Table of contents

In the subpicosecond time domain, the characterization and the control of ultrashort light pulses produced by means of femtosecond lasers are important problems. The low repetition rate of laser sources makes classical characterization methods, such as multishot autocorrelators, difficult to use. New different single-shot measurements of the temporal profile and of phase shifts introduced by nonlinear optical phenomena on femtosecond pulses are reported. These techniques are based on spatio-temporal and spectro-temporal transformations.

By considering the data for a 38 cm³ Ge(Li) detector from E_gamma =319.80 to 2598.80 keV, and for a 68 cm³ HPGe detector from E_gamma =223.430 to 3253.610 keV, it has been demonstrated that the contribution of the pair production process to the full energy peak efficiency (FEPE) of germanium detectors can be quite adequately accounted for in a semi-empirical way.

Numerical procedures for reconstructing dispersionless refractive profiles from a partial knowledge of the reflection coefficient are investigated. Particular attention is given to testing the procedures on profiles that are built up from piecewise linear components. A band-limited knowledge of the reflection coefficient is assumed to be the only input data. Effects of noise are also examined. The procedures investigated produce satisfactory results for simple profiles, but it is found that a band-limited knowledge of the reflection coefficient is insufficient when dealing with complex profiles.

A closed form solution for transient reflection from a lossy Lorentz medium half-space has been expressed as an infinite series of Bessel functions of the first kind and fractional order. The delta pulse plane wave excitation in the case of TM polarization has been considered. The variation of a reflected waveform with the angle of incidence, normalized collision frequency and normalized characteristic binding frequency of a lossy Lorentz medium are presented in several computer plots. The time domain Brewster angle is noted in the early time response. It is shown how the obtained transient response might be used for diagnosing the main characteristic parameters of the Lorentz medium.

The free electron laser on the basis of parametric (quasi-Cerenkov) instability is considered. Threshold conditions for this process are obtained. It is shown that rather rigid requirements for the particle beam are needed in the X-ray region.

Presents a theoretical model of a transversely excited CW CO₂ laser. The laser rate equations as well as the basic hydrodynamic and thermodynamic equations have been used to describe the whole laser system. The spatial distributions of small signal gain and saturation intensity along the gas flow direction have been calculated for a given set of initial conditions, namely, pumping rate, gas mixture, total pressure, gas temperature and gas flow velocity. The dependence of laser output power on discharge and flow parameters is in good agreement with the experimental data, confirming the usefulness of this model for understanding laser performance.

A free electron laser with a wiggler, which depends on the radial coordinate, is considered. The coupled wave equations and the growth rate in the laboratory frame are presented in terms of overlap integrals. Interactions which are not allowed under the assumption of a radially uniform wiggler are also found.

An analysis of the excitation of a space charge wave and a predominantly transverse magnetic mode in a cylindrical metallic waveguide filled with a relativistic electron beam is presented. The beam is guided by a uniform, axial magnetic field of arbitrary magnitude and passes through a spatially periodic, longitudinal electric (wiggler) field. A non-linear wave equation is derived in a form which includes the coupling of TM and TE modes due to the finite axial magnetic field. For the case of a weak and purely axial wiggler field, a solution of this wave equation is developed with each of the two excited waves expressed as a sum of three partial waves. A spatial growth rate formula is derived in a form appropriate for operation infinitesimally above cut-off of the TM mode. Numerical calculations indicate that the axial magnetic field magnitude should be somewhat larger than the value which maximizes the growth rate in order to achieve high efficiency.

The weld bead formed when a laser is used for penetration welding usually exhibits a ripple pattern superimposed on the surface. The reasons for this are not clear but the frequencies most likely to be observed will be related to the natural frequencies of oscillation of the weld pool when it is molten. Postacioglu et al. (see ibid., vol.22, p.1050, 1989) have already studied these frequencies in the case when the weld pool is almost circular and deep. When the welding speed is higher however, the weld pool is nearer to being ellipsoidal in shape. The frequency of the first asymmetric mode of oscillation is calculated, and is found to depend on the shape of the weld pool.

