[Ifeffit] large Eo's and neg. D-W factors

[Scott Calvin]

Hi Rich,

It depends on the information you want out. Here's where it's good to 
think about correlated effects. Sigma2's affect primarily the 
amplitude. They therefore correlate highly with coordination number, 
S02, normalization, sample quality (e.g. variable thickness in 
transmission; self-absorption in fluorescence), beamline effects 
(e.g. harmonics), and the like.

Other variables are primarily a question of phase: E0, delr, and the 
third cumulant, to name some of the more important ones.

Of course, some factors, such as the elemental identity of the 
scattering atom or the fraction of various phases that are present, 
affect both amplitude and phase.

If you're interested in knowing about the phase variables, then the 
amplitude variables are less crucial. A slightly wonky value can come 
from something else being slightly wonky; e.g. normalization. That's 
not ideal, and should make you nervous enough to poke at it a bit 
(for example, go back and look carefully at the normalization in 
Athena). But it's not nearly as serious as if you are interested in 
an amplitude variable. For example, if you're trying to fit 
coordination numbers, then a weird sigma2 is a big big red flag.

Finally, this is the kind of place where analyzing a standard of 
known structure is immensely helpful. Likewise, a temperature series 
can be very helpful for sorting out some of the correlated variables.

Hope that helps.

--Scott Calvin
Sarah Lawrence College

At 10:52 AM 5/2/2007, you wrote:


>Also, should I be worried if my Debye-Waller factors get small
>(0.0005-0.001) if I optimize my model in FEFF to a point that my delr's
>are on the order of 0.005-0.01 Angstroms?  This has been a concern of
>ours since we started doing EXAFS, especially since all of our
>measurements are at room temp.  We haven't seen this as much with our
>vanadium on silica samples, but it seems to be true with all of our
>titanosilicate samples