$M_3{M}_{45}{M}_{45}$ Auger lineshape measured from the Cu(111) surface: Multiplet term selectivity in angle-resolved Auger-photoelectron coincidence spectroscopy

The capability of the recently observed dichroic effect in angle-resolved Auger-photoelectron coincidence spectroscopy (DEAR-APECS) to disentangle individual multiplet terms has been exploited to study the lineshape of the $M_3{M}_{45}{M}_{45}$ Auger spectrum measured in coincidence with the $3p_{3/2}$ photoelectrons from the Cu(111) surface. The relevant multiplet structure of the two hole final state is determined with an unprecedented sensitivity, including a reliable experimental estimation of the energy of the ${}^{1}D$ multiplet term. Spectroscopic data for the $3p$ photoemission feature are also given and energy conservation applied to the photoelectron-Auger-electron pair has been successfully used in order to quantitatively explain energy shifts in coincidence spectra. Multiple-scattering calculations prove that the DEAR-APECS effect is not destroyed by diffraction effects and a simple model which combines atomic angular distributions and electron-diffraction modulations is provided in order to obtain a detailed understanding of the multiplet energy and intensity distributions in Auger spectra.

Figure 1

(a) $\text{Cu}3p$ photoemission spectrum (open circles). The solid line is the result of the best fit of the experimental spectrum obtained by using a Shirley-type integral background (dash-dotted line) and two Lorentzian functions (dashed lines) representing the two SO split components which have been convoluted by the experimental Gaussian broadening. (b) Single Cu $M_{23}{M}_{45}{M}_{45}$ Auger spectrum (dots with error bars) and best fit (solid line) obtained by the convolution of the experimental Gaussian broadening with four Lorentzian functions representing the multiplet contributions of the $M_3$ decay (solid line) and the $M_2$ decay (dashed line). Dash-dotted line is the background. Residuals relative to the error bars $\sigma$ of each data point are shown in the top part of both panels.

Figure 2

Cu $M_3{M}_{45}{M}_{45}$ coincidence Auger spectra measured in configurations (a) $AA$ and (b) $AN$ and (c) their sum are reported as open circles with statistical error bars. The solid line is the best fit obtained by using the multiplet terms (dashed lines) describing the Auger transition (see text for more). Residuals relative to the error bars $\sigma$ of each data point are shown in the top part of each panel.

Figure 3

Energy shifts occurring in APECS due to the energy conservation requirement. (a) Comparison between the Auger lineshape of the total coincidence spectrum (open rhombuses and fitting line) and the simultaneously acquired single Auger spectrum (dashed line). A shift in energy of about $-0.5\text{eV}$ between coincidence and single spectra is outlined. (b) Photoemission spectrum (open circles) superimposed with the best fit (thin solid line) with two Lorentzian components (thin dashed lines) and a Shirley-type integral background (dash-dotted lines). By multiplying each of the two Lorentzian peaks by the analyzer transmission curve (thick dashed line), one obtains the curves representing the accepted $3p_{3/2}$ photoelectrons (thick line) and the negligible amount of accepted $3p_{1/2}$ photoelectrons (thick dotted line). The former one results shifted by about $+0.5\text{eV}$ above the corresponding Lorentzian peak.

Figure 4

$m$ resolved electron-diffraction simulations with the MSCD code. The photoelectron polar distribution is shown in (a) at an azimuthal angle corresponding to the $⟨\overline{1}\overline{1}2⟩$ direction. Two Auger-electron polar distributions are shown in (b) and (c) along the $⟨\overline{1}\overline{1}2⟩$ and $⟨\overline{1}10⟩$ azimuthal directions, respectively. The total intensity (thick solid line) and the partial intensities for waves having $m=0$ (open circles and line), $m=-1$ (open down triangles and line), $m=1$ (open up triangles and line), $m=-2$ (minus symbols and line), and $m=2$ (plus symbols and line) are plotted as a function of the polar emission angle. Solid and dashed arrows indicate the positions of the analyzers in configurations $AA$ and $AN$, respectively.