Final-state Screening Effects Research Articles

Hard x-ray photoelectron spectroscopy as a probe of the intrinsic electronic properties of CdO

Hard x-ray photoelectron spectroscopy (HAXPES) is used to investigate the intrinsic electronic properties of single crystal epitaxial CdO(100) thin films grown by metal organic vapor phase epitaxy (MOVPE). The reduced surface sensitivity of the HAXPES technique relaxes stringent surface preparation requirements, thereby allowing the measurement of as-grown samples with intrinsically higher carrier concentration ($n=2.4\ifmmode\times\else\texttimes\fi{}{10}^{20}\phantom{\rule{0.222222em}{0ex}}{\text{cm}}^{\ensuremath{-}3}$). High-resolution HAXPES spectra of the valence band and core levels measured at photon energy of 6054 eV are presented. The effects of conduction band filling and band gap renormalization are discussed to explain the observed binding energy shifts. The measured bandwidth of the partially occupied conduction band feature indicates that a plasmon contribution may be present at higher carrier concentrations. The Cd 3$d$${}_{5/2}$ and O 1$s$ core-level line shapes are found to exhibit an increased asymmetry with increased carrier concentration, interpreted as evidence for final state screening effects from the carriers in the conduction band. Alternatively the core-level line shape is interpreted as arising from strong conduction electron plasmon satellites. The nature of these two competing models to describe core-level line shapes in metallic oxides is explored.

Photoemission and density functional theory study of Ge(100): Clean surface and Yb-induced (2×4) reconstruction

Clean and metal-adsorbed (100) surfaces of group-IV semiconductors, such as Si and Ge, often exhibit electronically and structurally similar reconstructions. However, the fundamental bulk properties of group-IV materials can have an impact on particular features of such systems, which are related, e.g., to final-state relaxation in photoemission and thus determine their spectral line shape. Here we have studied Yb/Ge(100)(2×4) reconstruction as well as clean Ge(100) surface by high-resolution photoelectron spectroscopy and ab initio calculations. An atomic geometry of both surfaces is thoroughly investigated. A detailed analysis of Ge 3d core-level photoemission, atomic origins of surface-shifted components, and final-state screening effects is presented. In particular, it is demonstrated that the core-hole screening plays an essential role in Ge 3d measurements, and that its amount in the complete screening model correlates well with the core-level binding energy of respective Ge atoms in the initial state. The results are discussed in the proper context of related reconstructions on Si(100).

Energy level alignment of cyclohexane on Rh(111) surfaces: The importance of interfacial dipole and final-state screening

Adsorption states and electronic structure of cyclohexane on clean and hydrogen-saturated Rh(111) surfaces were investigated by scanning tunneling microscopy and photoelectron spectroscopy. Monolayer cyclohexane molecules form an ordered superstructure on the clean Rh(111) surface. The energy level alignment of adsorbed cyclohexane depends on each adsorption site; molecular orbitals of adsorbed cyclohexane on the atop site have lower binding energies than those on the other sites. In contrast, it becomes insensitive to adsorption sites on the hydrogen-saturated Rh(111) surface. By preadsorption of hydrogen, all cyclohexane molecular orbitals are uniformly shifted to lower binding energy compared to those on the clean Rh(111) surface. The observed energy level alignment of cyclohexane on the Rh(111) surfaces is determined by the vacuum level shift and the final-state screening effects.

Theoretical and Experimental Study of the Electronic Structures of MoO3and MoO2

The geometric and electronic structures of MoO3 and MoO2 have been calculated using the generalized gradient approximation to density functional theory. The calculated cross-section weighted densities of states are compared with high-resolution X-ray photoemission spectra. There is very good agreement between the calculated structures and those determined previously by X-ray diffraction and between the computed densities of states and the present photoemission measurements. MoO2 is shown to be a metallic material, as is found experimentally, but the Fermi level sits in a distinct trough in the density of states. Satellite peaks found in core photoemission spectra of MoO2 are shown to derive from final state screening effects in this narrow band metallic material.

