VUV isochromat spectroscopy

A highly sensitive isochromat spectrometer employing a frequency selective helium iodine Geiger Müller photoncounter is described. The overall resolution of ±0.36 eV at an average quantum energy of 9.6 eV compares favourably with conventional set-ups. Sample data for a polycrystalline tantal target show satisfactory agreement with earlier experiments.     

Isochromat spectroscopy on CePd3

Isochromat spectra from CePd₃ recorded at quantum energies of 9.7 and 72.7 eV are presented. The data obtained at 9.7 eV show three different peaks and are by and large identical to previously published data for a quantum energy of 1486 eV. Only two of those three peaks are present in the 72.7 eV spectrum. We conclude that the missing peak is bulk in origin and derives from a Ce5d-Pd4d hybridization leading to a Ce5d split off level from the filled Pd d-band.     

Isochromat spectroscopy of photons emitted from metal surfaces in an STM

The pronounced variation of the intensity of photons emitted from the tunneling gap of an STM with respect to the applied bias voltage V_t is studied experimentally using simultaneous measurements of tunneling characteristics and photon emission. We show that the structure in isochromat photon spectra are determined by the following: bias-dependent changes in tunneling characteristics, density of initial and final states, and modifications of tip-induced plasmon modes. It is demonstrated that isochromat spectra provide a conclusive test for the inelastic tunneling mechanism. Coupling between tunneling electrons and tip-induced plasmon modes which gives rise to the intense photon emission observed is discussed within this model.     

X-ray and UV photoelectron spectroscopy study of Na-halogen-graphite intercalation compounds. Comparison between donor- and acceptor-graphite compounds

New results obtained with XPS and UPS on ternary Na-halogen-graphite intercalation compounds (GICs) complement previous ones obtained on binary alkali- and ternary alkali-oxygen-GICs yet published, to give a general overview about what can be learned from photo-electron spectroscopy. In addition to surface stoichiometry, the charge transfer, the Fermi level energy shift and the band structure are estimated thanks to the observation in UP spectra of the unoccupied Π^∗ level peak.     

Local structure of porous silicon studied by means of X-ray emission spectroscopy

2,3 X-ray emission spectra of porous silicon (P-Si) and of spark-processed silicon (sp-Si). Both types of Si-structure display strong photoluminescence in the visible range of the spectrum. Porous samples were prepared by anodization of n^-- and p⁺-Si-wafers. Whereas for the P-Si processed from p⁺-Si the presence of some amorphous silicon is detected, the X-ray emission spectra of porous Si prepared from n^--Si display a higher content of SiO₂. For spark-processed Si the Si L_2,3 X-ray emission spectra reveal a much stronger degree of oxidation which extends to depths larger than 10000 Å. Furthermore, the chemical state of silicon atoms of sp-Si measured at the center of the processed area is close to that of silicon dioxide, and it has an influence on the photoluminescence energy. Specifically, green photoluminescent sp-Si shows a higher degree of oxidation than the blue luminescent specimen. However, the depth of oxidation consistently decreases in areas with weak or no PL. Possible origins of the observed photoluminescence are discussed. Accepted: 6 March 1997     

Photoelectron spectroscopy investigation of defects on LiNbO3 surfaces

Results of UPS and XPS measurements performed on LiNbO₃ single crystals reduced by Ar⁺-ion bombardment are presented. The influence of reduction dose on the type and concentration of induced defects is studied. It is shown that Nb ions change their valency state from 5+to 4+for low reduction doses and to 3+and 2+for high reduction doses. Further it is shown that Nb⁴⁺ ions can be considered as bulk defects which manifest themselves in a broad emission band in the gap centred at 1.6 eV below the Fermi level and that defect ions Nb³⁺ and Nb²⁺ manifesting themselves in a narrow emission band at 3.9 eV below the Fermi level are present only in the outermost layers of the surface. An attempt is made to compare photoelectron spectra with optical spectra of LiNbO₃ and correctness of this procedure is discussed.     

The electronic structure of KMnO4 investigated by photoemission and electron-energy-loss spectroscopy

From photoemission and electron-energy-loss data the following picture of KMnO₄, with Mn^VII (with a formal charge state Mn⁷⁺ (3d ⁰)) tetrahedrally surrounded by four O^2–-ions, is deduced: strong covalent bonding between Mn^VII and O^2– leads to a considerable occupation of the Mn-3 d shell. The ground state of the (MnO₄)^–1 molecule is an orbital and spin singlet as seen by the absence of any multiplet splitting in the Mn core levels. The valence band shows a four peak structure extending form 4 eV to 8 eV below the Fermi energy. The first peak at 4.2 eV has mainly O-2p character. The remaining peaks are of strongly mixed Mn-3d/O-2p character due to the covalent bonding. This mixing decreases with increasing binding energy. The electron energy loss data show a variety of structures between 2 eV and 10 eV independent of the primary electron energy which defines them as dipole allowed charge-transfer transitions. An additional excitation at 1.8 eV decreases quickly in intensity with increasing electron energy which classifies it as a dipole or spin forbidden transition in the compound. This energy is close to the value of 1.6 eV reported for the activation energy observed in electrical transport data. The results are compared to quantum chemical molecular orbital calculations of the (MnO₄)^–1 molecule.Physics Department, Allahabad University Allahabad 211002, India     

