[Pw_forum] XSpectra on graphite slab

Alain Allouche alain.allouche at univ-provence.fr
Thu Mar 17 08:37:59 CET 2011


Merci Paolo, I should have though by myself... But even without  
London, after the right results of pw, XSpectra gives the message  
which I had not with diamondh

  -------------------------------------------------
  --- Polarisation and k vector [cartesian coordinates]----
xepsilon(:)= 1.00000000 0.00000000 0.00000000
xkvec(:)= 1.00000000 0.00000000 0.00000000

  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
%%%%%%%%%
      from main program : error #        63
      Wrong xiabs!!!
  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 
%%%%%%%%%

      stopping ...

using these data:

  &input_xspectra
     calculation='fermi_level',
     prefix='graphite',
     xread_wf=.true.,
     xiabs=1
  /
  &plot
  /
  &pseudos
     filecore='C.wfc',
  /
  &cut_occ
  /
6 6 1  0 0 0

I saw in the source that this error occurs when the atom's type is not  
found, but which one? I cannot understand
regards
Alain


Le 16 mars 11 à 23:12, Paolo Giannozzi a écrit :

> On Mar 16, 2011, at 16:22 , Alain Allouche wrote:
>
>> Can somebody explain me the curious behavior of pw.x accepting Ch
>> on  diamond and not on graphite, and  this "Wrong xiabs!!!" ??
>
>
> I'll do half of the job. It is the "london" option that makes the
> difference.
> Actually "Ch" is not acceptable syntax, or, more exactly: function
> "atomic_number" accepts "X", "Xn", n=0,...,9, "X-*", "X_*", where X
> is any atomic symbol. "Xh" confuses it. Since such routine is not
> actually
> called, except in some special cases (DFT-D is one such cases), this  
> has
> gine unnoticed until now
>
> P.
> ---
> Paolo Giannozzi, Dept of Chemistry&Physics&Environment,
> Univ. Udine, via delle Scienze 208, 33100 Udine, Italy
> Phone +39-0432-558216, fax +39-0432-558222
>
>
>
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum




More information about the users mailing list