New mechanism for electron-stimulated desorption of nonthermal halogen atoms from alkali-halide surfaces

Abstract

Fully angular-resolved kinetic-energy distributions of alkali and halogen atoms emitted due to electron-stimulated desorption have been measured for (100) and (110) alkali-halide surfaces. An unexpected strong directional emission of nonthermal halogen atoms along the (100) axis of the investigated crystals has been found which is in contradiction to the predictions of the previously proposed "Pooley model." We propose a new model involving diffusion of hot holes to the surface, and sudden localization of the excited hole at the surface of the crystal.PACS numbers: 79.20. Kz, 7i.35.+z, 78.55.Fv, 82.65.My Since the early work of Townsend and Kelly [1] and Palmberg and Rhodin [2], it is known that electron irradiation of alkali-halide surfaces results in the efficient desorption of alkali and halogen atoms. Several attempts have been made to construct a model of electron-stimulated desorption (ESD) in alkali halides [3][4][5][6][7][8]. In particular, the "Pooley model" has been applied as an explanation of ESD from alkali halides by Townsend [3,6]. The main idea of this model is that nonradiative decay of the bulk self-trapped exciton followed by a focused replacement sequence along the closely packed (110) chains of the halogen atoms is responsible for halogen desorption. Thus, this bulk-exciton model predicts strong directional emission of halogen atoms along the (110) directions of the crystal. Despite a rather wide acceptance of the Pooley model, we believed that a comprehensive experimental test of its predictions had not yet been performed. Early attempts to measure angular dependence of emitted species were done by means of a collector method [3,6,9]. Unfortunately this technique does not provide any information about either mass or charge of the different desorbing species that arrive at the collected deposit and stick with various probabilities. In addition, resputtering of the collector deposit by reflected primary electrons, strongly focused along surface crystallographic directions [10], is likely to influence such results.The first time-of-flight measurements for ESD of alkali halides were performed by Overeijnder et al. [11] for compressed powder samples. The only published energy distributions for halogen atoms desorbed from singlecrystal material were obtained by Postawa and coworkers [12,13]. In the work we report here, for the first time a direct correlation has been measured among the direction of ejection, the mass and translational energy of a desorbed particle, and the crystallographic structure of a well-characterized insulator surface. In particular, we present a direct comparison between angle-resolved energy spectra of bromine atoms taken for (100) and (110) surfaces of KBr. The data demonstrate that independent of the surface orientation the directional ejection of nonthermal halogen atoms takes place along the (100) axis of the bombarded crystals rather than along the (110) direction as predicted by the Pooley model. Our data strongly indicate the need for a new theoretical approach t...

Role of Excitons in Electron- and Photon-Stimulated Desorption of Neutrals from Alkali Halides

“…Angle resolved ESD measurements of KCl and KBr single crystals demonstrated a strong directional emission of the hyperthermal component along the ͗100͘ axis. To explain this result, the concept of hot hole diffusion in the bulk followed by hole localization at the surface was introduced [10]. Recent experiments on epitaxially grown alkali-halide thin films were explained in terms of hot excitons [11].…”

Dynamics of the defect-mediated desorption of alkali halide surfaces

“…It has been demonstrated in many experiments [20][21][22], that the TOF spectra of halogen atoms consist of two components: the thermal (which can be fitted with a Maxwellian distribution function) and the non-thermal one. The non-thermal peak, at energy of about 0.25 eV, corresponds to the halogen atoms emitted as a result of the decay of self-trapped excitons at the surface.…”

ESD thresholds for KCl

“…The details of this mechanism are still debated. It is commonly believed that emission of atoms is due to F and H centre formation, its subsequent diffusion to the surface and annihilation followed by thermal evaporation [8,17]. Desorption events are belived to take place most likely at step edges and corners because the amount of energy gained from defect annihilation is rather small.…”

“…However, structure of the yields measured for alkali halides cannot be explained completely by the Auger decay model. Furthermore, recent experiments have shown that the ESD could involve surface exciton decay as well [7][8][9]. Therefore for the unambiguous identification of the processes involved precise measurements of threshold energies are required.…”