Abstract
The data for the Si(111)7 surface obtained by ion-neutralization spectroscopy (INS) are found to be anomalous among those for a group of eight surfaces on six crystallographic faces of Si and Ge. This anomaly resides in the fact that it is necessary to shift the origin of the "fold" function, obtained during INS-data reduction, by 0.55 eV (of the unfold by 1.1 eV) before it is possible to obtain the correct physical unfold. Study of this phenomenon has led to a more general understanding of the folding, nonfolding, and cross-folding characteristics of the electron-energy distributions which a solid, such as a semiconductor, can present to an incident ion in the Auger-neutralization process. A folding electron distribution can provide two electrons within the initial-state lifetime (1014 sec) of the Auger process and a nonfolding distribution cannot. It is found that a folding distribution and a nonfolding distribution at an electronically homogeneous surface will produce, under certain circumstances, INS data which cannot be unfolded. Two folding distributions, if separated spatially as at an electronically inhomogeneous surface, can yield a function derived by INS methods from the observed kinetic energy distributions which also cannot be unfolded. Evidence is presented that the INS "fold" function derived for Si(111)7 is not unfoldable because there is an electronic superlattice at this surface which it is not unreasonable to take to be coincident with, and most likely produced by, the 7 × 7 atomic-structure superlattice which low-energy-electron diffraction shows to be present.