The ABINIT Users Mailing List ( CLOSED )

["张�s" <zhangting1980323@gmail.com>]

I have the same question

When calculating phonon, in abinis or abinip's main output file, there are asr imposed d-matrix, which follows the asr not imposed DDB data. Using these asr inposed d-matrix, one can get quite perfect phonon frequencies(Note that q should noe be at gamma point). My result is, the phonon freq calculated by anaddb with a q-mesh and the phonon freq calculated by the one-q-point d-matrix mentioned before, are quite similar.They agree within 1 cm-1.

But unfortunately, the phonon eigvector calculated with two method are quite different from each other. I don't know why, maybe it's natural? A small difference in D-matrix will cause a great difference in eigvec?
 

2006/12/7, yunhee.chang@kriss.re.kr <yunhee.chang@kriss.re.kr>:

Dear Abinit users,

When I calculate the dynamical matrix for getting phonon frequencies,
could I get the phonon eigenvector in linear response calculation?

In anaddb process, there is an "eivec" option,
but at that time I need several qpoint's DDB files.

Is it possible I can get the phonon eigenvector in only one qpoint's dynamical matrix calculations?

Please, give me any comments~!

Best regards,
Yunhee Chang