Table of contents

This paper describes a new method for fabricating a gas sensor composed of multi-wall carbon nanotubes (MWCNTs) using dielectrophoresis (DEP). MWCNTs dispersed in ethanol were trapped and enriched in an interdigitated microelectrode gap under the action of a positive DEP force that drove the MWCNTs to a higher electric field region. During the trapping of MWCNTs, the electrode impedance varied as the number of MWCNTs bridging the electrode gap increased. After the DEP process, the ethanol was evaporated and the microelectrode retaining the MWCNTs was exposed to ammonia (NH₃) gas while the electrode impedance was monitored. It was found that the electrode impedance was altered by ppm-levels of ammonia at room temperature. The ammonia exposure decreased the sensor conductance, while the capacitance increased. The sensor showed a reversible response with a time constant of a few minutes. The conductance change was proportional to ammonia concentration below 10 ppm and then gradually saturated at higher concentrations. Effects of the number of trapped MWCNTs on sensor response were also discussed.

The optical properties of polycrystalline copper subjected to tensile stress are monitored in situ and in real time using reflection anisotropy spectroscopy (RAS). It is shown that RAS allows investigation of the plastic regime. Here, in contrast to the Hooke's law regime, the stress-induced RAS lineshape is found to be dependent on the applied stress. The optical anisotropy in the visible region of the electromagnetic spectrum is directly proportional to the mechanical strain. An intense RAS peak observed at a photon energy of 4.0 eV is observed to saturate at a stress approximately equal to the yield stress of copper. This work demonstrates the potential of RAS as a nanomechanics monitor of materials under mechanical stress.

Doping ferromagnetic nickel–iron alloys with chromium causes the Curie temperature to be reduced. We have demonstrated that solid state solutions of Ni–Fe–Cr can be formed by implanting Cr ions into ferromagnetic NiFe alloy films, thus creating paramagnetic films. We find that the magnetic moment and coercivity decrease steadily with Cr⁺ dose, reaching zero at room temperature with the onset of paramagnetism. Using Cr⁺ implantation in conjunction with a lithographic mask we have patterned continuous Ni₈₀Fe₂₀ films into separate regions that are ferromagnetic and paramagnetic at room temperature. This magnetic patterning process may have applications for the manufacture of magnetic write heads, such as for the notching process used to constrain the stray field from the write gap.

Silver ions have been diffused into single crystals of β-BaB₂O₄ (commonly referred to as BBO). The crystals were wrapped in silver foil and then held in air at 760°C for periods of time ranging from 1 to 7 h. This treatment resulted in Ag⁺ ions occupying both Ba²⁺ sites and interstitial sites within the crystal. After the diffusion, the crystal was irradiated at room temperature with 60 kV x-rays. Three distinct paramagnetic defects, one holelike and two electronlike, were produced by the x-rays and observed with electron paramagnetic resonance (EPR) at temperatures between 15 and 80 K. The Ag⁺ ions substituting for Ba²⁺ ions trapped radiation-induced holes and became paramagnetic Ag²⁺ ions. At the same time, interstitial Ag⁺ ions trapped electrons and became paramagnetic Ag⁰ atoms. There are two distinct versions of these latter centres. The EPR spectra of the Ag⁰ atoms show resolved hyperfine from the ^107,109Ag isotopes, as well as from a neighbouring nucleus (possibly a Na⁺ impurity substituting for a Ba²⁺ ion). All the radiation-induced EPR signals were stable at room temperature for hours.

The change in confinement potentials in InAs/GaAs quantum dot (QD) nanostructures due to the interaction of strain fields from InAs QDs has been systematically investigated as a function of vertical stacking period in the light of the 'model solid' theory of Van de Walle and Martin using the strain information obtained from finite element analysis. As the stacking period (inter-dot separation) of InAs QDs decreases, in general, the interaction of strain fields in the nanostructure increases the direct band gap in most of the QD volume while a minor volume near the apex region shows a decreased band gap. A substantially close stacking of QDs results in a type-II behaviour along the stacking direction. In the inter-dot separation regime where the influences of the minor volume in the apex region, the type-II behaviour, and quantum mechanical coupling among QDs are not significant, it is anticipated that the closer stacking of QDs would yield an increased band gap and thus increased recombination energy for blue shift in photoluminescence spectra, as experimentally observed elsewhere recently.

