[Pw_forum] Relativistic semi-core potentials

[Dal Corso Andrea]

On Thu, 2008-05-22 at 12:28 +0100, Fernando Nogueira wrote:
> Dear Stefano,
> 
> 
> Sorry, I forgot to add it. My input is:
> 
> 
> ***************
>  &input
>     title='Cr',
>     zed=24.,
>     rel=2,
>     config='[Ar] 3d5.0 4s1.0 4p0',
>     dft='PBE',
>     iswitch=3
>  /
>  &inputp
>    lloc=0,
>    pseudotype=3,
>    file_pseudopw='Cr.UPF',
>    nlcc=.false.,
>    author='FNOG',
>  /
> 10
> 3P  2  1  2.00  0.00  1.60  2.40  0.5
> 3P  2  1  4.00  0.00  1.60  2.40  1.5
> 4S  2  0  1.00  0.00  1.60  2.40  0.5
> 4P  3  1  0.00  0.00  1.60  2.40  0.5
> 4P  3  1  0.00  0.00  1.60  2.40  1.5
> 3D  3  2  3.00  0.00  1.60  2.40  1.5
> 3D  3  2  0.00 -0.20  1.60  2.40  1.5
> 3D  3  2  2.00  0.00  1.60  2.40  2.5
> 3D  3  2  0.00 -0.20  1.60  2.40  2.5
> 3S  1  0  2.00  0.00  1.60  2.40  0.5

A few comments:
The local potential cannot be US so the two radii must be the same.
Then if you want to have a 4s with nonlocal projectors you need at least
two energies for the 4s or you have to request a NC 4s channel setting
two equal radii. 
With these large NC core radii the RRKJ has problems to find q_i for the
semicore states so I suspect it will be quite difficult to obtain a PP
with these radii but I have not extensive experience with PP with
semicore states. You can also experiment with the option tm=.true.

To make a pseudo with semicore states is still quite difficult with ld1.
In general to make a fully relativistic PP is not so difficult if you
have a good scalar relavistic PP. So I would start from the SR pseudo.

Hope this helps, although the problem of making good PPs with semicore
states with ld1 is not totally solved.

Andrea



> 
> Thanks,
> 
> 
> Fernando
> 
> On May 22, 2008, at 12:22, Stefano de Gironcoli wrote:
> 
> > Fernando Nogueira wrote:
> > > Dear pwscf users,
> > > 
> > > I have been trying to generate a relativistic ultrasoft potential
> > > for 
> > > Cr that includes semi-core electrons. So far I had no sucess: I
> > > keep 
> > > getting the same error message about a discrepancy in the number
> > > of 
> > > pseudo and all-electron electrons. Am I doing something wrong? Is
> > > this 
> > > combination available through the atomic code?
> > without seeing your input is quite difficult to have any ideas
> > about 
> > your problem...
> > stefano
> > 
> > > Thanks!
> > > 
> > > Fernando
> > > 
> > > 
> > > Fernando Nogueira
> > > Centre for Computational Physics and
> > > Physics Department
> > > University of Coimbra
> > > Rua Larga
> > > 3004-516 COIMBRA
> > > PORTUGAL
> > > 
> > > Phone: + 351 239410114
> > > Fax: + 351 239829158
> > > email: fnog at teor.fis.uc.pt <mailto:fnog at teor.fis.uc.pt>
> > > URL: http://nautilus.fis.uc.pt/~fnog/ <http://nautilus.fis.uc.pt/%
> > > 7Efnog/>
> > > 
> > > ------------------------------------------------------------------------
> > > 
> > > _______________________________________________
> > > Pw_forum mailing list
> > > Pw_forum at pwscf.org
> > > http://www.democritos.it/mailman/listinfo/pw_forum
> > > 
> > 
> > _______________________________________________
> > Pw_forum mailing list
> > Pw_forum at pwscf.org
> > http://www.democritos.it/mailman/listinfo/pw_forum
> > 
> 
> Fernando Nogueira
> Centre for Computational Physics and
> Physics Department
> University of Coimbra
> Rua Larga
> 3004-516 COIMBRA
> PORTUGAL
> 
> 
> Phone: + 351 239410114
> Fax: + 351 239829158
> email: fnog at teor.fis.uc.pt
> URL: http://nautilus.fis.uc.pt/~fnog/
> 
> 
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
-- 
Andrea Dal Corso                    Tel. 0039-040-3787428
SISSA, Via Beirut 2/4               Fax. 0039-040-3787528
34014 Trieste (Italy)               e-mail: dalcorso at sissa.it