The ABINIT Users Mailing List ( CLOSED )

[Nuno Galamba <ngalamba@cii.fc.ul.pt>]

I believe you misunderstood the meaning of nband.
You should check how many valence electrons are included in the 
pseudopotentials you are using. Then divide the sum of all the valence 
electrons of all atoms in your unit cell by two. You'll get a number close to 
the number of occupied states (depending on odd or even number of valence 
electrons). Add some more bands for virtual orbitals (bands) and I think your 
all set.

Four bands can only be occupied up to a max of 8 electrons. I am sure your 
unit cell has a lot more valence electrons.

Nuno Galamba











On Wednesday 19 April 2006 13:06, jmlv@imr.ac.cn wrote:
> Dear everyone:
>    I tried to get the equilbrium structure of TiNi unit cell (B2-structure,
> 2 atoms/unit cell). But I always got a ERROR notice.
>
>    First, I used the default of nband=1. the ERROR says:
>     scfcge: ERROR -
>   Potential-based CG line minimization not converged after 13 restarts.
>   Action: read the eventual warnings about lack of convergence.
>   Some might be relevant. Otherwise, raise nband.
>
>    Then I followed what the code told me, changing the nband to 4, the
> usual value for metals. There is an ERROR again, saying: chkneu: ERROR -
>   Initialization of occ, with occopt=   3
>   There are not enough bands to get charge balance right
>   Action: modify input file...
>   (check the pseudopotential charges, the variable charge,
>   and the declared number of bands, nband)
>
>    What am I gonna do? Is there something wrong with the setting of nband?
> Any suggestion would be appreciated. Thanks in advance.
>
>    The GGA-FHI pseudopotentials were used. The input file goes:
>
> # Relaxation of TiNi B2 phase unit cell
>
> #Definition of the unit cell
> acell  3*5.67        # Lattice constants.
> rprim  1.0 0.0 0.0   #
>        0.0 1.0 0.0   #
>        0.0 0.0 1.0   #
>
> #Definition of the atom types
> ntypat 2          # Type of constituent atoms.
> znucl 22 28       # Atomic number of each atom.
>
> #Definition of the atoms
> natom 2           # There are two atoms.
> typat 1 2         # One is Ti atom and another is Ni atom.
> xred              # Location of the atoms
>    0.0  0.0  0.0  # Triplet giving the REDUCED coordinate of atom 1.
>    0.5  0.5  0.5  # Triplet giving the REDUCED coordinate of atom 2.
>
> #Definition of the planewave basis set
> ecut  15.0        # Maximal kinetic energy cut-off, in Hartree
>
> #Definition of the k-point grid
> kptopt 1          # Automatic generation of k points, taking
>                   # into account the symmetry
> ngkpt 6 6 6       # This is a 2x2x2 grid based on the primitive vectors
> nshiftk 4         # of the reciprocal space (that form a BCC lattice !),
>                   # repeated four times, with different shifts:
> shiftk
>   0.5 0.5 0.5     # In cartesian coordinates, this grid is simple cubic,
> and 0.5 0.0 0.0     # actually corresponds to the
>   0.0 0.5 0.0     # so-called 4x4x4 Monkhorst-Pack grid
>   0.0 0.0 0.5
>
> #Definition of optimization parameters
> optcell    1         # Full optimization of cell geometry.
> ecutsm     0.5       # Default is 0.0. When optcell/=0, it should be larger
> than zero. strprecon  0.001     # Default is 1.0. But in this case, it is
> decreased. ionmov     2         # Use the Broyden algorithm.
> dilatmx    3.0       # Default is 1.0. Here it is increased to 1.1.
> ntime      30        # Max. number of Broyden "time steps".
> tolmxf     1.0E-5    # Stopping criterion for geometry optimization.
>
> occopt     3      #
> tsmear     0.04   #
> ixc        11     #
> nband      4      # Changed from 1(the default) to 4(usually for metals)
> after ABINIT's notice.
>
> #Definition of the SCF procedure
> nstep      100    # Maximal number of SCF cycles
> toldfe     1.0E-6 #
> diemac     1.0E6  #
>
> #Output control
> prtwf    0        # Do not creat the HUGE wavefunction file. Default is 1.

-- 
Nuno Galamba
Grupo de Fisica-Matematica da Universidade Lisboa
Lisboa
Portugal