Magnesium 2p and Auger Parameter Values

Magnesium 2p binding energy and modified Auger parameter values are shown in Table 1. It is worth noting that accounting for the difference between charge referencing procedures is vital for correct analysis of magnesium compounds (especially for MgO and Mg(OH)2) [1]. For MgO, the peak position for the Mg 2p transition is 49.4 eV when charge referenced to adventitious carbon at 284.8 eV, and 50.8 when referenced to the Mg 2p metal peak at 49.73 eV (or grounded). This has led to a lot of confusion regarding the Mg 2p peak positions to use for analysis. The results from a consistently analyzed dataset are shown in Table 1 together with compiled literature values [1]. Since the chemical sensitivity of the Mg 2p transition is low for magnesium the Auger parameter and anion signals are particularly important to consider for improved speciation.

Table 1: Experimental and literature values for the Mg 2p binding energy and the modified Auger parameter for different magnesium compounds. Subscripts and superscripts: “sc” denotes single crystal; “g” denotes grounded and no charge correction was performed; “Mg” denotes floated and the Mg 2p metal peak was used as a charge reference at 49.73 eV; “O 1s” denotes floated and the O 1s lattice peak of MgO or Mg(OH)2 was used as a charge reference at 531.1 eV and 532.7 eV, respectively. Samples with no superscript were floated, and adventitious carbon was used as a charge reference at 284.8 eV. 

The anion fitting parameters—binding energy, peak width, and line shape—obtained from fitting reference samples are reported in Table 2 from a consistently analyzed data set [1]. The O 1s signal can be used to separate MgO and Mg(OH)2 despite their overlapping Mg 2p signals. The O 1s peak position of Mg(OH)2 is +1.6 eV with respect to the main lattice peak of MgO. However, the peak position for Mg(OH)2 can overlap with other environments such as the MgO defective oxide and adventitious carbon, which may need to be accounted for in the analysis.

Table 2: Overview of anion fitting parameters, binding energy, peak width, and line shape, from reference data. Subscripts and superscripts: “sc” denotes single crystal;  “g” denotes grounded and no charge correction was performed; “Mg” denotes floated and the Mg 2p metal peak was used as a charge reference at 49.73 eV; “ref” denotes reference values when charge referencing to the O 1s lattice peak. Samples with no superscript were floated, and adventitious carbon was used as a charge reference at 284.8 eV. 

Reference:

[1] S.A. Skaanvik, J.D. Henderson, J.J Noël, and M. C. Biesinger, Surface and Interface Analysis, 57 (2025) 717-728.

Magnesium Induced Ghost Peaks

Ghost peaks occur when internal X-rays are produced within the sample, ejecting a detectable number of core-level photoelectrons. Since these internally produced X-rays have lower energy than the source X-rays, they eject photoelectrons with a lower kinetic energy (higher apparent binding energy). While ghost peaks in many cases appear with low intensity and therefore do not generally interfere with XPS analysis, they may cause confusion during peak assignment.

Ghost peaks are often observed for magnesium-rich samples from the Mg Kα emission lines (X-rays produced when a valence electron fills the Mg 1s core hole). The average photon energy of these lines is 1253.6 eV, and the position of the ghost peaks on the binding energy scale is dependent on the X-ray source. For example, the Mg Kα induced O 1s ghost peak is seen in the XPS spectra of Mg(OH)2 at 765.7 eV, when using an Al Kα source (1486.7 eV). This peak position is shifted by 233.1 eV, which is the energy difference between internal and source X-rays (Figure 1). 

Alternatively, this can also be calculated as follows: the Mg K X-ray induced peak for O 1s is found at a binding energy of ~ 765.7 eV, as the kinetic energy of these photoelectrons would be 1253.6 eV (Mg Kα X-rays) - 532.6 eV (O 1s B.E.) = 721.0 eV. Thus, they would then be seen as a peak at a binding energy of 1486.7 ( Al K(α) X-rays) - 721.0 eV = 765.7 eV. Similarly, the C 1s Mg X-ray induced peak would be at a binding energy of ~517 eV.

Also of note for spectra of magnesium-rich samples: Mg Auger structure can be found at ~242 eV, which is not a ghost peak and often not noted in most libraries (Figure 1, Mg KLL-2) and may cause confusion during survey scan peak assignment.  

Figure 1. Survey spectrum of Mg(OH)2 powder where the O 1s ghost peak is clearly visible at 765.7 eV. The figure is reproduced from reference [1]. 

Reference:

[1] S.A. Skaanvik, J.D. Henderson, J.J Noël, and M.C. Biesinger, Surface and Interface Analysis, 57 (2025) 717-728.

R.S.F. for Magnesium Auger Line

The CasaXPS/Kratos R.S.F. value (of 0.6) for the Mg KLL Auger line tends to significantly overestimate the amount of magnesium present compared to the quantitative values found from the Mg 2p and Mg 2s lines. A new R.S.F. value calculated from standard samples (magnesium metal and magnesium oxide) has been developed for use on samples where the Mg 2p or Mg 2s peaks are either too small or overlapped with other species. A value of 4.3 +/- 0.3 was found to work quite well. Use a background encompassing the main peak of the Mg KLL line extending from approximately 315 eV to 298-300 eV. Interference from strong C 1s loss structure can be mitigated by shifting the lower binding energy background endpoint to a slightly higher binding energy.

Mg KLL Auger R.S.F. = 4.3 +/- 0.3

Magnesium reference spectra used for this calculation and showing the full Mg Auger peak-set are presented in the link.