Hello, this happens when the system is a bad metal, with a pocket of valence electrons in a small Fermi surface. I.e., at some point in the Brillouin zone, the n-th band is lower in energy than the (n-1)-th band somewhere else, but these two bands do not touch. Just use a smearing for occupation and you'll solve the issue. Do not forget to recheck k-points convergence in the new case.
regards -- Lorenzo Paulatto - Paris On Feb 21 2021, at 7:05 pm, Mahmoud Payami Shabestari <[email protected]> wrote: > Dear QE-users/Developers, > > In DFT+U calculation of an AFM pure system, I used an input file in which I > had not used "occupation" card, but only a few bands more than half of the > number of electrons. Then, the SCF run was successful, and in the output I > get both HOMO and LUMO values correctly aligned, i.e., E_LUMO > E_HOMO. > However, when I add a metallic impurity (without changing the other > parameters), the SCF run successfuly ends and in the output now I get E_HOMO > > E_LUMO (!!). > I do not know how to interpret this result. Is there any miscalculation? > Any comments is highly appreciated. > > Best regards, > Mahmoud Payami > AEOI, Tehran, Iran > > Email: [email protected] > Phone: +98 (0)21 82066504 > > > _______________________________________________ > Quantum ESPRESSO is supported by MaX (www.max-centre.eu) > users mailing list [email protected] > https://lists.quantum-espresso.org/mailman/listinfo/users
_______________________________________________ Quantum ESPRESSO is supported by MaX (www.max-centre.eu) users mailing list [email protected] https://lists.quantum-espresso.org/mailman/listinfo/users
