Paper NM-TuP20
Decay Processes of Si 2s Core Holes in Si(111)-7×7 Revealed by Si Auger Electron Si 2s Photoelectron Coincidence Measurements

Tuesday, December 9, 2014, 4:00 pm, Room Mauka

The filling of an initial core hole by an electron from a higher subshell of the same shell is known as a Coster–Kronig transition, and Auger photoelectron coincidence spectroscopy (APECS), in which Auger electrons are measured in coincidence with photoelectrons with a fixed kinetic energy (KE), is an ideal tool for probing Coster–Kronig transitions because Auger electrons originating from a specific core ionization are detected .

However, studies of Coster–Kronig transitions have mainly been restricted to metal surfaces, even though Si surfaces are crucial for surface science applications and the semiconductor industry. Issues such as the assignments of the Si L₁VV Auger peaks, experimental determination of the branching ratio of the Si L₁L₂₃V and Si L₁VV Auger decays, and competition between the delocalization of the valence hole and the Si L₂₃V-VVV Auger decay are, to the best of our knowledge, largely unexplored. In this paper, we report on the decay processes of Si 2s core holes in a clean Si(111)-7×7 surface studied using coincidence measurements of the Si Auger electron and the Si 2s photoelectron.

We measured a coincidence spectrum measured by scanning the ASMA over a KE range of 20­–150 eV, with the DP-CMA fixed at a KE of 76.5 eV (Si-L₂₃VV-Si-2p APECS spectrum). Distinct Auger peaks are observed in a KE region of 50–92 eV with the maximum peak located at KE = 88 eV. We assigned these peaks to Si L₂₃VV Auger electrons emitted in the decay processes of Si 2p holes. Another coincidence spectrum was measured by scanning the ASMA over a KE range of 20­–150 eV with the DP-CMA fixed at KE = 26.3 eV, which corresponds to Si 2s photoelectrons (Si-Auger-Si-2s APECS spectrum). The maximum peak position and shape of the Si-Auger-Si-2s APECS spectrum in the AeKE = 50–92 eV region are almost identical to those of the Si-L₂₃VV-Si-2p APECS. Therefore, we assigned these Auger peaks to Si L₂₃VV Auger processes. This is direct evidence of Si L₂₃VV Auger processes being induced by Si 2s ionization.

The peaks of Si-Auger-Si-2s APECS in the AeKE = 20–50 regions were assigned to Si L₁L₂₃V Auger decays, while the peaks of Si-Auger-Si-2s APECS in the AeKE = 100–150 eV regions were assigned to Si L₁VV Auger decays. These results indicate that there are two nonradiative decay processes of the Si 2s core hole. The first is the Si L₁L₂₃V Coster–Kronig transition followed by delocalization of the valence hole and Si L₂₃VV Auger decay, while the second is Si L₁VV Auger decay. From the integrated intensity of the normalized Si L₂₃VV and Si L₁VV Auger peaks, we estimated the branching ratio of Si L₁L₂₃VV to Si L₁VV Auger processes to be (96.8 ± 0.4):(3.2 ± 0.4).