Table of contents

Back-reflection through a deep random phase screen is investigated for the case of an incoherently scattering reflector. General results are derived for simple imaging geometries which predict that, in scattering-limited situations, active object illumination can lead to a modest improvement in performance in comparison with passive imaging, but not to diffraction-limited operation. The predictions are explored with the help of an analytically tractable model.

In non-resonant stimulated Raman scattering under phase-matched conditions the Stokes-antiStokes coupling results in suppression of the exponential gain of both the Stokes and the antiStokes fields to zero. This phenomenon, known as parametric gain suppression, leaves the intensity of the laser beam passing through the medium practically unchanged, as the Stokes and the antiStokes processes cancel each other. The authors consider the effect of Stokes-antiStokes coupling on the gain of these two fields in the case of resonant stimulated Raman scattering (RSRS), when the pump laser is resonant with one of the intermediate states. The analysis shows that as one approaches the conditions for RSRS, the gain suppression due to the Stokes-antiStokes coupling under phase-matched conditions becomes smaller and smaller, and the gain of the Stokes intensity approaches the value for zero Stokes-antiStokes coupling. Furthermore, one finds that sufficiently close to resonance the antiStokes gain is largest under complete phase-matched conditions (although considerably smaller than the gain for the Stokes field under the same conditions), and it decreases to zero with increasing mismatch with a Lorentzian dependence of the gain on the mismatch parameter Delta k The cancellation of the parametric gain suppression for RSRS is a result of the fact that near resonance the non-linear susceptibility at the Stokes frequency becomes much larger than that at the antiStokes frequency.

Radiative transfer through a parallel-plane aerosol medium has been studied. The Mie theory is used to calculate the radiation transfer scattering parameters in the form of the extinction, absorption and scattering efficiencies. Averaged efficiencies over different size distributions for spherical particles of complex refractive index are calculated. The radiative heat fluxes through a plane medium of sodium and polydispersion aerosol are calculated.

The problem of calculating eddy currents and transformer signals in electromagnetic flowmeters is considered and a procedure for solving this is given. In this way Wyatt's theory (1966) relating base-line error to contact impedance effects at the electrodes is confirmed on the basis of the full field equations. A simple relation is shown to exist between the axial component E_z of the eddy current E field in the liquid and the potential U induced by a flat profile flow. Using this relation it is proved that there is no way of reducing base-line error by magnetic field shaping. The same relation between E_z and U provides the basis of a new technique for sensitivity measurement or self calibration of electromagnetic flowmeters.

Time dependent population densities of excited states of Ar and Ar⁺ have been measured by time and space resolved optical emission spectroscopy in an asymmetric parallel plate plasma reactor in Ar at an excitation frequency of 13.6 MHz. Time averaged ion densities in the plasma have been determined by a floating double probe. At pressures above 20 Pa and at distances from the powered electrode beyond the emission intensity maximum the population densities are modulated with essentially the fundamental RF frequency. In order to explain this experimental result the bulk plasma was modelled under the assumption that the local electric field is the sum of a time dependent part oscillating with the fundamental frequency and a constant part of similar size. For the calculations, the electron Boltzmann equation and a set of rate equations were simultaneously solved in a fully time dependent way. The observed modulations of the emission intensity of the spectral line of Arl at 516.23 nm and of the calculated level population of the emitting 6d(1/2)₁ state agree within the experimental errors.

Experiments to study negative ion densities in a radio-frequency CF₄ plasma have been carried out using a photodetachment technique. Electrons are photodetached from the negative ions using the pulse of a Nd-YAG laser at the tripled (355 nm) or the quadrupled (266 nm) frequency. The photodetached electrons are detected by a microwave method as a sudden increase of the electron density in the plasma. The negative ion density, which consists mainly of F^- is found to be typically four times higher than the stationary electron density at a pressure of 13 Pa, an RF power of 15 W and a CF₄ flow of 15 SCCM. The measured decay of the detached electrons after the laser pulse has been interpreted in terms of electron attachment and ambipolar diffusion. The results demonstrate the possibilities for use of this technique to evaluate attachment coefficients in active plasmas. The attachment rate constant for CF₄ is found to be (7+or-1)*10^-17 m³ S^-1 at RF powers of 15 W. The electron diffusion coefficient is 0.13+or-0.12 m² s^-1 at standard conditions of 1 Torr and 300 K.