Helium ion thermoluminescence (TL) fluence response is calculated as a function of particle energy from 1 to 10 MeV in the framework of the extended track interaction model (ETIM). The results of Monte Carlo calculations of absorbed dose as a function of the radial distance from the He ion track axis are employed for 1, 2, 2.6, 3.5, 4.95, 6.76 and 10 MeV He ions in condensed phase LiF. The radial dose is then transformed to radial occupation probabilities for the TL trapping and luminescent defect centres using optical absorption (OA) dose filling constants based on experimental gamma/electron OA dose response measurements. The radial defect occupation probabilities are used to estimate the track structure parameters (TSPs) r₁₀₀ (full occupation defining the track core), r₅₀ (50% occupation) and r_h (track extension). These, of course, may be different for the trapping centres, luminescent centres and competitive centres due to their different charge carrier trapping cross-sections leading to different dose-filling constants. He ion TL fluence response (linearity, supralinearity and saturation) is then modelled in the framework of the ETIM with the TSPs and dose filling constants as input parameters. It is illustrated that the maximum supralinearity f(n)_max increases smoothly with increasing particle energy due to the gradually increasing track radial extension and the increase in the number of electrons escaping the parent track, N_e, relative to the number recombining within the parent track, N_w.

Eutectic alloys of GaSb–FeGa_1.3 were prepared by the vertical Bridgman method. A microstructure with the needle-shaped metallic FeGa_1.3 phase oriented in a specific direction and uniformly distributed within the GaSb matrix was obtained. In GaSb–FeGa_1.3 eutectics, the electrical and thermal conductivity, thermal diffusivity, thermoelectric power and Hall coefficients were investigated in a wide temperature range. These properties were measured at different mutual directions of current, thermal flow, magnetic field and metal phase inclusions. The influence of metallic inclusions on these properties was revealed and the distinctive characteristics of electron and phonon processes were established.

A three-dimensional array of ∼55 nm diameter InGaAsN : Sb quantum dots (QDs) was nonlithographically fabricated starting with a MBE-grown InGaAsN : Sb/GaAs multiple quantum well wafer. The nonlithographic fabrication process relies on the use of a highly ordered nanopore alumina membrane for lateral patterning followed by a reactive ion etching. The dot size uniformity, spatial ordering and three-dimensional packing density are among the best reported so far. Nondestructive Raman scattering was employed to assess the material quality of the samples. While photoluminescence study points towards a moderately increased role of nonradiative transitions in the fabricated QD arrays, the dots show a relatively high emission intensity at room temperature, suggesting improved operation of long wavelength optical devices. Likely mechanisms contributing to the PL increase and the observed ∼50 meV red shift in the emission peak position are discussed. This study might be of great importance for engineering well controllable, ultra-high density, three-dimensionally arranged QD arrays for active optical devices.

We report studies on argon glow discharges established between flat disc electrodes, at pressure × electrode separation (pd) values between 45 and 150 Pa cm, with special attention to heavy-particle processes including heavy-particle excitation induced light emission. The discharges are investigated experimentally and also through self-consistent hybrid modelling. The comparison of the experimental and computed light intensity distributions verifies the correctness of the model, which gives a detailed insight into the discharge operation. The efficiency of heavy-particle excitation shows a universal dependence on the reduced electric field. At the higher pd values the scaling of electrical characteristics and light emission intensity with electrode separation is verified, however, additional processes (radial losses of charged particles and reduction of the active cathode area) result in the violation of scaling at the lowest pd value when the discharge tube diameter is kept constant.

