[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Fwd: Spin dependent LDOS calculations
Dear David,
On Tue, 6 May 2003, David Eustace wrote:
>
> Thought i'd try again with this one.
> >Sorry for bombarding the mailing list a bit today, but on a related note
> >i've been trying to get the spin-up and spin-down LDOS of CrN. With help
> >from Angel Garcia-Adeva (thank you), i'm running Feff8.2 with the AF part
> >of the code changed to:
> >
> >if(iph.eq.0) dmag(i,iph) = dmag(i,iph)
> >if(iph.eq.2) dmag(i,iph) = dmag(i,iph)
> >if(iph.eq.3) dmag(i,iph) = -dmag(i,iph)
> >if(iph.eq.1.or.iph.gt.3) dmag(i,iph) = 0.0
> >if(iph.eq.0) dmag(i,iph) = 0.0
> >
> >where potential 0 is my absorbing nitrogen atom (essential for xanes),
> >potential 2 is spin-up chromium,
> >potential 3 is spin-down chromium
> >potential 1 is non-absorbing nitrogen.
> >
> > From my previous email you can see i'm getting good results for the Xanes
> > region, but the LDOS agreement is not so good.
Could you explain, what do you mean by that? With what do you compare
the LDOS result? Did you use SPIN to get spin-dependent DOS?
> >I've followed the steps given in the manual, but i'm now at the stage
> >where i'm running out of variables to play with.
If you compare with some ground state calculations, you should do
calculations with NOHOLE. Also FEFF8 is not self-consistent code
with respect to spin on Cr. You can change that manually, by editing
subroutine getorb and changing the number of the variable ispn(24,i).
there is only one nonezero value in that array - 4. corresponding to
4 spin-up 3d electrons on Cr (iph=2) and 4 spin-down on Cr (iph=3).
This number should not be integer.
Also comparing to band structure calculation may require to do
calculation for bigger clusters.
> >
> >I was hoping anyone with a bit more experience of spin-dependent
> >calculations could point out any difficulties they've had to deal, any
> >unforeseen things that i'm probably missing. Any general advice would be
> >very welcome.
> >
> >Thanks again,
> >
> >Dave E
>
Best wishes
Alexei Ankudinov