[Ifeffit] difference spectra- Ifeffit Digest, Vol 159, Issue 23

[Bruce Ravel]

On 05/31/2016 03:34 PM, sb2c08 at gmail.com wrote:
> This is very helpful thank you. To give a little bit of context, the
> reaction is a solution of W(CO)6 irradiated at 266nm, to give
> W(CO)5*, it would be good to find the overall change of CO distance,
> especially as the multi scattering dominates the EXAFS. Just as a
> final point, the difference spectrum is taken from the normalised
> after and before laser spectra, so supposing the difference is fitted
> directly in Artemis (and coordination number is known/assumed), the
> amplitude factor will give an indication of the % change, supposing
> the Debye-Waller factor is similar to the ground state W(CO)6? But
> normalisation I find is the first difficult issue.

In general specific questions are better than general questions.  The 
way you posed your question initially seemed awfully open-ended and I am 
always unenthusiastic about open-ended questions.  A question about a 
specific material -- W(CO)6 in your case -- seems much more to the 
point.  Some of us here spend a lot of time trying to intuit what is 
meant by the questions posed on this list.  Specificity is much appreciated.

Are you actually asking me whether I think the sigma^2 will be the same 
for W(CO)5 as for W(CO)6?  I truly have no idea.

If you have been reading this list for a while, you will note that a lot 
of us so-called-experts make a big deal out of "interpretation" and 
"defensible results" and "prior knowledge" and the like.  The need for 
things like "interpretation" and "prior knowledge" is the reason that 
Artemis (and any program like it) requires a human driver.  While EXAFS 
fitting can be automated for certain kinds of well-defined problems, 
real research problems are really hard.  That's kind of why they are 
research problems!

Setting the W(CO)5 sigma^2 to its ground state value certainly sounds 
like a reasonable assumption.  I am hoping you know a heck of a lot more 
about the chemistry of your sample than I do -- whether that assumption 
is defensible is something that you will have to evaluate for yourself.

Ultimately, that is how hard data analysis is done -- you make a model, 
you test the model against the data, and you defend the results.  EXAFS 
never /solves/ structures.  Never.  I really mean that -- never.  But 
EXAFS analysis can be consistent with your data and part of a defensible 
story.  That's what you're doing here.

HTH,
B

-- 
  Bruce Ravel  ------------------------------------ bravel at bnl.gov

  National Institute of Standards and Technology
  Synchrotron Science Group at NSLS-II
  Building 535A
  Upton NY, 11973

  Homepage:    http://bruceravel.github.io/home/
  Software:    https://github.com/bruceravel
  Demeter:     http://bruceravel.github.io/demeter/