Table of contents

Experimental studies of heavy and highly charged ions have made remarkable progress in recent years. Today it is possible to produce virtually any ion up to hydrogen-like uranium; to study collisions of those ions with atoms, electrons, and solid surfaces; to excite such an ion and accurately measure the radiation emitted. This progress is largely due to the development of new experimental methods, for instance, the high-energy ion accelerators, laser-produced plasmas, advanced ion sources and ion traps (such as EBIS, EBIT, ECR, etc.), high temperature magnetically confined plasmas and heavy-ion storage rings.

The motivations for studies of collisions with highly charged ions and for the understanding of the structure of heavy atomic systems are multi-faceted. Besides of the basic scientific aspects which are mainly the subject of this symposium, much incentive is experienced by applications, e.g., the interpretation of spectra from space (solar corona, solar flares and hot stars), the modelling of stellar atmospheres, the diagnostics of fusion plasma impurities, and the development of X-ray lasers.

Since quite some time highly charged ions play a key role for high-precision metrology of atomic structure. These studies have been benchmarks for tests of advanced theories, including many-body theories of interelectronic correlations, relativistic and quantum-electrodynamic (QED) effects, effects due to the finite size of the nucleus and to parity non-conservation (PNC). The interest in QED effects in heavy ions has increased drastically in the last few years. The remarkable experiment on Li-like uranium, recently reported from Berkeley, has stimulated several groups to perform very accurate Lamb-shift calculations on such systems, and reports from three groups were given about such work. The agreement between the calculations as well as with experiment was generally very good, which implies that the problem of evaluating the first-order Lamb shift for any element is now essentially solved. The experimental accuracy is already so high that also higher-order QED effects become observable, and several groups are now active in trying to evaluate such effects from first principles.

Another related field where substantial progress has recently been made involves precision measurements of X-ray transitions. This has created an interest in the study of deep inner holes in heavy atoms, where large relativistic and QED effects appear. These effects are as large as in corresponding highly charged ions, but the interpretation requires that the many-body effects from the surrounding electrons are accurately extracted. This is a big challenge at present.

Atomic collision physics with highly charged ions has been dominated in recent years by the search for a possibility to describe electron-electron interaction within the dynamics of collisions. The experiments on multielectron transfer reactions with highly charged ions posed in this respect quite a challenge to the theory. The models developed to meet this were often based on methods and terminologies developed for describing the inter-electronic interactions in atomic structure. This caused many controversial discussions, also during this symposium. A new and fast rising field is the interaction of highly charged ions with solid surfaces. This may become an important link between atomic physics and condensed-matter physics, stimulated by the opportunity to study effects in coupled many-body systems present in the case when a large amount of electrons is transferred from the solid to each single ion. Furtheron, collision experiments with cooled ion beams in ion storage rings open new dimensions also for atomic spectroscopy. It appears possible that transition and binding energies can be measured in recombination of very heavy ions with a better quality than by conventional Auger electron or X-ray spectroscopy.

Obviously, it is not possible to cover all the fields mentioned here in a single symposium, and we had to concentrate on some aspects, mainly based on our own personal interests. Initially, two symposia were planned, one about heavy-ion spectroscopy with connections to collision phenomena and to atomic structure, and another with focus on relativistic and QED effects in heavy atoms and ions. Then it was realized that there was a considerable overlap between the two fields, both regarding the scientific contents and the people that would be involved. Furthermore, we felt that it might be stimulating also to bring people together, who are connected to different aspects of heavy-ion research and who do not have any natural contact in their research. Therefore, the idea was born to.merge the two symposia into one with the hybrid title Heavy-Ion Spectroscopy and QED Effects in Atomic Systems. There is always a risk with such an arrangement, namely that the symposium will split into two parts with only little communication between the two. We actively tried to avoid such a development by various arrangements, and we believe that we were reasonably successful in that respect.

