Erratum: Unusual secondary-electron-emission spectra: Surface-exciton sources for Si(111)7 x 7

Abstract

In many respects electron-beam excited secondary-electron-emission spectra (SEES) from the (111)7\ifmmode\times\else\texttimes\fi{}7 surface of 10-\ensuremath{\Omega} cm n-type silicon are like those from metals: The maximum in the secondary current is observed within a few eV of the threshold, and, with negative potential (with respect to chamber ground) applied to the sample, ${\mathit{V}}_{\mathit{s}}$, the spectra retain their shape while their detected kinetic energy increases by ${\mathit{eV}}_{\mathit{s}}$. The shapes of SEES from 10-\ensuremath{\Omega} cm p-type samples are different in a gross qualitative way: Secondary currents for kinetic energies out to 8 eV above threshold are greatly reduced, with the maximum in the distribution occurring at a kinetic energy of about 12 eV above threshold. With negative potential applied to the sample the shape of the spectra change, until at a value of ${\mathit{V}}_{\mathit{s}}$ of about -12 V the shape observed for the n-type samples is recovered. With more negative values of ${\mathit{V}}_{\mathit{s}}$ this part of the SEES retains its shape, while its detected kinetic energy increases by ${\mathit{eV}}_{\mathit{s}}$, again just like those for metals. In addition to these electrons, for -12${\mathit{V}}_{\mathit{s}}$-300 V another group of secondary electrons is observed to emerge from the sample, but from outside the beam landing area, and with energies close to chamber ground.The observation of secondary-electron emission at two distinct energies indicates that under these conditions this silicon surface is a nonequipotential surface. A model for all of these observations is presented in terms of mobile surface excitons, whose existence was previously demonstrated from angle-resolved energy-loss spectra. These long-lived excitons consist of a selvedge electron bound to a surface-state hole; the selvedge region lying between the bulk crystal and the image potential. A mode of formation of these surface excitons is postulated; electrons impinging on the surface from within the crystal, with energies greater than 1.5 eV above the bottom of the conduction band, can be captured by surface holes to form surface excitons. The unusual SEES is associated with the surface being p type. Bombardment of the (111)7\ifmmode\times\else\texttimes\fi{}7 surface of p-type silicon by electrons with energies between 40 and 300 eV is an effective way of producing long-lived surface excitons. It is suggested that this technique constitutes a source which can produce surface excitons for study by other probes. There is evidence for the formation of surface excitons in scanning tunneling, and in inverse photoemission spectra.