[gpaw-users] Charged setups

[Toma Susi]

> On 02.06.2020, at 12:21, Michael Walter <Michael.Walter at fmf.uni-freiburg.de> wrote:
> 
> Dear Toma,
> 
> Am Di., 2. Juni 2020 um 12:14 Uhr schrieb Toma Susi <toma.susi at univie.ac.at <mailto:toma.susi at univie.ac.at>>:
> 
>> 
>> Date: Tue, 2 Jun 2020 09:46:05 +0200
>> From: Michael Walter <Michael.Walter at fmf.uni-freiburg.de <mailto:Michael.Walter at fmf.uni-freiburg.de>>
>> To: Jens J?rgen Mortensen <jjmo at dtu.dk <mailto:jjmo at dtu.dk>>
>> Cc: gpaw-users <gpaw-users at listserv.fysik.dtu.dk <mailto:gpaw-users at listserv.fysik.dtu.dk>>,	shubham sharma
>> 	<shubhsharma024 at gmail.com <mailto:shubhsharma024 at gmail.com>>
>> Subject: Re: [gpaw-users] Charged setups
>> Message-ID:
>> 	<CA+Xf-=6QfQd=L9erL4geaVQZuSJ7ECm6Hp0v0n17T9+hXc9nZA at mail.gmail.com <mailto:CA+Xf-=6QfQd=L9erL4geaVQZuSJ7ECm6Hp0v0n17T9+hXc9nZA at mail.gmail.com>>
>> Content-Type: text/plain; charset="utf-8"
>> 
>> Dear Jens Jorgen,
>> 
>> Am Di., 2. Juni 2020 um 09:31 Uhr schrieb Jens J?rgen Mortensen <jjmo at dtu.dk <mailto:jjmo at dtu.dk>
>>> :
>> 
>>> On 6/2/20 9:14 AM, Michael Walter via gpaw-users wrote:
>>>> Dear all,
>>>> 
>>>> we are interested to implement charged setups.
>>>> It is not clear to us, where to put this as there are two generators (in
>>>> generator.py and generator2.py). The tests are not decisive in this
>>>> respect either:
>>>> 
>>>> gpaw/tests$ grep 'from gpaw.atom.generator' */*.py
>>>> corehole/test_h2o.py:from gpaw.atom.generator2 import generate
>>>> gllb/test_ne.py:from gpaw.atom.generator import Generator
>>>> gllb/test_ne_disc.py:from gpaw.atom.generator import Generator
>>>> setups/lithium.py:from gpaw.atom.generator2 import generate
>>>> zlow/test_lb94.py:from gpaw.atom.generator import Generator
>>>> 
>>>> Could someone clarify, please?
>>> 
>>> Add it to generator2.py.
>>> 
>>> I guess you have seen that one can already create setups with charged
>>> core-holes with generator2.
>>> 
>> 
>> thanks for the fast answer.
>> Yes, that is correct. We'll check, whether we can use the core-hole
>> approach for this (we need to change valence occupation).
>> 
>>> 
>>> Out of curiosity, what are these charged setups for?
>>> 
>> 
>> they are needed to get iterative Hirshfeld partioning and charges. See e.g.
>> dx.doi.org/10.1021/ct400694h <http://dx.doi.org/10.1021/ct400694h> | J. Chem. Theory Comput. 2013, 9, 4293?4299
>> 
>> Best,
>> Michael
> 
> That seems interesting, but how would charged setups work exactly for valence occupations in the PAW formalism? Would you essentially fix more of the orbitals, not just the core..?
> 
> I do not see a problem connected to PAW here. You just solve the atomic problem with a charge. Anions need special treatment for stabilization, though.
>  
> 
> Wouldn't CDFT also work?
> 
> Note, that the setups are meant for partitioning and not really for calculation. Whether it makes sense to use a charged setup in a real calculation, I do not know.
> 
> Best,
> Michael

Dear Michael,

Ah, I see now, yes of course for atomic calculation there is no problem. So would you use this just to parametrize dispersion corrections for ionic system, or are the other uses for iterative Hirshfeld partitioning?

Cheers,
Toma

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.fysik.dtu.dk/pipermail/gpaw-users/attachments/20200602/46b91d5c/attachment-0001.html>