Characteristic Loss Peaks Research Articles

Dielectric loss of ZnO varistor ceramics by variable temperature spectroscopy

In this paper， the dielectric spectra of ZnO varistor ceramic were measured by Novocontrol wide band dielectric spectrometer in the temperature range of -100 ℃—20 ℃ and frequency range of 10-2—106 Hz. It is found that the activation energy is 0.26 and 0.36 eV， respectively. After careful analysis of the measurement conditions， quantitative calculation results and other phenomena， three relaxations of cathode electron injection， dipole polarization and interface polarization are discarded. Thermally stimulated current（TSC） has three peaks， of which the higher temperature peak comes from thermionic polarization introduced by TSC itself， and other two peaks correspond to the characteristic loss peaks in dielectric spectra. Therefore， it is concluded that the characteristic loss peaks of ZnO ceramics originate from electronic relaxation of intrinsic defects.

Adsorbate-induced pinning of a charge-density wave in a quasi-1D metallic chains: Na on the In/Si(111)-(4x1) surface.

We find that foreign adsorbates acting as local impurities can induce a metal-insulator transition by pinning a charge-density wave (CDW) on the quasi-1D metallic In/Si(111)-(4x1) chain system. Our scanning tunneling microscopy image clearly reveals the presence of a new local 4x2 structure nucleated by Na adatoms at room temperature, which turns out to be insulating with a doubled periodicity along the chains. We directly determine a CDW gap energy Delta = 105+/-8 meV by identifying a characteristic loss peak in our high-resolution electron-energy-loss spectra. We thus report the first observation of a local impurity-derived Peierls-like reconstruction of a quasi-1D system.

Broadband Dielectric Relaxation of Organic Glass-Forming Liquids : Molecular Weight Dependence

In order to study the molecular weight dependence of the dielectric properties, the dynamics of the α -relaxation in both liquid and supercooled liquid of the dyhydroxyl propylene glycol monomer (PG) and its five oligomers (poly propylene glycol, PPG) has been investigated as the organic fragile glassforming materials in the frequency range from 10mHz to 10GHz by means of broadband dielectric spectroscopy. The frequency dependence of the complex dielectric permittivity is analyzed by the Harvrilliak-Negami (HN) expression. The temperature dependence of the characteristic loss peak frequency fmax for the α -process shows a Vogel-Tammann-Fulcher (VTF) behavior. The PG monomer is more fragile liquid than its oligomers because of the linear increase of the measure of non-linearity to 1/Mn.

Tantalum deposition on and reaction with the hydrogen terminated diamond (100) surface studied by Auger electron and electron energy loss spectroscopy

The evolution of the tantalum/diamond interface upon room-temperature Ta deposition on the (100) surface of a boron doped, synthetically grown diamond single crystal was monitored by Auger electron spectroscopy (AES), ionization loss spectroscopy (ILS), and electron energy loss spectroscopy (ELS). Characteristic loss peaks indicate carbide formation at the interface from very low coverages on, reflecting the strong interaction between tantalum and carbon. Thicker layers of TaC are formed during subsequent thermal annealing by diffusion of carbon into the tantalum film, at the same time the topmost diamond region is transformed into poorly ordered graphitic carbon.

Damping behaviour of dynamically cured butyl rubber/polypropylene blends

Dynamically cured blends of butyl rubber (IIR; isoprene-isobutylene rubber) and anhydride-grafted polypropylene (PP) of various component ratios and curative loadings were prepared by melt mixing. Evidence from dynamic mechanical analysis and phase-contrast microscopy suggested that these blends were heterogeneous, with the PP phase being continuous. Damping characteristics of these blends in the temperature range of −60 to 0°C were strongly affected by a liquid-state relaxation peak of the IIR phase. It was observed that the presence of a continuous PP phase tends to suppress the intensity of this characteristic loss peak and shift it towards lower temperatures; on the other hand, the crosslinking of IIR has little effect on the intensity of the broad loss peak but tends to shift it towards higher temperatures. These observations were interpreted in terms of effects of thermal misfit stress and crosslinking on the liquid-liquid transition of the IIR phase.

Simulation of electron/solid interaction and its application to quantitative analysis by Auger electron spectroscopy

AbstractThe simulation of electron/solid interactions is discussed, with emphasis on its applicability to the matrix correction in quantitative analysis by AES, and to its potential use in background substraction in AES. The Monte Carlo calculation models are first explained, which have been improved to allow the estimation of (i) back‐scattering correction factors, and (ii) energy profiles of back‐scattered electrons in the high‐energy region (i.e. elastically‐scattered electrons and the characteristic loss structures) and the slow secondary region. For this purpose, a dielectric approach is used for the treatment of the electron inelastic scattering in the solid. Then, the models are applied to samples with bi‐layer structures, and the change of Si Auger signals observed during sputter‐etching of a sample consisting of a Si thin film (7000 Å) on a W substrate is well explained. Changes in the energy distributions of back‐scattered electrons during the formation of very thin (5 to 100 Å) Cu films on a Si substrate are also estimated by the Monte Carlo calculation, together with the shapes of the elastic peak and the characteristic loss peaks.

