Table of contents

This review focuses on the peculiarities of quasiperiodic order for the properties of photonic and phononic (sonic) heterostructures. The most beneficial feature of quasiperiodicity is that it can combine perfectly ordered structures with purely point-diffractive spectra of arbitrarily high rotational symmetry. Both are prerequisites for the construction of isotropic band gap composites, in particular from materials with low index contrast, which are required for numerous applications. Another interesting property of quasiperiodic structures is their scaling symmetry, which may be exploited to create spectral gaps in the sub-wavelength regime. This review covers structure/property relationships of heterostructures based on one-dimensional (1D) substitutional sequences such as the Fibonacci, Thue–Morse, period-doubling, Rudin–Shapiro and Cantor sequence as well as on 1D modulated structures, further on 2D tilings with 8-, 10-, 12- and 14-fold symmetry as well as on the pinwheel tiling, the Sierpinski gasket and on curvilinear tilings and, finally, on the 3D icosahedral Penrose tiling.

We have used single crystal x-ray diffraction methods to establish the crystal structures of a compound in the Pd–Mn–Si system and in the Pd–Mn–Ge system. The title compounds have structures related to the Fe₂P structure type and are ferromagnetic with Curie temperatures above the room temperature. Density functional electronic structure calculations help to understand the nature of the local moment ferromagnetism in these compounds. However, neither the electronic structure calculations nor the magnetic measurements provide any evidence of half-metallic behaviour.

A new zinc-blende type photoelectronic single crystal, Hg_(3−3x)In_2xTe₃ (MIT), was grown by using the vertical Bridgman method. The structure of the MIT (x = 0.5) crystal was examined by x-ray powder diffraction. The lattice parameters were determined to be 6.2934 ± 0.0011 Å with a defect zinc-blende structure, which belongs to the group . The melting point of the MIT crystal was measured by using the differential scanning calorimetry. Hall effect measurements on electrical properties at room temperature show that resistivity, carrier density and mobility of MIT crystal were 4.79 × 10² Ω cm, 2.83 × 10¹³ cm⁻³ and 4.60 × 10² cm² V⁻¹ s⁻¹, respectively. The electron effective mass at the bottom of the conduction band and the top of the valence band was calculated. It was found that the Fermi level of the MIT crystal was about 0.008 eV higher than that of the mid-gap. The experimental carrier concentration was in good accordance with the theoretical result.

The semiconductor films of SrTiO₃, α-Fe₂O₃ and their heterojunction SrTiO₃/α-Fe₂O₃ were prepared on FTO (SnO₂ : F on glass) substrates by the spin-coating method and their structures and properties were characterized by XRD, SEM and photoelectrochemical measurements, respectively. It was observed that the photocurrent and the incident photon to current conversion efficiencies (IPCE or external quantum efficiencies) of the SrTiO₃/α-Fe₂O₃ heterojunction were higher than that of the single SrTiO₃ or α-Fe₂O₃ film, particularly under visible light irradiation. The band structures of SrTiO₃ and α-Fe₂O₃ were calculated. It was found that the charge carriers in α-Fe₂O₃ had large effective masses while the charge carriers in SrTiO₃ had small effective masses at the bottom of the conduction band and at the top of the valence band, which meant weak mobility of holes photo-generated in α-Fe₂O₃ and strong mobility of holes excited in SrTiO₃. As to the SrTiO₃/α-Fe₂O₃ heterojunction, the special valence band structures of SrTiO₃ improved the separation efficiency of charge carriers which had been photo-generated in α-Fe₂O₃. This led to the enhancement of photocurrent densities and external quantum efficiencies in the SrTiO₃/α-Fe₂O₃ heterojunction photoanode.

The effect of the frequency on the breakdown time in a straight discharge tube is investigated by means of a fluid model. The discharge tube is similar to a compact fluorescent lamp tube, containing argon at 3 Torr and mercury at a few Torr. The mechanism of breakdown at frequencies of the order of several 10 kHz is considered and related to breakdown at a dc voltage. During a negative potential on the powered electrode, an ionization wave traverses the tube in a way similar to that in a dc operated tube. During a positive potential on the powered electrode, the electric field in the part of the tube already traversed by the ionization wave is enhanced by negative charge on the inner wall of the tube. Although the ionized region does not extend during this phase, the ionization density increases substantially. Furthermore, we investigated the dependence of the breakdown time on the applied frequency and found that the breakdown voltage is independent of the frequency. This is shown to be consistent with experimental data.

