Table of contents

We report on the first use of undoped bulk-grown p-type GaSb as a low-temperature detector in composite bolometers for the detection of non-equilibrium phonons. In the ohmic regime, the detector's low-temperature resistivity measurements are consistent with the Mott variable-range hopping conduction characterized by ln R(T)∝exp (T₀/T)^1/4 with a temperature coefficient of (-0.42-3.46ln T)^-1 K^-1 in the temperature range 1.4 K⩽T⩽4.2 K. The detector showed a fast response time of approximately 1 µs and excellent reproducibility. In our opinion, the most promising areas of applications are in nuclear spectroscopy (low-energy x-ray and beta-particle detection) for which small-volume detectors can be used and the study of phonon dynamics in 2DEG structures.

The influence of temperature from 298 K to 353 K on the electrical conduction of low-density polyethylene is reported. Conduction current and differential scanning calorimetry (DSC) measurements show a correlation between the structural change due to the temperature treatment and the change in the activation energy of the conduction mechanism involved. When the temperature is greater than 318 K, an activation energy decrease from 0.94 eV to 0.68 eV and a DSC exothermic peak are observed. This temperature threshold is correlated with a facilitated detrapping of charges stored during polarization.

The analysis of the degradation mechanisms that affect solid-state devices represents a key factor in the era of VLSI technology. This paper presents an overview of how the low-frequency noise measurement (LFNM) technique has been applied to the study of the most frequent causes of failure in integrated circuits and discrete components. After discussing noise and fluctuations in electron devices, we present some results obtained by applying LFNM to the analysis of some degradation mechanisms. Although several other causes of degradation are taken into consideration (ageing, radiation, hot electrons), the focus is on two specific failure mechanisms: electromigration in interconnect lines and ultra-thin oxide breakdown in MOS capacitors. Experience has highlighted the need for dedicated measurement systems, thus an entire section of the paper outlines both the instrumentation required and the reduction of the effects of spurious noise sources and external disturbances.

We report the influence of stiffness of carbon nanotubes for probes of a scanning probe microscope on images. Multiwalled carbon-nanotube probes are fabricated by manipulation under the direct view of a scanning electron microscope. Using this manipulation, it is also revealed that a Hamaker constant of 60×10^-20 J for the van der Waals attraction is for a sidewall of the nanotube and the metallic surface at a vacuum of ~10^-3 Pa. The force curve measurements at a steep slope in air reveal the influence of the force acting not only on the tip of the probes but also on the wall of the tip. The origin of this effect is discussed in terms of the van der Waals attraction and the adhesive energy estimated from the force curve. This phenomenon is suppressed using a nanotube probe consisting of bundled nanotubes at the base to improve the stiffness for samples of high roughness.

X-ray phase-contrast topography has been improved by using phase modulation and computing techniques to determine lattice distortions inside near-perfect Si monocrystals. The technique presented is based on the detection of the spatially varying phase shift at the output of a monolithic x-ray interferometer of the triple Laue type using phase modulation and a multi-element detector. Information was obtained about the degree of perfection of crystals being used in the determination of the (220) lattice plane spacing of silicon and of the Avogadro constant.

TEM studies have been performed on 50 nm thick monolayer films of sol-gel derived anatase titania (TiO₂). It is found that a dense polycrystalline structure, having low porosity and consisting of nanocrystallites ranging from 5-20 nm in diameter, is the result of a sintering process involving temperature ramping up to 550 °C. Ohmic conduction is believed to be due to almost flat band conditions caused by partial depletion of carriers from the crystallites.

Silicon nanoparticles embedded in the host matrix of an Al₂O₃ thin film, with a thickness of about 260 nm, were prepared by using the pulsed laser deposition method. The laser target consisted of a small silicon wafer glued onto the surface of a circular Al₂O₃ plate, which was set to rotate uniformly when laser-ablating. TEM and EDS results showed that the films consisted of silicon nanoparticles in the form of nanocrystals with diameters of less than 6 nm dispersed in the amorphous Al₂O₃ matrices. A secondary ion mass spectroscopy depth profile of the silicon content of the film indicated that the silicon nanoparticles in the Al₂O₃ matrices were arranged in fairly well demarcated thin layers sandwiched between layers of the host material. FTIR results evidently suggested that the encapsulating amorphous Al₂O₃ material forms a good host for the prevention of atmospheric oxidation of the silicon nanoparticles. However, prolonged annealing can cause the silicon nanoparticle surface interfacing the host material to be oxidized by the oxygen mostly from the Al₂O₃ molecules and to be bonded to the Al or O-Al bonds.

