Table of contents

The rapid thermal processing (RTP) technique features dynamic control of temperature, which permits high heating and cooling rates that cannot be reached with conventional furnace treatments. In recent years, RTP has been increasingly applied to the processing of magnetic materials. The controllable heating profiles provide an approach for the deliberate construction of materials structure via expediting phase transitions and tailoring materials morphology. In this review, the history, principles and facilities of several types of RTP techniques are introduced, including laser and electron beam heating, lamp heating, Joule heating and pulse thermal processing techniques. RTP of various advanced magnetic materials are reviewed, including soft magnetic materials, spintronic materials, magnetic recording materials and nanocomposite hard magnetic materials. Advantages of RTP are highlighted and compared with conventional thermal treatments.

The soft x-ray absorption and magnetic circular dichroism spectra at the Co L_2,3-edges of [Co/Pd]₂₀ and [CoB/Pd]₂₀ multilayered films, which were sputter-deposited in Ar gas containing N₂ at room temperature, were measured to investigate the electronic and spin states of the films. The macroscopic magnetic properties, such as coercivity and saturation magnetization, and microstructural properties, such as grain size and preferred orientation, were markedly changed for the [Co/Pd]₂₀ film by the B doping and N₂ addition. The B doping and/or the addition of N₂ gas caused appreciable changes not only in macroscopic magnetic properties but also in microscopic properties, such as the spin angular momentum and orbital angular momentum, throughout the multilayered film. On the other hand, a new spectral feature at the high energy side of the Co L₃-edge peak in x-ray absorption spectra was found in only the films with N₂ added. Thus, the microscopic properties, such as electronic and spin states, of the Co 3d electronic state for the [Co/Pd]₂₀ and [CoB/Pd]₂₀ films were changed by only the sputtering process with additive N₂ gas.

The structure, magnetic properties and martensitic transformation behaviour were investigated for as-spun Co₅₀Ni₂₀Ga₃₀ ribbons under various heat-treatment conditions. When the as-spun ribbons were annealed at 400 °C, the saturation magnetization decreased from 49.5 to 40.1 emu g⁻¹ and the coercivity increased from 18 to 2622 Oe, accompanied by the disappearance of the martensitic transformation. Re-treating the annealed sample at different temperatures from 600 to 1200 °C, the martensitic transformation reoccurred and the T_M was adjusted between 132 and 249 K.

The CALPHAD (calculations of phase diagrams) method is used to examine the effects of applied magnetic fields on the α/γ phase boundary in the Fe–Si system in the paramagnetic state. The reported susceptibility data for pure Fe is first re-evaluated. The contributions to the total Gibbs energy of the ferrite (α) and austenite (γ) from the external fields are calculated based on the Curie–Weiss law and the re-evaluated susceptibility data. The Fe–Si phase diagram on the Fe-rich side as a function of applied field is calculated using the Thermo-Calc™ package. With increasing field strength, the γ loop shrinks monotonically; that is, the α/γ-Fe transition temperature increases while that for γ/δ-Fe transition decreases, albeit more slowly. Finally, in conformance with the existing CALPHAD databank, Redlich–Kister polynomials are proposed to account for the compositional and temperature dependence of the contribution to the total Gibbs energy from the applied field in the paramagnetic state in the range over which the Curie–Weiss law is obeyed.

The electrical transport and magnetoresistance (MR) properties of the composites (1 − x)La_0.67Sr_0.33MnO₃ (LSMO)/x Sb₂O₅ with different Sb₂O₅ molar percentages x were investigated. Experimental results show that the Sb₂ O₅ addition plays a key role in the electrical transport behaviour of the composites. Different transport and MR behaviours are observed for x ⩽ 2% and x⩾2%. Compared with pure LSMO, an enhanced MR effect for the composites is found over a wide temperature range under applied magnetic fields of both 0.3 and 3 T. Specially, a large and constant MR value of ∼18% is observed at room temperature, which is attractive for the study of applications. We argue that such an enhancement in MR is attributed to the magnetic disorder at the LSMO grain surfaces or boundaries caused by the Sb₂O₅ addition. Moreover, the resistivity (ρ) versus T curves for x = 1% and 2% samples fit well to the phenomenological model derived from spin-polarized tunnelling at grain boundaries and thermal activation. This result confirms that the grain boundaries have been modified in the composites with low Sb₂O₅ addition and provides information about the microscopic transport mechanism in the system.

