Section 3: Interaction of atomic, molecular and cluster ions with matter
Research in this field in the last few years has resulted in astonishing progress in the understanding of the dynamics of particle – matter interaction. Key to these advances were the availability of high-performance ion storage rings, of swift ion and cluster accelerators producing intense beams with excellent optical quality and of ion sources providing highly charged slow ions. Moreover, novel developments in target preparation techniques (atomic traps [1], supersonic jets) along with the recent breakthrough in atomic many-particle imaging techniques (such as recoil-ion and low-energy electron momentum spectroscopy [2]) have opened up the opportunity to study the collision dynamics in unprecedented detail. The fundamental goal is to probe the response of matter under the influence of strong and short pulses of electromagnetic radiation. Concepts related to the non-linear response of complex systems can be tested. Examples include the study of slow highly charged ions and clusters, of sub-attosecond pulses, and of strongly coupled non-ideal plasmas. Targets of increasing complexity ranging from individual atoms in dilute media and clusters to surfaces and bulk matter can be investigated. The latter play a crucial role for the understanding of the first stages of material modifications discussed in section 4. Spectacular improvement has been also achieved in the high level of accuracy reached to test fundamental aspects in atomic structure (QED), using other methods than those described in section 2. There is also a strong impact on many other scientific areas. Among those are astrophysics and astrochemistry, atmospheric physics, radiobiology and radiation chemistry. We limit ourselves to a brief discussion of a few examples of important studies recently performed.
3.1Interaction with dilute media
3.1.1Electronic excitation and charge exchange processes
At heavy-ion accelerators (GANIL, UNILAC/SIS) and at the ESR storage ring facility the interaction of highly charged ions with matter can be explored in a unique regime that is governed by virtual as well as by real photon fields, advancing our basic knowledge about the physics of strong fields. At high velocities it has been demonstrated recently that the sub-attosecond (t < 10-18 s), super-intense (1020 W/cm2) electromagnetic field, generated by the moving ion, can be interpreted as a pulse of virtual photons (multiply) exciting or ionising the target [3]. In this limit, two-electron processes such as simultaneous excitation of two K-shell electrons in Lithium forming hollow atoms [4] or double and multiple ionisation reactions at minimum momentum transfer [5,6] provide a unique tool to investigate dynamic electron correlations on ultra-short times scales that have not been previously accessible.
At relativistic ion energies, the electromagnetic interaction becomes strongly affected by the magnetic field component significantly influencing the population of excited states [7]. Real photon fields are involved when a target electron is captured into a bound state of a fast highly charged ion with the simultaneous emission of a photon. Angle-resolved photon detection during this time-reversed photoionisation process is a novel and, presently, the only approach to investigate photoionisation of highly charged ions at hundred keV photon energies [8]. Here, spin-flip contributions mediated by the magnetic component of the electromagnetic field have been unambiguously identified for the first time (see figure 3.1.1).
In the last years, C60 fullerene molecules have become fashionable targets for collisions with highly charged ions at low velocities because they bridge the gap between atoms and surfaces (for a review see [9, 10] and ref. therein). As in the case of ion-surface interaction (see section 3.3.3), the formation of hollow atoms is observed. Also, an image charge (induced by the polarisability of C60 molecules) causes an acceleration of the collision
1
Fig. 3.1.1: Angular distributions for the time-reversed photoeffect, i.e. radiative electron capture, measured for bare uranium ions at 88 MeV/u in collision with N2 (full line: complete relativistic calculations, dashed line: sin2 distribution, shaded area: spin-flip contributions) [8].
1
partners during the approach. Unfragmented fullerenes with charge states up to q=10+ have been observed after collision with slow Xe25+ [10]. A recent experiment has shown that the number of electrons active during the collision is even much higher (up to 6 times) than the final fullerene charge. For close collisions where the ion penetrates the cluster, the fullerene cage is destroyed through ionisation and excitation. The small-fragment yield presents the same characteristic oscillatory behaviour with the projectileatomic number as the energy transfer. This demonstrates that energy deposition into the electronic degree of freedom is responsible for the destruction of the C60 structure. Detailed studies of the fragmentation mechanisms and decay times have been undertaken to determine the relative importance of evaporation and fission at various excitation energies.