The symposium was organized in the form of four review sessions and four more specialized mini-symposia. The review talks covered topics like many-body theory, QED and PNC effects in atoms, electron correlation effects in atomic collisions, electron capture ion-atom collisions and ions in space. Experimental reviews were given about heavy-ion experiments at Berkeley and Darmstadt. The four-mini symposia were devoted to (i) atomic structure and (ii) QED effects, (iii) multi- electron transfer reactions, and (iv) PNC effects. In these cases efforts were made to mix people with the main background in atomic collision physics and atomic structure. The mini-symposium devoted to structure ended with a panel discussion about the future of atomic structure theory.

Of course, this book can only attempt to summarize the high level of knowledge existing in atomic spectroscopy and heavy ion collision phenomena. This became evident to us in the excellent lectures given during the symposium. Here we like to thank our distinguished authors for their enthusiastic cooperation in the task of transmitting the message of the Nobel Symposium 85 to the science community.

Some predominantly experimental considerations concerning both few electron highly charged ion spectra and inner-shell vacancy spectra are discussed in relation to the currently available theoretical approaches. For the cases of H-like and He-like spectra theoretical machinery appears to be well in hand particularly in comparison to the relatively sparse experimental database which is also of limited accuracy in most cases. Owing particularly to some impressive recent theoretical progress on the problem of nearly neutral ions containing inner-shell vacancies, the traditional x-ray data suitably filtered and strengthened by modern measurements, appear to offer some quite useful insights in spite of their formidable complexity. With the advent of new storage ring facilities and advanced technology ion sources as well as a new effort on transuranic x-ray spectra, one can reasonably foresee an upcoming period of fruitful activity.

The three possible types of atomic symmetry breaking, P not T, P and T and T not P, are reviewed in a unified way. A convenient two by two matrix format is used in the discussion. The current state of atomic theory and experiment is reviewed and the implications for particle physics are summarized.

Electron correlation effects are studied in the semi-classical approximation relevant for energetic ion-atom collisions. The basic formalism for dynamic electron correlation is presented in analogy with the corresponding stationary aspect of configuration mixing. The dynamic aspect of electron correlation is shown to manifest itself in dielectronic processes, which are two-electron phenomena induced by the electron-electron interaction. First-order perturbation theory is used to describe the time-dependent coupling of states which differ by two spin orbitals. For specific cases, analytic formulas are evaluated to estimate the strength of the dielectronic processes. A brief survey over the experimental studies of electron correlation in ion-atom collisions is given. Model calculations are discussed for the process of autoexcitation and dielectronic excitation in comparison with previous experimental results.

Highly ionized atomic physics systems are used to probe processes taking place in the upper solar atmosphere, a region that, when flares are included, encompasses some three orders of magnitude in the temperature (3 × 10⁴−3 × 10⁷K). A thorough review of spectral line intensities implies that plasmas in the different parts of the upper solar atmosphere are transiently ionizing and are consistently associated with increased electron density. A possible mechanism consistent with the observations is a current driven plasma compression.

Recent work on nonperturbative (in Zα) calculations of quantum electrodynamic (QED) effects in high-Z few-electron atoms is reviewed. Topics covered are precise calculations of the Coulomb self energy in hydrogenic atoms, nuclear size effects on the self energy, and electron screening effects on the self energy and vacuum polarization. Various computational methods are compared.

The theory of many-electron atoms is treated first from a many-body perturbation theory approach, and then in terms of Furry representation QED. The connection between the two approaches is shown to allow the precise definition of QED effects, and it is shown that the spectroscopy of highly charged ions provides an ideal way to study these effects. One-photon Feynman diagrams are evaluated for sodiumlike platinum in a non-Coulomb potential, and shown to give good agreement with experiment. The role of two-photon diagrams and the importance of their complete evaluation is discussed.

This article discusses our measurement of the 2²P_1/2-2²S_1/2 (lowest excited state to ground state) energy splitting in lithiumlike uranium, which has large quantum electrodynamic corrections. The reasons for using lithiumlike uranium are discussed. Our result, 280.59 ± 0.10 eV, is a benchmark test for current theory.

