[Pw_forum] anomaly in phonon frequencies as obtained using DFPT (ph.x) in a spin-orbit calculation

[Koushik Pal]

Dear sir,
  I calculated the phonon frequencies of a system using both the frozen
phonon method as well as DFPT (ph.x) in *fully relativistic regime
(spin-orbit interaction)* using QE verson 4.3. According to group theory
the 12 optical modes of the system are 2A1g + 2Eg + 2A2u +2Eu. I got
accurate results for the frequncies in the frozen phonon method (means, the
frequencies obeyed the symmetry rule as predicted by the group theory. Also
I visualized the eigenmodes which also agree well), but in the DFPT (linear
response theory), the obtained frequencies and eigenmodes do not obey the
above symmetry principles as predicted by group theory (though the
frequencies are near the frequencies as obtained in frozen-phonon method) .
It seems that the degeneracies in the frequencies are broken in DFPT. What
is the reason behind that? Please help. Please give references with
mathematical details (if available).

Thanks in advance.

=============
Koushik
Grad. student
Chemistry and Physics of Materials Unit
JNCASR
Bangalore , India
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20120717/fcdc6de9/attachment.html>