Table of contents

The engineering of electromagnetic modes at optical frequencies in artificial dielectric structures with periodic and random variation of the refractive index, enabling control of the radiative properties of the materials and photon localization, was first proposed independently by Yablonovitch and John in 1987. It is possible to control the flow of light in the periodic dielectric structures, known as photonic crystals (PC). As light waves scatter within the photonic crystal, destructive interference cancels out light of certain wavelengths, thereby forming a photonic bandgap, similar to the energy bandgap for electron waves in a semiconductor. Photons whose energies lie within the gap cannot propagate through the periodic structure. This property can be used to make a low-loss cavity. If a point defect, such as one or more missing periods, is introduced into the periodic structure a region is obtained within which the otherwise forbidden wavelengths can be locally trapped. This property can be used to realize photonic microcavities. Similarly, a line of defects can serve as a waveguide.

While the realization of three-dimensional (3D) photonic crystals received considerable attention initially, planar two-dimensional (2D) structures are currently favoured because of their relative ease of fabrication. 2D photonic crystal structures provide most of the functionality of 3D structures. These attributes have generated worldwide research and development of sub-μm and μm size active and passive photonic devices such as single-mode and non- classical light sources, guided wave devices, resonant cavity detection, and components for optical communication. More recently, photonic crystal guided wave devices are being investigated for application in microfludic and biochemical sensing. Photonic crystal devices have been realized with bulk, quantum well and quantum dot active regions.

The Cluster of articles in this issue of Journal of Physics D: Applied Physics provides a glimpse of some of the most recent advances in the application of photonic crystals. The modelling of PC defect-mode cavities are described by Zhou et al. Ye and co-authors describe the concept and realization of a novel 3D silicon-based spiral PC. It is, in fact, the only article on 3D PCs. The design and realization of ultra-high Q heterostructure PC nanocavities are described by Song and co-authors. The concept of self-collimation of light in PCs and its applications are presented by Prather and co-workers. Experimental and numerical studies on the negative refraction related phenomenon in 2D PCs are the subject of the next article by Ozbay and co-authors. The emerging subject of slow light generation, control and propagation in PCs is presented in the next two articles by Baba and Mori and by Krauss. Finally, the progress made in the development of PC microcavity lasers and electrically injected microcavity light emitters and arrays is described, respectively, by O'Brien et al and by Chakravarty et al.

It is hoped that readers will get a sense of the exciting developments and the possibilities presented by heterostructure photonic crystals and their devices from reading the articles in this Cluster.

A phenomenological dimensional reduction approach (PDRA) for the cavity characteristics in defect mode based photonic crystal (PC) lasers is presented. Based on the fully vectorial three-dimensional finite-difference time-domain (3D FDTD) technique, simultaneous enhancement and suppression in spontaneous emission and absorption were obtained in an absorptive photonic crystal slab (PCS) cavity. Effective index perturbation (EIP) was proposed for fast and accurate determination of the effective index and the dominant resonant cavity frequency in a 3D PCS structure, with two-dimensional (2D) FDTD simulation. Further dimensional reduction from 2D to one-dimensional planar cavity enables phenomenological modelling of lasing characteristics via the effective reflectivity calculation and rate equation analysis. Very fast and accurate results have been achieved with this PDRA approach. A high spontaneous emission factor and cavity quality factor Q were obtained in a single defect cavity, which led to over an order reduction in lasing gain threshold. The model offers a fast and accurate tool for the design and modelling of PC defect mode cavity based devices and aids the research in the proposed novel defect mode based devices such as ultra-compact light sources on Si and spectrally resolved PC infrared photodetectors.

Three-dimensional (3D) Si spiral photonic crystals have been realized by oblique angle deposition with a new approach of substrate 'swing' rotation. We demonstrate that, by using swing rotation, the arms of the spirals can be grown with a uniform diameter. Without swing rotation the arms of the spirals may overgrow due to a ballistic 'fan-out' phenomenon. The fabricated 3D Si spiral photonic crystals have a [0 0 1]-diamond symmetry which has been shown previously to have a large photonic band gap. Our optical measurements of reflectance have suggested the existence of such a photonic band gap in our structures centred at a wavelength of 2 µm.

In this paper, we review the fundamentals of heterostructures in two-dimensional photonic-crystal (PC) slabs and demonstrate ultra-high-Q nanocavities, which are key components for the realization of PC devices. Heterostructure PCs (hetero-PCs) consist of multiple PC waveguides with different lattice constants. Here we theoretically demonstrate the selective transmission and reflection characteristics of various hetero-PCs, including photonic-tunnelling structures. On the basis of these traits, we construct photonic double-heterostructure nanocavities and detail their optical properties, including the transmission characteristics of the input waveguides and the Q factors of the cavities. We show that ultra-high-Q factors (∼1 200 000) can be realized in the nanocavities and are limited by imperfections in the fabrication process. Ultra-high-Q photonic double-heterostructure nanocavities are important not only in the development of various photonic applications but also for advancing our knowledge of photonic heterostructures.

In this paper, we report on the development of the self-collimation phenomenon in photonic crystal structures for integrated optics applications. In addition, detailed numerical analysis, design procedures, fabrication and characterization techniques are included. Applications presented in this paper include: channelless waveguiding, orthogonal bending of light, tunable beam splitter, all-optical analog-to-digital converter, reconfigurable optical switch, chemical/gas sensor and a three-dimensional optical interconnect bus.

We present a review of our experimental and numerical studies on the negative refraction related phenomena in two-dimensional dielectric photonic crystals (PCs). By employing photonic bands with appropriate dispersion, the propagation of the electromagnetic wave through a PC can be controlled to a large extent, and diverse and completely novel electromagnetic phenomena can be generated. We perform the spectral analysis of the negative refraction arising from a convex TM polarized photonic band of a hexagonal PC. As a consequence of negative refraction, we demonstrate a photonic crystal flat lens, which has the ability to focus electromagnetic waves and provide subwavelength resolution laterally. Finally, a photonic crystal with an embedded source is shown to provide a highly directional beam, which can be utilized in certain antenna applications.

