Table of contents

The 19th International Colloquium on Magnetic Films and Surfaces (ICMFS 2006) was held on 14–18 August 2006 at the Sendai International Center in Sendai, Japan. The purpose of the Colloquium was to bring together scientists working on magnetic thin films and surfaces and to provide an opportunity for presentation and discussion of recent experimental and theoretical advances in the field. 285 scientists from 17 countries (Japan: 167, overseas: 118) participated in the Colloquium, as well as 6 family members. There were 56 oral and 178 poster presentations. The oral presentations consisted of 3 plenary talks, 23 invited talks and 30 contributed talks. The number of presentations by scientific category are as follows:

Spin dependent transport: 43 Magnetic storage/memory: 9 Magnetization reversal and fast dynamics: 15 Spin injection and spin transfer torque: 26 Magnetic thin films and multilayers: 71 High spin polarization materials: 17 Hard and soft magnetic materials: 3 Magneto-optics: 5 Characterization techniques for thin films and surfaces: 7 Exchange coupling: 13 Micro- and nanopatterned magnetic structures: 18 Micromagnetic modelling: 2

One of the characteristics of the present Colloquium is an increase in the number of presentations in the field of spin-electronics, as seen above. This Cluster Issue of Journal of Physics D: Applied Physics includes several important papers in this rapidly developing field. We believe that, in the future, the field of magnetic materials will maintain its popularity and, on top of that, other fields such as spintronics materials, materials related to life sciences and medicine and also materials related to the environment will be investigated further.

The ICMFS Conference started in London in 1964, and is now one of the world-wide conferences on magnetism. The Colloquium has been held in Japan four times now: the previous ones being the 5th ICMFS in the Mount Fuji area, the 10th at Yokohama and the 17th at Kyoto, which was organized by Professors Shinjo and Maekawa. The city of Sendai, where the 19th ICMFS was held, is the historical place for magnetism research in Japan. Kotaro Honda, who was a professor of Tohoku University in Sendai, laid the foundation for this research in Japan, and he was followed by Siji Kaya, Haraku Masumoto and Minoru Takahashi, also professors of Tohoku University. They continued the spirit founded by Honda and contributed greatly to the progress of research in magnetism. Therefore, it was a great pleasure for the organizers to have the ICMFS Conference come to Sendai.

The 19th ICMFS Colloquium was co-hosted by the Japanese Society for the Promotion of Science and the Foundation Advanced Technology Institute and supported by the Japanese Society of Applied Physics, the Magnetics Society of Japan, the Physical Society of Japan and the Japan Institute of Metals. This colloquium was also jointly held with the Conference on the Physics and Application of Spin-related Phenomena in Semiconductors (PASPS). All the members of the Organizing Committee would like to thank the members of the International Advisory Committee for scientific and administrative advice, and the Sendai Tourism and Convention Bureau, the Iwatani Naoji Foundation, the Asahi Glass Foundation and the Intelligent Cosmos Academic Foundation for their financial support. Without doubt, the Colloquium was a great success overall. The smooth and excellent running of the Colloquium would not have been possible without the assistance of the Program Committee and local members of the Colloquium.

We investigated the film performance and nanostructure of current-perpendicular-to-plane giant magnetoresistance (CPP-GMR) spin-valve film with a current-confined-path nano-oxide layer (CCP-NOL). By applying ion-assisted oxidation (IAO) for the CCP-NOL formation, we enhanced the MR ratio to 5.4% at a small RA value of 500 mΩ µm² for conventional Co₉₀Fe₁₀ layers. Furthermore, the use of bcc-Fe₅₀Co₅₀ also increased the MR ratio to 8.2% at a small RA value of 580 mΩ µm². A modified Valet–Fert model for the CCP-NOL showed that the MR enhancement by the IAO is due to the improvement in resistivity of the CCP, and that by Fe₅₀Co₅₀ is due to a larger spin-dependent interface scattering effect. Analysis by cross-sectional TEM and three-dimensional atom probe confirmed the formation of the CCP-NOL structure. A reliability test for test element devices showed almost no change even under acceleration stress. The CPP-GMR spin-valve film with the CCP-NOL is extendable to future high-density recording heads due to its potential for a higher MR ratio at a small value of RA.

Magnetic tunnel junctions (MTJs) with stacking structures of Co₂MnSi/Al–O/CoFe and Co₂MnSi/Al–O/Co₂MnSi were fabricated using a magnetron sputtering system. Co₂MnSi/Al–O/CoFe-MTJ and Co₂MnSi/Al–O/Co₂MnSi-MTJ exhibited extremely large tunnelling magnetoresistance ratios of 159% and 570%, respectively, at low temperature, indicating the half-metallicity of Co₂MnSi. We investigated the bias voltage dependence of tunnelling conductance (dI/dV–V) for each MTJ in order to clarify the band structure around the Fermi level in Co₂MnSi. The observed dI/dV–V curves for both MTJs reveal a clear half-metallic energy gap at 350–400 meV for Co₂MnSi, with an energy separation of just 10 meV between the Fermi energy and the bottom edge of the conduction band.