The current-voltage characteristics of Si/SiO₂ superlattices are numerically computed as a function of a number of physical parameters such as the number of Si wells, the Si and SiO₂ film thickness, the temperature, etc. It is assumed that the Si layers form two-dimensional systems, communicating to one another through tunnel SiO₂ barriers, without quantum coherence between one Si well and its next nearest neighbours. From the numerical results, we derive a number of simple conclusions regarding the optimization of such confined structures, in view of maximizing their electroluminescence efficiency.

We report on improved structural, crystallographic and electrical properties of epitaxial NdBa₂Cu₃O_7-δ (NBCO) films grown on SrTiO₃ by `off-axis' pulsed laser deposition (PLD). Transport and XRD studies show that the c-axis-oriented epitaxial films, with critical temperatures of 90-92 K, are mono phase and single-crystalline. Furthermore, very smooth, almost outgrowth-free surfaces and crystal coherences of up to 0.8 µm (to our knowledge the best value ever reported for high-T_c films) were obtained.

By comparing the photoluminescence intensity of a p-type crystal with that of a n-type crystal, we find that the p-side is the main light-emitting region in the double-n structure nitrogen-doped GaP (GaP:N) material grown by liquid phase epitaxy (LPE). From this experimental result, we have created a simple theory regarding the minority-carrier-current distribution in the lower donor concentration layer in the double n-structure GaP:N material. The theory describes that the interface of the lower, n^-, and the higher, n⁺, donor concentration LPE n-layers has a reflection effect on the holes diffusing from p- to n-type crystal. When we decrease the thickness of the n^--layer and the majority-carrier concentration ratio of the n^-- and n⁺-layers, the electron injection efficiency increases and the photoluminescence is correspondingly enhanced.

The structural modification upon milling in an intermediate step precursor of the sol-gel method for BaFe₁₂O₁₉ hexaferrite BaM production is discussed. The milling of the precursor diminishes the powder particle size, leads to a more homogeneous matrix and induces a solid-state transformation, from γ-Fe₂O₃ to α-Fe₂O₃. The induced modifications of the precursors change the magnetic and structural properties of the final BaM hexaferrite compared to the BaM obtained from the non-milled precursor.

The effect of temperature-dependent viscosity on the heat transfer over a continuous moving surface is studied. The boundary layer equations are transformed to ordinary differential equations containing a Prandtl number and a viscosity/temperature parameter. The velocity profiles, temperature profiles, the skin friction coefficient and the rate of heat transfer are computed and discussed in detail for various values of the viscosity/temperature parameter and Prandtl number.

This paper studies in experiment and theory the breakdown of argon, nitrogen, air and oxygen in a uniform dc electric field at different discharge gaps L, discharge tube radii R and cathode materials. At arbitrary geometric dimensions of the cylindrical discharge vessel and cathode materials the ratio of the breakdown electric field value to the gas pressure p at the minimum of the breakdown curves is shown to remain constant, (E_dc/p)_min≈constant. A modified breakdown law for the low-pressure dc discharge U_dc = f(pL,L/R) is obtained. That is, the breakdown voltage U_dc is shown to depend not only on the product pL, but also on the ratio L/R. A method is presented enabling one to predict a dc breakdown curve in the cylindrical discharge vessel possessing arbitrary L and R values from the measured data on the dc breakdown.

A new silicon hydride clustering model was developed to study the nucleation of particles in a low-temperature silane plasma. The model contains neutral silanes, silylenes, silenes and silyl radicals as well as silyl and silylene anions. Reaction rates were estimated from available data. Simulations were carried out for typical discharge parameters in a capacitive plasma. It was shown that the main pathway leading to silicon hydride clustering was governed by anion-neutral reactions. SiH₂ radical insertion was found to be important only in the initial stages of clustering, whereas electron-induced dissociations were seen to lead to dehydrogenation. Increased ion density (radiofrequency power density) leads to faster clustering due to increased formation of reactive radicals.

This paper deals with the theoretical analysis of plasma phenomena in the keyhole during laser welding. The mathematical model based on the kinetic equation for optical discharge is presented. The influence of the beam intensity and the keyhole radius on the plasma properties is shown. The temperatures of the plasma electrons, delivered by the model, correspond to the experimental data.