Light showing extremely slow propagation (known as slow light) provides various effects such as spatial compression of optical signals, buffering, convolution integral calculation, beam forming, and enhancement of optical absorption, gain, nonlinearity, and so on. To generate such light, very large material or structural dispersion is used. Photonic crystal waveguides are good candidates for many device applications since they can easily generate slow light at room temperature. This paper discusses the potential of slow light devices while explaining some critical issues, such as the delay-bandwidth product and the dispersion compensation.

The physical principles behind the phenomenon of slow light in photonic crystal waveguides, as well as their practical limitations, are discussed and put into context. This includes the nature of slow light propagation, its bandwidth limitation, the scaling of linear and nonlinear interactions with the slowdown factor as well as issues such as losses, coupling into and the tuning of slow modes. Applications in all-optical signal processing appear to be the most promising outcome of the phenomena discussed.

We report on our progress made in developing microcavity photonic crystal lasers. Both suspended membrane devices and devices formed in membranes bonded to sapphire are described.

Progress in the design, fabrication and characterization of electrically injected photonic-crystal quantum-dot microcavity light sources is described. In the devices investigated in this study, the carriers are injected directly into the photonic crystal microcavity, avoiding the surface state recombination in the photonic crystal pattern. A novel and robust air-bridge contact technology is demonstrated. Spectral linewidths ∼2–3 nm are observed from hexagonal microcavities of varying sizes in the output spectra of oxide-clad microcavity devices. Narrower linewidths ∼1.3 nm are observed from air-clad devices. Results on arrays of densely packed oxide-clad photonic crystal microcavities are also presented. Spectral characteristics of oxide-clad and air-clad devices are compared.

After more than 30 years of research, hard disk drives using perpendicular recording are finally commercially available. This review is a follow-up of a review written in 1999 and addresses the basic physics of perpendicular recording with special emphasis on the read and the write process and the magnetic aspects of the recording media. The paper also surveys various technical difficulties which prevented an earlier implementation of perpendicular recording. The paper closes with a short overview of alternative technologies that allow even higher storage densities.

Intermetallic compound Er₂Ni₁₇ crystallizes in the hexagonal Th₂Ni₁₇-type structure with the space group P6₃/mmc and lattice parameters a = b = 8.282 Å and c = 8.033 Å. The magnetization measurements show that it is ferrimagnetic below its magnetic ordering temperature of 150 K. The temperature variation of zero-field-cooled magnetization in a field of 200 Oe shows the compensation point at about 63 K. On the other hand, the field-cooled magnetization in low fields is found to be negative at low temperatures. The compensation temperature is found to shift to low temperatures with an increase in the field. These variations are explained on the basis of the two sublattice model. The intersublattice molecular field coefficient (n_RT) and the intersublattice exchange coupling constant (J_RT) have been calculated using the high-field free-powder method. The n_RT and J_RT values are found to be 1.4 Tf.u./μ_B and 5.2 K, respectively. The magnetocaloric effect has been measured in terms of the isothermal magnetic entropy change, using the magnetization isotherms. The maximum entropy change is about 2 J kg⁻¹ K⁻¹ for a field of 50 kOe.

The transition and thermal stability of disc-shaped ferromagnetic particles at the temperature of T = 300 K with a uniaxial anisotropy along the symmetry axis from a bi-domain to a single domain state has been studied. The nudge elastic band method was used to map the energy landscape and to calculate the energy barrier between the transition states. For single FePt disc-shaped particles with perpendicular anisotropy three transition configurations have been found: single domain, stripe- and stable bubble domains at zero applied field. The single domain configuration along the positive anisotropy axis is reached by an annihilation process of the domain wall and the all-down state by a complex domain expansion process. Magnetization configurations in two interacting discs show an increase in thermal stability compared with single disc systems, which is attributed to the interacting magnetostatic energy between the two particles.

Au/Co/Au trilayers are interesting for a range of applications which exploit their unusual optical and electronic transport behaviour in a magnetic field. Here we present a comprehensive structural and morphological study of a series of trilayers with 0–7 nm Co layer thickness fabricated on glass by ultrahigh vacuum vapour deposition. We use a combination of in situ electron diffraction, atomic force microscopy and x-ray scattering to determine the optimum deposition conditions for highly textured, flat and continuous layered structures. The 16 nm Au-on-glass buffer layer, deposited at ambient temperature, is found to develop a smooth (1 1 1) texture on annealing at 350 °C for 10 min. Subsequent growth of the Co layer at 150 °C produces a (1 1 1) textured film with lateral grain size of ∼150 nm in the 7 nm-thick Co layer. A simultaneous in-plane and out-of-plane Co lattice expansion is observed for the thinnest Co layers, converging to bulk values for the thickest films. The roughness of the Co layer is similar to that of the Au buffer layer, indicative of conformal growth. The 6 nm Au capping layer smoothens the trilayer surface, resulting in a surface roughness independent of the Co layer thickness.

La-substituted BiFeO₃ (BFO) thin films were formed by depositing sol–gel solutions on Pt/Ti/SiO₂/Si (1 0 0) structures. No secondary phase was observed by La-atom-substitution up to 20% in BFO thin films. The micro-structures of the films were changed with an increase in La concentration, which influenced the ferroelectric properties of the films. The dielectric constant of a BFO film was approximately 90 at 1 MHz and it increased to 125 in a 20% La-substituted BFO film, which was probably due to the size effect of grains. The substitution of La atoms in BFO films reduced the leakage current as well as remanent polarization. The leakage current contribution in the hysteresis loop was negligible when the loops were measured at above 20 kHz. The obtained saturated remanent polarization was 50 µC cm⁻² in a 5% La-substituted BFO film at 100 kHz.

Magnetic Barkhausen noise and hysteresis loops have been measured in the presence of various tensile stresses from three wire-shaped amorphous magnetic materials having different domain structures. Long duration large Barkhausen pulse indicates the presence of an inner core (IC) domain in positive magnetostrictive glass-coated Co_83.2Mn_7.6Si_5.8B_3.3 and Co_58.59Fe_4.41Ni₁₀Si₁₁B₁₆ microwires where magnetization takes place by ∼180° rotation of domain magnetization. This long duration Barkhausen pulse is missing in the case of low negative magnetostrictive Co_68.1Fe_4.4Si_12.5B₁₅ wire indicating a negligible volume of IC domain in it. Short duration pulses observed in all three samples suggest the presence of a multi-domain outer shell where small angle rotation and domain wall displacement are responsible for the magnetization process. The shapes of the hysteresis loops support the Barkhausen noise studies.