A study was made to verify the influence of the gas mixture flow on the iron sintering process with simultaneous surface enrichment of alloying elements by hollow cathode discharge. In this process, two independent cathodes formed an annular discharge: (1) a pressed cylindrical sample of iron powder, acting as the central cathode, was placed concentrically inside an external (hollow) cathode; (2) the external cathode, machined from a AISI 310 steel bar, acted both to confine the geometry of the plasma and as a source of alloying elements (Cr and Ni). The sintering was carried out at 1423 K, for a period of 7.2 × 10³ s, under a gas mixture of 80% Ar + 20% H₂ and a pressure of 399 Pa, at flow rates of 2 × 10⁻⁶, 5 × 10⁻⁶, and 8 × 10⁻⁶ m³ s⁻¹, with an inter-cathode radial space of 5.8 mm. The discharge was generated using a pulsed voltage power source with a 200 µs period. The gas mixture flow plays an important role both in the cleanliness of the sintering atmosphere (reflected in the electric power utilized to maintain the sample's temperature) and in the amount of metallic atoms deposited on the sample's surface (as a result of the sputtering and the oxidation/reduction process on the cathode surfaces).

A new non-invasive method for the determination of electron drift velocity has been investigated by measuring the light propagation signals in external electrode fluorescent lamps with two usual operation methods. At the luminance of 12 000 cd m⁻² with an operation frequency of 65 kHz, the electron drift velocity in accordance with the propagation speed of light emission is measured to be 2.72 × 10⁵ m s⁻¹. From the electron drift velocity, the electron temperature and the plasma density in the positive column are estimated to be kT_e ∼ 1.6 eV and n_e ∼ 4.3 × 10¹⁶ m⁻³, respectively, which agree well with data in the glow discharge plasma of typical fluorescent lamps.

This paper focuses on plasma etching and XPS surface analyses of new dielectric materials used in integrated circuits. We investigate by XPS surface modifications of methylsilsesquioxane low-k polymer (SiOC(H)) and amorphous hydrogenated silicon carbide (SiC(H)) when exposed to Ar, SF₆ and C₂F₆-based high density plasmas. Ar and SF₆ plasmas remove the carbonaceous groups from the surface leading to the formation of fluorinated top layers (SiOF and SiF-like layers) in SF₆ plasma. In the case of C₂F₆-based mixtures, the surface structure fits well with a two-layer model, consisting of a fluorocarbon top layer above a fluorinated interaction layer (SiOF or SiF) on the bulk materials (SiOC(H) or SiC(H)). Determination of top layer thicknesses from XPS data is discussed. We show that material etch rate is not correlated with the total modified thickness, in contrast to the top fluorocarbon layer thickness which is in good correlation with etch rates. Etch yields (etch rates divided by the ion flux) are calculated and are studied in order to draw material etch mechanisms in C₂F₆ based mixtures. We conclude that for both materials, etching mechanisms in C₂F₆/H₂ mixtures, and to a lesser extent in C₂F₆/Ar mixtures, are close to those of silicon in fluorocarbon plasmas and different from those of the conventional interlevel dielectric material SiO₂. On the contrary etching mechanisms in C₂F₆/O₂ mixtures are similar to those of silicon dioxide in fluorocarbon plasmas.

Plasma properties in a highly constricted arc were measured using spectroscopic techniques. The arc configuration studied is typical of those produced by plasma cutting equipment. Temperatures and electron densities in the arc were measured using several different emission lines and several different methods for property calculation. Factors such as fluctuations in the arc intensity, arc asymmetry and the validity of the assumptions used for property calculation were shown to affect the accuracy of the measurements. Selection of the temperature measurement method can influence the measurement results. For this arc configuration, methods which use oxygen ion emission lines provide the most accurate results. The measurement methods developed in this study were used to investigate differences in arc properties related to arc current and nozzle diameter. Large changes in the temperature and electron density due to shocks were measured in the under-expanded region near the nozzle exit.