Final-state screening dynamics in resonant Auger decay at the2pedge of vanadium

We investigated the resonant Auger process near the $\mathrm{V}$ $2{p}_{3∕2}$ edge in vanadium metal. Attention is centered on the onset of Auger decays and their behavior below the $2{p}_{3∕2}$ resonance. The $2{p}_{3∕2}3d3d$ decay has a crossover from the Raman-Auger to the normal Auger regime at the $2p$ ionization threshold. Meanwhile, Auger decays with core holes in the final state have normal Auger behavior even below the ionization threshold, the $2{p}_{3∕2}3p3p$ process being visible at 2.2 eV lower photon energy. The different resonant behavior of these Auger decays can be understood within the one-step model as final-state screening effects affecting the photoexcitation.

Satellite structure in the photoelectron spectra of transition-metal compounds

Abstract The electronic structure of high-spin CrCl3 and CrBr3 compounds has been investigated by means of X-ray photoelectron spectroscopy. Satellite features have been observed in the valence-band region and in Cr2p, Cr3s and Cr3p core-level spectra. Photoemission spectra have been analysed in terms of a simple cluster model, leading to estimates of the Coulomb correlation energy, charge transfer energy and hybridization strength. Satellites can be explained in general by a charge-transfer process, if the effective hybridization energy T eff (3.6-3.7 eV) is large enough. However, in the case of Cr3s core-level spectra, the main emission is assigned to unscreened multiplet transitions and final-state screening effects do not seem to contribute to the satellite structure.

Satellite structure in the photoelectron spectra of transition-metal compounds

Abstract The electronic structure of high-spin CrCl3 and CrBr3 compounds has been investigated by means of X-ray photoelectron spectroscopy. Satellite features have been observed in the valence-band region and in Cr2p, Cr3s and Cr3p core-level spectra. Photoemission spectra have been analysed in terms of a simple cluster model, leading to estimates of the Coulomb correlation energy, charge transfer energy and hybridization strength. Satellites can be explained in general by a charge-transfer process, if the effective hybridization energy T eff (3.6-3.7 eV) is large enough. However, in the case of Cr3s core-level spectra, the main emission is assigned to unscreened multiplet transitions and final-state screening effects do not seem to contribute to the satellite structure.

UHV-Cleaved NiO (100), CoO (100), and MnO (100) by He II UPS

The rocksalt transition-metal oxides whose He II UPS spectra are presented here have the same crystal structure but different cation d-electron configurations: Mn2+ is 3d5; Co2+ is 3d7; and Ni2+ is 3d8. There is currently a great deal of interest in the growth of thin films of these oxides, and it is important to determine the stoichiometry of such films. These UPS spectra are for ultrahigh vacuum-cleaved single-crystal surfaces having bulk stoichiometry. While interpretation of the spectra is complicated by final-state screening effects [V. E. Henrich and P. A. Cox, see The Surface Science of Metal Oxides (Cambridge University Press, Cambridge, 1994)], the increase in the number of occupied 3d orbitals in going from MnO to NiO is clearly visible in the size of the peak at low binding energy (which is the 0 eV reference for the abscissa in the accompanying figure).

Core-level X-ray photoelectron spectra and X-ray photoelectron diffraction of RuO 2(110) grown by molecular beam epitaxy on TiO 2(110)

We have measured Ru 3d, 4s, 4p and O 1s high-resolution core-level X-ray photoelectron spectra, along with Ru 3d and O 1s scanned-angle X-ray photoelectron diffraction angular distributions, for RuO 2(110). The surfaces were prepared by oxygen-plasma-assisted molecular beam epitaxial growth of RuO 2 on TiO 2(110). XPS spectral interpretation and the nature of the XPD scans strongly suggest that the complex line shapes are due to final-state screening effects, rather than the presence of Ru in oxidation states other than +4.

