Table of contents

It is shown that the manner in which enthalpy probes average temperature inhomogeneities gives rise to large deviations from the true mean. The dependence of the deviation on the flow régime can be used to obtain a better approximation as well as a measure of the extent of `unmixedness'.

Transverse excitation permits pulse operation of CO₂ lasers at high pressures. We have measured the gain of transversely excited lasers with linear and helical electrode arrangements with various gas mixtures at pressures up to 500 Torr and as a function of time after excitation. Gains in the range 1 to 6 dB m⁻¹ and time constants between 50 and 150 μs are observed. The results are of value in analysis of the behaviour and in the design of lasers of this type.

The theory of steady-state flow through slabs, hollow cylinders and spherical shells is formulated in terms of Fick's equation and developed for cases when the diffusion coefficient D is dependent upon positional coordinate r and concentration c, but where the functional forms of these dependencies are unknown. Graphical procedures for determining absolute values of the diffusion coefficient D, concentration profile c(r) and amount of diffusant within the membrane M_infinity are presented. Some consequences of reversing the direction of flow have also been investigated.

The effective recombination rate of a helium afterglow plasma, which is optically thick towards the resonance lines, is calculated from the coupled rate equations for the number densities of free electrons and of metastable atoms or molecules. The model employed is a neutral plasma consisting of one kind of ion and one kind of metastable. The ions are lost by electron-ion recombination only, with subsequent formation of metastables, which are then deactivated in collisions with free electrons or with other metastables: in the latter case one electron is regained to the free state. When the rate constants for these various processes are time independent, it is found that after a certain transition time a transient equilibrium between the number densities of electrons and metastables is attained. In a dense afterglow plasma, where the recombination coefficient may be large, the transient equilibrium density of metastables may become significantly higher than the quasi-equilibrium value obtained by equating the time derivative of the metastable density to zero, and the effective recombination coefficient may be reduced by much more than a factor of two.

The length of a floating segment along which a discharge can be maintained is mainly determined by the ratio of electron temperature to electric field. Metal segments several metres long appear possible for favourable values of this ratio. The effect of an axial magnetic field is to reduce the length of floating segment possible. The effects of biasing the segments positively or negatively is discussed. Some problems of discharge initiation are examined.

The equilibrium state of induction discharge plasmas in argon has been examined for a wide range of pressures (7·6 to 760 Torr) and with electron densities ranging from 3×10¹⁴ to 8×10¹⁵ cm⁻³. Grossly non-thermal conditions are obtained when the electron density falls below 10¹⁵ cm⁻³. For higher densities the approach to LTE (local thermal equilibrium) depends on the strength of the local heating fields and temperature gradients.

The conventionally used LTE criteria are found to give only order of magnitude predictions of the onset of experimental departures from LTE. Kinetic non-equilibrium arising from the heating fields is shown to be strongly affected by diffusion mixing of hot electrons over distances comparable with the plasma size. Even at atmospheric pressure diffusion-mixing distances in non-uniform fields are of the order of 2 mm.

Strong temperature gradients are also found to be important but only indirectly, and the observed departures from LTE are not adequately explained by the conventional criterion (d/T) (dT/dr)<<1. Instead ambipolar diffusion coupled with recombination appears to be a major factor. An LTE criterion based on this mechanism has been developed and fits the experimental results quite closely.

Recombination data needed to verify the ambipolar diffusion model have been obtained in a discharge with gas flow, and a virtually constant recombination time of 370 microseconds has been measured in the decaying plasma for conditions ranging from N_e=2×10¹⁵ and T_e=9000 K to 3×10¹⁴ cm⁻³ and 5500 K.

Measurements are reported for the transition probabilities of the 7635, 6416, 6032 and 4259 Å spectrum lines of neutral argon relative to the most probable value of 6·77×10⁻⁶s⁻¹ recommended by Weise. No systematic trend in the measured values was observed in the pressure range 760 to 76 Torr, or for electron densities ranging from 8×10¹⁵ to 2·5×10¹⁵ electron/cm³.

The question of collisional-radiative equilibrium is discussed in the light of measurements made at pressures down to 2 Torr, and it is shown that significant departures from LTE are to be expected only for pressures below 76 Torr.

A survey of published data on water and mercury has revealed that conventional methods of measuring compressibility are inaccurate. An analysis of the form of the discrepancies in the data has indicated that anisotropic distortion of the apparatus and air entrainment are the two principal sources of error. Means of eliminating these and other sources of error are described.

By utilizing the principles of accurate measurement outlined in this paper, the isothermal compressibility of liquids can be measured several times more accurately than has been achieved before.

The isothermal compressibility of mercury at 20 °C and 192 bar has been determined with an accuracy estimated at ±0·4%, which represents a considerable increase in accuracy over previous determinations.

The value obtained is within 0·3% of that derived from the velocity of sound by Davis and Gordon, although the accuracy claimed for the latter was only ±0·8%.

