Table of contents

A theoretical study has been made of the x-ray and electron scattering to be expected from molecular hydrogen. The methods of calculation have the common feature that they are based on expansions of the wave function, density and potential in Legendre polynomials. Simple self-consistent field wave functions having spherical symmetry are presented and, while the binding energy thereby obtained is poor, it is found nevertheless that they may be utilized with success in both the x-ray and the electron scattering calculations. The x-ray scattering is also discussed on the basis of an lcao molecular-orbital function and the `floating' function of Gurnee and Magee. It is found that the results of the various approximations used agree well among themselves and also with experiment, and some conclusions regarding the charge distribution in the hydrogen molecule may be drawn. For the elastic scattering of slow electrons, it is found that there is good agreement between the present results and earlier work of Massey and Ridley if exchange is neglected in both cases. However, it is found that electron exchange markedly affects the scattering cross sections, thus confirming the essential conclusions of Massey and Ridley, although the method employed in the present work is probably not sufficiently reliable to enable the precise effect of exchange to be determined.

A unique result for the absorption coefficient is obtained from calculations based on measurement of the decay-time using longitudinal and radial modes of propagation. Expressions are given for the energy losses in the viscous and thermal boundary layers at the walls of the resonator.

The method is applicable in the range of frequency divided by pressure of 1 to 45 kc/s per atmosphere. Calibration measurements have been made using argon and nitrogen: the relaxational contribution to the absorption has been evaluated in the case of cyclopropane.

The information available concerning the cross sections for elastic scattering of slow positrons by helium, neon, argon and hydrogen is analysed in terms of collision theory. Evidence is found that the polarization of the neutral atom or molecule by the incident positron is important in these collisions. A variational calculation of scattering by atomic hydrogen does not support this conclusion. Further methods of investigating the problem are suggested.

From microwave resonance measurements at 35 Gc/s (λ = 0.86 cm) on a nickel single crystal, the anisotropy constants K₁ and K₂ have been determined from 20 to 150°c. At 20°c, K₁ = K₂ = - 6.06 × 10⁴ ergs cm^-3 gives the best fit with the experimental data. g-values of 2.19 ± 0.02 were obtained for the nickel single crystal and for polycrystalline nickel and nickel-copper alloys containing up to 26 atomic per cent of copper. For nickel-manganese alloys, a decrease in g-value with increase in manganese content occurred, reaching 2.12 for the 13.5 atomic per cent manganese alloy.

Angular distributions have been measured of some protons groups emitted from targets of natural magnesium and ²⁶MgO when bombarded by 8.9 MeV deuterons. These distributions have been fitted with curves calculated from the deuteron stripping theories enabling assignments of parities and possible spins to be made to the states of the residual nuclei. The reduced level widths of these states have been calculated in absolute units.

Several proton groups were observed corresponding to previously unreported levels in ²⁵Mg and ²⁷Mg.

Rotational analyses have been made of six bands of the b³Σ-a³Π system and of the 0, 0 band of the c³Σ-a³Π system of BF. Contrary to an earlier analysis, appreciable Λ-type doubling is found in the ³Π state; its magnitude and sign show that this state is regular and not inverted. No predissociations or interactions with the singlet states have been observed.

Two new bands of aluminium hydride have been found in the far ultra-violet region of the spectrum at 2173 Å and 2101 Å. The rotational analysis shows that they are the (1, 0) and (0, 1) bands of the systems C¹Σ → A¹Σ and D¹Σ → A¹Σ respectively. The (0, 1) band at 2101 Å shows perturbation as well as predissociation. The rotational constants for the (0, 1) band and the (1, 0) band are calculated as follows:

D¹Σ → A¹, (0,1) band: B_v' = 6.56 cm^-1, B_v'' = 6.16 cm^-1, D_v' = 6.1 × 10^-4 cm^-1, D_v'' = 3.3 × 10^-4 cm^-1, ν₀ = 47665.6 cm^-1

C¹Σ → A¹, (1,0) band: B_v' = 5.87 cm^-1, B_v'' = 6.33 cm^-1, B_v' = 13 × 10^-4 cm^-1, B_v'' = 3.2 × 10^-4 cm^-1, ν₀ = 45922.45 cm^-1

This paper gives a calculation of the change in population of the states of the conduction band when a magnetic filed is applied. It is based on two main suppositions: first, that the conduction band states are shifted up or down by an energy d due to the interaction of their spins with the magnetic field; secondly, that the donor states are also shifted up or down in energy by the same amount d. The more important general properties of the model are discussed, and a numerical example is worked out in some detail. It is also shown that different results are obtained according to which of two different pictures of the nature of the donor levels one adopts. The first of these pictures is the usual one, which excludes the possibility of having more than one electron on a given donor (`localized' states). The second assumes as before that the levels of either spin are sharp in energy, but postulates that these levels can be populated in any manner consistent with the exclusion principle, as in the ordinary theory of metals (`non-localized' impurity states).

