Exactly, this could be the explanation. (Sorry, I mixed LDA/PBE energies. However, the situation could also be wrong with PBE, because I'd expect that the mBJ magnetic moment is bigger than the PBE moment, and the PBE functional does "not like" these large moments, thus favors the non-magnetic solution.
Still, do GGA+U for the total energies ... Am 06.04.2011 17:09, schrieb tran at theochem.tuwien.ac.at: > Hello, > > Actually, this is the LDA (and not PBE) exchange-correlation functional > which is used for the energy when MBJLDA is used for the potential > (option 28). > > If you want PBE for the energy you have to replace > > exu=2.d0*(exupls + eclsd) > exd=2.d0*(exdnls + eclsd) > by > exu=2.d0*(exuppb + ecpbe) > exd=2.d0*(exdnpb + ecpbe) > > in the part of the subroutine vxclm2.f which concerns option 28. > Then you recompile the lapw0 package. > Maybe, the antiferromagnetic state will be the lowest again? > > F. Tran > > On Wed, 6 Apr 2011, davgumo wrote: > >> Dr. Blaha, thank you very much for your comments. >> What is a bit confusing about my case is that my very first calculation with >> PBE identifies the ground state in some peculiar antiferromagnetic insulator >> (very small band gap though) this result was expected. Now when I try to >> improve the band gap by performing mBJ calcs, I find that this magnetic order >> which once I identified as the ground state is no longer lowest in energy. In >> fact, I found out that the nonmagnetic case is much lower in energy according >> to mBJ. >> >> Once again, it was expected that the ground state would be found as an >> antiferromagnetic insulator as shown with PBE. >> >> Do think that mBJ somehow has trouble determining the ground state for this >> crystal? >> >> I appreciate so much your help and the time you devote to all the users. >> >> By the way, we will soon have some experimental measurements about the band >> gap of this crystal, I hope the calculated mBJ band gap agrees with the >> experiment this would be very nice. >> >> David Guzman M >> Department of Physics& Astronomy >> California State University-Los Angeles >> (310)528-4841 >> >> On Apr 05, 2011, at 10:42 PM, Peter Blaha<pblaha at theochem.tuwien.ac.at> >> wrote: >> >> Clearly, mBJ is only a POTENTIAL, not an energy-functional. >> In fact, the mBJ option in WIEN2k uses the PBE-energy-functional, but >> of course with the electron density (and spin-density) corresponding >> to mBJ. >> >> The question whether you can trust the mBJ/PBE results more than the PBE/PBE >> results is not so easy to answer and may depend on the specific case. >> >> If your compound belongs somehow to the cuprate-family (with CuO2 planes >> and a Cu2+ ios) and you should get a significant Cu-moment. the chances are >> high >> that PBE does not get the correct groundstate (insulator ? magnetism ?), i.e. >> both the PBE-density and the PBE-energy are "wrong". >> >> The mBJ-density should be more reliable in this case and maybe ??? the >> PBE-energy for such a density is "better". >> >> At least I would "backup" the mBJ results by some LDA+U or hybrid-DFT >> calculations. >> >> Am 06.04.2011 05:38, schrieb davgumo: >>> Dear Wien users and Dr. Blaha >>> >>> I have been performing first-principles calculations on a recently >>> synthesized crystal structure CuZrTiO5. >>> This is a copper oxide the exhibits some potential to be a high Tc >>> superconductor under doping. >>> >>> First I was interested in finding the ground state, which according to PBE >>> calculations turns out to be in some antiferromagnetic order. After this I >>> was interested in >>> using mBJ to calculate an accurate value for the band gap, so I performed >>> mBJ calculation on this antiferromagnetic order where I found the lowest >>> energy. Recently I >>> decided to run mBJ calculations on all the possible magnetic cases for this >>> crystal. To my big surprise when I checked the energies calculated with mBJ >>> the lowest energy is >>> not longer on the magnetic configuration I found the ground state to be >>> according to the PBE calculation. >>> >>> My question is if mBJ calculations are meant only to improve the band gap >>> and will not provide the ground state of the system >>> >>> I appreciate any comments, hints, critics, etc that can help me to figure >>> this out, and get a better understanding of the mBJ potential. >>> >>> Thank you very much. >>> >>> I can provide more details if this is not clear. >>> >>> David Guzman M >>> >>> Department of Physics& Astronomy >>> California State University-Los Angeles >>> >>> >>> >>> >>> _______________________________________________ >>> Wien mailing list >>> Wien at zeus.theochem.tuwien.ac.at >>> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien >> >> -- >> ----------------------------------------- >> Peter Blaha >> Inst. Materials Chemistry, TU Vienna >> Getreidemarkt 9, A-1060 Vienna, Austria >> Tel: +43-1-5880115671 >> Fax: +43-1-5880115698 >> email: pblaha at theochem.tuwien.ac.at >> ----------------------------------------- >> _______________________________________________ >> Wien mailing list >> Wien at zeus.theochem.tuwien.ac.at >> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien >> >> >> >> _______________________________________________ >> Wien mailing list >> Wien at zeus.theochem.tuwien.ac.at >> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien -- 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/ --------------------------------------------------------------------------