Doubly excited states of atoms are not described by the independent electron approximation or the shell model. Various newly developed classification schemes are reviewed. It is shown that a certain class of doubly excited states exhibit collective modes similar to the rotor-vibrator motions of molecules and these states can be classified using molecular quantum numbers. In terms of these quantum numbers, new regularity and order in energy levels, radiative decay rates and Auger widths of doubly excited states are illustrated.

An attempt is made to identify the most important mechanisms responsible for the rearrangement of electrons during collisions between multiply charged ions and atoms at keV energies. It is discussed to which extent the influence of binding energy, angular momentum of heavy particles and electrons, and electron-electron correlation on the dynamic processes can be understood by means of simple models. Collision systems comprising only one, two, or many active electrons are distinguished, since rigorous theoretical treatments for such systems are well established, just being developed or not yet available respectively.

A report on the recent atomic physics program at GSI is given. The synchrotron storage cooler ring combination SIS/ESR allows us to produce, store and cool highly charged heavy ion beams up to U⁹²⁺. With the availability of such highly charged ion beams unique experiments have become possible, such as the observation of bound β decay of ¹⁶³Dy⁶⁶⁺ → ¹⁶³Ho⁶⁶⁺, the precision measurement of radiative electron capture and x-ray spectra in H-like and He-like heavy ions, and the investigation of dielectronic recombination for heavy systems. Besides this newly started program at SIS/ESR there has been some progress in the investigation of the correlated e⁺e⁻ lines observed in heavy ion collisions near the Coulomb barrier although a convincing explanation for this phenomenon is still lacking.

The investigation of nuclear-spin-dependent P-odd effects in atoms is the source of important and reliable information on P-odd nuclear forces. The upper limits on the T-odd, P-odd quark-quark and quark-gluon interactions deduced from the searches for the neutron and atomic electric dipole moments are comparable. The upper limits on the T-odd, P-even electron-electron, electron-nucleon and nucleon-nucleon interactions, as well as on some T-odd β-decay constants, extracted from the bounds on the dipole moments are much better than those following from the corresponding direct experiments. Extremely strict upper limits on the T-odd, P-even photon-fermion interaction are deduced from high-energy experiments.

A brief general survey is given of the experimental situation in atomic parity nonconservation and in the search for permanent electric dipole moments of the free neutron, atoms and molecules.

No abstract supplied

The path to accurate relativistic atomic structure calculations is lined with several points of interest or danger. Much of the non-relativistic many-body perturbation theory formalism, including the "coupled-cluster approach", can be taken over, but certain problems, such as the choice of electron-electron interaction, the need for projection operators and an apparent gauge dependence, must be dealt with. Results for the ground state energy of beryllium and also 2^1,3P-2¹S_o transition energies in beryllium-like Fe and Mo are presented as examples, where the accuracy is such that the calculation of QED effects in many-electron systems might be tested. In addition, the application to the study of additional perturbations in alkali-like system is discussed.

Convergence problems encountered in Multi-Configuration Dirac-Fock calculations using standard finite-difference techniques to solve differential equations are shown to originate from lack of positive-energy projection operators. A new numerical method is proposed which highlights the effects of the negative energy continuum, and can be used to project it out. It is shown that after projection all numerical problems vanish, and that very good convergence is obtained for high Z (up to 130), even when the Gaunt interaction is made self-consistent. Results for the ground state of two electron ions are presented.

No abstract supplied

Essentially exact results can now be obtained for the nonrelativistic energy and lowest-order relativistic corrections of helium. A combination of variational calculations and asymptotic expansion methods covers the entire singly-excited spectrum. Comparisons with recent high precision experiments allow the Lamb shift for the 1s2s¹S state to be extracted with an accuracy approaching that in one-electron He⁺(2s). Techniques for calculating the two-electron Lamb shift are discussed, including recent asymptotic expansion methods for the Bethe logarithm. Extensions to lithium and other three-electron ions up to U⁸⁹⁺ are briefly mentioned.