3.1.2Dissociation and fragmentation of molecules and clusters
Collisionally excited states of the transient molecular ion, populated before the dissociation, have been determined for the CO molecule shedding light on limitations of the Coulomb explosion model to reproduce the fragmentation dynamics [11]. Geometrical modifications during the break-up, time-sequencing of different processes and many-body Coulomb interactions have been explored [12] for CO2 and H2 [13], respectively. Of especial interest for radiobiology, fragmentation of H2O induced by highly charged ion collisions has been investigated [14]. Striking results have been obtained in the investigation of cluster-atom collisions.There, the use of clusters as projectiles offers many technical advantages.For instance, the fragmentation of hydrogen cluster ions of high kinetic energy colliding on helium atomsprovides information on phase transitionsin finite systems [15].Evidence is also obtained on the sensitivity of the multi-ionisation cross sections to the shapeof the clusters [16] as well as on the unexpectedly strong size dependence of non-dissociative rates in electron capture collisions [17]. Future studies will concentrate on the role of the cluster-electron polarisability and the internal vibrational energy.
3.1.3Collisions with trapped atoms
Atoms trapped by means of electric, magnetic and optical fields offer novel exciting possibilities for atomic collisions physics for various applications as well as for nuclear physics research. Recently, state-selective charge transfer measurements have been achieved down to 5eV/amu where experimental cross sections are found to be orders of magnitude higher than theoretical predictions [18]. Presently, a magneto-optical trap is implemented into the Heidelberg test storage ring TSR and experiments are in preparation to use this ultra-cold, K target as a beam profile monitor and for recoil-ion spectroscopy. First encouraging results, obtained at Groningen, Aarhus and at KSU, demonstrate that an unprecedented recoil-ion momentum resolution will be achievable with these instruments. Fascinating perspectives are envisaged using highly charged recoil-ions, emerging from collisions with fast ions, as ultra-low-energy secondary projectiles for collision processes at astrophysically relevant collision energies in the sub-meV range.
3.1.4Perspectives and technical developments
A major effort is dedicated to precisely prepare, control and, in many cases, to efficiently coolhighly charged atomic or clusters ions as well as neutral atoms and molecules. The aim is, on one hand, to reach ultimate precision in the investigation of atomic structure and collision dynamics. On the other hand, reliable data as well as new information will be provided for other fields of physics where the knowledge of fundamental collision processes is needed (nuclear physics, plasma physics, astro- and biophysics). At the upcoming TESLA-FEL (Tera Electron Volt Energy Superconducting Linear Accelerator- Free Electron Laser) in Hamburg, basic non-linear atomic processes in the interaction of this super-intense, short-pulse radiation with surfaces, bio-molecules, magnetic materials and condensed matter will be studied. First differential measurements on multiple ionisation of atoms and molecules in intense femtosecond laser fields open the door to such future experiments [19].
3.2Interactions with free electrons and plasmas.
Since Danared [20] first employed adiabatic expansion to lower the transverse temperature of electrons in coolers, collisions of atomic and molecular ions with free electrons of temperature around 1-10 meV are extensively explored in storage rings. The quality (density and energy spread) of the electron source is essential. New states of atoms, molecules and clusters (positively or negatively charged) are formed. Their study provides sensitive tests of refined theoretical predictions. These studies are related to important issues in astrophysics, chemistry and atmospheric physics. Free electron-ion collisions are also of prime importance in plasmas with potential applications related to Heavy Ion Inertial Fusion.