Polycrystalline, multiphase ceramic composites exhibiting voltage-limiting nonlinear current–voltage (I–V) relations with the nonlinearity coefficient (α) of 220–360 have been realized by co-sintering ZnO and La_0.6Sr_0.4MnO₃. The I–V characteristics are symmetrical on voltage reversal to the negative quadrant. By selecting the sintering temperature (1573–1673 K) and duration (1–4 h), the interdiffusion of Mn into ZnO and of Zn into La_0.6Sr_0.4MnO₃ could be controlled so that La and Sr segregation leads to the formation of p-type La_0.6Sr_0.4Mn_1−yZn_yO₃ (LSMZO) permeating along the intergrain regions of n-type ZnO_1−γ : Mn, thereby preserving interior n–p–n heterojunction characteristics at the grain boundaries. Possible mechanisms proposed for large nonlinearity in the I–V curves are (i) avalanche multiplication of minority carrier through impact ionization of majority carriers or (ii) Zener tunnelling of charge carriers across the depleted Schottky potential barriers prevailing at n-ZnO_1−γ : Mn/p-LSMZO (and p-LSMZO/n-ZnO_1−γ : Mn) interfaces on the left- and right-hand side of p-LSMZO intergrain phase, wherein the type of conduction mechanism is predominantly dependant on the transport properties and thickness of intergranular LSMZO phase. The I–V characteristics are modified under an applied magnetic field by way of decreasing the turn-on voltage leading to higher currents. Such magnetically tuneable electrical nonlinearity arises from the prevalence of spin-polarized conduction in the ferromagnetic insulating LSMZO which prevails at the intergrain regions of n-ZnO_1−γ : Mn. Magnetization studies of bulk LSMZO show the conversion from ferromagnetic metallic (y < 0.03) to ferromagnetic insulating (y = 0.03–0.08) and paramagnetic insulating (y > 0.08) phases on increasing the zinc content. However, the n-ZnO_1−γ : Mn grains remain paramagnetic insulator and the magnetically tuneable I–V curves are not dependent on the ferromagnetic ordering therein.

High-frequency magnetic and attenuation properties of ferrite/polymer composites were investigated for undoped and V₂O₅ doped W-type barium ferrites, BaCo_xZn_2−xFe₁₆O₂₇, with x = 1.0, 1.3 and 1.5. The results show that, as compared to the undoped samples, the permeabilities and are increased by about 50% and 40% for composites doped with 1.0 wt% of V₂O₅, due to the improved domain wall permeability. These composites have superior attenuation properties over the undoped ones and are, therefore, potential candidates for EM attenuation materials with low reflectivity and broad bandwidth covering S, X and Ku bands.

Neutron detection signatures in thin films of semiconducting boron carbide were measured as a function of three applied reverse bias voltages, 0, 1.57 V and 3.15 V. Thermal neutrons were detected at zero bias. As the bias was increased, an increase in efficiency and in peak channel positions was observed. These two significant effects have been attributed to an increase in charge collection at higher bias voltages.

Theories and methods of x-ray diffraction enhanced imaging (DEI) and computed tomography of the DEI (DEI-CT) have been investigated recently. But the phenomenon of projection offsets which may affect the accuracy of the results of extraction methods of refraction-angle images and reconstruction algorithms of the DEI-CT is seldom of concern. This paper focuses on it. Projection offsets are revealed distinctly according to the equivalent rectilinear propagation model of the DEI. Then, an effective correction method using the equivalent positions of projection data is presented to eliminate the errors induced by projection offsets. The correction method is validated by a computer simulation experiment and extraction methods or reconstruction algorithms based on the corrected data can give more accurate results. The limitations of the correction method are discussed at the end.

We propose photonic crystal (PC) multi-channel drop filters implemented using microstrip lines. Their performance is based on coupling between the cavity modes of microstrip PC cavities and the guiding modes of microstrip waveguides. The dropping frequency can be controlled by correct design of the cavity. The proposed multi-channel drop filter can be useful for implementing new types of functional microwave filters.

