Table of contents

Photon emission by light hydrogen-like ions interacting with intense, few-cycle Ti:sapphire laser pulses is calculated within the strong-field approximation of Lewenstein and co-workers, modified so as to treat the coupling of the atom with the incident field beyond the dipole approximation. The approach is tested by comparing with results obtained by solving the time-dependent Schrödinger equation numerically for a two-dimensional model ion. The effect of the magnetic field component of the driving pulse is significant beyond 2×10¹⁷ W cm^-2 peak intensity. It leads, in particular, to a saturation in photon emission with increasing laser intensity. As compared with emission polarized along the same direction as the incident field, emission polarized along its propagation direction is weaker by about two orders of magnitude at the maximum peak intensity considered, 3.6×10¹⁷ W cm^-2.

A new relation between the intrinsic parameters δ₁ and ξ₁ which describe the spin polarization of Auger electrons emitted in the decay of the oriented j = 1/2 vacancy is found. A similar relation for intrinsic parameters describing the resonant Auger photoemission from j = 1 states is discussed. The implication for a complete experiment in Auger decay is considered.

We present calculations of intense-field multiphoton ionization processes in helium at XUV wavelengths. The calculations are obtained from a full-dimensional integration of the two-electron time-dependent Schrödinger equation. A momentum-space analysis of the ionizing two-electron wavepacket reveals the existence of double-electron above threshold ionization (DATI). In momentum-space two distinct forms of DATI are resolved, namely non-sequential and sequential. In non-sequential DATI correlated electrons resonantly absorb and share energy in integer units of ℏω_laser.

A new enantio-selective process is considered: left- and right-handed molecules can scatter unpolarized electrons differently if true chirality is defined by the experimental geometry. This effect is caused by Coulomb forces resulting in significant asymmetries. The theoretical basis of this phenomenon is analysed within the framework of Barron's concept of chirality. Our conclusions are illustrated by numerical calculations for elastic electron collision from H₂S₂.

An extension of the GRASP92 (Parpia F A, Froese Fischer C and Grant I P 1996 Comput. Phys. Commun.94 249) multi-configuration Dirac-Fock (MCDF) program described previously (Perger W F and Idrees M 1995 Phys. Commun.85 389-97) is used for the calculation of the specific mass shift (SMS) of the helium ¹S ground state isoelectronic sequence. We also employ a multi-configuration Hartree-Fock (MCHF) method to calculate the ground state SMS for comparison with MCDF results. The SMS matrix elements for two-electron systems obtained from the relativistic program are shown to exhibit a trend: the larger the atomic number Z, the larger the relativistic contributions to the SMS matrix elements for the ions. The SMS matrix elements approximately vary as Z³ along the isoelectronic sequence from Z = 2 to 92. In addition, it is shown that the relativistic effects increase approximately as Z³/A² in the SMS values for all the ions considered confirming some previous observations (Parpia F A, Tong M and Froese Fischer C 1992 Phys. Rev. A 46 3717-24).

Excellent agreement is found between the present ab initio calculations and the available semi-relativistic calculations for the small values of Z along the helium-like ions. Furthermore, a large set of configuration state functions used in the calculations has revealed larger disagreements for high-Z ions between both relativistic (MCDF-optimized-level) and nonrelativistic (MCHF) calculations suggesting that the SMS for helium-like ions with Z>40 relativistic and correlation effects are increasingly important.

The emission spectra of Ne⁶⁺ and Ne⁷⁺ ions generated by beam-foil excitation in the range 18-48 nm show a number of lines arising from doubly excited levels. Hartree-Fock calculations using superpositions of quasi-degenerate configurations provided many of the identifications, as did interpolations along the lithium-like and beryllium-like isoelectronic sequences. From these observations, energies of Ne VII 1s2s2pnℓ and 1s2p²n'ℓ' configurations are derived for the n = 3 and 4 levels. This has led to a revision of several Ne VII levels previously reported in the literature. In addition, seven new 3d-4f transitions were found between doubly excited states in lithium-like Ne VIII**.

A quantum simulation of the thermally averaged translational-rovibronic spectrum of Li*He associated with the 3D(Σ,Π,Δ)←2P(Σ,Π) and 3P(Σ,Π)←2P(Σ,Π) transitions is performed, using as input data accurate potentials and dipole transition moment functions derived from ab initio calculations. A few inaccuracies of the potentials existing in the previous simulation of this spectrum are corrected, and previously neglected bound states are taken into consideration.