Tunnel magnetoresistance (TMR) and spin filter effect in magnetic tunnel junctions are studied by using numerical simulations based on the linear response theory. In Fe/MgO/Fe junction, TMR is calculated for both clean and disordered systems. Large MR ratio obtained in clean junctions is attributed to the momentum and symmetry filter effect of MgO barrier. It is shown that lattice distortions within MgO and at the interfaces decrease the MR ratio because the disorder weakens the filter effects and causes hybridization between Bloch wave functions having different momenta and symmetries. The spin filter effect in junctions containing a ferromagnetic insulator LaNi_0.5Mn_0.5O₃ is also investigated. It is shown that a large spin filter effect is obtained in LaNiO₃/LaNi_0.5Mn_0.5O₃/LaNiO₃ junctions although the tunnel barrier height is spin independent.

We present measurements of tunnelling spin polarization (TSP) of CoFeB alloys in Al/AlO_x/CoFeB tunnel junctions using superconducting tunnelling spectroscopy. We observe that though Co₇₂Fe₂₀B₈ (at.%) layers crystallize in the fcc structure, in junctions annealed above 300 °C the measured TSP degrades, indicating that in these junctions the TSP of amorphous CoFeB is higher than its crystalline counterpart.

The dc conductivity at T = 0 K of half-metallic diluted antiferromagnetic semiconductors and their super lattice is calculated by the use of the Kubo–Greenwood formula applied to the Korringa–Kohn–Rostoker (KKR) Green's function method and the coherent potential approximation in the framework of the local density approximation (LDA) of the density functional method (KKR-CPA-LDA). The calculated dc conductivity of superstructures composed of diluted antiferromagnetic semiconductors shows significant dependence on the magnetic structure of the superstructures. The mechanisms responsible for such behaviour are discussed on the basis of the first-principles calculation.

MgO-barrier-based magnetic double tunnel junctions including Au or Cr nanoparticles were prepared by molecular beam epitaxy, and their magnetotransport properties were investigated. A double junction sample including Au nanoparticles showed the Coulomb blockade effect and clear magnetoresistive hysteresis loops. The observed bias voltage dependence of the resistance and magnetoresistance (MR) suggested that the MR effects of 1–2% at high bias voltages were caused by spin accumulation in the Au nanoparticles. In the case of Cr nanoparticles, a double junction with relatively low sample resistance was obtained, showing a clear Coulomb threshold.

An experimental study of domain wall motion in Ni₈₀Fe₂₀ ring structures induced by current pulses as well as conventional magnetic fields is presented. Using constrictions we demonstrate that current-induced domain wall motion can be used to displace walls into parts of the structure where no pulsed currents are flowing. Measurements at variable temperatures between 2 and 300 K show that the fields necessary for wall motion decrease with increasing temperature, which can be explained by thermal activation. For the current-induced case we find, depending on the geometry and temperature range, that the current densities necessary for displacement can increase or decrease with rising temperature. This indicates that, in addition to thermal excitations, an intrinsic temperature dependence of the efficiency of the spin torque effect is present and leads to an increase in the critical current density with increasing temperature.

Current induced domain wall creation in CPP nanopillars made of Co/Ni films with strong perpendicular magnetic anisotropy is described. We find that stable domain wall states (DWS) can be nucleated with modest current densities of ∼10⁷ A cm⁻² and further controlled by current to restore the two uniform states. The reproducibility of both the creation and annihilation of such DWS is studied. Experiments have been performed several times for a given nanopillar, but also for different devices. The influence of temperature, size and shape of the devices as well as the distribution of magnetic properties on magnetization reversal is discussed.

The motion of a planar domain wall in a nanoscale ferromagnetic wire under electric current is studied, based on a microscopic description. In the adiabatic case, the domain wall is driven by spin torque (spin transfer), and there is an intrinsic pinning arising from hard-axis anisotropy energy. Effects of extrinsic pinning and spin relaxation (β-term) on the threshold current are discussed.

The effects of the current-induced Oersted field and of a hard axis applied field on spin transfer torque (STT) switching have been investigated by performing LLG micromagnetic simulations including a STT term and Gaussian thermal fluctuations. In parallel to anti-parallel switching at large currents, the C-like micromagnetic configuration induced by the Oersted field plays an important role in STT switching. In anti-parallel to parallel switching at large currents, the Oersted field induces a complicated micromagnetic configuration with several vortices. In both cases, the magnetization deviates considerably from macrospin behaviour. However, a macrospin-like behaviour is restored when a hard axis field is present.