The second derivative of a Langmuir probe characteristic is used to establish the electron energy distribution function (EEDF) in both a tandem and hybrid multicusp H- ion source. Moveable probes are used to establish the spatial variation of the EEDF. The negative ion density is measured by laser induced photo-detachment. In the case of the hybrid source the EEDF consists of a cold Maxwellian in the central region of the source; the electron temperature increases with increasing discharge current (rising from 0.3 eV at 1 A to 1.2 eV at 50 A when the pressure is 0.4 Pa). A hot-electron tail exists in the EEDF of the driver region adjacent to each filament which is shown to consist of a distinct group of primary electrons at low pressure (0.08 Pa) but becomes degraded mainly through inelastic collisions at higher pressures (0.27 Pa). The tandem source, on the other hand, has a single driver region which extends throughout the central region. The primary electron confinement times are much longer so that even at the lowest pressure considered (0.07 Pa) the primaries are degraded. In both cases the measured EEDF at specific locations and values of discharge operating parameters are used to establish the rate coefficients for the processes of importance in H^- production and destruction.

The ionization alpha , electron attachment eta and other electron transport coefficients for SF₆ and perfluorocyclobutane (c-C₄F₈) gas mixtures have been calculated by a three-term Boltzmann equation method over the E/N range 283-707 Td. A set of electron collision cross-sections that gives transport parameters in agreement with measurements for the pure gases is used. The calculation is performed for important experimental conditions, i.e. the steady-state Townsend (SST), pulsed Townsend (PT) and time-of-flight (TOF) methods. It is found that the greater the reduction in E/N the more pronounced is the peak of eta values of mixtures, e.g. at E/N=283 Td the peak appears at the SF₆ partial pressure k around 20%, though alpha values show monotonical variation with k, at a fixed E/N for the E/N range studied here. As a result, the E/N value at which alpha = eta , i.e. the limiting E/N of 20% SF₆ and 80% c-C₄F₈, can exceed the values for the respective pure gases. This fact suggests that this combination may be one desirable synergistic mixture since the limiting E/N is an important parameter for predicting the dielectric characteristics in electronegative gases.

The steady-state electron swarm parameters in a series of Penning gases, Ar/Ne, Kr/Ne, Xe/Ne, Hg/Ar and Hg/Kr mixtures, were analysed using a Boltzmann equation in which generation of secondary electrons through Penning ionization and ionization collision between two metastable Hg atoms, as well as electron-impact ionization, are property considered. The ratios of the Townsend first ionization coefficient to the field strength, alpha /E, calculated using the present equation in Ne gas with admixtures of Ar, Kr and Xe, were strongly influenced by Penning electrons. On the other hand, in Ar and Kr gases with Hg vapour, the ionization between two metastable Hg atoms was found to be the dominant ionization process at low E/N. The present alpha /E in every gas agrees well with experimental values over a wide range of alpha /E. The effective cross section of excitation for the metastable states of the rare atoms was derived by comparing the present alpha /E with the experimental values.

A kinetic model for the low-pressure oxygen positive column is presented and discussed. The model is based on the electron Boltzmann equation and the rate balance equations for the dominant heavy-particle species, which are solved simultaneously in order to take into account the coupling between the electron and the heavy-particle kinetics. The effects of vibrationally excited molecules, dissociated atoms and metastable states on the electron kinetics are analysed in detail. The predicted populations of O₂(X³ Sigma ), O₂(a¹ Delta ), O(³P), and O^- are shown to agree satisfactorily with previously reported measurements. A combination of this kinetic model with the continuity and transport equations for the charged species e, O^-, and O₂⁺ is shown to provide characteristics for the maintenance field that agree reasonably well with experiment.

