Scandium and lutetium surfaces studied by reflection electron energy-loss spectroscopy

The reflection electron energy-loss spectra of the (1 0 0) and (0 0 1) surfaces of Sc single crystals and the (0 0 1) surface of a Lu single crystal have been studied with primary energies in the range 50–2000 eV. Scandium is congeneric with lutetium and the loss spectra of the two elements are very similar in both the collective excitations and the interband transitions. Strong excitations observed at around 41 eV are attributed to 3p → 3d and 5p → 5d transitions in Sc and Lu, respectively. The loss data of Sc fit the characteristic energy-loss data of the other elements of the first group of transition metals. Oxygen adsorption and nitrogen adsorption on the (1 0 0) surface of Sc influence the loss spectra. The observed differences are correlated with density-of-states calculations for Sc, ScO and ScN.     

Electron energy-loss spectra of clean and gas-covered Ni(100) surfaces

Electron energy-loss spectra have been measured on Ni(100) surfaces, clean and following oxygen and carbon monoxide adsorption, at primary energies of 40–300 eV. The observed peaks at 9.1, 14 and 19 eV in the clean-surface spectrum are ascribed to the bulk plasmon of the 4s electrons, the surface plasmon, and the bulk plasmon of the coupled 3d + 4s electron, respectively, and the weak but sharp peak at 33 eV is tentatively attributed to the localized many-body effect in the final state. Assignments of the loss structures on the gas-covered surfaces have been attempted.     

Untersuchung über den einfluss von adsorbierten gasen auf die elektronen-energieverlust-spektren einer Ni(110)-oberfläche

Electron energy loss spectra on a (110) nickel surface exhibit characteristic changes upon adsorption of H₂, CO and O₂. The clean surface shows only the surface and bulk plasmon losses at 8 eV and 18 eV respectively. Adsorption of CO produces two new loss peaks at 13.5 eV and 5.5 eV. Loss peaks due to hydrogen adsorption at 15 eV and 7.5 eV show a strong correlation with the well known adsorption characteristics of this system. The oxygen induced losses are different for chemisorbed O on Ni and NiO. In any case the chemisorption-induced losses are well established for primary energies below 120eV. In the loss spectra with higher excitation energies only a drastic decrease of the surface plasmon loss peak-height is visible. If the new losses can be attributed to one-electron excitations from molecular orbital levels due to the chemisorption bond, with assumptions of the final state of the excited electron a determination of the postition of these levels can be made. In case of CO and H₂ reasonable results are evaluated.     

Electron energy loss spectroscopy of A {111} oriented aluminum single crystal

Recently, the electron energy loss spectroscopy (ELS) and secondary electron spectroscopy (SES) of the interaction of oxygen with polycrystalline aluminum were reported for primary energies in the range 30 ? E_p ? 250 eV. Two new transitions were resolved in the ELS spectra (one at 4 eV for clean aluminum and the other at 12 eV for oxide-covered aluminum) for low primary energies (E_p ~ 30 eV). In this paper we report on experiments utilizing a {111} oriented single crystal of aluminum that confirm the existence of these loss peaks for low primary energies and show that the 4 eV peak position for pure aluminum depends on the primary beam energy. This suggests that this low energy loss peak is due to direct nonvertical inter- and intraband transitions which differs from the previous assignment.     

Variation of electron energy loss spectra of Ni (100) with increasing primary energy

Electron energy loss spectra of clean Ni(1 0 0) show for the first time a 17 eV peak, which is attributed to an interband transtiion. All the observed peaks are shifted to higher energies as the primary electron energy E_p increases from 102 to 2045 eV. This shift is explained by a continuous decay in energy of the primary electrons inside the crystal. At E_p ? 700 eV, the decay takes place in the surface region of the crystal, while at E_p > 700 eV it takes place mainly in the bulk. The rate of decay increases with increasing temperature of the crystal between 300 and 900 K.     

Inner shell excitation,valence excitation and core ionization in NF3 studied by electron energy-loss and X-ray photoelectron spectroscopies

Electron energy-loss Spectroscopy (EELS) at impact energies of 2.5–3 keV has been used to obtain the electron excitation spectra for the N 1s (K-shell), F 1s (K-shell) and valence shell regions of NF₃. The inner shell spectra were recorded using small angle scattering (?1° ) while the valence shell spectrum was obtained at zero degree scattering angle. The inner shell excitation spectra show a strongly enhanced 1s→ δ^* type transition and continuum features which are typical for molecules with highly electronegative ligands. One of the peaks in an earlier published photoabsorption study of the N 1s region has been shown to be due to a N₂ impurity. The valence shell electron energy-loss spectrum shows a number of transitions which are considered to be mainly due to valence-valence type transitions, with also some evidence of Rydberg structure.The X-ray photoelectron spectra (XPS) of the N 1s and F 1s electrons along with their associated satellite structures have also been recorded using Al Kα (1486.58 eV) radiation. The vertical ionization potentials for the N 1s and F 1s electrons were found to be 414.36 (10) eV and 693.24 (10) eV, respectively. Both spectra exhibit a rich and different satellite structure. These “shake-up” features in the satellite XPS spectra are compared with continuum features of the inner shell electron energy-loss spectra and also with the valence shell spectrum.     

