There is a doubly-degenerate state at 34.84024 eV that is not present in the reference results. It might be a consequence of the change of default starting wavefunctions: a small randomization is added to the superposition of atomic wavefunctions. This breaks the symmetry and prevents iterative diagonalization from "missing" some states (something that in theory shouldn't happen, but in practice sometimes does, especially where high symmetry is present)
P. On Fri, 2013-04-12 at 20:42 +0800, Miao Gao wrote: > Dear all: > bands.x can be used to analyse the band symmetry, as > demonstrated in example04 in PP directory. > But when I run this example, I get different result > with the reference data for Gamma point. > --------------------------------------------------------------- > reference in example04: > xk=( 0.00000, 0.00000, 0.00000 ) > ... > Band symmetry, O_h (m-3m) double point group: > > e( 1 - 2) = 7.27272 eV 2 --> G_6+ > e( 3 - 6) = 13.29785 eV 4 --> G_8+ > e( 7 - 8) = 14.29146 eV 2 --> G_7+ > e( 9 - 12) = 16.11924 eV 4 --> G_8+ > e( 13 - 14) = 38.36105 eV 2 --> G_6- > e( 15 - 18) = 39.65390 eV 4 --> G_8- > ----------------------------------------------------------------- > my results (from QE-5.0.1 ! and QE-5.0.2) > Band symmetry, O_h (m-3m) double point group: > > e( 1 - 2) = 7.27271 eV 2 --> G_6+ > e( 3 - 6) = 13.29786 eV 4 --> G_8+ > e( 7 - 8) = 14.29147 eV 2 --> G_7+ > e( 9 - 12) = 16.11925 eV 4 --> G_8+ > e( 13 - 14) = 34.84024 eV 2 --> G_7- > e( 15 - 16) = 38.36105 eV 2 --> G_6- > e( 17 - 18) = 39.65390 eV &n! bsp; 2 --> ? > ----------------------------------------------------------------- > What is the reason for the differences ranging form 13-th band to > 18-th band? > Does it relate to different compiled libraries? > Any help will be appreciated. > > 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