We have investigated the effect of electrical conductivity of the constituents on the poling behaviour of the ceramic inclusions in 0–3 ferroelectric composites which comprise a dilute suspension of spherical particles uniformly distributed in the matrix material. A new model for the pyroelectric and piezoelectric properties in terms of the poling conditions (poling field and poling time) has been developed to include electrical conductivity. Simulated results show that conductivity plays an important role in the poling process. Properly increasing the conductivity of the matrix σ_m can enhance the polarization in the ceramic inclusion of the composite P_i, thereby making the poling of the composite more efficient. In contrast, higher conductivity of the ceramic inclusion σ_i results in lower polarization P_i, which is unfavourable to the poling of the composite. These results provide insights into the observed behaviour of 0–3 composites. The model predicts the pyroelectric and piezoelectric properties under different poling conditions, which agree well with the corresponding experimental data.

The true residual stress in La_0.8Ca_0.2MnO₃ (LCMO) thin films of various thicknesses deposited on STO substrates under the same deposition conditions was measured quantitatively by x-ray diffraction sin²ψ method. The truly strain-induced effect on the transport and magnetoresistance (MR) properties of LCMO films was investigated. The in-plane residual stress (σ₁₁) in the LCMO film is tensile, while the out-of-plane one (σ₃₃) is compressive. Moreover, the value of σ₃₃ is larger than that of σ₁₁. With increasing film thickness, the crystalline unit cell of the LCMO film reduces; also both the in- and out-of-plane components of the residual stress in the LCMO film decrease. It was found that the resistivity, T_MI and MR strongly depend on the in-plane tensile stress σ₁₁ (or/and the out-of-plane stress σ₃₃). With the increase in the in-plane stress σ₁₁ (or/and the out-of-plane stress σ₃₃), the values of resistivity and MR increase, while T_MI decreases. The truly strain-induced effect on the transport and magnetoresistance properties of LCMO film is discussed briefly.

This study is devoted to the formation of an n⁺p emitter for multicrystalline silicon (mc-Si) solar cells for photovoltaic (PV) application. The atomization technique has been used to make the emitter from H₃PO₄ phosphoric acid as a doping source. The doping emulsion has been optimized using several organic solvents. H₃PO₄ was mixed with one of these solutions: ethanol, 2-butanol, isopropanol alcohol and deionized water. The volume concentration of H₃PO₄ does not exceed 20% of the total volume emulsion. The deposit characteristics of the emulsion change with the organic solvent. H₃PO₄ : 2-butanol gives the best deposited layer with acceptable adherence and uniformity on silicon surface. Fourier transform infrared characterizations show the presence of organic and mineral phosphorous bonds in the formed layer. The obtained emitters are characterized by a junction depth in the range 0.2–0.75 µm and a sheet resistance of about 10–90 Ω/□. Such a low cost dopant source combined with a continuous spray process can effectively reduce the cost per Wp of the PV generator.

We investigate the modal characteristics of a telecommunication-oriented, phosphide photonic crystal vertical-cavity surface-emitting laser (VCSEL) as a function of the hole depth by using a three–dimensional, full vectorial plane-wave admittance method. The interplay of diffraction–waveguiding is considered and the optimal depth of the photonic crystal holes is determined, providing single mode operation with the smallest etching depth of the photonic crystal structure.

Assuming intracavity photon densities to be of Gaussian spatial distributions, the space-dependent rate equations of passively Q-switched intracavity Raman lasers are deduced for the first time for the pumping beams of Gaussian and top-head spatial distributions, respectively. The new rate equations are normalized and solved numerically to investigate the influences of the normalized initial population inversion density, normalized Raman gain coefficient, saturable absorber parameter, beam size ratio of pump to fundamental laser and loss ratio of the first Stokes to fundamental laser on the pulse parameters of the first Stokes. The results of the Gaussian and top-head pumpings show similar trends despite some discrepancies. The new theories and numerical results will help design passively Q-switched intracavity Raman lasers of high performance.

Complex current–voltage and electroluminescence (EL) investigations were performed for the single layered light emitting diodes on the base of the methoxy-modified pyrazoloquinoline (PQ) incorporated into polysilane matrices. The quantum chemical simulations performed for the isolated PQ chromophore molecule as well as for the complexes consisting of molecules and surrounding polymers show a principal role of dipole–dipole interactions for the appearance of the trapping levels substantially modifying carrier kinetics and the observed current–voltage and electroluminescent–current features. In particular, a significant change of the power index in the EL–current dependences versus the state dipole moments of the PQ chromophore determining chromophore–polymer chain interactions was established. The performed current–voltage investigations have shown that different power dependences at different voltages may indicate the inclusion of different carriers, which originate not only from the Shottky barrier jumps but also substantially due to trapping by the polaron and localized exciton states. The latter may also change the exciton kinetics and the diffusion lengths determining the power index of electroluminescent–current kinetics.

Metal–oxide–semiconductor capacitor structure with metal nanocrystals embedded in the gate oxide for the application of nonvolatile memory (NVM) is fabricated. Optimal process parameters are investigated and Au nanocrystals are adopted in this paper. High-frequency capacitance versus voltage (C–V) and conductance versus voltage (G–V) measurements demonstrate the memory effect of the structure which is shown to originate from the confined states of metal nanocrystals. Capacitance versus time (C–t) measurement under a constant gate bias is executed to evaluate the retention performance and an exponential decaying trend is observed and discussed. It is found that with respect to semiconductor counterparts, Au nanocrystals can provide enhanced retention performance, which confirms the high capacity of Au nanocrystals for NVM applications.

There is considerable difficulty in fabricating a reflector with embossing in an array substrate using a conventional single gap transflective fringe-field switching nematic liquid-crystal display. In order to solve this problem, we propose a new structure, which consists of a reflector on a colour filter substrate. The newly proposed structure with a complex field direction has problems such that the voltage-dependent transmittance and reflectance curves do not match each other, which necessitate a dual driving circuit. This paper reports the optimized electrode structure and calculated electro-optical results realizing a single gamma curve and high light efficiency.

