Re: [Ifeffit] Preparation help with fly ash sample with diluted Tl

Hi Dariusz, Thanks, those are good suggestions. I was assuming that they'd use a solid state (say 10+ element Ge) detector for concentrations at and below 10 ppm. A large diode detector might work OK -- they can often take higher total count rate at the expense of energy discrimination. I'd be happy to be shown a comparison, but my guess is that electronically removing the Fe and Compton scattering would be worth the effort for these concentrations. --Matt 2011/9/29 "Dr. Dariusz A. Zając" :

Dear Matt, this is very good introduction for fluorescence measurements on diluted samples... Like a tutorial... I just want to add small comments: as I understood the sample is an ash, so maybe will be possible to condense it, e.g. press it into the pellet. If there will be no possible to measure exafs (due to heavier elements), and is needed to focus on xanes, than I suggest to use non-energy dispersive detector, e.g. PIPS diode. It is very often more sensitive than regular fluorescence detectors, and can take higher count-rate. please correct me if I am wrong... kicaj

W dniu 11-09-29 17:02, Matt Newville pisze:

...

Dear Mengling,

The basic issues are:  will there be enough Tl fluorescence counts in the detector, and will you be able to see a decent edge.    From you spreadsheet of concentrations, the elements that stand out are:  Fe and to a lesser extent Zn, and Pb.

The Fe and Zn will dominate the fluorescence getting to your detector.  This hurts sensitivity to Tl because the solid-state detectors can handle a limited flux themselves (typically on the order of 100K Hz). The high Fe concentration means the Fe Kalpha will dominate the emission when you shine 12.7keV x-rays on the sample.  The simple-but-effective remedy is to put aluminum foil on the detector, as aluminum will absorb Fe fluorescence (at 6.4keV) about 4x more than the Tl fluorescence (at 10.3keV).   You might need a few hundred microns of aluminum foil, and may need to try a few different amounts to cut down the Fe counts sufficiently.

The other issue is Pb.  The Tl L3 edge is at 12.658keV.  once you hit the Pb L3 edge, at 13.1 keV, the Pb Lalpha fluorescence will dominate the Tl Lalpha fluorescence -- these lines (10.3 and 10.5 keV, respectively) are too close together to effectively filter electronically.   Even well below the Pb L3 edge, you'll start to see some Pb Lalpha fluorescence.  Since the Pb concentration might be 100x the Tl concentration, that would be something to look for, but there's not a lot that can be done about it.   Anyway, that really only limits the EXAFS, not so much the ability to get XANES.  The Tl concentrations are low, but I think that at a good APS BM line like 5-BM, getting Tl L3 XANES is just a matter of time.

Hope that helps,

--Matt

On Thu, Sep 29, 2011 at 9:39 AM, Mengling Yi Stuckman  wrote:

...

Dear Ifeffit Community,

I am currently working on project "Evaluating Speciation of Thallium in Coal Fly Ash". The project was accepted by APS and we will conduct our measurement on beamline 5-BM-D.

My fly ash sample has Tl concentration ranging from 0.5ppm to 13ppm. How can I know whether the sample can generate enough photon counts if using fluorescence mode? what would be the appropriate sample preparation procedure I should follow? Is there any interference I should pay attention to?

I understand that I should use Hephaestus, but I don't know how to do that if my sample contains lots of different oxides. I attached the elemental composition in this email. Most metals should be in form of oxides, or sulfates.

Any input would be highly appreciated,

Mengling

_______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

_______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit

_______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit