> 1- Does the dimension of Supercell for phonon calculation determine > by nq1, nq2,and nq3 in the phonon input file?
The parameters nq1, nq2,and nq3 is determine the Monkhorst-Pack grid for the DFPT calculation (phonon calculation) that done in q-space (you calculate the dynamical matrices explicitly and not the force constants). In DFPT there is no supercell but the convergence with respect to q-points is similar to the convergence with respect to the size of the supercell in supercell approaches. > 2- we know that the lowest three dispersion in the phonon spectra > show the three acoustic branches, but how can I find which one is > for TA, LA, and ZA? Phonons do not have just transverse or longitudinal polarization. There are phonons with polarization that is neither transverse nor longitudinal (although transverse and longitudinal polarizations are off-course possible). The most general polarization is elliptical. In general, the polarization vector is complex and you can know it from the calculation (after diagonalization of the dynamical matrices). You have this information in the dynamical matrices files which is output of the phonon calculation and also you have it in the output of matdyn.x in the file matdyn.modes. We recently publish a work in which we analyze the polarization of a complex acoustic phonon in the corner of a hexagonal Brillouin zone: http://journals.aps.org/prb/abstract/10.1103/PhysRevB.94.174305 You can maybe find it interesting. Uri Argaman Ben-Gurion University Israel
_______________________________________________ Pw_forum mailing list [email protected] http://pwscf.org/mailman/listinfo/pw_forum
