Accurate measurements of electron energies by field-emitter referencing

A new technique of electron Spectroscopy is described in which a standard electron spectrometer is used to compare the electron energy to be measured with the energies of electrons from a field-emitter reference source. Since the reference electrons are emitted at the Fermi level of the emitter tip, their total energy can be accurately measured with a high precision digital voltmeter. The measurements are automatically referenced to the absolute scale of energies and the need for calibration standards is eliminated. The application of the technique to X-ray photoelectron Spectroscopy (XPS) is described. Detailed analysis shows that the uncertainty in the field-emitter referenced XPS measurements is ± 0.06 electron volts.     

Attenuation lengths of low-energy electrons in solids

A compilation is presented of measured attenuation lengths of low-energy electrons in solids in the energy range (40 to 2000 eV) normally employed in X-ray photoelectron and Auger-electron spectroscopy. The techniques used to obtain electron attenuation lengths are summarized, and it is pointed out that the accuracy of measurement needs both to be defined adequately and to be improved for more meaningful intercomparisons of data and theory. An approximate expression is derived to predict attenuation lengths using either dielectric data (derived from optical or electron-energy-loss data) or average excitation energies estimated from electron binding energies for given materials at electron energies greater than about 500 eV. Good agreement is found between the predictions of this formula and some measured attenuation lengths (e.g. for Al, C, Mo, W) but further work is required to validate the formula and to extend it to lower electron energies.     

Direct experimental determination of the dispersion of a final state energy band

We report the first absolute and continous experimental determination of the dispersion of an unoccupied electron energy band in copper. This has been achieved by monitoring the extremal behaviour of binding energies in angular resolved photoelectron energy distribution curves obtained with a tunable light source. By this method the absolute value of the wave vector for transitions originating from the symmetry line ∑ was determined. The experiments confirm that the final state band deviates strongly from a parabola.     

In-situ photoelectron spectroscopy study of a TiS2 thin film cathode in an operating Na intercalation electrochemical cell

TiS₂ thin films were prepared and intercalated in UHV either chemically or electrochemically and investigated by photoelectron spectroscopy. The chemical reaction was induced by Na evaporation. For the electrochemical reaction, the film was deposited on a Na solid electrolyte and the voltage between the TiS₂ and a graphite layer on the back of the plate was controlled during PES investigations. With both methods the same effects on the substrate are observed. The Na Auger peak appears and increases at a kinetic energy typical for intercalated Na. Due to the electron transfer to the conduction band an increase of the electron density at the Fermi level is clearly observed. The progressive filling of higher energy states shifts the Fermi level as reference for the PES spectra, and as a result the S 2p core levels and valence bands are shifted to higher binding energies. A shoulder appears at the lower binding energy side of the Ti 2p peak, indicating a higher negative charge density on the Ti atoms. It is also shown how the in-situ electrochemical intercalation allows experimentally to extrapolate the variation of the ionic contribution to the battery voltage. Paper presented at the 6th Euroconference on Solid State Ionics, Cetraro, Calabria, Italy, Sept. 12–19, 1999.     

Photoelectron spectroscopy

The energy spectra of photoejected electrons provides a direct method for determining the electronic binding energies in atoms, molecules and solids. The binding energies may be used to identify a particular atom in a molecule or solid and, as well, to indicate from which orbital the electrons were removed. The relative intensities of peaks in the photoelectron spectrum reveal the nature of the chemical bond in a molecule and are a measure of the relative transition probabilities for ionization to a particular state. The effects of autoionization and angular distribution are discussed, and some applications are given.     

Angle- and energy-dependent core-level photoelectron energy loss studies in Al and In

Electron energy loss structures of Al and In core-level photoemission spectra, in particular surface and bulk plasmon losses, have been investigated as functions of photon energy (i.e., photoelectron kinetic energy). These studies utilized synchrotron radiation to provide a variable photon source in the ultra-soft X-ray region, thus allowing these loss processes to be studied at photoelectron kinetic energies for which the mean free path of the electrons is minimal. The Al plasmon loss structure was also studied with soft X-ray radiation in an angle-resolved mode, allowing the variation of effective photoelectron sampling depth with different electron take-off (collection) angles. These results for the relative intensity of the bulk and surface plasmons as a function of electron kinetic energy and electron exit angle are in qualitative agreement with the predictions of ?unji? and ?ok?evi?. The core-level binding energies of surface atoms have also been studied with the result that no significant shift has been observed with respect to bulk-atom core levels.     