The concentrations of O(¹D) and O(¹S) metastable atoms are measured in the positive column of a DC oxygen glow discharge by vacuum ultraviolet absorption spectroscopy at 115 and 122 nm respectively. In the investigated range of pressures p=0.2-2 torr and discharge currents I_d=10-80 mA, the concentrations of O(¹D) and O(¹S) atoms range from 10¹¹-10¹² and from 10¹⁰-10¹¹ cm^-3 respectively. In these conditions the main creation and destruction processes are discussed: the O(¹D) atoms are created by electron impact dissociation of O₂(X³ Sigma ) and O₂(a¹ Delta ), and excitation of O(³P), and lost through quenching by oxygen molecules. It is shown that it is necessary to add to the electron impact excitation another creation term of O(¹S) atoms to balance the destruction through quenching by O₂(X³ Sigma ), O₂(a¹ Delta ) and O(³P).

For pt.III see ibid., vol.23, p.526 (1990). The extensive collisional-radiative model is applied to plasmas generated by a RF coil in an atmospheric argon flow in order to investigate excitation mechanisms and departures from local thermodynamic equilibrium in several spatial positions in these discharges. Computations are carried out for various sets of input parameters, such as the electron kinetic temperature T_e, the atom temperature T_a, the electron number density n_e, the ground-state atom population n₁, the plasma column radius R and the escape factors Lambda _mn and Lambda _m, characterizing the non-equilibrium plasmas under consideration. The numerical modelling presented yields reliable information on the populations in the excited levels and on the validity of the so-called 'close to LTE concept' and the occurrence of an ionizing or a recombining regime in the locations investigated. The effect of the changes in the discharge parameters T_e, T_a, n_e, R and Lambda _1n on the excitation mechanisms, together with the role played by the inelastic atom-atom collisions and by the recombination flow of electrons from a continuum, are also shown.

The development of an arc leading to the steady-state condition has been investigated. Three types of transitions have been identified. Transition to a field dominated column and cathode arc within 100 mu s, transition to a thermally ionized column and field dominated cathode within 0.5 ms and transition to the final steady-state governed by the amount of heat gained as a function of time and processes at the cathode. V-I curves for cold arcs measured at 100 mu s after breakdown showed that the arc voltage is initially high, it is similar for both thoriated tungsten and copper cathodes and the arc has a positive resistance. Factors that affect the time taken to establish a steady state, such as cathode diameter, gap length, gas flow rate and the rate of current rise, have been examined. The results indicate two further advantages of having a power supply with a high rate of rise of current, both causing a faster transition of the arc to its steady state. The first is that the arc will initially have a large resistance and current simultaneously for a longer duration causing a larger amount of heat, the second is that there will be a higher rate of supplied energy during the whole of the transition period.

A fast backward-jump type arc motion is observed with an arc running on rails. It is shown by the use of conductivity probes that this motion does not correspond to a continuous displacement of an ionized high-temperature region. The model of current transfer between two parallel arcs is adopted to explain the jump motion, for which a high rising speed of the arc voltage is a decisive factor. The current transfer is considered to be different from the re-ignition of a post-arc current.