Band transitions in wurtzite GaN and InN determined by valence electron energy loss spectroscopy

Valence electron energy loss spectroscopy (VEELS) was applied to determine band transitions in wurtzite InN, deposited by molecular beam epitaxy on (0001) sapphire substrates or GaN buffer layers. The GaN buffer layer was used as VEELS reference. At room temperature a band transition for wurtzite InN was found at (1.7±0.2 eV) and for wurtzite GaN at (3.3±0.2 eV) that are ascribed to the fundamental bandgap. Additional band transitions could be identified at higher and lower energy losses. The latter may be related to transitions involving defect bands. In InN, neither oxygen related crystal phases nor indium metal clusters were observed in the areas of the epilayers investigated by VEELS. Consequently, the obtained results mainly describe the properties of the InN host crystal.     

Low energy excitations in KMnO4 with 1 eV<=ΔE<=10 eV from electron-energy loss spectroscopy

Electron-energy loss spectra of potassium permanganate (KMnO₄) with primary electron energies 25 eV<=E₀<=500 eV show 7 peaks in the energy-loss range 1 eV<=ΔE<=10eV and are successfully analysed with a superposition of 7 independent Gaussians. The intensity of these lines follow roughly a power-law dependence on the primary energy I∝E ₀ ^-α . There are two groups of lines, the first with an exponent α≈0.5, while the lines in the second group decay much stronger with increasingE ₀ corresponding to a value 0.9<=α<=1.3. The 4 lines in the first group are identified as dipole allowed transitions by comparison to recent first principle molecular-orbital calculations for the (MnO₄) molecule by H. Nakai et al. The dipole-allowed excitation spectrum obtained from this analysis agrees very well with these first principle calculations.     

Verification of CuInSe2 density of states features by reflection electron energy loss spectroscopy

Reflection electron energy loss measurements of CuInSe₂ single crystals are reported and when possible, the losses are compared with former results by Kazmerski and Jamjoum et al. Their identification of the transitions is confirmed and refined by spectrum simulations with total and partial densities of states from band structure calculations of Jaffe, Bendt, Zunger, and Bullett.New conduction-band maxima at energies not covered by the available calculations, are proposed and their consistency with the experimental loss energies is shown by another simulation with Lorentzian density peaks.     

Deconvolution of appearance potential spectra II

We have recently introduced a new method for the inversion of autoconvolution integrals [1]. In this paper, we describe an improvement of the previous algorithm which results from the consequent use of cubical spline functions. The treatment given here is mathematically more consistent and offers better numerical stability than that given earlier. The method is applied to the deconvolution of the rather narrow NiL _III appearance potential line. Problems of stability and error propagation are discussed in some detail.     

Determination of the entropy-factor of the gold donor level in silicon by resistivity and DLTS measurements

The parameters of the gold donor level inp-type silicon are determined from the temperature dependence of the free carrier concentration and from DLTS measurements. The entropy-factorX _T is determined to be 20±2. In addition, the capture cross-section for holes _p =5.5×10^–15 cm² and the reaction enthalpyH _T –H _v =0.35eV for the exchange of holes between the gold donor level and the valence band are obtained in the present investigation.     

The results of study of the anisotropy of free-electrons momenta distribution in Mg,Zn, and Cd by means of the positron annihilation method

The angular correlation of annihilation radiation was measured for single crystals of Mg, Zn, and Cd in seven crystallographic direction. The results of distribution of thez-component of the conduction electron momenta were compared with the crossectional-area distribution of the double Brillouin zone for two directions [0001] and . The results were analyzed also by the Mijnarend's method. The distributions of the electron momenta densities in the [0001] and [ ] directions are preliminarily discussed. Supported by Instytut Niskich Temperatur i Badan Strukturalnych Polskiej Akademii Nauk we Wrocławiu (Institute for Low Temperature and Structure Research, Polish Academy of Sciences in Wroclaw, Poland).     

Electron-hole pair production and the structure of ultraviolet isochromats

Isochromat spectra are generally considered to reflect the density of empty one-electron states in the anode material of the x-ray tube employed. Comparison of various measurements with theoretically determined density-of-states functions indicate that such a first-order picture applies only immediately above threshold. In this paper, we show that electron-hole pair production contributes considerably at higher energies above threshold. Taking this effect into account removes most of the existing discrepancy between theory and experiment in this region. Results are presented for the 5d transition metals hafnium, tantalum, and tungsten.     

