In 1927, E.P. Wigner proposed that in the event of weak spin-orbit coupling the only spectroscopic processes likely to occur were those for which the total electron spin angular momentum was conserved, namely ΔS = 0. This later became known as the Wigner Spin Conservation Rule. Its implications for radiative processes have been extensively investigated and it is now known that they adhere closely to the ΔS = 0 selection rule. In contrast, the application of the spin rule to binary collisional processes has been the subject of a small number of studies from which only tentative conclusions could be drawn as to its validity. To redress this imbalance, a systematic investigation was undertaken by the author for ion (keV)-particle (thermal) collisions, the results of which are presented in this thesis. In total, 164 reaction channels were identified of which the overwhelming majority (158) were found to be spin conservative. Furthermore, a modified version of the ΔS = 0 selection rule is proposed which states that if one collision partner undergoes a transition involving a multiplicity change a simultaneous transition in the other partner must conform to the selection rule ΔS = 0 or 1 with the proviso that singlet-singlet transitions are forbidden. During a charge stripping investigation of 4keV B^+,C^+ and Ar^+ ions, several features observed in the spectra using non-singlet target particles such as O_2 and NO, were not present when singlet target particles such as N_2 and Ne were used. To explain these discrepancies, a novel pseudo two step mechanism is proposed in which the ion undergoes a rearrangement stage prior to explusion of the electron. Metastable ions substantially influence collisional processes and to facilitate the former investigations, the fractional metastable content of six atomic ion beams (B^+, C^+, N^+, O^+, F^+ and Cl^+) formed by Electron Ionization are quantified.

This thesis describes a study of the fundamental collision processes which contribute to the spatial growth of pre-breakdown ionization current in argon in a uniform electric field, E, over a pressure range of 0.5 1 < p < 750 torr. Measurements were carried out on the growth of small photoelectric currents, Io, in purified argon over the P range of 5 <, P <, 1200 volt/cm.torr. Measurements of sparking potential as a function of pd and p were also made for argon. Following the analysis of Gosseries, measurements were carried out in order to establish the limiting value of E/p above which the electron swarm in the discharge space was no longer in a steady-state. In all the measurements of the pre-breakdown ionization current measured at constant E/p and constant p with varying d, was found to be in agreement with the Townsend growth equation, 1 0 exp(ad)1 - m/a [exp (cad) -1]. Using the above equation, accurate evaluations were made of the primary, a, and secondary, m/a, ionization coefficients. It was known from previous work that the values of ionization coefficients for rare gases were critically dependent on the state of purity of the gas, so that the ionization chamber was baked and pumped to an ultra-high vacuum base pressure ( -10-9 torr) before the measurements were carried out. The argon gas used was a standard grade argon of initial purity of 99.99%, further purified by passing the gas through a B.O.C. rare gas purifier. A mass-spectroscopic analysis of the sample used in the present work showed that impurity levels were of the order of 2-3 ppm. Analysis of the experimental data showed that over the range investigated the primary ionization coefficient is solely dependent on E/p, while the secondary ionization coefficient depends on p as well as E/P. A review of previous experimental and theoretical values of Cc/ p has been made and compared with the results obtained in the present work. The present values of a/p were lower than all the previous experimental results by at least 4% over the whole range investigated, and is attributed to the purer gas sample employed, compared with other workers. The values of the secondary ionization coefficient obtained show a minimum in their dependence on E/p. The decrease of m/a with increasing E/p at low E/p ranges up to 20 volt/cm.torr., indicated that the predominant secondary ionization process in argon was the emission of electrons at the cathode by the action of non-resonance photons produced in the gas body by three-body collision involving a metastable state and two ground-state atoms. The values of m/a in this range were found to be in good agreement, within the experimental error ( 30%), with theoretically calculated values with the assumption that the effect of non-resonance photons at the cathode was the dominant process. w/a values remained constant and independent of E/p over the range 20-200 volt/cm.torr., indicating that the predominant secondary ionization process was the action of positive ions. For E/p > 200 volt/cm.torr, W, increased as E/p increased, and this variation is attributed to the combination of the impact of positive ions and metastable states at the cathode. Sparking potential measurements made in purified argon show a deviation from Paschen's Law. At a particular value of pd, the values of Vs were higher at the lower potential. Finally, Gosseries plots were linear except for a slight upward deviation for E/p > 700 volt/cm.torr. at smaller electrode separations. This could be caused by a slight distortion due to the holes in the anode. This linearity indicates that the ionization coefficients are independent of d, and that the electron swarms are in equilibrium with the electric field over the whole discharge region.