The flow of a viscous incompressible fluid between two parallel discs which at time t are spaced a distance H(1- alpha t)^1/2 apart and are rotating with angular velocities proportional to (1- alpha t)^-1 has been studied. Account is taken of the heat transfer between the two discs. The governing Navier-Stokes equation and energy equation are transformed to a set of ordinary differential equations. Approximate and exact solutions to these equations giving temperature distribution are obtained in the case where the upper disc is isothermal and the lower disc is either isothermal or subject to a constant heat flux beta . It is found that the temperature of the fluid increases with the decrease of both the Prandtl number Pr and the parameter beta individually, keeping the other constant. It is established that the rotation of the two discs has a very small effect on the temperature of the fluid and the heat transfer process, while the rapid normal motion of the upper disc (R₂>or=1) has a dominant effect on the temperature of the fluid and the heat transfer process. The results show rather good agreement between the approximate and exact solutions.

The net emission coefficient of SF₆-N₂ mixtures was calculated for temperatures between 8000 and 24000 K and pressures between 1 and 8 bar. The calculation of the radiative transfer of the lines required a study of line broadening and was performed assuming that line overlapping had a negligible influence. The results show that the resonance lines, although they are strongly absorbed, make up a large part of the net radiation-about as much as the continuum. Also, the nitrogen lines are more strongly absorbed than the lines of sulphur or fluorine. At low temperatures the radiation of SF₆ is stronger than that of nitrogen. When T>12000 K the net radiation of the mixture is not proportional to the net radiations of the gases owing to variations in emissivity and line broadening with temperature and composition of the mixture.

The effect of a-Si:H thin film layers on the double-probe plasma measurements is experimentally and theoretically investigated. The results show that the thin film layer acts as a semiconducting layer or Schottky barrier between the metal-semiconductor interfaces depending on the film thicknesses. Hence, the non-linear effects on the double-probe characteristics are observed. Due to the a-Si:H thin film deposition, the plasma parameters determined by the double probes will underestimate or overestimate plasma density and electron temperatures.

In a 3.7 MPa electrode-stabilized high-pressure mercury discharge the temperature profile is determined by emission coefficient measurements of the 577 nm mercury line. Values of the net-emission U_e as radiation loss per unit volume are derived assuming the electric field strength decreases with increasing distance from the arc axis. The results can be fitted by an expression of the form u_e approximately p exp(-E/kT) where p is the pressure, and E=7.5 eV is an average excitation energy. The numerical value of E is in satisfactory agreement with earlier determinations of this quantity. The model calculations show that there is no significant current dependence of the axis temperature at currents above 1 A but that there is a strong correlation between the axis temperature and the field strength.

Voltage-current density (V-j) characteristics were studied in the cathode region of noble gas mixture glow discharges. Measurements were carried out in He-X (X=Ne,Ar,Kr,Xe) mixtures in plane cathode and hollow cathode arrangements. The partial pressure of the admixed gas was at a maximum of 2% of the total pressure. It was found that at constant voltage the current density increased by mixing Ar to He, while it decreased when Ne, Kr or Xe were added. Above a certain partial pressure value the current density was found to increase again in the case of Ne and Kr admixtures. In He-Xe it decreased, however, in the whole range investigated. These observations are explained, taking into account ionization processes and the differences in secondary electron emission coefficients.

Following an approach described recently for the calculation of the threshold intensities of breakdown in electronegative gases, the authors examine the influence of some physical processes (i.e. multiphoton ionization (MPI) and collisional ionization (CI) of the electronic excited states) on the ionization growth rate in these gases. Particular attention is devoted to oxygen at a pressure of about 10⁴ Torr, where a reasonable concentration of O^- ions is produced. Calculations of the EEDF are carried out for different cases of presence and absence of secondary ionization processes. The results show that the attachment losses drastically reduce the influence of the secondary ionization processes on the EEDF. This means that calculations of the EEDF can be carried out ignoring the effect of these processes.