Specific features of photoelectron emission from palladium clusters on graphite

Subthreshold photoelectron emission was observed to emerge from palladium nanoclusters formed on pyrographite surface under irradiation by photons in the energy range 3.1–6.5 eV. The average size of the palladium nanoclusters on the pyrographite surface was 50–80 nm, and the average height, 2–4 nm. Besides conventional photoemission from states below the Fermi level, photoelectron emission was observed at the energies of photons irradiating the surface 0.9 eV below the work function of the Pd surface. It is assumed that this emission is stimulated by direct electron transitions from Pd states below the Fermi level to the unfilled electron surface states formed in the Coulomb potential of image forces (image states) and, subsequently, into vacuum. This phenomenon is assumed to originate from the contact spot field generated above the surface which is nonuniform in work function. This assumption is supported by the calculations presented in the paper.     

Adsorption energy and geometry of physisorbed organic molecules on Au(1 1 1) probed by surface-state photoemission

The modification of the Au(1 1 1) Shockley-type surface state by an adsorbed monolayer of large π-conjugated molecules was investigated by high-resolution angle-resolved photoelectron spectroscopy (ARPES). We determined binding energy, band mass, and Rashba-splitting and discuss the results in the context of rare-gas adsorption on noble metals. This comparison allows the determination of the bonding strength of the adsorbates, found to be physisorptive with derived binding energies per molecule of 2.0 eV for perylene-tetracarboxylic-dianhydride (PTCDA) and 1.5 eV for naphtalene-tetracarboxylic-dianhydride (NTCDA). We will also present a superstructure model for the NTCDA/Au(1 1 1) system, deduced from low energy electron diffraction images (LEED) in combination with substrate band-backfolding.     

On the nature of the SO4/Ag(111) and SO4/Au(111) surface bonding

The nature of sulfate-Ag(111) and sulfate-Au(111) surface bonding has been investigated at the SCF + MP2 level of theory. Convergence of binding energy with cluster size is investigated and, unlike neutral adsorbates, large clusters are required in order to obtain reliable binding energies. In the most stable adsorption mode, sulfate binds to the surface via three oxygen atoms (C_3v symmetry) with a binding energy of 159.3 kcal/mol on Ag(111) and 143.9 kcal/mol on Au(111). The geometry of adsorbed sulfate was optimized at the SCF level. While the bond length between sulfur and the oxygens coordinated to the surface increases, the sulfur-uncoordinated oxygen bond length decreases. This weakening and strengthening of the bonds, respectively, is consistent with bond order conservation in adsorbates on metal surfaces. Although a charge transfer of 0.4 electrons towards the metal is observed, the adsorbate remains very much sulfate-like. The molecular orbital analysis indicates that there is also some charge back-donation towards unoccupied orbitals of sulfate. This results in an increased electron density around sulfur as revealed in the electron density difference maps. Analysis of the Laplacian of the charge density of free sulfate provides a suitable framework to understand the nature of the different charge transfer processes and allows us to establish some similarities with the CO- and SO₂-metal bondings.     

Interpretative methods in photoelectron spectroscopy

Experimental carbon and nitrogen core electron binding energies measured by high energy photoelectron spectroscopy are reported together with ab initio and semi-empirical molecular orbital studies for a series of organic pseudohalides. The derived wavefunctions are used in a discussion of molecular potential models which are employed in the interpretation of core photoelectron spectra. The application of localized wavefunctions in this context is described.     

Photoelectron spectroscopy on Pt atoms and clusters deposited on C(0001)

An experimental photoelectron spectroscopy study is presented highlighting several aspects of importance for the study of deposited metal clusters and particles with photoemission. It is shown that the Fermi level is the correct energy reference for the core level binding energies. The choice of different deposition conditions, well within the range of soft landing, has a strong impact on the outcome of the spectroscopic experiments. Single adatoms as well as clusters deposited with some excess energy display relatively narrow core level spectra at much lower binding energies than previously reported, even when atomic mass selection is not performed. In contrast, single sized Pt₁₉ clusters, deposited onto a thin Ar film before being exposed to the graphite surface show spectral broadening and shifts to higher binding energies. We discuss our results in terms of the cluster substrate interaction and the influence of deposition conditions on the metal adsorbate structure.     

Electron-energy bunching in laser-driven soft recollisions

We introduce soft recollisions in laser-matter interaction. They are characterized by the electron missing the ion upon recollision in contrast with the well-known head-on collisions responsible for high-harmonic generation or above-threshold ionization. We demonstrate analytically that soft recollisions can cause a bunching of photoelectron energies through which a series of low-energy peaks emerges in the electron yield along the laser polarization axis. This peak sequence is universal, it does not depend on the binding potential, and is found below an excess energy of one tenth of the ponderomotive energy.     