Various macromolecular parameters of air-dried Agave sisalana (sisal) fibre are investigated by small-angle X-ray scattering (SAXS) techniques considering it as a densely packed micellar system by applying the theories put forward by Vonk (1973). The authors show that the small-angle X-ray scattering intensities of de-waxed air-dried sisal fibre deviate from Porod's law, establishing that the fibre is non-ideal two-phase structure characterized by continuous variation of electron density at the phase boundary. Therefore the mean square electron density gradient < mod grad eta mod ²> in isotropic structure is proportional to the fourth moment integral s⁴/(s) ds of the SAXS intensity distribution and to the second derivative of the correlation function at the origin in reciprocal space where s is the radial coordinate in reciprocal space. Since sisal fibre has a two-space structure with unsharp phase boundaries, the above relations are used to find out the thickness E of the transition region. The most general approach to the small-angle X-ray scattering makes use of the correlation functions, which were introduced by Debye and Bueche (1949) and which contain implicitly all the information obtainable from the scattering curve. The scattered intensities and correlation functions are related by Fourier transformations. Other important physical parameters are obtained with the help of one- and three-dimensional correlation functions are D, the average periodicity transverse to the layer, S/V, the specific inner surface phi ₁, the volume fraction of matter, phi ₂, the volume fraction of void l_c, the length of coherence, l₁ and l₂, the transverse lengths, l_r, the range of inhomogeneity, f_c, the characteristic number and 2E/D, the volume fraction of transition layer.

The optical behaviour of surface-stabilized ferroelectric liquid crystal samples cannot be described using a uniform director profile. The simplest model based on the chevron layer structure and the surface alignment is a triangular director profile. Analytical expressions for the resulting transmitted light were calculated using the Jones matrix formalism and fitted to experimental data results of a smectic-C host made using an optical multichannel analyser (OMA). This simple model gives an excellent fit, with results that are consistent with X-ray measurements of the chevron structure.

A model of size-dependent melting of small spherical particles (with diameter larger than 2-4 nm) is used to estimate the variation of their melting temperature, when their size is varied. This variation depends on two bulk parameters: the bulk melting temperature and the energy of formation of intrinsic defects. It is estimated for 71 elements. A comparison with available experimental data shows a good or satisfactory agreement.

The structure of Langmuir-Blodgett multilayers of stearic, arachidic, behenic and 22-tricosenoic acids has been studied by RHEED and found to consist of a mixture of upright orthorhombic and tilted monoclinic nanometric grains. The upright orthorhombic grains are aligned with the b₀ axes parallel to the dipping direction while the tilted grains are azimuthally distributed by up to +or-70 degrees about the dipping direction. RHEED diffraction patterns appear as rectangular arrays of spots superimposed on parallel lines which extend in the a₀ or b₀ directions, indicating a much longer range of orientational order than of translational order. In the case of 22-tricosenoic acid the angle of inclination, theta , of the grains to the substrate normal is observed to vary continuously with the deposition pressure, theta decreasing as the pressure is increased. This behaviour has been associated with an area coverage mismatch which occurs on condensation of the bilayers to the solid phase on the substrate. By tilting over, grains are able to achieve greater substrate coverage, minimizing gaps in the structure and reducing the amount of free surface. For the saturated acids the inclination angle, theta , appears to be independent of the deposition pressure, falling into one of three tilt modifications. The low- and high-tilt arrangements have spacings characteristic of a monoclinic subcell (first setting) and correspond most closely to M(102) and M(101) structures. Electron diffraction patterns showing the 'medium' tilt display orthorhombic spacings. All these modifications show similar azimuthal spread about the dipping direction. No variation in grain orientation with increasing numbers of deposited layers is detected by RHEED in films thicker than two layers.

The photoelectric steady-state currents J_L in liquid Ar, liquid CH₄ and liquid N₂ were measured, changing the level of injection current. The current level was varied over a wide range using ultra violet light from a deuterium lamp. In liquid Ar and liquid CH₄, when the current in vacuum J_v (which is defined as an injection current) was less than 0.1 nA cm^-2. J_L was proportional to the external field E₀ at low fields, while at high fields J_L was proportional to E₀^1/2. However, by increasing J_V at low fields a space-charge-limited J_L was observed, with the linear relation between J_L and E₀^1/2 being unchanged at high fields. On the basis of measured curves of J_L against E₀, the distributions of space charge, field strength, and currents of electron and negative ion components within the gap were simulated. The space charge was presumed to appear due to accumulation of negative ions produced through attachment of electrons to oxygen impurities which might be included originally in gas samples. In liquid N₂, J_L was proportional to E₀^1.8-2, although it increased proportionally with J_V.