Swift heavy ion induced colour centres in polycrystalline LiF thin films deposited on silica substrate were studied. LiF thin films of 200 nm thickness, deposited by the electron beam evaporation method, and silica substrates were irradiated with 120 MeV Au⁹⁺ ions in the fluence range 5 × 10¹⁰–1 × 10¹³ ions cm⁻². Optical absorption, photoluminescence (PL) and glancing angle x-ray diffraction (GAXRD) techniques were used for the characterization of the irradiated films. Optical absorption spectroscopy was performed on irradiated LiF films by using irradiated silica substrates as reference sample. The resulting spectra show three bands at 260 nm, 380 nm and 445 nm corresponding to F, F₃ and F₂ colour centres, respectively. Absorbance of colour centres increases with fluence and seems to saturate at higher fluences. PL spectra, performed by an excitation source of 442 nm, show two broad bands of F₂ and colour centres. The intensity of both the centres increases upto 3 × 10¹¹ ions cm⁻² followed by some decrease. The GAXRD results show that the average grain size decreases systematically from 35 nm for the pristine film to 20 nm for the sample irradiated at a fluence of 1 × 10¹³ ions cm⁻².

In this paper we study optically stimulated luminescence (OSL) signals created by basic thermal transfer. The thermal stabilities of trapped charges associated with the source traps of thermal transfer and their related OSL traps, and trapped holes associated with non-luminescent centres (R centres) were studied for a sedimentary quartz, using isothermal annealing and pulse annealing methods. The thermally transferred charges are mainly from three groups of traps (source traps), shallow, medium and deep traps. Kinetic analysis indicates that the medium and deep source traps have trap depths of ∼1.1 eV and ∼1.5 eV, and lifetimes 7.6 × 10⁵ and 6.0 × 10¹1 years at 20 °C, respectively. Two kinds of OSL traps, one unstable and the other stable, are identified as accepting charges during thermal transfer. These OSL traps have different thermal stabilities from the main OSL traps, which are associated with the 325 °C TL peak. The results also indicate that the basic thermal transfer related OSL signals perhaps use different R centres from the dose-generated fast bleachable OSL signals.

The effect of H₂ post-metallization annealing (PMA) on the electrical behaviour of sputtered Ta₂O₅ layers on nitrogen ion implanted Si is investigated. Capacitance–voltage and current–voltage measurements at various temperatures (20–110 °C) are performed to analyse the electrically active defects and leakage currents before and after PMA. The high densities of oxide charge, interface and slow states typical of as-deposited stacks are found to be strongly reduced by one to two orders of magnitude after annealing. H₂ treatment affects both bulk Ta₂O₅ and interfacial layers but is more efficient in annealing electrically active defects in the nitrided layer. The effect of defect annealing on the leakage currents and conduction mechanisms is also discussed.

Thin films of silica containing 0.15 at.% Eu and 3.5 to 7 at.% Na were prepared by sol–gel chemistry and annealed at 600 ° C in air, then submitted to ion–exchange treatments in NaNO₃ : AgNO₃ for studying the effect of the plasmon resonance of Ag nanoparticles on the excitation of Eu³⁺ ions. When excited at other wavelengths in the UV other than that corresponding to the ⁷F₀–⁵L₆ transitions of the Eu³⁺ ions, the yield of the ⁵D₀–⁷F₂ emission is increased by a factor of the order of 10 in films containing Ag particles with respect to that in non-exchanged films.

An approximate analytical expression is obtained for the escape factors for thermionically emitting cathodes in atomic gases that is uniformly valid at all values of the reduced electric field. This expression is used for evaluation of the escape factors in neon, helium and mercury. An independent evaluation is performed by means of Monte Carlo simulations. The analytical results are in good agreement with the results of Monte Carlo simulations, both for reflecting and non-reflecting cathodes.

For a better understanding of the plasma/photocatalytic material interaction under plasma exposure, a study of the electrical properties of a cylindrical sinusoidal dielectric barrier discharge is performed with and without porous material containing TiO₂. The metallic inner electrode is in contact with the gas gap. First, the presence of porous material made of silica fibres coated with nanoparticles of TiO₂ leads to a strong increase of the injected energy for the same applied voltage. Then the time evolution of the current amplitude distribution function (CADF) shows two different peak populations on the positive half period (when the metallic inner electrode is positive). Apart from numerous low intensity plasma filaments (around 1 mA amplitude), much larger ones exist (around 1 A). These large current amplitude peaks are responsible for 50–70% of the injected energy depending on the presence of the photocatalytic material. The influence of 900 ppm of C₂H₂ as well as external ultraviolet irradiation on the CADF is also reported.