Zn_0.8Mg_0.2O metal–semiconductor–metal ultraviolet photodiodes were fabricated on quartz by radio frequency magnetron sputtering. Dark current, spectral responsivity and pulse response experiments were carried out for the device. The photodetectors showed a peak responsivity at 330 nm. The ultraviolet-visible rejection ratio (R330 nm/R400 nm) was more than four orders of magnitude at 3 V bias. The photodetector showed fast photoresponse with a rise time of 10 ns and a fall time of 170 ns. In addition, the thermally limited detectivity was calculated to be 3.1 × 10¹¹ cm Hz^1/2 W⁻¹ at 330 nm.

Two methods to measure the diffusion coefficient of a species in a liquid by optical interferometry were compared. The methods were tested on a 1.75 M NaCl aqueous solution diffusing into water at 26 °C. Results were D = 1.587 × 10⁻⁹ m² s⁻¹ with the first method and D = 1.602 × 10⁻⁹ m² s⁻¹ with the second method. Monte Carlo simulation was used to assess the possible dispersion of these results. The standard uncertainties were found to be of the order of 0.05 × 10⁻⁹ m² s⁻¹ with both methods. We found that the value of the diffusion coefficient obtained by either method is very sensitive to the magnification of the optical system, and that if diffusion is slow the measurement of time does not need to be very accurate.

Hydrogen sensing characteristics of single nanowires of ZnO, TiO₂ and WO_2.72 have been investigated by contact mode atomic force microscopy. All these nanostructures are able to sense hydrogen, but the WO_2.72 nanowire (40 nm diam) exhibits the highest sensitivity of 22 for 1000 ppm at 298 K. The WO_2.72 nanowire is also found to be good at sensing aliphatic hydrocarbons in the form of liquefied petroleum gas with a sensitivity of 15 for 1000 ppm at room temperature. A WO_2.72 nanowire with a diameter of 40 nm shows better sensing characteristics than a nanowire of 16 nm diameter.

Efficient white organic light-emitting diodes (WOLEDs) in which the blue and red emissions come from bis[(4,6-diflourophenyl)pyridinato-N,C^2'](picolinato)iridium(III) and bis[2-(2'-benzothienyl)pyridinato-N,C^3'(acetylacetonato)iridium(III) doped 4,4'-N,N'-dicarbazole-biphenyl (CBP) layers while the green emission comes from an ultrathin non-doped fac-tris[2,5-di(4-methoxyphenyl) pyridinato-C,N]iridium(III) [Ir(dmoppy)₃] sub-monolayer are demonstrated. The electroluminescent spectra of the devices can be fine tuned by the Ir(dmoppy)₃ thickness. The optimized device with 0.25 nm Ir(dmoppy)₃ shows a maximum current efficiency, power efficiency and luminance of 10.8 cd A⁻¹, 4.5 lm W⁻¹ and 12560 cd m⁻², respectively. Atomic force microscopy images reveal that the ultrathin non-doped layer is discontinuous. When sandwiched between the CBP and bathocuproine (BCP) layers, it forms a doping profile similar to the situation when the Ir(dmoppy)₃ is doped into CBP and BCP simultaneously.

Nanostructured tin oxides have been synthesized in the form of nano-pins using SnCl₂ · 2H₂O as precursors in a short period of 2.5 min by microwave plasma enhanced CVD. The resulting nano-pins have square cross sections of 50–60 nm at the pin head, 10–20 nm at the pin proper and less than 5 nm at the pin tail; the full length of the pin is about several micrometres. The optical properties measured by cathodoluminescence show strong indigo emission at 404 and 460 nm. The nucleation and growth are dominated by a self-catalytic vapour–liquid–solid mechanism, which has been studied in this work.

Er³⁺: Ca₃La₂(BO₃)₄ crystals with dimensions of ∅20 × 30 mm³ were grown by the Czochralski method. The spectral properties of these crystals were investigated and their spectroscopic parameters were calculated and analysed based on the Judd–Ofelt theory. The oscillator intensity parameters Ω_t were calculated to be Ω₂ = 7.18 × 10⁻²⁰ cm², Ω₄ = 3.27 × 10⁻²⁰ cm² and Ω₆ = 2.79 × 10⁻²⁰ cm². The emission cross-section of a Er³⁺: Ca₃La₂(BO₃)₄ crystal was estimated to be 9.06 × 10⁻²¹ cm² at 1535 nm. The fluorescence lifetime and radiative lifetime of the ⁴I_13/2 → ⁴I_15/2 transition are 0.63 ms and 2.68 ms, respectively. The Er³⁺-doped Ca₃La₂(BO₃)₄ crystals may be regarded as 1.55 µm laser gain medium materials.

Large transient photoconductivity was observed at room temperature in oxygen-deficient La_0.6Ca_0.4MnO_3−δ/Si heterojunctions. The increase in conductivity was over one order of magnitude while applying −2.5 V reverse bias under 532 nm continuous-wave laser illumination with a power of 40 mW. It was found that the photoconductivity and the photoresponse time depend on the laser illumination power, and a rise time of about 35 ns was observed when the junction was irradiated with a 532 nm laser pulse of 7 ns duration at the light power of 750 mW. Open-circuit photovoltage with a rise time of 13 ns was also observed. The results can be explained by photon-induced hole carriers in La_0.6Ca_0.4MnO_3−δ and are important for applications in manganite-based optoelectronic devices.

Efficiency of hydrogen peroxide production by an ac driven underwater capillary discharge was investigated quantitatively. The concentration of formed hydrogen peroxide was measured by a colorimetric method using a specific reaction between H₂O₂ and a titanium reagent. The amount of formed H₂O₂ increases linearly during the first hour of the discharge duration. The initial rate and corresponding total energy yield of H₂O₂ formation by the capillary discharge were determined for initial electrical conductivity of aqueous solution in the range of 100–500 µS cm⁻¹ and average applied power in the range 30–90 W. It comes out that the total energy yield of H₂O₂ formation, derived from the initial rate of H₂O₂ formation and the average applied power, increases linearly with average applied power and that solution conductivity has only a negligible effect on the energy yield. A maximum total energy yield of H₂O₂ formation of 0.9 g kWh⁻¹ was obtained for a 500 µS cm⁻¹ aqueous solution.

