Pb 4f photoelectron spectroscopy on mass-selected anionic lead clusters at FLASH

Abstract

4f core level photoelectron spectroscopy has been performed on negatively charged lead clusters, in the size range of 10-90 atoms. We deploy 4.7 nm radiation from the free-electron laser FLASH, yielding sufficiently high photon flux to investigate mass-selected systems in a beam. A new photoelectron detection system based on a hemispherical spectrometer and a time-resolving delayline detector makes it possible to assign electron signals to each micropulse of FLASH. The resulting 4f binding energies show good agreement with the metallic sphere model, giving evidence for a fast screening of the 4f core holes. By comparing the present work with previous 5d and valence region data, 2 the paper presents a comprehensive overview of the energetics of lead clusters, from atoms to bulk. Special care is taken to discuss the differences of the valenceand core-level anion cluster photoionizations. Whereas in the valence case the escaping photoelectron interacts with a neutral system near its ground state, core-level ionization leads to transiently highly excited neutral clusters. Thus, the photoelectron signal might carry information on the relaxation dynamics.

Modeling the core-hole screening in jellium clusters using density functional theory

“…The experimental results for the core-level binding energies as a function of the cluster size [16,17] follow the metal-sphere prediction down to a certain minimum cluster size and then deviate, ultimately approaching the single atom or ion result, which in the experiment with Pb − clusters was below the metal-sphere binding energy for the 5d and 4f core-shells (unlike for the 2p shell in Na above). A too rapid escape of the photoelectron would lead to an increased binding energy.…”

“…Option (ii) would manifest itself as an increased binding energy of the photoelectron as compared to the situation where all the relaxation energy goes to the photoelectron. Instead, the experimentally measured binding energies in [16,17] drop below the metal-sphere result, which ultimately must be so because the single-ion value (Pb − in [16,17]) lies below the metal-sphere line. Hence, one has to distinguish (at least) two effects here: the ability of the other electrons to screen and the ability of the photoelectron to pick up the relaxation energy due to screening.…”

“…Signatures in XPS spectra that would support option (iii) were not observed in experiments [16,17]. Instead, slow 'secondary' electrons are measured, which are likely generated via Auger decay of the core-hole after the photoelectron is already gone.…”

“…Hence, one has to distinguish (at least) two effects here: the ability of the other electrons to screen and the ability of the photoelectron to pick up the relaxation energy due to screening. The first one seems to be relevant in the experiments [16,17] so that indeed the metal-to-nonmetal transition in Pb − A around A = 20 could be probed. The situation for initially neutral Na clusters is quantitatively different.…”

“…Time-dependent DFT (TDDFT) (see [15] for a state-of-the-art account) is used to investigate the screening dynamics after the sudden removal of a 2p electron. We apply our model in this first study to Na clusters (instead of Pb − clusters in the experiments [16,17]) in order to keep the number of inner electrons in the embedded atom and thus the numerical effort manageable.…”