The electrical conductivity of polymer complexes of bis-2,6-diaminopyridinesulphoxide with CuCl₂, CuBr₂ and CuI₂ was measured at different frequencies and temperatures. It was found that the conductivity increases as frequency and/or temperature increases. The dielectric constant, dielectric loss and relaxation time of the polymer complexes were calculated.

Charge transfer almost almost always occurs when a metal touches an insulator. In general, repeated contact to the same place will transfer additional charge (charge accumulation). In many materials charge accumulation is clearly related to conduction. It is usually supposed that charge accumulation occurs because the charge deposited by the contact diminishes by some conduction process, and the surface states emptied of charge can be refilled on the next contact. The author describes a series of experiments which indicate that this is not the mechanism of charge accumulation in soda glass. He concludes from his observations that conduction influences charge accumulation because it modifies the interfacial electric field between the glass and the contacting metal; the amount of charge transfer is controlled by this interfacial field rather than by the number of accessible states.

Measurements of temperature dependence of the coercive field H_c are reported for Nd-Fe-B sintered magnets. The relationship between log H_c and log H_f was deduced from Gaunt's localized weak-domain-wall pinning model, which can be written in the following form, log H_c=log H_f+log(25 ((31rb²/25kT)-1)), where r is the domain-wall energy and b is the range of interaction between wall and pin. The values of log H_c were calculated using the above equation, and are in good agreement with experiment.

Proton spin-lattice relaxation times in the rotating and laboratory frames of five polysaccharides have been measured as a function of temperature and frequency. The proton second moment of glycogen has also been measured as a function of temperature. The results have been interpreted in terms of a model in which molecular motions are characterized by a log-normal distribution of correlation times. The T_1p results reveal a reorientation of CH₂OH groups. In the case of chitin, evidence of threefold reorientations of methyl groups has been found.

The electrical breakdown phenomenon in polyethylene for point-to-plane gap configurations has been investigated using a photo-optical current-measuring technique and a 760 ns rectangular high-voltage pulse generator. The waveform of prebreakdown current in polyethylene is essentially the same as those in liquid dielectrics. Discharge from a positive point is more intense and faster than that from a negative point. A linear relation to formative time lag and gap spacing is obtained for gap lengths greater than some critical value, indicating a constant propagation velocity in this region. The velocities are deduced to be 1.7 km s^-1 for a positive point at 50 kV and 0.38 km s^-1 for a negative point at 60 kV. These values are coincident with those of liquid dielectrics.

Steady-state space-charge and field distributions in insulators with needle-plane electrodes have been examined theoretically by making use of hyperboloid geometry and a relaxed version of Deutsch's assumption. An analytical solution has been obtained for the conventional space-charge boundary condition of infinite charge and zero field at the surface of the charge injecting needle contact. This solution has been extended to cover a range of injection boundary conditions of both space charge and electrical field magnitudes. It has been found that, in the bulk of the material, the forms of the space charge and field distributions are relatively insensitive to the injection conditions. The distribution of power density dissipated under complete space-charge limited current flow is reported and it has been found that a maximum in the power occurs close to, but not at, the needle tip.

Several breakdown events have been observed in ultrathin SiO₂ layers when subjected to constant-voltage stress. It has been shown that the I(V) and I(t) characteristics can be understood assuming that the sample is formed by a great number of independent capacitors (spots) connected in parallel. Each breakdown event corresponds to the dielectric failure of one of these spots. The failure rate of the spots is related to the failure rate of the samples by the well known area effect of the breakdown statistical distribution. It is demonstrated that the failure rate of the samples is related to the number of broken spots and that it can be directly obtained from the evolution of the current with time. So, the statistical distribution of the first breakdown of a set of capacitors can be obtained by provoking multiple breakdown events in only one sample during a constant-voltage stress. Experimental results are presented which demonstrate that this is a very powerful alternative technique to measure experimentally the statistical distribution of breakdown in wearout tests.