Ultrathin Fe films on Cu(1 0 0) are self-organized into stripes of opposite perpendicular magnetization. The process of self-organization involves stripe-nucleation and stripe-creep. We present images of nucleation and creep at the micrometre scale. These observations provide evidence of both quenched and self-induced disorder in a system with competing interactions.

In this work we show that the measurement of single magnetic reversal events is of critical importance in order to correctly characterize the switching of magnetic microstructures. Magnetoresistance measurements are performed on two pseudo-spin-valve ring structures with high enough signal to noise to allow the probing of single reversal events. Using this technique we acquire 'switching spectra' which demonstrate that the rings exhibit a range of variable reversal paths, including a bistable reversal mechanism of the hard layer, where the two switching routes have substantially different switching fields. The signature of the variable reversal paths would have been obscured in field cycle averaged data and in the bistable case would cause a fundamental misinterpretation of the reversal behaviour.

Injection and detection of non-local spin accumulation have been demonstrated in a series of nanoscale metallic lateral spin valves, fabricated by electron beam lithography. A distinct feature of non-local spin valves is that a pure spin current without charge flow can be generated. The measured spin signal is only sensitive to the relative orientations of the detector spin and the spin accumulation, and exhibits little charge background voltage. The injection polarizations of the interfaces and the spin diffusion lengths of the non-magnetic components have been determined by measuring the spin signals from a set of similar spin valves with varying injector and detector separations. In Py/Au/Py lateral spin valves, where Py is permalloy, the injection polarization across the Py/Au interface is 3%, and the spin diffusion length of Au is 63 ± 15 nm at 10 K. In Co/Cu/Co lateral spin valves, the injection polarization across the Co/Cu interface is increased to over 7%, due to better interfacial quality compared with Py/Au. The spin diffusion length of Cu is 200 ± 20 nm at 10 K, and > 110 nm at 300 K.

The influence of the pure spin current injection on a domain wall nucleation process in a magnetic wire is investigated by using the nonlocal spin injection technique. The pure spin current injection is found to assist the domain wall nucleation. The spin current responsible for nucleating the domain wall in the nonlocal configuration is comparable to the threshold spin current for the domain wall displacement in the direct current injection. The bias-field dependence of the switching current shows a complicated behaviour. This may be related to the detailed domain structure in the vicinity of the domain wall.

We have studied by hard x-ray scattering, soft x-ray resonant magnetic scattering and polarized neutron reflectivity the structural properties and the magnetization density profiles in multilayers of the Heusler phase Co₂MnGe with three different spacer layers, V, Au, and AlO_x and have found that the spacer layer has a significant effect on the magnetization profile inside the Co₂MnGe layer. In all cases the interlayers at the top and at the bottom of the Heusler layer exhibit a reduced magnetic moment, the thickness of which depends on the spacer layer and increases from 0.5 nm for V to 0.6 nm for Au and 1 nm for AlO_x.

In this work we report on an extensive and detailed study of exchange bias systems consisting of two ferromagnetic layers deposited with an antiferromagnetic layer of IrMn between them. Systems with two different ferromagnetic layers have been studied in which sample 1 had CoFe layers of different thicknesses and sample 2 had not only ferromagnetic layers of different thicknesses but also the composition of the upper ferromagnetic layers was changed from CoFe to NiFe. In both the samples the antiferromagnetic layer thickness was maintained constant at 5 nm, NiCr seed and capping layers (5 nm) were used on all samples. Such a system is of considerable interest as the properties of each ferromagnetic layer are affected by the same antiferromagnet. However differences in behaviour will occur due to the nature of the interfaces between the different layers as well as other parameters such as the ferromagnetic layer thickness. We have conducted a study of thermal activation effects in these systems where both or a single ferromagnetic layer can be reversed whilst the system is heated. We find that we can differentiate between bulk and interface effects indicating that the role of spin disorder at the interface is crucial in determining the final value of the exchange bias.

It is demonstrated experimentally that expanded face centred tetragonal (e-fct) Mn, which is characterized by an axial ratio c/a > 1, can be stabilized by epitaxial growth on ferromagnetic (FM) fct Co(0 0 1) films. At room temperature this new metastable Mn phase is antiferromagnetically (AFM) ordered for thicknesses above the 2.5 monolayer, as inferred from AFM/FM magnetic exchange interaction studies. This critical thickness does not depend on the magnetization direction of the FM layer. Even though for out-of-plane magnetized FM layers the coercivities are distinctly enhanced by the exchange interaction, no exchange bias can be detected. On the other hand, both coercivity enhancement and exchange biasing is observed for the in-plane magnetized FM/e-fct Mn bilayer. These results provide experimental evidence that the AFM spin structure of e-fct Mn at the AFM/FM interface is confined to the film plane.

