Table of contents

In response to the strong current theoretical interest in models of low-energy ionization in H⁺–H collisions, a crossed-beam coincidence counting technique previously used in this laboratory to obtain cross sections in the range 9–1500 keV amu⁻¹ has been successfully adapted to extend the measurements downwards in energy from 9 to 1.25 keV amu⁻¹. In addition, we have extended the triple-centre calculations of McLaughlin B M, Winter T G and McCann J F 1997 J. Phys. B: At. Mol. Opt. Phys. 30 1043, down to 0.5 keV amu⁻¹ where agreement with experiment is found to be satisfactory. Our new measurements and calculation are at variance with relative cross sections measured by Pieksma M, Ovchinnikov S Y, van Eck J, Westerveld W B and Niehaus A 1994 Phys. Rev. Lett. 73 46 in the range 1–6 keV amu⁻¹, but are in good accord with their calculations based on the hidden-crossings model including only S and T promotion. However, our measurements and calculation do not support the more recent calculations by Pieksma M, Ovchinnikov S Y and Macek J H 1998 J. Phys. B: At. Mol. Opt. Phys. 31 1267 in which a contribution from a radial decoupling mechanism is also included in the model.

We have performed an ab initio calculation of the hyperfine, P-odd and P,T-odd constants for the YbF molecule using a recently developed technique which allows us to take into account correlations and polarization in the outer-core region. The ground state electronic wavefunction of the YbF molecule is found with the help of the relativistic effective core potential method followed by the restoration of molecular four-component spinors in the core region of ytterbium within the framework of a non-variational procedure. Core polarization effects are included with the help of atomic many-body perturbation theory for the Yb atom. Our calculation of the isotropic hyperfine constant A agrees at the 3% level with the experimental value. The dipole constant (which is much smaller in magnitude), though better than in all previous calculations, is still underestimated by almost 23%. This error is reduced to 8% after a semiempirical correction for perturbation of the -shell in the core of Yb due to the chemical bonding made. Our value for the effective electric field on the unpaired electron is .

The extensive arc emission measurements of transition probabilities by Corliss and Bozman are known to contain errors due to flaws in the determination of level populations in the source. However, if no similar errors were present in the photometric calibration, then branching fractions from the same upper level deduced from these measurements should be valid. It is shown that the branching fractions for the transitions in Si I, Ge I, Sn I and Pb I can be accurately estimated using intermediate coupling amplitudes obtained from spectroscopic data, and thus provide a test of the validity of these measurements. The results of Corliss and Bozman are examined in the context of comparisons with these estimated values and with other measurements, and it is demonstrated that branching fractions from the same upper level obtained from these data can be quite reliable.

We use a simple potential well model to study the giant resonances in Ce, Sm, Eu and Gd. We show that photoionization of a d electron in a spherical potential well into a short-lived quasibound f state produces the observed asymmetric ion yield curves. In our approach a suitable normalization condition is applied to the final state wavefunction inside the well and it is assumed that the f electron escapes within a very short time from the well. We further show that our three-parameter model allows the resonance profiles to be parametrized according to the Fano profile of the photoionization cross section.

We report values for the single-photoionization cross sections to the ³S and ³P^o channels and for the total double photoionization cross section, σ⁺⁺(E), of the metastable He⁻ 1s2s2p ⁴P^o state, obtained from the implementation of the state-specific, complex eigenvalue many-electron, many-photon theory (MEMPT). The σ⁺⁺(E) rises from 0 at E = 0 to a maximum of about 1.0 × 10⁻²⁰ cm² at E ~ 3–5 eV and dies out around 30 eV. For this quantity, comparison with the measurements of Bae et al for E = 0.0–0.4 eV and with the semiclassical calculations of Crothers and Lennon, reveals disagreement. The effect of initial-state correlation in this case is quantitative but not qualitative.