In the present investigation, photon correlation spectroscopy in an extended form has been used for studying both the Rayleigh and Brillouin components of a dense fluid including their intensity ratio (Landau-Placzek ratio). As for the Brillouin component this amounts to spectroscopy in the ultrasonic region. Results for both phases of saturated n-butane in a broad temperature range are reported. They are compared with data from various sources. Although the approach to the critical temperature was limited (1.5 K for the liquid, 0.95 K for the vapour) the transition from normal to critical behaviour is distinctly marked.

Thermally stimulated luminescence glow curves were measured on CdS single crystals annealed in cadmium and sulphur vapour or air. The TSL glow curves were measured from liquid-nitrogen temperature to 500 K and nine glow peaks were observed at 110, 130, 190, 250, 280, 295, 320, 330 and 350 K with corresponding activation energies at 0.10, 0.18, 0.27, 0.32, 0.57, 0.82, 0.52, 0.78 and 0.83 eV. The activation energy and capture cross section for each of these peaks were calculated using a number of different analytical methods and compared with other data. The 190 K peak was the dominant feature observed in the crystal annealed in sulphur vapour so that the trap that lies 0.27 eV below the conduction band is associated with cadmium vacancies due to excess sulphur. For cadmium rich crystals, two characteristic traps appear at 280 and 350 K with depths of 0.57 and 0.83 eV. The 280 K peak was the dominant feature observed in the crystal annealed in cadmium vapour, associated with sulphur vacancies due to excess cadmium.

A thermal wave interference model is developed to analyse the response of coating to pulsed and stepped heating. The influence of the relative thermal properties of coating and substrate and the possibility of coating translucency are included in the analysis. The potential of transient techniques for the measurement of coating thickness is demonstrated and compared with the continuous wave technique. Experimental results obtained from plasma sprayed zirconia coatings are presented and compared with theoretical predictions.

The linear and nonlinear optical properties of high-quality single-crystal thin (3.3 mu m) MBE films of InAs on GaAs substrates are studied using the lambda =2.94 mu m output of a mode-locked Er³⁺ laser. The onset of absorption saturation occurs at pulse energy densities of approximately=100 mu J cm^-2, and a dynamic Moss-Burstein effect model gives a good quantitative explanation of the experimental data. Very rapid adsorption recovery times ( approximately=150 ps) suggest that interfacial recombination is the dominant relaxation mechanism, and indicate the potential for the use of these novel semiconductor heterostructures as fast passive optical shutters for mode-locking and Q-switching applications in the mid-infrared.

Cu_xS thin films appropriate for use as solar control coatings for architectural glazing applications have been deposited from chemical baths constituted from copper(II) nitrate or chloride, NH₃(aq), NaOH, triethanolamine and thiourea. At ambient temperature (25 degrees C), the duration of deposition ranges from 2 to 12 h, but at 50 degrees C, deposition can be considerably faster, from 1 h to 2 h 50 min. Cu_xS films deposited in this manner require air annealing at 150 degrees C for about 10 min, to reduce the integrated infrared transmittance, T*(IR), to about 10%. The corresponding integrated transmittance in the visible region, T*(vis), is about 30% and the integrated transmittance for AM2 solar spectra is about 20%. The optical transmittance spectra of the annealed films are peaked in the 0.55-0.575 mu m wavelength range, which provides a greenish yellow illumination inside the building under daylight, that corresponds to the peak in the spectral sensitivity curve of the human eye for photopic (daylight) vision. The reduction in sheet resistance of the Cu_xS films with the air annealing, from about M Omega Square Operator ^-1 to about 10-100 Omega Square Operator ^-1, ensures a low thermal emittance which is a requirement for high-efficiency solar control coatings. The issues involved in the optimization of the deposition conditions for large-area production of the coatings and the choice of protective polymer coatings are also discussed.

