Table of contents

The ²⁹Si(p, γ)³⁰Si reaction has been studied at six resonances below 1 MeV. From analysis of the results obtained spin and parity assignments have been made to six of the low lying levels, and tentative assignments to four more levels.

It is shown that these results do not readily fit in with what is expected on the Nilsson model.

The thermal neutron diffusion parameters for light water have been measured using the pulsed neutron source technique. It has been demonstrated that the value of the diffusion cooling coefficient is critically dependent on the physical shape of the experimental system. The experimental results have been analysed using two different values of the extrapolation length, since the value of this parameter is not accurately known. The discrepancies between previously published values of the diffusion parameters are largely explained by their dependence on system shape and extrapolation length.

Experimental cross sections of the elastic scattering of 1 to 20 MeV deuterons by various nuclei have been analysed using the optical model without spin-orbit forces. Two methods of predicting parameter values have been studied; the least squares method, with some modifications, was found more satisfactory. In most cases it is possible to obtain satisfactory agreement between theory and experiment, and trends in the parameters are shown.

Alternative simplified expressions have been deduced for f and g involved in the calculation of the true potential energy curves of such electronic states of diatomic molecules whose vibrational terms cannot be adequately represented by a quadratic in v + ½. These expressions have been used in calculating the true potential curve of the B³II_0u⁺ state of ⁷⁹Br⁸¹Br molecule and the results show a close agreement with those obtained by Rees (1947). The true potential energy curve of the A¹Σ+ state of LiH molecule has also been calculated by the present method and the results are found to agree quite well with those obtained by Fallon, Vanderslice and Mason (1960).

Two simple approaches of the method of `atoms in molecules' developed by Moffitt and Hurley have been applied to the lowest ¹Σ_u⁺ state of the hydrogen molecule. A trial wave function made up from a linear combination of a covalent (1s 2pσ) and an ionic structure (1s²) was used. The effect of different procedures used for selecting orbital exponents and mixing coefficients of covalent and ionic functions was studied and, in particular, the risk of using a minimization procedure was considered.

The influence of different electron energy distribution functions, whose mean energy is obtained from swarm experiment data, and of various collision cross section variations upon the calculated mobility is investigated. It is shown that the mobility is mainly determined by the total scattering cross section and is insensitive to the form of the distribution, provided the cross section does not vary too rapidly with electron energy. Calculations are carried out assuming the Maxwellian distribution to hold and, using representations close to the experimental cross section data, good agreement between calculated and experimental mobility is obtained for some but not all of the available data.

From the measurement of breakdown potentials in an electrodeless discharge in a magnetic field the values of γ_H, Townsend's second coefficient in a magnetic field, have been calculated and the curve between γ_H and H has been found to be hyperbolic in nature, for three different values of pressure, 10 μ, 25 μ and 57.5 μ, the magnetic field varying from 0 to 60 gauss. Assuming the expression derived by Sen and Ghosh in an earlier paper,

γ = A'(E/p) + (B'/(E/p)) + C',

which is valid if E can be regarded as constant, an expression has been deduced for the variation of γ_H with magnetic field in an electrodeless discharge. The formula thus deduced can explain the observed variation of γ_H with magnetic field and the quantitative agreement is also satisfactory. It is suggested that the incorporation of the variation of γ with magnetic field together with the concept of equivalent pressure will explain the observed experimental results better.

A model to describe the sorption of gas during the ion bombardment of solid surfaces has been developed in analytical terms, assuming any general form for the ion penetration probability. It is shown how continuous measurement of the quantity of gas sorbed, up to saturation conditions, can lead to a deduction of the form of the penetration function, and how investigation of the depth distribution at saturation can also lead to the form of this function. It is emphasized that in order to obtain the exact form of the penetration function from the depth distribution, either saturation conditions must be attained or very small quantities of gas must be sorbed.

The theory is also extended to include the case where a surface is continuously replenished rather than eroded by sputtering.

Penetration probability curves are deduced from various experimental studies and are found to be of the form anticipated from the Bohr theory of penetration. Possible modifications to the simple theory are also outlined.

Observations on the growth of the current pulses of electron avalanches in nitrogen, dry air and humid air, for pd-values in the range 100 to 2300 mmHg cm, are reported. A possible explanation for the delayed rise-time of the pulses in nitrogen and dry air is suggested.

