Dear Paolo: Thank you very much for helping. So there is no way I can do the lattice constance optimization right now with metaGGA functionals in QE, unless somebody can re-design the pp files. Is my understanding correct?
Best regards, Hong Tang __________________________________________ There isn't much to explain: nobody has ever written down and implemented the formula for stress in the case of meta-GGA. In QE meta-GGA is implemented (sort of) only for norm-conserving pseudopotential, so there is no need (bot that I know) to store the kinetic energy density in the PP file Paolo On Wed, Jun 28, 2017 at 4:34 AM, Hong Tang <tanghong992 at gmail.com <http://pwscf.org/mailman/listinfo/pw_forum>> wrote: >* Dear Paolo: *>>>>* Thank you so much for helping. *>>* I tried grep -n “stress” Si.rx.out *>>* What I got is below: *>>* [hongtang at login4 <http://pwscf.org/mailman/listinfo/pw_forum> results]$ grep -n "stress" Si.rx.out *>>* 1797: entering subroutine stress ... *>>* 1799: Message from routine stress: *>>* 1800: Meta-GGA and stress not implemented *>>* 3128: entering subroutine stress ... *>>* 3130: Message from routine stress: *>>* 3131: Meta-GGA and stress not implemented *>>* 4442: entering subroutine stress ... *>>* 4444: Message from routine stress: *>>* 4445: Meta-GGA and stress not implemented *>>* 5764: entering subroutine stress ... *>>* 5766: Message from routine stress: *>>* 5767: Meta-GGA and stress not implemented *>>* 7555: entering subroutine stress ... *>>* 7557: Message from routine stress: *>>* 7558: Meta-GGA and stress not implemented *>>>>* It shows MetaGGA and stress not implemented. Maybe this is the problem *>* there. *>>* I want to do structure optimization with metagga functional (TPSS here). In *>* VASP, I know the pseudo potentials (pps) of the elements should have *>* “kinetic energy density”. In quantum espresso, I don’t know what should be *>* in the pps. The pps I chose for Si is for tpss, as you can see from the *>* input file. *>>>>* Please explain to me a bit more clear how to fix the problem. I appreciate *>* your helps very much. *>>>>* Hong Tang *>>* Temple University *>>* Philadelphia, PA 19122 US* On Wed, Jun 28, 2017 at 10:37 PM, Hong Tang <tanghong...@gmail.com> wrote: > Dear Paolo: > > Thank you very much for helping. So there is no way I can do the lattice > constance optimization right now with metaGGA functionals in QE, unless > somebody can re-design the pp files. Is my understanding correct? > > Best regards, > > Hong Tang > > > > __________________________________________ > > There isn't much to explain: nobody has ever written down and > implemented the formula for stress in the case of meta-GGA. In QE > meta-GGA is implemented (sort of) only for norm-conserving > pseudopotential, so there is no need (bot that I know) to store the > kinetic energy density in the PP file > > > > Paolo > > On Wed, Jun 28, 2017 at 4:34 AM, Hong Tang <tanghong992 at gmail.com > <http://pwscf.org/mailman/listinfo/pw_forum>> wrote: > >* Dear Paolo: > *>>>>* Thank you so much for helping. > *>>* I tried grep -n “stress” Si.rx.out > *>>* What I got is below: > *>>* [hongtang at login4 <http://pwscf.org/mailman/listinfo/pw_forum> > results]$ grep -n "stress" Si.rx.out > *>>* 1797: entering subroutine stress ... > *>>* 1799: Message from routine stress: > *>>* 1800: Meta-GGA and stress not implemented > *>>* 3128: entering subroutine stress ... > *>>* 3130: Message from routine stress: > *>>* 3131: Meta-GGA and stress not implemented > *>>* 4442: entering subroutine stress ... > *>>* 4444: Message from routine stress: > *>>* 4445: Meta-GGA and stress not implemented > *>>* 5764: entering subroutine stress ... > *>>* 5766: Message from routine stress: > *>>* 5767: Meta-GGA and stress not implemented > *>>* 7555: entering subroutine stress ... > *>>* 7557: Message from routine stress: > *>>* 7558: Meta-GGA and stress not implemented > *>>>>* It shows MetaGGA and stress not implemented. Maybe this is the problem > *>* there. > *>>* I want to do structure optimization with metagga functional (TPSS here). > In > *>* VASP, I know the pseudo potentials (pps) of the elements should have > *>* “kinetic energy density”. In quantum espresso, I don’t know what should > be > *>* in the pps. The pps I chose for Si is for tpss, as you can see from the > *>* input file. > *>>>>* Please explain to me a bit more clear how to fix the problem. I > appreciate > *>* your helps very much. > *>>>>* Hong Tang > *>>* Temple University > *>>* Philadelphia, PA 19122 US* > >
_______________________________________________ Pw_forum mailing list Pw_forum@pwscf.org http://pwscf.org/mailman/listinfo/pw_forum