The emission spectrum corresponding to the 4p⁵4d^N+1 + 4p⁶4d^N–14f → 4p⁶4d^N transition array is concentrated in a narrow interval of wavelengths. That is due to the existence of an approximate selection rule and quenching of some lines by configuration mixing. Thus such emission of elements near Z = 50 is considered to be the main candidate for the EUV lithography source at λ = 13.5 nm. In the present work the regularities of these transition arrays are considered using their integral characteristics: average energy, total line strength, variance and interval of array containing some part of the total transition probability. Calculations for various ions of elements In, Sn, Sb, Te, I and Xe have been performed in a two-configuration pseudorelativistic approximation, which describes fairly well the main features of the spectra. The variation in the values of the main integral characteristics of the spectra with atomic number and ionization degree gives the possibility of comparing quantitatively the suitability of the emission of various ions for EUV lithography.

Space-charge dominated streamer discharges can emerge in free space from single electrons. We reinvestigate the Raether–Meek criterion and show that streamer emergence depends not only on ionization and attachment rates and gap length, but also on electron diffusion. Motivated by simulation results, we derive an explicit quantitative criterion for the avalanche-to-streamer transition both for pure non-attaching gases and for air, under the assumption that the avalanche emerges from a single free electron and evolves in a homogeneous field.

Optical emission spectroscopy was performed on a metal-halide lamp under micro-gravity conditions of the international space station. Several transitions of atomic and ionic Dy and atomic Hg have been measured at different lateral positions from which we obtained atomic and ionic Dy and atomic Hg intensity profiles. After Abel inversion, the calibrated radial intensity profile of Hg was used to calculate a radial temperature profile. By combining the radial temperature profile with the calibrated radial intensity profile of the additive, the absolute radial density profile of the total atomic and ionic density of Dy was obtained. The measurements showed a hollow density profile for the atoms and ions in the centre. In the outer parts of the lamp molecules were found to dominate. Lamps containing Dy showed contraction of the arc, which increased for higher powers. Measurements were duplicated at 1-g and showed less radial segregation than for 0-g. As the power was increased, the difference between 0-g and 1-g of the radial intensity, density and temperature profile were diminished.

This paper presents a simple phenomenological model and detailed simulation studies of the breakdown of a gas under the simultaneous action of electric and magnetic fields. In deriving expressions for the breakdown voltage variations of both the ionization coefficient and the secondary electron yield in a magnetic field are taken into account. Calculations were carried out by using the XOOPIC code with the old and the improved secondary emission models adjusted to involve the influence of the magnetic field on the secondary electron production. It was shown that the incorporation of the variation of the secondary electron yield with magnetic field leads to the better agreement with existing experimental results.

In this paper a large-scale pulsed corona system is described in which pulse parameters such as pulse rise-time, peak voltage, pulse width and energy per pulse can be varied. The chemical efficiency of the system is determined by measuring ozone production. The temporal and spatial development of the discharge streamers is recorded using an ICCD camera with a shortest exposure time of 5 ns. The camera can be triggered at any moment starting from the time the voltage pulse arrives on the reactor, with an accuracy of less than 1 ns. Measurements were performed on an industrial size wire-plate reactor. The influence of pulse parameters like pulse voltage, DC bias voltage, rise-time and pulse repetition rate on plasma generation was monitored. It was observed that for higher peak voltages, an increase could be seen in the primary streamer velocity, the growth of the primary streamer diameter, the light intensity and the number of streamers per unit length of corona wire. No significant separate influence of DC bias voltage level was observed as long as the total reactor voltage (pulse + DC bias) remained constant and the DC bias voltage remained below the DC corona onset. For those situations in which the plasma appearance changed (e.g. different streamer velocity, diameter, intensity), a change in ozone production was also observed. The best chemical yields were obtained for low voltage (55 kV), low energetic pulses (0.4 J/pulse): 60 g (kWh)⁻¹. For high voltage (86 kV), high energetic pulses (2.3 J/pulse) the yield decreased to approximately 45 g (kWh)⁻¹, still a high value for ozone production in ambient air (RH 42%). The pulse repetition rate has no influence on plasma generation and on chemical efficiency up to 400 pulses per second.

