Table of contents

The authors have calculated the relativistic autoionisation of the low-lying ⁴P/sub /J levels of Li using the state-specific theory with the Breit-Pauli Hamiltonian. The results provide a definitive explanation of the shorter lifetimes of metastable Li quartets above the Li⁺ 1s2s ³S threshold compared with the theoretical predictions from the radiative transition probabilities. This discrepancy between previous theory and experiment is explained in terms of the indirect mechanism of relativistic autoionisation, which is suddenly enhanced-as a result of symmetry, electron correlation and non-orthonormality effects-as the 1s2s ³S channel opens up. The explanation of the threshold effect is thus based on a configuration interaction mechanism and not on a low-energy electron emission effect recently proposed by Mannervik and Cederquist (1986).

For original paper see Nicolaides and Aspromallis, ibid., vol.19, p.L297 (1986). The authors discuss the recent theoretical results of Nicolaides and Aspromallis, which are in qualitative agreement with their experimental values and their interpretations of the threshold effect. From Nicolaides' and Aspromallis' limited presentation Mannervik and Cederquist feel it is still not possible to come to a general conclusion about the threshold effect observed in their previous work.

The photoion spectrum of Cs has been obtained over the wavelength 650-760 AA using synchrotron radiation as a light source. Rich structure has been found in the spectrum and identified mainly as a series converging to the ionic states 5p⁵(²P₁2/)6s and 5p⁵(²P₃2/)6s.

To calculate the triple differential cross section of dipolar (e, 2e) reactions in the vicinity of a binary peak the authors propose to use a form of Born approximation which effectively takes into account the orthogonality of the ground and ionisation atomic states if Q to 0.

The absolute coplanar asymmetric triple differential cross section data of Schlemmer et al. (1985) for the ionisation of atomic hydrogen at an incident electron energy of 150 eV and scattering angle of 16 degrees are employed to demonstrate that higher-order (n>2) terms of the scattering amplitude are quite important if the scattering angle is not too small. The calculations have been done in the second Born and the modified Glauber approximations with exchange.

Post collision interaction (PCI), usually considered as a threshold effect, is shown to persist at any non-relativistic impact energy in non-coincident particle impact experiments. As a result Auger lines obtained from these experiments are generally shifted in energy by an amount of the order of Gamma /(E_B/R)¹2/, where Gamma and E_B are the level width and binding energy, respectively, of the initial inner-shell hole. This effect has been neglected in the past; its implications are discussed.

The value of Z_eff, the effective number of electrons per molecule available to the positron for annihilation, is calculated for low-energy positron-hydrogen-molecule scattering using a scattering wavefunction containing terms in which the positron-electron distance is included linearly as a factor. The results at very low energy are much closer to the experimental value than any that have been obtained previously.

A dressed-atom analysis for third harmonic generation in Hg vapour is presented to explain the recent experiments of Normand et al. (1983).

With a canonical transformation, the Hamiltonian of a positively charged ion in a homogeneous magnetic field is transformed into the sum of three commuting operators and of a small term coupling the centre-of-mass and internal motions. The commuting operators are: (i) the corresponding Hamiltonian in the infinite-nuclear-mass approximation, (ii) a small additional Zeeman term, and (iii) a centre-of-mass operator. In the hydrogenic-ion case, the coupling term is treated as a perturbation and does not contribute to first order. Schematically, the perturbation parameter is the product of the reduced field B(2.35*10⁵ T) and the electron-to-nuclear mass ratio. The commuting operators provide an exact first-order centre-of-mass correction to the energy, valid for any field strength. Higher order corrections are calculated approximately for strong fields. Accurate quantum excesses are also determined with a simple variational basis.

Results of electron momentum spectroscopy measurements on lead vapour at a total energy of 1000 eV are presented in conjunction with a theoretical analysis using relativistic structure calculations. The 6p^-1 transition is seen to be split, resulting in transitions to the 6p₁ and 6p_3/2 ion states. The measured branching ratio for these transitions is reproduced when a multiconfiguration Dirac-Fock wavefunction is used for the target ground state. The transition involving ejection of an electron from the 6s subshell is also seen to undergo significant splitting. The splitting in this case is mainly due to final-state configuration interaction, although a multiconfiguration target state must again be used to obtain detailed agreement with experiment. The measured momentum probability distributions for the 6p and 6s electrons agree with the orbital momentum distributions given by relativistic Dirac-Fock wavefunctions, although distortion effects are evident at high momenta for the 6s^-1 transitions. The corresponding non-relativistic Hartree-Fock momentum distributions are in poorer agreement with the data. An inner valence 5d^-1 transition is also seen, but the cross section is too small for detailed comparison with theory.