Experimental results concerning the changes in temperature and particle densities in a SF₆ arc in extinction were compared with results obtained from modelling based on the hypothesis of thermal equilibrium. The model predictions for cooling rates were in agreement with experimental results. In particular, if the equilibrium is maintained, the time constant of the changes in electron density varies from 20 mu s to about 8 mu s when the arc diameter is reduced from 5 to 3 mm. Finally, a kinetic model showed that the mechanisms of electron/positive-ion recombination and of attachment are efficient enough to ensure the state of collisional equilibrium during arc extinction as long as the temperature is greater than 8000 K.

Investigations of the diffuse plasma expanding from cathode arc spots into a vacuum have revealed some unexpected properties calling for physical explanation. A theoretical model of such plasmas which is based on one-dimensional time-independent hydrodynamic two-fluid equations can be solved analytically in the form of asymptotic power series approximately describing the plasma parameters as functions of the variable s=(I/r)^2/5(current I, distance r). The main results give a quantitative decomposition of the force accelerating the ions into three partial forces caused by the electric field, by the ion pressure gradient and by the electron-ion friction, which are of comparable importance. Unlike the ions, the electrons are accelerated by the pressure gradient only, but are decelerated both by the field and by friction; all these forces cancel each other almost completely. The direction of the electric current is opposite to the direction of the field; therefore the resistance of the plasma is negative. The limitations of the validity of the derived solution are discussed, as are possible extensions and simplifications of the model.

Electron drift velocities for CH₄ gas determined by Monte Carlo simulations of Schlumbohm's (1965) and Frommhold's (1959) experiments, in which drift velocity is defined by the gap length divided by electron transit time, are compared with theoretical drift velocities such as the diffusion-modified drift velocity V_d of the steady-state Townsend, the average drift velocity of electrons W_v of the pulsed Townsend, the centre-of-mass drift velocity W_r and mean-arrival-time drift velocity W_m of the time-of-flight experiments when ionization is appreciable. The results show that the drift velocities deduced by the simulation of Schlumbohm's and Frommhold's experiments do not agree with any of the theoretical electron drift velocities, and suggest that the drift velocity from Frommhold's experiment coincides with an electron drift velocity represented by (W_r+W_m)/2, while the drift velocity from Schlumbohm's experiment assumes a value close to but slightly larger than W_m.

Describes a new method of apparent mobility measurement in a space charge flow field generated by corona discharge. The positive helium ion mobility at cryogenic temperatures down to 4.2 K is newly measured and the results obtained agree well with theoretical values for the atomic helium ion He⁺. The mobility of negative charge carriers is also obtained at 4.3 K. The results suggest the transition from free electron to electronic bubble in dense helium gas and support the discharge mechanism of a negative pulse corona in which an electronic bubble has the same space charge effect as a negative ion in electronegative gas.

The calculations cover the whole range of low electron densities to sufficiently high electron densities. Typical values of the ratios of the electric field to the gas density are in the range of 10^-16<E/N<10^-15 V cm². The results of the calculations show that as ionization degree n_e/N (n_e, N being, respectively, the electron and gas density) increases, the electron energy distributions move towards a Maxwellian distribution and the drift velocities are enhanced in the low E/N value region.

Current density-voltage (j-U) characteristics and vibrational excitation efficiencies of RF excited CO₂ laser discharges have been calculated in the frame of a self-consistent numerical discharge model. The model includes the continuity equations for electrons and positive ions Poisson's equation and the energy balance equation for the neutral gas. The results are compared with experiments. The slope of the j-U characteristic decreases with increasing frequency and with increasing pressure. The vibrational excitation efficiency increases with increasing frequency and decreases with increasing RF power and decreasing pressure.

PRAL, a projected range algorithm is widely used to predict the first few moments of the spatial distribution of ions implanted into amorphous targets of infinite extent. It is shown here that PRAL contains mathematical defects. These are corrected in the new Kent range algorithm (KRAL) which is shown to be compatible with LSS theory. KRAL and PRAL predict the same range but the new algorithm is superior in its predictions of lateral and longitudinal standard deviations when compared with moments derived from the Monte Carlo code TRIM using the same differential nuclear scattering cross sections and electronic energy loss mechanisms.