We present a new practical way to calculate the first order self energy in any model potential (local or non-local). The main idea is to introduce a new straightforward way of renormalization to avoid the usual potential expansion implying a large number of diagrams in higher order QED effects.

The renormalization procedure is based on defining the divergent mass term in coordinate space and decomposing it into a divergent sum over finite partial wave contributions. The unrenormalized bound self energy is equally decomposed into a partial wave (l) sum. For each partial wave the difference is taken and the sum becomes convergent. The comparably rapid asymptotic behaviour of the method is l⁻³. The method is applied to lithium-like uranium, and the self energy in a Coulomb field, the finite nucleus effect and the screened self energy is calculated to an accuracy of at least one tenth of an eV.

A new approach to the implementation of a mass-renormalization procedure in the evaluation of self-energy corrections to atomic energy levels in many-electron systems is presented. A partial-wave expansion is adopted for the bound- and free-electron propagator functions. All divergent quantities are eliminated from the calculation without the use of regulating parameters, yielding a convergent series of contributions to the self-energy component of the Lamb shift.

Inner shell transitions in medium to very heavy atoms are treated within a relativistic framework. The many-body part of the calculation includes full relaxation, correlation and the admixture of, sometimes degenerate, states with two vacancies and one excited particle. The Breit interaction is treated on equal footing with the Coulomb part of the electron-electron interaction through the whole calculation and the retardation beyond the Breit interaction is included in lowest order. The effect of the finite nuclear size is substantial and special care has been taken to use a correct nuclear mean square radius even for deformed nuclei. Hydrogenic radiative corrections (with finite nucleus effects) as well as screening contributions are included. Comparison with experiments over a wide range of elements show agreement within combined theoretical and experimental uncertainties.

We review ab initio calculations of the bound-state self-energy and vacuum polarization for resonant transitions of Li-like, Na-like and Cu-like ions with a degree of ionization of about ten or greater. The direct part of the Dirac-Fock potential is included to all orders in the rediative corrections, and exchange effects are added in first order. When combined with previous relativistic correlation calculations, the results agree in most cases with experiment, the most notable exception being the Cu isoelectronic sequence at low Z, where the third-order correlation calculations have probably not converged to experimental precision. For low and intermediate Z, the "screening" effect is found to scale closely as the normalization of the valence wavefunction at the origin, but at high Z deviations from this scaling principle can be significant.

The higher-order QED corrections for one- and two-electron heavy ions are analysed. The different methods of renormalization are discussed: the traditional potential-expansion method and the new one, based on the direct numerical subtraction of divergencies. The extraction of the reference state from the sums over the intermediate states in higher-order corrections is considered. It is shown, that this extraction leads to some special corrections to the energy.

The Low theory of the line profile is used for the calculation of the higher-order QED corrections and the nonresonant corrections distorting the Lorentz line profile are also discussed.

I summarize in this paper the present status of our experimental knowledge on the Lamb shift of high Z hydrogenlike ions. Some tentative prospect on the future improvements with the new large accelerators and ion sources are discussed and compared with the present accuracy of QED corrections.

The results on the 2S Lamb shift (2s²S_1/2−2p²P_1/2-level spacing) in hydrogen-like ions (Z = 15, 16) obtained by use of the laser resonance method are stated and discussed with respect to their significance for predictions obtained by theory. The properties of a new experimental technique - the autocollimation spectroscopy of broad resonances - are depicted. Finally we report the current status of preparation for a measurement of the 2S Lamb shift in ²⁸Si¹³⁺ and give an outlook to the future perspective of laser resonance experiments.

An outline of the treatment of field theoretical systems, as QED and QHD, in terms of the density functional approach is presented.