An original dc double anode plasma torch which provides a long-time and highly stable atmospheric plasma jet has been devised for the purpose of hazardous waste treatment. The arc fluctuations and dynamic behaviour of the argon and argon–nitrogen plasma jets under different operating conditions have been investigated by means of classical tools, such as the statistic method, fast Fourier transform (FFT) and correlation analysis. In our experiments, the takeover mode is identified as the fluctuation characteristic of the argon plasma jet while the restrike mode is typical in the argon–nitrogen plasma dynamic behaviour. In the case of pure argon, the FFT and correlation calculation results of electrical signals exhibit the only characteristic frequency of 150 Hz, which originates from the torch power and is independent of any change in the operating conditions. It indicates that the nature of fluctuations in an argon plasma jet is mainly induced by the undulation of the tri-phase rectified power supply. In contrast, besides the same low frequency bulk fluctuation, the dynamic behaviour of the argon–nitrogen plasma jet at high frequency (4.1 kHz) is ascribed to the rapid motion of both arc roots on the anode surface. In addition, it is found that each arc root attachment is rather diffused than located at a fixed position on the anode wall in the argon plasma jet, while constricted arc roots occur when nitrogen is added into argon as the plasma working gas.

The cathode spot movements on a polyacrilonitrile (PAN)-based carbon felt reinforced C/C composite and a three dimensional PAN-based carbon fibre reinforced C/C composite (3D-C/C) were investigated by a scanning electron microscope and a digital high-speed video camera. It was found that the carbon fibres have a higher ability to withstand the vacuum arc erosion than the carbon matrix. The cathode spot walks on the matrix, rather than on the carbon fibres. The cathode spot motion is controlled by the architecture of carbon fibres in C/C. The cathode spots move along the carbon fibres by a step-by-step manner rather than a random walk. The cathode spot tracks spread over a wide zone on the 3D-C/C surface parallel to the carbon fibre. The average arc spreading velocity is estimated to be about 0.9 m s⁻¹ and the transient arc spreading velocity is in the range of 0.54–4.5 m s⁻¹.

Nanoindentation-induced mechanical deformation in GaN thin films prepared by metal-organic chemical-vapour deposition was investigated using the Berkovich diamond tip in combination with the cross-sectional transmission electron microscopy (XTEM). By using focused ion beam milling to accurately position the cross-section of the indented region, the XTEM results demonstrate that the major plastic deformation was taking place through the propagation of dislocations. The present observations are in support of attributing the pop-ins that appeared in the load–displacement curves to the massive dislocation activities occurring underneath the indenter during the loading cycle. The absence of indentation-induced new phases might have been due to the stress relaxation via the substrate and is also consistent with the fact that no discontinuity was found upon unloading.

Electron microscopy has been used to determine directly the effect of artificially introduced roughness on the micromagnetic processes that occur in soft high moment multilayer films. The nanodefects, introduced to roughen the substrate, and the local magnetic domain structure were identified in the same images, leading to unambiguous information on the role played by the former. Characteristic of the micromagnetic state of the samples was a high density of 360° wall segments that showed quite remarkable resistance to annihilation. Following a description of the domain walls generated in both easy and hard axis magnetization cycles, a model is proposed for the way the observed domain walls respond when their ends are strongly pinned and, using this, we account for their continued existence over an extended number of magnetization cycles. Finally, the implications for device performance are discussed briefly.

Single-crystalline layered SnO₂ microrods were synthesized by a simple tin–water reaction at 900 °C. The structural and optical properties of the sample were characterized by x-ray powder diffraction, energy-dispersive x-ray spectroscopy, scanning electron microscopy, high resolution transmission electron microscopy, Raman scattering and photoluminescence (PL) spectroscopy. High resolution transmission electron microscopy studies and selected area electron diffraction patterns revealed that the layered SnO₂ microrods are single crystalline and their growth direction is along [1 1 0]. The growth mechanism of the microrods was proposed based on SEM, TEM characterization and thermodynamic analysis. It is deduced that the layered microrods grow by the stacking of SnO₂ sheets with a (1 1 0) surface in a vapour–liquid–solid process. Three emission peaks at 523, 569 and 626 nm were detected in room-temperature PL measurements.

