Table of contents

Current semiconductor technology, the so-called the molecule reaction based semiconductor manufacturing, now faces a very severe standstill due to the drastic increase of gate leakage currents and drain leakage currents. Radical reaction based semiconductor manufacturing has been developed to completely overcome the current standstill by introducing microwave excited high density plasma with very low electron temperatures and without accompanying charge-up damage.

The introduction of radical reaction based semiconductor manufacturing has made it possible to fabricate LSI devices on any crystal orientation Si substrate surface as well as (100) Si substrate surfaces, and to eliminate a very severe limitation to the antenna ratio in the circuit layout patterns, which is strictly limited to less than 100–200 in order to obtain a relatively high production yield.

We analyse the precessional magnetization switching of high perpendicular anisotropy nanostructures biased with an in-plane exchange field. Our calculation is made in the limit of non-damped macrospin ferromagnetic media. From the Landau–Lifshitz–Gilbert equation, we first derive the magnetic trajectories. Then, we give the analytical expressions of the time-dependent magnetization, the precession frequency and the switching time for any in-plane applied field pulse orientation. We show that the exchange bias field decreases significantly both the switching time and the reversal field. We find that the switching time is slightly dependent on the applied field orientation. Typical switching times less than 100 ps are predicted for a Co/Pt multilayer.

Thin films of magnesium doped SrRuO₃ (Mg-SRO) have been successfully prepared by laser ablation on SrTiO₃ (100) substrates. The stoichiometry, structural, electrical and magnetic properties of the films depend on the substrate temperature (T_s) and oxygen pressure ( ) during deposition. All Mg-SRO films are ferromagnetic, but the transport properties and remanent moments vary with T_s and . Metallic films are produced for T_s ⩾ 700 °C whereas insulating films can be produced with T_s = 500 °C with . A correlation between lattice parameter, stoichiometry and transport properties has been discovered: the larger lattice parameter correlates with the decreased Ru content and more insulating samples. Insulating samples prepared at T_s = 500 °C and have stoichiometry close to the SrMg_0.15Ru_0.85O₃ and are ferromagnetic Anderson insulators with transport properties consistent with variable range hopping.

In this study we report the effect of sintering temperature on the low field magnetotransport properties of La_0.7Ca_0.3MnO₃ manganite synthesized through the polymeric precursor route. The La_0.7Ca_0.3MnO₃ has been sintered at 600 °C (S6), 700 °C (S7), 800 °C (S8), 900 °C (S9) and 1000 °C (S10). X-ray diffraction confirms that phase formation starts at 600 °C. All the samples are single phasic having an orthorhombic unit cell. The lattice parameters decrease on lowering the sintering temperature (T_S). The crystallite as well as the grain size also show strong dependence on the sintering temperature. All the samples possess characteristic insulator–metal (T_IM) as well as paramagnetic–ferromagnetic (PM–FM) (T_C) transitions. The T_C varies in a small range [272–256 K] as a function of sintering temperature whereas the T_IM goes down from 267 K (S10) to 138 K (S6), a strong decrease of 129 K. This T_C–T_IM discrepancy is due to the fact that whereas the former is an intrinsic characteristic, the latter depends strongly on the extrinsic factors e.g. synthesis conditions, grain boundaries and associated disorders. For all the samples magnetoresistance (MR) shows strong dependence on T_S. The MR increases on lowering the temperature as well as on increasing the field with the occurrence of an intrinsic contribution around T_C. These variations of MR for all the samples have been explained in terms of the microstructural variations and spin-polarized tunnelling at low temperatures.

Several near-stoichiometric lithium niobate (LiNbO₃) plates were prepared by the vapour transport equilibrium technique. Domain reversal was carried out on these samples by electric field poling. A nonlinear dependence of switching field on the concentration of anti-site niobium ( ) ions was observed. This nonlinear relationship was explained by a two-dimensional model based on the dynamics of the domain wall. A parameter of θ_c, denoting the critical position of the domain wall round ions, was introduced to describe domain reversal. The domain wall energy per unit area for near-stoichiometric lithium niobate was also roughly calculated to be larger than 0.27 J m⁻².