The spectral profiles arising from different molecular channels are calculated with a high degree of accuracy, and the total absorption cross section summed over all the contributions is compared with the experimental spectrum. It is demonstrated that the inclusion of bound states considerably improves the agreement with the experimental values which finally seems to be almost very good. This result testifies to the accuracy of input data used for simulation.

We present a new form of explicitly correlated wavefunction whose parameters are mainly linear, to circumvent the problem of the optimization of a large number of nonlinear parameters usually encountered with basis sets of explicitly correlated wavefunctions. With this trial wavefunction we have succeeded in minimizing the energy instead of the variance of the local energy, as is more common in quantum Monte Carlo methods. We have applied this wavefunction to the calculation of the energies of Be ³P (1s²2p²) and Be^- 4S^o (1s²2p³) by variational and diffusion Monte Carlo methods. The results compare favourably with those obtained by different types of explicitly correlated trial wavefunction already described in the literature. The energies obtained are improved with respect to the best variational ones found in the literature, and within one standard deviation of the estimated non-relativistic limits.

We calculate and analyse spectra of electrons emitted from ultrarelativistic projectiles colliding with neutral atoms. Results of our calculations for the electron loss spectrum from 33 TeV Pb⁸¹⁺ colliding with Al atoms considerably differ from the shape of the electron loss spectrum measured in a recent experiment for Pb ions penetrating a solid Al target.

Electron pair densities and their moments associated with a set of multiconfigurational wavefunctions that systematically include nondynamic electron correlation have been constructed for the beryllium isoelectronic sequence. The obtained two-electron quantities are compared to the Hartree-Fock and nearly exact configuration interaction results. It is shown that the electron-electron counterbalance density identifies the non-dynamical (long-range) electron correlation. Hence, the wavefunction that maximizes the electron-electron counterbalance density contains all the non-dynamical electron correlation. This statement constitutes the proposed separation criterion. Explicitly calculated dynamical and non-dynamical Coulomb holes for the isoelectronic series of beryllium illustrate the feasibility of this approach.

Results of the quantum close-coupling calculations for the spin-changing transition in Ba(6s6p,¹P₁→³P₂) induced by collisions with argon are presented. The dependence on angular and collision energy of the calculated multi-structure differential cross-sections are discussed. The backward scattering is most prominent in a small energy range, particularly in the case of Π preparation of the initial Ba(¹P₁)-Ar molecular state. The similar characteristic features of differential cross-sections are observed in the experiment by Visticot et al (Visticot J P et al 1992 Phys. Rev. A 45 6371).

A theoretical analysis of laser-driven collisional ejection of inner-shell electrons is presented to explain the previously observed anomalous kilovolt L-shell x-ray emission spectra from atomic Xe cluster targets excited by intense sub-picosecond 248 nm ultraviolet radiation (McPherson A et al 1994 Nature370 631-4). For incident ponderomotively-driven electrons photoionized by strong above threshold ionization, the collisional ejection mechanism is shown to be highly l-state and significantly n-state (i.e. radially) selective for time periods shorter than the collisional dephasing time of the photoionized electronic wavefunction. The resulting preference for the collisional ejection of 2p electrons by an ionized 4p state produces the measured anomalous Xe(L) emission which contains direct evidence for (i) the generation of Xe²⁷⁺(2p⁵3d¹⁰) and Xe²⁸⁺(2p⁵3d⁹) ions exhibiting inner-shell population inversion and (ii) a coherent correlated electron state collision responsible for the production of double 2p vacancies. For longer time periods, the selectivity of this coherent impact ionization mechanism is rapidly reduced by the combined effects of intrinsic quantum mechanical spreading and dephasing - in agreement with the experimentally observed and extremely strong ~λ^-6 pump-laser wavelength dependence of the efficiency of inner-shell (2p) vacancy production in Xe clusters excited in underdense plasmas (Kondo K et al 1997 J. Phys. B: At. Mol. Opt. Phys.30 2707-16).

We consider radiative electron capture (REC) in fast nonrelativistic collisions of light atomic particles. In order to treat the process we adapted a formalism in which the REC is viewed as collision-stimulated transitions between electron states of different particles dressed by the radiation field. The formalism leads to frame-independent cross sections. Within the formalism the problem of defining the charge current operator does not arise. The results obtained for REC are close to those of Briggs and Dettmann (Briggs J S and Dettmann K 1974 Phys. Rev. Lett.33 1123).

