Without info. about smearing method, it is hard to tell. Some methods may give you incorrect forces, resulting in incorrect phonon. degauss and k-point mesh should be well tested TOGETHER.
Check out the following: http://caribeictp.uis.edu.co/lectures/R.Gebauer/Gebauer_Course2_KpointsMetals.pdf ---------------------------------------------------- Duy Le Postdoctoral Associate Department of Physics University of Central Florida. Website: http://www.physics.ucf.edu/~dle On Sat, Jun 22, 2013 at 9:05 AM, Miao Gao <miaogao at ruc.edu.cn> wrote: > Dear all? > > I am calculating phonon with QE-5.0.1. > A 20x20x10 k-mesh is used in scf calculation. Phonon wave vectors are > sampled > on 10x10x5 mesh. But when I set degauss=0.03Ry in scf run, a imaginary > frequency > was found for the first branch with omega= -105.6 cm^-1 at wave vector (0.2, > 0.346410162, 0.0). > But if degauss=0.05Ry in scf run, this imaginary frequency disappear and > become 132.1 cm^-1. > Meanwhile, the frequencies of all other branches change slightly for these > two degauss values. > So my question is which results should I trust, degauss=0.03 Ry or 0.05Ry? > > Btw: both lattice parameters and the inner coordinates of atoms are well > relaxed for > different degauss values. > > I am greatly indebted to you for your help. Thanks again! > > Best > Miao Gao > > --------------------------------------- > Ph.D. student of Department of Physics, > Renmin University of China. > --------------------------------------- > > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://pwscf.org/mailman/listinfo/pw_forum
