Table of contents

The electron spectrum of ²³³Pa has been studied over the energy range 4keV to 570keV. The energies of the observed conversion lines agree well with those (19keV and over) of Keller and Cork and confirm the existence of the weak 272 keV γ-ray. The low-energy spectrum includes L-conversion lines of the 28.7 keV γ-ray, and M, N and O conversions of the predicted 17.2 keV γ-ray.

The intensities of the observed lines are in good agreement with the observations (37 keV and over) of Elliott and Underhill. The main part of the continuum consists of three partial spectra with end-points at 568±5, 256±4, and 140±14 keV, and intensities of 5, 57 and 38% respectively. A decay scheme is proposed which requires that the partial spectrum of lowest energy is a close doublet.

²³³Pa yields a total of 2.0±0.15 electrons of 4 keV and over per disintegration.

The angular correlation of internally produced electron-positron pairs has been measured in the γ-transitions of several light nuclei. The measured correlation functions are compared with the theoretically predicted values and other information regarding the multipole character of the transitions.

The lack of agreement between different estimates of the cosmic radiation ionization intensity under similar conditions at sea level is pointed out, and arguments supported by experiment are advanced in favour of J. Clay's procedure for interpreting high-pressure ionization chamber data. An average specific ionization is obtained from the ionization intensity and a recently published value for the absolute particle intensity. The significance of the result is discussed and compared with the value computed from formulae for the energy loss by ionization.

A theoretical analysis of the statistical errors in cylindrical integrating ionization chambers is carried out, attention being given to the different types of ionizing particles encountered at background intensities. The conclusions are compared with experimental data and used to derive a parameter related to the specific ionization and `shower' properties of cosmic-ray particles. Independent considerations show that the value obtained for this parameter is satisfactory. Some practical suggestions are made concerning the minimization of statistical errors.

The energy of the proton beam from an air insulated Van de Graaff generator was measured absolutely by an electrostatic analyser, and the resonant energies for proton capture by ²⁴Mg and ²⁵Mg were determined by observing the positron yields. The following resonances were observed:

²⁴Mg (p, γ)²⁵Al 225.5 ± 0.3 and 418.4 ± 0.4 keV of half widths 6.0 ± 1.0 and 2.5 ± 0.5 keV respectively.

²⁵Mg (p, γ)²⁶Al 316.7 ± 0.7, 391.5 ± 0.5, 495.6 ± 0.6, 513.4 ± 0.7 and 530.4 ± 0.7 keV, of half widths 12.0 ± 1.0, 8.0 ± 1.0, 5.0 ± 1.0, 3.0 ± 1.0 and 3.0 ± 1.0 keV respectively.

The half lives of the positron decay of ²⁵Al and ²⁶Al were measured as 7.62 ± 0.13 and 6.68 ± 0.11 seconds respectively.

The spectrum of calcium was excited in a hollow cathode tube cooled with liquid hydrogen. The use of a sample of calcium enriched in the isotope 43 allowed hyperfine structures to be resolved in the are line λ6103 (4s4p ³P₀-4s5s³S₁) and the spark lines λ3933 (4s ²S_1/2-4p ²P_3/2) and λ3968 (4s ²S_1/2-4p ²P_1/2), from which an approximate value of - 1.2 nuclear moments (n.m.) for the nuclear magnetic moment was derived. The intensity ratio of the components of the are line was found to be affected by self-absorption due to the metastability of the lower term. After correction for this effect, the measured intensity ratio agreed best with a value of 7/2 for the nuclear spin. After publication of the value of the gyromagnetic ratio by Jeffries, the measurements of the width of splitting of the spark lines a lowed the nuclear spin to be determined conclusively as I = 7/2, in agreement with the predictions from the shell model and the value deduced from the strength of the nuclear resonance signal. The splitting of the ground term, 0.109±0.002 cm^-1, is 4% smaller than the value calculated from the gyromagnetic ratio by the use of the formulae of Goudsmit, Fermi, Segrè, Breit and Racah.

The variational method is used to obtain approximate wave functions for the 2pπ_u and 3dπ_g states of H₂+. Their accuracy is assessed by comparing the values of various quantities computed from them with the corresponding values computed from the exact wave functions. The dependence of the total kinetic and potential energies on the nuclear separation is briefly discussed.