[Pw_forum] Questions about band calculation

Richard Dong rcdong at gmail.com
Thu Oct 13 06:55:13 CEST 2011


Dear QE users,

I am a PhD student who is new to QE. I got some questions when I did some
simple test calculations on material OsO2.

First, my input file for "scf" calculation is:

&CONTROL
   calculation='scf'
   restart_mode='from_scratch',
   pseudo_dir='/espresso-4.3.2/pseudo/',
   outdir='./',
   prefix='OsO2'
/
&SYSTEM
   ibrav=6, celldm(1)=8.52551, celldm(3)=0.70731, nat=6, ntyp=2,
   nbnd=26, ecutwfc=30.0,ecutrho=300
   occupations='smearing',smearing='marzari-vanderbilt', degauss=0.05
/
&ELECTRONS
   diagonalization='david'
   mixing_beta=0.5,
/
ATOMIC_SPECIES
Os   190.23   Os.pw91-n-van.UPF
O    15.9994  O.pw91-van_ak.UPF
ATOMIC_POSITIONS crystal
Os   0.0   0.0   0.0
Os   0.5   0.5   0.5
O    0.3   0.3   0.0
O   -0.3  -0.3   0.0
O    0.8   0.2   0.5
O   -0.8  -0.2   -0.5
K_POINTS automatic
6 6 8 0 0 0

Then I did the "bands" calculation which just replaced the
calculation="bands" and set the K_POINTS path. Finally I runned bands.x and
plotbands.x to get the band structures. I actually got two strange flat
bands in Os 5d bands (plot is attached below). Does anyone know the reason
that can cause this kind of flat bands?

Another question is about the spin-orbital coupling. If I further add
spin-orbital coupling to my calculation, I am wondering if  I have to use
the full-relativistic pseudopoentials? If this is the case, is there any
tested version I can get or I have to generate by myself, since there is no
full-relativistic PPs on the QE website.

Thank you very much in advance.

Best regards,
Ruanchen Dong
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.democritos.it/pipermail/pw_forum/attachments/20111012/3e89c016/attachment-0001.htm 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: OsO2.pdf
Type: application/pdf
Size: 19109 bytes
Desc: not available
Url : http://www.democritos.it/pipermail/pw_forum/attachments/20111012/3e89c016/attachment-0001.pdf 


More information about the Pw_forum mailing list