[Pw_forum] [pw_forum]Problem with pseudopotential generation: the case of Po

[Axel Kohlmeyer]

On Wed, 25 Jul 2007, Zhiping Yin wrote:

ZY> Dear PWscf users,

dear zhiping yin,

ZY> I tried to generate a pseudopotential for Po with the ld1 code. I tried to
ZY> generate both an USPP and a NCPP.   The USPP and NCPP  were  generated with
ZY> no error/warning messages in the output files. However, when I tried to test
ZY> the generated pseudopotential, it ran into problems. I had no experience on
ZY> pseudopotential generation before.  Can someone help me to check if there is

under those circumstance, it is highly recommended to first try
to reproduce some existing and easy to test/compare pseudopotentials,
so that you can practice your skills in generating them.

ZY> any mistake with the input files? Thanks very much.

it would have been very helpful, if you had stated _what_ the
error was that you get.

ZY>  &test
ZY>   nconf=1
ZY>   file_pseudo='Po.rel.us.5d.RRKJ3.UPF',
ZY>   configts(1)='4f14. 5d10. 6s2 6p4'
ZY>  /

my guess would be that the error lies in specifying 4f14 here.
you didn't put any f electrons in the valence, why should 
you need to add them here?

please remember, that for transferability tests, you also
should test different electron configurations. see the input
examples in the atomic_doc/pseudo-test subdirectory.

cheers,
   axel.

[...]

-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
   Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.