Surface inverse photoemission

At a surface, not only can the atomic structure differ from the bulk, but electronic energy levels are present that do not exist in the bulk band structure. These are referred to as surface states . If the states are occupied, they can easily be measured with photoelectron spectroscopy (described in section A 1.7.5.1 and section Bl.25.2). If the states are unoccupied, a teclmique such as inverse photoemission or x-ray absorption is required [22, 23]. Also, note that STM has been used to measure surface states by monitoring the tunnelling current as a fiinction of the bias voltage [24] (see section BT20). This is sometimes called scamiing tuimelling spectroscopy (STS). 

A number of surface-sensitive spectroscopies rely only in part on photons. On the one hand, there are teclmiques where the sample is excited by electromagnetic radiation but where other particles ejected from the sample are used for the characterization of the surface (photons in electrons, ions or neutral atoms or moieties out). These include photoelectron spectroscopies (both x-ray- and UV-based) [89, 9Q and 91], photon stimulated desorption [92], and others. At the other end, a number of methods are based on a particles-in/photons-out set-up. These include inverse photoemission and ion- and electron-stimulated fluorescence [93, M]- All tirese teclmiques are discussed elsewhere in tliis encyclopaedia. 

Other techniques in which incident photons excite the surface to produce detected electrons are also Hsted in Table 1. X-ray photoelectron Spectroscopy (xps), which is also known as electron spectroscopy for chemical analysis (esca), is based on the use of x-rays which stimulate atomic core level electron ejection for elemental composition information. Ultraviolet photoelectron spectroscopy (ups) is similar but uses ultraviolet photons instead of x-rays to probe atomic valence level electrons. Photons are used to stimulate desorption of ions in photon stimulated ion angular distribution (psd). Inverse photoemission (ip) occurs when electrons incident on a surface result in photon emission which is then detected. 

The unoccupied electronic states of a solid can be experimentally explored by different techniques. The most commonly used are inverse photoemission, where low-energy electrons impinge on the surface of the solid, and the photon-based techniques ellipsometry, NEXAFS and constant-initial-state spectroscopy. Results derived from inverse photoemission spectroscopy might be questionable unless low-energy electrons (c. 10-20 eV) and low beam currents are used as in LEED... 

Using inverse photoemission, the unoccupied electronic states of solid surfaces are being studied. Here, instead of injecting an UV light onto the surface and analyzing the emitted electrons, an electron beam is injected onto the surface and the spectrum of the emitted photons is analyzed. Fig. 4.11 shows a summary of the results of photoemission and inverse photoemission of one of the most exhaustively studied surfaces, W(OOl) [Drube et al. (1986)]. As shown, strong surface states immediately below and above the Fermi level are observed. Both are of a character. 

The problem of first-principles calculations of the electronic structure of solid surface is usually formatted as a problem of slabs, that is, consisting of a few layers of atoms. The translational and two-dimensional point group symmetry further reduce the degrees of freedom. Using modern supercomputers, such first-principles calculations for the electronic structure of solid surfaces have produced remarkably reproducible and accurate results as compared with many experimental measurements, especially angle-resolved photoemission and inverse photoemission. 

ARPES) [37,38]. Inverse Photoemission has been developed to detect empty surface states [39]. By now more than thirty surfaces have been shown to sustain surface states, but the surface states of noble metal (111) surfaces [40-42] have become a model system for detailed studies. 

Both photoemission and inverse photoemission require reasonable sample conductivity and their application to hard insulators such as MgO and AI2O3 is problematic. Both techniques also involve the complication that inelastic electron energy loss processes become convoluted with electron emission or decay. This may give rise to spectral features in regions where none are expected fi om the density of states [24,25] and care must always be taken to exclude these features before considering assignment to surface states. 

Like Si( l ll), Ge(l ll) exhibits a complex set of reconstructions depending upon the surface preparation. Ge(lll) also forms a metastable (2x1) phase at low temperatures but there has been no structural determination to date. There is some indirect evidence from total energy calculations (Zhu and Louie, 1.991), angle-resolved pholoemission (ARPES) (Nicholls et al., 1984) and inverse-photoemission (Nicholls and Reihl, 1989), that the (2x1) structure is similar to the 7r-bonded chain model found for Si(lll)(2x 1). 

P.D. Johnson. Inverse Photoemission. In S.D. Kevan, editor, Angle-Resolved Photoemission Theory and Current Applications. Studies in Surface Science and Catalysis, Volume 74. Elsevier, New York, 1992. 

N.V. Smith and D.P. Woodruff. Inverse Photoemission from Metal Surfaces. Prog. Surf. Sci. 21 295 (1986). 

Fig. 3.8 a Photoemission (left) and inverse photoemission (right) spectra of Gd(OOOl) (from Weschke et al. [4]). The occupied part of the surface state appears in the PE spectrum (binding energy 100 meV) while the empty part is weakly visible in the IPE spectrum (binding energy tn —250 meV). b Tunneling spectrum measured on a sample similar to Fig. 3.6b above a Gd(0(X)l) island and above the first distorted ML at 293 K showing both spin parts of the surface state. Spatially resolved data at sample bias values indicated by small arrows c-e will be shown in Fig. 3.10. From [33], copyright 1998, reproduced with permission from World Scientific Publishing Co. Pte. Ltd...

Fig. 5.36 a Tunneling spectra as measured with a Fe covered probe tip above adjacent domains. An asymmetry of the dlIdU signal between the empty and filled part of the surface state can clearly be recognized. In contrast, variations in the dl/dU signal when measured with a pure W tip are always symmetrie inset), b Spin polarization of the tunneling current between an Fe covered probe tip and the Gd(0001) surface at T = 70 K filled circle) compared to spin-polarized inverse photoemission data of GdfOOOl) measured at T = liOK (asterisk) by Donath et al. [103] (reprinted with permission Irom [131]. Copyright 1999, American Institute of Physics)... 

---