The role of the microstructure of cathode materials in the geometrical parameters and dynamic processes of cathode spots is studied on the basis of investigation of the microscopic traces of vacuum arcs without an external magnetic field. The arc erosion pattern shows that spot craters locate on the phases with weak voltage withstanding, whose size, morphology and distribution determine those of the arc spots. The spot tracks over the cathode alloys with ultra-fine microstructure disperse to a greatly increased electrode area, suggesting that the spot motion is controlled by the microstructure of the electrode materials. A proposed spot dynamics model infers that the spot velocity may increase by two orders when the typical microstructure scale is less than a characteristic value, i.e. the spot fragment size. It is concluded that the microstructure, including the constituent phases and their size, is an important factor dominating the microscopic migration mechanism of arc cathode spots.

This paper considers the case where an electronegative gas is diluted by an electropositive one, as is typical in plasma processing applications, and examines the way in which the relevant parameters vary.

The impact of dilution is examined over a wide range of pressures (3 mTorr cm–300 Torr cm) and a wide range of degrees of ionization (10⁻⁶–10⁻²), for dilutions of 0%, 90% and 99%.

In order to be specific, the case of argon diluting chlorine is studied in detail. But comparison with experiment awaits such data, so only general conclusions can be reached.

A recent paper (Riboche et al 2003) gave a treatment of this problem as a generalization of earlier work by Lieberman (1988). The expressions produced are intrinsically complicated and there is a natural tendency to resort to numerical methods at an early stage. This note adopts an alternative approach and seeks to evaluate the integrals concerned in a systematic manner so as to obtain analytical expressions, albeit approximations, but ones with known accuracy.

An AC-driven, non-equilibrium, atmospheric pressure air plasma is generated within the gap separating a disc-shaped metal electrode and a water electrode. The typical OH emission intensity distribution is recorded by high-speed CCD camera. The temporal emission behaviour of OH(A–X) shows that the OH emission intensity stays relatively constant during most of the current cycle but decreases abruptly when the current is close to zero. The N₂ rotational and vibrational temperatures are obtained by comparison of the simulated spectra of C ³Π _u–B ³Π_g (Δv = −2) band transition of nitrogen with the experimental results. They are about 1800 K and 2600 K, respectively. The electron density in the plasma is measured utilizing the electrical conductivity. It is found to be 7 × 10¹² cm⁻³ in the centre of the discharge during the peak current of I_peak = 40 mA. The electron density follows a Gaussian radial distribution with a typical width of 0.35 mm.

Photo-induced non-self-sustained discharges are studied in point–plane gaps in air. Point electrodes made of different materials and with different non-metallic coatings (including TiO₂ and HfO₂) are tested with varying intensity of radiation and voltage. It is found that the quantum yield of photoemission (electrons per photon) changes with time and depends on the radiation intensity and applied voltage. It is assumed that the tunnelling of electrons through the barrier at the metal–dielectric interface and the finite capacitance of the electron reservoir at the dielectric–gas interface limit the current. The results of the model fit with the experimental findings.

Predictions are made of the electric field necessary for the onset of corona in air for spheres and cylinders from the general breakdown criterion, ; α' is the net ionization coefficient, r is the radius and Q is a constant. It is found that Q = 10⁴ gives a fair fit to experimental results for both wires and points, and for positive and negative corona, for radii varying from 0.01 to 20 cm. By assuming a linear variation of α' with the difference of the electric field from the critical field where ionization equals attachment, an analytic formula can be derived for the breakdown field for the onset of corona as a function of the radius of the sphere, which is similar to, but different from, Peek's formula for the onset of corona for cylinders or wires. The formulae also give the breakdown field as a function of temperature and pressure. For Townsend and Streamer breakdown mechanisms, values of Q are typically 3 × 10⁴ and 10⁸, respectively. Thus the value of Q derived from experimental values of the onset field for corona for both wires and points is significantly lower than would be expected from either the Townsend breakdown criterion or the Streamer breakdown criterion. It is suggested that the most likely collision process to explain the observed low value of Q is indirect ionization, for example by collisions between excited molecules.