3.2.1Studies with positive atomic ions
The increased resolution, obtained in recent years through improved cooling devices, has made dielectronic recombination an excellent tool for spectroscopic studies to improve our understanding of formation of atoms and the binding of their electrons. Our knowledge about atomic structure and Quantum Electrodynamical (QED) corrections to energy levels is almost exclusively obtained through photon spectroscopy (see section 2). Although it would hardly have been anticipated a few years ago, resonances in collision cross sections provide today an alternative access to highly accurate information on atomic energy levels. Several studies have been performed with highly charged Li-like ions and an energy splitting in Cu-like Pb, carrying a large QED correction, was determined with relative accuracy of ~10-5 [21]. Spectroscopic studies with accuracy in the measurements of energy splittings in the order of ~10-6 are envisaged for the future.
Ion-electron recombination studies also provide information on dynamical aspects of the collision process, when varying the electron energy. As an example, we show in Fig 3.2.1 the recombination spectrum of F6+ [22,23].
1
Fig.3.2.1. Low-energy electron-ion recombination rate coefficient measured at the TSR. (a) Result for F6+ ions, showing narrow dielectronic resonances. (b) Result for C6+ ions, due to radiative recombination only. The shaded area indicates theoretically expected recombination rates, for F6+ also including contributions from DR resonances (lines) [23]. represents an unexpected enhanced rate of recombination.
1
This work addresses in particular the long-standing question of enhanced radiative recombination (RR) at low energies [23, 24, 25]. Possible explanations include an enhancement of the electron density in the ion vicinity originating both from the external magnetic field and from the Coulomb field induced by the ions [23].
In the scenario of heavy ion inertial fusion, the energy deposited in a target by swift heavy ions is mainly converted into X-radiation. The efficiency of this X-ray converter depends on the spatial and time evolution of the beam energy deposition profile. When increasing the target temperature of low Z materials, one observes a large increase of the average ionization state of the ions. The stopping power and the energy straggling also undergo a dramatic increase. This feature is related to the fact that atoms of low Z targets are efficiently ionised at increasing temperature and thus form a plasma. The charge state evolution and the energy loss of heavy ions passing through a plasma show pronounced differences when compared to the passage through cold gases and solid state matter [26]. These results demonstrate that charge and energy fluctuations must be considered to predict correctly the profile of energy deposition.
Intense ion beams open new opportunities to investigate the interaction with dense plasmas and to study the hydrodynamic and radiative properties of beam heated matter. The first experiments made use of externally created plasmas. Today, dense plasmas in the temperature range of up to several 10 eV can be created directly by intense swift heavy ion beams (at GSI for instance) irradiating a light target (hydrogen). In this temperature and density regime, the potential energy of the particles, constituting the plasma, is of the order of their thermal energy. In these strongly coupled non-ideal plasmas, ion beam induced effects, such as hydrodynamics instabilities or "metallisation" of hydrogen, are evident [27]. Plasma temperatures of up to 300 eV and higher, where highly charged species prevail, can be explored in beam plasma interaction experiments with laser driven plasma targets [28]. Future developments will involve a petawatt high-energy laser for ion experiments (PHELIX) which is currently under construction at the accelerator facilities of GSI.
3.2.2Studies of positive molecular ions
Significant progress has been made in studies of molecular cations in storage rings. During the long storage time (many seconds), these molecules may cool down to the lowest vibrational state by slow infrared radiative transitions. In this way, well-defined molecular states can be studied[29].
When a positive molecular ion captures a free electron by exciting one of its bound electrons, it may respond by emitting an electron by autoionisation. However, the resulting electronic change normally yield a neutral molecule in an unstable, dissociating state which may lead to the formation of a number of neutral atomic and molecular fragments. In this case, the process is termed dissociative recombination, a process of great importance but very hard to treat theoretically.
