Table of contents

A high-energy pair spectrometer has been used to determine the ratio of the radiative to mesonic capture of pions stopped in liquid hydrogen. This measurement agrees well with the value expected from the charge-exchange scattering and photoproduction of pions.

The same apparatus described in a companion paper on hydrogen has been used to study the radiative capture of pions in liquid deuterium. The result agrees well with previous measurements and the value expected from the production of pions in nucleon-nucleon collisions.

The process of electron positron pair emission in nuclear electromagnetic transitions is investigated in the special case where the angle between the directions of emission of the electron and positron is small. It is shown that experimentally this process can be of value in determining the absolute emission rates of high energy γ-rays.

In a previous paper, a variational principle was introduced for 1 - p, the capture probability for neutrons slowing down in a homogeneous medium of infinite extent. In the present paper, the variational principle is used together with simple but accurate trial functions to obtain expressions for (i) corrections to the commonly used `narrow resonance' formula for capture and (ii) interference effects in the capture of neutrons by closely spaced resonances.

The mobility, thermoelectric power and weak field Hall coefficient have been calculated when the charge carriers of a polar semiconductor are scattered by an optical mode of vibration of the lattice. The collision operator derived by Fröhlich, and Howarth and Sondheimer was used, and the Boltzmann transport equation was integrated directly on an electronic computer. Results are given for 2θ ⩾ T ⩾ θ/2 and 5 ⩾ η ⩾ -2. (η is the reduced Fermi level, θ = ν₁h/k, and ν₁ is the frequency of the longitudinal optical mode.) These results are compared where possible with those obtained from a different method of calculation by Howarth, Lewis and Sondheimer. Rough values are also given for the Lorenz number and weak field Ettingshausen coefficient.

The diffusion of phosphorus into silicon has been studied and a relationship between the surface concentration of phosphorus and the vapour pressure of the phosphorus ambient is presented. It is shown that control of the surface concentration over the range 3×10¹⁵ atoms/cm³ to 5×10¹⁸ atoms/cm³ has been achieved. Factors setting the limits to the range are discussed and a method of raising the upper limit to 10²⁰ atoms/cm³ is indicated. The results are discussed in relation to the theory presented by Smits and Miller.

Excitation functions have been measured for the (γ, n), (γ, p) and (γ, pn+γ, 2n) reactions in ⁵⁸Ni and the (γ, p) reaction in ⁶²Ni, from threshold to 32 MeV. A preponderance of protons is found from ⁵⁸Ni with the ratio integral operator₀³² σ(γ, p)dE/integral operator₀³² σ(γ, n)dE=2.35 ± 0.20.

The shapes of the (γ, n) and (γ, p) cross sections are found to be very similar, peaking near 19 MeV with widths of about 4.3 MeV. The ⁵⁸Ni (γ, pn+γ, 2n) cross section increases monotonically from threshold to 32 MeV, and the integrated cross section is about 12% of the total which is integral operator₀³² [σ(γ, n) + σ(γ, p) + σ(γ, pn + γ, 2n)] dE=(0.84 ± 0.10) MeV barns.

The ⁶²Ni (γ, p) reaction is found to have an integrated cross section of (0.13 ± 0.02) MeV barns and to peak at 22 MeV, with a width of about 5.2 MeV.

The internal resonance, revealed by measurements of the complex permeability, has been investigated in nickel over the temperature range 5°-100°c. The resonant frequency is found to be a decreasing function of temperature and, over the range studied, the variation is roughly from 450 to 300 Mc/s in the case of a polycrystalline disc, and from 270 to 40 Mc/s in the case of a colloidal suspension of nickel particles. This behaviour is consistent with the view that the resonance is due to the internal magnetic field associated with the anisotropy energy and a quantitative interpretation is attempted using the available experimental values for the anisotropy constants.

