Table of contents

The study of autoionising resonances of a hydrogen atom in intense magnetic fields is formulated as a complex coordinate coupled-channel problem by designating each channel by a Landau quantum number which characterises a set of Coulomb levels below its continuum threshold. For magnetic fields in the range of 4.70*10⁹ to 1.175*10¹² G, the authors have calculated the positions and widths of several autoionising resonances below the first two excited Landau thresholds.

The autoionisation width of the 2s2p ³P⁰ level in He I has been determined through measurements of the widths of three different optical transitions, namely 2s2p ³P⁰-2p3p ³P^e, 2s3p ³P⁰-'2p3p' ³D^e and 2s2p ³P⁰-'2p4p' ³D^e giving the results 7.12+or-0.12, 8.36+or-0.08 and 8.32+or-0.15 meV respectively. As a tentative explanation for the extra broadening of transitions with ³D^e states as initial states the authors suggest that ³D^e states but not ³P^e states have an unusually large fine-structure splitting.

The angular asymmetry parameter beta for Kr 4s to epsilon p photoelectrons has been studied using synchrotron radiation (hv=31-63.5 eV). The present results show that earlier calculations based on the relativistic random-phase approximation (RRPA) do not satisfactorily describe the Kr 4s photoionisation process close to the Cooper minimum. This lack of agreement has also been found recently for the Xe 5s excitation. The experimental results are compared with a new, not yet published, theoretical approach which gives good agreement with experiment.

Photoionisation of Cs 6p and Cs 7p subshells has been calculated in the Hartree-Fock approximation (HF) using accurate wavefunctions. The Cs 6p cross sections are in agreement with other theoretical results and experiment near threshold. For Cs 7p the calculation reproduces the measured values of the asymmetry parameter beta and the cross section. Good agreement is obtained with the RPA result of Cherenkov (1981).

The alignment parameter of the M₃ subshell of gold has been determined by measurement of the polarisation of X-rays following proton impact ionisation in the 0.1-0.5 MeV energy range. Measurements have shown that the experimental dependence of alignment on collision velocity exhibits an oscillation in the 0.16-0.25 relative collision velocity range.

Total differential cross sections for K⁺-Ar collisions have been measured over a wide range of laboratory scattering angles, 5 degrees <or approximately= theta <or approximately=80 degrees , and at ion energies 100-350 eV. Beyond an angle of about 45 degrees , unexpectedly high cross sections have been found. The cross sections are attributed to Ar⁺ ions produced in charge exchange reactions between K⁺ ions and Ar atoms and also through autoionisation of doubly excited Ar^** atoms formed in the K⁺-Ar collisions. Elastic and inelastic differential cross sections have been analysed by a semiclassical treatment assuming three-state potentials The result shows the inelastic collisions with high probability are caused in the region of the closest approach R_c<or approximately=1.2 AA, where the repulsive potential V(R_c)>or approximately=41 eV.

The difference in electron impact excitation cross sections of various calculations has been discussed. The results of the authors are confirmed by means of the Bethe formula.

For pt.I see ibid., vol.14, no.18, p.3305-11 (1981). The double circular orbit method (DCOM) is proposed to extend the applicability of the generator coordinate method based on coherent states (COM) which has recently been developed by the authors. The efficiency of the DCOM is examined for the Morse, double minimum and cubic anharmonic potential systems and compared with that of the COM.

The relativistic electric multipole transition operators are considered, not specifying the gauge condition of the electromagnetic field potential. The nonrelativistic limit of the relativistic expressions obtained is found. The relativistic corrections to various forms of the electric dipole transition operators are investigated. The electric multipole transition operator contains an arbitrary parameter, the gauge constant, dependence on which can serve as one of the criteria for the accuracy of the wavefunctions used. Calculated examples of the theory, starting with the general expressions found, are presented. The methods described and the formulae presented allow the authors to investigate in an effective way the electric multipole transitions for atoms and their isoelectronic sequences, including very highly ionised atoms.