A nonperturbative quantum-electrodynamical approach to strong-field nonresonant multiphoton ionization is described and compared with other scattering models. This approach which is based on a work by Guo, Åberg, and Crasemann makes use of formal time-independent scattering theory and of availability of Volkov states for the construction of Green's operators. The semiclassical laser-field approximation is derived from the Dirac equation by considering the large-photon-density limit of a multimode electromagnetic field. A recent interpretation of the strong-field Kapitza-Dirac effect by Guo and Drake is treated as an example.

No abstract supplied

Accurate atomic many-body calculations of the parity nonconserving 6s → 7s dipole amplitude in atomic cesium are described. These calculations lead to the value

-0.905(9) × 10^-11i|e|a₀(-Q_w/N)

for the 6s → 7s amplitude. Combining this value with the measured amplitude leads to the value Q_W = 71.1 ± 1.6 ± 0.9 for the weak charge, where the first error is from the measurement and the second is from the calculation. Implications of this result for particle physics are discussed.

We discuss the possibility for search of correlation of nuclear spin with momentum of a photon Ik which is due to the parity nonconservation or to the nuclear electric dipole moment. The manifestation of this effect is the dependence of the nuclear magnetic resonance (NMR) frequency on the direction of a laser beam. For the electric dipole moment this is perhaps just a theoretical exercise because of broadening of the NMR line. However for the weak interaction, and especially for the measurement of nuclear anapole moment the suggested way looks promising.

In the quantum many-body systems that have a dense spectra of excited states (compound nuclei, rare-earth atoms, molecules, clusters, "quantum dots" in solids, spin systems) weak perturbation can be strongly enhanced. An example of a compound nucleus is considered in detail. The factor of enhancement of parity nonconserving (PNC) effects in neutron-nucleus reactions exceeds 100. Calculations of the mean square value of the matrix element between compound states show that the statistical mechanism explains the "random" part of the dynamical enhancement. Possible mechanisms of regular effect are considered. It is shown that the valence mechanism (PNC effect due to single-particle component of compound states and potential scattering) contradicts the results of measurements of PNC effects in low level density nuclei (n + ¹²⁴Sn, ²⁰⁷Pb, etc.). Calculations show also that the contribution of the valence mechanism is 100 times smaller than the observed PNC effect in ²³²Th. Any enhancement due to collective excitations (giant resonances) is also excluded. The limit on the strength of the neutron PNC weak potential is extracted from the measurements of neutron spin rotation in ¹²⁴Sn: |g_n| ⩽ 1 or |ε| ⩽ 10^-8. Possible mechanism of correlations among compound states components ("quasielastic" mechanism) is considered.

No abstract supplied

Basic Double Capture mechanisms are reviewed for collisions of highly charged ions with atomic target in the keV energy range. It is the main goal of this paper to clarify the relative role of the electron-nucleus and electron-electron interactions. First two independent electron models (IEM) are briefly described: the classical over barrier model and the quasi-molecular model. They account well for the population of configuration with two equivalent electrons. It is then shown that e⁻-e⁻ interaction is necessary to account for the population of doubly excited states with two highly non equivalent electrons. A special emphasis is given to a recent work on the stabilisation mechanisms responsible for true double electron capture.

We have measured cross sections for true double-electron capture (DC) and transfer ionization (TI) in slow Xe^q+-(Ar, He, Xe) collisions in the charge-state regime 15⩽q⩽44. Following the extended classical over-the-barrier model we assume that the first step in the population mechanism is the same for both processes and that the two transferred electrons initially go to a double Rydberg state. We define total probabilities for radiative stabilization as P_rad=σ_DC/(σ_DC+σ_TI). These quantities represent products of the branching ratios for radiative decay of all the individual intermediate states in the decay processes averaged over the possible cascade paths. Within the experimental uncertainties there is no difference for P_rad measured for two-electron transfer from He, Ar, or Xe. The common q-behaviour of P_rad consists of a strong increase in the region 26 < q < 36 to a level of about 0.35, a "flat top" for 37 < q < 42, and a rather steep decrease for q=43 and 44. Apparently, a substantial fraction of the intermediate states have high branching ratios for radiative decay. Electron emission seems to occur late in the cascade when one of the electrons is close to the core. We offer a qualitative explanation for these observations and discuss the possible influence from configuration mixing with strongly asymmetric doubly excited states at high q.

