[Wien] Question in *.in1 file

[Peter Blaha]

> I have a question in the *.in1 file. That is, if it is reasonable to set a small searching steps in the in1 file, where the default value given by WIEN2K code is zero?

I would not do that. The general rule is: The energy parameters should be as
close as possible to the corresponding occupied eigenvalues. For Fe the
4s and 4p states are mainly above EF, therefore the "search" gives those high
energies. On the other hand, for such delocalized states the E-dependency of
the radial wavefunction is weak and it should have only a minor influence on
E-tot (but it would have a big influence when you set the Fe-d parameter not
close to EF).

> Now I take the FCC-Fe (with Spin-Polization) as an example to present my question. I set a small searching step 0.005 instead of 0.000 for S-state and P-state, see the following in1 file, but, the last linearizing energy of P-state (1.845) shown in the SCF file is great than FERMI ENERGY, 0.78285. Could somebody explain the reason? My goal is to get a smooth energy vs. volume curve, and I get it at last with the small searching step 0.005 (see below). Then, if I use searching steps for all states in the in1 file?
>
> The input in1 file:
> ----------------------
> WFFIL        (WFPRI, SUPWF)
>  10.00       12    6 (R-MT*K-MAX; MAX L IN WF, V-NMT
>   0.46    5  0      (GLOBAL E-PARAMETER WITH n OTHER CHOICES, global APW/LAPW)
>  0    0.42      0.005 CONT 1   <====
>  0   -6.50      0.005 STOP 1
>  1    0.58      0.005 CONT 1   <====
>  1   -3.97      0.005 STOP 1
>  2    0.62      0.005 CONT 1
> K-VECTORS FROM UNIT:4   -8.0       2.5      emin/emax window
> ----------------------
>
> After convergence (energy change less than 0.000001 Ry), the linearizing energies shown in the scf file (for spin-up) are:
> ----------------------
>           OVERALL ENERGY PARAMETER IS    0.4600
>           OVERALL BASIS SET ON ATOM IS LAPW
>  ==>   E( 0)=    0.6650   E(BOTTOM)=    0.665   E(TOP)= -200.000
>              APW+lo
>           E( 0)=   -5.7300   E(BOTTOM)=   -5.855   E(TOP)=   -5.605
>              LOCAL ORBITAL
>  ==>   E( 1)=    1.8450   E(BOTTOM)=    1.845   E(TOP)= -200.000
>              APW+lo
>           E( 1)=   -3.2525   E(BOTTOM)=   -3.480   E(TOP)=   -3.025
>              LOCAL ORBITAL
>           E( 2)=    0.6225   E(BOTTOM)=    0.090   E(TOP)=    1.155
>              APW+lo


                                      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/
--------------------------------------------------------------------------