Plot of wavescans and peak markers

[Philipp Poeml]

Hi John,

I have a question about the wavescan plots. So, I choose some element, I make a peak search and this "calibrates" the wavescan, right? Now, it seems, that for this calibration the value in the xlines.dat is used. Could be 3.5 keV. Now, if I display peak markers for other elements in the plot, it seams, that the xray.mdb is used.

And now it comes: The values in xlines.dat and xray.mdb are sometimes way off. It took me a while to figure out. Because If I calibrate the spectrum on (example!) Fe Ka, the peak marker of Fe Ka would be off then. I already thought my spectrometers were broken. But then I realized that the two databases are not the same.

Any comments on this?

Thanks
Philipp

[Probeman]

I have a question about the wavescan plots. So, I choose some element, I make a peak search and this "calibrates" the wavescan, right? Now, it seems, that for this calibration the value in the xlines.dat is used. Could be 3.5 keV. Now, if I display peak markers for other elements in the plot, it seams, that the xray.mdb is used.

And now it comes: The values in xlines.dat and xray.mdb are sometimes way off. It took me a while to figure out. Because If I calibrate the spectrum on (example!) Fe Ka, the peak marker of Fe Ka would be off then. I already thought my spectrometers were broken. But then I realized that the two databases are not the same.

Any comments on this?

Yes. Science is not perfect (or even self consistent)!

The XLINES.DAT file values are from Cal Tech.  The XRAY.MDB file values are from NIST. Maybe they should talk to each other? 

Seriously, think of it this way- there is some disagreement in science, especially for the low energy emission lines.

[Philipp Poeml]

... and M lines...

[Probeman]

... and M lines...

Yes, those too!

In fact it is even worse than that!  The datasets are not even "self-consistent". For example, if we look closely at the latest NIST x-ray database and compare the Mg Ka and Kb emission lines in MgO using a TAP crystal (not exactly high resolution), here is the full scan showing both Mg Ka and Mg Kb lines:



But is we zoom in a little, we can see that even the NIST database by itself, doesn't match the actual data as seen here:



Of course some of this discrepancy could be due to chemical states, and I don't have any (relatively unoxidized) Mg metal data to compare with.