We report a newly observed phenomenon in a dusty plasma device of the \mbox{Q-machine} type. At low plasma densities the time required by the plasma to return to its no-dust conditions, after the dust dispenser is turned off, can be as long as many tens of seconds or longer. A tentative interpretation of this observation in terms of magnetized dust grains is advanced. It appears that an important loss mechanism of fine dust grains is by ion drag along the magnetic field lines. The effect of ion drag is somewhat counteracted by the -µ∇B force present when the magnetic field has a mirror geometry.

After current zero in a high-voltage SF₆ circuit breaker, the recovery voltage heats the remaining electrons, producing departures from thermal equilibrium that may modify the electrical conductivity of the medium and thus the interruption capability of the apparatus. In this study we present a two-temperature hydrodynamic model coupled with a kinetic model, applied to the extinction of a blown SF₆ arc in a two-dimensional simplified geometry. The free decay and the post-arc phase with the recovery voltage have been studied. The kinetic part of the modelling allows us to calculate the evolution of the electron number density. A simplified turbulence model has been added to study its influence on the post-arc phase.

The main results of this study reinforce the conclusions obtained taking into account only the chemical non-equilibrium: the departures from equilibrium tend to decrease the interruption capability predicted by the model. This effect is due to the weak energy exchange by elastic collisions between the electrons and the heavy particles, leading to an electron temperature higher than the gas temperature which favours the ionization of the medium submitted to a recovery voltage.

The phenol degradation processes by pulsed corona discharges are investigated under three kinds of discharge atmosphere (air, argon and oxygen). The temporal variations of the concentrations of phenol and the intermediate products are monitored by LIF spectroscopy. The species of the intermediate products are identified by spectral analysis. It is clarified that the oxidative gaseous reagents produced from O₂ and those from H₂O degrade phenol to intermediate products with comparable degradation rates. The degradation via the reagents from H₂O gives rise to the formation of molecules exhibiting fluorescence at 400-500 nm, in addition to dihydroxybenzene (DHB), while the degradation via the reagents from O₂ produces only DHB. The reagents from O₂ play an important role in the conversion of phenol to DHB.

Development of bioactive glasses for use as a coating on Ti6Al4V prostheses requires a better understanding of reactions at the bone/bioactive glass interface. Indeed, the bioactive glasses bond to bone through physico-chemical reactions. In vivo, an apatite rich layer is built up on top of a pure silica rich layer at the bioactive glass periphery. In this paper, we have studied Ti6Al4V cylinders coated with a bioactive glass and implanted in sheep femora for two, three and six months. At each time period, the samples were analysed with scanning transmission electron microscopy coupled with energy dispersive x-ray spectroscopy. In vivo, the bioactive glass dissolution led to the formation on its surface of spherical particles with different sizes. The distributions of Si, Al, Ca, P and Mg concentrations across the particles reveal precipitation of apatite with the incorporation of magnesium. Apatite precipitation is governed by diffusion through an Si layer and occurs under specific supersaturation conditions. Measurements of supersaturation for Ca and P demonstrate that the largest precipitates grow and the smallest dissolve. These results allow us to study the growth and dissolution rate of the apatite precipitates and their relevance to bioactivity. Particles with a radius twice the average radius (⟨R⟩) grow the fastest and, if the radius increases, the rate of growth decreases. Before three months, the growth of apatite precipitates (⟨R⟩≈1 µm) leads to the growth of a Ca-P interfacial layer. After three months, ⟨R⟩ is of the order of 0.5 µm, and the majority of the apatite layer dissolves. The effects of aluminium and magnesium on apatite generation are also studied.

Potential formation near a negative charge collecting substrate in a negative ion plasma was studied experimentally. We present the first observation of a potential double layer formed in front of a positively biased substrate in pure SF₆ discharge plasma. With the increase of SF₆ gas in Ar gas, the plasma potential reduced to negative potential, which is lower than the wall potential. The energy distribution function of the negative charges in SF₆ plasma predicted the presence of negative ions near the sheath edge and its disappearance at a position close to the substrate. This result implies that negative ions are kept inside the bulk plasma by the sheath electric field while few of them can reach the surface of the substrate.

Contributions of krypton ions and krypton metastables, as well as contributions of gamma-radiation and cosmic rays to the electrical breakdown in a krypton-filled tube have been analysed by use of the time delay method. The experimental results have shown that each of these contributions strictly depends on the afterglow period. The time delay method has enabled the determination of both the recombination time of positive ions and the lifetime of metastables, created in previous discharge. The obtained value of metastable lifetime is in agreement with its theoretical value. The dependence of the electrical breakdown on exposed dose rate of gamma-rays from the radiation source has also been investigated.