This paper includes cross sections and differential cross sections for the processes + ps(1s)→e + (n⩽2), assuming CPT symmetry and using ten partial waves with total angular momentum L ⩽ 9. Comparisons with previous calculations are made whenever feasible and good agreement is observed. It is noted that the forward peaks of the differential cross sections of the formation of (2p) atoms are much broader than those of the formation of S-state atoms. A strong secondary maximum of one of the (2p) formation processes is centred around 45°, in contrast to S-state processes at comparable energies where the differential cross sections are strongly peaked in the forward direction with θ < 40°. The combined differential cross sections of the two processes that lead to the formation of (2p) are also given for a number of energies.

Convergence problems in coupled-cluster iterations are discussed, and a new iteration scheme is proposed. Whereas the Jacobi method inverts only the diagonal part of the large matrix of equation coefficients, we invert a matrix which also includes a relatively small number of off-diagonal coefficients, selected according to the excitation amplitudes undergoing the largest change in the coupled-cluster iteration. A test case shows that the new inversion of partial matrix (IPM) method gives much better convergence than the straightforward Jacobi-type scheme or such well known convergence aids as the reduced linear equations or direct inversion in iterative subspace methods.

The outermost closed subshell photoionization cross sections from He, Li, Na and K atoms and asymmetry parameters β of the 2p photoionization from Na atoms are calculated applying both relaxed-orbital Hartree-Fock (ROHF) and sudden perturbation (SP) approximations. The suitability of both methods for the calculation of a single photoionization and shake-up cross sections is considered, and differences between these and frozen orbital approximations are discussed. The higher accuracy of the ROHF approximation with respect to that of SP is found near the threshold of ionization. The SP approximation can be used only for high photon energies. The weak dependence of the parameter β on the applied method (ROHF or SP) is noticed.

The process of electron tunnelling through a one-dimensional non-stationary barrier created by an atomic potential and an external electromagnetic bichromatic field consisting of coherent superposition of the fundamental laser field frequency and its second harmonic is considered and studied theoretically. Within the framework of the imaginary time method the analytical expression for the corresponding photoionization probability has been derived which is, in fact, the analogue of the well known Keldysh formula generalized for the bichromatic case.

Using five-laser resonance excitation via the intermediate state 5d6d ¹P₁, the La⁺ Rydberg states with n from 23 to 62 have been measured. Based on the eigenchannel quantum defects µ_α and transformation matrices U_iα, which are calculated from first principles by relativistic multichannel theory, the energy levels of the La⁺ Rydberg series are calculated within the framework of multichannel quantum defect theory and are in good agreement with the experimental values. With the calculated results we provide clear assignments for the experimental spectra.

Relative differential cross sections, at 0°, for electron-impact excitation of the b ³Σ⁺ state of carbon monoxide have been measured in the near-threshold energy region. A high-resolution crossed-beam double trochoidal electron spectrometer is used and the cross sections are measured directly, by detection of inelastically scattered electrons. Absolute differential cross sections at 0° are obtained, after normalization, in the energy region from threshold to 15 eV, for the first time. Integral cross sections are obtained by using relative angular distributions from previous measurements. The results obtained are compared with other available data.

The K-MM radiative-Auger transition probabilities for Ca, Ti and Cr were calculated using time-dependent perturbation theory. The atom plus the free radiation field are considered as being one system and the transition of one state of the atom to another results from the interaction between components of this system. The perturbing Hamiltonian is taken as the sum of the interaction between the electrons participating directly in the process and their interaction with the radiation field. Under this approach, the radiative-Auger process is described considering second-order terms in time-dependent perturbation theory series. The calculations were performed using screened hydrogenic wavefunctions to describe the state of the atom. The results obtained are compared with recent experimental data and other theoretical calculations.

Harmonic generation by an atom in a laser field is described by the three-step mechanism as proceeding via above-threshold ionization (ATI) followed by electron propagation in the laser-dressed continuum and subsequent laser-assisted recombination (LAR). The amplitude of the harmonic production is given by the coherent sum of contributions from different intermediate ATI channels labelled by the number, m, of absorbed laser photons. The range of m-values that give a substantial contribution is explored and found to be rather broad for high harmonic generation. The coherence effects are of crucial importance as they are responsible for the characteristic pattern of harmonic intensities with a plateau domain followed by a cut-off region. Due to the multiphoton nature of the process, an efficient summation of m-contributions can be carried out within the framework of the saddle-point method. The saddle points correspond to some complex-valued labels m = m_c associated with the intermediate effective ATI channels in the three-step harmonic generation process. The advantage of this approach stems from the fact that summation over a large number of conventional ATI m-channels is replaced by summation over a small number of effective m_c-channels. The equation governing m_c has a transparent physical meaning: the electron ejected from the atom on the first (ATI) stage should return to the core to make LAR possible. The effective channel labels m move along characteristic trajectories in the complex plane as the system parameters vary. In the cut-off region of the harmonic spectrum a single effective channel contributes. For lower harmonics, in the plateau domain, two effective ATI channels become essential. The interference of their contributions leads to an oscillatory pattern in the harmonic generation rates. The calculated rates are in good agreement with the results obtained by other approaches.

