Table of contents

Contributions of the isotope shifts due to the second order of the hyperfine structure interaction in the subconfigurations f⁶⁷F sp and f⁵ (⁶H-⁶F) ds² of SM I have been calculated. These shifts for ^147,149Sm are smaller than 0.1*10^-3 cm^-1.

Values of the critical electric field and energy of a highly excited hydrogenic ion in a strong uniform electric field are presented using the recent classical adiabatic theory of Banks and Leopold (1978). Comparisons are made with the data available in the literature. For highly excited states the classical theory should provide a realistic approximation.

Observations of the Stark structure of barium for Rydberg states in the region of n=12 are presented. The Stark manifold includes a valence state which perturbs the ¹D₂ series. General features of the interaction of the valence and Rydberg states in the presence of the electric field are explained by simple arguments based on configuration mixing.

Experimental determinations of total cross sections for the scattering of monoenergetic positrons in the range 20-1000 eV in neon and 200-1000 eV in argon are reported. The present results in both gases for positrons in the range 200-1000 eV lie close to, but in general slightly lower than, recently published theoretical data.

Variational calculations locate and identify resonances and new threshold structures in electron impact excitation of He metastable states, in the region of the 3³S and 3¹S excitation thresholds. A virtual state is found at the 3³S threshold.

Electron excitation cross sections for six autoionising configurations of K, Rb and Cs are calculated within the Born approximation for incident electron energies up to 600 eV. Semi-empirical atomic wavefunctions are used. Values of cross sections obtained are smaller than in previous Born calculations. Cross sections for excitation of d states significantly exceed cross sections for excitation of s and p states.

The lowest electronic transitions in H₂, CH⁺ and Be have been analysed in terms of the individual contributions to excitation energies and transition moments in a second-order polarisation propagator calculation and in a second-order diagrammatic calculation. These are based on both the full particle-hole propagator matrix and on the single-transition approximation. Collective effects, defined as sums of infinite series of irreducible self-energy diagrams, do not affect the excitation energies but change the intensities by up to 40 per cent. The cancellation effects among the second-order contributions to the excitation energy are substantial, indicating the importance of using methods using methods which are consistent through a certain order in the electronic repulsion.

Bounds to expectation values of various operators for atomic wavefunctions are derived. For expectation values like ( psi mod e^{2 alpha r} psi ) they are obtained by means of differential inequalities. Furthermore, a general upper bound to the spherically averaged atomic one-electron density is given. Two special cases are considered: a simple one depending only on the kinetic energy, and an exponentially decreasing bound, in which the estimates for expectation values mentioned above are utilised.

The wings of barium absorption lines for the series 6s²¹S₀-6snp¹P₁(n=7-11) have been measured with high resolution for barium in the presence of less than an atmosphere of xenon in the temperature range 1000-1100K. Tables of normalised linewing-spectra are presented. These spectra were analysed with both the unified Franck-Condon theory and a numerical evaluation of the correlation function of semiclassical line-broadening theory. Both analyses lead to similar barium-xenon potentials for these excited states, and demonstrate that the intense regions in the line wings arise from collisions occurring when the perturbing atom has penetrated within the valence-electron charge-density maximum of the excited barium atom. Comparisons of the potentials derived from these line contours are made with available calculations.

An exact classical method is developed for the study of the critical ionisation of highly-excited hydrogenic ions by slowly varying applied electric fields, under the assumption of adiabatic invariance of the classical action integrals. Exact classical values of the critical field, the critical energy of the atom and the separation constant are determined for a wide range of values of the parabolic action integrals. Simple analytic fits accurate to within one per cent of these values are presented. For applications to real atoms, effects of tunnelling and level crossing must be included. Corrections are also required for low-l states of excited non-hydrogenic atoms and ions.

Cross sections for production of the OH(A² Sigma ⁺ to X² Pi ) fluorescence in the lambda lambda 2800-3300 AA region by photodissociation of H₂O were measured using primary photons in the lambda lambda 760-1336 AA wavelength region. The maximum cross section is 1.55 MB, at lambda 1238.3 AA and the minimum is 0.2 Mb at lambda 760.4 AA. Combining these results with data published by other authors, the yield for the OH(X² Pi )+H(²S) process is found to be larger than 72% for the presently studied primary photon wavelengths in the lambda lambda 991.6-1335.7 AA region. Fluorescence in the lambda lambda 3060-3300 AA region, produced by photons at several selected wavelengths, was compared with theoretical synthetic rotational spectra of the OH(A² Sigma ⁺, nu '=0, J' to X² Pi , nu "=0, J") band.

The eigenphase formulation of Blatt and Biedenharn (1952) is applied to fine-structure transitions in ²P atoms colliding with ¹S perturbers. Consideration is given to the limit of weak spin-orbit interaction. If the eigenphases are equal to the phaseshifts for elastic scattering by the molecular potentials then the expression for the total cross section reduces to the expression derived in the elastic approximation. However, a numerical comparison for the Li(2p²P)+He(¹S) system shows that the elastic molecular phaseshifts are not good approximations to the eigenphases. Hence the elastic approximation cannot be reliable.

