[Ifeffit] Is good EXAFS data possible with 0.01 transmission coefficient

[Matthew Marcus]

Is most of the transmission loss due to the Mn  or the matrix (C, KOH, etc)?  Is that 0.01 above or below the edge?
If most of the absorption is due to the matrix with the Mn providing an edge jump of <1, then I'd say that you have a shot.
However, you become very sensitive to harmonics and glitches.  If the Mn provides an edge jump of much less than 1, them fluorescence
would be the way to go, if possible.  With thick samples in transmission, you have an issue with hole effect.  For instance, it's
possible that your 1% transmission comes entirely from 1% area of pinholes, with the actual material being almost completely
opaque.  If that happens, you get a reduced edge jump and reduced EXAFS amplitude, with distortions resembling overabsorption.

If you're in the thick-matrix scenario, a possible workaround would be to make a sample which contains only matrix and is as nearly as
possible exactly as thick as the real sample, and measure its transmission, and subtract.
	mam

On 7/25/2013 1:07 PM, Damon Turney wrote:
> Hi IFEFFIT community,
>
> I will make measurements soon at Brookhaven's NSLS to do EXAFS, and my
> sample is a powder mixture of carbon powder, MnO2 powder, and
> potassium hydroxide liquid.  The transmission coefficient of 6.5keV
> x-rays (the Mn K edge is at ~6.5keV) through the sample is ~0.001 to
> 0.01.  I am told that the ionization chamber detectors can easily
> detect the x-ray beam after this ~0.01 transmission loss (by
> increasing the gain on the detector), but I would like to ask the
> IFEFFIT community if there will be other problems with the EXAFS
> technique when the transmission coefficient is so low.  Does anybody
> have comments?
>
> Much thanks -- if you have any info I greatly appreciate it!
> Damon Turney
> City College of New York
> _______________________________________________
> Ifeffit mailing list
> Ifeffit at millenia.cars.aps.anl.gov
> http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
>