin up a 1 spin down
        occ3   1 1 1 1                      # Occupation numbers for spin up state and spin down state.  ---> 1st row spin up,  2nd row spin down
               1                            # dimensions (nband(1)+nband(2)+...+nband(nkpt))*nsppol

       xred3   0.5 0.0 0.0                  # The starting values
       nsym3   1                            # not spherical symmetry
     spinat3   0.0 0.0 1.0                  # Initialisation of spin

      typat3   2                            # This is N

#-----------------------------------------------------------------

      nsppol   2                            # Spin polarized calculation
      ntypat   2                            # # of atom types
       znucl   31 7                         # atomic numbers of Ga and N
       ntime   500                          # max number of iteration
     optcell   0                            # acell cannot change
      tolmxf   1.0d-4                       # Stopping criterion for the geometry optimization

# SCF calculation parameters

      toldfe   1.0d-4                       # stopping criteria for SCF
       nstep   500                          # maximal number of SCF cycles

# Definition of K-points

      kptopt   0                            # Enter the k points manually
        nkpt   1                            # Only one k point is needed for isolated system,
                                            # taken by default to be 0.0 0.0 0.0
      diemac   2.0                          # macroscopic dielectric constant (speeds up SCF calculations)

ebousquet
Posts: 469
Joined: Tue Apr 19, 2011 11:13 am
Location: University of Liege, Belgium

Re: Ga-N dimer

Post by ebousquet » Fri Mar 22, 2019 5:20 pm

Dear zav_16,
It looks indeed strange that Ga-Ga and N-N cases are working fine but not Ga-N but after quick look I do not see what could be wrong in your input.
Are the reference you mention the same for the three Ga-Ga, N-N and Ga-N calculations? Are these references using the same exchange-correlation? Do you have only one reference to compare with?
Best wishes,
Eric

Locked