The time transients of vibrational/rotational excitation up to v = 6 level of the ground electronic state of nitrogen were measured in a positive column during the 1 µs pulsed electric discharges, and in the afterglow. The peak current densities were up to 25 A cm⁻² at the gas pressure of 5 Torr. During the active discharge the energy was transferred mainly into vibrational levels, primarily above v = 1, resulting in highly non Boltzmann distribution. The vibrational distribution became Boltzmann-like approximately 80 µs after the discharge pulse and remained of this character until the next pulse. Rapid rotational heating was observed immediately following the cessation of the discharge current. The rotational temperature had risen from about 300 K during the discharge to more than 3000 K within the first 80 µs of the afterglow and returned later to ambient (300 K) in less than 100 µs. Its peak coincided with the time when the redistributed vibrational populations closely resembled the Boltzmann distribution. The vibrational populations could be described by a single temperature (approximately 3000 K) from 100 µs till nearly 1 ms after the discharge. All vibrational bands could be well described by the same rotational temperature at all times. Standard, coherent anti-Stokes Raman spectroscopy was used in all measurements.

Pinus caribea (pine) litter flame is a weakly ionized medium. Electron–neutral collisions are a dominant form of particle interaction in the flame. Assuming flame electrons to be in thermal equilibrium with neutrals and average electron–neutral collision frequency to be much higher than the plasma frequency, the propagation of microwaves through the flame is predicted to suffer signal intensity loss. A controlled fire burner was constructed where various natural vegetation species could be used as fuel. The burner was equipped with thermocouples and used as a cavity for microwaves with a laboratory quality network analyser to measure wave attenuation. Electron density and collision frequency were then calculated from the measured attenuation. The parameters are important for numerical prediction of electromagnetic wave propagation in wildfire environments. A controlled pine litter fire with a maximum flame temperature of 1080 K was set in the burner and microwaves (8–10.5 GHz) were caused to propagate through the flame. A microwave signal loss of 1.6–5.8 dB was measured within the frequency range. Based on the measured attenuation, electron density and electron–neutral collision frequency in pine fire were calculated to range from 0.51–1.35 × 10¹⁶ m⁻³ and 3.43–5.97 × 10¹⁰ s⁻¹, respectively.

The inactivation of bacteria by plasma discharges offers the unique benefits of short treatment times, minimal damage to the objects being sterilized and minimal use of hazardous chemicals. Plasmas produce reactive fluxes of ions, atoms and UV photons from any given precursor gas and are expected to be a viable method for such sterilization applications. The plasma based inactivation of harmful biological systems is, however, not yet widely used, because any validation is hampered by the limited knowledge about the interaction mechanisms at the interface between a plasma and a biological system. By using quantified beams of hydrogen atoms, argon ions and UV photons, the treatment of bacteria in a typical argon–hydrogen plasma is mimicked in a very controlled manner. As an example the inactivation of endospores of Bacillus atrophaeus is studied. It is shown that the impact of H atoms alone causes no inactivation of bacteria. Instead, the simultaneous impact of atoms and low energy ions causes a perforation of the endosporic shell. The same process occurs during plasma treatment and explains the efficient inactivation of bacteria.

Segregation of elemental Dy in a DyI₃–Hg metal-halide high-intensity discharge lamp has been observed with x-ray induced fluorescence. Significant radial and axial Dy segregation are seen, with the axial segregation characterized by a Fischer parameter value of λ = 0.215 ± 0.002 mm⁻¹. This is within 7% of the value (λ = 0.20 ± 0.01 mm⁻¹) obtained by Flikweert et al (2005 J. Appl. Phys.98 073301) based on laser absorption by neutral Dy atoms. Elemental I is seen to exhibit considerably less axial and radial segregation. Some aspects of the observed radial segregation are compatible with a simplified fluid picture describing two main transition regions in the radial coordinate. The first transition occurs in the region where DyI₃ molecules are in equilibrium with neutral Dy atoms. The second transition occurs where neutral Dy atoms are in equilibrium with ionized Dy. These measurements are part of a larger study on segregation in metal-halide lamps under a variety of conditions.

Axial segregation in metal-halide lamps is caused by a complex interaction between convection and diffusion and is not yet fully understood. By enhancing convection, by placing the lamp in a centrifuge, the effect of convection on axial segregation was studied. The centrifuge caused the lamp to be accelerated between 1g and 10g. Optical emission spectroscopy was performed on a metal-halide lamp while placed in the centrifuge. Several transitions of atomic and ionic Dy, and atomic Hg were measured at different lateral positions from which we obtained atomic and ionic Dy and atomic Hg intensity profiles. Atomic lateral profiles of Dy at different axial positions in the lamp were used for the calculation of Fischer's axial segregation parameter. The theoretical model of the Fischer curve, which shows the axial segregation parameter as a function of convection, was verified along the full range by measuring lamps of different fillings and geometry. Moreover, the radial temperature profile of the arc for the different accelerations was determined.

Thin films were produced by the reactive thermal evaporation of pure silver oxide (AgO) in a background of molecular oxygen. The effects of the deposition rate and oxygen partial pressure on the structural, chemical, electrical and optical properties of the films were investigated. The films were characterized using x-ray diffraction, x-ray photoelectron spectroscopy, electrical resistivity and normal-incidence transmittance and reflectance. The resulting films were found to be mainly metallic with a small oxide component that increased with the oxygen partial pressure.

Ba(Zr_0.2Ti_0.8)O₃ (BZT) and 2 mol% Mn additional doped BZT (Mn–BZT) thin films were deposited on Pt/Ti/SiO₂/Si by the pulsed laser deposition (PLD) technique under the same growth conditions. X-ray diffraction scans showed that both films were polycrystalline and preferentially (1 1 1)-oriented, and an enhanced crystallization effect was obtained after Mn doping. The parallel-plate capacitors of Au/BZT/Pt and Au/Mn–BZT/Pt were prepared to investigate the electric properties, respectively. The remanent polarization and the coercive electric field for Mn-doped BZT film were both smaller than those of undoped BZT film. Furthermore, Mn-doped BZT film exhibited a higher dielectric constant of 460 at zero bias, larger dielectric tunability of 69.0% and lower dielectric loss of 5.0‰ under an applied electric field of 615 kV cm⁻¹ than those of undoped BZT film. The figure of merit for preferentially (1 1 1)-oriented BZT thin film was greatly enhanced from 94 to 138 by Mn doping. The enhanced dielectric behaviour by Mn doping could be mainly attributed to the decrease in electron concentration and oxygen vacancies and the reorientation of the defect complex.

