[Wien] mBJ + U

[Peter Blaha]

mBJ is too "weak" to act sufficiently on the localized 4f states.

So most likely "mBJ + U" gives the better solution, but probably the U 
should be smaller than in LDA+U, because mBJ already shifts the f-states 
a bit.

Why should mBJ + SO not work ?
Maybe the standard kinetic energy density is ill defined when SO is 
added, but I'd expect it should still be possible ...

Am 08.02.2012 15:18, schrieb Pablo de la Mora:
> Dear Wien-users,
> I have a system with a 4d and a 4f elements.
> This system is an insulator, so my approach would be to use the mBJ
> potential.
>
> The normal calculation gives a small gap with the 4f peak above but
> close to the gap.
> Using the mBJ potential opens the gap considerably but moves the 4f peak
> into the gap and Ef is on the 4f peak. The calculation does not seem to
> converge. One could say that the system is an insulator since the the 4f
> orbitals are quite localized.
> If I put the mBJ potential plus a Hubbard U then the gap remains the
> same, but the 4f peak moves down in energy outside the gap.
> So, what is the correct solution? mBJ or 'mBJ + U'?
>
> I have a further problem, I want to add spin-orbit coupling, but it
> seems to have problems with the mBJ potential.
>
> Yours
>
> Pablo
>
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien

-- 
Peter Blaha
Inst.Materials Chemistry
TU Vienna
Getreidemarkt 9
A-1060 Vienna
Austria
+43-1-5880115671