Hello all,
This morning I added the Bruker EDS detector that is installed on our CAMECA microprobe to DTSA-II and ran some simulations of pure copper to get an idea of how well the results matched up with spectra collected on the instrument. The simulated spectra had approximately 5 times the counts (in the characteristic lines) of the spectra collected from a copper standard on the instrument. I checked my detector and simulation parameters, but couldn't find anything that would explain such a large discrepancy.
I then remembered that the EDS detector is equipped with a three-position aperture (large, medium, small) that is used to reduce the count rate going into the detector to avoid excessive dead times when high beam currents are used. The default position is medium which is what I had been using to collect the spectra that I was comparing to the DTSA-II simulations. I changed to the large aperture setting and collected a new spectrum on the copper standard. This new spectrum proved to be a much better match against the DTSA-II simulated spectra.
The difference in the detector incoming count rate between the medium aperture setting and the large aperture setting is about five times on the copper standard. Out of curiosity, I reduced my simulation probe-dose by a factor of five and got a simulated spectrum that reasonably matched the spectra I had collected using the medium aperture.
Long-term, I would like to leave the aperture on the medium setting so I'm trying to identify potential problems I might encounter when using spectra collected with the medium aperture in DTSA-II. So far, these are my thoughts and questions:
- I think spectrum quantification should still work, so long as standards and specimens are both collected using the medium aperture setting to avoid any errors due to intensity loss, but maybe I'm overlooking something.
- What is the best way to approach simulations? Reduce the probe dose as I did before? Find out what the aperture sizes are and change the detector area in DTSA-II to match? Something else?
- Is compensating to get more realistic simulation results even a good idea? Would different specimens require different probe dose compensations?
All advice is appreciated. Thank you!
Essentially an aperture reduces the effective detector area. The best bet is to create a detector model with an area consistent with the apertured area.
So long as you use the same aperture to collect standards and unknowns, standards-based quant should work as per normal.
Nicholas