Three important issues should be mentioned here (see [29]). One concerns to H3+, a species that plays an important role in interstellar chemistry. The rate coefficient for dissociative recombination of this ion was measured at the CRYRING storage ring where also the branching ratios (H versus H2 production) were determined. At the ASTRID storage ring, an imaging technique was applied to measure the kinetic energy release of the atomic fragments in the dissociative recombination of O2+. The dissociative recombination leads to a substantial population of oxygen atoms in the metastable 1S state. The resulting 1S 1D transition is responsible for the green light emission of the ionosphere of the Earth. In a recent work at the TSR storage ring, Coulomb explosion together with the fragment imaging technique was used to obtain rate coefficients for individual initial vibrational levels (=0,1,2,..) of the HD+() ion. This work provides a very critical test for the understanding of the dissociation dynamics of small molecules. Recently, the first studies of dissociative recombination of twofold positively charged molecular ions, where the dissociation follows repulsive curves of the mono-cation, have also been carried out [30].
3.2.3Collisions involving negative ions
Most atoms and molecules form stable negative ions and many of these ions play an important role in man-made plasmas and in nature (e.g. in the interstellar space and in planetary atmospheres). The electron-impact detachment cross section of negative ions has been investigated at the ASTRID storage ring.
1
Fig. 3.2.2. Electron-impact detachment cross sections of NO2- as a function of the electron energy in the ion rest frame obtained at the ASTRID storage ring [31]. Two dianion states of positive energy are revealed.
1
The aim is to understand the dynamics of the detachment process and to study the possible formation of small doubly charged negative ions (dianions), a new form of ions in the gas phase where electronic correlation is expected to be extremely important. It appears that atomic dianions do not exist. However, many small molecular ions form highly unstable dianions. This formation is revealed by the presence ofresonances in the scattering cross sections. Figure 3.2.2 shows an example with NO2- where two resonances are visible [31]. Ab initiocalculations show that the lower one corresponds to the dianion ground state.
3.2.4Perspectives
The magnetically confining heavy-ion storage rings are widely used for atomic and molecular ions where the mass to charge ratio is not too big. However, very heavy molecular and cluster ions cannot be stored at sufficiently high energy to match the speed of the electrons of the electron cooler in the merged beams configuration of the rings. Crossed-beam setups are being prepared for studies of electron interactions with gas-phase bio-molecular ions and cluster ions in new electrostatic storage rings [32]. Newly developed electrostatic traps have also been used to investigate molecular or cluster ions [33].
3.3Interactions with condensed matter: bulk and surfaces
Apart from quantities of immediate practical relevance such as mean charge states and stopping power of the projectile, an improved microscopic understanding of the complex array of interaction processes of ions with solids has been the focus of recent investigations. In addition to their own fundamental interest, these studies have impact on many different subfields in physics. For example, the knowledge of the projectile electronic state in matter permits to optimise the injection of high-intensity beams for radioactive beam production or for spallation neutron sources and to better control beam-induced material modification. The latter includes track formation and flux pinning in high-temperature superconductors, selective sputtering, and fundamental concepts of the non-linear response of matter to strong fields (see section 4).
3.3.1High velocity ions inside solids
As a swift ion penetrates a solid, it undergoes a large number of collisions with ionic cores as well as valence and conduction band electrons (“electron gas”) leaving behind a track of electronic excitation while simultaneously changing its charge state and state of excitation.
While the basic properties of the stopping power at high energy are well described by the perturbative theories of Bethe and Bohr, recent studies with highly charged ions, at GANIL (Caen) and at the Van de Graaff CN accelerator of Legnaro National Laboratories (Catania), have added new features. Measurements of absolute electron emission cross sections have been performed with large detector arrays (ARGOS), initially designed for nuclear physics studies. One of them is the production of ultrahot electrons [34] in the target with energies beyond the classical binary encounter limit in both forward and backward directions. They have been identified as a signature of the so-called shuttle acceleration, originally proposed by Fermi to explain high-energy cosmic rays and later invoked for the acceleration of ionised electrons and atoms in particle - solid collisions [35].