Table of contents

Relativistic effects have been included in non-relativistic term energies in some recent work by relating to non-relativistic radial integrals. It is seen that the need for a non-relativistic term energy is principally one of labelling the eigenstates of the atomic system according to some single-configuration coupling scheme, which could be achieved by a coupling transformation from the basic jj-coupled schemes used in relativistic atomic structure calculations.

On the basis of the authors' previously determined highly accurate numerical wavefunctions they investigate, in quantitative terms, the effect of spin-orbit coupling on low-lying states of the hydrogen atom in magnetic fields of arbitrary strength. They stress the point that, in spite of the expected minute corrections of Coulomb binding energies, spin-orbit coupling produces narrow anticrossings of levels having identical parity and projection m_j of the total angular momentum. In this way they find the correlation diagram (between the low- and high-field quantum numbers) of the physically 'real' hydrogen atom to be quite different from that derived by Simola and Virtamo (1978) for the idealised non-relativistic hydrogen problem with infinite proton mass.

Ab initio calculations of the position and structure of the 4d⁹5s²5p² configuration of Sn II suggest that the recent interpretation of the 4d photoelectron spectra is in error.

The 4d photoelectron spectrum of tin recently measured is now believed to have been due to lead. The tin experiment was performed after the lead experiment using a completely new oven chamber, nozzle and ceramic rods but keeping the high-frequency coil, the cooling jacket and the oven support. Most probably during the authors' attempt to record the tin spectrum the nozzle of the oven chamber was blocked by tin but lead was evaporated from the cooling jacket and oven support (perhaps also from the high-frequency coil) polluted with lead.

The relativistic R-matrix method is used to calculate the total cross section for the photoionisation of mercury for photon energies in the range 10.43 eV (the ionisation potential of mercury) to 14 eV. These cross sections reveal a wealth of structure in this energy region. The asymmetry parameter, beta , for the angular distribution of the photoelectrons is also calculated.

One-electron capture into N⁴⁺(1s2ln'l') configurations, with n'=2 to 4, has been observed by electron spectrometry when a N⁵⁺(1s2s³S) multicharged ion beam encounters an He or H₂ target, at low collision velocity ( nu =0.37 au) within single-collision conditions. Contributions of other 1s2l metastable states and of the 1s² ground state may be disregarded. A small indication of two-electron capture by 1s2s³S ions (1s2s³S)3l3l' configurations is also seen.

The six dissociation and two charge exchange channels populated by single collisions of He₂⁺ molecular ions have been studied at 400 and 800 keV. The use of the grid method enabled the separation of undissociated and dissociated channels having equal energies and charges. Both He₂⁰ and He₂²⁺ molecules have been observed.

The three normalised Stokes' parameters characterising the decay of the electron impact excited d³ pi _u ( nu ₁=0, N₁=1) rotational level of H₂ were determined in an electron-photon coincidence experiment at 25 eV impact energy and 60 degrees scattering angle. The low degree of polarisation (coherence) which was observed could be caused by both the averaging over the magnetic sublevels of the ground state of the target and the fine and hyperfine relaxation process. The average angular momentum transfer perpendicular to the scattering plane from the electrons to the molecules was found to be near zero in this excitation process.

A general method is given for the calculation of products of electrostatic operators, using the formalism of coupled creation and annihilation operators. Such products appear both in inner products (on which the concept of operator orthogonality is based) and in perturbation theory (where the summations can be evaluated explicitly in the case of so-called 'weak' perturbations). A second-order calculation for the l^N configuration is given in detail. In addition, third-order calculations are discussed with stress on the simplifying cancellations that occur between various contributions; an example of a class of perturbations of interest is worked out. The present method has, when compared with previous calculations, the advantage of leading in a natural and a relatively quick way to the final equations. Moreover, the operators are directly classified according to their pure n-particle character (customarily, n is restricted to 1, 2, 3). Such a classification is indispensable in ab initio calculations of parameters.

Semi-empirical pseudopotentials including core-valence correlation have been adjusted to properties of the single-valence-electron ions X⁺ (X=Be to Ba). The pseudopotentials are tested in valence CI calculations for the ionisation potentials (IP) and dipole polarisabilities ( alpha ) of the neutral atoms. Agreement with experiment to about 0.02 eV is achieved for the IP; the computed alpha are probably more accurate than present experimental data.

Taking the H_alpha line as a test case, the authors give a proper treatment of the broadening including energy splitting and spontaneous emission effects. At low perturber densities (from 10¹⁰ to 10¹⁴ cm^-3) the profile is calculated in the line centre within the impact approximation. Each collision is well described by a semiclassical perturbational approach. The energy separation of the sublevels for n=2 and n=3 produces a splitting of the line into seven components and modifies the collisional cross sections. These components, well resolved at perturber densities lower than 10¹² cm^-3, are gradually mixed for increasing densities.

