[CPMD-list] Pt uspp potential incompatible with cpmd

[Axel Kohlmeyer]

On Fri, 6 Apr 2007, Li Chen wrote:

LC> Dear all,

dear li,

LC> I used uspp 736 to generate an Pt vdb pseodopotential and reformed it to a
LC> formatted one. But cpmd complains this potential "incompatible with CPMD".

the VDB support in CPMD currently requires that the pseudopotential
has the same number of projectors for each angular momentum. the 
Pt psp in the vdb library does not match this requirement and thus
the error message (older versions of CPMD would acutally continue 
and produce bogus results). it is not a problem of the format.

LC> Then, I generated a Carbon vdb pp from uspp736, and this time it worked.
LC> 
LC> Then I tried gold, and got the "incompatible with CPMD" error again. But the
LC> au_vdb_***.psp potential which I got from cpmd website _is_ compatible with
LC> CPMD.

yes. because it is a _different_ pseudopotential, i.e. the parameter
set for the generation of the potential has been modified to meet
the CPMD requirements.

LC> So..I'm not really sure where I went wrong... Is there any way that I can
LC> make it compatible?

see above. please note, that this is a nontrivial procedure. 
well, changing the parameters is simple, but the resulting 
pseudopotential can easily cause bogus results, if the changes
are not done with care and the potential thoroughly tested.

LC> Also, I think cpmd does not support kpoints for vdb(??), is there somewhere

that is (mostly) correct. there used to be an experimental version 
with k-point support, but it never got folded back into the 
development tree (it was based on a much older version of CPMD,
and the merging would require a _lot_ of work. as far as i know
none of the currently active CPMD developers has the time and 
interest(?) to put in the effort).

you may want to have a look at the quantum espresso 
and/or abinit packages instead.

LC> that I can get a norm-conserved pp for Pt?

there should be norm-conserving platinum pseudopotentials
in the updated goedecker library maintained by the CP2k
project at:
http://cvs.berlios.de/cgi-bin/viewcvs.cgi/cp2k/potentials/Goedecker/cpmd/

cheers,
   axel.
LC> 
LC> 
LC> Thanks a lot,
LC> Good day,
LC> 
LC> Li
LC> 

-- 
=======================================================================
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.