The microstructure and m of a material composed of an oriented lamella stack are described. The material used for these studies was an S-B-S block copolymer exhibiting a lamellar morphology. By processing this material, using a novel injection moulding procedure, large specimens can be produced in which the lamellar microphases are uniformly oriented. This provides a unique opportunity for studying the anisotropy of the mechanical properties of a well characterized lamella stack, in which the two phases are molecularly bonded together. The experimental data are compared with predictions derived from composite mechanics. Excellent agreement between theory and experiment is found and this provides information concerning Poisson's ratio for the rubbery phase.

A modification of the conventional capacitive dilatometer, a four-rod assembly, was used to measure the thermal density variations in (Ar)_1-x(N₂)_x alloys with x=0.22, 0.71 and 1. The results are in favour of complete solid state miscibility, contrary to the recent findings of Yantsevich et al. (1986). The potential use of the four-rod capacitor as a device to detect molecular electric quadrupole relaxations at audio frequencies is discussed.

Solid thermal conductivities of silica aerogels with densities between rho =62 and 262 kg m^-3 were measured within the temperature range T=5 to 300 K. Instead of the usual four-point bar method, a flat plate geometry was employed. Radiative contributions could thus be reliably controlled. Determination of the spectral infrared optical absorption coefficient and use of suitable approximations for the equation of heat transfer allowed separation of the contributions from radiative transport and solid conduction. Thermal contact resistances between the specimen and the boundaries could be eliminated by a variation of the pressure load onto the specimens. Between 10 and 100 K the measured solid thermal conductivity scales with temperature according to lambda varies as T^alpha , with alpha =17+or-0.2. Above 100 K the temperature dependence agrees well with that for non-porous SiO₂ (vitreous silica). The absolute value of lambda , however, is reduced by a geometrical factor g, which strongly depends on density; g is found to be between 10² and 10³ for aerogels.

The authors report room temperature measurements of the surface potential dark decay in amorphous poly(N-vinylcarbazole) films over the field strength range 0.6-2.6*10⁸ V m^-1. A model of trap-controlled hopping is used to describe the experimental curves. The decay curves are influenced, not only by shallow, but also by deep traps. The model parameters such as the effective hole mobility ( approximately=10^-16 m² V^-1 s^-1 corresponding to the mobility of charge carriers controlled by shallow traps deeper than 0.55 eV), the hopping distance ( approximately=0.6 nm) and the capture and release rates of charge carriers in a deep hole trap were determined by curve fitting. The effective hole mobility as well as the trapping time were found to be weak functions of field strength. The depth of the deep trap was estimated to be 0.9 eV.

Reversible light-induced fatigue of two-layer organic photoconductors consisting of a charge generation layer (CGL) and a charge transport layer (CTL) was studied by time-of-flight (TOF) photoconductivity measurements. The CGL contains 50% X-metal-free-phthalocyanine in a polyester and the CTL is a solid solution of 33% p-diethylaminobenzaldehyde-diphenylhydrazone in a polyester. Prolonged light exposure under applied fields above 10⁵ V cm^-1 causes remarkable changes in electro-photographic properties and the TOF current pulse shape. After light exposure the TOF current increases and the drift velocity of holes through the CTL decreases compared with those before light exposure. From the experimental results, it is concluded that light-induced fatigue of the two-layer photoconductor is a consequence of the build-up of the negative space charge caused by the electron trapping in the CGL and the positive space charge caused by the hole trapping at the CGL-CTL interface. The TOF current waveform strongly reflects the internal potential profile modulated by these space charges. A rest in the dark reverses the TOF current waveforms. The recovery process of TOF waveforms shows that holes trapped at the interface are first released, and then electrons trapped in the CGL are released thermally.

Charge transfers very easily across a metal-SiO₂ interface even though the charge is completely immobile in the oxide. To a first approximation, the charge transfer is such as to maintain a certain fixed field across the interface; as one should expect if there were a high density of electron states on the insulator surface, able rapidly to exchange electrons with the metal. However, detailed investigation shows that the interfacial field is not maintained exactly constant but depends to some extent on the previous history; for example, an electric field influences charge transfer even if it is switched off before the contact is broken.