Angular distribution of the Xe M₄–N_4,5N_4,5 Auger transitions has been determined from an experiment which was carried out by 3.6 MeV proton impact. In a comparison with existing relativistic calculations for the α₂ parameters a good agreement was found which confirms the validity of relativistic theory in describing the core–core Auger transitions.

We present theoretical calculations for two-photon detachment rates and asymmetry parameters of H⁻ for energies below 1 eV. The calculated asymmetry parameters compare very well with values determined in a recent experimental work. This is the first time that this kind of comparison has been made for H⁻, which reinforces the predictive value of previous theoretical works on the two-photon detachment of H⁻ for which experiments are still lacking.

We have measured, for the first time, subshell-resolved L - K ( - K, - K and - K) electron transfer cross sections for electron transfer from the L-shell of the target to the vacant K-shell of the projectile. These cross sections are derived by measuring the charge-state dependence of the target L x-rays from a thin Au target using bare and H-like sulphur ions. In addition, the subshell-resolved ionization cross sections are also derived from the analysis of the L x-ray production cross sections. We found that the vacancy production cross section for the subshell is much smaller than that for strongly bound subshell and the - K transfer cross section is also much smaller than the - K transfer cross section. These observations are contrary to the prediction of first-order theories.

Triple differential cross sections (TDCS) for double photoionization in H₂ and D₂ (for comparison also in helium) have been measured for equal energy sharing, E₁ = E₂ = 10 eV. In agreement with previous observations by the Reddish group and in comparison to the equivalent helium case, our D₂ data confirm a `narrowing' effect in the correlation factor which influences the observed TDCS lobes, as well as a `filling-in' effect at relative angles θ_1,2 = 180^°. However, we do not confirm the previous differences in the magnitude of D₂ and H₂ TDCS values. Instead, we get equal TDCS strengths and, in comparison to helium, reasonable agreement with the corresponding total cross section ratio for double photoionization in He and D₂.

Two-photon absorption processes resulting in two-electron ionization of He and H⁻ are calculated using a time-dependent close-coupling expansion on a two-dimensional radial lattice. The ground state atom or ion is subjected to a 10-cycle pulse of linearly polarized light with peak intensities ranging from 10¹³ to 10¹⁶ W cm⁻². Probabilities are calculated for one-photon single ionization, two-photon single ionization and two-photon double ionization. The ratio of non-sequential double ionization to single ionization following two-photon absorption is found to be somewhat larger than the similar ratio following one-photon absorption.

The fermion-molecular-dynamics (FMD) method is applied to negative-muon capture by the hydrogen atom. It is found that the FMD results are similar to those of the classical-trajectory Monte Carlo (CTMC) method, whose validity was confirmed previously by the most accurate existing method, classical - quantal coupling (CQC). Thus FMD is validated as a quasiclassical method for capture and will be applicable to multi-electron atomic and molecular targets, for which the CTMC approach is not suitable. Actually the FMD results agree slightly better with the CQC results and suggest an interpretation of the very low-energy and adiabatic-ionization-threshold regions where the quantal (CQC) cross sections differ most from the classical (CTMC) results.

We use a simple two-electron model introduced previously in the literature (Watson et al 1997 Phys. Rev. Lett. 78 1884) to study the process of non-sequential double ionization of atoms. In particular, we discuss the phenomena of non-sequential double ionization of helium and we show that there is a definite relationship between the dipole radiation of the outermost (strongly ionized) electron and the ionization of the inner electron. We also investigate in detail the effect of quantum interference of the outer-electron wavefunction on the non-sequential ionization.

This paper describes the determination of the energy averaged over the atomic states, with a definite spin, derived from a general configuration composed of equivalent and nonequivalent electrons. Using a simpler approach, we extend previous determinations which were limited to the equivalent electron case. The procedure, which is based on our method for calculating traces of replacement operators over model spaces, leads to a formula which is appropriate for computer evaluations.

