[Pw_forum] Reg: Convergence of doped supercell

[Suresh A]

Dear All,
             I have carried out convergence and vc-relaxation for undoped
2*2*1 supercell of anatase.Then I replace one oxygen by one nitrogen and
did scf calculation to know the energy and forces due to doped nitrogen.
While runing code asks me to include smearing methods as follows.

 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
     Error in routine electrons (1):
     charge is wrong: smearing is needed
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%



so I include randomly gaussian smearing method with 0.01 width without
doing convergence with respect to smearing width  in my input file


 &CONTROL
                       title = '1anatase' ,
                 calculation = 'scf' ,
                restart_mode = 'from_scratch' ,
                      outdir = '/home/slave1/Desktop/n-doped-supercell/' ,
                  pseudo_dir = '/home/slave1/Desktop/n-doped-supercell' ,
                      prefix = 'anatase' ,
               etot_conv_thr = 1.0d-4 ,
               forc_conv_thr = 1.0d-3 ,
                     tstress = .true. ,
                     tprnfor = .true. ,
 /
 &SYSTEM
                       ibrav = 6,
                           A = 7.579318 ,
                           B = 7.579318 ,
                           C = 9.650253 ,
                       cosAB = 0 ,
                       cosAC = 0 ,
                       cosBC = 0 ,
                         nat = 48,
                        ntyp = 3,
                     ecutwfc = 30 ,
                     ecutrho = 300 ,
            exxdiv_treatment = 'none' ,
              occupations = 'smearing' ,
                     degauss = 0.01 ,
                    smearing = 'gaussian'
 /
 &ELECTRONS
                    conv_thr = 1.0d-10 ,
                 mixing_beta = 0.7 ,
 /
 &IONS
                ion_dynamics = 'bfgs' ,
               ion_positions = 'from_input' ,
                 phase_space = 'full' ,
           pot_extrapolation = 'atomic' ,
           wfc_extrapolation = 'none' ,
                     upscale = 100.d0 ,
                   bfgs_ndim = 1 ,
            trust_radius_max = 0.80 ,
            trust_radius_min = 1d-3 ,
            trust_radius_ini = 0.5d0 ,
 /
 &CELL
               cell_dynamics = 'bfgs' ,
                 cell_factor = 1.2 ,
                 cell_dofree = 'all' ,
 /
ATOMIC_SPECIES
   Ti   47.86700  Ti.pw91-sp-van_ak.UPF
    O   15.99940  O.pw91-van_ak.UPF
    N   14.00600  N.pw91-van_ak.UPF
ATOMIC_POSITIONS crystal
Ti    0.0000000000    0.0000000000    0.0000000000
Ti    0.2499985872    0.2499985872    0.4999971721
Ti    0.0000000000    0.2499985872    0.249998522
Ti    0.2499985872    0.0000000000    0.7499958221
Ti    0.0000000000    0.4999971744    0.0000000000
Ti    0.2499985872    0.7499957616    0.4999971721
Ti    0.0000000000    0.7499957616    0.249998522
Ti    0.2499985872    0.4999971744    0.7499958221
Ti    0.4999971744    0.0000000000    0.0000000000
Ti    0.7499957616    0.2499985872    0.4999971721
Ti    0.4999971744    0.2499985872    0.249998522
Ti    0.7499957616    0.0000000000    0.7499958221
Ti    0.4999971744    0.4999971744    0.0000000000
Ti    0.7499957616    0.7499957616    0.4999971721
Ti    0.4999971744    0.7499957616    0.249998522
Ti    0.7499957616    0.4999971744    0.7499958221
O    0.0000000000    0.2499985872    0.0430194597
O    0.0000000000    0.0000000000    0.2069790644
O    0.2499985872    0.2499985872    0.2930179727
O    0.0000000000    0.2499985872    0.4569776784
O    0.2499985872    0.0000000000    0.5430166658
O    0.2499985872    0.2499985872    0.7069763714
O    0.0000000000    0.0000000000    0.7930152798
O    0.2499985872    0.0000000000    0.9569748844
O    0.0000000000    0.7499957616    0.0430194597
O    0.0000000000    0.4999971744    0.2069790644
O    0.2499985872    0.7499957616    0.2930179727
O    0.0000000000    0.7499957616    0.4569776784
O    0.2499985872    0.4999971744    0.5430166658
O    0.2499985872    0.7499957616    0.7069763714
O    0.0000000000    0.4999971744    0.7930152798
O    0.2499985872    0.4999971744    0.9569748844
O    0.4999971744    0.2499985872    0.0430194597
O    0.4999971744    0.0000000000    0.2069790644
O    0.7499957616    0.2499985872    0.2930179727
O    0.4999971744    0.2499985872    0.4569776784
O    0.7499957616    0.0000000000    0.5430166658
O    0.7499957616    0.2499985872    0.7069763714
O    0.4999971744    0.0000000000    0.7930152798
O    0.7499957616    0.0000000000    0.9569748844
O    0.4999971744    0.7499957616    0.0430194597
O    0.4999971744    0.4999971744    0.2069790644
O    0.7499957616    0.7499957616    0.2930179727
N    0.4999971744    0.7499957616    0.4569776784
O    0.7499957616    0.4999971744    0.5430166658
O    0.7499957616    0.7499957616    0.7069763714
O    0.4999971744    0.4999971744    0.7930152798
O    0.7499957616    0.4999971744    0.9569748844
K_POINTS automatic
  4 4 2   0 0 0

and my doubts are
1.)  Should  i do convergence test with respect to smearing
methods(gaussian M-V M-P) and with respect to smearing width for N-doped
supercell as well as for  pure supercell to compare them?

2) I did convergence test for anatase unitcell and took ecut and k-point
values to supercell exactly as anatase unitcell. Is it correct?

3)How to do convergence test for doped supercell because  one do
convergence test for unitcell and take converged values exactly as unitcell
to supercell, now third atom comes in the supercell but it is not in the
unitcell?.

4) Ti   47.86700  Ti.pw91-sp-van_ak.UPF
    O   15.99940  O.pw91-van_ak.UPF
    N   14.00600  N.pw91-van_ak.UPF
Is my way of choosing pseudopotential correct? Because Ti atom is produced
with scalar relativistic correction other twos or not. And sp semicore
states of Ti atom are in valence.



              With Regards,
A.Suresh,
Project Fellow,
Madurai Kamaraj University,
Madurai.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20140527/6fb0ee0d/attachment.html>