Table of contents

We report the effect of ultraviolet annealing at 400 °C of tantalum pentoxide films photo-deposited at 350 °C from tantalum ethoxide and nitrous oxide mixtures using 172 nm radiation. A significant reduction in the leakage current density of the films has been found. This is attributed to the formation of SiO₂ on the surface of the Ta₂O₅ and the reduction of suboxides, oxygen vacancies and defects in the films. We find excellent agreement between the two techniques of x-ray photoelectron spectroscopy and Fourier transform infrared spectroscopy for detection of the unexpected formation of a thin SiO₂ layer on the surface of the Ta₂O₅, which implies that Si atoms outdiffuse from the Si substrate through Ta₂O₅ films into the overlaying Ta₂O₅ film. Effective dielectric constant value of 20 and very low leakage current densities in capacitors incorporating the oxide around 7.7 × 10^-9 A cm^-2 at a voltage of 3 V, compare very favourably with films grown by alternative higher temperature techniques. This low-temperature, atmospheric-pressure UV annealing process offers a simple, but very effective, method for the fabrication of high-quality Ta₂O₅ films with low leakage currents for practical use in high-density dynamic random access memories.

An investigation has been carried out into the possibility of using polarization methods to characterize clouds of irregular particles. Measurements were made at scattering angles between approximately 115° and 135°, where it was expected that polarization effects would have maximum variation with shape. It was found that the angular variation of any polarization property was similar for most particles. Further, as a consequence of reciprocity, the useful parameters were limited to the scattered intensities for two incident polarization states and one cross-polarized intensity. The only useful correlation appeared to be between cross-polarization and the `roundness' of the particles, although different correlations are observed between particles which are inherently isotropic and those which are anisotropic. The main conclusion of the study is that polarization can be used to measure the average roundness of particles in a cloud and has the potential to discriminate between crystals of different structure.

Single phase Ba(Mg_1/3Ta_2/3)O₃ films were prepared on Si (100) substrates and a fused silica substrate by pulsed laser deposition. The structure and annealing properties of these films were studied by the x-ray -2 scan. The wavelength dependence of the transmittance of the films deposited on fused silica was determined. The refractive index was calculated from the transmission spectrum. The quadratic electro-optic effects of the film were successfully measured to be 8.35 × 10^-18 m² V^-2 by the differential detection method at 6328 Å.

The reaction of Si (100) surfaces at T = 950 °C with radicals of methane obtained in a low-power-density glow discharge plasma, has been studied by combining in situ surface science techniques (x-ray photoemission spectroscopy and high electron energy diffraction) and ex situ analytical techniques (atomic force microscopy and infrared absorption). An analysis of C 1s and Si 2p core-level shifts combined with the examination of the valence-band curves showed that the obtained buffer layers were stoichiometric. For long carbonization times (>30 min) the formation of a carbon rich surface was observed. To understand the mechanism of hetero-epitaxial silicon carbide (SiC) buffer layer growth, the early stage of SiC nucleation was observed by atomic force microscopy and reflection high-energy electron diffraction. The results suggest that three-dimensional epitaxial islands nucleate at the earliest growth stage followed by a further Volmer-Weber growth until the formation of a carbon rich surface. The growth mechanism of the SiC buffer layer is discussed on the basis of a reported model.

The development of a thin liquid film under non-uniform rotation and in the presence of transverse magnetic field has been studied numerically by using the finite-difference technique under the assumption of a planar interface. Similarity variables were used to transform the axisymmetric Navier-Stokes equations into a set of coupled, nonlinear, unsteady, partial differential equations. The time-dependent free surface was mapped into a finite fixed computational domain. It is shown that the rate of film thinning slowed down by increasing either the Hartmann or Eckman number. It is also observed that a small change in the Eckman number has a stronger effect in film thinning than that for the Hartmann number. Furthermore, it is found that a faster rate of thinning can be obtain if the spinner starts impulsively and then increases its spinning rate continuously.

We investigated the broadening of spectral lines in a high-pressure zinc discharge. The contributions of resonance, van der Waals and Stark broadening are calculated and compared with broadening constants derived from side-on measurements. The energy balance is solved for a lamp containing about 1 bar metallic zinc and 2.25 bar Ar as a starting rare gas. The axis temperature of the discharge derived from the contours of the spectral lines is 5400 K as compared to model calculations yielding 5370 K. It is shown, that the contribution of radiation diffusion to the thermal conductivity is dominating at plasma temperatures above 5000 K, i.e. at temperatures, where the discharge is assumed to be in local thermal equilibrium. In a similar way as for mercury discharges, broadening of the resonance line at = 213.8 nm (Hg: = 253.6 nm) is dominant with respect to radiation diffusion, but also transitions to the triplet S4P3P and to the singlet S4P1P must be taken into account. The energy balance is solved using the measured line broadening data, resulting in a good agreement of the temperature profile with that derived from side-on measurements.