Diffusion of Au and its effects on structural, electrical and optical properties of CdTe films fabricated by the close-spaced sublimation technique have been investigated. Diffusion of Au was studied in the range 400–550 °C using energy dispersive x-ray fluorescence analysis. Au-doped and un-doped CdTe films were characterized by x-ray diffraction (XRD), electrical and optical absorption measurements. The temperature dependence of the diffusion coefficient of Au in CdTe films is described as D = 4.4 × 10⁻⁷exp(−0.54 eV/kT). The mechanism of Au diffusion in polycrystalline CdTe films is attributed to the fast migration of Au along grain boundaries with simultaneous penetration into grains and settling on Cd-vacancies. It is supposed that the weak influence of Au diffusion on XRD patterns of CdTe films can be explained by dispersal of Au atoms preferentially on Cd-vacancies owing to proximity of the covalent radius of Au and Cd. Au atoms, placed on Cd-vacancies (Au_Cd) during fast cooling from diffusion temperature to room temperature, show an acceptor behaviour with an energy level about of E_v + 0.2 eV. The nature of this level is discussed.

Diamond multilayer film with nanocrystalline diamond film coated on a layer of diamond nanocone film, was synthesized on silicon substrates by using in situ graphite etching in a hot filament chemical vapour deposition reactor. Many small and sharp protrusions on the surface of the multilayer film were observed using scanning electron microscope and atomic force microscope. The multilayer film exhibits lower turn-on field and larger emission current than flat diamond films and diamond nanocone films. Field electron emission measurements at different residual gas pressures indicate that emission current of the multilayer film shows no obvious degradation and the Fowler–Nordheim (F–N) plots parallel to each other when the residual gas pressure is increased from 10⁻⁷ to 10⁻⁵ Torr. At a higher pressure of 10⁻³ Torr, obvious degradation of emission current was observed. The slope of the F–N plot becomes larger, indicating that the work function is increased. The possible reason is that the surface of the multilayer film is fully covered by residual gas molecules.

Room temperature photoluminescence (PL) from plasma-polymerized hexamethyldisiloxane (PP-HMDSO) thin films deposited on silicon wafers has been investigated as a function of both the applied RF power and the monomer flow rate. Films were deposited in a low pressure–low temperature remote plasma ignited in a 13.56 MHz hollow cathode discharge reactor, using pure HMDSO as a monomer and Ar as a feed gas. The substrate temperature during the deposition was as low as 40 °C and the total pressure was about 0.03 mbar. Optical emission spectroscopy (OES) has been used as in situ tool for monitoring the different chemical species present in the plasma during deposition processes. The deposited PP-HMDSO films showed a strong, broad 'green/yellow' PL band. The RF power and the flow rate of the HMDSO monomer are found to have a significant impact on the PL intensity of the deposited film. The changes in the chemical bonding of the film as a function of deposition parameters have been investigated by using the Fourier transform infrared (FTIR) spectroscopic analysis and are related to PL and OES results. The 'green/yellow' PL band is ascribed to chemical groups and bonds of silicon, hydrogen and/or oxygen constituting the films, in particular, SiH, SiO bonds and silanol Si–O–H groups.

In the present study, the elastic properties of Ti_1−xAl_xN (0 ≤ x ≤ 0.75) hard coatings were studied by ab initio calculations. The bulk (B), shear (G) and Young's (E) moduli and Poisson's ratio (υ) were calculated for isotropic materials. In addition, the anisotropy of the elastic moduli (G₁₂, G₂₃, E₁₁, E₂₂) and the Poisson's ratios (ν₁₂, ν₂₁, ν₂₃) are described with respect to particular crystallographic planes. The results of the calculations enable us to relate the experimentally observed increase in hardness to an increasing Al content, at least within the range 0 ≤ x ≤ 0.5, to an increase in G and E and to a decrease in B. The calculated longitudinal Young's modulus E₁₁(1 1 1) of Ti_1−xAl_xN (0 ≤ x ≤ 0.4) is in good agreement with experimental data taken from the coatings with the (1 0 0) preferred orientation.

This paper investigates the combined torsional buckling of multi-walled carbon nanotubes (MWNTs) coupling with radial pressures. The analysis is based on the continuum mechanics model, and the effect of the van der Waals interaction between adjacent tubes is taken into account. A buckling condition is derived for determining the critical shear membrane force under combined torsional buckling, which clearly indicates the role of radial pressures. The critical shear membrane force and the buckling mode are worked out for three typical MWNTs subjected to various internal pressure or external pressure. It is shown that the effect of internal pressure or external pressure on the critical shear membrane force for combined torsional buckling of MWNTs is related to the types of MWNTs. This effect is strong for thin MWNTs, moderate for thick MWNTs and small for solid MWNTs. Numerical results also indicate that the buckling mode corresponding to the critical shear membrane force of MWNTs is unique and only dependent on the structure of MWNTs. In particular, for combined torsional buckling of MWNTs with very small internal pressure or external pressure, the buckling mode is just that for the corresponding pure torsional buckling.