The major contribution to the polarisability of titanium which comes from the 4s orbital is examined using perturbation theory. The polarisability is expressed in terms of the second-order energy correction. The ground state and appropriate first-order correction to the wavefunction are determined variationally as a superposition of configurations. Only configurations containing 4p and 5p functions are included. A careful study of the optimum np polarising and correlating functions is made.

For each alkali metal (M) in the gas phase, the difference Delta sigma (M) between the nuclear shielding of the neutral atom and of the first positive ion is computed from non-relativistic Hartree-Fock theory. The contributions to Delta sigma (M) arising from electron correlation are derived from the Hellmann-Feynman theorem by combining Dirac-Fock and experimental ionisation potentials and are shown to be small. It is also shown that Delta sigma (M) must be positive in contrast to the reported experimental result for Delta sigma (Cs). The results of the present theory agree with experiment for Delta sigma (Rb) and two previous methods for calculating Delta sigma (M) are shown to be completely untrustworthy. The reported experimental value for Delta sigma (K) is seven times larger than the present theoretical result whilst that reported experimentally for Delta sigma (Cs) is not even positive. It is concluded that the discrepancies between theory and experiment for both Delta sigma (K) and Delta sigma (Cs) greatly exceed any errors in the present calculations, and hence that the experimental values should be re-examined.

A simple method which allows one to study the influence of all relaxation processes and the pumping rate on the velocity distribution of the two-level atom populations is reviewed. In particular, the author presents a treatment that consistently includes the effects of phase and velocity relaxation on the atomic coherences. This method is applied to describe the drift of absorbing atoms-the effect which is also named light-induced gas diffusion. The results of numerical calculations for the drift velocity are presented and the role of the atomic coherences in the description of this phenomenon is pointed out. The dependence of the drift velocity on various parameters is discussed and the best experimental conditions for observing this effect are suggested.

Cross sections and angular asymmetry parameters for the photoionisation of Ag4d¹⁰5d²D_1/2 and Cu3d¹⁰4d²D_1/2 states are calculated in the Hartree-Fock approximation with full exchange considered. Accurate single-configuration wavefunctions from MCHF are used for the initial excited states. The asymmetry parameter, beta , shows the anticipated rapid variation in the vicinity of the l to l-1 and the l to l+1 minima in the cross sections. Hence, the predicted multi-minima in the photoionisation cross sections of excited atomic d subshells can also be investigated experimentally through the measurement of beta . The authors infer that in the threshold region correlations are small and affect only the sigma _{epsilon p} partial cross section. Significant disagreement exists between the Hartree-Fock and the Hartree-Slater approximations regarding the prediction of the positions of the minima.

If solid layers of condensed Ar or Kr are bombarded with thermal Xe or Kr atoms in the metastable ³P₂ state, the mixed rare-gas excimers ArKr^*, ArXe^* and KrXe^* are formed on or within the surface. From the observed vacuum UV emissions between 110 and 190 nm, the following binding energies with respect to dissociation for these heteronuclear excimers can be derived: ArKr^* (106+or-20) meV, ArXe^* (38+or-3) meV and KrXe^* (138+or-50) meV. While in the case of KrXe^* the binding energy reported by other authors is confirmed, the existence of ArKr^* and ArXe^* is unequivocally established for the first time. From the obtained vacuum UV spectra, the potential parameters for the three heaviest rare-gas excimer systems are deduced.

The rotational intensity distribution of the N₂ photoelectron spectrum excited by the He I resonance line (584 AA) is obtained using the deconvolution technique and taking account of the regularity in the rotational transitions. It is found that the cross section sigma (J"=0 to J'=0) for the transition from the N₂(X¹ Sigma _g⁺; nu ^"=0,J") state to the N₂⁺(X² Sigma _g⁺; nu '=0,J') state is 2.66 Mb and those for sigma (J"=0 to J'=2) and sigma (J"=0 to J'=4), 3.86 and 0.28 Mb, respectively.

Transition probabilities, partial and total photoionisation cross sections for CH₄ and HF are obtained from a discrete basis set calculation by using the Stieltjes imaging technique for the continuum oscillator strength densities. Bound and continuum electronic eigenfunctions in the ground-state equilibrium geometry are described in the independent channel approximation using large basis sets of integrable functions centred on the heavy atom. Besides the usual STO functions, hydrogenic Rydberg functions and diffuse oscillatory functions are used to improve the description of the bound excited orbitals and the continuum orbitals respectively. The results in the discrete and continuous energy regions compare well with the available experimental data, confirming the validity of the approach for this class of molecules.