A newly developed technique for the three-dimensional x-ray reconstruction of binary images is discussed. It allows one to reduce extremely (about 10-100 times) the number of required projections and views and to suppress the main artifacts. The technique is based upon the solution of an imposed operator equation using special types of functionals, which are also discussed. The steps for its solution provide the zero-level approximation, the computer simulation of the radiographic process and the use of the a priori knowledge about structural features in general form. They give the priority to planar and/or volumetric images in the matrix. The efficiency of corresponding functionals and algorithms is compared to the maximum-likelihood (ML) estimation using both simulated and real radiographic data. The introduced prior functionals are discussed in terms of Markov random fields and Gibbs statistics used for Bayesian image reconstruction with higher-order models as priors.

This study explores how to enhance both the power and the efficiency (at optimum power) for the family of reciprocating Stirling-like heat engines (including Carnot) with ideal regeneration and either finite or infinite heat reservoirs. To accomplish this, a generalized mixed-mode, heat-transfer approach is developed and employed at cycle thermal ends. Additionally, the difference in the effect on power optimization of operating with either interactive or non-interactive finite reservoirs is examined.

The approach developed makes use of a unique time symmetry procedure to minimize cycle time. This procedure ensures the proper allocation of the hot- and cold-end, heat-exchanger capacitance. Furthermore, time symmetry ensures the concurrent employment of the first and second laws of thermodynamics. This in turn guarantees the minimization of internal entropy generation and the maximization of specific cycle work (for a given set of operating temperatures).

The general case formulation produces a semi-decoupling of the solution expressions for the upper and lower power-optimized operating temperatures. This semi-decoupling allows the use of various combinations of single and mixed-mode, heat-transfer arrangements at the hot and cold thermal ends of these cycles. Hence, semi-decoupling permits examination of the power and work potentials of different combinations of heat modes under power-optimized conditions. The results point conclusively to the strong preference of using the radiation mode (with or without transfer by the linear mode, but better with as little as possible) in the high-temperature-end, heat-exchange process and of using essentially only linear modes at the low-temperature end of the cycle (i.e. only as small a percentage as possible by radiation). Such cycles are found capable of exceeding the one-half ideal work limit of power-optimized cycles that exclusively employ linear modes at both thermal ends (i.e. W_opt greater than ½W_rev). Also, the efficiency at optimum power of these cycles exceeds the Curzon and Ahlborn efficiency (i.e. _opt greater than 1-(T_L/T_H)^0.5). No other combination of heat-transfer modes is found capable of exceeding either the Curzon and Ahlborn efficiency or the ½W_rev limit for power-optimized cycle work for such cycles.

Finally, comparing the effects of interactive and non-interactive heat reservoirs on power optimization shows that the heat-exchanger operation (outlet temperatures) should also be included in the overall optimization procedure.

A new method is proposed for producing ¹⁸F, a positron emitter, via ¹⁸O(p,n)¹⁸F reactions with fast protons from the interaction of a relativistically-intense short-pulse laser with an underdense plasma layer. The fast ion concentration contributing to the nuclear reaction is estimated on the basis of a two-dimensional particle-in-cell simulation. The instantaneous production rate of ¹⁸F is found to be two orders of magnitude larger than by the standard method using a cyclotron.

The problem of the steady heat transfer in a hollow cylinder subject to intense heating on its walls is studied and solved analytically. The heat supplied to the inner cylinder wall is supposed to be generated by the action of a laser beam on a gas circulating inside the hollow cylinder, acting as the cathode of an arc-welding device. The inverse problem is also solved and analysed for a variety of heat flux profiles. The analytical solution is compared with former numerical approaches of the inverse problem.

Prebreakdown electron emission between extended metal electrodes is experimentally studied with time resolution of 100 ns. Switching to higher emission levels is observed and could be provoked by mechanical shocks. This is found to be analogous to spontaneous switching effects. Elementary time constants of current jumps are 1-3 µs while the whole switching process lasts up to 40 ms, thus ruling out impact of macroparticles. The breakdown voltage is found to be correlated with the emission level. Therefore, switching on electron emission can explain breakdowns with delay times >1 ms. Still longer times are found at the deposition of metal vapour plasma originating from neighbouring arc gaps. In some cases, transient increases of the emission occur that last several seconds. Therefore, in vacuum interrupters late breakdowns after current zero can be caused by mechanical shocks without particle impact and also by deposition of residual vapour at the cathode.

Prebreakdown electron emission between extended copper electrodes is experimentally studied with a time resolution <100 ns. When taking the current-voltage characteristics sufficiently quickly it is found that after a voltage increment the emission can relax to a lower level with time constants <1 ms. As a consequence, Fowler-Nordheim plots become flat, yielding values of the microscopic field enhancement that are too high. During the deposition of a metal-vapour plasma that originates from neighbouring arc gaps, a steady growth of the emission is observed, which leading to a considerable reduction of the voltage, down to 50% of the initial value. The results are discussed in terms of metallic-field electron emission from atomic structures that are subjected to surface migration under the influence of the electric field. The migration processes can be activated by mechanical shocks. This explains the shock effects found in the preceding paper in this issue.

