The ABINIT Users Mailing List ( CLOSED )

[wei cheng <chengwei0@hotmail.com>]

Hi,
I tried to calculate phonon dispersions of 8 atom Si, but without symmetry.
I always get following errors using mrgddb. How can I do?

read the input derivative database number     2
 compare the current and input DDB information
 chki8 : ERROR -
  Comparing integers for variable  nkpt.
  Value from input DDB is        36 and
        from transfer DDB is        64.
  Action : check your DDBs.

My inputs are:

   chkprim 0
   ndtset   10

  getwfk1   0            # Cancel default
  kptopt1   1            # Automatic generation of k points, taking
                         # into account the symmetry
    nqpt1   0            # Cancel default
  tolvrs1   1.0d-18      # SCF stopping criterion (modify default)
  rfphon1   0            # Cancel default
 
     nqpt   1            # One qpt for each dataset (only 0 or 1 allowed)

     qpt2   0.0 0.0 0.0
     qpt3  0.0 0.0 0.0
     qpt4  0.0 0.0 0.5
     qpt5  0.5 0.0 0.0
     qpt6  0.0 0.5 0.0
     qpt7  0.0 0.5 0.5
     qpt8  0.5 0.0 0.5
     qpt9  0.5 0.5 0.0
     qpt10 0.5 0.5 0.5

    iscf2   -3         # Need this non-self-consistent option for d/dk
  kptopt2   2          # Modify default to use time-reversal symmetry
  rfphon2   0          # Cancel default
  rfelfd2   2          # Calculate d/dk wave function only
  tolvrs2   0.0        # Cancel default for d/dk
  tolwfr2   1.0d-22    # Use wave function residual criterion instead


  getddk3   2          # d/dk wave functions from last dataset
  kptopt3   2          # Modify default to use time-reversal symmetry
  rfelfd3   3          # Electric-field perturbation response only

   getwfk   1          # Use GS wave functions from dataset1
   kptopt   3          # Need full k-point set for finite-Q response
   rfphon   1          # Do phonon response
  rfatpol   1 8        # Treat displacements of all atoms
    rfdir   1 1 1      # Do all directions (symmetry will be used)
   tolvrs   1.0d-8     # This default is active for sets 3-10

    acell   3*10.25         # This is equivalent to   10.61 10.61 10.61
    rprim   1.0  0.0  0.0   # In lessons 1 and 2, these primitive vectors
            0.0  1.0  0.0   # (to be scaled by acell) were 1 0 0  0 1 0  0 0 1
            0.0  0.0  1.0   # that is, the default.

   ntypat   1         # There are two types of atom
    znucl   14        # The keyword "znucl" refers to the atomic number of the
    natom   8         # There are two atoms
    typat   1 1 1 1 1 1 1 1        # The first is of type 1 (Al), the second is of type 2 (As).
                      
     xred   0.0  0.0  0.0
            0.5  0.5  0.0
            0.5  0.0  0.5   
            0.0  0.5  0.5
            0.25 0.25 0.25
            0.75 0.75 0.25
            0.75 0.25 0.75
            0.25 0.75 0.75

    nband   18        

      ixc   1             # LDA Teter Pade parametrization

     ecut   30.0           # Maximal kinetic energy cut-off, in Hartree

    ngkpt   4  4  4         
  nshiftk   1              # Use one copy of grid only (default)
   shiftk   0.0 0.0 0.0    # This gives the usual fcc Monkhorst-Pack grid

     iscf   5          # Self-consistent calculation, using algorithm 5
    nstep   25         # Maximal number of SCF cycles
   diemac   12.0        # Although this is not mandatory, it is worth to

---

Windows Live™: Keep your life in sync. Check it out.