Homogeneous single-phase Cu(In_0.75Ga_0.25)(Se_1−yS_y)₂ alloys were prepared by the reaction of magnetron sputtered CuIn_0.75Ga_0.25 precursors to a H₂Se/Ar and H₂S/Ar gaseous atmosphere under well-defined experimental conditions. X-ray diffraction analysis of the films showed characteristic chalcopyrite peaks with a high degree of symmetry, indicative of homogeneous rather than compositionally graded material. Glancing incident angle x-ray diffraction revealed virtually no variation in the lattice parameters through the entire depth of the alloys. X-ray photoelectron spectroscopy depth profiling confirmed the homogeneous distributions of the respective elements through the entire depth of the pentenary semiconductor alloy under optimized selenization/sulfurization conditions.

The accuracy of the commercially used, simplified scalar optical effective frequency method (EFM) is verified by comparing its results with those obtained using more accurate vectorial models represented by the method of lines (MoL). Both EFM and MoL methods are applied to the modern telecommunication oriented GaAs-based oxide-confined vertical-cavity surface-emitting diode laser (VCSEL) emitting 1.3 µm radiation from the InGaAsN/GaAs double quantum well active region. Although both methods are applied to the specific laser structure, the conclusion remains valid for a broad range of different VCSEL designs. The scalar EFM approach has been found to remain surprisingly often quite exact; nevertheless, it fails for structures violating its fundamental assumption of plane wave propagation within the laser cavity. Therefore, a noteworthy discrepancy between modes calculated with the aid of scalar and vectorial approaches occurs for weaker guidance, i.e. for higher-order modes, smaller aperture diameters, thinner oxidation layers and/or apertures shifted from the antinode position. These cases need an exclusive use of the more exact but also more time-consuming vectorial algorithms.

High birefringent liquid crystals are attractive for photorefractive applications but device operation, usually restricted to the Raman–Nath regime, can be degraded by small phase shifts between the optical and refractive index gratings and coarse grating spacings with narrow beam intersection angles. Applying external electric fields and having to tilt the cell at an angle to the grating k-vector cause further complications. In this paper, two-beam coupling in hybrid photorefractive cells comprising a nematic liquid crystal layer between inorganic photorefractive windows is described. Photorefractive properties are determined by the windows, while the liquid crystal amplifies the overall refractive index modulation. Bragg matched liquid crystal gain coefficients >1600 cm⁻¹, grating periods <300 nm and wide beam intersection angles with perfect 90° phase shifts between the optical and refractive index gratings at normal incidence without the need for an external field are demonstrated.

ZnO single-crystal microtubes were fabricated by an optical thermal evaporation method. The microtubes showed hexagonal hollow morphology with well faceted side surfaces, with cross-sectional dimensions of 20–100 µm and lengths of several millimetres. Air flow is found to be the critical experimental parameter for the formation of ZnO microtubes with different cross-sections. Under the excitation of 325 nm, photoluminescence of the microtubes was obtained with a peak wavelength of 492 nm.

We report on a novel phenomenon occurring when ultrashort laser pulses are diffracted in a Bragg grating recorded in a photo-thermo-refractive glass: two intense beams at the third harmonic of the incident laser frequency appear between the direct and the diffracted laser beams. A high conversion efficiency of about 0.04% at an incident intensity of 5 × 10¹¹ W cm⁻² in a 1 mm thick grating is attributed to self-phase matching for the third harmonic generation via Cherenkov radiation from the interfering third order polarization of the incident and diffracted beams. It is assumed that this is a general phenomenon for diffraction in a volume grating.

Transmission spectra of TiO₂ films grown by atomic layer deposition on fused silica have been analysed by using Lorentz dispersion and a model consisting of two sublayers inside a film. It has been shown that the deposition process parameters significantly influenced the refractive index gradient in the film growth direction. The films grown at a lower flow of the carrier gas showed a 35 nm thick sublayer with low refractive index at the silica substrate and a layer with a higher refractive index at the film surface. The films deposited at higher carrier gas flow had a higher refractive index at the substrate and a lower refractive index at the film surface. The results also demonstrate that by using a parameter modelling one can obtain some information about the internal structure of a film even if there are no clearly defined interference fringes in a transmission spectrum.

