[Pw_forum] Pseudopotential with semicore states

[Stefano de Gironcoli]

Paolo or some other may correct me but this is how I think it is.

In order to generate the screened pseudopotential the KS equation must be 
inverted (for ONE pseudowavefunction per channel) and this is done using 
the semicore states that are nodeless.
In order to unscreen the potential one need the pseudocharge that is 
computed from all occupied states (both semicore and valence states).

Stefano de Gironcoli


On Mon, 9 Oct 2006, e kb wrote:

> Dear all,
>  I am trying to generate a pseudopotential with semicore states,
>  INPUT_LD1 file was very useful (thanks)
>  but there is this sentence that i still couldnt understand what is exactly meant:
>   (at Section 3.1 )
>   "...use semicore states ONLY to produce the PP, use semicore
>   AND valence states (if occupied) to make the unscreening."
>
>  It might be a very easy question, but i really couldnt get.
>  Why should i use ONLY semicore states to produce PP,
>   why not the -occupied- valance states  at that step?
>
>  Thanks,
>  Emine Kucukbenli
>
>
> ---------------------------------
> Get your email and more, right on the  new Yahoo.com