Metal/Langmuir-Blodgett film/metal structures based on amphiphilic charge-transfer materials have been fabricated and the DC conductivity perpendicular to the plane of the film has been measured. Linear current-voltage behaviour was observed for small applied biases. In contrast, the high-field (>10⁵ V cm^-1) current against voltage characteristics exhibited a log (current) against voltage^0.5 dependence, similar to that measured for fatty acid multilayers. No evidence for rectification has been found in structures containing both donor and acceptor materials. However, devices incorporating a tetracyanoquinodimethane derivative invariably possessed output voltages with no applied bias; a significant dispersion in the low-frequency capacitance was also noted for these structures.

The electrical response of dielectric materials is often graphically represented by means of a plot in the complex impedance (or admittance) plane. In many cases these impedance plots are semicircles. An equivalent circuit consisting of a capacitor (or a constant phase element) and two resistors is often used to describe such a semicircle. It is shown here that even if the experimental data are lying on such a semicircle they do not necessarily correspond to that equivalent circuit because of a different frequency dependency. In addition it is shown that the Kramers-Kronig transforms can be very helpful but are not decisive for the determination of whether experimental errors are made.

For aqueous solutions, the relation between ion-ion interaction the ionic activity coefficient and the rate of water tree development is considered. It is argued that a decrease in the activity coefficient may lead to a decrease in tree growth. Experimental results with 0.1 M solutions of LiCl, NaCl, KCl, RbCl, MnCl₂, and FeCl₃ show agreement with the proposed model. However, results with 0.5 M solutions show that the controlling influence of ion-ion interactions may be offset by the obstructive presence of hydration sheaths.

Chemisorption has been used to immobilize uniform, low-defect density monolayers of aminopropylsilane and of d-biotin on evaporated gold substrates. The quality of the monolayers has been confirmed by surface potential measurements and by copper decoration. Avidin has been immobilized to these monolayers by (i) crosslinking to the aminopropylsilane with glutaraldehyde and (ii) binding directly to the biotin ligand. The changes in surface potential observed during each immobilization step are shown to be related directly to the molecular structure of each chemisorbed layer. Significantly, when the avidin is immobilized on the biotin monolayer the tetrameric protein is orientated with one pair of biotin binding sites on the upper surface of the protein monolayer. This allows the bifunctional ligand, bisbiotin, to be bound to the protein giving the possibility of attaching further protein layers to form molecular organizates suitable for molecular electronic and molecular sensing applications.

It has been shown that the points of inflection of a thermoluminescence peak can indicate its order of kinetics. Also a new set of expressions is derived to evaluate the thermal activation energy of a thermoluminescence peak of any order of kinetics using the points of inflection. The validity of these expressions is discussed by calculating the thermal activation energy of a number of numerically computed thermoluminescence peaks using the present sets of expressions and comparing them with those calculated using the formulae of Land.

An attempt has been made to model the ZnS:Mn,Cu powder DC electroluminescent (DCEL) display system by means of an equivalent circuit. The object in so doing has been to define a framework for an objective measure of the effect of device parameters on the performance of any DCEL display, and thus reduce the subjectivity which presently dominates DCEL display development. The display has been investigated as a two-port network in both the forward and reverse directions. The parameters measured at the input port are current and voltage while the parameter measured at the output port is luminance. From investigation of the relationship between these parameters it has been possible to define the system input impedance, output transfer function and a feedback transfer function of voltage produced by incident light on the display.

Manganese doped (0 to 10%) zinc sulphide films have been prepared by vacuum evaporation. The films are polycrystalline and the lattice constants do not change with Mn mol.%. The optical transmission has been recorded in the wavelength range 300-900 nm. The optical band gap and refractive index have been evaluated. The band gap of these films varies from 3.48 eV (undoped) to 3.70 eV (10% Mn). The refractive index dispersion data have been fitted to the single oscillator model. The average oscillator strength S₀ (1*10¹⁴ m^-2) is constant with Mn doping and the refractive index dispersion parameter E₀/S₀ varies from 6.75*10^-14 eV m² (undoped) to 6.24*10^-14 eV m² (10% Mn). The variations in band gap and refractive index dispersion parameter have been discussed.