A simple model is proposed to account for temperature dependence of mobility values reported for n-hexane. A relation similar in form to that of Stokes and Walden is deduced, in which the essentially coulombic interaction between the charge carriers and host liquid is represented by an effective viscosity. The concept can be extended to yield an experimentally confirmed relationship between the charge carrier diffusion coefficient and the measured liquid viscosity.

Radio-frequency (55 MHz) breakdown characteristics of air have been studied in the presence of a parallel low-intensity magnetic field over the pressure range of 5 to 115 mTorr. The breakdown potential has been found to decrease with increase in magnetic field intensity. This decrease is most marked at the lowest pressure. It has further been observed that, although the variation of breakdown potential with pressure shows a broad minimum in the presence of lower intensity magnetic fields, this variation becomes almost linear beyond 80 G. Thus an appreciable reduction in diffusion loss of electrons under the above mentioned conditions has been observed experimentally.

A Langmuir ion gauge has been used to measure the rate of evaporation of caesium from a tri-alkali photocathode transferred from its activation cell into high vacuum. The variation with temperature has been studied over the range 22-40 °C, and a value obtained for the latent heat of evaporation. The loss of caesium was found to be relatively slow at room temperature, about 5 × 10⁴ atoms cm⁻² s⁻¹, and it is concluded that the photocathode is stable in an alkali-free environment.

The experimental method employed is thought to be inherently more reliable than that used by other workers, since practically all free caesium, which could condense on the gauge and influence the results, is excluded from the test cell.

A method is described for estimating dislocation damage in crystals as a result of spark cutting. The method allows direct comparison of dislocation arrangement in the same area before and after cutting and is applied to estimate the maximum depth of damage (mdd) in cadmium single crystals using a dislocation etch on the (101bar above 0) plane.

The investigation suggests that previous work on other metals has underestimated the mdd and in addition shows:

1. The moving-wire cutter commonly used to section metal crystals frequently makes mechanical contact with the workpiece, with resultant shorting and mechanical abrasion.

2. A simple fixed-wire cutter produces a more satisfactory and damage-free cut than the moving wire.

3. Using this fixed wire, the mdd is a function of the diameter of the wire.

Using the lowest available spark energy and a wire of 0·001 inch diameter, the maximum depth of damage in cadmium can be reduced to 0·1 mm.

Energies of up to 3 kJ g⁻¹ were deposited in thin aluminium foils by means of a pulsed electron beam with a duration of 35 ns. The subsequent expansion of the foils was observed using high-speed photography and compared with that predicted from numerical solutions of the equations of motion. The velocities observed were about double those calculated. The limitations of the expansion equation of state used in the calculations are discussed.

The variation of the spall strength of aluminium with temperature was derived from the measured energy deposition thresholds to spall foils of different thicknesses. It was found to fall from 12·5×10⁸ N m⁻² (12·5 kbar) at 350°C to about 4×10⁸ N m⁻² at 600°C with some evidence for a strength in the solid-liquid phase of a fraction of a kilobar.

Freshly cleaved (100) cleavages of KBr single crystals grown in the laboratory have been etched in an etchant containing 10 ml absolute ethyl alcohol, 0·5 ml n-butyl alcohol, 0·1 ml formaldehyde and 0·1 ml distilled water. Both clockwise and anticlockwise spiral etch pits are produced. By etching (100) matched cleavage pairs in the same conditions, one-to-one correspondence in number, position and shape but not the sense of some of the spirals has been established. By re-etching a cleavage face having spiral etch pits in dislocation etchant, an etch pit is formed exactly at the centre of each spiral, indicating thereby that the spiral etch pits nucleate at the sites of dislocations terminating on the crystal face. The implications are discussed.

A new model is proposed to explain the voltage-current characteristics of evaporated silicon oxide films in the whole range of electric field intensity, in which one sort of Poole-Frenkel effect is used and the current conduction is thought of as due to the carrier jumps over the coulomb potential wall from the occupied Poole-Frenkel sites to the empty ones. The voltage-current characteristics are calculated and compared with existing experimental data. We found good agreement between the calculated results of this model and the experimental results.

Hall coefficient and conductivity measurements have been made on undoped and oxygen-doped gallium phosphide grown by the liquid encapsulation technique over the temperature range 80 to 400 K. The electrically active impurity in undoped crystals grown from a silica crucible is silicon. The thermal ionization energy of this donor is estimated at 82 meV, which is in good agreement with the optically estimated value. The ionization energy decreases with donor concentration and, to a first approximation, is given (in eV) by E_D = 0·082 - 0·021 × 10⁻⁶N_D^{fraction one-third} where N_D is the donor concentration (cm⁻³).

