Table of contents

This communication amplifies a recent publication and shows that the fluid model is not adequate to give the full detail of the plasma-sheath structure for an electronegative plasma in the parameter range where stratification develops with two distinct plasmas separated by a potential barrier. A generalization of the Tonks-Langmuir model does not suffer the same deficiencies.

We show that the liquid crystal triphenylene derivative 2,3,6,7,10,11-hexapentyl-oxytriphenylene acts as a high-resolution electron beam resist. Using pentanol as a developer, positive behaviour was observed for electron doses greater than ~300µC cm^-2 at 20 keV. At higher doses (>2.5 mC cm^-2), the resist rapidly assumes negative tone behaviour. With the developer monochlorobenzene, only negative behaviour was observed, with a sensitivity of ~2.5 mC cm^-2 at 20 keV. The resist allows relatively facile definition of 14 nm patterns (negative tone) with a 30 keV electron beam and without the need for any complex pre-irradiation preparation or post-irradiation processing of the resist.

Since the discovery of the giant magnetoresistance effect and more recently its application in magnetic recording technology, the interest in spin-valve (SV) structures used in devices such as magnetoresistive sensors and random access memories, has increased greatly. As the size of these devices becomes smaller and smaller, the need to investigate the local microstructure and the micromagnetic behaviour of SV materials becomes obvious.

High-resolution electron microscopy (HREM) analyses have been carried out to investigate the microstructure of these metallic layered films. The choice of the materials used is crucial for the resulting magnetic properties. After a summary of the most common configurations found in the literature and some comments on their advantages and disadvantages, we will present some HREM studies which have clarified the differences in magnetic properties between top and bottom SVs. The growth conditions, the use of a seed layer and the thermal behaviour of SVs annealed at different temperatures will be discussed. In addition, some magnetostatic effects have been explained by microstructural considerations.

In addition to the HREM experiments, one of the techniques enabling micromagnetic studies at the micron scale to be carried out is Lorentz transmission electron microscopy (LTEM). This technique, which allows the magnetic domain structure of a magnetic material to be observed in situ, has been improved over the past few years. Very recently, the development of in situ magnetizing experiments in LTEM has enabled us to apply simultaneously an external field as well as a current through an SV element during the observation of the magnetization reversal. As a result, both electronic properties, via the giant magnetoresistance (GMR) curve, and local magnetic properties, via observation of the domain structure, can be analysed and correlated. Furthermore, the use of a mapping technique which allows quantitative analysis of the in-plane magnetization of the SV element, based on the analysis of Foucault images, has shown a clear correlation between the resistance values and the domain structure of the element. Such facilities have also resulted in a better understanding of the behaviour of various SV elements under real operating conditions. In particular, the effect on the reversal mechanism of the current density, the stray-field coupling at the edges of the element for different shapes of elements and the current direction through the element have been carefully studied.

The Fourier theory of heating is not applicable to the short-pulse type of laser heating due to the assumptions made in the theory. On the other hand, two-equation and kinetic theory models offer an improved solution to the problem. Consequently, the present study compares the predictions of one-equation (Fourier heating model), two-equation, and kinetic theory models for the laser heating pulses of 10^-9, 10^-10 and 10^-11 s lengths. The physical significance of the predictions are described and the discrepancies among the findings are discussed. It is found that all the models employed in the present study predict similar temperature profiles in the substrate for a nanosecond laser heating pulse. As the pulse length shortens such as to 10^-10 and 10^-11 s, the one-equation model predicts excessive temperature rise in the surface vicinity; however, two-equation and kinetic theory models predict similar temperature profiles. In this case, electron temperature rises rapidly while the lattice temperature increase slows down.

Silicon films were deposited at low temperature by remote plasma-enhanced chemical vapour deposition with H₂ and He plasma gases. The effects of H₂ and He plasma gases and plasma power on the film growth kinetics were studied. The composition, microstructure and optical properties of the deposited Si films were characterized by x-ray photoelectron spectroscopy, Fourier-transform infrared spectroscopy, ultraviolet-visible spectroscopy, Raman spectroscopy, x-ray diffraction, transmission electron microscopy and atomic force microscopy. It was found that He plasma is effective in increasing the deposition rate and reducing the hydrogen content in the film. H₂ plasma supplies excess atomic hydrogen which helps the formation of crystalline Si. The Si films deposited in H₂ plasma are polycrystalline with very small needle-like grains, which are perpendicular to the substrate and distributed uniformly in the thickness of the film. The structure of the Si films deposited in He plasma is amorphous. The crystallization mechanisms are discussed in terms of gas phase species and film surface reaction. It is believed that atomic hydrogen plays an important role in crystalline Si deposition. A linear relationship was found between the hydrogen concentration and optical band gap.