The effect of a tunnel charge transport in the near-surface region of silicon on the electrical characteristics of MOS structures with a 2–3 nm insulator layer is studied theoretically. An equilibrium condition for the substrate is assumed. The cases of an Al and polySi gate are considered. The possibility of a 'double' (in Si and through SiO₂) tunnelling expands the energy range of transported particles, which increases one of the components of the total tunnel current. The proposed model allows for the improved simulation of gate current in MOSFETs, which is especially important for highly-doped substrates.

Theoretical and experimental investigations of the temporal behaviour of the column plasma of square-wave-pulsed low-pressure glow discharges in mixtures of helium and xenon are presented. In the framework of the time-dependent, radially averaged model the equations for the external electrical circuit accounting for an approximate treatment of the electrode regions, rate equations for 13 xenon and 3 helium states and the time-dependent Boltzmann equations of the electrons are self-consistently solved. The experimental studies comprise time-resolved measurements of the electrical characteristics and of the axis densities of the lowest excited states of xenon. The main aspects of the self-consistent model and of the electrical and spectroscopic measurements are given and the results for the He–Xe mixtures containing 2% of xenon in a discharge tube with a diameter of 17.5 mm and an electrode distance of 20 cm at applied voltages between 140 and 250 V and gas pressures between 1 and 4 Torr are presented. The agreement between experimental and theoretical data is generally good during the entire pulse period. The analysis shows that the velocity distribution function of the electrons is subject to large structural alterations as a result of the periodic change between the on-phase of the discharge, driven by the electric field, and the off-phase when chemo-ionization process gets large importance. However, the chemo-ionization processes are not efficient enough to act as a reservoir of charged particles under the conditions considered. Concerning the excited xenon atoms it is found, in particular, that their temporal behaviour is significantly influenced by their mixing due to electron collision processes for optically forbidden transitions in the afterglow period.

Laser-produced Sn plasmas are at present a major contender in the challenge to find a suitable replacement for the currently used excimer-laser technology, which has wavelengths of 248 and 193 nm, and that is utilized in projection lithography. These wavelengths are to be superseded by soft x-ray sources in the 13.5 nm wavelength regime for utilization in extreme ultraviolet lithographic (EUVL) technologies. To date, considerable international efforts have been channelled into the experimental realization and optimization of various tin based EUV sources. Therefore, in order to compliment these experimental accomplishments we have undertaken a spatio-temporal study of the free electron number density, atomic number density, average charge state and expansion kinetic energy of Sn and SnO₂ plasmas. This has been achieved by coupling the collisional radiative equations to the one-dimensional Lagrangian fluid dynamic model MED103 (MEDUSA), thus obtaining the spatial and temporal histories of the aforementioned variables within a laser-produced plasma of spherical geometry, generated using a Gaussian laser pulse at 1064 nm. The evolution of ion stages Sn⁴⁺ to Sn¹³⁺ within a fluid cell is also presented. In addition, the dependence of the Sn fractional ion populations upon the atomic number density within variable composition plasmas of binary mixtures formed from Sn and oxygen and Sn combined with samarium is investigated. The overwhelming influence of both the atomic and free electron number densities within these plasmas is highlighted.

Mass spectrometry (MS) and optical spectroscopy have been used to study the residual gas formed during hydrogen plasma debinding of polypropylene from powder injection moulded parts. Mass spectra show that the residual gas formed during the debinding process presents a large number of aliphatic compounds and a certain difficulty in their identification. However, it was possible to identify the presence of propylene, methane, acetylene and ethane. Optical spectroscopy also shows the presence of CH radicals. The end point of the debinding process can easily be monitored by following the temporal evolution of the CH radical and the different ions by optical spectroscopy and MS, respectively.