Second-order perturbation calculations for the hydrogenic Zeeman effect are performed using the effective Hamiltonian of Delande and Gay (1984) and the hydrogenic basis separable in parabolic coordinates. Manifolds characterised by magnetic quantum numbers mod m mod =n-1, n-2, n-3, n-4 and an arbitrary principal quantum number n allow for a completely analytic determination of the first two Rayleigh-Schrodinger coefficients. For highly excited states. When mod m mod <or=n-5, the calculations of the coefficients include a numerical diagonalisation within the given subspace with fixed n and m. Comparisons with variational results are presented and the application of the method to calculation of zeroth-order bound-bound transition matrix elements is discussed.

Model potential curves for low-lying Sigma and II states of the Ca⁺-He and Mg⁺-He systems were calculated and used to obtain fully quantal broadening and shift parameters for the Ca⁺4P_1/2,3/2 to 4S and Mg⁺3P_1/2,3/2 to 3S spectral lines for a range of temperatures 400-3000 K. Comparison with experiment shows satisfactory agreement for the Ca⁺ lines but a 33% discrepancy for the only Mg⁺ measurement, that for the k line at 556 K. Possible experimental origins for this discrepancy are discussed. Simple semiclassical dynamics is shown to be satisfactory for the high temperature Mg⁺ and Ca⁺ widths. Rates for fine-structure changing collisions were evaluated over the same temperature range.

The optical-pumping theory of the passage of polarised radiation through a resonant medium was developed in the irreducible tensorial formalism. Transitions with arbitrary momenta J_1.2 are considered in the cubic approximation. For transitions with small values of momenta (J₁=1 to J₂=0; J₁=¹/₂ to J₂=¹/₂) the vector form of the equation system in density-matrix notation allows the problem to be solved exactly for light intensity. In the general case of an elliptically polarised wave, the equation systems for the interaction between circular components of polarisation of light in a resonant medium are obtained. Numerical calculations for deformation and rotation of the polarisation ellipse in a medium subject to saturation effects are also carried out. Irrespective of the initial state the polarisation tends either to a linear or to a circular form as it propagates, depending on times of collisional relaxation of orientation and alignment. The polarisation of linearly and circularly polarised waves does not change as the waves propagate, but the values of the saturation power for these polarisations differ considerably.

In a light-induced drift experiment the drift velocity is dependent on the change of the diffusion cross section of a particle upon optical excitation. The authors discuss an approximate method for deriving diffusion cross sections from interaction potentials. The method is used to calculate the diffusion cross section of all alkali-noble gas pairs, the alkali being in the ground state or in the first excited state. Implications for light-induced drift in alkali-noble gas mixtures are discussed.

An apparatus is described in which merged beams of ground-state Ba⁺ ions and energy-resolved radiation from the Daresbury Synchrotron Radiation Source were used to obtain absolute cross sections for the photoionisation of Ba⁺ in the wavelength range 420-775 AA (29.5-16.0 eV). Approximately 75 peaks were resolved, the largest of which corresponded to cross sections greater than 10^-15 cm²; this indicated very large contributions from inner-shell excitation-autoionisation. Systematic errors were estimated to be less than +or-15% and the energy resolution ranged between 20 and 4 meV. The energies of some of the peaks can be correlated closely with series limits of inner-shell transitions of neutral barium measured by Connerade et al. (1979) and with energies of autoionising states of Ba⁺ calculated by Hansen (1974-5) and by Griffin et al. (1984) in Hartree-Fock approximations. The energies of some transitions also correspond quite closely with absorption lines observed by Roig (1976) by flash pyrolysis of barium vapour.

The first high-resolution VUV absorption spectrum of hydrogen telluride is reported. Rydberg series of p and d types are assigned whereas the s-type Rydberg series is noticeably absent. Rotational band structures observed in the p and d series are illustrated. The spectrum of deuterium telluride in the VUV is also presented for a study of isotope effects. To the authors' knowledge the most accurate values of the first ionisation potentials for these molecules are given.

L X-ray production cross section for ⁴He ion bombardment of Ag, Cd, In, Sn, Sb, Te and I have been measured in the energy range 0.6-5.0 MeV. Subshell ionisation cross sections have been extracted and compared with the predictions of the binary encounter approximation with corrections accounting for binding and Coulomb retardation effects (BEABC), the semiclassical approximation with binding, energy loss, Coulomb deflection and relativistic corrections included (SCACBR) and the perturbed stationary state theory including energy loss. Coulomb deflection and relativistic corrections (ECPSSR). For the L₂ and L₃ subshells a good agreement is obtained with the ECPSSR predictions and, for the higher incident energies, with the BEABC ones, while the SCACBR theory overestimates the experimental data. For the L₁ subshell the agreement is worse; especially for the lower incident energies (E(⁴He)<2 MeV) the present way of calculating theoretical corrections turns out to be inadequate to reproduce the experimental findings.

