The ABINIT Users Mailing List ( CLOSED )

["Sergey Lisenkov" <proffess@yandex.ru>]

Dear Abinit authors and users,

 I am performing test calculations of SrRuO3 cubic structure (typical 
metallic ferromagnet). But the convergence of the total energy versus ecut 
parameter is not smooth: 
Ecut (Ha)         Total energy (eV)
10            -4.59474385225766E+03
15            -4.67199387156161E+03
20            -4.70104353549418E+03
25            -4.71164792907798E+03
30            -4.71331050600648E+03
35            -4.71490606273784E+03
40            -4.71467977186843E+03
45            -4.71344010962546E+03
50            -4.71464118709542E+03
55            -4.71602398195787E+03
60            -4.71574494759242E+03
65            -4.71555642052624E+03
70            -4.71583844398062E+03
75            -4.71621802205639E+03
80            -4.71591485993982E+03
85            -4.71627156970371E+03
90            -4.71468636042394E+03
95            -4.71453759557438E+03
100           -4.71552711772037E+03
105           -4.71482898978000E+03

I tried different occupation options (occopt from 3 to 7, tsmear 0.1 eV), but 
the results are the same. Here is the part of my input:

optcell 0
ionmov  2
ntime  200         # Number of ionic steps
ecutsm  0.5

#Definition of the k-point grids
kptopt 1          # Option for the automatic generation of k points, taking
                  # into account the symmetry
nshiftk 1
shiftk  0.5 0.5 0.5

ngkpt 18 18 18

iscf      7     #  Pulay mixing of the potential based on the npulayit 
previous iterations
npulayit  10

# Gaussian smearing and degauss

occopt    7     # Gaussian smearing
tsmear 0.1 eV   #  degauss in Ha which is equal 0.1 eV (taken from PRB 71, 
014113 (2005)
nband  36

# Because this is a spin-polarization calculation,
# spin on atoms are given
spinat   0.0 0.0   0.0
         0.0 0.0  10.0
         0.0 0.0   0.0
         0.0 0.0   0.0
         0.0 0.0   0.0

nsppol   2  #  spin-polarized case
nspinor  1  #  scalar wavefunction
nspden   2  #  scalar density



#Definition of the unit cell, rhombohedral lattice

acell 3*3.93  angstrom         # lengths of the primitive vectors


rprim   1.0000   0.0000   0.00000
        0.0000   1.0000   0.00000
        0.0000   0.0000   1.00000

#******************initial atomic coordinates********************
xred                # coordinates in rprim
              0.0000000000E+00  0.0000000000E+00  0.0000000000E+00
              5.0000000000E-01  5.0000000000E-01  5.0000000000E-01
              0.0000000000E+00  5.0000000000E-01  5.0000000000E-01
              5.0000000000E-01  0.00000000000000  5.0000000000E-01
              5.0000000000E-01  5.0000000000E-01  0.0000000000E+00
#*****************************************************************

Definition of the atom types
ntypat 3             #  Three types of atoms
znucl 38 44 8        #  Bismuth, iron and oxygen


#Definition of the atoms
natom 5              # There are 10 atoms
typat 1 2 3 3 3     # 1 Sr, 1 Ru and 3 O



#**********************************************************************
#Definition of the planewave basis set
ecut: 10 ecut+ 5

#Definition of the SCF procedure
nstep 100          # Maximal number of SCF cycles

toldff  5.0d-6    # Will stop when this tolerance is achieved on forces
tolmxf  5.0d-5    # precondition the SCF cycle. The model dielectric
                  # function used as the standard preconditioner
                  # is described in the "dielng" input variable section.
                  # Here, we follow the prescription for bulk silicon.





ixc 1      # LDA-CA with spin-polarization


Any hints?

 Thanks,
   Sergey