[Pw_forum] projector for local channel in ultrasoft pseudopotential

[Gabriele Sclauzero]

Dear Alaska,

asubedi wrote:
> Dear all,
> 
> I am a complete newbie trying to create an ultrasoft pseudopotential
> for bismuth. I would like to use l=1 for the local part and I also
> want projectors for it.  So I have (excluding other valence electrons)

Why do you want to do this?

Do you understand the difference (theoretical and practical) between local part and 
non-local part of the pseudopotential?

Given an electronic shell, for instance 6p, either you describe it with projectors or you 
include it in the local part.
The last option is a reasonable choice only for some shells as far as I know (either s 
electrons or empty shells). Not sure for the 6p shell of Bi, which is in valence and 
partially occupied.

Maybe someone else can give a more correct explanation than mine.


HTH

GS

> 
> 6P  2  1  2.00  0.00  3.10  3.20  0.5
> 6P  2  1  0.00 -0.50  3.10  3.20  0.5
> 6P  2  1  1.00  0.00  3.10  3.20  1.5
> 6P  2  1  0.00 -0.50  3.10  3.20  1.5
> 
> However, the code gives an error:
> 
>      from ld1_readin : error #         1
>      only one local channel
> 
> Am I right in concluding that ld1.x cannot make ultasoft
> pseudopotential for the local channel? I checked the formula for
> decomposition of pseudopotential to local plus nonlocal part. For the
> nonlocal part, summation is over l=0 to lmax and does not exclude
> l_local.
> 
> Thanks,
> Alaska
> UTK and ORNL

-- 


o ------------------------------------------------ o
| Gabriele Sclauzero, PhD Student                  |
| c/o:   SISSA & CNR-INFM Democritos,              |
|        via Beirut 2-4, 34014 Trieste (Italy)     |
| email: sclauzer at sissa.it                         |
| phone: +39 040 3787 511                          |
| skype: gurlonotturno                             |
o ------------------------------------------------ o