Experiments on sparking potentials in hydrogen at low pressures (10-100 mmHg) have shown that the sparking potential is not just a function of the product of gas pressure and sparking distance (Paschen's law), but decreases as the gas pressure is decreased. Values of the generalized secondary coefficient deduced from these results therefore show a pressure dependence which is approximately linear. The loss of excited molecules by collisions of the second kind proposed by Davies, Dutton and Llewellyn Jones seems inadequate to account quantitatively for these results. The loss of excited molecules by collision with the anode is suggested as an additional mechanism.

This paper discusses the stability of longitudinal electrostatic oscillations in plasmas of finite dimensions. It is shown that in general the effect of simple boundary conditions is to increase the domain of stability. Thus, a cylindrical plasma of radius a, consisting of electron and ion streams of equal temperatures and surrounded by a coaxial conducting boundary of radius b, where b >> a, is stable for all stream velocities when λ_D > 0.53 a(In b/a)^1/2, where λ_D is the Debye length. In the case of a plane-shaped plasma of semi-width a, between parallel plane-conducting boundaries at distance b, the equivalent condition is λ_D > [2(ab-a)²]^1/2. More generally, for multi-stream plasmas, in which the component streams have arbitrary temperatures and densities, it is possible to give a sufficient condition of stability for all stream velocities. The stability of three-stream plasmas is briefly discussed, and it is shown that the experimental results of Looney and Brown, and their failure to observe any excitation of plasma oscillations, are in accord with the stability criteria for plasmas of finite dimensions. Attention is drawn to the similarity between plane shaped plasmas and auroral arcs, and also the similarity between cylindrical shaped plasmas and auroral rays. It is suggested that the difference of the stability conditions for plane and cylindrical shaped plasmas may account for the tendency of auroral arcs to disrupt into rayed structures.

The energy lost from ZETA discharges, in deuterium and argon by radiation, plasma particles and non-thermal electrons, has been investigated experimentally over a wide range of conditions. Two definite regimes of energy loss have been established. These regimes are determined principally by the pressure at which the machine is operated. At high pressures (5 microns) line radiation from impurities is the dominant loss process and under certain discharge conditions accounts for all the energy fed into the discharge. Theoretical considerations are presented to show that the observed impurity content of 10¹¹ atoms per cubic centimetre can account for the radiation loss. At lower pressures (0.5 micron) the energy is lost by particles, firstly by plasma which is not efficiently confined and secondly by fast electrons which are accelerated at the termination of the current pulse and strike the tube walls with about 8 kev of energy.

The results reported are related to energy loss observations made by workers with other stabilized pinch devices.

Crystals of cadmium fluoride have been grown from the melt by the Stockbarger process up to 15 cm in diameter and 12 cm in length. Special difficulties are the tendency of the melt to evaporate and also to dissociate, and for a `crust' to form on the crystal. It is possible to choose an operating pressure (2 mmHg) such that the undissociated material stays in the crucible, while any free cadmium produced by dissociation can evaporate, giving a crystal transmitting down to about 2060 Å. Furnace details and operating conditions are described, and the effect of temperature gradient discussed.

Pure crystals of CdF₂ are found to scintillate with 7.1 nanoseconds decay time and an efficiency which is low but sufficient for the purpose of high energy spectroscopy. The decay time and the efficiency are temperature dependent; at - 178 °C the former increases to 120 nanoseconds and the latter to 14 times its room temperature value. The crystals are optically clear from 2200 Å; the scintillation emission band extends from 5000-5800 Å. Large crystals have been used to detect cosmic mu-mesons and the resolution obtained is in reasonable agreement with the expected value. The short radiation length (1.6 cm) of CdF₂, combined with the fact that large (e.g. 14 cm diameter) crystals can be grown, offers possibilities for high-energy gamma-ray spectroscopy, and the fast decay time may allow the investigation of fundamental particle reactions.

The effects of radiation damage on the scintillation efficiency of anthracene crystals are shown to be comparatively small. They cannot account for the low efficiency observed for particles of high specific ionization, even where the integrated effects of prolonged irradiation are very marked. Some evidence is found for a reduction in scintillation efficiency in impure anthracene, where emission occurs only from impurity centres, but this is a co-operative effect between excited states, and is not connected with permanent radiation damage.

Detailed examination of the scintillation pulse shape in anthracene leads to the hypothesis that the slow components (similar 1μ sec decay time) are due to short range interactions between pairs of excitons. These components are observed in the fluorescence decay due to pulsed ultra-violet excitation when the intensity is increased to produce an exciton density approaching that near the track of an ionizing particle. A number of deductions concerning the scintillation pulse shape in impure anthracene are borne out by experiment.

