Search

We analyze experimentally and theoretically the angular distribution of electrons backscattered from Cu(001) at medium energies. We present polar intensity plots (PIP's) of backscattered electrons with important qualitative differences with PIP's of x-ray photoemitted electrons acquired in similar conditions. In particular, our PIP's of backscattered electrons present peaks, ascribed to diffraction effects in the incident part of the electron trajectory, and a symmetry that are not observed in the PIP's of photoelectrons. The theoretical analysis has been made using a model presented recently to explain the change of the angular pattern in passing from low to medium energies. We present calculations performed including curved-wave-front and multiple-scattering effects that are in very good agreement with the experiments. The results show that the main structures in the PIP's are produced by forward-focusing interferences (similar to those of photoelectron diffraction) in either the incoming or the outgoing parts of the electron trajectory, and by interferences between waves with important forward scattering in both the incoming and outgoing parts of the electron trajectory.

We studied the electronic structure of ${\mathrm{CaMnO}}_{x}$ $(2.66l~xl~3.00)$ using x-ray photoemission and O $1s$ x-ray-absorption spectroscopy. Analyzing the spectra of the two end members with the configuration-interaction cluster model, we have determined all the main parameters of the electronic structure. We have found that the ground states of ${\mathrm{CaMnO}}_{2.5}$ and ${\mathrm{CaMnO}}_{3}$ are highly covalent, with approximately the same number of ligand holes per oxygen atom: 0.22--0.23. The main separations between the Mn $3d$ bands closest to the Fermi level are of the order of 3 eV and the band gaps are of the charge-transfer type $(Ug\ensuremath{\Delta})$. The main effects of the oxygenation are the disappearance of the occupied ${e}_{g\ensuremath{\uparrow}}$ band and an approximately rigid shift of about 1 eV to lower energies of all the other Mn $3d$ bands. Finally, we discuss the most probable causes for the absence of a metallic phase in ${\mathrm{CaMnO}}_{x}.$

The ``elastic'' backscattering of electrons from crystalline surfaces presents two regimes: a low-energy regime, in which the characteristic low-energy electron diffraction (LEED) pattern is observed, and a medium-energy regime, in which the diffraction pattern is similar to those observed in x-ray photoemission diffraction (XPD) and Auger electron diffraction (AED) experiments. We present a model for the electron scattering which, including the vibrational degrees of freedom of the crystal, contains both regimes and explains the passage from one regime to the other. Our model is based on a separation of the electron and atomic motions (adiabatic approximation) and on a cluster-type formulation of the multiple scattering of the electron. The inelastic scattering events (excitation and/or absorption of phonons) are treated as coherent processes and no break of the phase relation between the incident and the exit paths of the electron is assumed. The LEED and the medium-energy electron diffraction regimes appear naturally in this model as the limit cases of completely elastic scattering and of inelastic scattering with excitation and/or absorption of multiple phonons. Intensity patterns calculated with this model are in very good agreement with recent experiments of electron scattering on Cu(001) at low and medium energies. We show that there is a correspondence between the type of intensity pattern and the mean number of phonons excited and/or absorbed during the scattering: a LEED-like pattern is observed when this mean number is less than 2, LEED-like and XPD/AED-like features coexist when this number is 3\char21{}4, and a XPD/AED-like pattern is observed when this number is greater than 5\char21{}6. \textcopyright{} 1996 The American Physical Society.

We present a theory of the elastic scattering of electrons from crystalline surfaces that contains both low-energy-electron-diffraction (LEED) effects at low energies and x-ray-photoelectron- and Auger-electron-diffraction (XPD/AED) effects at intermediate energies. The theory is based on a cluster-type approach to the scattering problem and includes temperature effects. The transition from one regime to the other may be explained as follows: At low energies all the scattered waves add coherently, and the intensity is dominated by LEED effects

The combination of the two-dimensional layered structure and the semimetallic character of graphite is expected to produce a $\mathrm{C}\phantom{\rule{0.2em}{0ex}}1s$ photoemission spectrum of the (0001) surface with only one highly symmetric component. This simple view has been challenged by a recent study, performed on highly oriented pyrolytic graphite (HOPG), in which the existence of two components separated by $120\phantom{\rule{0.3em}{0ex}}\mathrm{meV}$ has been reported [T. Balasubramanian et al., Phys. Rev. B 64, 205420 (2001)]. Since HOPG is not perfectly crystalline, in this study, we present a comparison of high-resolution $\mathrm{C}\phantom{\rule{0.2em}{0ex}}1s$ spectra measured in HOPG, natural graphite crystals, and crystalline graphite films grown on $6H\text{\ensuremath{-}}\mathrm{Si}\mathrm{C}(0001)$. Our results demonstrate clearly that the spectrum in graphite(0001) has just one component; the second component at $120\phantom{\rule{0.3em}{0ex}}\mathrm{meV}$ is observed only in HOPG and must, therefore, be associated with defects present in this not perfectly crystalline material.

