[Pw_forum] Problem in VC-relaxation with fixed atomic positions

[Marcos Veríssimo Alves]

Hi all,

I am trying to perform a cell optimization for graphene in which I would
like to fix the atomic positions so as to have arbitrary bond lengths, and
see what happens to the in-plane cell vectors. Thus I would like to keep
the atomic positions fixed and let the in-plane cell vectors change.

I am using vc-relax with cell_dofree="2dxy" and I am (theoretically) fixing
atomic positions (specified in Angstrom) with "0 0 0" after the cartesian
coordinates. The problem is, the coordinates do not remain fixed during the
cell optimization:

ATOMIC_POSITIONS (angstrom)
C        0.000000000   0.000000000   7.000000000    0   0   0
C        1.508583432   0.000000000   7.000000000    0   0   0
--
ATOMIC_POSITIONS (angstrom)
C        0.000000000   0.000000000   7.000000000    0   0   0
C        1.553137965   0.000000000   7.000000000    0   0   0
--
ATOMIC_POSITIONS (angstrom)
C        0.000000000   0.000000000   7.000000000    0   0   0
C        1.583982256   0.000000000   7.000000000    0   0   0
--
ATOMIC_POSITIONS (angstrom)
C        0.000000000   0.000000000   7.000000000    0   0   0
C        1.592677072   0.000000000   7.000000000    0   0   0
--
ATOMIC_POSITIONS (angstrom)
C        0.000000000   0.000000000   7.000000000    0   0   0
C        1.597484048   0.000000000   7.000000000    0   0   0
--
ATOMIC_POSITIONS (angstrom)
C        0.000000000   0.000000000   7.000000000    0   0   0
C        1.599092143   0.000000000   7.000000000    0   0   0
--
ATOMIC_POSITIONS (angstrom)
C        0.000000000   0.000000000   7.000000000    0   0   0
C        1.598936919   0.000000000   7.000000000    0   0   0
--
ATOMIC_POSITIONS (angstrom)
C        0.000000000   0.000000000   7.000000000    0   0   0
C        1.598936919   0.000000000   7.000000000    0   0   0

My input is as follows:

&control
    calculation='vc-relax'
    restart_mode='from_scratch',
    prefix='graphene',
    pseudo_dir = '/home/mverissi/pseudos_espresso',
    outdir='./'
/
&system
    ibrav=0,
    celldm(1)=4.073139044,
    nat=2,
    ntyp=1,
    nspin = 1,
    ecutwfc = 28.0,
    ecutrho = 252.0,
    occupations='smearing',
    smearing='methfessel-paxton',
    degauss=0.001,
    nbnd=10,
/
&electrons
    conv_thr = 1.0e-9,
    mixing_beta = 0.7
/
&ions
    ion_dynamics='bfgs'
/
&cell
    cell_dynamics = 'bfgs',
    cell_dofree = '2Dxy',
/
ATOMIC_SPECIES
C   12.0107    C.pbe-rrkjus.UPF
ATOMIC_POSITIONS {angstrom}
C    0.0     0.0   7.0    0  0  0
C    1.42    0.0   7.0    0  0  0
K_POINTS {automatic}
       24   24   1  0  0  0
CELL_PARAMETERS {alat}
   0.866025404   0.5  0.0
   0.866025404  -0.5  0.0
   0.000000000   0.0  6.0

Am I making some silly mistake here? The version of Espresso in use is
5.0.2. Sorry if this has already come up, but I couldn't find anything
similar to my problem.

Best regards,

Marcos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20140605/abf2466b/attachment.html>