In addition to the quantized molecular structure that constrains the kinematics of atom-molecule collision, constraints arising from energy and angular momentum (AM) conservation strongly influence the magnitude and probability distribution of rotational transfer due to collisions. However, once the nature of these constraints is established, it is possible to predict quantitatively, using quantized momentum mechanics, the outcome of inelastic and reactive collisions. The existence of these constraints is most clearly portrayed in plots of relative velocity versus final rotational AM . These represent channel opening under the conditions: (i) energy conservation, (ii) simultaneous energy and angular momentum conservation, the latter via a specified maximum effective impact parameter and (iii) angular momentum conservation via this same is generally set to be the half bond length (HBL) of the diatomic, as suggested by experimental and theoretical considerations. Each of these conditions may constrain the rotational transfer process. The plots also give indications of conditions under which backward or forward scattering will occur and when long-lived complexes will form. For sharply defined velocity distributions, scattering angle peaks are readily calculated and these agree well with experimental data. When energy constraints dominate, the value of may be less than HBL and this reduced value may be obtained from the kinematic equations. This permits wider usage of the AM theory of rotational transfer. AM constraints effectively reduce the number of channels that appear to be accessible from energetic considerations. The plots may also be used to analyse reactive collisions and inelastic processes in polyatomic molecules for data in which the angular momentum change is directed principally along one of the molecule's inertial axes.

The influence of nonlinear interference effects (NIEF) on the radiation redistribution functions is investigated in terms of the atomic density matrix formalism. The ion motion is considered within the model microfield method. The particular calculations have been performed for the spectral doublet , of helium-like ion in hot dense plasmas (T = 350 eV, ). Taking account of both NIEF and ion dynamics is essential in the line centre and the nearest wings of the scattered radiation.

We consider the non-relativistic free-free absorption of a photon in the presence of a Coulomb field and calculate the momentum transferred to the resulting ion for polarized photons and isotropic initial electron angular distributions. Our calculations, carried to the quadrupolar order, end in an integral which is solved numerically. We show that the final ion always recoils in the opposite direction to the incident photon and that the transferred momentum may be much greater than the photon momentum. Our results are consistent with those of bound-free absorption processes in the limit where the initial bound state tends to the ionization threshold.

Effective collision strengths for O II ions are evaluated from electron-impact excitation collision strengths of singly ionized atomic oxygen ( O II) calculated in the close-coupling approximation using the ab initioR-matrix method at the Breit-Pauli approximation level. The lowest 21 fine-structure levels arising from the 11 , LS target states are included in the calculation. The effective collision strengths for all forbidden transitions between the ground configuration levels are obtained by averaging the electron collision strengths, for a wide range of incident electron energies, over a Maxwellian distribution of velocities. Results are presented for electron temperatures in the range 2000-1000 000 K, applicable to many laboratory and astrophysical plasmas. We believe that the results are the most accurate to date and so should be used in preference to previous values available in the literature.

The yield and angular distribution of the Auger electrons from the lowest transitions in 3p photoexcited ions are measured in the photon energy range 31-34 eV. Both are strongly influenced by the resonance structure of the excitation. An analysis of the observed angular distribution is given on the basis of the resonant cascade model.

We present a theoretical study of strong laser-atom interactions, when the laser field parameters are subjected to random processes. The atom is modelled by a two- and a three-level system, while the statistical fluctuations of the laser field are described by a pre-Gaussian model. The interaction of the laser-target is treated nonperturbatively by using the calculation method based on the Hermitian Floquet theory. Our aim consists in studying the kinetics of atomic relaxation induced by laser noise. In the resonant case with electric field strengths small with respect to the atomic unit electric strength, the present nonperturbative results are in agreement with those obtained within the rotating wave approximation of Eberly et al and Wodkiewicz et al for an atom modelled by a two-level system. We discuss some examples which demonstrate the destruction of atomic coherence by the noise, the regime of relaxation to the equilibrium state and the optical analogue of motional narrowing. We also give new results for two- and three-level systems, and for a strong laser field at exact resonance, in the case of phase, amplitude noises. The case of fluctuation due to collisions is also discussed. Our numerical results indicate that ionization effects, in the presence of laser noise, can lead to important modifications of the populations for strong laser-atom interactions. The changes in the ionization rates generated by the noise are also investigated.