The present paper shows that in the case of a micro-discharge modelling using the hydrodynamics assumption, the second order fluid model involving the complete electron momentum conservation equation must be used in order to better quantify the radical formation in a micro-discharge applied to air pollution control. The present results show large differences in the micro-discharge parameters (such as velocity and electron density) between the three tested hydrodynamics models: the classical first order model using the local electric field approximation and two second order models using the local energy approximation with or without the drift–diffusion approximation. The tests have been carried out in the case of a wire-to-plane corona reactor filled with a typical flue gas (76% N₂, 12% CO₂, 6% O₂, 6% H₂O) at atmospheric pressure and ambient temperature. The simulation of the micro-discharge dynamics is performed using a 1.5D numerical streamer model coupled with a simple chemical kinetics model involving 31 species (charged and neutral particles in their fundamental or metastable state) reacting following 29 selected chemical reactions.

In this paper, the transportation of nitrogen atoms in a flowing post-discharge of molecular nitrogen at atmospheric pressure is studied. The axial variation of the density of nitrogen atoms is modelled and compared with measurements. A relatively high initial density, i.e. several 10¹⁴ atoms/cm³, is obtained by this approach, showing also that the main losses are caused by volume and surface recombination during transportation. A low value is found for the recombination probability of the used polyamide surface (around some 10⁻⁴) allowing therefore quite long transportation times and distances.

The synchronous generation of the surface streamer discharges from two electrodes in air is investigated in terms of wavelength and intensity of the discharge light as a trigger for the synchronization of the surface streamers. It is found that the light emitted from the first streamer causes the photoelectron emission and the photoionization in the vicinity of the opposing electrode, triggering synchronous discharge of the second streamer. The wavelength of the light contributing to this phenomenon is shorter than 250 nm, with a remarkable rise in synchronization probability from 164 nm to 115 nm. This is considered to be due to the existence of the photoabsorption windows for oxygen in this vacuum ultraviolet region. These results suggest that the photon energy required to supply initial electrons sufficient to trigger the surface streamers on an acrylic resin insulator is at least 5 eV and more enhancement is found at 8 and 10 eV in this study.

Electrostatic hybrid resonance oscillation modes are investigated in semi-bounded magnetized dusty plasmas. The surface wave dispersion relations are obtained for the orientation of the magnetic field using the specular reflection boundary condition with the plasma dielectric function. The result shows the existence of upper- and lower-hybrid resonance oscillation modes in semi-bounded magnetized dusty plasmas. It is found that the frequencies for the electrostatic upper- and lower-hybrid resonance oscillations are independent of the orientation of the applied magnetic field. In addition, the hybrid resonance frequencies are found to decrease as we increase the ratio of the electron plasma frequency to the electron cyclotron frequency.

The possibility that the pink afterglow (PA) of a flowing nitrogen discharge occurs as a result of recombination of N(⁴S) atoms is evaluated and discussed, based on a detailed kinetic model for a microwave discharge and post-discharge. The present simulation shows that the states responsible for the emission of the PA cannot be created via an indirect mechanism initiated with atomic recombination. Alternatively, it is indicated that the PA may have its origin in non-resonant vibration–vibration energy-exchange processes between molecules, which lead to an overpopulation of high levels of the vibrational manifold.

The identity and the energy distributions of positive and negative ions electrostatically extracted from the liquid phase in an ionic liquid ion source (ILIS) are analysed with a time-of-flight mass spectrometer and a multi-grid retarding potential analyzer. Accurate energy measurements using ionic liquids in an externally wetted configuration are reported for the first time. Droplet-free beams are produced using the ionic liquid 1-ethyl-3-methylimidazolium bis(triflouromethylsulfonyl)amide (EMI-Im) in which the solvated ions (EMI-Im)_nEMI⁺ and (EMI-Im)_nIm⁻ with n = 0,1,2 are observed. The small ion source size and the energy distribution widths and deficits of a few electronvolts are quite similar to those of liquid metal ion sources, confirming that ILIS can be used in applications requiring highly focusable beams, e.g. sub-micron ion lithography. Measurements also suggest that solvated ions with n ⩾ 1 exhibit post-extraction fragmentation into lighter species at a rate increasing with their original degree of solvation. About 10% of the total beam current is carried away by metastable species that break up almost immediately after extraction while inside the emitter accelerating region.