Vacuum emission of ballistic electrons from Au-SiO₂-Si structures has been investigated for oxide layers of various thicknesses d_ox = 13-500 nm. In particular for very thick layers and high field strength 5 MVcm^-1 we measure electron energies up to 200 eV, that is 80% of the maximum ballistic drive energy E_ball = eU_S. In order to enable such high-energy electrons to escape into vacuum very thin and deliberately non-perfect Au layers (d_Au = 8-12 nm) should be used as a top semitransparent electrode. Obviously, no energy stabilization of emitted electrons is observed. This is probably caused by defect agglomeration and the forming of defect channels. These defect channels should allow a nearly-ballistic electron transport until they are `burned-out' and `quenched-off' in a macroscopic breakdown process. What remains is damage in the form of `dead craters'.

In this paper, a Stoner-Wohlfarth approach to magnetic hysteresis is used as a model for a ferromagnetic composite rod consisting of single domain `grains' with random orientations embedded in a non-magnetic matrix. The rod is considered to be under torsional stress, which means that the rod at each point on its outer surface is experiencing biaxial stress along axes at 45° relative to the rod axis, with one stress equal to and the other equal to -. A magnetic field is considered to act parallel to the rod axis. The modified Stoner-Wohlfarth model of Callegaro and Puppin is adapted to this biaxial stress configuration. Results computed using the adapted model are presented and compared to results computed previously that were obtained by applying the domain wall pinning model to the torsional stress situation. It was found that the adapted model results are not very sensitive to the maximum value of the field, while the opposite is true for results from the domain wall pinning model. Secondly, the coercivity increased with torque, instead of the opposite situation with the domain wall pinning model, where coercivity decreased with torque. Reasons for this behaviour are presented in the paper. Differences related to symmetry properties of the model are also discussed. Finally, the model exhibits the possibility of both first- and second-order transitions in the hysteresis curve for different values of torque.

In beam lines and in synchrotrons various types of magnets of high-field quality are needed. The field quality of iron magnets mainly depends on the pole profile. In this work we have used the linear programming technique to optimize the pole profile for obtaining a very high-field quality while keeping the pole width small. The use of this method has been demonstrated in dipoles, quadrupoles and sextupoles.

The microstructure in a granular Cu₈₈Co₁₂ alloy was investigated by means of x-ray diffraction (XRD), a field ion microscope with atom probe (AP-FIM) and a field emission gun-transmission electron microscope (FEG-TEM) with energy dispersive x-ray spectroscopy (EDS). Giant magnetoresistance (GMR) and magnetic properties were studied to clarify the relationship between the GMR, magnetic properties and microstructures. The morphology of nanometre-sized cobalt granules within the grains were directly determined utilizing AP-FIM and FEG-TEM with EDS. Most cobalt granules are spherical and some others are plate-like in shape. GMR was observed in alloys with an average granule size ranging from 1.5 to 6 nm in diameter. TEM investigations showed that larger Co-rich precipitates formed at grain boundaries. The magnetic properties of the granular alloy were interpreted using the microstructure information and the dependence of GMR on magnetization was discussed.

MnBi_xAl_0.15 (0.4x0.9) thin films with an Al protective layer have been prepared using e-beam evaporation. The Kerr rotation as a function of the Bi concentration, x, has been investigated. Compared with MnBi_x thin films, it is found that a large enhancement of Kerr rotation appears in the MnBi_xAl_0.15 thin films with 0.4x<0.7, but no enhancement of Kerr rotation appears when x is greater than 0.7. When x = 0.5, a maximum Kerr rotation of 2.75° is observed at 633 nm for the MnBi_0.5Al_0.15 thin film, which is much larger than that of 1.56° for the MnBi_0.5 thin film. The possible reason for the enhancement of Kerr rotation is discussed.