A detailed TEM investigation on the microstructure of TiN/Si₃N₄ nanocomposite coatings, which is believed to be responsible for the coatings' remarkable mechanical properties, was carried out. Parallel simulation utilizing two-dimensional TiN/Si₃N₄ nanomultilayered coatings was further performed to study whether the variation of Si₃N₄ interlayer thickness has an influence on the coatings' microstructure and mechanical properties. The results revealed that, in nanocomposite coatings with high hardness, Si₃N₄ tissue has a thickness of about 0.5–0.7 nm and exists in the crystalline state. Low-energy coherent interfaces are formed between Si₃N₄ and neighbouring elongated TiN grains. For TiN/Si₃N₄ nanomultilayered coatings, Si₃N₄ modulation layers with thicknesses less than 0.7 nm were also found to crystallize and form coherent interfaces with the neighbouring TiN layers; at the same time, the hardness of the coatings is remarkably enhanced. When its thickness exceeds 1.0 nm, Si₃N₄ transformed its growth mode into amorphous and the coherent interfaces were damaged; as a consequence, hardness enhancements in the coatings vanished. The similarity of the microstructure and the mechanical properties response between nanocomposites and nanomultilayered coatings indicates that the crystallization of Si₃N₄ as well as the formation of coherent interfaces between TiN and Si₃N₄ is the main reason for the hardening of the nanocomposites.

Copper deposition was investigated on electrochemically etched mesoporous silicon epilayers of pore diameter 7 nm, porosity 45% and surface area 500 m² cm⁻³ by immersion plating in an aqueous solution containing Cu²⁺ ions. When the layers were immersed in a 0.1 M CuSO₄ · 5H₂O aqueous solution for different time-periods ranging from 1 to 60 min, elemental copper was deposited on the surface. Fourier transform infrared (FTIR) spectroscopy in transmittance mode was carried out on copper deposited porous silicon (PS) samples to infer the changes in the surface bonding of the PS. It was clearly observed from x-ray diffraction patterns and FTIR spectra that the reduction of copper ion on PS was accompanied by gradual and finally complete oxidation of the PS epilayers surface. Scanning electron microscopy has been used to study the morphology of the copper aggregates at the PS epilayers surface.

In this work, a method has been developed to easily determine the effective diffusion coefficient (D_e) of a solute in a permeable membrane using electronic speckle pattern interferometry. Fringes are introduced parallel to the direction of diffusion during the diffusion process and D_e can be calculated by simple measurements on the interference pattern. For a fast and convenient determination of D_e, a mathematical expression has been derived from the analytical solution of diffusion in two media separated by a resistance. The D_e obtained when fringes are introduced is in agreement with that obtained when fringes are not introduced. The effect of temperature variation on the optical path of the reference and the object beams has also been investigated. The error introduced into the calculation of D_e, when the temperature oscillation is not taken into account, has been compared for the case when fringes are not introduced during the diffusion experiment and the case when fringes are introduced. In the first case, the relative error can be greater than 100%. Interestingly, in the latter case, the error caused by temperature oscillation is considerably reduced, and no error is introduced if the temperature changes homogeneously over the whole diffusion cell used for the diffusion experiment.

A 275 µm thick GaN layer was directly grown on the SiO₂-prepatterned sapphire in a home-built vertical hydride vapour phase epitaxy (HVPE) reactor. The variation of optical and structure characteristics were microscopically identified using spatially resolved cathodeluminescence and micro-Raman spectroscopy in a cross section of the thick film. The D ⁰ X_A line with the FWHM of 5.1 meV and etch-pit density of 9 × 10⁶ cm⁻² illustrated high crystalline quality of the thick GaN epitaxial layer. Optically active regions appeared above the SiO₂ masks and disappeared abruptly due to the tapered inversion domains at 210–230 µm thickness. The crystalline quality was improved by increasing the thickness of the GaN/sapphire interface, but the region with a distance of 2 µm from the top surface revealed relatively low quality due to degenerate surface reconstruction by residual gas reaction. The x-ray rocking curve for the symmetric (0 0 2) and asymmetric (1 0 2) reflections also showed good quality and a small wing tilt of the epitaxial lateral overgrowth (ELO) GaN.

We have used scanning tunnelling microscopy and spectroscopy (STM/STS) to investigate the electronic properties of ultrathin silicon oxide films on silicon, e.g. surface band gap and band offset versus silicon oxide thickness near the interface. The length of the interfacial electronically transitional region is measured to be about 0.9 nm, in which the silicon oxide surface band gap increases gradually with thickness and its evolution is meticulously observed. Both the conduction and valence band offsets between silicon oxide and silicon increase by about 1.0 eV with only a ∼0.3 nm steric difference in this region. Furthermore, in an oxide thinner than ∼0.6 nm, oxide gap states can be discerned from STS which is attributed to the electronic states extending from the silicon substrate over the Si oxide. On the basis of local layer-resolved electronic properties, we corroborate that a usable oxide thickness should be no less than ∼0.9 nm on the Si(1 1 1) substrate, which corresponds to only three initial oxide layers.

Metal copper films with thicknesses from several nanometres to several micrometres were deposited on the surface of cenosphere particles by the magnetron sputtering method under different working conditions. An ultrasonic vibrating generator equipped with a conventional magnetron sputtering apparatus was used to prevent the cenosphere substrates from accumulating during film growth. The surface morphology, the chemical composition, the average grain size and the crystallization of cenosphere particles were characterized by field emission scanning electron microscopy (FE-SEM), inductively coupled plasma–atom emission spectrometer, x-ray photoelectron spectroscopy and x-ray diffraction (XRD) analysis, respectively, before and after the plating process. The results indicate that the copper films were successfully deposited on cenosphere particles. It was found from the FE-SEM results that the films were well compacted and highly uniform in thickness. The XRD results show that the copper film coated on cenospheres has a face centred cubic structure and the crystallization of the film sample increases with increasing sputtering power.