Whisker growth plays an important role in determining the performance and lifetime of micro/nano electronic devices. In this work, we analyse the contribution of lattice diffusion to the whisker growth in a bilayer structure. An analytical solution of the growth rate is obtained as a function of the average chemical stress and the whisker size. The growth rate is inversely proportional to the whisker size for the growth of a whisker over a semi-infinite crystal, while it is proportional to the thickness of the surface layer for larger ratio of the whisker size to the thickness of the surface layer. The analytical result provides us a rational basis to evaluate the growth behaviour of metallic whiskers and nanowires controlled by the lattice diffusion.

Kinetic studies were carried out on CaWO₄ thin films prepared by chemical bath deposition. It was found that the deposition temperature and the pH value of the solution can significantly affect the reaction velocity constant k. From these kinetic studies we can clarify the dependence of experimental parameters on the growth of thin films and provide a theoretical view to control the process of the deposition.

Composite materials consisting of powders of a standard hydride forming an intermetallic compound (LaNi₅) dispersed into a hydrogen-permeable elastomer (polysiloxane) have been produced. The H₂ storage capacity and the hydrogen desorption kinetics have been measured with composite samples having a metal content of 50 and 83 wt%. At a metal content of 50 wt%, the composite material consists of separated LaNi₅ particles embedded into the polymeric matrix: this composite sample shows low H₂ storage capacity that is attributed to the chemistry of the metal–polymer interface limiting the H₂ absorption by the metallic particles. The amount of absorbed hydrogen significantly increases at the metal content of 83%, when a percolative distribution is assumed by the metallic particles which are connected through a high volumetric density of metal–metal interfaces. The rate limiting process in the hydrogen desorption kinetics is attributed to the H diffusion process in the metallic fraction of the composite.

The absorption of water from a slurry into an absorbent substrate is analysed using Sharp Front theory. The analysis describes the relationship between the sorptivity S of the substrate, the desorptivity R of the slurry and the transfer sorptivity A between slurry and substrate, and leads to the relationship 1/A² = 1/R² + 1/S². Experimental data are presented which validate this equation for the practically important case of the absorption of water from soft mortar mixes by fired clay bricks. A unique feature of the experimental work is the measurement of the desorptivity of the mortars at a pressure equal to the wetting front capillary pressure of the clay brick substrate. Analysis of the experimental data also enables, for the first time, the calculation of the capillary potential at the slurry/substrate interface. The analysis has relevance to many aspects of ceramic and mineral processing, industrial filtration and construction engineering.

Hydrated cement pastes with different water-to-cement ratios have been investigated by the small-angle scattering neutron technique. Special attention has been paid to the fractal nanostructure of the calcium silicate hydrate (C–S–H solid gel) and its basic building particles, i.e. nanometric globules. The inner stability of these particles has been tested and shown to be sufficiently persistent to withstand all spatial regroupings of the overall nanostructure caused by variations of w/c ratios.

This paper describes the use of interferometric photothermal microscopy for measuring the thermal diffusivity, crystallographic orientation and elastic anisotropy of a microscopic homogeneous volume of matter. This purely optical technique makes use of an intensity-modulated and focused laser beam to periodically heat the surface of the sample tested. An interferometer is used to detect the thermal expansion. After describing the experimental setup, we explain the inversion scheme allowing the determination of the local thermal diffusivity from the periodic sample surface displacement map, as well as the crystallographic orientation and elastic anisotropy in the case of cubic materials. These parameters are measured with a spatial resolution of a few cubic micrometres.

This work presents a study of the thermal lens technique to quantify the thermo-optical coefficients ds/dT (optical path change with temperature), thermal diffusivity and conductivity of tellurite glasses doped with Yb³⁺. The glasses present binary, ternary and quaternary compositions. Yb³⁺ ions allow the determination of the quantum efficiency, radiative and experimental lifetime of the ²F_5/2 → ²F_7/2 Yb³⁺ transition, as well as the fraction of energy converted into heat. It is pointed out that ds/dT is strongly dependent on the glass composition. It is shown that the quaternary composition presents the highest K, D and low ds/dT suggesting that it could be used to develop high temperature photonic devices.

