Dear Dr. Wang, can you expand a little how you arrived to this results? also in the panasis code I found this comment [1]:
here is the correct algorithm for making the eigenvectors real: abbreviated form: one needs to extend the supercell and multiply by phase factor and add together as complex conjugates until they are real (this is already implemented somewhere in isotropy--email stokes more later...? I don't understand where complex conjugate part comes from. [1] http://danse.cacr.caltech.edu/packages/dev_danse_us/parnasis-0.5.tar.gz thanks, Alexandr. >? > From: xirainbow <nkxirainbow at gmail.com> Sent: Wednesday, June 5, 2013 8:50 > PM wrote: > Dear Alexandr Fonari: > ? ? ? ? ? I think the real displacement at q!=0 ?is : displacement = > real(eigenvector*exp(i*k*r-omega*t)). real() means the real part, > omega is frequency, t is time, r is the equilibrium position of atoms. >? >? > On Thu, Jun 6, 2013 at 2:12 AM, A F <af3_pw_forum at yahoo.com> wrote: > > Hello pw_forum, > > > > I have a question with regard to the normal modes at non G-point (q/=0). > > As discussed previously [1], phase is random. > > Thus my question is, how one can obtain displacements in real space from > > those complex displacements? > > > > > > 1. http://qe-forge.org/pipermail/pw_forum/2006-August/079408.html > > > > = = = = = = = = = = > > Alexandr Fonari, > > graduate student, > > Georgia Institute of Technology. > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://pwscf.org/mailman/listinfo/pw_forum >? >? >? > --? > ____________________________________ > Hui Wang > School of physics, Fudan University, Shanghai, China