The shapes, intensities and end-points of the three strong components of the ¹⁴⁴Pr beta spectrum have been measured with a prolate spheroidal field spectrometer and coincidence techniques. A spin of 0- has been assigned to the ground state of ¹⁴⁴pr on the basis of these measurements. It is shown that the pure axial vector correction factor for a ΔI = 0 (yes) transition provides a good fit to the ground state spectrum. The spectrum of the β-transition to the 695 kev level in ¹⁴⁴Nd has a unique first forbidden shape.

The problem of electron-phonon coupling is investigated using the Born-Oppenheimer expansion. The expansion is carried out in this paper to second order in the expansion parameter (m/M)^¼. It is shown that to this order the expansion can be conveniently generated by a canonical transformation. The advantage of the canonical transformation is that a direct comparison can be made with the usual theories of the electron-phonon interaction. The frequency spectrum, and matrix elements of the electron-phonon interaction, of a monovalent metal are calculated and shown to be essentially identical with those derived by Nakajima and Bardeen and Pines. To second order in (m/M)^¼ the effects on the specific heat and spin susceptibility are calculated and found to be negligible at all temperatures.

Some aspects of the behaviour of group II, IV and VI elements as impurities in InSb, InAs, GaSb and GaAs have been studied. It is of particular interest to know whether these behave as donors or acceptors. In some cases, which are discussed, e.g. magnesium in GaAs, abnormal behaviour is observed.

An exhaustive study of sonoluminescence has been made at a frequency of 16.5 kc/s. The temperature dependence of the luminescence from 15 pure liquids and some aqueous solutions has been determined in the range 4°C to 80°C. The intensity of luminescence was correlated with various parameters; the best correlation was found with (surface tension)²/(vapour pressure). Photographs have been taken showing that sonoluminescence generally occurred as a single discrete flash once every sound cycle. This flash appeared to occur within a tenth of a period from the sound pressure minimum with non-volatile liquids like water; with volatile liquids, the flash apparently occurred shortly before or at the sound pressure maximum. Sometimes a secondary flash occurred shortly before or after the main flash. No sonoluminescence was detected from any liquid when the cavitation bubbles collapsed, i.e. shortly after the sound pressure was a maximum.

The results would appear to substantiate the theories that sonoluminescence is caused by electrical microdischarges occurring within the cavitation bubbles and by the photochemical recombination of dissociated molecules at the nascent surfaces of these bubbles.

It is suggested that an adequate explanation of the switching current in barium titanate single crystals can be given by a domain model which includes an effective wall mass and a `viscous' opposition, but no depolarizing field and no specific representation of impurities or lattice imperfections. The form of the switching current and also its variations with the magnitude of the switching field can be derived, if the domain growth is controlled by the nucleation process, and if the nucleation rate lies within certain limits.

A method is developed for determining the angular-dependent solution of the Fermi-Thomas equation for the potential within a body-centred or face-centred cubic lattice, calculations being performed for iron. By comparison with the spherically symmetric solution of the form obtained by Slater and Krutter it is possible to obtain estimates of the errors involved in the use of a spherically symmetric potential field. For iron the correction to the potential causes an increase of 0.67 ev in energy of the Fermi level, a decrease of 1.32 ev in the potential energy barrier between atoms and an increase of 3.0 ev in the total energy per atom. The total energy per atom of any metal, in the angular-dependent case, is reduced to a sum of integrals over the surface of the atomic polyhedron.

When two Fabry-Perot etalons are placed one in front of the other, and the length of one is a small multiple of that of the other, fringes are seen in white light and have been used in the past to compare the lengths of etalons. In the study reported in this paper these fringes have been observed photo-electrically while the optical length of one etalon is varied by changing the pressure of the air inside it. It is shown that the variation of intensity with path difference is the sum of two Fourier integrals involving an intensity function and the phase shifts at the reflecting surfaces. By forming cosine and sine Fourier transforms these quantities can be obtained separately and a correction applied for the error introduced into the comparison of lengths by the change of the phase shifts with wavelengths. It is shown that phase shifts can be measured more accurately by this method than by other interferometric means. It is possible to compare a metre length with a 20 cm length to about 1 in 10⁸.