The measurements were performed with a Versaprobe spectrometer from Physical Electronics, using monochromatized AlK alpha radiation at 1486.6 eV and a pass energy of 23.5 eV. This results in a full width at half maximum (FWHM) of 0.58 eV for the Ag3d5/2 core level of a clean Ag sample. No in-situ cleaning was performed prior to the XPS measurement. Charge neutralization was done by a simultaneous irradiation with electrons and low energy (20 eV) ion beams. The measurement spot had a diameter of 200 µm. The take-off angle was 45°. After a first measurement, in-situ bombardment with Ne ions with 4 keV was additionally performed in the XPS chamber, in order to induce structural changes in the films. Mathematical analysis of the Ar2p core level peaks was performed with the UNIFIT software, using a convolution of Lorentzian and Gaussian line profiles. The doublet separation between the Ar2p3/2 and Ar2p1/2 was fixed to 2.1 eV. The width of the Lorentzian component, which is associated to the lifetime of the photohole, was fixed to a value of 0.12 eV. The Gaussian widths of both doublet components (2p3/2 and 2p1/2) were the same. The BE scale was fixed by placing the C1s peak of adventitious carbon at 284.7 eV. This resulted in a BE mean value of 74.0 plus/minus 0.3 eV for the position of the Al2p3/2 core level in all the magnetron-deposited Al2O3 films.

Figure 3 (a) shows sample 30 (a.i. as introduced and after sputtering), (b) sample 34 (a.i. as introduced and after sputtering) and (c) sample 38 (a.i. as introduced and after sputtering)