On Wednesday 16 July 2008 15:27:27 Scott Calvin wrote:
Bruce--I'm not following your critique of the third panel. Where's the confusion? I see the disorder term, and I see the mean-free path term, but I don't see them inappropriately labeled or confused with each other (aside from the common use of "Debye-Waller factor"). Unless perhaps it's just that the spacing and arrows in the label "From ab- initio calculations or from reference compounds" is confusing? (i.e. it kind of looks at first like the slide claims that the mean-free path comes from a reference compound)
Unless you redefine sigma^2 such that it has units of 1/angstrom^2 and it subsumes the 2 we usually put in that exponential, the expression exp( -k^2 / sigma^2 ) is incorrect. I concur that one is always free to define ones terms. But redefining what sigma means in the exafs equation can only lead to confusion. (BTW, When you analyze your data by comparison to an empirical standard, I would presume that the mean free path is considered to be chemically transferable.) B -- Bruce Ravel ------------------------------------ bravel@bnl.gov National Institute of Standards and Technology Synchrotron Methods Group at NSLS --- Beamlines U7A, X24A, X23A2 Building 535A Upton NY, 11973 My homepage: http://xafs.org/BruceRavel EXAFS software: http://cars9.uchicago.edu/~ravel/software/exafs/