[Pw_forum] does cell_dofree work?

Pablo Aguado paguado at gmail.com
Mon Aug 24 14:03:44 CEST 2009


Dear all,

I'm playing around with the cell_dofree variable and I don't manage to
make it work. Currently I'm trying to relax only the c lattice vector
of a tetragonal cell together with the atomic coordinates, so I use
the following:

&control
    calculation  = 'vc-relax'
    restart_mode = 'from_scratch'
    pseudo_dir   = '***/pseudo/'
    outdir       = '***/tmp'
    forc_conv_thr = 1d-3
 /
 &system
    ibrav=6
    celldm(1)=6.8
    celldm(3)=1.1
    nat=5
    ntyp=3
    nbnd=28
    ecutwfc=30.0
    occupations = 'fixed'
 /
 &electrons
    diagonalization = 'cg'
    conv_thr = 1d-8,
    mixing_beta=0.2,
 /
 &ions
    ion_dynamics = 'bfgs'
    bfgs_ndim = 5
    ion_positions = 'from_input'
 /
 &cell
    cell_dynamics = 'bfgs'
    cell_dofree = 'z'
 /
ATOMIC_SPECIES
  Pb   207.2      Pb.pz-d-van.UPF
  Ti    47.867    Ti.pz-sp-van_ak.UPF
  O     15.9994   O.pz-van_ak.UPF
ATOMIC_POSITIONS crystal
  Pb    0.000    0.000    0.040   0 0 1
  Ti    0.500    0.500    0.530   0 0 1
  O     0.000    0.500    0.500   0 0 1
  O     0.500    0.500    0.000   0 0 1
  O     0.500    0.000    0.500   0 0 1
K_POINTS {automatic}
  6 6 6 1 1 1

However, when I run PW, all the lattice vectors are relaxed. I've
checked the mail list but it's not clear to me whether this variable
is actually fully implemented or not.

Am I doing something wrong?

Thanks in advance,

Pablo


More information about the Pw_forum mailing list