Spin-resolved scanning tunnelling microscopy measurements were performed for Co double-layer nanostructures on Pt(1 1 1). High-resolution spectroscopy shows that the electronic structure is highly inhomogeneous due to dislocations induced by the lattice mismatch between Pt and Co. Spin-polarized measurements allow a magnetic state dependent determination of out-of-plane magnetized islands. In the presence of external fields we have observed two different types of Co double-layer islands which exhibit a surprisingly large coercivity depending on their preparation.

This paper presents an overview of slow light in semiconductor heterostructures. The focus of this paper is to provide a unified framework to summarize and compare various physical mechanisms of slow light proposed and demonstrated in the past few years. We expand and generalize the discussions on fundamental limitation of slow light and the delay–bandwidth product trade-off to include gain systems and other mechanisms such as injection locking. We derive the maximum fractional delay and compare the differences between material dispersion and waveguide dispersion based devices. The delay–bandwidth product is proportional to the square root of the device length for a material dispersion based device but has a linear relationship for a waveguide dispersion based device. Possible scenarios to overcome the delay–bandwidth product limitation are discussed. The prospects of slow light in various applications are also investigated.

We present a method and model for the direct and continuous separation of red and white blood cells in plasma. The method is implemented at the microscale using a microfluidic system that consists of an array of integrated soft-magnetic elements embedded adjacent to a microfluidic channel. The microsystem is passive and is activated via application of a bias field that magnetizes the elements. Once magnetized, the elements produce a nonuniform magnetic field distribution in the microchannel, which gives rise to a force on blood cells as they pass through the microsystem. In whole blood, white blood cells behave as diamagnetic microparticles while red blood cells exhibit diamagnetic or paramagnetic behaviour depending on the oxygenation of their haemoglobin. We develop a mathematical model for predicting the motion of blood cells in the microsystem that takes into account the dominant magnetic, fluidic and buoyant forces on the cells. We use the model to study red/white blood cell transport, and our analysis indicates that the microsystem is capable of rapid and efficient red/white blood cell separation.

Interactions in oligonucleotide functionalized ferrofluids were investigated by measurements of frequency dependent complex magnetization in a superconducting quantum interference device. The ferrofluids consisted of aqueous suspensions of single-stranded oligonucleotide functionalized 130 nm sized magnetic beads, suitable for further use in magnetic biosensor applications based on changes in the Brownian relaxation behaviour of oligonucleotide-coated beads. The interbead interactions were found to depend strongly on the surface coverage by oligonucleotide molecules. At low surface coverage, aggregates of beads within the ferrofluid were formed, most likely due to interbead SS-crosslinking reactions. When the surface coverage increased, the entropic interbead repulsion arising from the thermal motion of oligonucleotide molecules began to stabilize the ferrofluid by preventing aggregation, and the crosslinking probability decreased. At a surface coverage, of ∼40 oligonucleotides per bead, determined by a radioactive labelling analysis, an optimal configuration was obtained for which the ferrofluid behaved as consisting of almost non-interacting beads. At higher oligonucleotide functionalization degrees, the high coverage induced less thermal motion flexibility of the oligonucleotide chains which decreased the entropic repulsion effect. This in turn led to a higher crosslinking probability, as well as an increase in the hydrodynamic size of the beads. Thus, the ferrofluid sample of nearly non-interacting nature, in which an optimal degree of stabilization and the highest Brownian relaxation frequency are achieved in the low surface coverage region, may be considered as the most appropriate one for further use in magnetic biosensor applications, from both a functional and an economic point of view.

The effect of added Ag on the electro-magnetic properties and enhanced room temperature magnetoresistance in (La_0.7Ca_0.2Sr_0.1MnO₃)_1−x/Ag_x composites has been studied systematically. According to the results of x-ray diffraction, scanning electron microscopy and magnetic measurement, we can suggest that Ag segregates at the grain surface of La_0.7Ca_0.2Sr_0.1MnO₃. The results of electronic measurements show that the intrinsic insulator/metal transition temperature (T_P) does not change, which is in accordance with the results for the Curie temperature (T_C). It is very interesting to note that the magnetoresistance at room temperature is enhanced, which is encouraging for potential applications. These phenomena can be explained by considering that Ag, which segregated at the grain boundaries or surfaces, does not change the intrinsic structure of La_0.7Ca_0.2Sr_0.1MnO₃ and lead to a modification of grain boundaries.

Thin-film transistors (TFTs) with transparent amorphous zinc indium tin oxide (ZITO) channel layer are demonstrated. Optical transmission of the channel layer is approximately 85% in the visible portion of the electromagnetic spectrum. The channel layer is formed via rf magnetron sputter deposition and then furnace annealed in air. Peak incremental mobilities of 5–19 cm² V⁻¹ s⁻¹ and turn-on voltages of −4 to −17 V are obtained for devices annealed post-deposition at 100–300 °C, respectively. Current–voltage measurements indicate n-channel, depletion-mode transistor operation with excellent drain current saturation and a drain current on-to-off ratio greater than 10⁶. ZITO is one example of an emerging class of high performance TFT channel materials involving transparent amorphous multicomponent oxides composed of heavy-metal cations with (n − 1)d¹⁰ns⁰ (n ⩾ 4) electronic configuration.

