[Wien] Fermi surface not 4 fold symmetric for a cubic material with space group Pm3m

[Anup Shakya]

Dear Prof Blaha,

I have performed non magnetic spin polarized calculations using
runsp_c_lapw -orb. I have not shifted the k-mesh for FS calculations. I
couldn't use the command x kgen -so for FS calculations as I got the
following error.
forrtl: severe (24): end of file during read, unit 20, file/case.ksym
error command: /home/kbmaiti/wien2k/kgen kgen.def failed.
So I used x kgen 30000 kpoints no shift
x lapw1 - up
x lapwso
cp case.outputso case.output1
x lapw2 - so - fermi
I don't think there is any problem with the FS calculations as it is
matching with the bandstructure calculations. The only problem is the
lowering of the symmetry with SOC from 48 to 16 number of symmetry
operations though the structure is still cubic for magnetization direction
along 0 0 1 axis. Due to this lowering of symmetry the band structure plot
along G-X and X-G is different. But experimentally if one measures the
bandstructure or the FS of this material it should be the same because
experimentally the structure is cubic. So my question is whether non
magnetic spin polarized calculations with U and SOC is at all reliable to
match or analyze the experimental results. The only reason for doing non
magnetic spin polarized calculations was to take into account the strong
correlation of R 4f states. I am confused. Please clarify.

Sincerely,
Anup Pradhan Sakhya





On Mon, Nov 5, 2018, 1:23 PM Anup Shakya <npshakya31 at gmail.com wrote:

> Dear All,
> I have done non-magnetic GGA+SOC+U calculations with U and J for a
> material containing rare earth element. Without SOC and U the Fermi surface
> is 4 fold symmetric but when I performed GGA+SOC+U calculations the Fermi
> surface is not 4 fold symmetric. Before applying SOC the position of the
> atoms were as follows:
> Atom X Mult =1  0.0      0.0       0.0
> Atom Y Mult 6  0.207   0.5      0.5
> After SOC the a,b,c values and the symmetry i.e, Pm3m is the same as
> before but now the Y positions has split into two non-equivalent atoms as
> shown below.
> Atom Y 2 Mult 4 0.207   0.5  0.5
> Atom Y 3 Mult 2 0.5      0.5   0.207.
> The position and the multiplicity of X atom is unchanged. I think probably
> because of SOC the equivalent atoms have changed into non-equivalent atoms
> and led to the anisotropy in the band structure along the G-X and X-G
> direction and also 4 fold symmetry is not there in the Fermi surface
> calculations. But if I want to compare the obtained band structure or the
> Fermi surface with the ARPES results then it doesn't make sense because the
> Fermi surface should be 4 fold symmetric. I do not know what to do. If you
> need any more information then please let me know. Please provide some
> suggestions as to preserve the 4 fold symmetry along with GGA+SOC+U
> calculations. Any suggestion would be very much valuable for me.
>
> Sincerely,
> Anup Pradhan Sakhya,
> TIFR, India
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20181105/478168b6/attachment.html>