[Pw_forum] Problem with ibrav=0 in supercell calculations

[Mark Mazar]

Your CELL PARAMETERS flag is missing its underscore.

On Fri, Jun 4, 2010 at 2:57 PM, DHRUV SINGH <dhruv.singh at gmail.com> wrote:
> Hi everyone,
> I am trying to do a supercell calculation of InAs (zinc blende structure)
> using the following input file(ibrav=0 and specifying cell parameters with
> cubic symmetry)
>  &control
>     calculation='scf'
>     restart_mode='from_scratch',
>     tstress = .true.
>     tprnfor = .true.
>     prefix='inas',
>     pseudo_dir = '/home/ba01/u110/singh36/espresso-4.1.3/pseudo/',
>     outdir='./tmp'
>  /
>  &system
>     ibrav=0,
>     celldm(1)=11.05,
>     nat=  8,
>     ntyp= 2,
>     ecutwfc =25.0
>     ecutrho=250.0
>  /
>  &electrons
>     conv_thr =  1.0d-8
>     mixing_beta = 0.7
>  /
> ATOMIC_SPECIES
>  In  114.818  In.pz-bhs.UPF
>  As  74.921  As.pz-bhs.UPF
> CELL PARAMETERS{cubic}
>  1.0 0.0 0.0
>  0.0 1.0 0.0
>  0.0 0.0 1.0
> ATOMIC_POSITIONS {crystal}
>  In 0.00 0.00 0.00
>  As 0.25 0.25 0.25
>  In 0.50 0.50 0.00
>  As 0.75 0.75 0.25
>  In 0.00 0.50 0.50
>  As 0.25 0.75 0.75
>  In 0.50 0.00 0.50
>  As 0.75 0.25 0.75
> K_POINTS {automatic}
>  4 4 4 0 0 0
> Strangely enough an scf run does not read the cell parameters card and I get
> an error
>  Waiting for input...
> Warning: card CELL PARAMETERS{CUBIC} ignored
> Warning: card  1.0 0.0 0.0 ignored
> Warning: card  0.0 1.0 0.0 ignored
> Warning: card  0.0 0.0 1.0 ignored
>  %%%%%%%%%%%%%%%%%%%%%%%%%
>      from read_cards : error #         1
>      ibrav=0: must read cell parameters
>  %%%%%%%%%%%%%%%%%%%%%%%%%
>      stopping ...
> --------
> I have run a few calculations in the past with the cubic supercell
> specifying different values of cell parameters and they work fine. Also the
> calculation runs fine with ibrav=1 (and not specifying the cell parameters).
> However, I am not able to pin down the mistake I am making here (Are we not
> allowed to specify a free lattice if it has equal cell parameters?).
> Thanks,
> Dhruv Singh
> School of Mechanical Engineering, Purdue University
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
>