[Pw_forum] Empty bands with hybrid functional
On Tue, 2013-12-03 at 12:22 -0800, Alexandr Fonari wrote: > What is the right way to get empty bands when employing hybrid functional, > either: > > a. First run SCF with no empty bands and then again SCF with: > restart_mode='restart' > nbnd = occupied_plus_empty_bands unlikely to work, for a stupid reason: when restarting, the code expects nbnd bands, and this is larger than the number of written bands > or b. Run SCF with no empty bands and then NSCF with: > nbnd = occupied_plus_empty_bands I am not sure it works, and even if it does, not sure it takes less time than c.) below. The diagonalization at fixed potential should include a fixed exchange potential as well. Not sure this is done. > or c. I've missed something and there is other way. c. run a SCF calculation with some more empty bands. Wasteful but safe. P. > > thanks a lot, > Alexandr Fonari, > grad student, > Georgia Tech. > > ___ > Pw_forum mailing list > Pw_forum at pwscf.org > http://pwscf.org/mailman/listinfo/pw_forum -- Paolo Giannozzi, Dept. Chemistry&Physics&Environment, Univ. Udine, via delle Scienze 208, 33100 Udine, Italy Phone +39-0432-558216, fax +39-0432-558222
[Pw_forum] Empty bands with hybrid functional
Wasteful but safe sounds good. Thanks Paolo. > On Tuesday, December 3, 2013 4:06 PM, Paolo Giannozzi uniud.it> wrote: > On Tue, 2013-12-03 at 12:22 -0800, Alexandr Fonari wrote: > > > What is the right way to get empty bands when employing hybrid functional, > > either: > > > > a. First run SCF with no empty bands and then again SCF with: > >? ? restart_mode='restart' > >? ? nbnd = occupied_plus_empty_bands > > unlikely to work, for a stupid reason: when restarting, the code expects > nbnd bands, and this is larger than the number of written bands > > > or b. Run SCF with no empty bands and then NSCF with: > >? ? nbnd = occupied_plus_empty_bands > > I am not sure it works, and even if it does, not sure it takes less > time than c.) below. The diagonalization at fixed potential should > include a fixed exchange potential as well. Not sure this is done. > > > or c. I've missed something and there is other way. > > c. run a SCF calculation with some more empty bands. Wasteful but safe. > > P. > > > > thanks a lot, > > Alexandr Fonari, > > grad student, > > Georgia Tech.> > > > ___ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://pwscf.org/mailman/listinfo/pw_forum > > -- > Paolo Giannozzi, Dept. Chemistry&Physics&Environment, > Univ. Udine, via delle Scienze 208, 33100 Udine, Italy > Phone +39-0432-558216, fax +39-0432-558222 > >
[Pw_forum] Empty bands with hybrid functional
Dear QE community, What is the right way to get empty bands when employing hybrid functional, either: a. First run SCF with no empty bands and then again SCF with: ? ? restart_mode='restart' ? ? nbnd = occupied_plus_empty_bands or b. Run SCF with no empty bands?and then NSCF with: ? ? nbnd = occupied_plus_empty_bands or c. I've missed something and there is other way. thanks a lot, Alexandr Fonari, grad student, Georgia Tech.