Two-dimensional hydrodynamic numerical modelling of a cathode-directed streamer discharge at a constant anode voltage of 100 kV in a 13 cm long gap filled with a nitrogen–oxygen mixture at atmospheric pressure is performed. The dynamics of main discharge parameters is analysed and quite good agreement with experimental measurements is obtained.

It is shown that, when the electron loss processes in the streamer plasma are neglected, the discharge dynamics changes insignificantly: the electric field in the streamer head changes by less than 1% and the streamer propagation velocity changes by 7%, while the anode current and the electric field in the streamer channel change by a factor of about 2. The calculated anode current is found to coincide with the experimental one only when kinetic processes in the streamer channel are taken into account.

The inception probability and the streamer length of pulsed positive corona discharges is determined in argon, nitrogen, oxygen and air. This study is performed in a 25 mm point-plane gap at a pressure of 1 bar. The lowest voltage at which a discharge in argon starts is 3 kV but only with an inception probability of 1%. At 5 kV the corona discharge in argon transforms into a spark with a probability close to 100%. The inception probability of corona discharges in all molecular gases used here as a function of the voltage is identical, starting with 1% at 4 kV and going up to 100% at 9 kV. The streamer lengths are quite different for these gases, nitrogen requiring the lowest voltage for streamers to cross the gap and oxygen the highest. This is probably due to electron attachment in oxygen. A remarkable result is that in air streamers bridge the gap at 8 kV, but spark breakdown occurs only above 26 kV. This property makes it relatively easy to obtain powerful pulsed corona discharges in air.

The influence of plate surface roughness in precise ellipsometry experiments is studied. The realistic case of a Gaussian laser beam crossing a uniaxial platelet is considered. The expression for the transmittance is determined using first order perturbation theory. In this frame, it is shown that interference takes place between the specular transmitted beam and the scattered field. This effect is due to the angular distribution of the Gaussian beam and is of first order in the roughness to wavelength ratio. As an application, a numerical simulation of the effects of the roughness of a quartz surface at normal incidence is provided. The interference term is found to be strongly connected to the random nature of the surface roughness.

ZnO films were electrodeposited on porous silicon templates with different porosities. The photoluminescence (PL) spectra of the samples before and after deposition of ZnO were measured to study the effect of template porosity on the luminescence properties of ZnO/porous Si composites. As-prepared porous Si (PS) templates emit strong red light. The red PL peak of porous Si after deposition of ZnO shows an obvious blueshift, and the trend of blueshift increases with an increase in template porosity. A green emission at about 550 nm was also observed when the porosity of template increases, which is ascribed to the deep-level emission band of ZnO. A model-based band diagram of the ZnO/porous Si composite is suggested to interpret the properties of the composite.

W–Cu–W multilayer metallic coatings are designed and deposited by dc magnetron sputtering on an Fe substrate. Correlations between the deposition parameters, such as target power and Ar gas pressure, and the film characteristics are investigated. Especially, deposition parameters for a dense W–Cu multilayer coating are discussed. The coatings exhibit small grain sizes and a dense surface structure for high target power and low argon pressure, leading to dense and well adhesive films.

A new ternary compound, HgBrI, having a variable, large bandgap was investigated using different nondestructive methods. The homogeneity of the crystal composition along the growth axis was given from absorption edge measurements performed on different parts of the crystals using classical optical measurements. The atomic composition was determined by x-ray energy dispersive spectroscopy. Based on photoluminescence and reflection measurements using Confocal Scanning Laser Microscopy, the exact position and dimensions of the inhomogeneity regions were established.