50-period SiO₂/SiO_x layers have been prepared on Si wafers by ion beam sputtering deposition and subsequently annealed to form Si nanocrystals (NCs) in the SiO_x layer. The integrated photoluminescence (PL) intensity of the samples with x ⩽ 1.6 increases on decreasing the temperature down to a certain temperature named as T_C, but below T_C, it decreases. T_C is maintained at 95 K, almost irrespective of x. The rate of decrease of the PL below T_C gets slower as x increases up to 1.6, and at x = 1.8, the PL intensity shows a monotonic increase with decreasing temperature. This phenomenon is strongly related to the change in the temperature dependence of the PL decay rate at around T_C as x varies. These results can be explained by the fact that the PL behaviours are increasingly influenced by the NCs' environment such as the defect states of the Si NCs/SiO₂ interfaces as x increases.

Nanocrystalline Ba_0.97Ca_0.03SO₄ : Eu having a grain size of 58 nm has been prepared and a comparative study has been done with its corresponding microcrystalline form. The thermoluminescence (TL) glow curve of the nanophosphor has a prominent peak at 161 °C and a very small hump at 225 °C. The TL sensitivity of the nanophosphor is about 0.1 times that of the microphosphor. Kinetics parameters of the TL peaks of the nanophosphor are obtained and are compared with those of the microphosphor. The TL response of the nanophosphor is linear in the dose range 1 Gy–20 KGy, which is much wider than that of its corresponding microcrystalline form. The glow curve shape and structure also do not change in the linear dose range and, since the TL response is linear at higher doses, the nanophosphor is quite well suited for high dose measurements. Photoluminescence emission spectra of the nano- and the microphosphors are also studied and their results correlated with other experimental findings.

Current–voltage (I–V) and capacitance–voltage (C–V) characteristics of He-ion irradiated Pd/n-Si₀₉Ge_0.10 Schottky contacts have been measured in the temperature range from 100 to 300 K. Schottky barrier properties such as the Schottky barrier height (Φ_bn) and ideality factor (n) have been studied as a function of temperature. The degree to which their characteristics deviated from the ideal case increased as the temperature decreased. A decrease in Φ_bn and an increase in n with decreasing temperature are observed. Additionally, linear dependence between the so-called temperature factor T₀ and temperature as well as between Φ_bn and n are shown. This type of strong temperature dependence indicates the presence of a large degree of lateral inhomogeneities of the barrier height, resulting from the He-ion irradiation induced defects and traps which produce a variation in the number of free carriers. The presence of electrically active defects introduced by He-ion irradiation at and below the Si_0.90Ge_0.10 surface support this interpretation.

Analysis of available and developed data on phase matching in AgGaGe_xS_{2(1 + x)} (x = 0, 1) crystals is carried out. Nanosecond AgGaS₂ type I optical parametric oscillator with a continuously tunable range 2.65–5.29 µm is demonstrated pumped by a Q-switched Nd : YAG laser. An output pulse energy of up to 0.56 mJ at 4 µm is recorded. Phase matching of second harmonic generation in both crystals is represented. Best sets of Sellmeier equations for two crystals are determined.

We demonstrated high-efficiency red organic light-emitting diodes (OLEDs) employing a europium complex, Eu (III) tris(thenoyltrifluoroacetone) 3,4,7,8-tetramethyl- 1,10-phenanthroline (Eu(TTA)₃(Tmphen)), as an emitter and a blue electrophosphorescent complex, Iridium (III) bis[4,6-di-fluorophenyl-pyridinato-N, C²] picolinate (FIrpic), as an assistant dopant codoped into 4,4-N,N-dicarbazole-biphenyl (CBP) host as an emissive layer. A pure red electroluminescence (EL) only from Eu³⁺ ions at 612 nm with a full width at half maximum of 3 nm was observed and the EL efficiency was significantly enhanced. The maximum EL efficiency reached 7.9 cd A⁻¹ at 0.01 mA cm⁻² current density, which is enhanced by 2.8 times compared with electrophosphorescence-undoped devices. The large improvements are attributed to energy transfer assistance effects of FIrpic, indicating a promising method for obtaining efficient red OLEDs based on rare-earth complexes.

