Table of contents

Structure and oscillator strength calculations are performed for the Cr II ion in intermediate coupling. Certain important correlation configurations contribute large numbers of levels which makes the problem very demanding in computing resources. A method of selecting a 'natural' basis for the algebraic expansion states is demonstrated. This method results in very substantial reduction in the size of the problem and allows one to investigate large relativistic mixing problems that might otherwise be intractible. The method is applied to the level structure arising from mixing the 4p ⁶D^o, ⁶P^o and ⁴P^o terms in Cr II. Results are reported for oscillator strengths for transition from these levels to the levels arising from 3d⁵⁶S and 4s ⁶D terms.

A simple correlated wavefunction (with explicit dependence upon the interelectronic distance r₁₂) is presented. The use of new screened hydrogen-like orbitals allows the determination of an analytic wavefunction having an exact behaviour when the two electrons are close to the nucleus, or when an electron moves at large distances from the other one. A scale parameter q_e has been introduced in the Coulombic term q_e/r₁₂. In a variational approach, the authors have calculated average values for helium-like systems, showing an overall accuracy about 0.1% on the energies, with a three-parameter monoconfigurational wavefunction.

It is found that the Fock operator is not necessarily dilatation analytic in symmetry changing situations. This leads to an unreliability of self-consistent fields to predict symmetry (e.g., the correct point group of a molecule). The analysis shows that the non-analyticity is particularly relevant for open-shell systems. Possibilities to circumvent these difficulties are briefly mentioned. The discussion applies also to the complex rotation method.

Two-colour photodetachment of a negative ion modelled by a three-dimensional delta-function potential is theoretically investigated. The high-frequency low-intensity field is closely tuned to the detachment energy of the ion, the low-frequency high intensity field creates a ponderomotive potential of 2.6% of the detachment energy and at the same time sidebands. The two fields are circularly polarized with either equal or opposite polarizations. In accordance with the Wigner threshold law, for equal and opposite polarizations, the sideband-induced thresholds are above or below the one-photon detachment threshold, respectively. The one-photon threshold is markedly visible for opposite polarizations at the position shifted by the full ponderomotive potential. For equal polarizations and low frequencies it is not noticeable at all.

The theory of backscattering in low-energy collisions of electrons with one-electron fluorine ions has been evaluated and compared with experimental data derived from the binary-encounter peak in the spectrum of electrons ionized in 19-30 MeV collisions of fluorine ions with simple gas targets. The comparison is excellent.

The cusp electrons for electron capture to the continuum (ECC) of H⁺ ions impacting on He, Ne and Ar at 30, 50, 70 keV were measured. The double differential cross sections were analysed by Legendre polynomials to obtain the multipole moments of the cusp electron distribution. The results indicated that for k>1, the tendencies of mod B₁⁰/B₀⁰ mod and mod B₂⁰/B₀⁰ mod respectively decrease and increase smoothly with decreasing energy of ions H⁺ as in the results of Knudsen, Andersen and Jensen (1986) and with increasing mass of target as in the results of Yu and Lapicki (1987). However, the cross point of the curves for the tendencies of mod B₁⁰/B₀⁰ mod and mod B₂⁰/B₀⁰ mod is at k approximately=1.3, different from k approximately=2 for Knudsen et al. The values of mod B₂⁰/B₀⁰ mod show a dependence on the target.

Exponents of Gaussian-type basis functions have been optimized for electron-molecule scattering purposes. The criterion for optimization was to obtain the best least-squares fit to six Bessel functions j_l(k_h^(l)*r) representing the continuum functions. The values for the radial momentum k_h^(l) are defined by the boundary conditions for the Bessel functions to have vanishing radial derivatives at r=20 au. For each l=0, 1 and 2, Gaussian basis sets of eight functions have been optimized. The results are of excellent quality. It is therefore concluded that usual atomic Gaussian basis sets, augmented by these functions, can be sufficient in electron-molecule scattering calculations, such as R-matrix calculations, for example.

Positron-atom inelastic processes are formulated in terms of the random phase approximation (RPA). The theory is applied to the excitation of both the 2¹S and 2¹P levels of He. The experimental values of the literature, attributed to excitation to the 2¹S level, are shown to be in excellent agreement with theoretical values for transition to the 2¹P state.

The dynamics of the muon spin in muonium during repeated electron spin exchange with spin-1/2 paramagnetic species is investigated. The full spin states of the system at t after n spin exchange collisions at t₁, t₂, . . ., t_n are explicitly calculated in terms of the phase factor introduced at each spin exchange and muonium hyperfine transition frequencies. The expectation value of the muon spin polarization at t is then averaged over statistical time distributions of t₁, t₂, . . ., t_n for a fixed n and over all possible numbers of collisions (n) between t=0 and the observation time t. The present work treats the exchange of unlike spins (spin flip exchange) and like spins (spin non-flip exchange) on the same footing and allows one to identify clearly the physical origin of observed muon spin relaxations and of the empirical parameter introduced by Nosov and Yakovleva (1965) to account for muon spin depolarizations.

Recursion relations are derived for the non-relativistic Coulomb matrix elements of negative integer powers of r. Explicit relations are given for the bound-free elements but generalization to the bound-bound case and to the free-free case is straightforward. Using the analytically known values for two monopole matrix elements and a directly computed value for one quadrupolar matrix element, one is then able to get for each negative integer power of r the Coulomb matrix elements with all allowed values of angular momentum. The numerical efficiency of the method is checked and turns out to be excellent.

Formulae are derived for the angular distribution of fluorescence following Doppler-free two-photon absorption, in terms of the polarization and propagation vectors of the photons involved. The relationship to another, more indirect, approach is briefly discussed.