Electrical resistivity and thermoelectric measurements were made on ferrites of nominal compositions Ni₁Fe_1.98Mn_0.02O₄ and NiFe_1.88Mn_0.02O₄ prepared at various sintering temperatures and at oxygen pressures of up to six atmospheres. The results show that in certain composition ranges the manganese atoms suppress the easy conduction mechanism involving ions of the same element in different valence states. The detailed relationship between electrical properties and sintering pressures of specimens is found to be unexpectedly complex and suggests that hitherto unsuspected factors determine the conduction process.

The principal magnetic susceptibilities of single crystals of potassium permanganate have been measured from room temperature down to liquid oxygen temperature. The results on magnetic anisotropy have been discussed in relation to the fine structure studies of the crystal by x-ray methods. The feeble paramagnetism of the crystal has been attributed to the high frequency paramagnetism of Van Vleck. This paramagnetism has been found to vary slightly with temperature.

It is shown that the different antiferromagnetic arrangements of ordered magnetic spins in the identical lattices of α-Fe₂O₃ and Cr₂O₃ can be explained by the overlap of suitably directed oxygen p orbitals and the differently populated d orbitals of the Fe³⁺ and Cr³⁺ ions.

Percival's (1960) analysis for a flat sheet is extended to thin (compared with the skin depth) shells. Specific application is made to axially symmetric waves on a circular, cylindrical shell placed in an axial magnetic field. It is concluded that magnetically supported waves of sufficient length should be observable if the shell can be weakened in such a way as to keep the hoop stress at or near zero.

The diffusion of zinc and cadmium into selenium-doped indium antimonide has been studied. The results obtained indicate the importance of outdiffusion and the presence of a surface rate limitation. Estimates are made for the diffusion constants of the two materials.

Electron spin resonance, presumably from electrons or holes bound to appropriate vacancies, has been detected in neutron-irradiated polycrystalline BeO. A preliminary study of the behaviour of the intensity and line-shape of the electron spin resonance signal with increasing radiation dose, and on annealing, has been made. Considered as an anisotropic line, the electron spin resonance signal prior to annealing has the following g values (approximate only): g_II = 2.0036 ± 0.001, g_⊥ = 2.0110 ± 0.001.

However, the change of line-shape of the electron spin resonance signal after annealing tends to indicate that the initial signal may come from more than one type of spin centre.

By means of the Mössbauer effect, the nuclear quadrupole splitting of the first excited state of ¹¹⁹Sn has been measured in several tin compounds and a comparison made of the isomer shifts of the ground state transition. The isomer shift is interpreted in terms of a change in the mean square deformation of the nucleus <β²> arising mainly from the zero-point vibrational motion. The difference in <β²> between the excited and ground states is found to be +6 × 10^-4. From estimates of the field gradient at the ¹¹⁹Sn nucleus a value of approximately-8 × 10^-26 cm² has been assigned to the quadrupole moment of the 3/2+ state. Values of the two measured quantities are calculated using a model of a single particle coupled to a vibrator. The inclusion of pairing correlations of the nuclear force is necessary to obtain agreement with the experimental results.

Thin lens equations are derived for the correction of the two aberrations of astigmatism and distortion, resulting from the anamorphotic effect in an anamorphotic telescope, and it is shown that these two aberrations can only be corrected in such a system of three thin cylindrical lenses of the available optical glasses, when the angular magnification of the telescope in the first principal section is very close to unity.

The results are presented of an experiment carried out on the CERN synchrocyclotron to measure the polarization in free neutron-proton scattering at an effective energy of 140 ± 2 MeV. The neutron energy spectrum was roughly triangular in shape with a base extending from 100 to 190 MeV. Asymmetries were measured by using a solenoid to precess the neutrons. The neutron beam which was used had a polarization of 0.226 ± 0.010 as determined by small-angle scattering from uranium and lead targets. Measurements of asymmetries in the scattering of the neutrons by liquid hydrogen targets were made at 14 centre-of-mass angles between 20° and 160° using multiple detectors and an electronic system which reduced the systematic uncertainties normally present in the experiments. The results agree well with recent measurements by Kuckes and Wilson using a different technique. Agreement is also found with the phase shift sets YLAN3 and YLAN3M of Hull, Lassila, Ruppel, McDonald and Breit.