Estimates of energies of core exchange

Energies of core exchange, δ, for some first row elements have been estimated from thermodynamic data and experimental core electron binding energies. Values of energies of core exchange extend the equivalent cores method to allow the calculation of absolute binding energies as well as shifts. In some cases it is possible to estimate the differences in binding energy without requiring heats of formation of gaseous cations. The values of δ obtained from LCAO-MO-SCF calculations show similar trends although the absolute values may differ by a few electron volts.     

Relative energies of surface and defect states: ab initio calculations for the MgO (001) surface

We present the results of calculations of the energy levels of defects at the (001) surface of MgO relative to the top of the valence band and values of defect ionisation potentials and electron affinities. The calculations were made using an embedded cluster method in which a cluster of several tens of ions treated quantum mechanically is embedded in a finite array of polarisable and point ions modelling the crystalline potential and the classical polarisation of the host lattice. The calculated ionisation potential of the ideal surface, which fixes the position of the top of the valence band with respect to the vacuum level, is about 6.7 eV. This value is used as a reference for positioning the energy levels of three charge states of a surface anion vacancy, which are also calculated as ionisation energies with respect to the vacuum level. The surface and vacancy electron affinities are calculated using the same method. As a prototype low-coordinated surface site, we have considered a cube corner. Our calculations predict the splitting of the corner states from the top of the surface valence band by about 1.0 eV. Both unrelaxed and relaxed holes are strongly localised at the corner oxygen ion. The ionisation energies and electron affinities of the corner anion vacancy are calculated. The electrons in the F and F⁺ centres at the corner are shown to be significantly delocalised over surrounding Mg ions.     

Energy level alignment between C60 and Al using ultraviolet photoelectron spectroscopy

The energy level alignment between C₆₀ and Al has been investigated by using ultraviolet photoelectron spectroscopy. To obtain the interfacial electronic structure between C₆₀ and Al, C₆₀ was deposited on a clean Al substrate in a stepwise manner. The valence-band spectra were measured immediately after each step of C₆₀ deposition without breaking the vacuum. The measured onset of the highest occupied molecular orbital energy level was located at 1.59 eV from the Fermi level of Al. The vacuum level was shifted 0.68 eV toward lower binding energy with additional C₆₀ layers. The observed vacuum level shift means that the interface dipole exists at the interface between C₆₀ and Al. The barrier height of electron injection from Al to C₆₀ is 0.11 eV, which is smaller value than that of hole injection.     

Monte Carlo simulation of absolute secondary electron yield of Cu

A Monte Carlo simulation model of electron interaction with solids that includes cascade secondary electron production has been used to study secondary electron emission from Cu. An optical dielectric function was used to describe electron energy loss and the associated secondary electron excitation. From the simulation, the absolute primary energy dependence of the secondary yield and the energy distribution of secondary electrons has been obtained. We have compared the calculated true secondary yield and total secondary yield curves with experiment. Good agreement has been found only for those experiment data obtained in ultrahigh vacuum. PACS 79.20.Hx; 79.20.Ap; 02.70.Uu     

AES: energy calibration of electron spectrometers. IV. A re-evaluation of the reference energies

Reference kinetic energies for Auger electrons were published by the National Physical Laboratory (NPL) in 1990. These energies are traceable via measurements of X-ray photoelectron binding energies, referenced to the Fermi level position in the spectrum, and values for the X-ray energies involved. Revised values of both the Fermi level position and the X-ray energies have now been obtained. Use of these new values leads to an increase of the recommended reference kinetic energies for AES by 0.07 eV from the previous values.     

Electron beam-adsorbate interactions on silicon surfaces

The interactions which occur between electron beams in the energy range 0.5–2.5 keV, with currents of 0.1–1.0 microA and various adsorbates (H₂, CO, CH₄ and C₂H₄) on silicon surfaces have been investigated. The accumulation of beam induced dissociation products on the surface has been monitored by Auger spectroscopy, and the extent of electron stimulated desorption of neutral molecules has been determined mass spectroscopically. Thermal desorption spectra for various gases have also been obtained in order to compare adsorption behaviour with and without the presence of an electron beam. It is concluded that serious experimental errors may occur when LEED and AES are used in adsorption studies, particularly where comparatively weak binding energies are involved.     

Auger and direct electron spectra in X-ray photoelectron studies of zinc,zinc oxide,gallium and gallium oxide

Auger and direct electron spectra from Zn, ZnO, Ga and Ga₂O₃ have been studied with X-ray photoelectron spectroscopy (ESCA). The chemical shift between zinc electron binding energies in Zn and ZnO is very small, whereas the zinc Auger electron signals are separated by 4.3 eV. In gallium, the oxide and metal signals are separated by 1.9 eV, but the Auger electron energy shift is three times as large. Thus the Auger signals are more sensitive to the chemical environment than the direct electron signals, which is the same relation as earlier observed for copper and copper oxides.