[Pw_forum] Question on Fractional occupation used to generate Mg.pw91-np-van.UPF

[Derek Stewart]

Hi everyone,

I would like to use the PW91 Mg pseudopotential for some calculations, 
but I noticed something odd in the occupation listed at the beginning of 
the pseudopotential.

nl pn  l   occ               Rcut            Rcut US             E pseu
2P  2  1  6.00     10.00000000000      2.00000000000     -3.69842610300
3S  3  0  1.00     10.00000000000      2.30000000000     -0.51455069900
3P  3  1  0.75     10.00000000000      2.00000000000     -0.23733335900
3D  3  2  0.00     10.00000000000      2.00000000000     -0.02607571400

The ground state configuration of Mg is 1s2 2p6 3s2 , however this 
pseudopotential uses 1s2 2p6 3s1 3p0.75 with 0.25 electrons missing.

Later on in the file, it does specify the total valence as 8 electrons.

Does someone know why this approach was taken in generating the 
pseudopotential?  Is this done to provide a better pseudopotential for 
ionic materials where Mg will have a positive charge?

Thanks,

Derek


-- 
################################
Derek Stewart, Ph. D.
Senior Research Associate
http://sites.google.com/site/dft4nano/
250 Duffield Hall
Cornell Nanoscale Facility (CNF)
Ithaca, NY 14853
stewart (at) cnf.cornell.edu
(607) 255-2856