The ABINIT Users Mailing List ( CLOSED )

[ADAMS Donat <donat.adams@cea.fr>]

Hello everyone

In order to calculate the U parameter in LDA+U I put a small potential 
on Fe using the keyword
natvshift=5   and
atvshift=-0.00367   -0.00367   -0.00367   -0.00367   -0.00367
              -0.00367   -0.00367   -0.00367   -0.00367   -0.00367
This should pull in electrons on the specified atom. But the final 
response is about one magnitude too small (The shift in the occupation 
on that particular atom compared to the calculation without shift should 
be about 0.1 e where it is 0.01043 e) . What could be the reason?

The inputfile is given below. The computation is done on bcc 
ferromagnetic iron with 2 atoms in the unit cell. The symmetry is 
lowered in order to allow differences in the occupation on the first and 
the second atom. LDA+U is activated but has no effect (except from the 
printing of the occupation number) as U and J are by default =0. The 
lattice parameter is optimized to result in a pressure abs(P) < 1 GPa. 
The calculation is converged concerning the size of tsmear.

Which factors could enhance the redistribution of the electrons?

Thanks.

Donat

=========================================

# abi.in:
     acell  3*5.19741
   chkprim  0                  # 0: do not check if uc primitive
     rprim  1.00 0.00 0.00
            0.00 1.00 0.00
            0.00 0.00 1.00

#Definition of the atom types
    ntypat  1
     znucl  26

#Definition of the atoms
     natom  2
     typat  2*1               # atomic types
      xred
            0.0 0.0 0.0
            0.5 0.5 0.5
      ecut  16                  # Energy cutoff
 pawecutdg  40                  # pawecutdg > 2*ecut
     nband  25                  # Fe_2 minband=17

#Definition of the SCF procedure

#Spin polarization
   nsppol  2                    #1 unpolarized / 2 polarized
   spinat  0 0 5   0 0 5

#Definition of the k-point grid
    kptopt  1                    # 1:  autoomatic generation of k points
     ngkpt  6 6 6                # nxnxn
   nshiftk  1
    shiftk  0.5 0.5 0.5

#Smearing
    occopt  4
    tsmear  0.001 Ha

#DFT+U
   usepawu  1                    # 1 at lim dble cnt / 2 rnd m fld dle cnt
     lpawu  2                    # ang moments corrrected
 dmatpuopt  2                    # more accurate for comp of orbital occ
      nsym  6
symrel       1  0  0   0  1  0   0  0  1       0  1  0   1  0  0   0  0  1
             0  0  1   1  0  0   0  1  0       1  0  0   0  0  1   0  1  0
             0  1  0   0  0  1   1  0  0       0  0  1   0  1  0   1  0  0

nstep 50
irdwfk 1
toldfe 10d-9 eV

natvshift 5
 atvshift     -0.00367   -0.00367   -0.00367   -0.00367   -0.00367
              -0.00367   -0.00367   -0.00367   -0.00367   -0.00367
               0.00000    0.00000    0.00000    0.00000    0.00000
               0.00000    0.00000    0.00000    0.00000    0.00000