CaCu₃Ti₄O₁₂ ceramic samples sintered at different temperatures were studied by measurements of microstructural and dielectric properties. Electron microscopy for thermally-etched samples confirmed a microstructure with subgrains (SG) buried in a crystalline grain. The detailed dielectric properties were shown to be conformable with the double-barrier-layer capacitance model. The complex electric modulus analysis revealed unambiguously each contribution of the SG interiors, SG boundaries and grain boundaries to the dielectric constant. Our result clarified how the dielectric properties change with increasing sintering temperature. It was concluded that grain boundaries rather than SG boundaries played the key role in the low-frequency giant dielectric constant in CCTO ceramics.

(Bi_1.5Zn_0.5)(Zn_0.5Nb_1.5)O₇ (BZN) films with different thicknesses and preferred orientations have been fabricated on Nb doped SrTiO₃ substrates by pulsed laser deposition. As the thickness increases, the permittivity increases, and the dielectric loss decreases, while the tunability only has a little variation. The asymmetric behaviour of the electric field dependent permittivity reduces gradually with the increasing thickness, which should be attributed to the decrease in the effect of the interfacial layer between the dielectric film and substrate (electrode). Furthermore, compared with the (1 0 0) oriented BZN film, BZN film with (1 1 1) preferred orientation exhibits high dielectric loss.

Flow electrification of polymer melts is an important side effect of polymer processing. The studies dealing with this phenomenon are few and most of the scientific work has been focused on flow electrification of aqueous and insulating Newtonian liquids. From these previous results it is well established that the flow electrification in Newtonian liquids is a consequence of the formation of an ionic double layer. Convection of this layer induces the electrification of the liquid at the outlet of the pipe. In these models, the key parameters governing the flow electrification are thus the intrinsic electrical properties of the polymer and the flow characteristics. In this work, we reconsider the assumptions made previously and we propose a new approach to modelize the flow electrification in the particular case of non-Newtonian polymer materials in laminar flow conditions. We establish that a key parameter for the electrification quantification in the polymer melt is the shape of the velocity profile. Additionally, in some cases, we show that a slip velocity at the polymer/die wall interface must be considered to accurately describe the electrification. As a consequence, we deduce that the slip velocity at the interface can be calculated by measuring the electrification: this work gives an alternative manner to measure the slip velocity during polymer flow.

AC-impedance spectroscopic studies in the temperature range 550–700 °C are carried out on undoped and Mn doped PZN-PT single crystals grown by the flux method. The variation of dielectric permittivity with temperature at different frequencies shows normal ferroelectric and relaxor-like dependence for the doped and undoped crystals, respectively. Temperature-dependent spectroscopic modulus plots reveal a much broader peak for PZN-4.5PT + 1%Mn compared with that for PZN-4.5PT, which is different from the dielectric behaviour of the doped one.

Complex modulus imaginary part (Z'') versus real part (Z') plots fit well with one semicircle thus indicating only bulk contribution. The relaxation observed in the spectroscopic plots was assigned to mobile relaxor species such as oxygen vacancies and ions. No such relaxation could be observed for PZN-4.5PT + 1%Mn in the dielectric measurements. For both undoped and Mn doped crystals, the conduction behaviour was modelled by the universal dynamic response equation and by the NTC (negative temperature coefficient) materials resistance-temperature behaviour. A large difference in behaviour was found between the two single crystals such as the thermistor coefficients and the activation energy values, which could explain the increase in the thermal stability observed in the Mn doped PZN-PT single crystals by many studies.

A simple model is proposed and tested for simulations of ceramic processing by microwave heating. The model is based on a piecewise constant approximation of the material properties and makes it possible to separate and analyse different effects caused by the sample shape and the dependence of the material properties on temperature. Specifically, the simulation results demonstrate that microwave heating of an alumina sample can be very sensitive to a variation of its dielectric constant with temperature. For different geometries, there is a similarity in the dependences of the thermal state characteristics (temperature drop across the sample, amount of dissipated power and electric field amplitude at the sample centre) on maximal temperature. It is shown also that a temperature drop between the sample centre and surface can be strongly enhanced in the case of a spherical sample irradiated symmetrically by microwaves.

Nanocrystalline, monophasic, B-site isovalent substituted BaTiO₃ with the representative formula, Ba(M_xTi_1−x)O₃ [M = Sn, Zr] have been prepared by a simple direct-precipitation method. The crystalline phase and particle size of the nanopowders are examined by XRD and transmission electron microscopy, respectively. The influence of the isovalent substituent on the Curie temperature (T_c) and the nature of the phase transitions have been determined. The dielectric properties of the solid solutions prepared by this new method are reported. Factors influencing T_c in the solid-solution systems are also reported.

This paper proposes to apply constructal theory to the optimization of ionic transfer by electrokinetics through porous media. By using an external electrical source (potential difference, or current), the ions from the pore solution of a saturated porous medium can be transferred (channelled) in an accelerated fashion, while the direction of their transport can be controlled by the electrodes position and polarity. The constructal law of maximization of flow access is used to optimize the electrokinetic process in two ways: (i) in time and (ii) in space. In (i), the ionic transport is shown to be driven by a diffusive mechanism before convection due to electrical effects which dominate the transfer. Constructal theory explains how the combination of the two mechanisms corresponds to an optimization of transport in time. In an application to ionic decontamination, the optimal location of the electrodes is determined from the constructal law by setting the diffusive ionic flow rate equal to the ionic flow rate due to electrical effects.

The shock response of the ferroelectric ceramics PZT and PSZT is monitored using maganin stress gauges in longitudinal orientation. Comparison with previously published data shows good agreement between our PZT and similar compositions. A reduction in wave speed below the Hugoniot elastic limit (HEL) with increasing impact severity is explained in terms of the degree of phase transformation between the ferroelectric and antiferroelectric phases. In contrast, the PSZT wave speeds show the opposite trend, although we believe that a similar mechanism is operating. The HELs of both materials are determined. Although the PSZT has a significantly higher strength, comparison with established trends of the HELs of this class of material with density suggests that the strength of these materials is controlled by density rather than composition.