Low-temperature method of cleaning <Emphasis Type="Italic">p</Emphasis>-GaN(0001) surfaces for photoemitters with effective negative electron affinity

The changes in the chemical composition, atomic structure, and electronic properties of the p-GaN(0001) surface upon chemical treatment in an HCl-isopropanol solution and vacuum annealing are investigated by x-ray photoelectron spectroscopy, high-resolution electron energy-loss spectroscopy, and low-energy electron diffraction. It is demonstrated that a considerable part of the surface gallium oxide is removed upon chemical treatment of the GaN surface. Subsequent annealing of the surface under vacuum at temperatures of 400–450°C leads to a decrease in the residual carbon and oxygen contamination to 3–5% of the monolayer. The preparation of a clean p-GaN(0001) surface with a (1×1) structure identical to that of the bulk unit cells is confirmed by the low-energy electron diffraction data. The cesium adsorption on the clean p-GaN surface results in a decrease in the work function by ～2.5 eV and the appearance of an effective negative electron affinity on the surface. The quantum efficiency of the GaN photocathode at a wavelength of 250 nm is equal to 26%.     

Synchrotron-radiation photoemission study of the ultrathin Ba/3C–SiC(111) interface

Electronic structure of the Ba/3C–SiC(111) interface has been detailed studied in situ in an ultrahigh vacuum using synchrotron radiation photoemission spectroscopy with photon energies in the range of 100–450 eV. The 3C–SiC(111) samples were grown by a new method of epitaxy of low-defect unstressed nanoscaled silicon carbide films on silicon substrates. Valence band photoemission and both the Si 2p, C 1s core level spectra have been investigated as a function of Ba submonolayer coverage. Under Ba adsorption two induced surface bands are found at binding energies of 2 eV and 6 eV. It is obtained that Ba/3C–SiC(111) interface can be characterized as metallic-like. Modification of both the Si 2p and C 1s surface-related components were ascertained and shown to be provided by redistribution effect of electron density between Ba adatoms and both the Si surface and C interface atoms.     

The electron energy loss spectrum of tin at low primary beam energies

The electron energy loss spectra of clean and oxidised tin have been measured for primary energies in the range 100–1000 eV. The structure found for clean tin is similar to published spectra recorded with low primary energies except for the presence of a small peak at 4.7 eV. Differences between the reported volume plasmon energy recorded with low and high primary beam energies were noted and a model is presented to explain these differences in terms of plasmon dispersion.     

Photoemission studies of single crystals of titanium carbide

The electron distribution in the valence band from single crystals of titanium carbide has been studied by photoelectron spectroscopy with photon energies h?ω = 16.8, 21.2, 40.8 and 1486.6 eV. The most conspicious feature of the electron distribution curves for TiC is a hybridization between the titanium 3d and carbon 2p states at ca. 3–4-eV binding energy, and a single carbon 2s band at ca. 10 eV. By taking into account the strong symmetry and energy dependence of the photoionization crosssections, as well as the surface sensitivity, we have identified strong emission from a carbon 2p band at ? 2.9-eV energy. Our results are compared with several recent energy band structure calculations and other experimental data. Results from pure titanium, which have been used for reference purposes, are also presented.The valence band from single crystals of titanium carbide have been studied by means of photoelectron spectroscopy, with photon energies ranging from 16.8 to 1486.6 eV.By taking into account effects such as the symmetry and energy dependence of the photoionization cross-sections and surface sensitivity, we have found the valence band of titanium carbide to consist of two peaks. The upper part of the valence band at 3–4 eV below the Fermi level consists of a hybridization between Ti 3d and C 2p states. The C 2p states observed in our spectra were mainly excited from a band about 2.9 eV below the Fermi level. The APW^5–9, MAPW¹⁰ and EPM¹¹ band structure calculations predict a flat band of p-character between the symmetry points X′₄ and K₃, most likely responsible for the majority of C 2p excitations observed. The C 2s states, on the other hand, form a single band centered around ?10.4 eV.The results obtained are consistent with several recent energy band structure calculations^{5–11, 13} that predict a combined bonding of covalent, ionic and metallic nature.     