Plasma temperature and electron density have been measured by optical emission spectroscopy in a transferred direct current arc with graphite electrodes. The measurements were realized for the arc running in a 1.7 cm³ s^-1 argon flow at atmospheric pressure, a current of 200 A, a voltage of 48±2 V and with a distance of 85 mm between the electrodes. Specially designed electrodes were developed to minimize displacements of the arc because of movements of the electrode spots. In this way, the measurement errors have been greatly diminished. The measurements were made supposing an optically thin collisionally dominated plasma of axial symmetry. Density distributions of excited argon atoms and ions on their energy levels were determined to evaluate plasma temperature. The half-widths of spectral lines were used to reach the electron density in the plasma. Spatial distributions of the measured parameters are presented for the arc positive column and near the electrodes. Maximum values of 1.5 × 10¹⁷ cm^-3 and 13 500 K were found close to the cathode for the electron density and temperature, respectively. The data and possible reasons for some of the variation found in the observed results are discussed.

In this paper, we experimentally investigated an effect of streamer polarity on the reduction of NO_x by using a barrier discharge plasma reactor with a multipoint-to-plane geometry. A high-frequency sine wave voltage generator and a repetitive pulse generator, which generated narrow and high-voltage pulses were used to remove nitrogen oxide (NO_x). By the optical spectrum measurement, it is clarified that electrons with high energy, more than 10 eV, are produced in the discharge region at atmospheric pressure. The intensity ratio of the first negative system of N₂⁺ (391.4 nm, threshold energy 18 eV) to the second positive system of N₂ (337.1 nm, threshold energy 11 eV) neither depends on the applied voltage nor the frequency, but it depends on streamer polarity. Further, it is shown that the value is higher in a positive streamer than in a negative streamer. The use of a multipoint-to-plane electrodes caused the streamer polarity interchange every half cycle of the sine wave of the applied voltage, while in a case of the parallel plane electrodes, the positive streamer was only generated in the discharge region. The simulated gas and the exhaust gas from a diesel engine generator were used to confirm the NO_x removal performance of the reactor.

The transport characteristics of electrons and holes in an alternating block copolymer, poly(1,8-octanedioxy-2,6-dimethoxy-1,4-phenylene-1,2-ethenylene-1,4-phenylene-1,2-ethenylene-1,4-phenylene-1,2-ethenylene-3,5-dimethoxy-1,4-phenylene), have been investigated by current-voltage measurements in metal/polymer/metal sandwich structures. Strong evidence that the hole current is space-charge limited with a single discrete set of shallow traps and the electron current is space-charge limited with an exponential distribution of traps was obtained. A hole trap density of ~1 × 10¹⁷ cm^-3 and a hole trap energy of ~0.1 eV relative to the highest occupied molecular orbital level were determined. The characteristic energy of the trap distribution, ~0.05 eV, and the trap density <3.7 × 10¹⁷ cm^-3 for electrons were also estimated. It was found that the hole current became trap-free, space-charge limited, when using Au, Pd or Ni (ϕ_Au≈ϕ_Pd≈5.1 eV) instead of Cu as the anode, indicating that the hole trap energy must lie betweenϕ_Cu(≈4.65 eV) and ϕ_Au. A hole mobility of ~10^-6 cm² V^-1 s^-1 in this copolymer was determined.

The electro-physical surface properties of polymer composites of ultrahigh-molecular-weight polyethylene containing 5, 10 and 15 vol% carbon black filler were investigated before and after irradiation by gamma-rays of a ⁶⁰Co source. Measurements of the specific surface conductivity _s, surface dissipation factor tan_s, volume dissipation factor tan_v, and the maximum surface potential U_max after charging with positive or negative ions by dc corona discharge, were carried out at room temperature in air of 65% humidity.

U_max was found to have a strong dependence on irradiation dose especially at doses ranging up to 1 Mrad. Also, _s exhibits a substantial decrease of up to seven orders of magnitude at doses from 100 krad to 1 Mrad. These effects are supposedly linked with the radiationally stimulated destruction of carbon chains on the polymer composite surface. On the other hand, at doses higher than 1-2 Mrad there is an extraordinary high rise of the specific surface conductivity of the samples containing 5 vol% filler which reaches nine orders of magnitude and more. This is attributed to the formation of chains of carbon nanoclusters in the near-to-surface layer of the polymer composite.

A probe beam deflection technique was utilized to measure the propagation of a shock wave and material vapour plume generated during excimer laser ablation of aluminium samples. The measured transit time of the laser-induced shock wave was compared with the prediction based on an ideal blast-wave model, using the Sedov-Taylor solution. The prediction of the incident laser energy converted into the laser-induced gasdynamic flow utilizing this blast-wave model overestimated the efficiency, even under conditions when the measured shock-wave velocity follows the correct model relation. The propagation of material vapour was measured from the deflection of the probe beam at later times. The propagation velocity of material vapour ranged from 20-40 m s^-1 with a greater velocity near the target surface.

The correct title of this paper is as above, with the word 'contact' and not 'contract' as was published originally.