Fineet al.Reply:

Abstract

Fine et al. Reply: In our Letter [1], we reported on the formation of three autoionizing states of neutral, 2p core-excited Na ‫ء0‬ 2p 5 3s3l ͑l s, p, d͒ observed in the electron spectra of ion-bombarded sodium halide surfaces. We proposed that the formation of these Na ‫ء0‬ states, due to electron capture by collisionally excited moving excitons Na ‫ء1‬ 2p 5 3s, 3p, 3d [2], involves Landau or Zener-type resonant electron transfer in Na ‫ء1‬ collisions with lattice halogen ions. Evidence for such a mechanism comes from measurements obtained on slightly halogen-depleted NaCl surfaces (Fig. 1) which show that near surface Cl 2 is involved in the neutralization of the above excitonic states [1]. Such moving Na ‫ء0‬ atoms can, clearly, autoionize inside the solid or, if they are sputtered, outside.In the preceding Comment [3] and in [4], Baragiola acknowledges that the formation of Na ‫ء0‬ does involve electron capture by Na ‫ء1‬ but disputes our collisional electron-capture model. He suggests that the neutralization process involves Na ‫ء1‬ and weakly bound surface electrons and explicitly rules out any electron transfer from halogen ions. We believe that his neutralization concept [3,4] is not consistent with our experimental data [1] for the following two reasons.(1) The electron signals of the three Na ‫ء0‬ transitions are very intense with an estimated yield for NaCl of 2 3 10 23 electrons/projectile ion [5]. This very high yield is consistent with the molecular dynamics (MD) simulation of the Na ‫ء1‬ excitation or neutralization (by electron-transfer collisions with Cl 2 ) process. The MD yield is 10 24 [5]. It is difficult to imagine that such a high

Bombardment induced electron-capture processes at sodium halide surfaces

“…This situation is quite different in wide-bandgap ionic solids [ 11 , 18 , 19 , 41 ] where, because of the highly localized nature of the valence electrons, there are essentially no conduction-band electrons. It is then possible to obtain discrete deexcitation transitions from a wider range of excited levels [ 41 ] and, because of the absence of conduction-band electrons, deexcitation transitions that involve localized electron capture also can occur.…”

“…The concept of localized collisional deexcitation has not been specifically addressed in descriptions of inelastic collision processes in solids. Our recent measurements [ 18 , 19 ] of the nonradiative electron deexcitation spectra at ion-bombarded surfaces of sodium halide crystals can be interpreted in terms of a new collisional deexcitation model which we propose and in which electron capture plays a critical role. These spectra, consisting of three characteristic sodium lines in the 25 eV to 35 eV region, are quite different from the single, intense ion-induced transition which is typically observed at 26 eV on metallic sodium [ 20 – 25 ].…”