Thin films

[Mike Jercinovic]

OK, I finally have to get into thin film analysis a bit.  I am sure I will struggle with it before figuring it all out, but basically I need to get started and also to get some idea of the expected precision/accuracy of the technique as implemented via PfE.  The samples in question will not require anything sophisticated (Stratagem processing etc) - thin Al, a few hundred nm thick maybe, on a polymer substrate.  There is probably something obvious in the documentation about how to go about setting this up, but I don't see this yet either.  This is an obvious plea for help.

MJ

[John Donovan]

Hi Mike,
For thin film work PFE is normally utilized to acquire multiple voltage acquisition (MVA) where the same points are measured using two or more electron beam energies. Because about 80% of my EPMA work is on thin films for material scientists, I have implemented a very easy way to automate the acquisition of this data.

Specifically you simply create a sample setup (and note that oxygen and other non-metals can become a significant issue depending on how the films were deposited and also the native oxide growth time scales), containing the elements of interest.

For example, a student might need to characterize a Bi-Te film on a silicon substrate.  So I would normally tune for Bi, Te, Si and O.  Then using the New Sample button create, say, three identical samples except that the beam energy is specified at 10, 15 and 20 keV. The exact beam energies may need to be different for particularly thin (< 50 nm) or particularly thick (> 400 nm) films, but 10, 15 and 20 keV beam energies are usually good for films in the range of 50 to 200 nm.

Then one goes to the Analyze! window and select all three samples and click the Add To Setup button. This specifies the three samples as "Sample Setups" which can be assigned to the standards (and unknowns of course).  The sample setup assignment is performed in the Automate! window using the Multiple Setups button. Once this is done you will note that the New Sample Basis option is now Use Multiple Digitized Sample Setups.

Then simply start the standard and unknown automation and you will see that the program will acquire each digitized set of points at the three specified beam energies.

[John Donovan]

For data processing I usually use the Output | Save STRATAGem Format menu to create an import file for STRATAGem* which using the method of Pouchou and Pichoir can iterate the intensities from the three beam energies to calculate both thickness and composition, by using the Import | STRATAGem* Import Format menu. This gives excellent accuracy for the composition, but the thickness is dependent on the accuracy of the density specified in STRATAGem*.

But..., it should also be noted that if you have a film on a low Z substrate (Si is usually ok), you can simply run Probe for EPMA and use the Use Particle/Thin Film Calculations button in the Calculation Options button in the Analyze! window. This is the method of Armstrong but assumes little or no contribution (of fluorescence and/or backscatter) from the substrate to the measurement.

The Armstrong strategy is to simply adjust the thickness of the film until you get a decent total and that should be your thickness (you'll need to be sure the density of the sample is correctly specified). If you acquired intensities at three beam energies you cannot combine them as in PAP, but you can perform this calculation on each of the three samples and see if you get similar results. Yes, it is more crude than the PAP method, but works surprisingly well also.

* © Copyright 1993-2016 SAMx

[Mike Jercinovic]

Thanks a million John.  That's very helpful.

Mike

[John Donovan]

Also see this related thread on modeling thin film intensities in CalcZAF (Standard.exe) using the Penepma 2012 GUI:

http://smf.probesoftware.com/index.php?topic=57.0

[Probeman]

Here's a link to the thin film paper for when you have the same element(s) in the film and the substrate:

http://epmalab.uoregon.edu/pdfs/Determination%20of%20Ni-Si%20Ultra-thin%20Films,%20Phung,%20et.%20al.,%202008.pdf

Basically one needs to "fix" the thickness using x-ray reflectivity (XRR) when elements are duplicated in the films and/or substrate.