This paper describes influences of interfacial properties on second-harmonic generation of Lamb waves propagating in layered planar structures. The nonlinearity in the elastic wave propagation is treated as a second-order perturbation of the linear elastic response. Due to the kinematic nonlinearity and the elastic nonlinearity of materials, there are second-order bulk and surface/interface driving sources in layered planar structures through which Lamb waves propagate. These driving sources can be thought of as forcing functions of a series of double frequency lamb waves (DFLWs) in terms of the approach of modal expansion analysis for waveguide excitation. The total second-harmonic fields consist of a summation of DFLWs in the corresponding stress-free layered planar structures. The interfacial properties of layered planar structures can be described by the well-known finite interfacial stiffness technique. The normal and tangential interfacial stiffness constants can be coupled with the equation governing the expansion coefficient of each DFLW component. On the other hand, the normal and tangential interfacial stiffness constants are associated with the degree of dispersion between Lamb waves and DFLWs. Theoretical analyses and numerical simulations indicate that the efficiency of second-harmonic generation by Lamb wave propagation is closely dependent on the interfacial properties of layered structures. The potential of using the effect of second-harmonic generation by Lamb wave propagation to characterize the interfacial properties of layered structures are considered. Some experimental results are presented.

In the case of typical gas sensors responding by gas adsorption on the surface, it is very difficult to detect inert gases which have low chemical adsorption energies under normal conditions. Carbon nanotubes (CNTs) are well known as electron emitters in applications like field emission displays. In this work, we fabricate a physical gas sensor (or vacuum sensor) using CNTs with compactness and reliability of the fabrication process and discuss current–voltage characteristics as a function of the vacuum pressure. The proposed sensor, based on an electrical discharge theory known as Paschen's law, works by figuring changes of the discharge current depending on the concentration of gases which can ionize through collisions with electrons emitted from CNT emitters.

Growth of nano-crystalline silicon (nano-Si) films from hydrogen-diluted SiCl₄ by the plasma enhanced chemical vapour deposition technique at 250 °C has been studied through mass spectrometry, Langmuir probe diagnostic techniques and Raman spectra measurements. The effects of the hydrogen dilution ratio on the relative densities of SiCl_n (n = 0–2) in SiCl₄/H₂ plasma are investigated. The average electron energy (E_e) and electron density (N_e) in SiCl₄/H₂ plasma increase with the hydrogen dilution ratio till their maxima at 9.25 eV and 3.7 × 10⁹ cm⁻³, respectively. A suitable hydrogen dilution ratio R (0.4–0.67) is beneficial for the formation of SiCl_n (n = 0–2) radicals because E_e and N_e both have maxima in this range. More SiCl_n (n = 0–2) radicals will improve the deposition rate and film quality. In addition, hydrogen radicals play an important role in the surface reaction process. The reaction of H and SiCl_n (n = 0–2) on the growing surface is beneficial for creating dangling bond sites and prompting the film growth, and the exothermic reaction exothermic energy of H with Cl on the film-growing surface results in the increase of the effective temperature of the film growth.

It has been observed that the microwave annealing of doped silicon wafers in the multimode cavity is accompanied by a specific temperature rise in the near-edge region of the wafer. Experimental investigation and theoretical analysis suggest that the effect is not a result of the microwave irradiation non-uniformity but occurs due to the diffraction of electromagnetic waves on the edge of a thin conducting plate. The level of local overheating depends on the polarization and propagation direction of the incident electromagnetic wave. It is most pronounced in the case when the wave vector is parallel to the plate surface but perpendicular to the plate edge. A method of the plate screening has been suggested to suppress the edge effect and improve the temperature uniformity over the plate during heating. The efficiency of the method has been confirmed by a FDTD numerical simulation of the microwave field near an edge of the plate irradiated isotropically in the multimode cavity.

The temperature dependence of the VUV transmission of synthetic fused silica is of interest for commercial applications as well as for fundamental research. In this work the transmission properties of Suprasil 2 from Heraeus with an absorption edge at very low wavelengths is investigated. The absorption edge of this quartz glass shifts from 170 to 180 nm between 789 and 1129 K. The Urbach rule is discussed for the characterization of the temperature dependent transmission curves. The results are applied to the diagnostics of the Hg 185 nm line from a high pressure mercury discharge lamp.

