Moin,
I discovered a problem in the calculation of the ILDOS (plot_num=10) in pp.x,
version 4.2.1, when ultrasoft pseudopotentials are used.
In principle, setting emin to -\infty and emax to E_{Fermi}, the ILDOS should
be identical to the total charge density (except for small differences due to
the smeared occupations around E_Fermi in the density).
This is true if one uses norm-conserving pseudopotentials.In the case of ultrasoft pseudopotentials, the resulting ILDOS values are of an unreasonable range (-41 to +240), while the density is in good agreement with the ncpp-case (only differences coming from the differnces in valence/core partition). Is this a known problem? -- Mit freundlichen Gr??en Henning Glawe Max-Planck-Institut f?r Mikrostrukturphysik Weinberg 2, 06120 Halle (Saale), Germany http://www.mpi-halle.de/~theory Phone: +49-345-5582-613 Fax: +49-345-5511223 Email: glawe at mpi-halle.de
