Questions about MAN background use

[Panseok]

Hi John,

I had to update 35 elements! I was in a hurry yesterday and missed updating a couple of elements.

Thank you,
Panseok

[John Donovan]

Hi John,

I had to update 35 elements! I was in a hurry yesterday and missed updating a couple of elements.

Thank you,
Panseok

No worries!  I miss things all the time, but fortunately PFE never does!   

[John Donovan]

A colleague of ours asked us recently: why don't we make the "Z Fraction" averaging method the default for MAN background fits?

Well, it's clearly a better method especially for high Z compounds that have elements with large differences in their A/Z ratios, but the thing is, we hesitate to modify a software default because it could cause confusion when users expect a certain behavior.

So yes, you *should* use the Z Fraction MAN averaging method, but since this is more of an "education" issue I will show here why you should. Even for materials with relatively low average Z values.

Here is a plot of Mn Ka continuum values for standards that do not contain Mn:



Note the variance of the MAN background fit. Now here are the same continuum measurements but now calculated using the Z Fraction averaging:



So, yes we really should be using the Z Fraction averaging method for MAN fitting. And for that matter we should also be using the Z Fraction DAM backscatter method in our matrix corrections:

Donovan, John, et al. "An improved average atomic number calculation for estimating backscatter and continuum production in compounds." Microscopy and Microanalysis 29.4 (2023): 1436-1449.

https://smf.probesoftware.com/index.php?topic=1566.0

But we're not going to change the default behavior of the software just yet. But we ask that users should utilize this more accurate method averaging when using MAN background corrections.

Note that if you make a new probe run and load a file setup from an old run which had this Z based averaging method selected, the software will of course utilize this Z based averaging method automatically.

https://smf.probesoftware.com/index.php?topic=106.0

[AndrewLocock]

Are MAN background curves linear or quadratic?
This is the question that came up with a colleague the other day.

As our lab is running sulfides this week, with MAN backgrounds for both standards and unknowns, we investigated this topic.

Over a large range in Z (generally from boron metal to bismuth metal), the MAN backgrounds are quadratic.
This is for a range in X-ray energies, from Si K-alpha (1.74 keV) to Au L-beta-one (11.44 keV).
The data for the MAN backgrounds were acquired at 20 kV, 20 to 30 nA, beam diameter 10 microns, and count time 30 s on peak. Six or seven points were acquired on each material.

I have attached a PDF of screen captures of the MAN background fits for 18 elements (file: AAN bkgd fits Mar31_2026.pdf).  You cannot view this attachment.

Also attached is a PDF that shows a table comparing the percent-relative-deviations of the quadratic and linear fits, organized in order of increasing X-ray energy (file: Table MAN bkgd fits.pdf).  You cannot view this attachment.

In summary, the quadratic functions fit the MAN background data much better than the linear curves, for a large range in Z.

Admittedly, if one were to look only a small range in Z, a linear approximation might be acceptable.

Cheers, Andrew

[John Donovan]

Are MAN background curves linear or quadratic?

In summary, the quadratic functions fit the MAN background data much better than the linear curves, for a large range in Z.

These MAN curves are very impressive. Nice work!

I should note that Andrew is utilizing our "Z fraction" method for fitting these MAN curves. This must be selected manually as described here:

https://smf.probesoftware.com/index.php?topic=4.msg14006#msg14006

The physics justification for this method can be found here:

https://academic.oup.com/mam/article-abstract/29/4/1436/7224307

Donovan, J., Ducharme, A., Schwab, J. J., Moy, A., Gainsforth, Z., Wade, B., & McMorran, B. (2023). An improved average atomic number calculation for estimating backscatter and continuum production in compounds. Microscopy and Microanalysis, 29(4), 1436-1449.