Electrical transport properties of cobalt doped polyaniline in an aqueous ethanol medium were investigated in the temperature range 77 ⩽ T ⩽ 300 K, applying magnetic fields up to 1 T in the frequency range 20 Hz–1 MHz. The room temperature dc resistivity increases with increase in Co content. The dc resistivity and magnetoresistivity of these samples have been interpreted in terms of the variable range hopping theory. The frequency dependence of conductivity has been described by a power law σ(ω) ∝ ω^S. The value of s is found to be temperature dependent, which shows a decreasing trend with temperature. The correlated barrier hopping model is the most likely mechanism for the electron transport. The different physical parameters were calculated from the experimental data.

This study deals with the improvement of electric field-induced thickness strain of polyurethane (PU) elastomer films by carbon black (C) nanopowder incorporation in the polymer matrix. Different carbon volume concentrations—0.5, 0.7, 1 and 1.5%—have been tested. Weak-field dielectric and resistivity measurements revealed that a percolative effect is not induced by carbon filling up to 1.5 vol%. Thickness strain measurements showed that both pure PU and C/PU composite films exhibit similar strain variations which are not governed only by electrostatic forces (Maxwell stress) and/or electrostriction forces. The highest strain amplitude value observed was obtained for 1% C composite thin film (S_a = −7.4% at E = 17.8 kV mm⁻¹). In comparison the highest S_a for pure PU thin film was −6.7% at E = 37.5 kV mm⁻¹). In the case of thick samples, the thickness strain was not enhanced by C loading, which strongly suggests interfacial space charge effects in pure PU film, confirmed by the frequency dependence of strain level.

Application of electromagnetic stirring (EMS) during continuous casting shears off the dendrites from the solidification front to produce billets with a non-dendritic microstructure. In the present study, a systematic approach to the scaling analysis of momentum, energy and species conservation equations pertaining to the case of the solidification of a binary alloy in the presence of EMS is outlined. With suitable choices of non-dimensionalizing parameters, the governing equations coupled with appropriate boundary conditions are first scaled, and then the relative significance of various terms appearing in them are analysed. In the physical domain two regions are identified, one where the electromagnetic forces play a dominant role in the momentum equations, and the other where the inertia and viscous effects play major roles. Using the scaling predictions, the influence of various processing parameters on the system variables can be utilized for the selection of appropriate electromagnetic forces to shear off the dendrites from the solidification area. For the sake of assessment of the scaling analysis, the predictions are validated against corresponding computational results.

Discharges between water drops are important in the ageing of hydrophobic outdoor insulators. They may also be important in the processes leading up to flashover of these insulators in high pollution conditions. This paper considers discharges between drops when a limited alternating current is available, as experienced by an ageing insulator in service. A phenomenon is identified in which the length of a discharge between two drops is reduced through a particular type of distortion of the drops. This is visually characterized as a liquid protrusion from each of a pair of water drops along the insulator surface. This process is distinct from vibration of the drops, general distortion of their shape and the very fast emission of jet streams seen in very high fields. The process depends upon the discharge current, the resistivity of the moisture and the hydrophobicity of the insulation surface.

Single-wall carbon nanotubes (SWNTs) with anionic or cationic coatings have been prepared by exploiting the ability of certain surfactants to form a monolayer shell around the nanotube. The presence of electrically charged functional groups on the surface of the SWNT allows thin film deposition to proceed via the electrostatic layer-by-layer method. This self-assembly process was monitored using the quartz microbalance technique and Raman spectroscopy, while the morphology of the resulting thin layers was studied with atomic force microscopy. A variety of different architectures has been built up. In one arrangement, a single species of a modified SWNT (anionic or cationic) was alternated with a passive polymer to form a composite structure. A 'superlattice' architecture comprising alternating anionic and cationic modified nanotubes was also fabricated. The in-plane and out-of-plane dc conductivities of the films were measured at room temperature and contrasted with reference architectures (i.e. those containing no nanotubes). The results showed clearly that the incorporation of SWNTs into the multilayer assemblies provided electrically conductive thin films. It is suggested that the current versus voltage behaviour, particularly in the out-of-plane direction, is controlled by quantum mechanical tunnelling of carriers between the nanotubes.

