On Sunday 05 November 2006 06:38, levy@msri.org wrote:
Dear List,
I have noticed in several fits that Artemis -- i.e., feff -- reports for a SINGLE shell (with absorber = scatterer = Hg, edge = L3) an almost-node in the envelope of chi(k) around 6 inverse angstroms. See the image at
http://pjm.math.berkeley.edu/users/levy/files/HgNode.gif
and the FEFF output after my signature.
I've done my best to account for this behavior, but I simply don't have a physical intuition for why the F(k) factor should dip to zero or almost zero. Note that this is not a beat caused by two similar frequencies being added; the Fourier transform does of course show two peaks -- which is how my curiosity was aroused -- but the example figure comes from a SINGLE path, with 12 identical scatterers all at the same distance from the absorber.
Looking forward to your two cents...
Silvio, Look at figures 6 and 8 in the Rehr-Albers Rev. Mod. Phys review of XAS theory: Reviews of Modern Physics, Vol. 72, No. 3, July 2000 Figure 8 clearly shows the same phenomenon in real data. Figure 6 shows the reason for it. At some value of k, there is a minimum in the scattering amplitude. Multiplying an F(k) with that shape by a pure sine wave will have a similar effect as the interference between two sine waves with wavelengths such that they beat at the minimum of F(k). This is one of the many reasons that chi(R) is not a radial distribution function. So why is there a minimum in F(k)? Well, the simple, fairly naive explanation is that the wavelength at that energy is well matched to the size of the scatterer and the photoelectron tunnels right through. You'll find that most heavy backscatterers display this behavior. HTH, B -- Bruce Ravel ---------------------------------------------- bravel@anl.gov Molecular Environmental Science Group, Building 203, Room E-165 MRCAT, Sector 10, Advanced Photon Source, Building 433, Room B007 Argonne National Laboratory phone and voice mail: (1) 630 252 5033 Argonne IL 60439, USA fax: (1) 630 252 9793 My homepage: http://cars9.uchicago.edu/~ravel EXAFS software: http://cars9.uchicago.edu/~ravel/software/exafs/