General formulations of the quantal theory of collisions between two atomic systems are available and the relevant sections are summarized in Chapter I. The results of Born can then be applied in Chapter II to the problem of electron loss from fast H passing through its parent gas at impact energies lying between 2.5 kev and 10 Mev. The sensitivity of the cross sections for which either of the resultant H atoms is excited to the 2p state, to the choice of bound and free wave functions is examined and the total loss cross section is compared with data on electron loss in H2 and H. The same theory is used in Chapter III to study the excitation and ionization of Lithium by fast electron and proton impact at incident energies up to 1 kev and 1 Mev respectively. The ionization cross sections obtained using Seaton's method (1959) yield comparison material, and the excitation calculation so far tested for sensitivity to choice of orbital parameters. In Chapter IV the rather different methods used for treating slow collisions between electrons and positive ions are employed to gain information on the diffusion and viscosity cross sections and hence on the transport coefficients, in particular the electrical and thermal conductivities. Various numerical and analytical approximations are obtained for these quantities, in the ranges 10 °K - 10 K and 10 -101 per cm3 of temperature and electron number density respectively It is shown that the electrical conductivity passes through a maximum at a temperature which is possibly in the laboratory range for high number densities. The numerical results are summarized in a series of tables.

In view of the recent developments in the theory of the possible stationary states of the helium atom, and the hearing on this theory of recent experimental work on the critical potential differences for the production of radiation and of ionisation in helium, it was thought that a spectroscopic investigation of the excitation of helium under different potential differences might he fruitful. An investigation of the minimum potential differences required for the stimulation of the different lines of the ordinary helium line spectrum, and also of the enhanced lines and the helium hand spectrum, was therefore made under various conditions of gas pressure and intensity of bombarding electron stream. In particular the possibility of exciting lines in the visible region of the spectrum, without the occurrence of ionisation, was examined. The conditions for the appearance of the band spectrum and for its maintenance, were investigated with special reference to the recent suggestion of Franck and Knipping that it originates from He2 molecules which are produced by the combination of pairs of abnormal helium atoms resulting from 20.4 volts electron impacts with normal helium. The experimental results which have been obtained are as follows: 1.(a) There are no genuine differences in the minimumvoltages required for the excitation of different individual lines in the ordinary series spectrum of helium. (b) The minimum voltages at which these lines were everexcited (as distinct from maintained) was 20.4 volts and this only with a high gas pressure and a dense bombarding electron stream. In a large number of observations the lines were notexcited until 25.2 volts (i.e. the ionisation voltage) was reached. (c) Under very special conditions the series lines weremaintained at potential differences as low as 13 - 14 volts. There seems, moreover, to be no reason for supposing that they cannot be maintained below this value. The minimum voltage for the excitation of 4686, a prominent line of the enhanced system of helium, has three differentwell defined values under different experimental conditions. Two of these values, 80 volts, and 54.2 volts correspond respectively to the voltages required theoretically (and confirmed experimentally) for the removal from the helium atom of both electrons simultaneously, and for their removal by two separate electron impacts. The third value, 50.8 volts, corresponds to the energy required by Bohr's theory for the removal of the remaining electron in an already ionised helium atom to orbit 4, the line 4686 being, on the same theory, the radiation which is emitted when the second electron in such an atom falls from orbit 4 to orbit 3.III. (a) The band spectrum of helium is not always present when the ordinary series spectrum is excited though the minimum voltage at which the band spectrum was ever excited was the same as the corresponding minimum for the ordinary series spectrum. (b) The band spectrum was never observed in the absenceof the ordinary series lines, but when these were maintained at voltages below 20.4 volts the band spectrum could be maintained below this value also. (c) The band spectrum became more intense relatively to the ordinary line spectrum as the pressure was increased, and at the same time the lowest voltage at which it could be seen was reduced. The results given under I lead to the conclusion that the occurrence of ionisation is essential for the production of the ordinary line series spectrum of helium. This conclusion is difficult to reconcile with some of the recent conclusions of Franck and Knipping, but it is in complete agreement with the results of the work of McLennan and Ireton, and of Foote and Meggers on certain metallic vapours. In contrast with this the results given under II lead to the conclusion that in the case of the helium positive ion, further ionisation is not essential for the production of the enhanced lines of helium. The results of the investigationof the conditions for the excitation of the enhanced line 4686 are in accord with Bohr's general theory of radiation. The results given under III lead to the conclusion that the hand spectrum originates from some system whose production depends upon the presence of abnormal helium atoms, and they therefore support the view that it originates from He2 molecules. The results of the present research alone afford no particular reason for concluding that the abnormal atoms concerned are those resulting from 20.4 volts electron impacts, but considered in conjunction with the results of a recent investigation by Horton and Davies, they lend strong support to this suggestion.