The conventional worldwide accepted method of computerized glow curve deconvolution based on the general order kinetics formalism has two fatal defects in systems where the trapping levels (two or more) have non-zero retrapping probability. The first one is ignoring the thermal connectivity between thermoluminescence (TL) peaks. This arises from the fact that under such a situation electrons trapped at one trapping level, once activated, can be retrapped in another thermally connected level via the conduction band during the recording of the glow curve. The other is the impossibility of obtaining a global minimum, in fitting the experimental TL with the theoretical one with existing techniques.

This paper aims to provide answers to these defects. The first one can be overcome by resorting to rigorous analysis using appropriate mathematical rate equations describing the flow of charge carriers. Though the second defect cannot be overcome completely, one can obtain a reasonable fit, which may not be unique. The algorithm is tested for synthetic as well as experimental glow curves.

The effects of alloying (Cr, Zr, Nb, V, W, Mo and Al) on the trend of elastic properties of TiN-based nitrides have been investigated using ab initio density functional theory calculations within the generalized gradient approximation. The calculated elastic coefficients of pure TiN are in good agreement with experimental results and compare favourably with other theoretical calculations. The Cauchy pressure (C₁₂–C₄₄) and the ratio G/B of shear modulus (G) and bulk modulus (B) are used to assess the brittle/ductile behaviour of TiN accompanying alloying additions. The difference between C₄₄ and G/B or Cauchy pressure in evaluating the brittle/ductile behaviour of nitrides is identified. The results demonstrate that alloying with Mo and W can reduce G/B to near 0.5 and reverse Cauchy pressure from negative values to positive values. These results indicate that Mo and W could improve the ductility of TiN, with Mo achieving the best effect.

Closed form, approximate stress solutions for simply supported cubic and orthotropic plates subjected to a lateral pressure are presented and compared with a series solution. The accuracy of the approximate solutions depends on the degree of anisotropy. For cubic plates exhibiting significant anisotropy, the approximate solution is within ∼10% of the series solution. For a graphite–epoxy composite plate of orthotropic symmetry, the approximate solution is within 10% of the series solution for the maximum stress at the plate centre and within 23% for the minimum stress at the plate centre. Single crystal plates tend to exhibit a large tangential stress where the stiff directions intersect the plate edges, implying that the edges of brittle single crystal test specimens must either be prepared better than those of isotropic specimens or have more of the plate overhanging the support.

Using high performance Brillouin spectroscopy we present a new technique, which enables us to perform acoustic microscopy with a spatial resolution of about 1 µm. This technique, called Brillouin microscopy, is tested on several bulk- and film-like polymer samples.

We report on investigations into the transport of positive thallium ions from gas-to-liquid and conversely liquid-to-gas, across the interface, in xenon. These studies showed that for a variety of drift field strengths and liquid depths, it was possible to transport the thallium ions from the gas phase through to the liquid. However, extraction of the positive ions from the liquid to the gas was not achieved. The difference in the two outcomes is thought to be due to the varying degrees of clustering of neutral xenon around the thallium ions in each case.

The feasibility of improving the conversion efficiency of a thermoelectric converter by employing interfaces between materials exhibiting Fermi gas (FG) and Fermi liquid (FL) behaviour has been studied. Thermocouples consisting of a semiconductor and a strongly correlated material have been fabricated and the Peltier heat measured over the temperature range 15–330 K. A number of materials possessing different types of strong electron correlation have been synthesized including the heavy fermion compound YbAl₃, manganite La_0.7Ca_0.3MnO₃ and high-T_c superconductor YBa₂Cu₃O_7−δ. n- and p-Bi₂Te₃-based solid solutions as well as n-Bi_0.85Sb_0.15 solid solution have also been synthesized and used as materials exhibiting FG properties. Experimental measurements of the Peltier heat were compared to the results of calculations based on preliminary measured thermoelectric properties of materials and electrical contact resistance at the interfaces. The potential of employing FG/FL interfaces in thermoelectric energy conversion is discussed.