The two-dimensional hydrodynamics of ion extraction from a non-uniform quasi-neutral plasma, in an electrostatic field, has been simulated numerically. Experimentally, tunable pulsed lasers with a spatially Gaussian profile produce a non-uniform plasma through stepwise photo-excitation and photo-ionization or multi-photo-ionization processes. For such a non-uniform plasma, the ions were assumed to be initially super-Gaussian with a finite gradient density profile at the boundaries. Poisson's equation was solved simultaneously with the equations of mass and momentum, assuming the Maxwell–Boltzmann distribution for electrons. In these calculations the dynamics of the electric potential, ion density and velocity, and ion current density were assessed and the extraction time was estimated. Three separate regions of the plasma, i.e. the ion sheath, the transition region (pre-sheath) and the quasi-neutral plasma were considered in these calculations. The time history of the pre-sheath front as well as the ion current density across the cathode was also examined. The results show that while the extraction time remains constant for both the super-Gaussian and the uniform plasma, the pre-sheath velocity, ion current density and the ion density are significantly different.

The discharge conditions required for low-temperature plasma sterilization were investigated using low-pressure surface-wave plasma (SWP). The discharge conditions for both continuous wave (CW) and pulse-modulated SWPs in low-temperature sterilization of Geobacillus stearothermophilus with a population of 1.5 × 10⁶ and 3.0 × 10⁶ were studied by varying the microwave input power from 500 W to 3 kW, and the effective plasma treatment time from 40 to 300 s. Results showed that sterilization was possible in a shorter treatment time using a higher microwave power for both CW and pulse-modulated SWPs. Pulse-modulated SWPs gave effective sterilization at a temperature roughly 10 to 20 °C below that of CW SWPs under the same average microwave power.

A two-dimensional numerical model of an atmospheric pressure glow (APG) discharge was developed. The simulation of the APG in helium with some nitrogen impurities successfully reproduced the discharge evolution during the breakdown process observed in experiments. The results show that the breakdown first appears at the central region of the discharge followed by the axial and radial propagation of the ionization wave. The space charge induced electric field is the main reason for the radial propagation. Penning ionization is the dominant ionization mechanism in the discharge. It leads to an elevated pre-ionization level before the discharge pulse that is essential for maintaining a uniform discharge. Increasing the driving frequency favours a transition from a filamentary to a uniform glow discharge. An increasing number of filaments with driving frequency are observed. The coalescence of the filaments at sufficiently high frequency leads to a uniform discharge. Decreasing the dielectric permittivity of the barriers limits the discharge current and leads to a shift of the discharge towards a uniform Townsend discharge.

The stress in a single-layer continuous deposition of amorphous silicon dioxide (SiO₂) film is compared with the stress within multiple-layer intermittent or 'stop–start' depositions. The films were deposited by helicon activated reactive evaporation (plasma assisted deposition with electron beam evaporation source) to a 1 µm total film thickness. The relationships for stress as a function of film thickness for single, two, four and eight layer depositions have been obtained by employing the substrate curvature technique on a post-deposition etch-back of the SiO₂ film. At film thicknesses of less than 300 nm, the stress–thickness relationships clearly show an increase in stress in the multiple-layer samples compared with the relationship for the single-layer film. By comparison, there is little variation in the film stress between the samples when it is measured at 1 µm film thickness. Localized variations in stress were not observed in the regions where the 'stop–start' depositions occurred. The experimental results are interpreted as a possible indication of the presence of unstable, strained Si–O–Si bonds in the amorphous SiO₂ film. It is proposed that the subsequent introduction of a 'stop–start' deposition process places additional strain on these bonds to affect the film structure. The experimental stress–thickness relationships were reproduced independently by assuming a linear relationship between the measured bow and film thickness. The constants of the linear model are interpreted as an indication of the density of the amorphous film structure.

A dispersively aligned carbon nanotube array has been achieved between metal electrodes by electric-field assisted manipulation of surface modified single-wall carbon nanotubes (SWCNTs). Surface modification of SWCNTs with octadecylamine (ODA) molecules creates well-dispersed SWCNT solutions and effectively obviates the entanglement between deposited SWCNTs. The alignment effect of SWCNTs is improved by aligning them with an ac electric field of high frequency in a high-volatility solvent. The good electrical contact between the aligned SWCNTs and the metal electrodes is obtained by adopting a UV-light radiation method for removing the grafted ODA on the aligned SWCNTs.