Band structure calculations for Heusler phase Co2YBi and half-Heusler phase CoYBi (Y=Mn, Cr)

We have studied the electron structure and magnetic properties of Heusler phase Co₂YBi and half-Heusler phase CoYBi (Y=Mn, Cr) by using the full-potential linearized-augmented plane-wave (FLAPW) method. Co₂MnBi and Co₂CrBi are predicted to be half-metallic magnetism with a total magnetic moment of 6 and 5 μ_B, respectively, well consistent with the Slater-Pauling rule. We also predict CoMnBi to be half-metallic magnetism with a slight compression. The gap origin for Co₂MnBi and Co₂CrBi is due to the 3d electron splitting of Mn (Cr) and Co atoms, and the gap width depends on Co electron splitting. The atom coordination surroundings have a great influence on the electron structure, and consequently the Y site in the X₂YZ structure has a more remarkable electron splitting than the X site due to the more symmetric surroundings. The investigation regarding the lattice constant dependence of magnetic moment shows that the Co magnetic moment exhibits an opposite behavior with the change of the lattice constant for Heusler and half-Heusler alloys, consequently leading to the different variation trends for total magnetic moment. The variation of total and atom magnetic moment versus lattice constant can be explained by the extent of 3d electron splitting and localization of Mn (Cr) and Co atoms for both the series of alloys.     

Site Preference and Alloying Effect of Excess Ni in Ni-Mn-Ga Shape Memory Alloys

The formation energies and electronic structures of Ni-rich Ni-Mn-Ga alloys have been investigated by firstprinciples calculations using the pseudopotential plane wave method based on density functional theory. The results show that the alloying Ni prefers to occupy the Mn site directly in Ni9Mn3Ga4 and to occupy the Mn site and drive the displaced Mn atom to the Ga site in NigMn4Ga3, which is in accordance with the experimental result. According to the lattice constants and the density of states analyses, these site preference behaviours are closely related to the smaller lattice distortion and the lower-energy electronic structure when the excess Ni occupies the Mn site. The effect of Ni alloying on martensitic transformation is discussed and the enhancement of martensitic transformation temperature by Ni alloying is estimated by the calculated formation energy difference between austenite and martensite phases.     

Deconvolution of appearance potential spectra

Soft x-ray appearance potential spectra from solids carry information on the local density of unoccupied states in the surface region of the sample. The intensity of the emitted characteristic radiation is, according to a simple model, proportional to a convolution product of the transition densities of the incoming electron and the ionized core electron. In a first approximation, the two transition densities are proportional to the density of states in the unoccupied part of the conduction band. Attempts to retrieve the density of states from measured spectra must rely on deconvolution procedures. In this paper, we introduce a new simple method. It is essentially a stabilized version of the Hagstrum-Becker method which appears to work quickly and reliably. As an example, the method is applied to an ironL _III appearance potential spectru. Comparison of the results with a theoretical calculation of the density of states and with measured Bremsstrahlungsisochromats obtained by Turtle and Liefeld strongly supports the self-convolution model in this case.     

A comparison between X-ray absorption spectroscopy and Bremsstrahlung Isochromat Spectroscopy: The empty states of Pd−Al alloys and Pd2Si

We present X-ray absorption spectra (XAS) obtained with synchrotron radiaton at the PdL _2,3 edges of Pd–Al alloys and Bremsstrahlung Isochromat Spectra (BIS) at 1,486.7 eV of Pd–Al alloys and of Pd₂Si. The XAS and BIS results for the alloys are very similar indicating that the effect of the core potential (e.g. many-body, electronhole excitation) is negligible. The BIS results on Pd₂Si show differences with respect to Pd–Al due to the more localized bonds. The comparison between XAS and BIS data in Pd₂Si is still an open problem and might indicate a more subtle role of the core-hole potential.     

Impedance spectroscopy of a poly(p-phenylenevinylene)-based light-emitting diode device

A poly-p-phenylenevinylene (PPV)-based light-emitting diode (LED) device was studied by means of comprehensive ac impedance measurements. Separate regions of frequency-dependent responses, the bulk and depleted (interface) regions, with distinct electrical properties, were identified and characterized. The ac J-E characteristics revealed the space-charge limited conduction mechanism, which was specifically assigned to the depleted region. Besides, a direct estimate of the trap density in the bulk region was obtained from the applied voltage dependence of the charge concentration.     

Density and well width dependences of the effective mass of two-dimensional holes in (100) GaAs quantum wells measured using cyclotron resonance at microwave frequencies

Cyclotron resonance at microwave frequencies is used to measure the band mass (m_b) of the two-dimensional holes (2DHs) in carbon-doped (100) GaAs/ Al_xGa_1−xAs heterostructures. The measured m_b shows strong dependences on both the 2DH density (p) and the GaAs quantum well width (W). For a fixed W, in the density range (0.4×10¹¹ to 1.1×10¹¹ cm⁻²) studied here, m_b increases with p, consistently with previous studies of the 2DHs on the (311)A surface. For a fixed , m_b increases from 0.22m_e at to 0.50m_e at , and saturates around 0.51m_e for .