The ABINIT Users Mailing List ( CLOSED )

[Marek Veithen <Marek.Veithen@ulg.ac.be>]

Kristopher Andersen wrote:

In trying to generate a pseudopotential using the PBE GGA functional
for Lithium (not available on the web site) with the Fritz-Harber
Institute code, I'm not sure how to choose the value of fchrg, the
amplititude of the core charge.

The help file Infos/Psp_infos/psp6.info says the following, but offers
no advice on how to choose fchrg. (It goes into some depth on how to
choose rchrg.)

 fchrg: amplitude of core charge (if =0 non core-correction performed,
        if >0 core-correction is applied). The specific value of fchrg
        has no physical meaning in fhi98pp scheme of core correction.

In the .cpi output file, there are values of the core charge and it's
derivatives on the radial mesh. Is it appropriate to set fchrg equal
to the value of the core charge at the core cutoff radius (rchrg) one
would choose for the model core?

I was hoping that looking at the fhi pseudopotential for Na, which is
available on the web site, might shed some light on this, but
apparently that pseudopotential doesn't include nonlinear XC core
corrections. I was under the impression that for sodium these would be
needed?

Any help or advice is appreciated.
 

Dear Kristopher,

when generating a pseudopotential for Li using the fhi98PP code
I adopted a different solution:
I used both the 1s and 2s electrons as valence electrons.
Of course, the so generated pseudopotential is quite hard and
requires a large ecut (about 45 hartree) but it works well in all aplications
I did up to now.

I you want, I can send you my pseudopotential file as well as
a few informations on the thests I performed to check its
transferrability.

Best regards,

Marek
 

-- 
Marek VEITHEN
Universite de Liege
Institut de Physique, Bat. B5
Allee du 6 aout, 17
B- 4000 Sart Tilman
BELGIUM

Phone : ++(32) (0)4-366.36.12
Fax   : ++(32) (0)4-366.29.90
E-mail: Marek.Veithen@ulg.ac.be