The paper is aimed at the fractal analysis of the creeping discharge patterns propagating over different solid insulators immersed in mineral oil, under negative impulse voltage using a point-plane electrode arrangement. The considered solid insulators are circular samples of different thickness made of polycarbonate, phenoplast resin (Bakelite) and glass. Two methods are used to estimate the fractal dimension, namely the box counting method and the fractal measure relations method. It is shown that the creeping discharges propagate radially; the shape and the length of these discharges depend on the types of solid insulators and their thicknesses. By using the box counting method, we show that the discharge patterns present a fractal dimension D which depend on the thickness of the solid samples (e) and the type of insulator. D decreases when e increases. This suggests the possible implication of capacitive effects on the propagation phenomena of creeping discharges. So D increases with the dielectric constant of the insulator; it is the highest for glass and the lowest for polycarbonate; D is in between for phenoplast resin. This dependency of D on the solid insulators and their thicknesses reveals the existence of a relation between the fractal dimension and the physical parameters. The fractal measure relations method gives a fractal dimension equal to the Euclidian dimension (D = 2) due to the fact that the density of branches ρ(r) remains constant whatever the radius r of the circular domain within which ρ(r) is determined. ρ(r) depends on the thickness and the nature of solid insulating material. The value of the density of branches increases when the thickness of the solid insulating material decreases.

The structural, electrical and optical properties of RF sputtered In₂O₃ : Sn (ITO) thin films and the effect of post-deposition annealing have been studied. The thickness ranges from 225 to 862 nm. X-ray diffraction, scanning electron microscopy (SEM) and atomic force microscopy (AFM) experiments were performed to study the structure and the surface morphology of these samples. We found that thinner films have a ⟨100⟩ texture and as the film grows the preferred orientation changes from ⟨100⟩ to ⟨111⟩. The lattice parameters are found to be larger than the bulk value, indicating that the samples are under a tensile stress. The grain size increases with increasing thickness. SEM images show a dense granular structure with grains having different shapes and sizes. From AFM images, the average surface roughness (rms) was estimated to be 3.89 nm. The energy gap was found to decrease from 3.65 to 3.50 eV as t increases from 225 to 866 nm. Annealing experiments were done, in the air, at temperature T in the 100–500 °C range. We found that the ⟨111⟩ texture becomes stronger after the annealing treatment. A large increase of the grain size with increasing T is observed. The lattice constant decreases with T to become closer to the bulk value, i.e. annealing seems to relieve the stress present in the as-deposited films; T = 400 °C seems to be the best temperature to obtain practically a stress free sample. We observe a large decrease in the electrical resistivity ρ after annealing. The lowest ρ value (16 × 10⁻⁴ Ω cm) was noted in the 699 nm thick sample annealed at 500 °C. The decrease of ρ seems to be the consequence of a larger grain size and a stronger ⟨111⟩ texture.

The dc and ac conductivity of polycrystalline SeSm_0.005 bulk samples have been measured under vacuum in the temperature range 363–93 K. The samples displayed dielectric dispersion in the frequency range 50 Hz–80 kHz. The calculated values of the exponent s of the function σ = Aω^s showed that correlated barrier hopping is the suitable model to describe the ac conduction mechanism. Calculation of the real dielectric constant (ε'), loss factor (ε'') and loss tangent (tan δ) are given in the studied frequency and temperature ranges. The loss factor displayed a loss peak that gives direct evidence of the existence of a Debye relaxation type. Also, the arc shape of Cole–Cole diagrams has been used to determine and discuss the optical (ε_∞) and static (ε_s) dielectric constants besides the macroscopic relaxation (τ₀) and molecular relaxation (τ) times.

The surface morphology and electrical conductivity of poly(3-octylthiophene) (P3OT) films, in both their pristine and doped states, have been studied by scanning electron microscopy and by measuring the conductivity at room temperature and the variation of conductivity with temperature in the range 10–300 K. Pristine P3OT film exhibits a mat-type structure whereas ferric chloride doped P3OT film shows conducting domains in the range 40–80 nm. The room temperature dc conductivities of pristine and doped P3OT films are ∼1 × 10⁻⁸ S cm⁻¹ and 8.2 × 10⁻⁴ S cm⁻¹, respectively. The temperature dependence of dc conductivity in the region 77–300 K, where hopping conduction dominates, is well described by Mott's three-dimensional variable range hopping transport.