This thesis is in two parts, the first portion of which describes experiments which were carried out to detect a positive photo-electric emission from potassium. The result of these was that if such an emission occurs it is of the order of 10 times the negative emission at most, i.e. a thousand times as small as the value found by Dember, who claimed to have definite evidence Of a photo-electric effect. The major portion concerns the two or wore threshold frequencies of potassium. A brief historical account is given ending with the work of Richardson and Young in which they predict for the sensitised potassium surface a threshold at 10,000 AU and another at 30,000 AU. Precautions were taken to obtain a clean surface in a high vacuum, the apparatus being of a form which made the saturation of the photo-electric currents easy and the potassium was subjected to approximately black body radiation. Using Richardson's formula for the saturated currents, the work functions, b, and the corresponding threshold frequencies were determined. The values of the frequencies found were 7,100 A.U., 10,000 A.U. 21,000 A.U. A tentative explanation is given of certain Hysteresis phenomena which were observed.

The first and major part of this thesis deals with electron capture collisions, while the second part is concerned with excitation of atomic hydrogen by electron and proton impact. In Chapter 1 the formal theory of scattering is used to obtain an exact expression for the cross section for electron capture by a structure-less particle in atomic hydrogen, and the Born series of approximations is introduced via a formal expansion of the Green's operator for the system. Chapter 2 begins with a discussion of the OBK approximation; relevant experimental investigations are then briefly described, a number of Born approximation calculations are discussed and a detailed description is given of some impact parameter methods which have proved useful in this field. The chapter ends with remarks on some very recent work. The impulse approximation is derived in Chanter 3 and the evaluation of cross sections for the processes and is described in detail. Results for these processes are presented and are compared with the values obtained by other authors. The behaviour of electron capture cross sections in the high energy limit forms the subject of Chapter. Most of the standard approximations are discussed but the main emphasis is placed on the Born and impulse approximations, for which the analysis is given in detail. The application of the impulse approximation to the excitation of atomic hydrogen by electron and proton impact is discussed in Chapter 5. It is shown that the results obtained by Akerib and Borowitz for electron impact are unreliable. New results are given which are in poor agreement with experiment and reasons for the discrepancy are discussed.

In this thesis we develop and investigate the use of a semi-classical impact parameter method for treating the excitation of atoms by electron impact. Our particular interest is forbidden, non-exchange transitions. Chapter 1 contains a brief description of the method along with reasons for considering such a method. Previous work employing the same general ideas is reviewed, as are other theoretical calculations and experimental results of direct relevance to this thesis. The paper of Seaton (1962) which applies the impact parameter method to optically allowed transitions is reviewed in some detail as our treatment of the subject is based largely on this work. In Chapter 2 we derive expressions for the transition probabilities and introduce the idea of the averaged probabilities. Cross section formulae are given which are applicable to dipole and quadrupole transitions and which include a strong coupling form. The formulae are generalized to apply to an incident point particle of arbitrary mass and charge. Chapter 3 contains formulae for the averaged probabilities for arbitrary interaction potentials and these are applied to the case where the potential is a sum of spherical tensor operators. Expressions for cross sections, applicable to arbitrary transitions induced by electron or proton impact, are given and the explicit form evaluated for the case of hydrogen. In Chapter 4 are presented results of calculations of cross sections for certain forbidden transitions in hydrogen and helium, both for electron and proton impact. We examine the validity of the approximations made and discuss the choice of the cut-off. Chapter 5 contains an evaluation of the method as well as suggestions for improving it and for further applications. The derivation of certain results used in the thesis is presented in the appendices.

Several calculations have been carried out using the transition operator technique. The radiative decay in a two-level quantum system with excited state coupled by an external perturbation has been investigated and an expression obtained for P(t), the probability of finding the atom in its excited state at time t. This is seen to possess a steady value for yt >> 1 and a third term which decays at nearly half the rate of its second. Next radiative decay of an atom with two close-lying excited states is considered and |b_q|², the spectral density for spontaneous emission from the uppermost level, calculated. This shows that the proximity of the upper two levels makes the line-shape non-Lorentxian. Heisenberg equations of motion were then derived for the transitions operators of an ℓ-level atom undergoing radiative decay, the possibility of overlapping pairs of levels being ignored. These master equations were then used to obtain the spectral profiles of certain atomic lines, employing the so-far avoided Markoff approximation. First the spectral density of the scattered radiation was found for transitions between levels 5P_3/2 and 4S (ground state), and 6S and 5P_3/2, in the potassium atom when driving fields of arbitrary strength coupled levels 6S and 4P_3/2, and 4P^3/2 and 4S. Triple-peaked profiles were obtained. Spectral profiles for emission were then derived for transitions between levels 3P and (ground state), and 3P and in the hydrogen atom when a driving field of arbitrary strength coupled levels The former profile was double the latter quadruple-peaked. Lastly master equations were derived for an atom, with two excited levels undergoing radiative decay, when allowance was made for thepossibility of their overlapping. A driving field of arbitrary magnitude coupled the uppermost and ground states. The spectrum for the scattered light resulting from transitions between the uppermost and ground state was found and graphs computed for various field strengths and separations of the excited states.

