[Pw_forum] Different optical mode frequency at the same q-vector

[dfsgdv dfsdf]

Hi,

I am trying to generate phonon dispersion of a complicated structured
material.
I succeeded all the steps until I got the final dispersion relations, but
somehow it doesn't look quite correct.
I will explain my problmes with a simpe example.

Assume I am doing the phonon dispersion along Gamma to Z(straight up
(0,0,1) from the origin) and Gamma to X(straight right (1,0,0) from the
origin) .
The dispersion relation along Gamma to Z produced reasonable 6 different
frequency modes (Mine actually had 15 modes), but Gamma to X was the weired
one. The problem was that the Gamma to X even had different frequencies at
Gamma.

To figure out the problem, I first calculated the frequencies at Gamma
(0,0,0), and it was (0,0,0,4,4,5) which was the same as the frequencies
that I've got in the Gamma to Z. However, when I added one more q-vectors
(0.1,0,0) to the frequency generation input file (test.matdyn.in), the
frequencies at Gamma changed to (0,0,0,3,4,5) and the next one was (0.1,
0.1, 0.2, 2.9, 4.1, 5.1). They seemed to have continuousness between these
two, but adding another q-vector caused even the frequencies at Gamma being
changed, and it didn't look so reasonable.

So basically my problem is that I have generated two phonon dispersion
relation: Gamma to Z and Gamma to X, but the optical mode frequencies at
Gamma in the two plots are different.

I wonder if there's any parameter that I have to adjust to force them to be
consistent or any idea that I can try.
I hope this explains my problem well, so I can get any advice.

Thank you.

Kris
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20121004/318c46b6/attachment.html>