Effective Teaching Methods for EPMA

[Probeman]

All,
I've started this topic in order to share our most useful and interesting ideas and resources for teaching EPMA to students. I hope students will chime in as well to tell us what they deem most effective also!

I'll start by a small example of something that I have found to be valuable in teaching students the difference between the emission energy and the edge energy.

I usually start by turning off the automatic peak ID in the EDS software and acquiring a number of spectra of pure metals at 20 keV. Then I get the students to identify the peaks and all families (K, L and M), if they are visible, to confirm the ID.

Finally I go to Mo metal and acquire a spectra and ask them to identify it and they usually do as seen here:



Then I ask: but why don't we see the Mo Ka line?  Usually some bright young person finally speaks up and asks what the critical excitation energy is for Mo Ka and yes, lo and behold, it is just over 20 keV!

Then I have to explain why Thermo NSS still shows the peak marker for Mo Ka when the beam energy is 20 keV!



Not to worry, Bruker Esprit also shows the Mo Ka lines when at 20 keV as seen here:



At least Thermo and Bruker should draw these "non-excited" KLM markers using dotted lines or something- don't you think?

Anyway, if anyone is interested, my EPMA class notes and PPT files are here:

http://epmalab.uoregon.edu/lecture.htm

and I'd be pleased if others who teach a similar course would share their tips and resources also in this topic!

[Probeman]

One of my student test questions is usually along the lines of how does keV, atomic number and density affect BSE production?

They usually get the first two right, but often fail with the last (density). So I ran some Penepma simulations recently for gold at a variety of densities and plotted the results so I can show them and explain why this might seem counter-intuitive (for one thing the nuclei are already very far apart so multiplying that (vacuum) distance doesn't affect BSE rates).

See attached.

[Probeman]

This is a good challenge for students near the end of the term to get them to put things together a little more in their heads:



Think about what x-rays are produced in pure Fe...

Attached below is a pdf if you want to use it in your teaching.

[John Donovan]

I came across this bibliography for EPMA that looks to be useful, though it may not have been updated the last 5 years or so from the looks of it:

http://www.nendai.nagoya-u.ac.jp/CHIME/epmadoc.html

[Probeman]

This is a one slide explanation of absorption and fluorescence in the context of matrix corrections originally from John Fournelle (see attached pdf). I think it is a very effective presentation for students.

Faculty: what resources do you all use for your matrix correction lectures?
Students: do you find this slide a useful explanation? Why or why not?

[John Donovan]

I've recently (06/24/2014) made some nice changes to the Penepma GUI in Standard.exe. This application is available *for free* along with the EPMA utility CalcZAF.exe, as described here:

http://probesoftware.com/smf/index.php?topic=81.0

The latest version of Standard.exe makes it even easier to model secondary fluorescence couples and inclusions as described here:

http://probesoftware.com/smf/index.php?topic=59.msg1382#msg1382

Though for most analytical situations involving a straight boundary, the Penfluor/Fanal approach (also in Standard.exe) is usually much preferred to greatly reduce calculation time as described here:

http://probesoftware.com/smf/index.php?topic=58.msg214#msg214

[Probeman]

One of my student test questions is usually along the lines of how does keV, atomic number and density affect BSE production?

They usually get the first two right, but often fail with the last (density). So I ran some Penepma simulations recently for gold at a variety of densities and plotted the results so I can show them and explain why this might seem counter-intuitive (for one thing the nuclei are already very far apart so multiplying that (vacuum) distance doesn't affect BSE rates).

See attached.

For estimating/comparing BSE of compounds, the important question is: how should we calculate "average Z"?   Since from physics we know atomic weight does not affect BSE production (see attached), clearly weight fraction averaging, as is commonly utilized, is a close but not exact proxy, for calculating the average Z of compounds...

We proposed in the paper below a modified electron fraction averaging (which includes a correction for nuclear screening by inner orbital electrons), but my guess would be that we should probably utilize some kind of an "elastic scattering cross section" fraction, but that isn't nearly as convenient as simple mass fraction!

Above is a post that shows the effect of different densities in the bulk Au BSE calculation (see the attached plot):

http://probesoftware.com/smf/index.php?topic=126.msg508#msg508

Hint: there shouldn't be a density effect on BSE for bulk materials and Penepma demonstrates this nicely.