Table of contents

It is pointed out that the temperature variation V_D=a-bT of the diffusion potential, as deduced from the forward characteristic of rectifiers, is in broad agreement with theory. Hence formulae are derived for the temperature variation of the zero resistance of a rectifier having an arbitrary distribution of impurity centres, and subject to either diode or diffusion theory Interpretation of experiments with the aid of these formulae leads to the temperature dependence of (a) the effective mass of current carriers in germanium rectifiers, and (b) of the mobility of current carriers in selenium rectifiers The latter curve has the same form as that obtained from Hall effect measurement, and it is pointed out that it is of broadly the same shape as the corresponding curves for high resistivity samples of silicon and germanium It is shown that two d c characteristics of the same rectifier, but corresponding to different temperatures, can intersect Formulae for the applied voltage and the temperature rise which lead to the onset of thermal instability are deduced for all impurity centre distributions, and for diode and diffusion theory A general theory of current creep in rectifiers is outlined, and good agreement obtained with experiments on selenium rectifiers

The paper describes an experimental investigation of breakdown in selenium rectifiers. Cooling from about 110°c to -74°c caused the breakdown voltage to decrease continuously, and at the same time anomalous behaviour was observed in the current-voltage characteristics When high voltages were applied, the current decreased as the temperature increased, but at low voltages rise in temperature caused the current to increase.

The effect of changing the thermal dissipation constant was similar to that caused by a change in ambient temperature Reduction in the dissipation constant, effected by placing the rectifier in a vacuum, was observed to increase the breakdown voltage, but the leakage current at low voltages also increased

None of the available theories appears to account satisfactorily for the observations.

Experimental studies have been made of the distribution in size of output pulses from a scintillation counter detecting monoenergetic alpha-particles. Separate investigations have been made of contributions to the pulse size spread by the photomultiplier and by the scintillating crystal It is shown that the main effect is due to statistical fluctuations in the number of photoelectrons produced at the photomultiplier cathode. The effect of the multiplying stages on these fluctuations is investigated and also the effect of flaws in the detecting crystal The frequency distribution of the pulses is usually found to follow the gaussian form

It is shown that for an electron immersion objective under the condition of constant cut-off voltage the emission current is uniquely defined by the cross-over potential. There is thus some theoretical foundation for the empirically observed law that the current at zero modulation voltage depends only on the three-halves power of the cut-off voltage

At wavelengths outside the photographic range the determination of the optical constants of absorbing media involves the measurement of intensity changes on reflection at the surface of the medium. A review of previous methods for their determination is made and the experimental difficulties are indicated. A method is described which surmounts some of these difficulties by avoiding the direct measurement of reflection coefficients. The accuracy to be expected from the method and an example of its use are briefly discussed

Expressions are derived, in terms of the functions normally used in optical designing, for the variation with conjugates of the primary aberrations of any symmetrical system with isotropic object and image spaces. The possibilities of invariance of the aberrations, or their simultaneous correction for more than one magnification, are discussed

An experimental study is made of the velocity and absorption of ultrasonic waves in carbon dioxide and ethylene at frequencies between ½ and 2 Mc/s and at pressures up to 100 atmospheres in a variable-path acoustic interferometer At pressures above 10 atmospheres the results are independent of frequency (in this range) in the gaseous phase. The velocity and absorption coefficient reach a minimum near the condensation point, but especially at the critical point itself have high and indeterminate values when the change of phase takes place The ultrasonic viscosities calculated from these measurements are of the order of 1000 timesthose derived from transpiration or low-frequency oscillation experiments.

The contact of electrolytes has a marked effect on the mechanical behaviour of single crystals of cadmium and zinc which have been exposed to the air This is due to the disruption, or formation, of a thin surface film, probably of oxide or hydroxide. If ordinary single crystal cadmium wires are loaded so as to flow very slowly, the contact of solutions containing free cadmium ions increases the rate of flow a similar effect exists in the case of zinc treated with solutions of zinc salts If the surface of the wire is cleaned by preliminary thermal evaporation there is no such effect, establishing that it is due to a surface film normally present. Solutions of cadmium or zinc nitrate stop the flow and raise the critical shear stress appreciably, which is attributed to the formation of a surface film, probably of hydroxide With the cadmium nitrate there is a temporary increase of flow before the cessation, which is attributed to the cadmium ion, for there is no such effect with the zinc nitrate With polycrystalline wires there is no electrolyte effect The work described supports the view that a thin oxide or hydroxide film increases the mechanical resistance of single crystal wires, and that with cadmium and zinc wires such a film is normally present.