This paper reports on the determination of the temperature dependence of the critical reduced electric fields (E/N)_cr in SF₆ and mixtures of SF₆ with C₂H₆ at gas temperatures ranging from 293 to 1500 K. The (E/N)_cr values are derived from the quasi-stationary burning voltages of a self-sustained volume discharge in these gases after their being irradiated by a pulse CO₂ laser. The mechanism of a laser-induced heating of the mixtures treated, and evaluation of the gas temperatures established due to absorption of the laser energy, are described in detail. The temperature dependence of (E/N)_cr established for SF₆ is compared with those previously obtained in other studies. A topological approach based on Thom's catastrophe theory is involved to predict and qualitatively describe all possible structurally stable features, which can only be expected in the behaviour of (E/N)_cr on changing the time of the thermal action on SF₆ from vanishingly small values to infinitely large ones. On this basis, some inferences about the peculiarities of atomic fluorine generation on heating SF₆ are drawn.

Stability of different modes of steady-state current transfer to cathodes of high-pressure arc discharges is investigated in the framework of the model of nonlinear surface heating. An eigenvalue problem governing perturbations has been formulated and its properties analysed. It is shown that the spectrum of this problem is real, therefore the change of stability can occur only at bifurcation points, in which different modes of steady-state current transfer join each other. Analytical expressions describing behaviour of the increment in the vicinity of bifurcation points have been derived. A general pattern of stability of the diffuse mode of current transfer to axially symmetric cathodes and of 3D spot modes branching off from the diffuse mode is established. Theoretical predictions are found to be in agreement with trends observed in the experiment and it is shown that the recently published comparison of the theory and the experiment on the spot mode needs to be revisited.

The influence of the content and nature of impurities on the rotational temperature of three probe molecules (OH, NH and ) in a helium plasma created in a microwave resonant cavity is investigated by the synthetic spectra method and by mass spectrometry. OH is found to be the only reliable probe to obtain a satisfactory estimation of the gas temperature. In our experimental conditions, depending on the impurity, the gas temperature varies strongly and at 500 ppm, for instance, a difference of 700 K is possible (∼30% of the mean temperature).

The characteristics of homogeneous discharge are studied and compared with those of dielectric barrier discharge (DBD) by measuring their electrical discharge parameters and observing their light-emission phenomena. Several factors, including the voltage amplitude, gap distance, electrode configuration and barrier thickness, are experimentally studied to see their effects on the transition of the homogeneous discharge from the stable mode to the unstable mode. The results show that the homogeneous discharge exists only under certain conditions. The important role of the mesh for obtaining the homogeneous discharge is identified through experiments. It is found that the voltage range for maintaining a stable discharge is wider when the barrier thickness is smaller, the gap distance is shorter and the mesh number is greater. It is thought that the corona discharge produced by the wire mesh together with the property of the PET barrier can cause accumulation of charges on the barrier surface, which may provide initial electrons leading to the burst of the homogeneous discharge in the whole air gap when a certain threshold value of the electric field (breakdown value) is exceeded. In view of this assumption, the experimental results for the effects of these factors on the stability of the homogeneous discharge can be qualitatively interpreted.

The paper presents results of a high-pressure, non-self-sustained crossed discharge–M = 3 supersonic laser cavity operation. A stable and diffuse pulser–sustainer discharge in O₂–He flows is generated at pressures of up to P₀ = 120 Torr and discharge powers of up to 2.1 kW. The reduced electric field in the dc sustainer discharge ranges from 0.6 × 10⁻¹⁶ to 1.2 × 10⁻¹⁶ V cm². Singlet delta oxygen (SDO) yield in the discharge, up to 5.0–5.7% at the flow temperatures of 400-420 K, was inferred from the integrated intensity of the (0, 0) band of the O₂(a ¹Δ → X ³Σ) infrared emission spectra calibrated using a blackbody source. The yield increases with the discharge power and remains nearly independent of the O₂ fraction in the mixture (in the 10–20% range). Static pressure and temperature measurements in the supersonic cavity show that a steady-state M = 3 flow in the cavity can be sustained for up to 20 s, at the flow temperature of T = 120 ± 15 K. The results suggest that the measured SDO yield exceeds the threshold yield at the cavity temperature by up to a factor of 2.5. PLIF iodine vapour visualization in the supersonic cavity, which showed the presence of large-scale structures, suggests the need to improve iodine vapour mixing with the main oxygen–helium flow.

Results are presented that have been obtained while operating the graphite hollow cathode duoplasmatron ion source in dual mode under constant discharge current. This dual mode operation enabled us to obtain the mass and emission spectra simultaneously. In mass spectra C₃ is the main feature but C₄ and C₅ are also prominent, whereas in emission spectra C₂ is also there and its presence shows that it is in an excited state rather than in an ionic state. These facts provide evidence that C₃ is produced due to the regeneration of a soot forming sequence and leave it in ionic state. C₃ is a stable molecule and the only dominant species among the carbon clusters that survives in a regenerative sooting environment at high-pressure discharges.

