[Wien] Band gap as a function of effective U - LDA+U(SIC)

[pieper]

Dear Jihoon Park,

I don't have that much of an experience myself with orb, and you give too
little information, but my first guess would be that you didn't really call
orb and lapwdm. Maybe a few hints where to look can help:

Does your dayfile show a call to orb and lapwdm?
Did you generate case.indm? Since your case is magnetic you most probably
need case.indmc!
Do you use w2web? Did you switch on orb and dm there?

And finally: Does your ferrite structure REALLY contain 11 inequivalent Fe
atoms?

Best regards,

Martin Pieper

On Tue, 29 Jan 2013 09:46:02 -0600, Jihoon Park
<maximumenergyproduct at gmail.com> wrote:
> Dear Users,
> 
> 
> I have been trying to see the band gap dependence on Ueff for strontium
> ferrite.
> Therefore, three different Ueff were used in "case.inorb" as below.
>
----------------------------------------------------------------------------------------------------------------
>   1  11  0                     nmod, natorb, ipr
> PRATT  1.0                    BROYD/PRATT, mixing
>   1 1 2                          iatom nlorb, lorb
>   2 1 2                          iatom nlorb, lorb
>   3 1 2
>   4 1 2
>   5 1 2
>   6 1 2
>   7 1 2
>   8 1 2
>   9 1 2
>   10 1 2
>   11 1 2
>   1                              nsic 0..AFM, 1..SIC, 2..HFM
>    0.28 0.00        U J (Ry)   Note: we recommend to use U_eff = U-J and
>    J=0
>    0.28 0.00        U J
>    0.28 0.00
>    0.28 0.00
>    0.28 0.00
>    0.28 0.00
>    0.28 0.00
>    0.28 0.00
>    0.28 0.00
>    0.28 0.00
>    0.28 0.00
>
----------------------------------------------------------------------------------------------------------------
>   1  11  0                     nmod, natorb, ipr
> PRATT  1.0                    BROYD/PRATT, mixing
>   1 1 2                          iatom nlorb, lorb
>   2 1 2                          iatom nlorb, lorb
>   3 1 2
>   4 1 2
>   5 1 2
>   6 1 2
>   7 1 2
>   8 1 2
>   9 1 2
>   10 1 2
>   11 1 2
>   1                              nsic 0..AFM, 1..SIC, 2..HFM
>    0.52 0.00        U J (Ry)   Note: we recommend to use U_eff = U-J and
>    J=0
>    0.52 0.00        U J
>    0.52 0.00
>    0.52 0.00
>    0.52 0.00
>    0.52 0.00
>    0.52 0.00
>    0.52 0.00
>    0.52 0.00
>    0.52 0.00
>    0.52 0.00
>
----------------------------------------------------------------------------------------------------------------
>   1  11  0                     nmod, natorb, ipr
> PRATT  1.0                    BROYD/PRATT, mixing
>   1 1 2                          iatom nlorb, lorb
>   2 1 2                          iatom nlorb, lorb
>   3 1 2
>   4 1 2
>   5 1 2
>   6 1 2
>   7 1 2
>   8 1 2
>   9 1 2
>   10 1 2
>   11 1 2
>   1                              nsic 0..AFM, 1..SIC, 2..HFM
>    0.76 0.00        U J (Ry)   Note: we recommend to use U_eff = U-J and
>    J=0
>    0.76 0.00        U J
>    0.76 0.00
>    0.76 0.00
>    0.76 0.00
>    0.76 0.00
>    0.76 0.00
>    0.76 0.00
>    0.76 0.00
>    0.76 0.00
>    0.76 0.00
>
----------------------------------------------------------------------------------------------------------------
> 
> But, I do not see the band gap changes, even though Dr. Novak has done
the
> same thing and reported that the band gap gets larger if larger Ueff is
> used "PRB 2005".
> Please give me some idea what I am missing.
> 
> 
> All my best,
> Jihoon Park

-- 
Dr. Martin Pieper
Karl-Franzens University
Experimentalphysik
Universitätsplatz 5
A-8010 Graz
Austria
Tel. +43-316-380-8564