The γ-rays in the decay of ¹¹⁰Ag have been observed using scintillation counters. A simplified decay scheme is presented. Spin assignments for the levels of ¹¹⁰Cd and mixing ratios of the γ-rays are obtained from the comparison of angular correlations in the double and triple cascades.

The effects of phonon-electron drag on the Nernst effect are examined using general methods described in an earlier paper. Following an account of the physical processes involved in this effect, there is a calculation of the Nernst coefficient in some simple cases. It is shown that the sign of the Nernst coefficient can be altered by a phonon current as has been found experimentally in germanium. Furthermore, a quantitative comparison of the theory with these experiments shows satisfactory agreement.

The viscosity, density and specific resistance of a number of tin based and zinc based binary alloys have been determined over various temperature ranges starting at a few degrees above the liquidus and extrapolated back to the melting point. The variation of these properties with increasing alloy additions has been found to be non-linear and minima have been observed near eutectics and limits of solid solubility.

The paper gives the results of these measurements on Zn-Al, Zn-Pb, Zn-Sn, Sn-Zn, Sn-Ag, Sn-Cu, Sn-Ni and Sn-Pb systems.

The 421 ± 4 keV proton capture resonance in ²⁴Mg has been studied employing scintillation counters. The resonance has a width of less than one kilovolt, and the relative intensities of the resulting gamma-radiations of energies 2.71, 2.26, 1.81, 1.36 and 0.89 MeV are as 1: 1.5: 0.4: 0.5: 2.2 respectively. Coincidence experiments show that the 2.71 MeV level in ²⁵Al decays to the ground, 0.453 and 1.81 MeV states. The 1.81 MeV level decays to the ground and first two excited states. The second excited state at 0.94 MeV decays in part through the emission of a 0.49-0.45 MeV γ-ray cascade.

Measurements of the angular distribution of the gamma rays establish the spin of the resonance level to be fraction three-twos⁺, and the ground state transition to be a mixture of M1 and E2 radiations. The 2.26 MeV transition to the first excited state is probably pure M1. The spin of the 1.81 MeV state is indicated to be fraction five-twos⁺, which is consistent with the results of triple angular correlation measurements of the 0.90-1.36 MeV gamma-ray cascade.

Measurements of the Hall effect, resistance, magnetoresistive effect, and thermoelectric power have been made for in all over fifty specimens prepared by sintering compressed powder blocks of cadmium oxide. For all specimens the Hall effect has been measured at room temperature, and this provides a fairly accurate estimate of the conduction electron density. The results are compared in each case with the theoretical expressions for conduction in polar crystals given by Howarth and Sondheimer (1953) and Lewis and Sondheimer (1955). Of particular interest is the variation with temperature of the Hall coefficient. Changes are observed which are not ascribed to changes of electron concentration. The complete comparison gives values of the effective electron mass m and the characteristic temperature for longitudinal optical vibrations Θ_E.

The heat of transport of helium II has been measured accurately from 1.15°k to 2.1°k in order to compare it with the calorimetric entropy data. The measurements were made with a thermally isolated glass vessel, which was connected to an isothermal liquid helium bath by a tube packed with fine powder. The heat of transport was determined (i) from the temperature and hydrostatic pressure changes in the vessel during thermomechanical flow, and (ii) from the heat input necessary to produce a measured flow rate. The results agree well with the indirect measurements of the fountain effect by van den Meydenberg, Taconis, Beenakker and Wansink and with the calorimetric data of Kramers, Wasscher and Gorter. The agreement with Kramers et al. is within the experimental error up to 1.8°k but at higher temperatures the present measurements are slightly higher.

Oscillations in the temperature of the vessel were observed even during thermomechanical flow, provided that the rate was subcritical. As expected, the oscillations were damped out above the critical rate. When the flow was just critical, a pronounced peak in the temperature-time graph was observed at the beginning of the flow, which appears to be due to an unexplained irreversible process.