Fourier transform-infrared studies have been carried out on the non-stoichiometric Ni-Zr substituted barium hexaferrite synthesized through the citrate precursor method. We observed approximately 30 bands, almost twice as many as reported for stoichiometric samples. These are attributed to the shifting of the vibrational modes under lattice distortion, which produces significant reduction in anisotropy and coercive fields without affecting magnetization.

A transformation of a plane electromagnetic wave in a medium with permittivity and conductivity changing in time as a finite sequence of rectangular periodic pulses is considered. An exact solution of the problem is derived and the influence of the conductivity on the wave parametric amplifications is investigated. A transition to intermittent chaotic behaviour of the secondary waves is shown to occur and the Hurst index is numerically estimated.

A unified electrodynamic approach to the guided-wave excitation theory is generalized to the waveguiding structures containing a hypothetical space-dispersive medium with drifting charge carriers possessing simultaneously elastic, piezoelectric and magnetic properties. Substantial features of our electrodynamic approach are: (i) the allowance for medium losses and (ii) the separation of potential fields peculiar to the slow quasi-static waves which propagate in such active media independently of the fast electromagnetic waves of curl nature. It is shown that the orthogonal complementary fields appearing inside the external source region are only associated with a contribution of the potential fields inherent in exciting sources. Taking account of medium losses converts the usual orthogonality relation into a novel form called the quasi-orthogonality relation. Development of the mode quasi-orthogonality relation and the equations of mode excitation is based on the generalized reciprocity relation (the extended Lorentz lemma) especially proved for this purpose that allows for specific properties of the space-dispersive active media and the separation of the potential fields. The excitation equations turn out to be the same, in form, whatever the waveguide filling, including both the time-dispersive (bi-anisotropic) and the space-dispersive media. Specific properties of such media are reflected in a particular form of the normalizing coefficients for waveguide eigenmodes. It is found that the separation of potential fields reveals the fine structure of the interaction between the exciting sources and mode eigenfields: in addition to the exciting currents (bulk and surface) interacting with the curl fields, the exciting charges (bulk and surface) and the double charge (surface dipole) layers appear to interact with the quasi-static potentials and the displacement currents, respectively.

We analyse the well known Tamm problem, examining the motion of a charge in a finite space interval, with a velocity exceeding the velocity of light in the medium. By comparing Tamm's formulae with the exact ones we prove that the former do not properly describe the conditions for Cherenkov radiation. We also investigate Tamm's formula cos = 1/_n, defining the position of the maximum of the field strengths' Fourier components for the infinite uniform motion of a charge. Numerical analysis of the Fourier components of the field strengths shows that they have a pronounced maximum at cos = 1/_n only for the charge motion in the infinitely small interval. As the interval is increased, many maxima appear. For the charge motion in an infinite interval there is an infinite number of maxima of the same amplitude. Quantum analysis of Tamm's formula leads to the same results.

Radiation guiding properties in a sheet-beam free-electron laser are investigated. The relative roles of refractive guiding and gain guiding have been identified by scanning the wiggler strength from the Raman to Compton limits. A gradual transition of the eigenmode confinement property has been observed with a power law W~^s, where W is the eigenmode width and is a diffraction parameter. The power s decreases from ~0.71 in the Raman limit to 0.6 in the Compton limit.

When the series of voltage pulses is applied to a gas-filled discharge tube, one observes that a breakdown and the subsequent discharge will influence the probability of the next breakdown. Repeating the measurement of the breakdown delay time many times we can determine the distribution of and subsequently decompose breakdown delay time into a formative time lag and the statistical delay time. Our measurements show that even a measurement taken 10 min after the last breakdown is influenced by it. The processes in a volume being unimportant due to a flow mode of operation, it is proposed that the memory effect is caused by some surface processes. The experiments are carried out in a discharge tube with two plane-parallel electrodes separated by 2 cm in a pressure range of several Torr. In this paper, dependences of both statistical delay time and formative time lag on the repetition rate, discharge duration and anode voltage are presented.