Using angle resolved photoemission we show that bandlike states in ${\mathrm{K}}_{6}{\mathrm{C}}_{60}(110)$ films at $40\phantom{\rule{0.3em}{0ex}}\mathrm{K}$ survive in spite of the strong electron-phonon coupling expected in fullerides that should give rise to polaron formation. The momentum dependence and the total bandwidth of the dispersing bands contributing to the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) features agree quite well with one-electron band structure calculations. While a crude estimation of the dimensionless electron-phonon coupling constant $\ensuremath{\lambda}$, from the temperature dependence of the LUMO linewidth in photoemission spectra, seems to place ${\mathrm{K}}_{6}{\mathrm{C}}_{60}$ in an intermediate polaronic regime, the expected strong coupling to ${K}^{+}$ modes remains uncertain.

We have analyzed the $\sqrt{3}\ifmmode\times\else\texttimes\fi{}\sqrt{3}$ $R30\ifmmode^\circ\else\textdegree\fi{}$ reconstruction of the Si-terminated $6H\text{\ensuremath{-}}\mathrm{Si}\mathrm{C}(0001)$ surface using high-resolution photoemission and photoelectron diffraction. We have acquired a large set of Si $2p$ and C $1s$ spectra in the form of azimuthal scans at different polar angles and photon energies. In the case of the C $1s$ core level we have confirmed that one of the surface components reported previously is due to surface contamination. However, in the spectrum of the contamination-free surface, reported here for the first time, we have uncovered a hitherto hidden new component and, therefore, the number of surface components remains at two. We consider different geometries for the surface reconstruction and conclude that only the $\mathrm{Si}\text{\ensuremath{-}}{T}_{4}$ model is compatible with the existence of two surface components in both the Si $2p$ and C $1s$ spectra. Additionally, we show that multiple-scattering cluster diffraction (MSCD) calculations based on this model reproduce correctly all the intensity modulations. These calculations have also been used to determine the main structural parameters. We have also analyzed the data in the light of the new interpretation of the Si $2p$ spectrum proposed by Anisimov et al. [Phys. Rev. B 61, 1752 (2000)]. These authors have suggested that new $2p\text{\ensuremath{-}}3p$ exchange interactions arising in the Mott-Hubbard ground state could completely change the spectrum of the Si adatoms. We have accordingly fitted the Si $2p$ spectra with a standard $2{p}_{3∕2,1∕2}$ doublet representing the bulk component and a triplet representing the emission from the adatoms. Although these two components allow a perfect fitting of all the spectra, we have found that the intensity modulations derived with this model are in complete disagreement with the MSCD calculations, which thereby rules out this interpretation.

We present x-ray diffraction, x-ray photoemission spectroscopy, magnetization, and transport results on the cerium-doped manganite La 0 . 4 7 Ce 0 . 2 0 Ca 0 . 3 3 MnO 3 . We have achieved good quality, single-phase polycrystalline samples by the nitrate decomposition route. A Rietveld analysis of the x-ray spectrum revealed a distorted structure with a small Mn-O-Mn apical angle of 152.6°. We found that both 3+ and 4+ charge states of cerium ions coexist, introducing an additional source of cation disorder in the A site of the perovskite structure. The low-temperature magnetization measurements, together with the high-T susceptibility results, indicate a magnetically frustrated behavior below T F 70 K. The zero-field resistivity shows, at T M 145 K. the typical metal-insulator transition found in manganite systems, though a temperature for a long-range ferromagnetic order transition does not obviously exist. The magnetoresistance (MR) is large, with a maximum value close to 100%. Strong hysteresis effects are observed in the MR vs field curves, that correlate well with the magnetization vs temperature measurements. We discuss these observations and show that the resistivity results are closely related to the magnetic behavior of the system.

We studied the changes in the electronic structure of ${\mathrm{LaNiO}}_{3\ensuremath{-}\ensuremath{\delta}}$ across the metal-insulator transition. The technique used in the study was mainly $\mathrm{O} 1s$ x-ray absorption spectroscopy (XAS). The experimental spectrum of ${\mathrm{LaNiO}}_{3.00}$ was analyzed in terms of a cluster-model calculation. The spectrum of ${\mathrm{LaNiO}}_{3.00}$ presents a sharp peak at a threshold that corresponds to ${3d}^{8}L\stackrel{\ensuremath{\rightarrow}}{}c{3d}^{9}$ transitions. Analysis of this peak indicates that the charge carriers in ${\mathrm{LaNiO}}_{3.00}$ contain considerable oxygen character. The intensity of this peak decreases in the spectrum of ${\mathrm{LaNiO}}_{2.75}$ and disappears almost completely for ${\mathrm{LaNiO}}_{2.50}.$ This suggests that the metal-insulator transition is related to the disappearance of the charge carriers and the ensuing band-gap opening. The peak in the ${\mathrm{LaNiO}}_{2.75}$ compound is split due to the presence of two nonequivalent crystallographic sites. The metal-insulator transition in this compound is tentatively attributed to potential disorder between these two sites.

The polarization of the electrons emitted by a KrF excimer laser (pulse duration 16 ns) from polycrystalline iron has been measured as a function of pulse energy. The results are well accounted for by an optical-heating model where the spin and lattice systems have identical temperatures.

We have measured energy distributions of M/sub 2,3/NN Auger electrons of K produced during low-energy Ar and K bombardment of K-implanted surfaces. The spectra present bandlike and atomiclike components. We have assigned the atomiclike peak to reflected and/or sputtered K atoms in the autoionizing state K L3sS3pV4sS (SP) decaying to K . We have not observed any evidence of excitation of autoionizing states of Ar.