We analyse spreading of a density front in the Küntz–Lavallée model of porous media. In contrast to previous studies, where unusual properties of the front were attributed to anomalous diffusion, we find that the front evolution is controlled by normal diffusion and hydrodynamic flow, the latter being responsible for apparent enhancement of the front propagation speed. Our finding suggests that results of several recent experiments on porous media, where anomalous diffusion was reported based on the density front propagation analysis, should be reconsidered to verify the role of a fluid flow.

We present a methodology for the stress–strain analysis of a film/substrate interface by combining crystallographic and continuum modelling. Starting from measurements of lattice parameters available from experimental observations, the heterostructure is recast initially in the form of a crystallographic model and finally as a continuum elastic model. The derived method is capable of handling dense arrays of misfit dislocations as well as large areas of the interface between two crystal structures. As an application we consider the misfit dislocation network in the GaN/Al₂O₃ interface region through determination of strain relaxation and associated residual stresses. Our calculated results are referred back to and found to be in good agreement with the experimental observations of misfit dislocation arrays obtained from high resolution transmission electron microscopy.

In a recent paper, Nemchinsky presents an analysis of the 'rocket' effect on the particle dynamics under conditions of thermal plasma spraying and claims that this effect is quite substantial with iron particles evaporating in an argon plasma flow as the calculation example. The same problem is re-examined in this communication by considering the intense evaporation of particles as a combined heat and mass transfer process and treating the 'rocket' force as a part of the total drag force acting on the evaporating particle. It is shown that the 'rocket' force caused by the non-uniform distribution of the evaporated-mass efflux from the evaporating particle makes up only a small percentage of the total drag force and thus is not important in determining the particle dynamics for the studied case.

The heating history of a droplet during its flight can be divided into two phases: (a) the initial phase when evaporation, although it occurs, does not change the heat balance of the droplet much (the case considered in our previous paper and (b) the final phase when the cooling due to evaporation balances the heat flux from the plasma. The later phase is considered in Chen's comment. In our reply, a very straightforward consideration demonstrates that even in the final phase of the droplet flight, the 'rocket' effect can be significant.

The molecular continuum radiation of high intensity discharges (HID) containing the rare-earth iodides TmI₃, HoI₃ and DyI₃ in quartz burners have been investigated. For the case of TmI₃ a more detailed analysis has been performed. Two main molecular emission contributions have been identified, being located in the arc core and in the mantle region. These molecular emissions turned out to deliver half of the total emission of this type of discharge. The mantle emission was identified as originating from the diatomic TmI. Two distinct upper energy levels at about 4 eV have been determined, being separated about 0.4 eV from each other. A pragmatic approach has been chosen to describe the spectral molecular emission coefficient as a sum of three Gaussian profiles.

The proposed method allows one to take into consideration molecular emission for modelling of HID containing rare-earth iodides.

In this paper we present the results of physical investigations on mercury-free automotive discharge lamps—also known as the so-called Xenon lamps. In this type of lamp we replace Hg by ZnI₂ for environmental reasons while keeping the standard metal halide mixture consisting of NaI and ScI₃. In order to obtain excellent lamp performance, plasma properties and wall temperature distributions of the discharge vessel (burner) are analysed as a function of burner geometry and lamp filling. Special focus is on the control of arc bending and arc diffusity to match the optical requirements using standard reflectors. In addition, wall temperatures need to be adapted in the case of Hg-free operation. From physical simulations and experimental studies it is concluded that elliptically shaped burners are preferred with respect to optical and thermal lamp properties as compared with cylindrical geometries. This work is part of a cooperative project and has been supported by the German Federal Ministry of Education and Research, Germany.

In the context of developing a mercury-free lamp for automotive lighting an optimization of lamp design and plasma radiation was performed. In contrast to existing quartz lamps, a new lamp design was chosen consisting of a sapphire capillary combined with ceramic parts and pure tungsten electrodes. The lamps consist of metal halides and between 1 and 20 bar of xenon. They were operated with an electronic ballast at input powers between 20 and 40 W. Besides an optimization of the tube materials, thermography and plasma diagnostics were performed to understand the processes inside the lamp and to find optimal operation conditions. In addition to experimental diagnostics, a simulation of self-reversed sodium spectral lines was performed to verify plasma parameters, particularly the xenon pressure which could not be determined from the experiment. Additionally, tendencies of the influence of single components could be estimated by modelling.

