> I guess I misinterpreted the info at the UG ("LAPW1" section, "Input" > subsection, almost at the end of that subsection where it is explained > the "Line 4"), you are totally correct, it is only needed when I'm > interested at unoccupied states. > However, including a proper LO ("l"-line twice) at the La-4f (in order > to avoid linearization errors or try to reduce it) should not affect > in any drastic way the total energy, right? I would like to know how > may I choose the two energies for this LO?, should I pick the same > value for both?
No, you must pick two energies which differ by at least 0.5-1.0 Ry > Basically what I tried to perform with this scheme was the description > of the La-atom in the system without f-states, as discussed before in But then do not wonder that calculations with/without 4f states yield different results. "Removing" the 4f states is a "brutal" approximation. If you do not trust the GGA description of La 4f states, the cleaner solution is to use LDA+U for these states, which will shift them a bit higher in energy. (use -orb, not -orbc) > this mailing list: > http://zeus.theochem.tuwien.ac.at/pipermail/wien/2009-March/012264.html > ECHO{ > Peter Blaha pblaha at theochem.tuwien.ac.at > Tue Mar 3 14:54:33 CET 2009 > > * Previous message: [Wien] La atom > * Next message: [Wien] La atom > * Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] > > Just put the 4f energy parameter to a "large" value (eg. +2.0). This "removes" > the (true) 4f states from the basis. > > Laurence Marks schrieb: >> Does anyone know the trick to calculate a La atom in PBE without >> f-states? The only thing I can think of is to try and setup an orbital >> potential for the f's and use -orbc to suppress them. >> -- P.Blaha -------------------------------------------------------------------------- Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna Phone: +43-1-58801-15671 FAX: +43-1-58801-15698 Email: blaha at theochem.tuwien.ac.at WWW: http://info.tuwien.ac.at/theochem/ --------------------------------------------------------------------------