[Wien] LDA+U

[Peter Blaha]

> I'm using fcc Ni  example to learn how to do LDA+U and I have Wien2k_03
> version.
> First I do a spin polarized calculation using LDA :
> runsp_lapw   -cc  0.0001
>
> and then
> runsp_lapw -orb -cc  0.0001
>
> The calculation stops after the first cycle  without any error messages,
> however when
> I start again I get the message
>
> STOP:  LAPW0 END
> STOP:  ORB   END
> STOP:  ORB   END
> if: Badly formed number

a) Are you sure your version of WIEN2k_03 is a current one ? Eventually
   update the scripts in "SRC"
b) I think to remember that we had this problem on the mailinglist before.
   On some particular computer (SGI ?) it did not understand the statement
          if ( "$so" == "-so" && -s $file.dmatud ) then
   Try to comment this if statement in runsp_lapw

>   1  1  0                     nmod, natorb, ipr
> BROYD 0.40                    BROYD/PRATT, mixing
>   1 1 2                          iatom nlorb, lorb
>   0                              nsic 0..AFM, 1..SIC, 2..HFM
>    0.590 0.07                    U J (Ry)

Never use BROYD anymore in case.inorb. We are mixing the density matrices
in mixer. Instead use:

PRATT  1.0                    BROYD/PRATT, mixing




                                      P.Blaha
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-15671             FAX: +43-1-58801-15698
Email: blaha at theochem.tuwien.ac.at    WWW: http://info.tuwien.ac.at/theochem/
--------------------------------------------------------------------------