High-resolution energy loss spectra (ELS) for the production of N₂(B³ Pi _g) and CO(a³ Pi ) in collisions of He⁺ on N₂(X¹ Sigma _g⁺) and CO(X¹ Sigma ⁺) have been obtained in the 200 eV<or=E<or=1 keV energy range. The present study of vibronic excitation at low energies has been made possible by the use of a microchannel-plate detection device. The measured vibrational distributions show a non-Franck-Condon (FC) behavior for E<500 eV in the case of N₂. In the case of CO, deviations from FC distributions are found in the entire energy range. To explain the differences between the ELS of N₂ and CO at E>500 eV, the authors invoke in the case of CO long-range charge dipole interactions that perturb the entrance channel prior to (and/or the exit channel after) the electronic transition. For E<500 eV the deviations from FC distributions in both the N₂ and CO cases are tentatively attributed to perturbations caused by near-resonant charge exchange channels.

A simple and straightforward approach to an absolute emission cross section measurement of the He⁺ 2²P to 1²S line at 304 AA following the simultaneous ionisation-excitation of He by electron impact is presented. The undispersed He and He⁺ impact radiation in the extreme ultraviolet (XUV) region ( lambda <or=600 AA) is detected by a channel electron multiplier after passing through an XUV filter. A 304 AA emission cross section of (6.2+or-1.7)*10^-19 cm² at 200 eV impact energy is extracted utilising the known He cross sections, the measured He/He⁺ ratio and the wavelength-dependent transmission of the XUV filter and quantum yield of the photon detector. In the regime of high impact energies the emission cross section is found to be inversely proportional to the incident electron energy as expected for the optically forbidden two-electron simultaneous ionisation-excitation process.

The cross section for the ionisation process e+Ni⁺ to 2e+Ni²⁺ has been measured for electron energies from threshold up to 780 eV using the electron-ion beams technique. The total error at a 90% confidence level is estimated to be +or-5% for energies greater than 18 eV. There is evidence that the target ion beam contained a substantial fraction of ⁴F and ²F metastable ions as well as ²D ground-state ions. However, since the ions were produced in a sputter ion source, the measured cross section is directly relevant to ionisation of Ni⁺ ions sputtered from the walls of the fusion plasma device. The semi-empirical formula of Lotz (1969) grossly overestimates the cross section, whereas the scaled Born cross section of McGuire (1977) is reasonably accurate at energies above 60 eV.

The process of dielectronic recombination is studied in terms of the standard treatment of a free-bound dipole radiative transition between stationary states. The initial free state of electron-ion resonant elastic scattering is analysed using the Fano formulation for discrete-state-continuum configuration mixing. The n to infinity divergence in the energy-averaged partial dielectronic recombination cross sections, which arises in the isolated- resonance approximation, is eliminated by requiring continuity with the partial excitation cross sections at the excitation threshold. Detailed calculations for Mg⁺(3s-3p), Ca⁺(4s-4p), B²⁺(2s-2p) and C³⁺(2s-2p) are compared with experimental results. In most cases there is reasonable overall qualitative agreement, but uncertainties in the final-state distributions involved in the measurements prevent a fully quantitative comparison. The effects of external fields on the dielectronic recombination process are also discussed in the context of the present method.

The ²P_3/2-²P_1/2 fine-structure interval in hydrogenic iron and zinc has been determined in a beam-foil spectroscopy experiment. A novel technique utilising a dual-arm Johann spectrometer was used which allowed the measurements to be performed with high precision and high efficiency. The fine-structure splitting was determined to be 21.5+or-0.4 eV in hydrogenic iron and 38.2+or-0.8 eV in hydrogenic zinc compared with theoretical values of 21.2 and 37.9 eV, respectively, given by Mohr (1983). The result obtained for iron is in better agreement with theory than the previously reported measurements of Briand et al (1983-4).

It is shown that for a wide class of plasma experiments the spectral line broadening caused by ions has to be described in the impact (rather than the quasi-static) approximation. The theory of ion impact broadening (IIB), taking place against a background of electron impact broadening, is developed. It is demonstrated that, in a general case, broadening at the expense of interaction with the ion subsystem can, in a non-trivial manner, depend on parameters characterising the electron subsystem. Detailed calculations are made for IIB of the resonance doublet of hydrogen-like ions Ar XVIII, Br XXXV, Kr XXXVI, I LIII, Xe LIV, which may be added to a DT mixture for diagnostics in experiments on an inertial fusion programme. The results of calculations show that, for these purposes, below subthermonuclear and thermonuclear temperatures the Stark broadening of the red component of the resonance doublet of the indicated ions can be used.