Many-body perturbation theory has been applied to compute expectation values of the one- and two-electron interactions which yield the relativistic corrections (Abragam-Van Vleck terms) to the atomic g_J factors. A two-electron term due to the motion of the atomic nucleus is also considered. The present calculations are based on analytic expansions of the single-particle states, and a main purpose of the work is to investigate the capability of this method for computations of complex atomic properties. The examples considered are the ground states of the alkali atoms up to rubidium, the ground states of the first-row atoms, and chlorine. The calculations are complete to third order and some fourth-order diagrams are also included. For the first-row atoms energy lowerings ranging from 82-92% of the experimental correlation energies were obtained, showing that a considerable amount of correlation has been included. The computed g_J factors are compared with high-precision experimental values and reasonable agreements for the troublesome cases of nitrogen and the fluorine ²P_1/2 state are among the present achievements. Correlation effects were in several cases found to be of importance for the computed g_J factors.

For pt.I see ibid., vol.16, no.16, p.2891-912 (1983). Many-body perturbation theory has been applied to compute the spin-orbit coupling in the ground states of O, F, Al, Si, S and Cl, and in the first excited ²D state of N and P. The present calculations are based on analytic expansions of the single-particle states and are complete to third order with some important fourth-order diagrams included. In the cases of N, O, Si, P and Cl the present results are found to account for about 80% of the discrepancy between experimental and Hartree-Fock values. For the more troublesome cases of F, Al and S about 60% of the discrepancies are recovered. Core polarisation effects are found to yield the dominant contribution in all cases except fluorine. Correlation contributions were, however, in most cases also found to be of considerable importance. The remaining discrepancies between theoretical and experimental results seem to be mostly related to neglected high-order correlation diagrams.

The static dipole polarisabilities of Xe, Lu, Hg⁺, Hg, Tl and At have been determined from finite-field SCF calculations within a valence-electron relativistic effective potential formalism. The effect of the self-consistent inclusion of spin-orbit coupling on the calculated polarisabilities have been investigated by comparing j-averaged (L, Lambda ) results with j-dependent (J, Omega ) calculations. The atoms selected provide a range of closed-shell, particle and hole states for comparative study. The results suggest that self-consistent inclusion of spin-orbit coupling is far more important for the particle states than for the hole states. In the case of Xe, results are presented from calculations using relativistic effective potentials based on both Hamiltonian-consistent and shape-consistent pseudo-orbitals. The shape-consistent approach is found to produce polarisabilities in better agreement with previously determined all-electron Hartree-Fock values. In addition, for Hg the shape-consistent approach yields a polarisability in excellent agreement with all-electron numerical Dirac-Fock calculations.

The mean static dipole polarisability of scandium is calculated. Perturbation theory was used so that the polarisability is expressed in terms of the second-order energy correction. The ground-state and necessary first-order correction to the wavefunction were determined variationally as a superposition of configurations. The importance of the various polarising orbitals and the length of the configuration expansion are examined.

It is proved that in magnetic fields B>or approximately=10⁷ T, which are characteristic of neutron stars, the lifetime of an electron in an excited Landau level (N>or=1) bound to a proton becomes identical to that of an electron moving freely along the magnetic field lines, viz. the lifetime with respect to cyclotron transitions. The associated widths are found to be larger than or at least comparable with the Coulomb binding energies (depending on the magnetic field strength) whence the nature of these states as discrete states is obliterated. The authors show that the anomalous magnetic moment of the electron and the finite mass of the proton cause additional level shifts comparable in magnitude with the Coulomb energies. Furthermore, they investigate thermal line broadening and find that, as a consequence of the pronounced anisotropy of the effective masses for motion parallel and perpendicular to the field, thermal broadening is strongly reduced for emission perpendicular to the field.

The authors have measured photoelectron angular distributions for resonant two-photon ionisation of metastable Ne(3s ³P₂) via Ne(3p ³D₃) at 0.52 eV photoelectron energy. They derive: (i) the alignment of the Ne(3p ³D₃) state; (ii) the ratio of the reduced matrix elements for the 3p to epsilon d and 3p to epsilon s transitions and (iii) the difference of the phaseshifts in the final channels. For the absolute cross section at 0.52 eV they find (2+or-1)*10^-18 cm². At four different photoelectron energies between threshold and 0.7 eV they have studied the dependence of the total cross section on the angle between the polarising and ionising laser fields. These results are analysed to give the ratio of the above mentioned matrix elements at these energies.