Undoped crystals grown from a boron nitride crucible and oxygen-doped crystals grown from a silica crucible have similar electrical properties. The room-temperature carrier concentration in these crystals lies between 5 × 10⁹ and 2 × 10¹⁵ electrons/cm³ and the thermal ionization energies range from 0·20 to 0·55 eV. The impurities responsible for these levels have not been identified although it seems likely that oxygen is involved.

The Hall mobility and thermoelectric power of n- and p-type Cd_xPb_1−x Te alloys with x = 0·03, 0·06 and 0·11 have been measured at temperatures within the range 77 to 450 K for carrier concentrations within the range 10¹⁶ to 10²⁰ cm⁻³. The results have been interpreted using a two valence band model. The energy separation between valence bands was found to decrease with alloying and this had the important consequence that direct information could be obtained concerning heavy-mass hole conduction.

The Hall mobility, thermoelectric power, magnetoresistance and optical absorption coefficient of Mg_xPb_1−x Te alloys with 0less-than-or-eq, slantxless-than-or-eq, slant0·06 have been measured at temperatures within the range 77 to 450 K. It has been found that alloying PbTe with MgTe increases the direct energy gap and the resulting transport properties are similar to those of Cd_xPb_1−x Te alloys.

Measurements of Hall mobility, thermoelectric power and transverse Nernst-Ettingshausen coefficients have been made on n- and p-type Sn_xPb_1-xTe, Cd_xPb_1-xTe, Mg_xPb_1-xTe, Cd_xPb_1-xSe and Mg_xPb_1-xSe alloys with 0less-than-or-eq, slantxless, similar0·1. The measurements covered the temperature range 100 to 400 K and the alloys had carrier concentrations within the range 5 × 10¹⁷ to 1 × 10¹⁹ cm⁻³. To explain the results several scattering mechanisms were considered and the non-parabolicity of the energy bands was taken into account. It was found that the charge carrier scattering in all the materials under the conditions of experimentation was predominantly by acoustical phonons.

An equation to fit the magnetic hysteresis loop is derived, based on harmonic analysis ideas. This equation has the advantages of fitting the entire B-H loop and being easy to fit and handle with conventional computer library programs. It has applications in electrical machine calculations and, in particular, leads to new methods of non-destructive testing and measurement in ferromagnetic materials.

The variation of magnetization produced by the simultaneous application of magnetic fields and stresses (both compressive and tensile) has been studied for a range of iron-carbon alloys that have been subjected to various metallurgical treatments. For annealed, low carbon content materials in a small magnetic field, the magnetization-stress curves are unsymmetrical with respect to tension and compression. In the applied fields and stresses used, the asymmetry becomes less pronounced both for increase in residual internal stress and for increase in carbon content until, for cold-worked, high carbon content materials, curves are obtained that are sensibly symmetrical.

Up to now, two mechanisms have been used to explain the experimentally observed changes in magnetization, namely, that of stress-induced pressure p acting on 90° domain walls and that of large-scale changes in domain structure due to stress. It is demonstrated that these two mechanisms alone cannot account for the present observations. However, qualitative agreement is obtained over the whole range of alloys when account is taken of a third mechanism, that of stress-induced changes in the opposition term that must be in balance with p for domain wall equilibrium.

The mechanism of light reflection from diffusing materials is considered as a photon diffusion process. Equations from neutron diffusion theory are adapted to give reflection factors in terms of the scattering and absorption cross sections of the diffusing material. Measurements of the reflection spectra from blue and green ground glass are used to calculate the absorption spectra of the original glass using diffusion theory. Good agreement is obtained between the absorption spectra measured by transmission on the original glass and that calculated from the reflectance spectra of the ground glass. Comparison with the Kubelka and Munk coefficients of scatter and absorption calculated from the same data shows that diffusion theory gives a more accurate description of the absorption process. The variation of the scattering cross section and scattering coefficient of Kubelka and Munk with wavelength is similar for the ground glasses considered.

A simple procedure to determine the elements of a 3×3 matrix which describes the reflection of light by natural surfaces composed of soil, sand, etc. is described. The procedure is based on the determination of the Stokes vector of the light reflected by the sample under different conditions of illumination, either with unpolarized or linearly polarized light. It is found that, in the principal plane defined by the direction of illumination and the normal to the mean surface, the reflection matrix is analogous in form to that of a specular reflector. The individual matrix elements are, however, suitably modified to account for the lack of complete polarization of the reflected light under all conditions of illumination.

The observed diffiraction line profile from cold-worked materials is a convolution of instrumental profile and true profile resulting from the effects of lattice imperfections. In integral breadth analyses these profiles are usually described by Gaussian or Cauchy distribution functions. In order to examine the exact nature of instrumental and also true profile, several combinations of Gaussian and Cauchy forms have been considered following linewidth relations of Ruland in some filed copper- and silver-base alloys and the results have been compared with those of Fourier analysis along with integral breadth analyses where the instrumental profile takes an intermediate form. Comparing the results it is concluded that the instrumental profile is best approximated by a Gaussian or intermediate parabolic form.