Photoemission, core levels

The calculation of the core levels of LajCuO discussed above was performed within the cluster approximation wherein the dependence of the spectral shape on the oxygen band-width is missed out. In order to include the effect of the oxygen 2p derived band on the spectral shape, the core-level photoemission calculation has to be performed within the impurity model which includes the O 2p band-width (instead of a single O 2p level). For reasonable values of the hybridization strength, t, the range of values of A is such that the Cu appears close to the bottom of the... 

Hollinger, G. and Himpsel, F. J. Probing the transition layer at the Si02-Si interface using core level photoemission. Applied Physics Letters 44, 93-5 (1984). 

From the perspective of this symposium, analysis of the atomic dynamics and electronic structure of surfaces constitutes an even more exotic topic than surface atomic geometry. In both cases attention has been focused on a small number of model systems, e.g., single crystal transition metal and semiconductor surfaces, using rather specialized experimental facilities. General reviews have appeared for both atomic surface dynamics (21) and spectroscopic measurements of the electronic structure of single-crystal surfaces (, 22). An important emerging trend in the latter area is the use of synchrotron radiation for studying surface electronic structure via photoemission spectroscopy ( 23) Moreover, the use of the very intense synchrotron radiation sources also will enable major improvements in the application of core-level photoemission for surface chemical analysis (13). 

Valence and core-level photoemission experiments were carried out in the IFW Dresden under the following conditions. Unless otherwise stated, the valence band photoemission data were recorded at room temperature using He Ia radiation (21.22 eV) with a total energy resolution of 100 meV, and the core-level photoemission data, using monochromatised Al Ka radiation (1486.6 eV) with a total energy resolution of 0.4 eV. All such data are angle-integrated in nature ( 4° or more). 

In this section, we will present and discuss results from Sc2 C84, which is the most widely studied dimetallofullerene to date. Early scanning tunnelling microscopy [26] and transmission electron microscopic [27] investigations provided evidence in favour of the endohedral structure of this system, which was later confirmed by x-ray diffraction experiments utilising maximum entropy methods [28]. Before experimental data from this system were available, the Sc ions were predicted to be divalent from quantum chemical calculations [29]. Subsequent data from vibrational spectroscopy [30,31], core-level photoemission [32] and further theory [33] on this system were indeed interpreted in terms of divalent Sc ions. 

In the lower panel of Fig. 10 we show the Sc 2p core-level photoemission spectrum of a UHV-prepared film of Sc2 C84 [34]. Analogous data, but from films prepared by dropping from a solvent solution in air, were used in [32] to argue that, as the binding energy of the Sc 2p spin-orbit split components is lower than in Sc203 (a trivalent Sc standard), the Sc ions in the endohedral fullerene are divalent. 

Fig. 10 Core-level photoemission spectra of (top) Sc3N C80 [36] and (bottom) Sc2 C84 [34] recorded using Al Ka radiation...

Aware of the fact that the absolute binding energy in core-level photoemission of transition-metal systems can be strongly influenced by screening and charge-transfer effects in the photoemission final state, we turned to x-ray absorption as a highly direct probe of the valency of transition-metal and rare-earth systems [35]. 

Fig. 16 K 2p core level photoemission spectra for the system Kx(Sc3N C80) as a function of increasing K intercalation level, x. The labels T and 0 refer to signals originating from K ions situated in interstitial sites of a tetrahedral or octahedral type, respectively...

Lin DS, Ku TS, Chen RP (2000) Interaction of phosphine with Si(100) from core-level photoemission and real-time scanning tunneling microscopy, Phys. Rev. B 61 2799-2805... 

De Nadai C, van der Laan G, Dhesi SS et al (2003) Spin-polarized magnetic circular dichroism in Ni 2p core-level photoemission. Phys Rev B 68 212401... 

A systematic semiempirical study of the core-level photoemission spectra of a wide range of 3d transition-metal compounds has been carried out (Bocquet et al., 1992,1996). The values for U and A obtained from a simplified Cl cluster model analysis are demonstrated in Figure 7.2. As can be inferred from the graphs, the heavier 3d transition metal compounds shown in the figure are expected to be charge-transfer insulators, whereas the compounds of the fighter metals are generally expected to be of the Mott-Hubbard type. 

As discussed earlier, it is now possible to make and study deposits of monosized, highly dispersed, transition metal clusters.(S) In this section we summarize results from the first measurements of the valence and core level photoemission spectra of mass selected, monodispersed platinum clusters. The samples are prepared by depositing single size clusters either on amorphous carbon or upon the natural silica layer of a silicon wafer. We allow the deposition to proceed until about 10 per cent of the surface in a 0.25 cm2 area is covered. For samples consisting of the platinum atom through the six atom duster, we have measured the evolution of the individual valence band electronic structure and the Pt 4f... 

Since the bulk of the studies reviewed here were carried out in the authors laboratories using LEED and core-level photoemission spectroscopy (CLS), we preface our discussion with short accounts of the LEED and CLS methods in our implementation. These are illustrated by brief accounts of their application to the clean Al(l 11), (100), and (110) surfaces, which serve to define the starting points for the studies of alkali metal adsorption on these surfaces. 

The use of core-level photoemission spectroscopy for investigating the geometrical structure of overlayers and surface alloys is based on the fact that... 

The final characteristics of the interface can continue to be driven as additional monolayers are added. Strain fields due to lattice mismatch and grain boundary phenomena can play an important role. Usually, the interface reaches a stable configuration within 3 to 10 monolayers for most metals on semiconductors. Interfaces are often metastable, however, and the interdiffusion is accelerated at increased temperature or under biased conditions. The application of core level photoemission to this problem confirmed what had already been deduced with Auger spectroscopy, but on a finer dimensional scale. An overriding fact of metal-semiconductor interfaces is the intrinsic aspects of interfadal interdiffusion and interaction. These two aspects of the chemistry often tend to overwhelm issues that might be deduced from the clean separate surfaces. The interdiffusion can result in interface layers with radically different characteristics from the supposed components. 

---