The deposition of PbS thin films from chemical baths is significantly accelerated when exposed to solar radiation or other sources of illumination. The enhanced deposition rate is seen to be a cumulative effect of an increase in the bath temperature due to photothermal conversion as well as to photoactivated deposition at the PbS film. For deposition employing a 1:1 ratio of Pb²⁺:thiourea (TU) in the bath, a film thickness of approximately=0.09 mu m, corresponding to an integrated reflectance in the visible region of approximately=17% and producing a purple appearance in reflected daylight, is obtained in 40 min under solar radiation of 750 W m^-2 as opposed to 80 min inside the laboratory (26 degrees C) and to 55 min when the bath temperature simulates the variation in temperature (i.e. up to 36 degrees C) observed in the bath under solar radiation. The time lag between the three cases can be varied by varying the bath composition. Novel applications of the photoaccelerated deposition process for producing a variably reflective glaze (golden, purple, bluish, etc.) on the glass surface for decorative coatings and to obtain pleasantly reflective latent images, such as a purple image on a golden background, are presented.

The static yield stress of magnetic colloidal suspensions is measured as a function of the applied magnetic field in a controlled stress rheometer. The experimental results are analysed with the help of the magnetization curve for different volume fractions. The experimental yield stress scales as the square of the magnetization but its magnitude is always lower than the one predicted on the basis of interparticle forces. The nature of the material used for the plates of the rheometer cell, as well as the surface roughness, is shown to have a large effect on the measured yield stress.

A hydrodynamic flow in circular free suspended films of octyloxybenzoic acid (OOBA) in nematic phase before transition to smectic C (110 degrees C<or=T<or=115 degrees C) was observed. The conclusion is made that the flow is generated by the nonlinear temperature dependence of the surface tension which changes its slope at about 115 degrees C from negative to positive.

Reproducible and well defined small hysteresis loops of rhomboid shape were measured on thin uniaxial magnetic garnet films. Observation of the domain structure showed that the change in the magnetic moment described by these loops was due solely to elementary translations of domain walls. The authors have called the half-width of these loops the domain wall coercive field, H_dw; this parameter differed substantially from the half-width, H_c of the major saturation-to-saturation hysteresis loops. They called the rhomboid small loops the domain wall coercive loops (DWCL) and showed that H_dw obtained from DWCL was identical to the coercive field value measured by the low frequency domain wall oscillation method. The slope of DWCL coincided with the initial susceptibility of stripe domain structures and thus made it possible to calculate the basic parameters of the sample, giving results in agreement with other established methods of measurement of these parameters.

The X-band EPR spectra of iron-containing VPI-5 were recorded at room temperature and liquid nitrogen temperature. An intense signal at g/sub ///=2.0 and g_{perpendicular to} =6.0 characteristic of high-spin Fe³⁺ ions in axial symmetry was observed at both temperatures. A splitting of these components due to superhyperfine interaction indicated the presence of these Fe³⁺ ions in the network-forming position. An anisotropic line at G=4.3 and weak superhyperfine lines centred at g=1.7 characteristic of high-spin Fe³⁺ ions in orthorhombic symmetry were also observed at liquid nitrogen temperature. These signals may also be attributed to the presence of Fe³⁺ ions, most probably in the network-forming position. Lastly, the very weak signal at g=3.1 was tentatively assigned to traces of occluded compounds of iron outside the framework.

The flow stress of many metals is known to vary more strongly with strain rate epsilon above a critical value epsilon _c than below it. An analysis of previous authors' measurements on copper indicates that the value of epsilon _c varies systematically with specimen size. An estimate of inertial stress shows that inertia may be an important mechanism behind this observation. Qualitative arguments suggest that deformation velocity, friction and wave propagation could also contribute.