The photoemission of atomic Fe, Co and Ni has been obtained in the photon energy range 50-80 eV. The strong coupling of the 3p⁶3dⁿ4s² to 3p⁵3dⁿ⁺¹4s² excitations with the 3p⁶3dⁿ4s² to 3p⁶3d^n-14s², 3p⁶3dⁿ4s ionisation of the outer 3d and 4s shells leads to a resonant enhancement of the photoemission lines above the 3p threshold. For Ni the contributions of the 3p⁶3d⁸4s to 3p⁵3d⁹4s² and 3p⁶3d⁹4s to 3p⁵3d¹⁰4s transitions can in part be separated.

The experiment demonstrates at low pressure the resonant excitation of the ³P₁ atomic state of xenon by three-photon absorption; as the pressure increases the xenon molecules in the ground state intervene in the absorption process. The emission spectra present the VUV continua characteristic of the excimer. In the same experimental conditions but with observation along the excitation axis, the frequency tripling phenomenon is seen close to the resonance line at 147 nm and the authors have measured new values of the phase matching parameter C_Xe between 145 and 147 nm.

The authors consider the absorption of energy from a laser by a molecule that has been excited into the lower quasi-continuum, with particular reference to SF₆ The theory is analytic and purely classical, following the general prescription of E.V. Shuryak. They find stochastic energy absorption, but of a different magnitude from that of Shuryak. The energy diffusion constant is independent of the applied field, in disagreement with experiment. This disagreement is unlikely to be due to quantal effects. They find that if rotation is incorporated into the original model, the picture is completely changed and there is an explicit dependence of diffusion on field strength, but no detailed analysis is available.

An expression for the energy transfer between two particles moving with relativistic velocities has been given on the basis of a binary-encounter description of elastic scattering. By means of a Lorentz transformation of the differential scattering cross section in the center-of-mass system the differential cross section and the total cross section have been established for relativistic particle velocities. Expressions have been derived that relate stopping power and straggling for a projectile penetrating a medium in internal motion to the stopping parameters of a medium at rest. These expressions contain the velocity distribution of target particles, kinematic factors and the Moller speed.

The authors present a new and powerful method for the evaluation of the relativistic ionisation form factor in momentum space. Limiting cases and generalisations are discussed.

Associative ionisation in slow collisions between Na*(np) levels and the Na(3s) ground state is studied in a crossed-beam experiment. Excited levels are prepared by laser photoexcitation, and ionic species are analysed by time-of-flight mass separation. Rate coefficients for associative ionisation, measured for principal quantum numbers ranging from n=5 to 15, exhibit a pronounced on n with a maximum of approximately 3*10^-9 cm³ s^-1 near n=11. Comparisons are made with the theoretical model developed by Duman and Shmatov, and Mihajlov and Janev.

Energy-loss spectra and different elastic and inelastic scattering functions for 540 eV single electron capture collisions of Ar²⁺ and 1250 eV Ar³⁺ ions with helium are reported and interpreted in terms of the avoided crossings of potential energy curves. Calculations are made of the proportions of long-lived excited states of these ions present in the beam, which is produced in an oscillating electron ion source.

Electron capture cross sections have been computed for Ti⁴⁺+H to Ti³⁺+H⁺ and Ti³⁺+H⁺ to Ti⁴⁺+H, where the reactants were in their ground states. The energy range investigated was 0.1 to 10 keV amu^-1. The impact parameter perturbed-stationary-state method incorporating electron translation factors were used to calculate the cross sections. The molecular wavefunctions were generated using the pseudo-potential method. At 1 keV amu^-1 ( nu _rel=4.4*10⁷ cm s^-1), representative electron capture cross sections for the Ti⁴⁺+H and Ti³⁺+H⁺ reactions are 2*10^-15 cm² and 1*10^-17 cm², respectively. An appreciable energy dependence was observed for both cross sections.