A mercury-free high-pressure discharge in a quartz vessel is investigated which achieves a high luminous efficacy of more than 90 lm W⁻¹ and a good colour rendering index of more than 70. It is shown that the favourable properties are achieved by a combination of Xe and AlI₃ as mercury is substituted with an admixture of TlI and TmI₃. The molecular radiation of the mono-iodide TmI, which is present in large areas of the constricted temperature distribution of the plasma, dominates the spectral radiant flux.

Electric-discharge plasmas in gallium-iodide vapours are experimentally found to convert 40% of input electric power into ultraviolet and visible radiation (200–800 nm). The conditions are a weakly ionized positive column consisting of 5–10 Torr argon, and the gallium-iodide vapour is formed by heating condensed gallium-iodide to 100–120 °C. The input power density is 50–100 mW cm⁻³. The plasma is contained in a sealed silica tube and excited by an external radiofrequency antenna. Computational analysis and plasma diagnostics lead to a quantitative understanding that gallium atoms are formed by electron-impact dissociation of gallium-iodide compounds that evaporate into the plasma volume, and that further electron collisions excite the gallium atoms, which then decay by photon emission. High efficiency is possible only because several photons are emitted per dissociation event, and because nonradiative power channels such as electron-impact elastic heating and vibrational excitation are not dominant. The dissociated species recombine on the wall to reform the species that evaporates. The plasma properties change discontinuously as the molar ratio of iodine to gallium (I/Ga) in the system crosses the values I/Ga = 3 and I/Ga = 2, consistent with the thermodynamic properties of condensed gallium-iodide compounds.

Present-day computational techniques provide a possibility of evaluating properties of macrosystems using ab initio quantum chemistry and theories of elementary processes. Physical and chemical phenomena on very different timescales have to be taken into account (excitation, emission, chemical reactions, diffusion) at different levels of refining. This refining covers a very wide region of parameters starting from the structure of species up to the macro chemical mechanism of their conversion. This multilevel approach is described in detail in the paper and includes interaction and data transfer between different levels of phenomena description. In the framework of the approach, unknown properties of molecules, ions and atoms (structure, potential energy curves, transition dipole moments) are calculated based on quantum-chemical methods. The calculation results are used to evaluate rate characteristics of physical and chemical processes. The developed kinetic state-to-state scheme is then used to calculate the macro properties of the system under investigation. As an example of the multilevel approach, the emission properties of the Ar–GaI₃ positive column discharge plasma were calculated using the Chemical Work Bench computational environment. The calculations yield the electron energy balance and emission efficiency as functions of plasma parameters.

Ac-driven low-pressure He–Xe lamp discharges have been studied by experiment and self-consistent modelling in a wide parameter range. The calculated periodic behaviour of the electrical characteristics, the axis densities of the lowest excited xenon atoms and the output power of VUV radiation are compared with the results of electrical measurements and of tuneable diode laser absorption spectroscopy. Special attention is paid to the experimental studies of the behaviour of the xenon states Xe(1s₃) and Xe(1s₂). The influence of the pulse frequency on the discharge and output characteristics is discussed regarding the corresponding results for a dc discharge. Using the verified model, optimum discharge conditions with respect to high efficiency and output power of the VUV radiation have been revealed.

As the environmental awareness of people becomes stronger, the demand for mercury-free light sources also becomes stronger. The authors have been keeping their interest in developing mercury-free discharge lamps, especially in low-pressure xenon as the most promising UV sources to substitute mercury. In this paper the authors report the effect of auxiliary external electrode (AEE) on the pulsed xenon fluorescent lamps. In this research two types of structures are used. One is the standard type, which has one anode and one cathode inside the lamp. The other is the double anode type, which has two anodes and one cathode inside the lamp. By using the AEE the luminance and the efficacy are improved simultaneously. The luminance is improved by about 129% for the standard structure type and 20% for the double anode type. The efficacy is improved by 60% for the standard type and by 63% for the double anode type. We call this effect/method the Jinno–Motomura effect/method (JM-effect/method). Regarding the result of the xenon metastable atom distribution, measured by LIF, this effect is attributed to the plasma potential control and enlargement of the positive column by AEE.

