Table of contents

It is pointed out that the matrix C whose elements C_nm are the van der Waals long-range force constants between atoms n and m can have no negative eigen-values. This fact can be used to give a sequence of increasingly accurate upper and/or lower bounds for the force constant between two atoms when all other elements of the force constant array are known. The simplest of these bounds, C_nm <or= (C_nnC_mm)^1/2, which relates the interaction between atoms n and m to the geometric mean of the like-atom interactions, provides bounds of a useful quality whenever the dynamic polarizability curves of atoms n and m are nearly proportional, as in the series of rare gas atoms. Some simple bounds are also derived for the van der Waals three-body forces.

A calculation of wave functions for helium and its isoelectronic series is described. Using these wave functions, calculations have been performed for the line strengths involving transitions connecting the doubly excited series of levels with the 2 ³S state of helium.

Hartree-Fock Slater integrals and semi-empirical Slater parameters for the configurations 3d⁹4s5s and 3d⁹4s4p of Cu I have been obtained and their perturbations discussed.

Absolute excitation cross sections for the emission of the first negative bands of O₂⁺ under electron impact on O₂ have been obtained. The peak cross section for the b ⁴Σ_g^- state was found to be 32·5 × 10^-18 cm² at 100 eV. The results are compared with proton and other electron impact data. Some preliminary results on the second negative system are also given.

Ionization and fragmentation by 5-45 keV He⁺ ions during passage through thin targets of H₂, N₂, O₂, CO, CO₂ and CH₄ have been investigated by the mass spectrometric analysis of the secondary ionic products of single collisions. The method ensured that all secondary product ions were collected and recorded with high and uniform efficiency. Fragmentation processes are found to play a dominant role in secondary ion production for all the cases considered in the present energy range. Below about 15 keV, it is shown that charge transfer takes place mainly through dissociative processes. Dissociative ionization is also shown to be significant, particularly for targets of H₂, N₂ and O₂, at unexpectedly low impact energies. The energy dependence of the various cross sections for the production of fragment ions is not at all in accord with the predictions of the simple adiabatic maximum rule.

A predissociation of the upper electronic state (A) of the 3680 capital A, rinf transition in CrH is suggested to explain the weakness of the system at low pressures. The relatively stronger appearance of the corresponding deuteride system indicates that the predissociation occurs by a tunnelling process.

An extension of the hydrodynamic theory of radiation-driven breakdown wave propagation is described in which ionization is treated explicitly in the conservation equations.

A cubic equation relating the instantaneous velocity to the absorbed laser flux density is obtained. This reduces to the result obtained by Ramsden and Savic in 1964 and Raizer in 1965 if ionization energies are neglected.

Calculated results from the present theory are compared with those obtained by neglecting ionization. For the régime of interest in laser-induced breakdown of gases the effects of ionization are seen to be significant.

It is shown that previous theories relating transport and ionization coefficients to fundamental collisional processes between electrons and atoms are incomplete, and can, in some cases, lead to considerable error. The influence of diffusion and injection of new electrons into the discharge through ionization are taken fully into account in the development of a more rigorous Boltzmann equation, which is then solved numerically. A total excitation cross section is found which gives good agreement between experimental values of Townsend's primary ionization coefficient and theoretically predicted values over almost the whole range of E/p₀ where experiments have been carried out. The cross section covers a range of electron energy up to 50 eV, and is about 20% lower than the experimental cross section of Maier-Leibnitz.

New Monte Carlo methods of calculation for tracing the motion of electrons in gases are described. Two problems have been treated by these methods, and the techniques applied to electron swarms in neon. Firstly, the probability of back-scatter of electrons to the cathode has been estimated for values of E/p₀ between 10 and 200 V cm^-1 torr^-1 (E is the electric field and p₀ the pressure reduced to 0 °C) and for emission energies up to the lowest excitation potential. Secondly, Townsend primary ionization coefficients α/p₀ and drift velocities have been evaluated for values of E/p₀ from 20 to 400 V cm^-1 torr^-1. It is shown that the use of the ergodic hypothesis is not valid, yielding values of α/p₀ which are up to 40% too high in neon, and that entire avalanches must be simulated before meaningful results can be obtained. The present results agree well with those obtained recently at Swansea using the Boltzmann equation, indicating that the Lorentz approximation employed in the latter method leads to only small errors in estimates of the mean properties of the electron swarms.

The equilibrium of the swarm with the field is considered in detail. It is shown that, for uniform fields, deviations from exponential growth of current with increasing electrode separation occur if the interelectrode voltage is not at least three times the mean energy of electrons impinging on the anode.

The fluorescence emission from purified argon at atmospheric pressure has been analysed spectrophotometrically by the use of calibrated optical filters. Over 90% of the total emission has been found to be contained within the continuum centred at 1250 Angstrom. By examining the effect of an applied electric field on the continuum centred at 2250 Angstrom the origin of this emission is suggested as being above the lower energy limit for dimeric ion formation at 14·7 ev. No evidence has been found for the presence of columnar dissociative recombination along the alpha-particle track.

In preparation for the derivation of correspondence identities for the Rutherford scattering problem, the relevant classical theory is developed. Convenient conventions for classical canonical transformations are introduced. Classical O(1, 3) symmetry properties are discussed. Classically accessible and inaccessible regions of momentum space are defined, and, for the latter, generalized classical paths and the corresponding action functions are obtained by analytic continuation.

A complete correspondence identity is obtained for the electron-proton system, whereby the non-relativistic quantum dynamics of the system is obtained from solutions of the corresponding classical problem and their analytic continuation given by the authors in the previous paper. The kernels of the spectral operator I_E = δ(E - H) in momentum and symmetric representation are obtained as sums over classical action functions for all non-zero real energies E. A general derivation of a scattering cross section from a spectral operator is presented, and applied to this system: the long-range distortion appears naturally. By this means and alternatively in terms of transition operators it is shown how the correct quantum-mechanical differential scattering formula follows from classical Rutherford theory. Complete correspondence identities are discussed. Quantum-mechanical barrier penetration is obtained through analytic continuation of classical action functions. A model of the system based on classical electron orbits is an improvement on the Bohr-Sommerfeld model.

The techniques of data analysis employed in molecular beam measurements of total cross sections and interaction constants are reviewed. In addition, a critical assessment is made of the experimental methods used, which has enabled a proper comparison to be made between all published experimental data and theoretically derived interaction constants.

The spectrum emitted as a result of proton impact on an argon target has been investigated, and cross sections determined for the collisionally induced emission of the 4200 Angstrom Ar I line and the 4431 Angstrom Ar II line. The cross section for the 4431 Angstrom line is 8·0 × 10^-20 cm² at 150 keV and decreases linearly with increasing energy. The 4200 Angstrom emission has a cross section of similar magnitude, but decreases less rapidly with energy.

Franck-Condon factors are presented for thirteen bands of the vprime = 0 progression of the N₂ fourth positive band system.

An electronic band system due to a hitherto unidentified diatomic emitter AlSe has been observed in thermal emission. The new spectrum, which is constituted of red-degraded bands in the spectral range λλ 3900-4610, is most probably due to the transition A ²Σ -> X ²Σ, analogous to the visible spectrum of AlS identified by McKinney.

The rotational analysis of the (0,0), (0,1) and (0,3) bands of the 2250 Angstrom system of the BiF molecule has been carried out and the molecular constants determined. The transition is found to be of the ¹II-³Σ^- type.