I'm trying to fit the Mg2p peak. In the literature, just like for Al2p, I observed that the Mg2p peak is fitted with a single peak (not with 3/2 and 1/2 components). Is this because AlK-alpha cannot resolve the spin-orbit component? or is there any other reason?
As always,
Thanks for your help
In the Scofield cross sections, the 1/2 and 3/2 component have the same energy within the significant digits (https://www.osti.gov/servlets/purl/4545040/) so the effect is very weak for Mg.
The 2p3/2-2p1/2 splitting decreases with decreasing Z number. For Si it is ca. 0.6 eV while for Al - ca. 0.4 eV. Mg is next in line so very likely the splitting is less than the resolution of your instrument.
The doublet separation is reported to be 0.28 eV https://srdata.nist.gov/xps/XPSDetailPage.aspx?AllDataNo=33318
If the FWHM of the peaks is >> the separation then as people have said, you may not be able to observe the splitting in the raw data. E.g. for Al 2p, metallic spin-orbit split components may be resolved but oxide peaks (with increased FWHM) appear as a single peak. However, this doesn't mean you should fit as one peak! The physics dictates there are two spin-orbit split peaks in Mg 2p and I'd recommend it good practice to always fit this, you will likely need to constrain a fit to make two peaks (area ratio 1:2 for a p-orbital) split at ~ 0.3 eV.
- Badges
- Science topic
- Similar topics
- Philosophy
- Philosophy Of Science
- Reasoning