We have investigated the effect of large in-plane compressive strain on the electrical transport in La_0.88Sr_0.12MnO₃ in thin films. For achieving large compressive strain, films have been deposited on single crystal LaAlO₃ (LAO, a = 3.798 Å) substrate from a polycrystalline bulk target having average in-plane lattice parameter a_av = (a_b + b_b)/2 = 3.925 Å. The compressive strain was further relaxed by varying the film thickness in the range ∼6–75 nm. In the film having least thickness (∼6 nm) large increase (c = 3.929 Å) in the out-of-plane lattice parameter is observed which gradually decreases towards the bulk value (c_bulk = 3.87 Å) for ∼75 nm thick film. This shows that the film having the least thickness is under large compressive strain, which partially relaxes with increasing film thickness. The T_IM of the bulk target ∼145 K goes up to ∼235 K for the ∼6 nm thin film and even for partially strain relaxed ∼75 nm thick film T_IM is as high as ∼200 K. This enhancement in T_IM is explained in terms of suppression of Jahn–Teller distortion of the MnO₆ octahedra by the large in-plane compressive strain. We observe a large enhancement in the low field magnetoresistance (MR) just below T_IM in the films having partial strain relaxation. Thick films of 6 and 20 nm have MR ∼14% at 3 kOe that almost doubles in 35 nm film to ∼27%. Similar enhancement is also obtained in the case of the temperature coefficient of resistivity. The near doubling of low field MR is explained in terms of delocalization of weakly localized carriers around T_IM by small magnetic fields.

Multi-component gas diffusion in the continuum flow regime is often modelled using the Stefan–Maxwell (SM) equations. Recent advances in lattice Boltzmann (LB) mass diffusion models have made it possible to directly compare LB predictions with solutions to the SM equations. In this work, one-dimensional (1D) and two-dimensional (2D), equi-molar counter-diffusion of two gases in the presence of a third, inert gas is studied. The work is an extension and validation of a recently proposed binary LB model for components having dissimilar molecular weights. The treatment of inflow and outflow boundary conditions (for specifying species mole fractions or mole flux) is developed via the averaging of component velocities before and after collisions. Results for one and two spatial dimensions have been compared with analytic and numerical solutions to the SM equations and good agreement has been found for a wide range of parameters and for large variations in molecular weights. A novel molecular weight tuning strategy for increasing the accuracy has been demonstrated. The model developed can be used to model continuum, multi-component mass transfer in complex geometries such as porous media without empirical modification of diffusion coefficients based on porosity and tortuosity values. An envisioned application of this technique is to model gas diffusion in porous solid oxide fuel cell electrodes.

A gap is one of the most important issues to be solved in laser welding of a micro butt joint, because the gap results in welding defects such as underfilling or a non-bonded joint. In-process monitoring and adaptive control has been expected as one of the useful procedures for the stable production of sound laser welds without defects. The objective of this research is to evaluate the availability of in-process monitoring and adaptive control in micro butt welding of pure titanium rods with a pulsed neodymium : yttrium aluminium garnet (Nd : YAG) laser beam of a 150 µm spot diameter. It was revealed that a 45 µm narrow gap was detected by the remarkable jump in a reflected light intensity due to the formation of the molten pool which could bridge the gap. Heat radiation signal levels increased in proportion to the sizes of molten pools or penetration depths for the respective laser powers. As for adaptive control, the laser peak power was controlled on the basis of the reflected light or the heat radiation signals to stably produce a sound deeply penetrated weld reduced underfilling. In the case of a 100 µm gap, the underfilling was greatly reduced by half smaller than those made with a conventional rectangular pulse shape in seam welding as well as spot welding with a pulsed Nd : YAG laser beam. Consequently, the adaptive control of the laser peak power on the basis of in-process monitoring could reduce the harmful effects due to a gap in micro butt laser welding with a pulsed laser beam.

An extension of a series solution of the non-linear fin problem with temperature-dependent thermal conductivity [Kim and Huang 2006 J. Phys. D: Appl. Phys.39 4894–901] is done to include a temperature-dependent heat transfer coefficient as well. It is assumed that the heat transfer coefficient is expressed in a power-law form and the thermal conductivity is a linear function of temperature. The algorithm is based on the Taylor series solution. The proposed approximate solution is successfully compared with the solutions for the Adomian decomposition method (ADM) and numerical algorithm. Also, it has been shown that the Adomian decomposition solution is just an approximation of the series solution given in this work, and therefore the present algorithm is better than ADM.

The optimization of transverse gradient coils is one of the fundamental problems in designing magnetic resonance imaging gradient systems. A new approach is presented in this paper to optimize the transverse gradient coils' performance. First, in the traditional spherical harmonic target field method, high order coefficients, which are commonly ignored, are used in the first stage of the optimization process to give better homogeneity. Then, some cosine terms are introduced into the series expansion of stream function. These new terms provide simulated annealing optimization with new freedoms. Comparison between the traditional method and the optimized method shows that the inhomogeneity in the region of interest can be reduced from 5.03% to 1.39%, the coil efficiency increased from 3.83 to 6.31 mT m⁻¹ A⁻¹ and the minimum distance of these discrete coils raised from 1.54 to 3.16 mm.

In a recent paper, Wang et al (2006 J. Phys. D: Appl. Phys.39 3030) have reported the effects of hydrogen dilution ratio on the relative densities of SiCl_n (n = 0–2) radicals in a SiCl₄/H₂ plasma generated in a RF low pressure discharge. This comment argues that the radical detection method proposed and used by Wang et al is not appropriate to estimate radical concentrations and that the experimental results obtained with it consequently seem to be unreliable.

It is argued that the new method for measuring neutral radicals in RF low pressure discharge plasma presented by Wang et al (2006 J. Phys. D: Appl. Phys.39 3030) is correct. The feasibility and reliability of this new method are discussed using the latest experiment carried out by Wang et al (2007 Int. J. Mass. Spectrosc.261 25) where details of their proposed new method are also given.