Re: comparing results of LDOS and XANES calculations

["John J. Rehr" <jjr@leonardo.phys.washington.edu>]

Hi Lydia,

 To compare the XAS against the LDOS, one should pay attention to the
appropriate selection rules. It's not the total DOS, but the projected
DOS that matters.

 Thus the XAS for a K-shell core level (s-initial state) will have
l=1 (p-final states), so you should compare the xmu.dat col 2 and 4
against ldos00.dat col 1 and 3 (with FEFF8.2) or rhol00.dat col 1 and 4
(with FEFF8.0).

 For L2,3 core levels (p initial state) the final state is predominantly
l=2 (d-final states) so you should compare with the dDOS. OK?

  Cheers,
  John Rehr



On Fri, 3 Dec 2004 ALNemir@aol.com wrote:

> Dear Mr. Rehr,
>
> I have plotted together the spectroscopic data (xmu.dat file) and the local
> densities of states data (rholNN.dat files) using your answer to my original
> question.
> There is one rholNN.dat file written for each entry on the POTENTIALS list,
> and each file has 4 columns of LDOS data. To obtain the total DOS data I have
> plotted I summed the 4 columns in each rhol.dat file and summed that data from
> all rhol.dat files.
> The result is plotted in the attached file. While the match is far from
> perfect, for most of the peaks there is at least a correlation in position, and for
> one (at ~ 25eV) also in amplitude. However that is not the case with the
> main, highest peak (at ~ 10 eV). There is nothing in the local DOS that would
> explain it, save to a peak to the left (at ~ 11 eV).
>
> I was wondering if you might guide me in the right direction.
>
> Thank you,
>
>
> Lydia Nemirovsky
> Research Assistant
> EXAFS Laboratory
> Department of Physics and Astronomy
> Hunter College