I think I understand your question now, though, and can answer it with a clear example.
Suppose sample A is:
1% iron metal by mass
2.5% hematite (Fe2O3) by mass
the rest is not iron compounds at all
Sample B is:
2% iron metal be mass
5% hematite by mass
the rest is not iron compounds
Sample C is:
1.5% iron metal by mass
2.5% hematite (Fe2O3) by mass
the rest is not iron compounds
The normalized spectra of samples A and B will be identical if there are no other differences between them.
The normalized spectrum of C will be different from the other two, because the ratio of hematite to iron metal is different.
The ratio of hematite to iron can be determined from the normalized XANES of each sample.
Hopefully that helps.
--Scott Calvin
Sarah Lawrence College
On Oct 29, 2008, at 2:31 PM, Jens Kruse wrote:
Thanks for your replay and sorry for the confusing question. but I think it not getting better.
Scott, you are right LCF would be helpful but even with lot of standards it is hard to get a meaningful result from low concentrated samples at the N k-edge.
I think my problem is than I don`t understand the physical meaning of a edge step normalized spectra. Is it the spectrum 1 x-ray photon would produce? Is it right that I loose the information of the absolute concentration after normalization so that I still can identify relative proportions of different compounds with in a sample but I can`t compare different spectra in an absolute way anymore? The extreme case would be a comparison of pure and a diluted spectra.
jens