Ar by EPMA

[Dan R]

Someone recently asked me if EPMA could detect Ar in a thinfilm. I did not think that this was possible because of the P10 gas -- however I do have some Xe detectors. Additionally, a solid standard doesn't exist but maybe there's a way using the interpolated standard intensities with PFE? Am I crazy or has anyone tried something like this?

[John Donovan]

If you merely want to detect if Ar is present or not, a high precision wavescan should be sufficient.

If you want to quantify the Ar, you should utilize the "virtual standard" feature in Probe for EPMA:

https://probesoftware.com/smf/index.php?topic=179.0

[Brian Joy]

I did not think that this was possible because of the P10 gas -- however I do have some Xe detectors.

The identity of the counter gas does not matter.  Keep in mind that continuum radiation can also have energy = 2.96 keV, and these X-rays have no problem ionizing the P-10 gas.

[Anette von der Handt]

There have been people studying Ar in basaltic glasses by EPMA. You might want to check out these papers for more information and maybe the authors can provide more details.

This study used the virtual standard feature in PFE as John suggests and peaked on Ar-bearing materials they had synthesized:
Jackson, C.R., Williams, C.D., Du, Z., Bennett, N.R., Mukhopadhyay, S. and Fei, Y., 2021. Incompatibility of argon during magma ocean crystallization. Earth and Planetary Science Letters, 553, p.116598.

Here they used a calibration curve approach (if only they had PFE)  :
Fabbrizio, A., Bouhifd, M.A., Andrault, D., Bolfan-Casanova, N., Manthilake, G. and Laporte, D., 2017. Argon behavior in basaltic melts in presence of a mixed H2O-CO2 fluid at upper mantle conditions. Chemical Geology, 448, pp.100-109.

No information on EPMA standardization. The Fabbrizio paper states that they used one of their materials as standard:
Schmidt, B.C. and Keppler, H., 2002. Experimental evidence for high noble gas solubilities in silicate melts under mantle pressures. Earth and Planetary Science Letters, 195(3-4), pp.277-290.

The biggest problem I see is getting your hand on such an Ar-bearing material to at least peak on unless the abundance is high enough in the sample already.

[Probeman]

The biggest problem I see is getting your hand on such an Ar-bearing material to at least peak on unless the abundance is high enough in the sample already.

Yes. This is the hardest part (if one already has Probe for EPMA- which Dan does!).   

One solution to this problem is (when tuning up the two elements on either side of the Ar Ka peak, say Cl Ka and K Ka), to simply set your Ar Ka spectrometer position to the same *offset* you find on your Cl and K peaks. Or at least the average offset of the two if they are different.

[Probeman]

There's also a feature in the StartWin app which allows one to calibrate the spectrometers using known peaks. This is useful if the difference between the actual on-peak positions and the theoretical on-peak positions are not constant over the range of the spectrometer.

Here is a screen shot from the PFE Reference manual:



This feature is accessed using the Peak/Scan Options button in StartWin (from the Count Acquisition window) and clicking the Calibrate Parameters button.

Here is some more discussion on this feature:

https://probesoftware.com/smf/index.php?topic=797.msg5015#msg5015

Note that you need to have the UseMultiplePeakCalibrationOffset flag set in the Probewin.ini file for this to work. Once you have the spectrometer(s) calibrated using multiple peaks, Probe for EPMA will automatically utilize this calibration curve to calculate the peak position when selecting a new element.

[Probeman]

Also it just occurs to me, if Ar trace accuracy is an important question, at least finding a zero blank for Ar to test accuracy should be easy!     E.g.:

https://probesoftware.com/smf/index.php?topic=29.msg387#msg387

Though maybe not for Ar in a thin film!

I also seem to remember that Cal Tech had an Ar impregnated glass or feldspar(?) standard at one time I think.  Paul Asimov might know more.

[Paul Carpenter]

Yes, Caltech had an Ar standard at one time. It was a graphite planchet that had been implanted with Ar. The implantation was accurately known and determined using a known flux and time for implantation. It is possible there was more than one and a calibration curve was generated but we only measured one set of samples. Today, using PFE, it would be entered as a standard and used just like other reference materials. Because the Ar was neutral, it was very stable in the graphite both at room conditions and in vacuum and no other precautions were necessary.

Paul

[Dan R]

Thanks everyone, i'll attempt your recommendations in the next few weeks... first order of business is to determine whether I can distinguish a peak for Ar between 2 different samples -- a control and an unknown (although both are technically unknowns).