The effects of oxygen plasma treatment on epitaxial ZnO thin films grown by molecular beam epitaxy were studied. The Au–ZnO–In junction exhibiting ohmic behaviour before the treatment gradually changes to a Schottky junction with the increase in oxygen plasma treatment time. The crystallinity and the surface microstructure did not change to any great extent after the treatment. However, the x-ray photoelectron spectroscopy studies show the removal of conductive OH layer from the surface of ZnO films and the current–voltage characteristics of Au–ZnO–In junction exhibit the rectifying behaviour after oxygen plasma treatment. The fabricated Au–ZnO–Au ultraviolet (UV) detector was successfully tested and was observed to be sensitive to the two UV sources used. The photoresponsivities of the UV detector for the irradiation of two different power densities 350 (λ = 356 nm) and 420 µW cm⁻² (λ = 254 nm) are 13.5 A W⁻¹, 15.3 A W⁻¹ at 5 V and 128.9 A W⁻¹, 160 A W⁻¹ at 10 V, respectively.

Charge-trap flash- (CTF) memory structures have been fabricated by employing IrO₂ nanodots (NDs) grown by atomic-layer deposition. A band of isolated IrO₂NDs of about 3 nm lying almost parallel to Si/SiO₂ interface is confirmed by transmission electron microscopy and x-ray photoelectron spectroscopy. The memory device with IrO₂NDs shows much larger capacitance–voltage (C–V) hysteresis and memory window compared with the control sample without IrO₂NDs. After annealing at 800 °C for 20 min, the ND device shows almost no change in the width of C–V hysteresis and the ND distribution. These results indicate that the IrO₂NDs embedded in SiO₂ can be utilized as thermally stable, discrete charge traps, promising for metal oxide-ND-based CTF memory devices.

Ba₂FeNbO₆ (BFNO) is an antiferromagnetic insulator with a Néel temperature T_N of 25 K. We have systematically studied the effect of substrate on the electrical transport of this compound under different growth conditions. We find that while the bulk and thin films grown on LaAlO₃ and NdGaO₃ substrates are insulating, the film grown on SrTiO₃ substrate shows metallic behaviour. The anomalous behaviour of this compound on SrTiO₃ is attributed to the defect conduction of SrTiO₃ induced by the film growth conditions.

We demonstrate the calculation of the refractive index profile (reverse engineering) of an inhomogeneous optical coating based on in situ measured optical transmission spectra. A full triangular algorithm has been applied for refractive index profile calculation. The obtained profile is consistent with deposition rate recordings obtained during film deposition from quartz crystal monitors. The proposed methodology is also applicable for reverse engineering tasks in traditional high-low-stacks. As an additional feature, the method may be utilized for re-calibrating quartz monitors used for deposition control either after or during the deposition experiment.

Nd-doped LiNbO₃ (Nd : LiNbO₃) thin film with c-axis preferred orientation has first been fabricated on a SiO₂/Si(1 1 1) substrate by pulsed laser deposition. A smooth surface and a sharp interface have been affirmed by scanning probe microscopy and transmission electron microscopy, respectively. The fluorescence at around 902.5, 1087.5 and 1379.3 nm was excited with a diode laser at 808 nm, and the optical and spectral characteristics and waveguide behaviours were investigated as well. The results show that the highly c-axis oriented Nd : LiNbO₃ thin film on SiO₂/Si(1 1 1) substrate may have actual application in integrated optics.

We study the formation mechanism of rims created around femtosecond laser ablated craters on glass. Experimental studies of the surface morphology reveal that a thin rim is formed around the smooth craters and is raised above the undamaged surface by about 50–100 nm. To investigate the mechanism of rim formation following a single ultrafast laser pulse, we perform a one-dimensional theoretical analysis of the thermal and fluid processes involved in the ablation process. The results indicate the existence of a very thin melted zone below the surface and suggest that the rim is formed by the high pressure plasma producing a pressure-driven fluid motion of the molten material outwards from the centre of the crater. The numerical solutions of pressure-driven fluid motion of the thin melt demonstrate that the melt can flow to the crater edge and form a rim within the first nanoseconds of the ablation process. The possibility that a tall rim can be formed during the initial stages of the plasma is suggestive that the rim may tilt outwards towards the low pressure region creating a resolidified melt splash as observed in the experiments. The possibility of controlling or suppressing the rim formation is discussed also.

A c-axial oriented (Bi_1.5Zn_0.5)(Zn_0.5Nb_1.5)O₇ thin film has been grown on a (0 0 1) Nb doped SrTiO₃ substrate by pulsed laser deposition. The permittivity, dielectric loss and tunability of the c-axial oriented film are 187, 0.002 and 6% (at 750 kV cm⁻¹ biasing), respectively, indicating a figure of merit of 30. Moreover, an asymmetry behaviour is observed in the dc electric field dependence of permittivity, which could be attributed to the asymmetry of top and bottom electrodes.