Scanning Low Energy Electron Loss Microscopy (SLEELM): Au on Si

High resolution scanning low energy electron loss imaging of a well characterised system of Au on Si is demonstrated in reflection geometry at a primary electron energy E p of 680 eV and spatial resolution of 3μm. Characteristic losses below 30 eV are chosen to examine contrast between two regions (gold and silicon dominated) respectively. By differencing between a maximum in a loss intensity and an adjacent trough, high quality Si images are obtained with contrast superior to L 2,3VV Auger electron images produced in comparable times. It should be noted that, in contrast to scanning Auger microscopy, the spatial resolution of SLEELM is limited only by the profile of the incident beam. An advantage of low electron energy loss imaging is that the depth resolution may be varied by altering the primary electron energy, with the possibility of bridging the gap between Auger microscopy and the X-ray microprobe analyser. Disadvantages of the technique lie in the weakness of the elastic backscattering which is essential to observe the losses in reflection geometry, and the possibility of overlapping low energy characteristic loss peaks, an effect which is illustrated here for the Au losses. It is shown that electron probes with characteristics rather different from those currently used in Auger microscopes are required to optimize the technique, but that SLEELM has the potential to be a very useful form of surface microscopy which will complement other surface imaging techniques.

Scanning low energy electron loss microscopy (SLEELM)

SUMMARYHigh resolution scanning low energy electron loss imaging of a well characterized system of W on Si is demonstrated in reflection geometry at a primary electron energy Ep of 740 eV and spatial resolution of 3 μm. Characteristic losses below 60 eV are chosen to examine contrast between two types of regions (tungsten and silicon dominated, respectively). By differencing between a maximum in loss intensity and an adjacent trough, high quality chemical images are obtained with contrast far superior to Auger electron images produced in comparable times. It should be noted that, in contrast to scanning Auger microscopy, the spatial resolution of SLEELM is limited only by the profile of the incident beam. In addition, the technique is readily capable of chemical analysis at the surface of bulk specimens. A further advantage of low electron loss imaging is that the depth resolution may be varied by altering the primary electron energy, with the possibility of bridging the gap between Auger electron microscopy and X‐ray microprobe analyser. Possible disadvantages of the technique lie in the weakness of the elastic backscattering which is essential to observe the losses in reflection geometry, and the possibility of overlapping low energy characteristic loss peaks, an effect which is, however, less bad than for low energy Auger peaks. It is shown that electron probes with characteristics rather different from those currently used in Auger microscopes are required to optimize the technique, but that SLEELM has the potential to be a very useful form of surface microscopy which will complement other surface imaging techniques.

X-ray photoemission studies of atom implanted matrices: Ni in carbon

Nickel, in low concentration, has been implanted in carbon foil. Both the Ni 2p3/2 and the L3M45M45 Auger levels are shifted 0.9 eV relative to nickel foil, while the valence band level is shifted by 1.2 eV. The characteristic energy loss peak associated with the core and valence levels of bulk nickel is observed on the core and valence levels of implanted nickel. An analysis is presented to show that the characteristic loss peak is consistent with 3d→4s or 4s→4p shake up processes. Model-based spectra are consistent with an initial state configuration... 3d94s1 for implanted nickel atoms.

Secondary electron energy spectra of single-crystal Fe(110) at various emission angles

Secondary electron energy spectra from Fe(110) surfaces were measured at various emission angles with a fixed angle of incidence to study the escape process. All spectra which are normalized to constant peak value agree with each other at Ep=400 and 1055 eV. It may be concluded from these results that the escape process of secondaries in a single crystal of iron is approximately the same as that in a polycrystalline material. Hence it is possible, in the first approximation, to calculate with the free-electron model the yield and the extent of diffusion of secondaries which determines the quality and resolution of a scanning electron microscope. No difference was found in the backscattered energy spectra between the anomalous transmission and absorption condition for the primary beam of 1·0 keV. These energy spectra consists mainly of large characteristic loss peaks and energy spectra of rather gradual slope.