Spectroscopic study of plasmons in ion-irradiated single-walled carbon nanotubes

The electronic structure of single-walled carbon nanotubes was experimentally investigated using x-ray photoelectron spectroscopy, reflection electron energy-loss spectroscopy, and Auger electron spectroscopy. A shake-up satellite structure observed near the C 1s core-level lines in the x-ray photoelectron spectra at high binding energies in the range 284–330 eV due to excitation of π and π + σ plasmons was studied. The effect of irradiation by 1-keV argon ions on the shape of the spectra was analyzed. The shape of the C 1s satellite spectra was found to be sensitive to Ar⁺ irradiation in the electron energy loss range 10–40 eV corresponding to excitation of π + σ plasmons. Auger spectroscopy revealed the presence of argon on the surface of ion-irradiated samples. The argon content increased to ～4 at. % with increasing irradiation dose. An analysis of the results obtained and their comparison with the data available in the literature led to a qualitative conclusion that the bond angles of the carbon atoms making up the walls of single-walled carbon nanotubes are distorted at sites exposed to Ar⁺ irradiation.     

Comparison of the electron energy loss spectrum of erbium metal with ErO2andErH2

Electron energy loss spectra of metallic erbium, Er under different exposures of oxygen at room temperature, and Er deposited in an atmosphere of H₂ are presented in both N(E) and $\text{dN}\text{dE}$ form for primary energies in the range 100–1000 eV. Resonant excitations associated with the 5p and 4d levels in Er show little environmental dependence, and are largely intraatomic in character. In contrast the main plasmon peak shifts to higher energy on exposure to oxygen or hydrogen, and the spectrum of one electron excitations at low energies alters with a decrease in metal losses around 3.5 eV accompanied by a build up of valence band transitions at 8–9 eV. There is no evidence of a stable chemisorption phase under oxygen exposure, but the results are consistent with rapid oxygen incorporation into subsurface layers and oxide formation.     

Adsorption of hydrogen on palladium single crystal surfaces

The adsorption of hydrogen on clean Pd(110) and Pd(111) surfaces as well as on a Pd(111) surface with regular step arrays was studied by means of LEED, thermal desorption spectroscopy and contact potential measurements. Absorption in the bulk plays an important role but could be separated from the surface processes. With Pd(110) an ordered 1 × 2 structure and with Pd(111) a 1 × 1 structure was formed. Maximum work function increases of 0.36, 0.18 and 0.23 eV were determined with Pd(110), Pd(111) and the stepped surface, respectively, this quantity being influenced only by adsorbed hydrogen under the chosen conditions. The adsorption isotherms derived from contact potential data revealed that at low coverages θ ∞ √p_H2, indicating atomic adsorption. Initial heats of H₂ adsorption of 24.4 kcal/mole for Pd(110) and of 20.8 kcal/mole for Pd(111) were derived, in both cases E_ad being constant up to at least half the saturation coverage. With the stepped surface the adsorption energies coincide with those for Pd(111) at medium coverages, but increase with decreasing coverage by about 3 kcal/mole. D₂ is adsorbed on Pd(110) with an initial adsorption energy of 22.8 kcal/mole.     

Structure of the clean Ta(100) surface

The clean Ta(100) surface and some aspects of hydrogen adsorption have been studied by LEED and AES. The thorough examination of LEED patterns did not provide any evidence for an atomic reconstruction of the clean surface over the entire temperature range investigated, 150–600 K. The r-factor analysis used for comparison between measured and calculated I–V spectra yields a contraction of the topmost layer spacing of about 11% and an expansion of the second layer spacing of about 1% compared to the bulk value. The hydrogen adsorption does not induce any superstructures, but small hydrogen exposures lass then 1 L influence I–V spectra substantially.     

Layer-dependent shifts in ionization potential and auger energies for Kr/Cu (110)

Photoelectron spectra of the krypton 3d core-levels and MNN Auger spectra of krypton mono- and multilayers on Cu(110) have been recorded with synchrotron radiation. The Kr-3d line is found to shift to higher binding energies by 0.73 eV for first- and second-layer adsorption, respectively. This value is much larger than the work function decrease for Kr mono-layer adsorption, Δø = ?0.29eV. The shift in Auger line energies is found to be about three times larger than the 3d line shift. These observations can be readily explained in terms of image-charge screening of the hole states.     

