[QE-developers] PAW for Meta-GGA Functionals and UPF Format Changes

Paolo Giannozzi p.giannozzi at gmail.com
Tue Aug 31 09:23:51 CEST 2021


On Tue, Aug 31, 2021 at 7:39 AM Kyle Bystrom <kylebystrom at gmail.com> wrote:

> Hello QE developers,
>
> I noticed that Meta-GGA functionals in QE are currently only supported for
> Norm-Conserving pseudopotentials, not USPP or PAW. I had seen previous
> discussion saying that this is because Meta-GGAs are too numerically
> unstable for PAW, but I was wondering if this might be changing now that
> more stable functionals like r2SCAN have been developed. Are there any
> plans to add capability to perform Meta-GGA calculations with PAW?
>

it is not that simple: with PAW and USPP, the kinetic energy density has an
"augmentation" term. It can be computed but it seems to me that its
implementation is more akin to a full-fledged project than to something
that can be done in the spare time.


> Relatedly, I was wondering if the major overhaul
> <http://www.quantum-espresso.org/pseudopotentials/unified-pseudopotential-format>
> of the UPF format mentioned on the website might include the ability to
> store core kinetic energy density for use in Meta-GGAs.
>

For the time being, the "major" overhaul is limited to the definition of a
xml schema (see
https://www.quantum-espresso.org/resources/road-map/qe_pp-0.99.xsd.xml) and
does not touch its contents. The format might for sure be  extended to
include atomic kinetic energy density, but somebody has to generate the
needed quantities.

Paolo


> Thank you,
> Kyle
> --
> Kyle Bystrom
> PhD Student, Applied Physics
> Harvard School of Engineering and Applied Sciences
> _______________________________________________
> developers mailing list
> developers at lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/developers
>


-- 
Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
Univ. Udine, via delle Scienze 206, 33100 Udine, Italy
Phone +39-0432-558216, fax +39-0432-558222
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/developers/attachments/20210831/1137a350/attachment.html>


More information about the developers mailing list