Quantification of Ni $L_{2,3}$ core-hole relaxation pathways utilizing Auger photoelectron coincidence spectroscopy

Ni $LVV$ Auger spectra, in coincidence with the corresponding $2p_{1/2}$, $2p_{3/2}$, and $6\text{eV}$ satellite photoelectrons, have been used to examine electron correlation and itinerance effects in Ni. In coincidence with the $2p_{3/2}$ core level, the Auger spectral shape is represented by localized $3d^8$ and itinerant valence final states with an additional $3d^7$ Auger shake-up contribution. The spectra in coincidence with the $6\mathrm{eV}$ satellite probe the decay of localized $2p^{5}3d^9$ double hole states, leading to $3d^7$ final states. It is found that a fraction of the double hole states delocalize before the Auger decay. A similar delocalization is observed for the double hole states produced by the $L_2{L}_3{M}_{45}$ Coster-Kronig process, and the delocalization rates have been determined.

Figure 1

(a) The map depicts Ni $2p$ photoelectrons in coincidence with $LVV$ Auger electrons after the subtraction of the accidental coincidence counts. The dashed lines connect points with a constant sum of Auger and photoelectron kinetic energies (note that a binding energy scale is used for the photoelectrons in the figure). Features A and B correspond to the normal Auger process. Feature C is associated with the Auger decay connected to the $6\text{eV}$ photoelectron satellite. Feature D corresponds to the Auger decay following an $L_2{L}_{3}V$ CK process. The side plots display the Auger spectrum (b) and the photoelectron spectrum (c) obtained by integration of the map. The singles spectra are shown as dashed lines. The spectra are normalized to the background intensity at high kinetic energies. This normalization is used to show the improved signal to background ratio in the coincidence data. The colored regions indicate the three regions chosen for closer analysis.

Figure 2

(a) $LVV$ Auger electron spectrum measured in coincidence with $2p_{3/2}$ photoelectrons. The fit is based on atomic multiplet calculations and includes the $3d^8$ (gray) and $3d^7$ (blue) final states for the atomic fraction. The bandlike contribution (yellow) is represented by a self-convoluted valence band. (b) and (c) show spectra measured in coincidence with the $6\mathrm{eV}$ satellite and the $2p_{1/2}$ photoelectrons, respectively. The fits consist of the atomic parts of the fit in panel (a) (gray), the bandlike contribution (yellow), the Auger vacancy satellite (red), and an additional $3d^7$ contribution (blue).

Figure 3

The photoelectron spectrum in coincidence with $LVV$ Auger electrons are shown in black. The Shirley background is based on the data that is shown in cyan. Blue area under the 6 eV satellite ($58\%$) depicts the maximum background originating from the $L_3$ photoelectrons causing a $L_{3}VV$-like fraction in the Auger spectrum in coincidence with the $6\mathrm{eV}$ satellite. The green area shows the intrinsic electrons, which cannot be explained by the background. The cyan area under the $L_2$ edge shows the maximum background originating from the $L_3$ and from the $6\mathrm{eV}$ satellite ($65\%$). Consequently the area shown in red is the intrinsic signal.

Figure 4

Background corrected $LVV$ Auger spectrum in coincidence with the photoelectrons between the $L_2$ and the $6\mathrm{eV}$ satellite (862–$866\mathrm{eV}$ binding energy). In black the full fit is shown. The gray area corresponds to the atomic $L_{3}VV$ final states, the orange to the bandlike final states. In red the Auger vacancy satellite is shown. The blue area corresponds to the $3d^7$ final states.