Excimer emission at 172 nm was observed from xenon discharges generated between a perforated anode, with opening dimensions in the sub-millimetre range, and a planar cathode. A thin dielectric layer 100–250 µm in thickness, with the same size opening as the anode, is aligned with the anode opening and used to separate the electrodes. Devices with this structure are referred to as cathode boundary layer (CBL) discharge or micro-hollow cathode discharge devices, depending on the surface structure of the cathode. The emission intensity and efficiency of these devices are pressure- and current-dependent. Typical power densities and internal efficiencies (ratio of excimer radiant power to electrical input power) are 0.5–1.5 W cm⁻² and 3–5%, respectively. In the current range between normal and abnormal mode operation, the CBL discharge shows regularly arranged filaments (self-organization). Optimum emission of the excimer radiation is observed at the transition from the normal glow mode to self-organization. The resistive current–voltage characteristic in the self-organization region allows the operation of multiple CBL devices in parallel without individual ballast, but with an excimer emission slightly off the maximum value. The measured decrease of the excimer emission to about 10% of its initial value after approximately 250 h of continuous operation seems to be caused by the increasing contamination of xenon, through minor leaks in the discharge chamber and/or the outgassing of chamber components. Refilling the chamber with fresh gas after such an extended operation resulted in full recovery of the discharge with respect to excimer emission. The results suggest the possibility of generating extended lifetime, intense, large area, planar excimer sources using CBL discharges in sealed discharge chambers including getters.

Flat lamps, having radiating areas as large as 15 × 15 cm² and comprising arrays of Al/Al₂O₃ microcavity plasma devices, have been fabricated and characterized in the rare gases. Sealed arrays of devices with microcavities having diamond-shaped cross-sections yield lamps with an overall thickness of ∼800 µm (of which ∼500 µm is the quartz output window) and luminance and luminous efficacy values greater than 1600 cd m⁻² and 10 lm W⁻¹, respectively, are observed in preliminary experiments in which microplasmas in a Ne/20%Xe mixture illuminate a commercial green phosphor in a transmission arrangement with the ∼10 µm thick phosphor film situated immediately above the array. Efficacy values well in excess of 20 lm W⁻¹ are expected when the array design and microcavity-phosphor geometry are optimized. Fully flexible arrays sealed in polymeric packaging have also been demonstrated, thereby providing access to new opportunities for lighting in which lightweight arrays are mounted onto curved surfaces.

This special Cluster of articles in Journal of Physics D: Applied Physics covers the subject of mercury-free discharges that are being investigated by different light source researchers, as an alternative to existing mercury-containing lamps.

The main driving force to move away from mercury-containing discharge light sources is connected to the environmentally unfriendly nature of mercury. After inhalation or direct contact, severe mercury exposure can lead to damage to human brain cells, the kidneys, the liver and the nervous system. For this reason, the use of mercury in products is becoming more and more restricted by different governmental bodies.

In the lighting industry, however, many products still make use of mercury, for different reasons. The main reason is that mercury-containing products are, in most cases, more efficient than mercury-free products. For a realistic comparison of the environmental impact, the mercury-contamination due to electricity production must be taken into account, which depends on the type of fuel being used. For an average European fuel-mix, the amount of mercury that is released into the environment is around 29 μg kWh^-1. This means that a typical 30 W TL lamp during a lifetime of 20,000 hours will release a total of about 20 mg mercury due to electricity production, which exceeds the total mercury dose in the lamp (more and more of which is being recycled) by a factor of 5–10 for a modern TL lamp.

This illustrates that, quite apart from other environmental arguments like increased CO₂ production, mercury-free alternatives that use more energy can in fact be detrimental for the total mercury pollution over the lifetime of the lamp. For this reason, the lighting industry has concentrated on lowering the mercury content in lamps as long as no efficient alternatives exist.

Nevertheless, new initiatives for HID lamps and fluorescent lamps with more or less equal efficiency are underway, and a number of them are described in this special issue. These initiatives may in time offer realistic alternatives for mercury-containing discharge lamps as the efficiency gap with existing products is getting smaller.

At the same time, new applications for radiation sources are becoming more important, and in some of them the presence of mercury has other disadvantages besides the environmental aspects. Since in most cases mercury is used in the form of a saturated vapour, the mercury pressure is dependent on the ambient temperature, which means that mercury-containing lamps often show a slow increase to the steady-state light output or a strongly reduced output in cold environments, which is undesirable in many applications. For this reason also, different options for light sources without mercury are being investigated, and a number of them can be found in this special issue.

This collection of papers gives a good overview of the different technologies that are currently being investigated as alternatives to existing lamp technologies, and will surely inspire others to reduce the use of mercury for lighting applications.