In relation to the newly discovered laser cooling mechanisms, the authors have studied the momentum diffusion of an atom with a j (ground)-j+1 (excited) state transition (j not=0), interacting with two counterpropagating sigma ₊- sigma _- plane waves. A correlation between successive photon absorptions via the multi-sublevel internal atomic structure leads to a large increase of the momentum diffusion coefficient. In a simple probabilistic picture this enhanced diffusion is interpreted as a random walk with effective steps much larger than the one-photon recoil momentum k. For j=1 the authors report the results of a complete calculation (i.e. with a quantum description of the internal atomic state) showing a marked dependence upon the velocity of the atom as well as on the parameters of the laser field (detuning and power).

The spectra emitted in the decay of the first excited states of the (RgA)⁺ ions with Rg=Ar, Kr or Xe and A=Li, Na, K, Rb or Cs have been investigated. The emission of these excimer-like ions is in the vacuum ultraviolet region and hence they are of interest for possible lasers operating in the VUV. Each ion decays by emission of 3 or 4 continua. Based on recent ab initio calculations by Markert et al. the transitions are assigned to the decay of the Omega =1 and 0⁺ levels of the (RgA)⁺ ions correlating with the asymptotic states Rg⁺(²P₃2/)+A. In the (ArA)⁺ ions the doublet expected for the (RgA)⁺ ions with a Rg⁺(²P₁2/)+A asymptotic state was also observed.

The impact broadening rates of Rb ns Rydberg levels by He, Ar and Xe atoms measured by Weber and Niemax (1982) and by Ne and Kr atoms observed by Thompson et al. (1987) are quantitatively explained over the whole range of the principal quantum number n. As Rb ns Rydberg levels (quantum defect delta =3.13) are very close to the nearly hydrogenic high l levels, inelastic collisions make as important a contribution to the broadening rates as elastic collisions. The authors derive the cross sections for the elastic and inelastic using a combination of Kaulakys' (1984) and Lebedev and Marchenko's (1985) impact theories. Good agreement between the theoretical and experimental broadening rates is achieved over the whole range of n.

Experimental differential cross sections for single and double ionization of He by impact of fast protons and deuterons are presented. The cross sections are measured at projectile scattering angles in the interval 0.27-4.2 mrad and for energies between 0.1 and 2.0 MeV amu^-1. Pure ionization events, in which the electrons that leave the He atom go into continuum states, are experimentally distinguished from ionization reactions associated with charge transfer. The results on pure single ionization by protons show how a structure in the differential cross section at 0.55 mrad, first observed by Kamber et al. (1988) at 3 and 6 MeV, gradually appears as the proton energy is increased beyond about 500 keV. Results on double ionization of He at small impact parameters are presented in the form of ratios between cross sections for double and single ionization. Using reasonable models for the dependence on impact parameter of the probabilities for pure single ionization or for single-electron capture, the authors extract energy-dependent characteristic impact-parameter ranges for the reactions.

For pt.I see ibid., vol.23, p.2309 (1990). The previous paper proposed an approach to the calculation of electron transfer reactions in ion-atomic collisions based on the Alt-Grassberger-Sandhas three-body equations for the purely Coulomb interaction. In the present paper the equations derived in I are solved numerically for concrete charge-exchange reactions. The effective potential of the problem is taken in its lowest order approximation which corresponds to the 'pole' electron transfer amplitude. The total and partial cross sections and a few other characteristics of the reactions H⁺+H to H+H⁺, H⁺+He(1s²) to H+He⁺(1s), He²⁺+H to He⁺+H⁺ and Li³⁺+H to Li²⁺+H⁺ are calculated in a wide range of collision energies from 0.1-100 keV amu^-1.

A second-order continuum distorted-wave-eikonal initial state approximation in the straight-line version of the impact parameter formalism is introduced by means of the Dodd and Greider (1966) formulation. Differential and total cross sections for K-K shell charge transfer are compared with other theoretical results as well as with experimental data. Moreover, absolute values of the real and imaginary parts of the transition amplitude are analysed.

Discusses one-quantum bremsstrahlung accompanying fast electron scattering on neutral atoms in a laser field. The linear coefficient of radiation absorption is obtained and the conditions for the latter to acquire negative value are determined. It is shown that the coefficient of radiation absorption determined by bremsstrahlung in fast electrons scattering on neutral atoms is 10²-10³ times greater than the similar coefficient corresponding to the scattering on ions.

The close-coupling approximation (CCA) has been used to investigate positronium formation into the n=1 and n=2 levels for the positron-hydrogen system. The solution has been obtained by solving integral Lippmann-Schwinger equations in momentum space. The authors have used two basis sets (H(1s,2s,2p)+Ps(1s,2s,2p) and H(1s,2s,2p)+Ps(1s,2s,2p)), which make some allowance for short- and long-range correlation effects in both the incident and positronium channels. These are the first close-coupled results for Ps(n=2) formation. The effect of adding polarization in the positronium channel is found to be significant for Ps formation. A new feature of the present calculations is the use of Gaussian basis functions in the evaluation of the rearrangement matrix elements.

The authors apply their previously-developed semi-empirical potential method in the calculation of the dominant high order many-body correlation effects in thallium which can be modelled by a central potential. The authors use this method to evaluate the thallium parity non-conserving electric dipole 6p_1/2 to 7p_1/2 and the 6p_1/2 to 6p_3/2 transitions. To estimate the accuracy of the calculation the authors also calculate the parity conserving E1 transition matrix elements and the hyperfine interaction constants. For the 6p_1/2 to 7p_1/2 parity non-conserving transition the authors obtain -7.97*10^-11 (1+or-0.07) (-iea₀Q_W/N) and for the 6p_1/2 to 6p_3/2 parity non-conserving transition -31.6*10^-11 (1+or-0.09) (-iea₀Q_W/N).