The rise probabilities of positive corona streamers initiated by a flux of negative ions are investigated under the combined voltage (dc + ac) in point-to-coaxial cylinder discharge gaps in atmospheric air. A method is developed to detect and to estimate quantitatively an after-effect of the preceding streamer on the rise probability of the following one during the next cycle of the ac voltage. The method is based on the statistical analysis of sequences of streamers. The dependence of the intensity of the after-effect on dc and ac voltage, ac frequency, pre-ionization level, air humidity and airflow velocity is investigated. The causes of the after-effect of streamers are discussed.

A collisional radiative model is presented for mercury discharges with electron temperatures between 0.75-2 eV and electron densities between 10¹⁸-10²⁰ m^-3. Such plasma parameters are encountered in a number of modern light sources, such as mercury-operated induction lamps and the compact fluorescent lamp. The analytical top model has been used, which allows the majority of the non-equilibrium levels to be taken into account implicitly. As a result, indirect ionization processes involving highly excited atomic mercury states are taken into account in spite of the relatively low number of levels (19) which has been considered.

The influence of higher atomic mercury levels on the ionization rate coefficient has been carefully analysed, and has been found to contribute significantly. Furthermore, a consistent means of quantifying the production of radiation by the plasma will be presented by introducing the specific effective emissivities. These enable one to express the total radiated power in terms of the densities of the transport-dominated states and the electron density and temperature. These coefficients, as well as the net coefficients of ionization and recombination, will be presented and discussed, enabling their usage in plasma transport models.

The aim of this paper is to study the effects of the magnitude and direction of air flow rate on the characteristics of the corona discharge generated in a wire-duct reactor. Two air flow directions were attempted: one parallel to, and the other normal to the discharge wire in the reactor. The corona current-voltage characteristics correlative with ozone generation were studied under different modes of applied voltage. The results have shown that dc and ac corona current-voltage characteristics depend significantly on the magnitude of the air flow rate in the reactor, whatever the direction of the air flow. The ozone concentration generated by either dc, ac or pulsed coronas with parallel flow is higher than that generated with normal flow with the same discharge conditions. The ozone concentration generated by pulsed coronas is higher than that generated by either dc or ac coronas with the same discharge conditions irrespective of the polarity of the pulsed corona.

Significant problems in deep laser spot welding are the formation of welding defects, and particularly porosities in the case of tantalum joining. In this study, we investigate and model porosity formation and trapping. Two types of porosity are observed. (1) Small porosities are round shape bubbles of less than 250 µm diameter. These defects may come from gas bubbles generated following hydrogen rejection during solidification or they may come from bubbles induced by both an intense evaporation inside the keyhole and a turbulent flow in the molten pool. (2) Large porosities are voids generated in the bottom of the welded zone. They arise from a lack of matter inside the weld as if the molten metal had not had enough time to fill up the keyhole completely before it solidified.

We elaborated a numerical model for the comprehension of small porosities trapping by comparing the calculated spot weld solidification time with the rise time of bubbles in the liquid phase. We also studied the melt flow back into the keyhole in order to explain the formation of the large voids observed. These two models are in good agreement with experimental observations, in particular with x-ray radiographs achieved during the solidification of tantalum spot welds.

We present a method for the determination of the average depth of the activity of Cs-137 in soil, exploiting the information contained in the low-energy part of an in-situ spectrum. The predictive power of the method is evaluated by comparing the results with laboratory measurements of the activity profile and it is shown that the proposed approach yields results of comparable quality in a much shorter measurement time. The method relies on the analysis of the entire spectrum of in-situ gamma-rays from a single spectrum, including the broad peak of scattered photons at the energy between 20 and 80 keV. It is based on a simple notion that the deeper the source is buried beneath the surface, the more numerous the photons scattered in the soil, compared to the corresponding peak of gamma-rays of the original energy. Monte Carlo calculations are necessary for this approach and were based on the GEANT system developed at CERN, Geneva.

