Colleagues, Please advise whether, following a converged LDA+U and mBJ calculation, the Fermi energy to be inserted into case.insp after the sequence:
x lapw1 -c -p -up -band x lapw1 -c -p -dn -band x lapwso -c -orb -p -up x lapw2 -c -so -p -qtl -up -band x lapw2 -c -so -p -qtl -dn -band is to be found from case.scf2up/dn (grep :FER case.scf2up), not from previously saved case.scf (cf. paragraph 8.3.2, line 9, efermi in the Wien2k manual). The -band calculation uses different number of k-points read from case.klist_band (e.g. 61 k-points for the hcp list) than the preceding mBJ calculation with fewer k-points. Is this the reason for the difference between :FER from case.scf and case.scf2up/dn? Would the difference be removed if case.klist and case.klist_band were identical? Thanks, Kamil Klier Lehigh University ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program.
