[Pw_forum] A question about generating NC-pseudopotential for Fe

[Nicola Marzari]

Dear Xinjie,


maybe you could keep the 3s in the core, and have only the 3p in 
valence. Then, you end up with a 14 e- psp with a s channel tuned
for the 4s; you could test this against the 16 e- ultrasoft 
pseudopotentials from the pwscf database Fe.pbe-sp-van.UPF and 
Fe.pbe-sp-van_mit.UPF (the latter one is tuned for ionic configurations
- i.e. Fe 2+ and 3+).

A very good test for accuracy are the energy splitting of the multiplets 
- I'll send you a table of what we have obtained.

Best,

			nicola

Paolo Giannozzi wrote:

> On Tuesday 10 May 2005 17:21, Xinjie Wang wrote:
> 
> 
>>My purpose is to generate a NC-PP for Fe as good as possible regardless
>>of how hard it is.
>>In this case of pseudotype=2, rel=2, can I use more than one projector
>>(not state) per angular momentum to get good results?
> 
> 
> My understanding is that you can use the semicore (3s) state and another 
> (unbound) state at an energy not too close and not too far from the energy
> of the former. Presently you cannot use both the semicore and the valence 
> (4s) state: they have to be orthogonal to each other, i.e. the valence state
> has to have a node, and this is not presently implemented.
> 
> I have always used only the semicore states in such cases, but the
> resulting PPs are not exceedingly accurate: they work very well in
> the energy region around semicore states, less so around valence
> states. Of course it would be preferrable to use more than one 
> projector (i.e. state, bound or unbound) per angular momentum
> 
> Paolo
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum

-- 
---------------------------------------------------------------------
Prof Nicola Marzari   Department of Materials Science and Engineering
13-5066   MIT   77 Massachusetts Avenue   Cambridge MA 02139-4307 USA
tel 617.4522758  fax 617.2586534  marzari at mit.edu  http://nnn.mit.edu