The classical trajectory Monte Carlo method has been used to calculate capture and ionization cross sections and final-state electron momentum distributions in collisions to show how the collision mechanisms evolve over the range of intermediate projectile speeds . A circular Rydberg target was aligned in two orientations defined by the orbital angular momentum vector of the electron being either antiparallel or perpendicular to the incident momentum vector of the projectile. Aligning the target atom in this manner helps us to single out specific collision processes otherwise obscured in a randomly oriented atom. A plot of the capture and ionization cross sections shows a crossover in the capture and ionization curves as soft collisions get replaced by harder, more head-on collisions, which leads to more ionization events than captures. The ratio of ionization cross sections, , indicates a further change in collision dynamics around . It is believed that the collision dynamics for become dominated by the geometric size of the target. Plots of the ionized electron final momentum spectra for both alignments, projected onto the projectile-projectile collision plane, show specific types of two- and three-body interactions and show their evolution as a function of . The presence of these interactions can be identified and monitored by observing electrons captured to the continuum, backward scattered and saddle-point electrons, and binary collision rings and how each interaction evolves as a function of .

The Coulomb-Born approximation has been employed to study charge transfer cross sections in the case of collisions of (q = 1-3) and (q = 1-4) with atomic hydrogen in its ground state, respectively, within the energy range of . The interaction of the active electron with the incoming projectile ion has been approximated by a model potential. Cross sections for capture into different sub-shells have been given in tabular form. Computed total capture cross sections compare favourably with existing available results.

We present results of experimental measurements and theoretical calculations for the spin asymmetry, ), of the triple differential cross section. Spin-polarized lithium atoms were ionized by spin-polarized electrons of 54.4 eV incident energy. The geometry was coplanar with one detector at a fixed angle ( or ). Three data sets were obtained for ratios of shared energy = 8.8, 2.4, and 1. For the most asymmetric set the convergent close-coupling method gives best agreement, whereas for the symmetric set the distorted-wave Born approximation seems to be in better agreement.

Polarization of the radiation from Zn and Cd atoms (n = 4 for Zn and 5 for Cd, respectively) was observed after electron impact with energy slightly above the threshold excitation energy. The observing direction was perpendicular to both the incident electron beam and the atomic beam. All of the triplet radiation (J = 0, 1 and 2) was slightly polarized. The observed fractional polarization degrees were and corresponding to J = 0, 1 and 2 spectra in the final states of the Zn atom and , 4.3 and for the Cd atom. These values could be explained by assuming populations in the sublevels and of the states deviating from an equal mean value of for each sublevel. It is interesting and reasonable that the deviations determined independently for different states coincided with each other quite well. The deviated values of fractional population for the sublevel were 0.36 for Zn and 0.37 for Cd, respectively.

We report results from an ab inito calculation of low-energy electron scattering by ozone molecules. Applying the Schwinger multichannel method with pseudopotentials, we calculated elastic integral, differential and momentum transfer cross sections in an energy range from 6 eV up to 30 eV. We compare our results with available theoretical results and with available experimental data. In particular, our integral cross section shows two shape resonances, for and symmetries, which agree with the theoretical results of Sarpal et al, of Lee et al and with the observations of Allan et al.

A projectile elastic close-coupling approximation with electron exchange is employed to study Ps-H scattering using the basis set [Ps(1s) + H(1s,2s,2p)] at low and medium energies. The s-, p- and d-wave phase shifts are reported below 5.1 eV along with the integrated cross sections for elastic (also quenching) and target inelastic transitions. The present prediction is compared with other theoretical predictions and found to be consistent. The conversion ratio approaches the value 0.25 with the increase of incident Ps energy.