It is shown that the deviation from the ideal Hollomon relation in describing the stress–strain behaviour is characteristic of all materials at low strains. The Ludwigson relation describing the deviation from the Hollomon relation at low strains is critically analysed and it is shown that the deviation at low strains is a consequence of some unknown 'plastic strain equivalent' present in the material. Stress strain curves obeying an ideal Hollomon relation as well as that of a structurally modified (prior cold worked) material were simulated and compared. The results show that the yield strength and the flow strength of a material at constant strain rate and temperature are dictated by the magnitude of the 'plastic strain equivalent' term. It is shown that this component need not necessarily mean a prior plastic strain present in the material due to prior cold work alone and that prior cold work strain will add to this. If this component is identified, the stress–strain behaviour can be adequately described by the Swift relation. It is shown that in both formalisms, the strain hardening index is a function of the yield strength of the material.

We have studied the electromagnetic wave propagation in a two-dimensional anisotropic metamaterial (AMM) realized by a periodic structure composed of an anisotropically loaded transmission line (TL) network. The dispersion relation of the TL metamaterial is analysed by rigorous periodic TL theory, which shows hyperbolic dispersion surface in phase space below a certain critical frequency. We have constructed an interface between an isotropic normal medium and an AMM with a properly designed normal TL mesh and the loaded TL network, respectively, and analysed the electromagnetic wave propagation in such a system using microwave circuit simulations. We directly demonstrate the negative refraction of energy flow and the positive refraction of the wave vector, as well as the partial focusing phenomenon in such AMM, have good agreement with the theoretical prediction. When the dissipation is included in the simulation on actual microstrip line structure, we find that the dissipation in the TL metamaterial affects only the magnitude of the propagating energy, not the refraction phenomenon.

A randomly mixed model is developed for the prediction of the effective thermal conductivity of a multi-phase system. The proposed model is based on the assumption that the smallest part of the phases is a cube, and all the cubes are randomly dispersed in the space. The effective thermal conductivity therefore can be found numerically from thermal conductivities and volume fractions of the components, using the principle of heat conduction in anisotropic media. The prediction does not depend upon empirical parameters and the algorithm is easy to perform in a personal computer. The validation of the proposed model is tested by several types of moist porous media with various porosities and degrees of saturation. Compared with the experimental data of the soils and building materials, the proposed model can give a fine prediction of the moist porous media for porosity less than 0.6. Finally, the deviations between the predicted and experimental results for high porosity are also analysed.

The main objective of this paper is to study the electric field distributions around a standard post insulator for the power frequency, lightning impulse (1.2/50 µs) and switching impulse (250/2500 µs) voltages under icing conditions, and these were computed numerically using the finite element method. Thin glaze coating on the insulator requires a very large number of elements for finite element analysis because of the open boundary around the ice-covered insulator. To reduce the number of elements and hence computation time, the region between the domain of interest and infinity was modelled by a form of Kelvin transformation. The simulation results, confirmed by laboratory experiments, show that while a lightning impulse is the limiting factor in the design of a resistive glazed insulator under clean conditions, a switching impulse is the limiting factor under icing conditions.

Nonlinear optical properties of colloidal Au and poly(vinylpyrrolidone) (PVP) composite films were investigated using the Z-scan technique with 25 ps pulses at 532 nm near the surface plasmon resonance frequency of Au nanoparticles. Experimental results show large optical nonlinearities of the composite films with the real and imaginary parts of the third-order nonlinear optical susceptibility χ⁽³⁾ being 4.7 × 10⁻¹⁰ esu and 2.5 × 10⁻¹⁰ esu, respectively. The obtained χ⁽³⁾ value of Au/PVP composite films is comparable or higher than that of the organometallic complexes reported. Furthermore, a strong optical limiting response was observed and a possible application of the composite films as an optical limiter of picosecond laser pulses was discussed.