The form of characteristic energy-loss spectra

1. Expressions have been obtained for electron-path distribution functions in the form of elastic sifting matrices and have been used for deriving formulas for the intensities of characteristic-loss peaks (with certain assumptions and for the spectrum-recording conditions indicated). 2. An experimentally observed spectrum of characteristic energy losses consists of individual maxima, i.e., “peaks.” The position and form of peaks on a spectrum envelope result from the superposition of peaks of various types of loss. For this reason, to determine true losses, particularly when there are several types of them, it is necessary to resolve an experimentally observed spectrum envelope into its components. 3. In reflection spectra, it is possible to detect only a few peaks of mutliple and composite losses. With increasing magnitude of energetic losses, the spectrum envelope gradually reaches a plateau, which, depending on the experimental conditions (value of Δ/B), is either almost horizontal or inclined. The plateau may be raised in those loss-spectrum zones whose energy is comparable with the energy of the electrons hitting the specimen. 4. Model reflection and transmission spectra calculated for aluminum using the formulas derived and the parameters indicated closely approximate in shape to experimentally observed spectra. 5. Spectra calculated for zirconium, niobium, and molybdenum on the assumption that there are two types of volume plasma loss, corresponding to the existence of two groups of valence electrons, are very similar in shape, when suitable peak parameters are chosen, to experimentally observed spectra.

Inelastic Scattering Cross Sections for 20-keV Electrons in Al, Be, and Polystyrene

Measurements are reported of the inelastic scattering cross sections associated with the characteristic energy losses of 20-keV electrons transmitted through thin films of Al, Be, and polystyrene. These measurements have been made using two techniques which overcome several sources of systematic error in previous measurements. For materials which have narrow characteristic loss lines (such as Al), it is possible to establish the form of the differential cross section and the generalized oscillator strength for the \ensuremath{\approx} 15-eV Al plasmon energy loss for scattering angles between 0 and 20 mrad, the cutoff angle. The differential cross section has been used in repeated two-dimensional folding calculations to correct the intensity measurements of the multiple plasmon losses at the larger scattering angles where unambiguous intensity measurements could not be made. For materials which have broad characteristic loss peaks, such as polystyrene, it is not possible to establish the variation of generalized oscillator strength $f(q)$ with momentum transfer $q$. If it can be assumed that $f(q)\ensuremath{\approx}f(0)$ for small $q$, a cross section for a given loss can be obtained from an energy-loss spectrum measured at one scattering angle. The cross-section measurement accuracy is improved by a comparison of two energy-loss spectra, that of a standard (such as Al) and that of a material of unknown loss cross section, which have been obtained at zero angle under the same measurement conditions. Cross sections of (1.5\ifmmode\pm\else\textpm\fi{}0.4)\ifmmode\times\else\texttimes\fi{}${10}^{\ensuremath{-}18}$ ${\mathrm{cm}}^{2}$ and (1.5\ifmmode\pm\else\textpm\fi{}0.5)\ifmmode\times\else\texttimes\fi{}${10}^{\ensuremath{-}18}$ ${\mathrm{cm}}^{2}$ have been measured for the excitation of the \ensuremath{\approx} 15-eV and \ensuremath{\approx} 19-eV plasmon energy losses in Al and Be, respectively, using the former measurement technique. These values are in good agreement with the cross sections expected theoretically. The second measurement technique has been used to obtain cross sections of (2.4\ifmmode\pm\else\textpm\fi{}0.8)\ifmmode\times\else\texttimes\fi{}${10}^{\ensuremath{-}20}$ ${\mathrm{cm}}^{2}$ and (1.0\ifmmode\pm\else\textpm\fi{}0.2)\ifmmode\times\else\texttimes\fi{}${10}^{\ensuremath{-}18}$ ${\mathrm{cm}}^{2}$ for the 7-eV and 21-eV energy losses, respectively, in polystyrene. The corresponding oscillator strengths $f(0)$ are in good agreement with those deduced recently by LaVilla and Mendlowitz. Values were also obtained for the Al loss-dispersion constants, $\ensuremath{\Delta}E(\ensuremath{\theta})=14.8+(1.39\ifmmode\pm\else\textpm\fi{}0.07)\ifmmode\times\else\texttimes\fi{}{10}^{4}{\ensuremath{\theta}}^{2}+(1.2\ifmmode\pm\else\textpm\fi{}0.2)\ifmmode\times\else\texttimes\fi{}{10}^{7}{\ensuremath{\theta}}^{4}$ (energy loss $\ensuremath{\Delta}E$ in eV, scattering angle $\ensuremath{\theta}$ in radians), and for the critical wave vector for plasmon excitation ${k}_{c}=1.46\ifmmode\pm\else\textpm\fi{}0.07$ ${\mathrm{\AA{}}}^{\ensuremath{-}1}$.