Dear Prof Blaha,
thank you very much for your advice. Now the structure looks as you
said. So after creating the struct file sgroup found the right symmetry
I have now another question. The bandstructure of my slab looks now as
known from literature.
Now I would like to run a second SCF
My thunderbird released the message earlier then expected... :) I tried
to set up the .inso file like:
WFFIL
10 1 0 llmax,ipr,kpot
-10. 1.5 emin,emax (output energy window)
0. 0. 0. direction of magnetization (lattice vectors)
2
Dear Mr Marks,
thank you for the fast replay. I have general questions regarding your
points.
ad0) I will try several configurations. The main point is that between
the slabs the charge density must be zero in order to avoid interaction
between the surfaces.
ad2) As fas as I understand the
The procedure is ok. But you should accept the changes of sgroup !
It will move the origin so that you have inversion symmetry.
Am 08.09.2010 12:15, schrieb Bartosz Slomski:
Dear Mr Blaha,
indeed my structure does not represent a surface along [111].
I made a mistake using structgen/octave.
After executing x sgroup with my initial Pbslab_34_34.struct I get the
warning: !!! Bravais lattice has changed.
sgroup found: 12 (C 2/m) [unique axis c] cell choice 2.
When I accept the structure of x sgroup, then the resulting structure
(Pb495_after_symmetry.struct) looks completely
So many things
0. 6 layers of Pb and 12 of vacuum is strange. I normally use 2/3
crystal, 1/3 vacuum.
1. Don't ignore the changes Wien2k suggests -- use them
2. You will need to do a minimization of the positions before you can
look at the band structure.
3. Use a centro-symmetric structure,
Oh, and use 120 (not 60) for the angle, avoids more problems apart
from being correct (60 is wrong!)
On Tue, Sep 7, 2010 at 9:20 AM, Laurence Marks L-marks at northwestern.edu
wrote:
So many things
0. 6 layers of Pb and 12 of vacuum is strange. I normally use 2/3
crystal, 1/3 vacuum.
1.
Have you looked how the generated structure looks like ? (xcrysden)
It is not correct.
I'm not the expert with structgen and thus cannot tell you, where the error is.
In any case, an fcc-111 surface is very easy (although one cannot use
directly the supercell program).
Have a look into
8 matches
Mail list logo