The plasma produced by focusing the output of a high-powered, neodymium glass laser onto plane, metallic targets has been used as a source for studying the grazing incidence spectra of the elements in the periodic table from chromium to nickel. Over 400 new identifications are presented, the vast majority with transitions belonging to the n k n-1 k+12s 2p - 2s 2p transition arrays in the 0 I to Be I isoelectronic sequences. Of particular interest are identifications of inter combination lines in the N I to Be I sequences, which are observed for the first time in laboratory spectra of fourth period elements. The new data, taken together with other published results, allow a complete n k n-1 k+1 listing of the 2s 2p - 2s 2p transitions in the elements chromium to nickel, from which the appropriate term schemes have been determined. In addition to the n = 2 - 2 transitions, the wavelengths of some sodium-like transitions of the elements chromium to nickel are presented. The laser system used in this study has an output power of, typically, 50 GW in pulses of a few nanoseconds duration. Special features of this system are its high repetition rate and the reproducibility of its output and these features have been exploited in experiments to distinguish between the emission from different ionisation stages. A full description of the laser system is given. Consideration is, also, given to the various atomic structure calculations which have been used to give wavelength predictions of n = 2 - 2 transitions. The emphasis is placed on Hartree-Fock methods. The importance of the n = 2 - 2 transitions is evidenced by the variety of plasmas in which they occur and their uses both as a diagnostic of certain plasma parameters and for comparison with atomic structure calculations. Their occurrence in different plasmas is illustrated by the application of the term scheme analysis to tokamak and solar flare plasmas, both of which are of topical interest. Predictions of the allowed and forbidden, n = 2 - 2 transitions of metallic impurity elements which are expected to have sufficient intensity to be observed in tokamak discharges are listed and some new identifications in solar flare spectra are presented. Of particular note amongst these predictions and identifications are those of then forbidden transitions within the 2s 2p configurations, some of which fall in the near-UV and visible spectral regions, where high resolution spectroscopic techniques can be used. An example of a diagnostic use is the measurement of electron densities from line intensity ratios and this method is used to estimate the electron densities of some solar flare plasmas. In addition, the application of this method to high density plasmas, which is being investigated by other authors, is discussed. Finally, some of the observed wavelengths are compared with the results of atomic structure calculations.

The eigenfunction expansion method is used to describe the scattering of low energy electrons by atoms or ions having any number of incomplete subshells. Hartree Fock wave functions are used to describe the target system and allowance is made for the inclusion of any number of excited electronic configurations in the expansion. The continuum electron wave functions are given as the solutions of coupled integro differential equations with prescribed boundary conditions. Expressions for the photoionisation cross sections are also derived within this approximation. The formalism presented in this thesis uses the technique of angular momentum recoupling which greatly simplifies the evaluation of the potential terms which appear in the equations and the expressions for the photoionisation cross sections. This has enabled us to write, for the first time, a computer program for the calculation of the various electron scattering and photoionisation cross sections for a general atomic system, which requires as input only the Hartree Fock functions for the bound orbitals and the parameters needed to specify the terms to be included in the eigenfunction expansion. A program which calculates the necessary recoupling coefficients for any given expansion is described.

The close-coupling approximation has been used to compute cross sections for a wide range of processes (electron impact, photodetachment and photoionization). Errors in previous formulations have been pointed out and corrected. The results are compared with previous calculations and experiments and are correlated to recent work on the effects of configuration inter-action. A description of the computer code used is also given.

Two low energy positron beams have been designed, built and their performance characteristics evaluated. Both are magnetically guided systems using a combination of a solenoid and Helmholtz coils. Slow positrons are produced by the moderation of fast positrons, from a 22Na source, in annealed polycrystalline tungsten mesh with efficiencies greater than 2 x 10^-4. The original beam, built to high vacuum specifications, has been incorporated into a fully automated microcomputer controlled Doppler-broadening spectrometer system. The newer beam line is built, to UHV specifications, into a liquid helium cryostat. It also has a vertical geometry making the whole system far more versatile and allowing the study of liquid surfaces. Doppler-broadening analysis has been applied to measurements taken using both beams applied to pure metals (Mo and Ga). A two-state model incorporating diffusion of thermal positrons back to the surface has been found to be inadequate at low incident positron energies. In this regime epithermal positron and positronium emission is significant. A model of simple back-scattering of epithermal positrons was found to be successful in fitting the experimental lineshape parameters. Both fast and slow positron techniques have been applied to the study of inert gas precipitates in metals. Using conventional methods a detailed Doppler-broadening study has been carried out on the anealing of bulk Cu samples containing 3 atomic % Kr in the form of a high concentration of solid precipitates at 300K. Melting of the Kr and bubble growth are clearly seen. Deconvolution of the annihilation lineshape indicates that positrons are trapped at the Cu-Kr interface. Slow positrons have been used to profile the defect distribution of Mo implanted with a high dose of Kr ions. Using the simplest model of a step-function distribution there is reasonable agreement with the expected Kr profile.