Table of contents

Strains in multivalley semiconductors can destroy the strict equivalence of the valleys that is demanded by cubic symmetry. Significant changes in the properties of a semiconductor may result. A proposed implementation of quantum computing with donor atoms in silicon would suffer from alterations of the donor wave functions caused by strains that are produced by fabrication processes. Deliberately straining the silicon to an extent that removed all but one valley from participation in the lowest donor state, would prevent further changes in the wave function by strain. The strain required can be achieved with established technology for depositing silicon on SiGe alloys.

We show that a combination of photoresist patterning and self-assembly of polymer spheres can be used to generate ordered arrays of silicon nanopillars in selected surface regions. The size and shape of patterns in the resist layer are shown to regulate the nucleation and growth of ordered monolayers of spheres, which are employed as templates for metal deposition and reactive ion etching. The method may be useful for applications in photonics requiring structural order on multiple length scales.

The Preisach model formalism has been applied to analyse hysteresis measurement results for evaluating fatigue damage in Fe-C alloys caused by cyclic fatigue loading. Hysteresis loops and differential permeability curves were measured at various stages of the fatigue life of the samples. The parameters which were built by means of the PMF and the classical hysteresis magnetic parameters (such as saturation magnetization, coercive field and others) were studied as a function of the fatigue lifetime. The present results show that the Preisach model analysis can be used to improve the sensitivity of magnetic hysteresis measurements for non-destructive evaluation of the accumulation of fatigue damage in steel components.

Magnetic properties and microstructure of melt-spun ribbons of RE₆₀Fe₃₀Al₁₀ alloys with RE{} = {}Nd, Sm, Dy, Gd and Y were studied in detail. High coercivity values in the range of MA m^-1 were observed at low temperatures for amorphous ribbons. Presence of Fe-rich clusters and nanoscale rare-earth crystallites in the amorphous matrix in the ribbons were revealed by high-resolution transmission electron microscopy studies. The magnetic transition temperatures were estimated experimentally and compared with fitting results based on the cluster ferromagnetism model (Wang L et al 2001 Phys. Rev. B 64 214410). Possible mechanisms for the magnetic behaviour observed due to the presence of Fe-rich magnetic clusters are discussed.

The porous nanosized poly-crystal LiNbO₃ film on the fused SiO₂ substrate was prepared by using r.f. sputtering method. The microstructure and the third-order optical nonlinearity of the film were investigated. The size of the micro-crystal grains was in the range of 110-160 nm, and the average pore size was 80 nm. The magnitude of ultrafast response χ⁽³⁾ of the sputtered thin LiNbO₃ film at 830 nm wavelength was estimated to be as large as 1.9×10^-8 esu, which is 4 orders of magnitude larger than that of LiNbO₃ single crystal.

Spray pyrolysis was used to prepare polycrystalline thin films of undoped Ga₂O₃, as well as those doped separately with Eu³⁺, Tb³⁺ and Tm³⁺. The energy gap values of the as-grown films and those annealed at 600\r{}C and 900\r{}C were found to be 4.75, 4.48 and 4.44 eV, respectively. Films containing Eu³⁺ and Tb³⁺ exhibited red and green cathodoluminescence (CL), respectively. The CL of Ga₂O₃ : Tm mainly showed a broad band in the blue-green region, which resulted from the emission by both the Ga₂O₃ host and Tm³⁺. The broad blue-green emission band was divided into three Gaussian peaks at 424 nm (2.92 eV), 497 nm (2.49 eV) and 526 nm (2.36 eV). The CL intensity of undoped Ga₂O₃ thin films depended on the annealing ambient and temperature, suggesting that it is associated with the presence of oxygen vacancies. Mechanisms responsible for the broad blue-green emission of undoped Ga₂O₃ thin films were explored.

A non-stationary reactive gas dynamics model in a mono-dimensional geometry, including radial mass diffusion, gas temperature variation and chemical kinetics, is developed in this paper. The aim is to analyse the spatio-temporal evolution of the main neutral species involved in a corona discharge used for NO pollution control in polluted air at atmospheric pressure and ambient temperature. The present reactive gas dynamics model takes into account 16 neutral chemical species (including certain metastable species) reacting following 110 selected chemical reactions. The initial concentration of each neutral species is obtained from a 1.5D electrical discharge model. The gas temperature variations are due to direct Joule heating during the discharge phase, and also result from the delayed heating due to the relaxation of the vibrational energy into a random thermal energy during the post-discharge phase. The simulation conditions are those of an existing experimental setup (anode voltage of 10 kV in the case of a point to plane geometry with an interelectrode distance of 10 mm). The obtained results show that the diffusion phenomena and the gas temperature rise affect quite well the gas reactivity and the neutral species evolution. This allows us to better understand the different reaction processes and transport phenomena affecting the NO concentration magnitude inside the discharge channel.

Plasmas were produced from polymers like polystyrene when excited in air or nitrogen by a Nd : YAG 1064 nm intense pulsed laser, and the dominant emission peaks were in the ultraviolet ranging from about 384 to 389 nm, emitted from CN radicals in the plasmas. The intensity of these ultraviolet emission peaks was found to change approximately with the cube of the pumping laser power, suggesting a three-photon induced emission process.