The Druyvesteyn distribution function, an example of a non-Maxwellian distribution for electrons, is used to study the effects of electron energy distribution functions (EEDFs) on the floating potential of an isolated particle under thin plasma sheath conditions by comparing the results with a Maxwellian distribution, and an analytical expression x_f^D2(x_f^M)^1/2 is obtained, where x_f^D and x_f^M are the normalized floating potentials with Druyvesteyn and Maxwellian electron distributions, respectively. Finally, the results are extended to a whole family of EEDFs with the functional form f() = A_p^1/2exp(-B_p^p), where p is an arbitrary number p1.

In this paper we discuss the use of generalized Saha and Guldberg-Waage equations for calculating the composition of a two-temperature SF₆ plasma in a pressure range 0.1-1.6 MPa, for an electron temperature in the range 300-20 000 K. We compare two sets of two-temperature laws and we recommend the expressions proposed by the team from Eindhoven. Furthermore, knowing the great influence of the choice of the excitation temperature on the plasma composition, we propose, on the basis of a chemical kinetics approach, a simple general relation allowing the calculation of this temperature versus the electron temperature, the heavy particle temperature and the electron number density. The application of these relations to a real case of a two-temperature SF₆ arc plasma gives a realistic radial profile of the electron number density.

A Townsend-type discharge generated in a modified ionization cell with symmetrical short gaps between a high-resistivity semiconductor plate and two planar electrodes is studied. The spatial stabilization of the gas discharge cell with a planar GaAs semiconductor plate is studied over a wide range of gas pressure values 101-342 Torr. The characteristics of the system are obtained both without and with illumination of the semiconductor plate with light of a particular wavelength range to control the photoconductivity of the material. The semiconductor material is found to stabilize the discharge. It is shown that generation of carriers by a gas discharge establishes a positive feedback. A qualitative discussion of this effect is given, which includes avalanche formation in a system having high-resistivity semiconductor for the Townsend discharge region. Recording of the current-voltage characteristic between parallel-plane electrodes is realized.

Analytical results are presented concerning the thermophoretic force acting on a spherical particle suspended in the near-wall region of a low-pressure plasma processor. Because the mean free path length of gas molecules is not negligibly small in comparison with the characteristic dimension of the plasma processor, for this case, a complex temperature variation will occur near the processor wall which can be approximately treated by using a temperature jump at the wall associated with a linear temperature distribution extended from the wall to bulk gas. The temperature jump at the processor wall has been included in the present analysis to determine the temperature of the molecules coming from the bulk gas and incident upon the suspended particle. It is shown that the thermophoretic force acting on the near-wall particle depends not only on the temperature gradient within the bulk gas but also on the magnitude of temperature jump at the processor wall. Dependent on the gas type, the thermophoretic force acting on the near-wall particle may even be greater than its counterpart without accounting for the near-wall effect.

Experimental studies of a Laue-Laue double-crystal monochromator for x-radiation have been conducted. It has been shown that a high-frequency ultrasonic wave can be used to adjust reflectivity of the monochromator and significantly increase its maximum throughput. In the particular case of a quartz Laue-Laue double-crystal monochromator the throughput can be increased more than six times with simultaneous broadening of the monochromator rocking curve width of three-and-a-half times. The new monochromator system is of potential interest for x-ray applications that require adjustment of the flux and energy resolution.

When thermoplastic elastomers (TPEs) such as styrene-isoprene-styrene or styrene-butadiene-stryrene copolymers in an aligned lamellar or hexagonal morphology are stretched perpendicularly to the plates or rods, then above a critical stress the lamellae or cylinders buckle to form a chevron or zig-zag structure. We examine this instability both analytically and using finite element analysis. In the analytic work we treat the elastomer as if it were a homogeneous anisotropic material and describe the chevron formation in terms of strain fields at length scales larger than that of the microphase pattern. We find that in order to do this one must respect the underlying microphase structure in two respects: (i) in terms of a statement as to how the material anisotropy rotates with the material; and (ii) in terms of additional `material moduli' which represent couplings between the macroscopic strain fields and deformations that occur at the microphase length scale. We define moduli which relate to hard phase bending and to displacement of the soft phase relative to the hard phase. The analytical results are tested by comparison with finite element models which solve for the microscopic strain field, and which allow the examination of post-buckling behaviour. We find that for perfectly aligned TPEs there is a geometric instability towards a sinusoidal buckling profile, which evolves into the chevron shape on further strain beyond the instability. The buckling is associated with a sharp turnover in the stress-strain curve. We also set our work in the context of treatments of similar buckling instabilties found in the fields of structural geology and liquid crystals.