SnS-Cu_xS thin-film combinations deposited on glass substrates are found to offer sought-after solar control characteristics for architectural and automobile glazing applications with a choice of colours in reflection-bluish, greenish yellow, golden purple etc. The coatings were obtained by successive chemical depositions of SnS and Cu_xS thin films. The nontoxicity of SnS and Cu_xS films, their stability and the good adhesion of SnS films to glass substrates and that of Cu_xS films to polymer coatings are attractive features for the production of laminated or other types of solar control glazings for architectural and automobile windows which would be acceptable from legal, technical and aesthetic points of view.

The authors present results mainly on the variation of the work function phi _ad, with the coverage theta of La on W, on the (011), (111) and (112) planes and the (161) region using probe-hole field emission microscopy. On the (011) plane, initially, phi _ad decreases monotonically with an increase in the coverage theta , and attains the minimum at about theta =1. On the other planes, phi _ad increases initially, and then decreases. Attempts have been made to estimate the adsorbate-substrate dipole moment and the charge transfer between them on the (011) plane. The results are discussed in the light of the geometrical structure of the adsurface.

It has been demonstrated experimentally that Ga LMI sources can be modulated at relatively high frequencies using a laser beam to thermally heat the emitter tip. Using a 10 mu m laser spot size, experimental LMI source response times of 1.5 mu s were achieved. A computer simulation of the process indicates that with a 1 mu m laser spot size, response times as short as 10 ns should be possible. A study of the temperature dependence of the current suggests that the modulation is additive and that the dominant thermal effect is a change in surface tension of the liquid metal at the tip.

Stress amplitude equations are found for a generalized design of an ultrasonic fatigue dumbbell specimen which has the usual cylindrical gauge and shoulders but is of unspecified gauge fillet profile. Specimen design equations are found for the maximum stress amplitude S and minimum resonant half-length l. The known result that for the minimum specimen length condition, the stress magnification factor M= alpha , the diametric ratio of the gauge and shoulders, is shown to be independent of the gauge fillet profile. Special cases including those pertaining to the stud dumbbell, and notched specimens are also considered.

Cuticles and wax from the leaves of Citrus aurantium L. have been investigated by using NMR, DSC, X-ray diffraction and gas chromatographic methods. The wax, consisting mainly of 1-alkanols and n-alkyl esters, has chain lengths ranging from 25 to 53 carbon atoms with an average chain length of 34 carbon atoms. The X-ray results show that the wax is to a large extent ( approximately 80%) amorphous and that the crystalline fraction consists mainly of the n-alkyl esters with an average chain length of 43.5 carbon atoms. The activation energies for the different molecular motions in the wax have been extracted from the NMR spin-lattice relaxation time measurements. The activation energy associated with the reorientation of the chains in the amorphous fraction is much lower than that in other natural and Fischer-Tropsch waxes. The molecular reorientations in the cuticle membranes can best be described in terms of a model in which the correlation time has a lognormal distribution. No major difference could be detected between the DSC thermograms of the sample of cuticular membranes and the sample of matrix membranes from which the wax was extracted. Similarly, the temperature and frequency dependence of the spin-lattice relaxation times of these samples are equal. It is concluded that the presence of wax molecules does not influence the molecular dynamics of the matrix membranes noticeably.

Electrical conduction in composite materials is not isotropic and this is also the case in sedimentary rocks or other naturally occurring conductors. In dealing with such problems it is usually assumed that the medium is laterally uniform in both its electrical and magnetic properties. However, situations arise where the medium may be anisotropic in the two principal horizontal directions in addition to the vertical direction. The author presents a concise analysis of this case where an idealized half-space model is adopted to characterize the anisotropic region. The source is taken to be a vertical magnetic dipole, with time harmonic excitation, located in the adjacent isotropic half-space. It is shown that the radiation field is elliptically polarized except in the planes of symmetry of the structure.