This paper reports the conductive and ferroelectric properties of RTP crystals that are important for the fabrication of quasi-phase-matched RTP devices. The frequency dependence of the conductivity is shown to be represented by a Debye-function in the lower frequency range ({<}10 kHz) and to obey a power law in the high frequency range. The spontaneous polarization and coercive field are found to be 0.198±0.005 C m^-2 and 2.65±0.2 kV mm^-1, respectively, by measuring the switching current of RTP under domain reversal. Ferroelectric hysteresis loops are obtained in which the effects of a previously unreported dependence of the ac conductivity on the amplitude of the applied electric field are clearly visible. The domain structure of as-grown crystals and the change in the domain structure under a square waveform of 3 kV mm^-1 are revealed through the use of optimized etching conditions.

Pyroelectric array detectors in 8.0-14.0 µm thermal imaging cameras require a means of area modulation to generate a change in signal intensity on the detector. Currently this is achieved by mechanically `chopping' the image using a rotating blade, which introduces moving parts to the system and has the drawbacks of size and image degradation as camera motion changes the blade's rotational speed. Adequate modulation has been achieved using a novel method in the 8.0-14.0 µm region by introducing moderate levels of excess carriers to suitably prepared germanium by excitation from a diode laser source. The modulation produced is greatly superior to previously published attempts to produce a broad optical bandwidth, polarization-insensitive large-area modulator. The process utilizes intervalence band transitions from the light-hole to heavy-hole band, requiring excitation power densities in the order of W cm<sup>-2</sup>. A depth of modulation of from 95.4 to 4.7% transmission at 10 µm using a total power density of 5.6 W cm<sup>-2</sup> from 980 nm diode lasers has been achieved over a 1 cm<sup>2</sup> aperture. The `on' state loss at this wavelength is limited by the single-layer quarter-wavelength AR coating properties, and may be eliminated through use of a multilayer coating. Techniques used to optimize the induced absorption are detailed, and a carrier diffusion model is presented which is used in temporal transmission decay experiments to extract the surface recombination velocity and bulk carrier lifetime parameters.

The x-ray sensitivity of a high-resistivity photoconductor sandwiched between two parallel plate electrodes and operating under a constant field is analysed by considering charge carrier generation that follows the x-ray photon absorption profile and taking into account both electron and hole trapping phenomena but neglecting recombination, bulk space charge and diffusion effects. The amount of collected charge in the external circuit due to distributed generation of electrons and holes through the detector is calculated by integrating the Hecht collection efficiency with Ramo's theorem across the sample thickness. The results of the model allow the x-ray sensitivity to be calculated as a function of the applied field, detector thickness and electron and hole ranges (µτ), given the field and energy dependence of the electron and hole pair creation energy, W_±, and the energy spectrum of incident x-ray radiation. The sensitivity model was applied to stabilized a-Se that is currently used as a successful x-ray photoconductor in the recently developed flat panel x-ray image detectors. Recent free electron-hole pair creation energy versus electric field data at room temperature and appropriate electron and hole drift mobilities were used to calculate the sensitivity for monoenergetic x-rays at 20 and at 60 keV. For the 20 keV radiation, it was shown that a typical detector thickness of 200 µm (4 × attenuation depth at 20 keV) with currently attainable electron and hole trapping parameters in a-Se was operating optimally, the sensitivity of which can only be increased by further increasing the applied field. With the receiving electrode positively biased, the sensitivity was much more dependent on the hole lifetime than electron lifetime. The absence of hole transport results in a reduction in sensitivity by a factor of about 4.4, whereas the absence of electron transport results in a sensitivity degradation of only 22%. The ratio of hole trapping limited sensitivity to electron trapping limited sensitivity is about 0.3. For a detector of thickness 200 µm operating at 10 V µm^-1, the maximum sensitivity is about 220 pC cm^-2 mR^-1, and this sensitivity degrades by more than 10% when either the electron lifetime falls below ~20 µs or the hole lifetime falls below ~5 µs. When the hole lifetime is very short so that the sensitivity is substantially reduced, the sensitivity versus thickness dependence at a given field exhibits a maximum (an optimal thickness) that is less than that for full absorption. In the case of 60 keV x-ray photons, it is more useful to examine the sensitivity as a function of detector thickness given the practical bias voltage limit. The sensitivity versus thickness behaviour for a given bias voltage exhibits a maximum, that is an optimal thickness, that is less than that for nearly full absorption. Electron lifetimes longer than ~200 µs and hole lifetimes longer than ~10 µs do not significantly affect the sensitivity.

The minus sign in equation 11 of this article was incorrectly placed outside the opening square brackets, instead of between them. The corrected version follows: