[Pw_forum] Questions on ibrav in pw.x

Yantao Wu ywu at g.hmc.edu
Wed May 29 03:21:53 CEST 2013


Thank you! It makes much more sense now.


On Tue, May 28, 2013 at 1:47 AM, Axel Kohlmeyer <akohlmey at gmail.com> wrote:

> On Tue, May 28, 2013 at 9:57 AM, Yantao Wu <ywu at g.hmc.edu> wrote:
> > Dear QE,
> >
> > I'm wondering what ibrav means in pw.x. My confusion is that if you can
> > specify atomic positions freely, isn't the structure of the solid already
> > implied by the positions? So why do we need the extra variable of ibrav.
> Or
> > does ibrav carries more meaning than just the initial lattice structure
> of
> > the solid?
> >
> > Another related question: When I ran vc-relax, I specified ibrav = 2
> (fcc)
> > and the atomic positions as follows:
> >
> > ATOMIC_POSITIONS {alat}
> > Al   0.0      0.5     0.5
> > Al   0.5      0.0     0.0
> > Al   0.25     0.25    0.25
> > Al   0.0      0.5     0.0
> > Ti   0.75     0.75    0.75
> > Ti   0.5      0.0     0.5
> > Cr   0.0      0.0     0.0
> > Cr   0.75     0.75    0.25
> > Fe   0.5      0.5     0.0
> > Fe   0.75     0.25    0.25
> > Co   0.0      0.0     0.5
> > Co   0.25     0.75    0.25
> > Co   0.75     0.25    0.75
> > Ni   0.25     0.75    0.75
> > Ni   0.25     0.25    0.75
> > Ni   0.5      0.5     0.5
> >
> > QE erred and complained that atom #1 and atom #6 overlap, which they
> > apparently don't from the first sight. When I set ibrav = 1 (sc) and kept
> > the same atomic positions, the complaint disappeared. So does anyone know
> > why did the error occur?
>
> because of the difference between the primitive cell and the
> conventional cell. sc and fcc have the same conventional cell, but the
> primitive fcc cell contains only a quarter of the atoms. obviously you
> entered atoms in the conventional cell.
> to be more explicit. the primitive cell of, say, an fcc metal contains
> just one atom (one corner) and the basis vectors are not orthogonal,
> whereas in the conventional cell, the basis vectors are orthogonal and
> you have 4 atoms (one corner and 3 faces).
>
> Q-E expects that you enter coordinates in the primitive cell only.
> yet, since the sc primitive cell is identical to its conventional
> cell, you can use an sc lattice to enter an fcc lattice geometry in
> the conventional cell.
>
> look it up in a crystallography text book (and on the web) and you'll see.
>
> axel.
>
> >
> > I appreciate your replies with all my gratitude.
> >
> > Yantao "ignorant student" Wu
> >
> > _______________________________________________
> > Pw_forum mailing list
> > Pw_forum at pwscf.org
> > http://pwscf.org/mailman/listinfo/pw_forum
>
>
>
> --
> Dr. Axel Kohlmeyer  akohlmey at gmail.com  http://goo.gl/1wk0
> International Centre for Theoretical Physics, Trieste. Italy.
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://pwscf.org/mailman/listinfo/pw_forum
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20130528/18e447c8/attachment.html>


More information about the users mailing list