We have applied optical emission spectroscopy together with computerized tomography (CT) to study the transition between E and H modes in inductively coupled plasma. Measurements were performed in pure Ar and in CF₄(5%)/Ar mixture in the pressure range between 5 and 300 mTorr. It was found that the radial and azimuthal CT images of the E (capacitively coupled) mode are quite different from those of the H (inductively coupled) mode. The E mode structures show clearly the presence of capacitive coupling, while the structure of the H mode corresponds to that of the predicted induced fields. The differences diminish at low pressures due to the non-local transport of electrons. The contributions of direct and stepwise excitation are separated by using two different energy levels, 2p₁ and 2p₉, of Ar. In addition, the measurements obtained in electronegative gas mixture with large losses of electrons of moderate and small energy show more clearly the kinetics of excitation in the two modes and the regions where high-energy electrons are produced.

In the technique of plasma immersion ion implantation, plasma ions are implanted into a target upon pulsed charging to high negative voltages. Using hybrid particle-in-cell computer simulation, we study the effects of pulse frequency and duty cycle on the ion flux onto the target. We find that incomplete plasma recovery between the pulses affects the ion current maxima occurring at the beginning of the high-voltage pulses; these strongly decrease for small pulse-off times. Also, for large duty cycles, the implantation current becomes dominated by the saturation current of the steady-state sheath.

A two-dimensional hydrodynamic model for an N₂ inductively coupled thermal plasma (ICTP or thermal ICP) at atmospheric pressure was developed using reaction rates without the chemical equilibrium (CE) assumption. Particle composition distribution in the N₂ ICTP was derived by solving the mass conservation equations for each of the particles, considering diffusion, convection and production terms. The electrical conductivity, mass density and diffusion coefficient were calculated at each of the calculation steps with the derived particle composition distributions. Using this model, the influence of gas flow rate on chemical composition distribution was investigated. The dependence of mass flow of N atom on gas flow rate was obtained. From the result, a large deviation from CE in the distribution of the particle composition was found, especially near the wall of the ICTP.

Thin films of amorphous Se_90-xSb₁₀In_x and GeSe have been prepared by electron beam evaporation technique. The effect of irradiation of GeSe and Se_90-xSb₁₀In_x (x = 0, 5, 10, 15) thin films on the optical energy gap and dc activation energy has been investigated for different doses of ¹²C⁵⁺ and ¹⁰⁷Ag¹⁴⁺, respectively, at 10¹² and 10¹³ ions cm^-2. The optical band gap of GeSe decreases from 1.70 to 1.43 eV on irradiation with a dose of 10¹³ ions cm^-2. Similar change was observed in Se_90-xSb₁₀In_x thin films. The value of dc activation energy also decreased with the increase in radiation dose. The results are explained by considering the energy deposited due to the electronic energy loss upon irradiation with heavy ions, which modifies the properties of the material.

Polytetrafluoroethylene films have been deposited for use as low dielectric constant materials. Deposition is performed through pyrolysis of hexafluoropropylene oxide (HFPO) to produce CF₂, which can then polymerize and deposit as a thin film. The variation of CF₂ concentration as a function of reactor conditions has been characterized by ultraviolet absorption spectroscopy. CF₂ concentration is observed to go through a maximum with respect to both pressure and pyrolysis temperature when it is present in large amounts (~10¹⁴ cm^-3). A one-dimensional model including known kinetic reactions for HFPO decomposition and CF₂ recombination and multi-component diffusive transport has been applied to the parallel plate system. The result is seen to overestimate the measured concentration and does not capture the maxima observed versus pressure and temperature. An additional mechanism of particle formation, by CF₂ insertion into (CF₂)_n oligomers, has been introduced to produce a kinetic model that explains the CF₂ concentration measurements.

This work presents a theoretical calculation of the I-V characteristics of an n-p junction. The total current across the n-p junction is presented as the superposition of currents due to each region of the junction. It results in ideality factor values between 1 and 2. Threshold voltage values, related to the low and high level injection in each region, are introduced. The theoretical model is used to describe the experimental emitter-base I-V characteristics of a bipolar transistor. The effects of the electrical ageing are taken into account as a creation of a defect layer in the base, near the junction. A maximum error of 6% between the theoretical model and experiment is obtained.

The effect of post-growth annealing and magnetic field on the two-way shape memory effect (TWSME) of Ni₅₂Mn₂₄Ga₂₄ single crystals was investigated. We found that samples subjected to a two-step post-growth annealing can provide a better TWSME than that of directly quenching from the parent phase. By way of comparison, the martensitic microstructures of samples, subjected to two different post-annealing treatments, were studied by means of optical and electron microscopy. Moreover, we found that the magnitude and direction of TWSME can be changed optionally by selecting the strength and direction of the magnetic field. The work output by the sample on a fixed magnetic field of 1.2 T was calculated to be 12 kJ m^-3, which is comparable to that of Terfenol-D actuators, being of order 10 kJ m^-3reported by Clark et al (1988 J. Appl. Phys.63 3910).