Electronic excitations on clean and adsorbate covered Pd(111) by angle resolved electron energy loss spectroscopy

Electronic excitations on clean and adsorbate covered Pd(111) have been investigated by angle resolved electron energy loss spectroscopy. Primary energies in the range of 50–1000 eV were chosen for strong specular reflection to emphasize elastic diffraction-before-loss processes. The clean Pd spectra are compared with optical data, and good correspondence is found for the optical limit (q ? 0). The loss features are interpreted in terms of plasmon resonances and interband transitions within the framework of a recent band structure calculation. Virtually no dispersion is observed for the intrinsic Pd losses, but vertical interband transitions decay fast in the dispersive (q ≠ 0) spectra. Two adsorbate systems are investigated in this study: CO in a disordered adsorbate layer and bromine in a well-ordered $\text{(}\text{3×}\text{3 )R30°}$ structure. Adsorbate derived loss features are generally prominent in the nonspecular (q ≠ 0) spectra. While no dispersion is seen for the intramolecular 13.5 eV excitation of adsorbed CO, dispersion up to 1 eV is found for the Br 4p derived loss feature of the ordered overlayer. This is discussed in terms of a two-dimensional adsorbate band structure of bromine.     

Low energy electron spectroscopy of Pt (100) and Pt (100)/CO

Fine structure in the n_vi, _VIIVV spectrum of clean Pt (100) has been observed, and interpreted as “band like” in origin rather than quasi-atomic. Differences in the dependence of the Auger yield on primary beam energy are observed between the N_VI, _VIIVV and O_IIIVV peaks, and are associated with anomalies in the dependence of the inner shell ionization crossection of the 4f level. Low energy electron loss spectra on the clean surface have been investigated at primary energies in the range 71–774 eV and at angles of incidence of the beam 0–60°. The results are related to high energy loss and optical data, and assignments are given for inter-band and plasmon losses. With approximately $\text{3}\text{4}$ of a monolayer of CO on the surface there is a prominent additional loss at around 13.5 eV, which is interpreted as a one electron transition from a σ state below the d band to available states several electron volts above the Fermi level.     

Evaluation of plasmon-loss spectra of molybdenum using elastic-peak electron spectroscopy

Plasmon-loss spectra of clean polycrystalline molybdenum surfaces have been determined in the primary energy range E_p = 50–3000 eV. Spectra a distributions (nonderivative mode). A simplified model is described for evaluating plasmon-loss spectra using elastic-peak electron spectroscopy, as de of elastically reflected electrons is determined by integrating the N(E) spectrum of secondary and backscattered electrons. The ratio of the ar (23–24 eV) to that of the elastic peak is Pλ, the product of the probability for creating a volume plasmon loss and the inelastic mean free pat second plasmon-loss peak is (Pλ)². Evaluation of our experimental plasmon-loss spectra gives Pλ = 0.4–0.5 for E_p > 500 eV. Th constitutes ～50% of all losses determining the IMFP, interband loss processes being important in the remainder. For the low energy range, E_p found. For E_p < 100 eV, no volume plasmon-loss peak could be detected in our N(E) spectra. The simplified model proves to be valid fo plasmon-loss peak (11–12 eV), i.e., such that N_pls/N_e ? 10^?2. Some results are presented concerning surface plasmon losses as molybdenum surface.     

Study of chemical states at intergranular fracture planes of iron-phosphorus alloys by auger and electron energy-loss spectroscopies

Auger and electron energy-loss spectroscopies (ELS) have been used to study chemical states at the fracture surfaces of iron-phosphorus alloys. The transgranular (TG) and intergranular (IG) fracture planes have similar Auger P LVV transition energies and similar Fe 3p loss energies. These indicate that phosphorus and iron atoms are dispersed atomically in layers segregated at IG planes. The energy separations between the inter-and intra-atomic Auger transition peaks in P LVV show that the TG and IG planes and Fe₃P have similar energy separations between the P 3p and Fe 3p levels. The layers segregated at IG planes are estimated to be equivalent to a monolayer, as indicated by the dependence of the loss energy of the valence-conduction-band transition on primary-electron kinetic energy from 200 to 500 eV. The phosphorus concentration in the segregated layers is saturated to ~ 20 at.% above a bulk phosphorus concentration of 2 at.%.     

Variation of relative intensities between surface and bulk plasmon losses due to crystal orientations for aluminium in low energy electron reflection loss spectroscopy

The contrast change of secondary electron images due to the crystal orientations is observed by the ultra high vacuum scanning electron microscope (UHV-SEM) for crystal grains of clean surface of polycrystalline Al in the primary energy E_p of 200 eV to 5 keV. The low energy electron loss spectra are measured by the cylindrical mirror analyzer. The relative intensity ratio between surface and bulk plasmon loss spectra was dependent on the crystal orientations. The SEM images taken by the surface and bulk plasmon signals at E_p = 230 eV show the inverse contrast depending on the grains. The inversion of the relative intensities between the surface and bulk plasmon losses is explained qualitatively by taking into account of variation of the penetration depth of the incident beam caused by the electron channeling.