Adlayer Core-Level Shifts of Random Metal Overlayers on Transition-Metal Substrates

We calculate the difference of the ionization energies of a core-electron of a surface alloy, i.e., a B-atom in a A_(1-x) B_x overlayer on a fcc-B(001)-substrate, and a core-electron of the clean fcc-B(001) surface using density-functional-theory. We analyze the initial-state contributions and the screening effects induced by the core hole, and study the influence of the alloy composition for a number of noble metal-transition metal systems. Data are presented for Cu_(1-x)Pd_x/Pd(001), Ag_(1-x) Pd_x/Pd(001), Pd_(1-x) Cu_x/Cu(001), and Pd_(1-x) Ag_x/Ag(001), changing x from 0 to 100 %. Our analysis clearly indicates the importance of final-state screening effects for the interpretation of measured core-level shifts. Calculated deviations from the initial-state trends are explained in terms of the change of inter- and intra-atomic screening upon alloying. A possible role of alloying on the chemical reactivity of metal surfaces is discussed.

Ab initio calculation of surface core-level shifts for transition metal surfaces

The surface core-level shift (SCLS) for a number of smooth surfaces of 4d-transition metals Mo, Rh, Pd and Ag was calculated within two different approaches using the full-potential LMTO method. The first approach, the initial state approximation, estimates the SCLS from the position of the core eigenvalues of atoms at the surface relative to those of the bulk. The second approach treats a surface or bulk atom with a core hole as an impurity by means of a supercell. This approach is in principle exact within the local-density approximation and thus includes final-state screening effects. The results show that the screening of the core hole is of different nature for Ag when compared to the transition metals.

Final-state screening effect in the 3s photoemission spectra of Mn and Fe insulating compounds.

The satellite structures of 3s core-level photoemission spectra of Mn and Fe dihalides and mono-oxides have been studied systematically. We found that exchange splitting between 3s holes and 3d electrons, intrashell electron correlation, and final-state screening (charge-transfer satellite) effects all contribute to their satellite structures. We extended the existing model of 2p core-hole satellite structures for transition-metal compounds [Park et al., Phys. Rev. B 37, 10 867 (1988)] to the case of 3s core holes by including the exchange interaction and the intrashell electron correlation effect. The intrashell electron correlation effect is included by introducing two parameters, the energy separation and the coupling strength between 3s3${\mathit{d}}^{\mathit{n}}$ and 3${\mathit{p}}^{2}$3${\mathit{d}}^{\mathit{n}+1}$ states. With this model, we were able to explain very well the 3s spectra of the Mn and Fe insulating compounds studied here, and which were consistent with their 2p core-level spectra analyses. We observe that the importance of final-state screening effects in the core-level spectra depends on the ligands. As a result, 3s energy splittings for very ionic compounds such as ${\mathrm{MnF}}_{2}$ and ${\mathrm{FeF}}_{2}$ can be well understood by exchange splitting alone, but that 3s splittings in other compounds in general are not directly related to the 3d local magnetic moment of the ground state. We found that the change of 3s splittings in more covalent compounds is mostly determined by the final-state screening due to the different values of 3s-hole--3d-electron Coulomb attraction Q depending upon the spin of the final state, rather than the exchange energy between 3s holes and 3d electrons.

Auger and x-ray absorption studies of solid molecular oxygen

We have studied the x-ray absorption spectrum (XAS) and the Auger electron spectrum (AES) of condensed molecular oxygen in the vicinity of the O 1s (K) core level. The XAS spectrum, which is similar to that previously reported for gas phase O2, has sharp peaks at about 530.8 and 540.3 eV, which we attribute to excitations from the 1s level into π* and σ* states, respectively. We assign a third peak at 543.3 eV to Rydberg transitions. Compared with gas phase O2, the σ* and Rydberg states are shifted to higher energies, an effect that we attribute to solid state effects in the condensed layer. The AES spectra of the ionized molecule and excited molecule are both shifted relative to the spectra of gas phase O2, but by different amounts, 10 and 2 eV, respectively, which may be due to different final state screening effects. A tentative assignment of the features in the AES spectra is also made.

Inverse photoemission of adsorbed xenon multilayers on Ru(001): Refutation of final-state screening effects.

In this Letter we describe photoemission and inverse photoemission spectra of adsorbed xenon multilayers on Ru(001). Electron energy-loss spectra of xenon adsorbed on gold taken by Demuth, Avouris, and Schmeisser are included in the discussion. The observed layer-dependent shifts of the inverse photoemission spectra closer to the Fermi level clearly invalidate image screening effects as being the dominant cause of these shifts but support a floating of the adsorbed Xe potential well as a whole with the surface potential in the initial state.

Final-state screening of core holes in metals.

Core-level photoemission is often employed to extract chemical information concerning the atomic site and this requires assumptions concerning the balance between initial-state chemical and final-state screening effects. With this in mind, we have calculated the binding energy of the 4f-italic level in gold metal using local-density total-energy calculations. We find a binding energy of 83.7 eV in good agreement with the experimental value of 83.9 eV. We are thus able to test several less complete computational schemes. For example, confining the response to a single site yields 84.3 eV, an error of 0.6 eV, while limiting the screening to the atom and its near neighbors produces an error of 0.07 eV. In additions, we have calculated the heat of solution of mercury in gold. This may be related via a Born-Haber cycle to the core-level binding energy assuming local screening and the equivalent core approximation. This yields a binding-energy error of 0.15 eV, which, while small, is of significance. Finally, we have explicitly decomposed the binding energy into initial- and final-state contributions and find a relaxation energy in gold of 17.7 eV compared to 13.1 eV for the isolated atom.

The electronic structure of Bi 2− xGd xRu 2O 7 and RuO 2: A study by electron spectroscopy

The bismuth gadolinium pyrochlore ruthenates Bi 2− x Gd x Ru 2O 7 have been studied in relation to RuO 2 by the techniques of X-ray and ultraviolet photoelectron spectroscopy (XPS and UPS) and high-resolution electron-energy-loss spectroscopy (HREELS). The composition-dependent metal-to-semiconductor transition in the pyrochlore system is mirrored by the progressive decrease with composition parameter in (i) the density of electronic states at the Fermi energy in UPS, (ii) the plasmon frequency in HREELS, and (iii) the probability of screening a Ru : 3 d core hole in XPS. These changes are too gradual in themselves to pinpoint the transition, but are generally consistent with transport and X-ray diffraction data that indicate a metal-to-nonmetal transition at x = 1.55 mediated by an interplay between disorder and correlation-induced electron localization. Comparison of the results for the pyrochlores with those from RuO 2 suggest that the fine structure in the Ru : 3 d spectrum of the latter material, previously believed to arise from differing oxidation states at the surface, should in fact be attributed to final-state screening effects in a stoichiometric material. Our conclusion is confirmed by the signals associated with Ru : 4 d electrons in XPS, UPS, and HREELS: each of these three techniques appears to probe a conduction-electron concentration essentially equal to its bulk value. In particular, UPS confirms details of the band structure of RuO 2 not obvious from previous photoemission experiments.

Electronic and structural study of Na 2/3WO 3(110) by XPS, UPS, EELS and LEED

The (110) surface of a perovskite sodium tungsten bronze with composition close to Na 2/3 WO 3 has been studied by X-ray and ultraviolet photoemission spectroscopy (XPS and UPS), electron-energy-loss spectroscopy (EELS) and low-energy electron diffraction (LEED). The chemical composition at the surface differs little from that in the bulk, but sodium ion ordering along the [1̄10] direction leads to a (3 × 1) superstructure not evident in bulk diffraction experiments. The UPS data give no indication of surface electronic states in the bulk bandgap, although a surface plasmon loss peak is seen. By contrast fine structure in the tungsten 4f region does not arise from plasmon excitation and must be ascribed to final state screening effects. The screening mechanism is discussed in relation to the UV photoemission spectra.

Summary Abstract: Initial and final state screening effects on the Auger line shapes of silicon

Summary Abstract: Initial and final state screening effects on the Auger line shapes of silicon

Mixed valence and 4f surface chemical shift in CeN

Resonant photoemission has been used to study the 4f state in single crystal Ce chalcogenides and pnictides. For mixed valent CeN a surface binding energy shift of the 4f state of about 1 eV has been observed. The apparent binding energy of the 4f state in Ce pnictides and Ce chalcogenides differs by as much as 3 eV. Systematic trends in the binding energy are observed. It is suggested that final state screening effects due to 5d electrons, which show the same trends, are responsible for the observed effects.