In this thesis, the spatial and temporal growth of ionization currents is investigated for electrode systems where the applied field is distorted, due to either the proximity of the enclosing discharge chamber, or the accumulation of space-charges in the discharge region. In charters 2 to 5, a two-dimensional model for determining the D.C. breakdown voltage of an electronegative gas in a nominally uniform field electrode system is described. For three separate experimental arrangements, the computations successfully predict the occurrence of anomalous breakdowns, which have been observed at larger gap spacings. In chapters 6 and 7, a one-dimensional model for computing the temporal growth of ionization in a helium discharge at low pressure between parallel-plane electrodes is described, and the computations predict time lags which are in excellent agreement with the experimental values of D.K. Davies et al. For one set of experimental values, the numerical model is used to compute the current growth through to the steady-state glow discharge regime. By allowing for the non-equilibrium nature of the electrons in the cathode-fall region, the computations are able to predict all the essential features of the normal glow discharge in helium. In chapter 8, a one-dimensional model for simulating Trichel pulse corona in CC2 at low pressure is described, and results are presented for a discharge in a hyperboloid cathode/plane anode electrode configuration. The computations give results that are in good agreement with experiment, and the size and duration of the first Trichel pulse is successfully predicted. In particular, the accumulation of the negative ion space-charge leading to the quenching of the Trichel pulse and the subsequent interpulse region is successfully modelled.

High resolution translational energy loss single-electron capture spectra have been obtained for B<SUP>2+</SUP>, C<SUP>2+</SUP> and N<SUP>3+ </SUP> in collision with H<SUB>2</SUB>, N<SUB>2</SUB>, O<SUB>2</SUB>, CO and NO molecular targets and He, Ne, Ar, Kr and Xe atomic targets for a range (2 - 8 keV) of collision energies, with a total of 134 reaction channels identified. Splitting of the upper [<SUP>2</SUP>P<SUB>1/2</SUB><SUP>0</SUP>] and lower [<SUP>2</SUP>P<SUB>3/2</SUB><SUP>0</SUP>] <I>J</I> states of Kr<SUP>+</SUP> and Xe<SUP>+</SUP> targets was observed, with a novel statistical method for the calculation of the <I>J</I> state populations being employed to find the gradient and centre of the reaction window for these systems. Experimentally determined total and partial collision cross sections have been calculated for the N<SUP>3+</SUP> collision systems using a specially designed energy loss spectrometer with a wide angle of acceptance. When compared to various theoretically determined total cross sections the extended classical overbarrier model was found to be closest to the observed results, other models failing to take into account the large number of electron core violating reaction channels. The C<SUP>2+</SUP> collision system was found to consist of two distinct reaction windows, separated by the nature of the entrance channel. A semi-empirical model has been postulated to account for this observation. The contribution from a specific reaction channel to a single electron capture spectrum was found to be dependent upon spin conservation, position within the reaction window and electron core conservation. The contribution by collisional broadening to the ultimate resolving power attainable by a sector instrument has been studied by the use of a short path length collision cell. By precisely registering the cell on the instrumental focal plane it was found that collisional broadening could be essentially eliminated, the major resolution-limiting component at this level being thermal broadening. A theoretical model for collisional broadening has also been developed.

No description available.

No description available.

No description available.

No description available.

This thesis is concerned with two sensitive and selective methods for detecting low concentrations of molecules and reactive atoms. The methods are described in detail, and in particular emphasis is given to the design, construction and operation of systems to implement Resonance Ionization Spectroscopy (RIS) and Laser Induced Fluorescence (LIF) for detecting photodissociated alkali atoms and alkali halide molecules. Alkali halide molecules are often treated as model cases by both theoreticians and experimenters because of their unique properties which distinguish them from other molecules such as the avoided crossing of their lowest two excited potential curves with the ground state potential curve. Molecular beams of sodium iodide, lithium iodide, and caesium iodide, representative for alkali halides, were used as a rich source to study the dissociation and the detection of these molecules. In the RIS experiments, a pulsed tunable UV laser was used to photodissociate NaI, LiI and CsI. Na, Li, and Cs atoms were detected by two-step RIS. This is the first time that RIS on alkali halides has been performed in a molecular beam. The photodissociation of NaI was studied as a function of the UV-wavelength. The result has shown that the dissociation cross section for NaI at the first UV maximum (325nm) is of the same magnitude as the absorption cross section determined by Davidovits et al, 1967. A variety of different experiments were conducted to establish the sensitivity of the technique, and detection limits of 5x10<SUP>6</SUP> Na-atoms cm<SUP>-3</SUP>, 10^6 Li-atoms cm^-3 and 3x10<SUP>6</SUP> Cs-atoms cm<SUP>-3</SUP> were measured. To detect NaI molecules, a molecular beam of NaI was used to demonstrate the technique of LIF. However, no spectrally resolved molecular LIF was observed; explanations for the failure are provided, and in this context factors that affect the detection sensitivity of RIS and LIF are discussed.

No description available.