The shape and width of lines observed in time-of-flight mass spectroscopy applied to molecular photofragmentation contain information on the initial momentum and energy distributions of the fragment ions. The problem of extracting this information is discussed in perspective. Usual determinations of a single kinetic energy parameter may, depending on the line shape, lead to quite biased results. A simple and intuitively representative solution is suggested instead. It is shown that, under the conditions of isotropic emission and absence of kinematical discrimination, the mean kinetic energy is proportional to the variance of the line distribution. This parameter is rather insensitive to noise and easy to correct for instrumental and thermal broadening, as is demonstrated on spectra of the decay CS₂²⁺→S⁺ + CS⁺, observed at various energies of the mother ion.

The controlled deflection of Rydberg atoms in an inhomogeneous electric field is demonstrated and shown to be in agreement with predictions based on a well-defined Stark map. Krypton atoms in a supersonic beam are excited in a two-colour multiphoton process to selected Stark states with n = 16-19 and undergo deflections in the field of an electrostatic dipole. The spatial distribution of the deflected atoms is monitored by ion imaging. The atoms travelling ~10 cm in the beam direction after excitation are deflected from the beam axis by distances of up to 3 mm, the interaction with the field taking place over 40 µs. Simulations of the trajectories agree well with the experimental results, demonstrating that the detected Rydberg atoms are those which have not undergone decay processes in the 40 µs applied field.

Raman interaction of a trapped ultracold ion with two travelling wave lasers has been used extensively in ion-trap experiments. We solve this interaction in the absence of a rotating wave approximation (RWA) by a continued fraction, without considering the restriction of the Lamb-Dicke limit and the weak excitation regime. Some interesting characteristics of the ion-trap system, particularly for the ion outside the weak excitation regime are found. Finally, a comparison of our results with the solution under the RWA is made.

We analyse measurements of the Stark-broadened profile of the He II Balmer-α transition at densities and temperatures up to n_e = 3×10¹⁸ cm^-3 and k_BT_e = 30 eV, respectively, and compare the results with calculations and existing experimental data. Theoretical spectra calculated according to the frequency fluctuation model (FFM) agree within the measured spectral range with the experimental ones when ion dynamics is taken into account, whereas calculations according to the model microfield method (MMM) underestimate the width. Concerning the Stark shift, the present experimental values are consistent with existing shifts measured at lower temperatures but due to the experimental error bars it is not possible to distinguish between predictions based on the MMM or the standard theory, even though these predictions differ by a factor of two.

We have measured the total cross sections for single charge exchange in He²⁺-He⁺, Ne²⁺-Ne⁺ and Ar²⁺-Ar⁺ collisions at centre-of-mass energies of 1.8-14.8, 1.8-10.8 and 2.8-7.3 keV, respectively, using an intersecting beam technique. The results for He²⁺-He⁺ collisions are compared with existing measurements and calculations. For Ne²⁺-Ne⁺ and Ar²⁺-Ar⁺ the measured cross sections are compared with results from new close-coupling calculations based on a one-electron model potential description of the collision system.

Experimental branching fractions (BFs) of Mo II, ranging in wavelength from 1970 to 4370 Å, have been measured from intensity calibrated spectra recorded with the Lund UV Fourier transform spectrometer (FTS). Radiative lifetimes for 10 levels have been measured using the method of laser-induced fluorescence (LIF). Combining BFs with new as well as previously measured lifetimes, 16 in total, oscillator strengths of 91 lines were derived. Seven transitions are resonance lines involving the ground state. The BF results are compared with calculations made with the Cowan code and the f-values are compared with previously published data. Improved wavelengths from an ongoing term analysis are also reported.

We present the computation of the two-photon transition matrix element between vibrational states of H₂⁺ or D₂⁺ of ¹S^e symmetry (i.e. two J = 0 vibrational levels of the 1sσ_g electronic ground state). The method uses very accurate fully non-adiabatic wavefunctions of the non-relativistic problem. It is first applied to the calculation of the static polarizabilities; our results for the ground state are in excellent agreement with the literature, with an improved accuracy. The method is applied to the evaluation of the two-photon transition probabilities and light shifts. We also discuss the feasibility of a two-photon spectroscopy experiment in H₂⁺.