The process by which diamonds wear when they are facetted using traditional diamond polishing has been the subject of much debate. The absence of a plausible wear mechanism to describe the process has been due in part to the lack of reliable data; a multitude of variables are involved with the process. In recent years, the enhancement of sophisticated analytical techniques has led to the discovery that diamond undergoes a phase transformation during polishing, although the process by which this occurs is unclear. One of the more interesting observations seen during diamond polishing is the occasional emission of light from a polishing diamond. Since this emission is a direct consequence of physical processes at the polishing interface, it provides us with detailed and direct information about electronic processes occurring as diamond bonds that are broken and/or transformed. These light emissions have loosely been described as triboluminescence, although no detailed study of the phenomenon occurring in diamond has ever been published. This paper describes an investigation into the light emitted during fracture (such as that obtained when diamond is scratched) and during polishing. In each case, the spectral emissions are different. During fracture, the overall spectral envelope is dominated by light emission due to frictional sliding and to a lesser extent by photoluminescence, demonstrated by the appearance of the 'N3' centre. During polishing, it is suggested that the mechanism by which light emission occurs is via electroluminescence.

The mechanical properties of 93%Pb(Zn_1/3Nb_2/3)O₃–7%PbTiO₃ (PZNT93/7) single crystal were evaluated by microindentation and nanoindentation techniques. The (1 1 1) facet of the crystal exhibited excellent combination mechanical properties (Vickers microhardness Hv, fracture toughness Kc and elastic modulus Er) than those of (0 0 1) facet. The Hv on (1 1 1) and (0 0 1) facets were 3.71 GPa and 3.00 GPa, respectively. In-plane fracture toughness anisotropy was observed on both the (1 1 1) and (0 0 1) facets, the Kc value was low in the direction parallel to the domain wall but large in the direction perpendicular to it. The Er of (0 0 1) and (1 1 1) facets was independent of the load, Er_(1 1 1) was measured to be about 130 GPa and Er_(0 0 1) was about 110 GPa. The above results implied that (0 0 1) PZNT crystals would face a more severe challenge in device application due to their weaker combination mechanical properties but outstanding piezoelectric properties than those of (1 1 1) PZNT crystals.

The accurate measurement of the electrical conductivity of low conducting liquids is an important issue in many industrial applications. The lack of repeatability is a common problem to the available procedures and commercial techniques. In this paper, we present a device to measure the electrical conductivity of low conducting liquids. The variable inter-electrode gap allows us to assure the existence of an ohmic regime, since only under ohmic regime conditions is the value of the conductivity meaningful.

An explicit solution for the longitudinal and transverse polarizability of the asymmetric dielectric intersecting double sphere is obtained as a rapidly converging series of integral operators, which can be implemented efficiently, for example, in Java Applet. This article generalizes the results of the paper (Pitkonen M 2006 J. Math. Phys.47 102901) to the asymmetric case and also allows the permittivities of the spheres to have different values.

In this paper, the authors discuss one- and three-dimensional space charge distributions in glass fibre/epoxy resin composites. By the conventional pulsed electroacoustic (PEA) method, only a one-dimensional distribution of the average charge over a whole area parallel to the two electrodes can be observed. Therefore, the authors have developed a new PEA system capable of measuring a three-dimensional space charge distribution. Using this system, they measured the charge distribution in glass fibre/epoxy resin composites made of lattice-woven glass fibre and epoxy resin. It has become clear that spatial variation in signal intensity observed depends on the internal structure of the composite. There appear repetitious positions where a high charge density is observed on the same lateral cross section along the vertical direction in the composite. Such positions are consistent with the intersections of the glass fibres. Accumulation of mobile charge carriers or appearance of polarization charge due to mismatch of the ratio of the conductivity and permittivity between the glass fibre and the epoxy resin is thought to be responsible for the PEA signals.

We investigate acoustic phonon transport and thermal conductance at low temperatures in a quantum wire with rectangle scatters periodically placed in the quantum channel. It is found that the transmission spectra of zero mode exhibits a series of resonant peaks–valleys structures, and the transport valley gradually develops into stop-frequency gap by increasing the number of the period. The number of resonant peaks or valleys between two nearest gaps is just twice as large as the number of the period. The result also indicates that the thermal conductance is sensitive to the number of the period and structural parameters, and the change is more pronounced in the structure with smaller number of period.

A pulsed laser to generate ultrasound has been used with an electromagnetic acoustic transducer (EMAT) to successfully detect ultrasonic waves for the fast online inspection of steel billets. Simulations of the ultrasonic signals have been performed by the finite element method to explain the laser-generated ultrasonic field and to compare it favourably with the signals detected by a water-cooled EMAT. With an understanding of the nature of the generated ultrasonic field, the separation between the laser and EMAT can be optimized for internal defect inspection using bulk waves and surface defect inspection, using Rayleigh waves, as the pulsed laser source can generate both these wave modes simultaneously.