We investigate the interaction of multiply charged ions with metallic surfaces. Simulations are presented displaying the formation of hollow atoms when slow highly charged ions approach the surface. It is shown that above surface neutralization proceeds via hollow-atom formation. Relaxation of the multiply excited states to the ground state occurs only subsequent to the close encounter with the topmost atomic layer. Recent experimental data on K Auger emission and image acceleration confirm this scenario.

A comparison is made of the electronic processes which occur when a multiply charged ion is approaching an atomic target on the one hand or a metal surface on the other hand. In both cases three collision phases can be identified: those of attraction, of electron capture and of decay in the vacuum; in case of surface collisions two more phases show up: the touch down on the surface and the decay inside the solid. Results from various groups, especially regarding the analysis of ejected electrons, are discussed to clarify similarities and differences between the different collision systems.

No abstract supplied

No abstract supplied

Based on the concepts of the classical barrier model and the analytical properties of the potential energy curves of the one-electron diatomic molecular ion, a direct physical description of slow ionizing collisions between highly charged ions and atoms is given. New expressions for the energies of the quasi-stationary states localized on the top of the potential barriers are used to derive simple expressions for the ionization cross sections in the limit of high charge states. A simple scaling in reduced variables is possible for the low energy ionization cross section in this limit.

Relativistic many-body perturbation theory and a qualitative estimation of non-perturbative effects are used to calculate low energy scattering and annihilation of a positron with noble gas atoms. The existence of a virtual level (resonance) and an annihilation enhancement factor that is produced by the strong e⁺ e⁻ correlation, is demonstrated for heavy atoms (e.g. Xe and Rn). For molecules, (e⁺ Xe_n, e⁺ Rn_n, n ⩾ 2) the virtual level can transform into a real bound level for the positron. These results can explain the very high annihilation rate and non-linear dependence on atomic density observed in measurements of positron decay rates in dense xenon gas. The absence of such effects in a e⁺ -Ne system completes the picture.

The logarithmic part of the Lamb shift, the contribution of the relative order α³ log (1/α) to the atomic state energy, is related to the usual infrared divergence. This fact allows one to calculate easily such corrections in positronium, and derive the recoil and electron-electron terms in the Lambshift Hamiltonian in many-electron atoms. Logarithmic energy corrections of the next order, α⁴ log (1/α), are of a different, relativistic nature. Their calculation is reduced to the ordinary perturbation theory for the nonrelativistic Schrödinger equation. The perturbation operators have the Breit-type structure and are found by the calculation of on-mass-shell diagrams. For positronium, the calculated log-arithmic correction survives only in n³S₁ states and constitutes (5/24)mα⁶ log (1/α)/n³. Logarithmic corrections of the relative order α² log (1/α) to the positronium decay rate are also of the relativistic origin and can be easily computed within the same approach. Arguments are presented in favour of a large numerical factor in the (α/π)² correction to the positronium decay rate.

This report describes some aspects of the use of animated interactive computer graphics for analysis of ion-atom collisions, as shortly presented at the Nobel Symposium 1992. The applications discussed in more detail are concerned with geometrical aspects of the collisions, i.e. alignment and orientation of the collision products, studied in the framework of the semiclassical theory of ion-atom collisions. One of the aims of this presentation is to show that these techniques are available without unreasonable additional expenses.

Contributions to the electron Lamb shift in highly charged ions are summarized. Recent theoretical developments as well as current experimental results are considered. Special emphasis is laid on higher-order vacuum polarization corrections as well as on the nuclear size effect on the electron self energy.