We have applied the non-adiabatic phase matrix method for vibrational scattering processes to a system which supports an extremely broad (short-lived) shape resonance. This method provides a unified, systematic treatment of the range of physical effects that influence the collision as the nuclear geometry varies, from strongly resonant to purely background scattering. For this implementation we have chosen the prototypical shape resonance in scattering below 10 eV. We compare vibrational cross sections and from the present non-adiabatic theory to results from fully converged body-frame vibrational close-coupling calculations and from purely adiabatic studies using the energy modified approximation.

The elastic scattering of positrons from beryllium and magnesium is studied by using a simple model potential to represent the interaction between the positron and target atoms. The model is tuned by adjusting the free parameters to reproduce previous predictions of the positron binding energies of the and ground states. An indication of the theoretical uncertainty in the derived phase shifts and cross sections is given by running a series of calculations with slightly different model potentials. The estimated scattering lengths are and for Be and Mg, respectively. The total cross section for positron-magnesium scattering constructed by adding the inelastic cross section computed with many-body perturbation theory to the present elastic cross section is broadly consistent with the experimental data.

A frozen core model of atomic copper is developed with the core consisting of the 1s-3d shells. The model is validated by predictions of the electron affinity and binding energies of neutral Cu. The model is used to predict the existence of an electronically stable ground state of with a binding energy of 0.005 518 Hartree and a spin-averaged annihilation rate of .

We have performed a numerical study of multiple scattering in fractal aggregates of different fractal dimensions D. We have found a behaviour which does not correspond to previous work on random media. Our theoretical approach allows us to understand the structure of the backscattered intensity and shows that the coherent backscattering peak due to time-reversal invariance is drowned in other types of interference.

Extreme ultraviolet line intensity ratios for Ne VII involving transitions in the range are calculated for application to high-temperature plasmas, using electron-impact excitation rates generated with the R-matrix code. These results are compared with line ratios measured from both -pinch and tokamak plasmas, and in general the agreement is found to be good. However, in a number of cases there are discrepancies between theory and experiment, which may be due to blending.

An input mistake was found in the outer region step of our calculations of the potential energy curves of NO at some internuclear distances. The resulting errors, although small, affect the reported positions of the higher intruder states, which are slightly lowered in energy. The corrected results for the NO symmetries ²Σ⁺ and ²Π, corresponding to figures 4 - 7, are given below. Corrected numerical data for these and the other figures given in our paper can be obtained from our world wide web server on http://www.tampa.phys.ucl.ac.uk/moldata.

The small changes in our results does not alter the discussion, except for the second paragraph of page 1982 which refers to the disappearance of the resonance 3sσ at R = 2.350 a₀. This resonance now does not disappear.

Figure 4. Potential energy curves of the ²Σ⁺ states of NO. Full light curves; Rydberg states. Only Rydberg states 3 - 6s σ and 3 - 6p σ are plotted. * mark the positions of the intruder states; full/broken curves join diabatically these states where bound/resonant. Resonances labelled nlσ and n/π are members of the Rydberg series converging to NO⁺(a ³Σ⁺) and NO⁺(b ³Π) respectively.

Figure 5. Effective quantum number, v, of the NO(²Σ⁺) bound states as a function of internuclear length. Included are the Rydberg states of l. The l character of the states is indicated in the extremes. ×, s; °, p; Δ, d; , f. * mark the positions of the intruder states.

Figure 6. Potential energy curves of the ²Π states of NO. Full light curves; Rydberg states. The dots in the X ²Π state are the grid of geometries calculated. Only Rydberg states 3 - 7p π are plotted. * mark the positions of the intruder states; full/broken curves join diabatically these states where bound/resonant. The resonances 3sσ and 3pπ are Rydberg states of NO⁺(b ³Π) and NO⁺(a ³Σ⁺) respectively.

Figure 7. Effective quantum number, v, of the NO(²Π) bound states as a function of internuclear length. Symbols are the same as figure 5.