The crossed ion and electron beams technique has been used to make an absolute measurement of the cross section for the ionisation process e+Al⁺ to 2e+Al²⁺ at electron energies ranging from below threshold up to 750 eV. The compounded random and systematic errors are estimated to be about +or-5% at energies in excess of about 25 eV. The cross section for the ground-state ion is determined from the measured data using a small correction that allows for the presence of about 9% of metastable ions in the Al⁺ beam. Comparison of the present data with the predictions of Lotz shows that the latter work appreciably overestimates the ground-state cross section at energies above 35 eV. The scaled Born prediction of McGuire is in closer agreement but it also exceeds the measured cross section above 35 eV. The shape of the Coulomb-Born-exchange cross section of Moores is in fair agreement but the magnitude is lower than the present data.

Electron-excitation collision strengths have been calculated for transitions between the ten lowest levels of Ca XVII (2s²¹S₀, 2s2p ³P_0,1,2, 2s2p ¹P₁, 2p²³P_0,1,2, 2p²¹D₂, 2p²¹S₀). At high impact energies, where all the channels are open, the calculation was carried out in the LS-coupling approximation by means of the R-matrix method. Transitions between the fine structure levels were then determined by application of a unitary transformation to the LS-coupled K-matrices. At low impact energies, where some of the channels may be closed, an extension of the R-matrix method was employed to take account of relativistic effects directly in the scattering equations. In general, results are in good agreement with recent distorted-wave calculations. Electron-excitation rates are given for a range of electron temperatures.

For pt.I see ibid., vol.15, no.12, p.1929-37 (1982). An electron impact spectrometer incorporating a position-sensitive multidetector has been used to study the excited electronic states carbon monoxide in the energy region 11 to 20 eV. A series of peaks previously assigned to one vibrational progression have been resolved into two progressions, one of which has been assigned to a transition to a previously unobserved state. At high impact energy and small scattering angle the spectra observed are in good agreement with previous photoabsorption data. At high scattering angle members of Rydberg series resulting from spin-forbidden transitions have been observed in the energy region below the first ionisation potential. At higher excitation energies several new spectral features have been observed, some of which have been assigned to Rydberg series.

Cross sections for rotationally elastic and inelastic scattering of slow electrons by spherical (CH₄) and asymmetric top (H₂O) molecules are calculated in the adiabatic-nuclei-rotation (ANR) approximation. The elastic scattering amplitude is derived in the molecule-fixed frame and the corresponding K matrix is obtained by solving a set of coupled differential equations; in which the total interaction potential includes an ab initio molecular core potential from the ground state near Hartree-Fock wavefunctions, an exchange term in the free-electron-gas-exchange (FEGE) model plus the orthogonalisation procedure and the distortion of the target through a recently proposed nonparametric polarisation potential of Jain and Thompson (1982). Symmetry properties and the selection rules for the asymmetric rotor in general are discussed. In order to avoid convergence problems for the molecule with a permanent dipole moment, an alternative two-stage hybrid S-matrix theory of Collins and Norcross (1978) along with the closure formula of Crawford and Dalgarno (1971) is adopted to obtain converged cross sections (in which proper use of the unitarised Born approximation is made). Results are compared with recent available measurement and calculations; there is a qualitative agreement for CH₄ with recent calculations performed in a different model. In the case of H₂O, some difficulties are discussed in comparing directly the authors' data with recent experimental cross sections.

For pt.IV see ibid., vol.14, no.8, p.1353-671 (1981). Absolute cross sections for dissociative recombination of OH⁺, H₂O⁺, H₃O⁺ and D₃O⁺ have been measured using a merged electron-ion beam apparatus through an energy interval from 0.01 to 1 eV. The recombination cross section for the diatomic OH⁺ is small, one of the smallest measured by the authors for a diatomic ion. The cross sections for the polyatomics H₂O⁺, H₃O⁺ and D₃O⁺ are large and nearly identical. Like most other polyatomic ions they exhibit nearly an E_cm^-1 energy dependence below approximately 0.1 eV. At higher energies the slope of the cross section versus electron energy is much steeper with an energy dependence near E_cm^-2. Cross sections for D₃O⁺ are less than those for H₃O⁺ indicating the presence of a small isotope effect.