This paper reports the development of optimal vascularization for supplying self-healing smart materials with liquid that fills and seals the cracks that may occur throughout their volume. The vascularization consists of two-dimensional grids of interconnected orthogonal channels with two hydraulic diameters (D₁, D₂). The smallest square loop is designed to match the size (d) of the smallest crack. The network is sealed with respect to the outside and is filled with pressurized liquid. In this work, the crack site is modelled as a small spherical volume of diameter d. When a crack is formed, fluid flows from neighbouring channels to the crack site. This volume-to-point flow is optimized using two formulations: (1) incompressible liquid from steady constant-strength sources located in every node of the grid and from sources located equidistantly on the perimeter of the vascularized body of length scale L and (2) slightly compressible liquid from an initially pressurized grid discharging in time-dependent fashion into one crack site. The flow in every channel is laminar and fully developed. The objectives are (a) to minimize the global resistance to the flow from the grid to the crack site and (b) to minimize the time of discharge from the pressurized grid to the crack site. It is shown that methods (a) and (b) yield similar results. There is an optimal ratio of channel diameters D₂/D₁ < 1, and it decreases as the grid fineness (L/d) increases. The global flow resistance of the grid with optimized ratio of diameters is approximately half of the resistance of the corresponding grid with one channel size (D₁ = D₂). The optimized ratio of diameters and the minimized global resistance depend on how the grid intersects the crack site: this effect is minor and stresses the robustness of the vascularized design.

Effects of compressive stress on the dielectric properties of PMN–PT ceramics were investigated. The ceramics with the formula (1 − x)Pb(Mg_1/3Nb_2/3)O₃-(x)PbTiO₃ or (1 − x)PMN–(x)PT (x = 0.1–0.5) were prepared by a conventional mixed-oxide method and then characterized with x-ray diffraction (XRD) and scanning electron microscopy. Dense perovskite-phase PMN–PT ceramics with uniform microstructure were successfully obtained. The dielectric properties under compressive stress were observed at stress up to 230 MPa using a home-built compressometer. The experimental results revealed that the superimposed compression stress significantly reduced both the dielectric constant and the dielectric loss tangent of 0.9PMN–0.1PT ceramic, while the changes were not as significant in the other PMN–PT ceramic compositions. In addition, the dielectric properties were considerably lowered after a stress cycle. Since change in the dielectric properties with applied stress was attributed to the competing influences of the intrinsic and the extrinsic contributions, the observations were mainly interpreted in terms of domain switching through non-180° domain walls, de-ageing, clamping of domain walls and the stress induced decrease in the switchable part of spontaneous polarization.

Scanning force microscopy is used to investigate microscale ferroelectric domains at the +Z surface of near-stoichiometric lithium tantalate (SLT) single crystals. Both line and lattice patterns of microdomains are fabricated at the +Z surface of SLT wafers, which are switched by scanning these wafers with the external field exceeding the coercive field. The +Z polar surface in these patterns is etched by a mixture of acid aqueous solution (HF : HNO₃ = 1 : 1), which reveals the precise microdomain structure. The analysis of the experimental data indicates that the rearrangement of the ferroelectric domain structure occurs during the chemical etching process; the effect of both defect and crystal structure on the microdomain growth process is also discussed. The present results are helpful for us to realize the fabrication of the expected microdomain morphology.

Thermal creep occurs in anisothermal gas microflow. It is highly desirable to understand the creep effect on the flow and heat transfer characteristics for developing natural convective microflow. In this study, we investigate the steady developing natural convective flow in an open-ended vertical parallel-plate microchannel with asymmetric wall temperature distributions. The boundary-layer equations subject to the boundary conditions with respect to dynamic pressure at the channel entry as well as higher-order jump temperature and slip velocity with thermal creep along the channel surface are employed. The mathematical model and the numerical code are validated through available macroscale work. Numerical solutions of high-order slip coefficient, slip/jump, velocity, pressure, temperature, flow rate, flow drag and heat transfer rate are presented for air at the standard reference state with complete accommodation. It is found that thermal creep has significant effect on the high-order slip effect and the flow and thermal fields. The creep effect is to increase the flow rate; moreover, valuable reduced flow drag and enhanced heat transfer are obtained.

A reference was inadvertently omitted at the end of the second paragraph on page 50. The text should be amended to read as:

Optical third harmonic generation during femtosecond pulse diffraction in a Bragg grating was first observed and described in [24].

The new reference [24] is:

[24] Smirnov V I, Juodkazis S, Dubikovsky V, Hales J, Zel'dovich B Ya, Misawa H and Glebov L B 2002 Resonant third harmonic generation by femtosecond laser pulses on Bragg grating in photosensitive silicate glass Proc. CLEO 2002 (Long Beach, CA) paper CTuG7