For pt.I see ibid., vol.19, p.401 (1986). Absolute cross sections for electron loss to the continuum (ELC) have been obtained for 0.8-2.8 MeV H₂⁺ and He⁺ ions incident on H₂, He, Ne and Ar gases. The target consisted of a jet of gas from a multicapillary array, thereby providing a well localised and directed gas beam. A 30 degrees parallel-plate analyser having a FWHM resolution of (dE/E)=10^-3 was used to study the energy spectra of the ejected electrons as detected in a cone of solid angle 1.56*10^-3 sr about the forward direction. The cross sections were obtained by numerically integrating the ejected-electron velocity spectra over a width of 0.25 au on either side of the centre of the cusp-shaped peak. When defined in this way the cross sections are found to decrease slowly with increasing projectile velocity, in agreement with an available theoretical calculation and other experimental results.

Cross sections for K-shell ionisation for natural ⁶⁰Nd, ⁶²Sm and ⁶⁹Tm targets bombarded with 0.8-2.6 MeV protons have been determined relative to the known cross sections for Coulomb excitation of the nuclear collective states. The results are discussed with reference to those from other proton-induced ionisation experiments on Nd and Sm targets, and to the theoretical models for the ionisation process.

L-subshell ionisation cross sections were determined experimentally for Dy and Yb bombarded by 1-3 MeV protons. Thin carbon-backed targets were employed and the L X-rays emitted were measured with Si(Li) X-ray detector, the efficiency of which was measured independently. Target thicknesses were determined using the Rutherford back-scattering technique. The measured cross sections and their ratios were compared with the predictions of the PWBA and the C^EPSSR theories, with particular attention being paid to the energy dependence exhibited by the cross sections. Factors which may account for the significant discrepancies observed between experiment and theory are noted.

The technique of laser-induced fluorescence (LIF) following electron impact excitation is used to examine the neon ³P₀ and ³P₂(M/sub /J) and the helium 2¹S metastable cross sections near threshold. In each case resonance structure is observed. The branching ratio of resonance features into the two neon metastable channels is discussed.

A new approach to calculate the exact eikonal (unrestricted Glauber) direct and exchange amplitudes is proposed and applied, as a test case, to the elastic scattering of electrons and positrons by atomic hydrogen. The six-dimensional integrals of the eikonal direct and exact exchange amplitudes have been reduced to simple one-dimensional integrals, in contrast to the double integrals obtained so far. These one-dimensional integrals have been evaluated numerically using a quadrature method. Both the post and the prior forms of the exchange integrals have been computed and no post-prior discrepancy is noted for elastic scattering. Numerical results have been compared with some existing theoretical results and experimental findings.

Using a high-resolution electron impact monochromator combined with a penetrating-field analyser for efficient detection of electrons of nearly zero energy, the first high-resolution electron impact spectrum of N₂O was obtained. The threshold spectrum is presented over a wide energy range for 1.5 to 17.5 eV. Decay of a low-energy shape resonance into the vibrationally excited ground state was observed. Several new optically forbidden and previously unobserved Rydberg states were detected.

The dissociative attachment cross sections of NH₃ and ND₃ in the 5.0-6.0 eV region are found to consist of a series of overlapping peaks characteristic of the nu ₂ (inversion mode) vibrational quanta. The total scattering cross section is also structured at these energies as observed by electron transmission spectroscopy. Comparison of the results with optical absorption and electron energy loss data for excitation of the lowest Rydberg state of NH₃ suggests that fragment ions are produced through the negative-ion analogue of the predissociation process known to occur in this state. The authors assign the electronic state of the relevant temporary negative ion to the configuration ((. . .)(la"₂)¹NH₃⁺)(3sa'₁)*(3sa'₁)'²A"₂(D_3h). The relative vibrational spacings in the cation, the Rydberg state of the neutral and the anion provide evidence for the nature of the electron correlation in the latter.

The R-matrix method, which has proved successful in the study of electron-molecule collisions, is used to calculate, ab initio, eigenphases and elastic cross sections for the impact of positrons on diatomic molecules. Good agreement is obtained at several levels of approximation with previous calculation on the e⁺-H₂ problem, provided sufficient orbitals are included in the calculation to allow for short-range charge polarisation effects. Calculations on e⁺-N₂ reproduce the magnitude of the observed cross section between 0.1 and 0.5 Ryd, but are too low outside this range. This is explained at low energy by the neglect of long-range polarisation and at high energy by the absence of inelastic effects in the calculation.

The theory of electric-field-induced second-harmonic generation is developed within the framework of quantum electrodynamics. Using a recently developed exact averaging procedure to account for the anisotropy introduced to a dipolar molecular fluid by the static electric field, the polarisation dependence of the process is investigated. A number of experimental arrangements are suggested which allow a discrimination to be made between the two mechanisms involved: the third-order optical route facilitated by the field-induced anisotropy, and the fourth-order electro optical route. This allows measurement of the corresponding non-linear optical and electro-optical tensor parameters. The full temperature and field-strength dependences are presented in detail, and it is shown that in two particular polarisation cases, that of helicity conversion, and where the static field and pump polarisation are perpendicular to the polarisation of the harmonic, the usual approximation for the orientational average is incomplete and produces misleading results.