An analysis is presented of the range of validity of the approximations currently used to calculate cross sections for fine-structure transitions at thermal energies. The particular example considered is the ²P_1/2-²P_3/2 transition in C⁺-H collisions. For a given set of interaction potentials of CH⁺, the full quantum-mechanical close-coupling method is used to provide a reference for comparison with a number of approximate dynamical models-semiclassical sudden approximation, quantal elastic and Omega -conserving methods. All of the dynamical models examined proved to be very satisfactory. On the other hand, the calculated cross sections are sensitive to the interaction potentials, especially for intermediate and long internuclear distances where the presently available data are not sufficiently reliable; errors in the calculated cross sections may attain 40%.

The molecular framework of Ne-Ne collisional excitation processes in the moderate energy range is reported. Diabatic-like orbitals, fulfilling the diabatic correlations of Lichten (1967) are calculated in two different approximations termed frozen-orbital and rotated-orbital methods. The coupling terms and potential energies computed in both approximations show that the excitation mechanism proposed by Brenot et al. (1975) is not consistent with the experimental findings. Quantal differential cross section calculations performed in the frozen-orbital and in the rotated-orbital approximations show the consistency of these two diabatic approaches. These calculations also show that the excitation processes are primarily induced at the crossings between the promoted 4f sigma _u molecular orbital and empty Rydberg sigma _u molecular orbitals. The investigation of secondary mechanisms following the sigma _u sigma _u primary transitions demonstrates that the one- and two-electron processes favour the Ne(2p⁵3p) levels.

The deviations from a Lorentzian shape of the high-energetic atomic line tail observed in heavy-ion collisions can be explained by perturbed initial conditions which are due to vacancy production. The present calculation for the Ni+39 MeV Ni system shows good agreement with the experimental result of the Dubna group.

Cross sections for the accidentally resonant charge-transfer process He²⁺+H(1s) to He⁺(2s)+H⁺ and total cross sections for the process He²⁺+H(1s) to He⁺( Sigma n,l)+H⁺ involving capture into all final bound states of He⁺ have been measured for 4-343 keV ³He²⁺ ions. A tungsten tube furnace was used to provide a target of highly dissociated hydrogen. The results greatly extend the energy range of previous measurements and provide data of increased accuracy thereby permitting a better assessment of the range of validity of theoretical descriptions of He²⁺-H collisions. Corresponding cross sections for capture in H₂ have also been determined. Cross sections for the direct excitation process He⁺(1s)+H to He⁺(2s)+H( Sigma ) have also been measured for the first time within the same energy range.

The total (He⁺,He) and (Ar⁺,Ar) charge-exchange cross sections have been measured in the energy range 1 keV<E_lab<10 keV, using the condenser-plate method. Special emphasis was placed on accurately controlling and measuring the target gas pressure, and this resulted in an uncertainty of less than 0.5% in the relative values of the cross section and 2.5% in their absolute values. The measured cross sections fit the expression ( sigma )¹2/=k₁-k₂ ln nu to within the experimental errors, and no oscillatory behaviour was observed. The cross section for (He⁺,He) has been calculated in the JWKB approximation using published ² Sigma _g⁺ and ² Sigma _u⁺ potential curves. The calculated and experimental values agree within 10%. The experimental data have been analysed in order to obtain the cross section for electron loss in neutral (Ar,Ar) collisions, and this cross section was found to increase monotonically from (0.39+or-10%)*10^-16 cm² at 1.5 keV to (2.47+or-10%)*10¹⁶ cm² at 10 keV.

Distorted-wave approximations for s to s excitation in electron-atom collisions are presented. In particular, it is shown that two methods due to Bransden and Winters (1975) are extensions of the well known distorted-wave first and second Born approximations (DWB1 and DWB2). The results of several distorted-wave calculations for the direct excitation of the 2s state of hydrogen by 100 and 300 eV electrons are used. These include the DWB2 approximation, for which a straightforward method of calculation is proposed, and its extension by Bransden and Winters. The results demonstrate the importance of distortion in both initial and final states. It is also found that higher-order corrections to the distorted waves in the DWB1 and approximation have a small effect compared with that arising from the distorted-wave second Born term.

In a crossed-beam experiment absolute cross sections for ionisation by electron impact are measured for the processes Mg+e to Mg⁺+2e and Mg+e to Mg²⁺+3e. The Mg⁺ and Mg²⁺ ions are separated by a time-of-flight method. Electron energies are varied between 30 eV and 280 eV. Cross sections at the maximum of the ionisation functions are 7.8*10^-16 cm² (Mg⁺) and 9*10^-18 cm² (Mg²⁺).

Measurements as functions of incident energy within the range 6 to 16 eV have been made of the yields of both metastables and ultraviolet photons following electron impact on carbon monoxide and nitrogen molecules. The energies of resonances in the excitation functions have been obtained from comparison with features of known energies in the noble-gas atoms. The configurations and binding energies of the resonances have been considered in terms of a model in which two Rydberg electrons couple together strongly in the field of the positive ion core.