The perturbation caused by a vertical thermal resistance to the alternating current temperature field in a homogeneous sample is analysed by using a plane-wave spectrum for which the reference plane is perpendicular to the crack. This unusual Fourier analysis gives rapid access to the temperature field in the frequency domain at any place in the sample. This can thus furnish the source term for a complete thermoelastic local response and complete the available surface temperature data.

A study of the photo-induced and thermally activated tautomerization of 2-(2,4-dinitrobenzyl) pyridine (-DNBP) by solid state nuclear magnetic resonance spectroscopy (NMR) is presented. Whereas similar high-resolution ¹H and ¹³C spectra were obtained from polycrystalline powders before and after ultraviolet light irradiation, the corresponding spectral data obtained from PVC films doped with -DNBP show significant changes. The -DNBP phototautomer lifetimes in PVC films (2.4 h or 6 d) were determined from the proton NMR signal recovery. Scanning electron microscopy (SEM) and differential scanning calorimetry of the doped films show that, in addition to matrix isolated molecules, -DNBP is incorporated into the polymer matrix in the form of nanocrystalline aggregates. At high guest concentrations, the crystallites can be modified by controlled solvent evaporation, as monitored by NMR and SEM. The variation of the ¹H spectrum with temperature shows the presence of at least two stable conformational isomers in the solid samples.

The composite structure of glow peak 5 in LiF:Mg,Ti (TLD-100), has been investigated using optical bleaching by 310 nm (4 eV) light. The decay of composite peak 5 and the growth of peak 4 are described by two exponential components of the same mean lives for both peaks. Using computerized glow curve deconvolution and contrary to previous investigations, it is determined that peak 4 does not decrease in intensity at long bleaching times and is therefore not associated with the 4 eV absorption band. The fast and slow components of the exponential decay are associated with the bleaching of peaks 5a and 5, respectively. Optical bleaching as a function of the temperature of post-irradiation annealing reveals that the conversion efficiency of peak 5a to peak 4 is very high, at a constant value of approximately 30%, whereas the conversion efficiency of peak 5 is no greater than a few per cent. The behaviour of the conversion efficiency as a function of the temperature of the post-irradiation anneal is explained in the framework of the spatially correlated trapping centre/luminescent (TC/LC) centre model. In this model, the TL of peak 5a results from the geminate recombination of a locally trapped electron-hole pair: peak 5 results from recombination via electron diffusion in the conduction band following thermal release of a singly-trapped electron in the TC/LC structure. Peak 4 arises from a singly-trapped hole in the TC/LC structure. The high conversion efficiency of peak 5a to peak 4 arises from direct optical ionization of the electron in the electron-hole pair, leaving behind a singly-trapped hole (glow peak 4), a direct mechanism not subject to competitive processes. Optical ionization of the `singly-trapped' electron (peak 5), however, can lead to peak 4 only via a multi-stage mechanism involving charge carrier transport in the conduction band, a mechanism subject to competitive processes, which leads to an order-of-magnitude decrease in the observed efficiency

We have developed a new design for a three-dimensional (3D) micro electrode octopole, driven by rotating electric fields in the MHz range. It allows simultaneous caging and rotation of single cells or particles with a size between 5 and 30 µm. The conflict between particle caging and the induction of particle rotation has been solved by breaking the symmetry of the electrode arrangement. The octopole consists of two layers of planar electrodes processed on glass and separated by a 40 µm thick polymer spacer, which forms a flow channel. Unlike planar four-electrode devices, the 3D octopole electrode cage can trap micro objects in free suspension. In addition, we have developed a new method for the automatic detection of single particle dynamics (MOSPAD). We also give calculations for dielectric forces and torque moments and a comparison of differences to four electrodes structures.

The discussion of bound states in section 2.2 of our paper 1993 J. Phys. D: Appl. Phys.26 798-803 was incorrect. A proper treatment is presented here.