Universität Bielefeld, Fakultät für Physik,
Research Group "Physics of Atoms and Clusters"

Ion-impact induced multiple ionization and fragmentation of molecules and clusters


The ion-induced fragmentation of molecules and clusters is a process of fundamental importance in various areas of science and technology ranging from the physics and chemistry of upper planetary atmospheres to the understanding of radiation damage to biological tissue. Experiments in which all fragment ions emitted after a particular collision are detected in coincidence, can not only provide valuable information about the state of the intermediate multiply-charged molecular ion, they also shed light on the excitation and fragmentation dynamics. In our experiments we use a time- and position sensitive multi-particle detector to establish correlations between the charged fragments from a particular molecular break-up. In case of a Coulomb-fragmentation where all fragements emerge as positive ions this allows a kinematically complete study of the fragmentation process.



Dipl. Phys. Michael Ehrich
Dipl. Phys. Stephan Hülsmann
Prof. Dr. Nikolay M. Kabachnik
Dipl. Phys. Imke Küster
Dr. Axel Reinköster
Dr. Bärbel Siegmann
Dr. Udo Werner

Principle of experiment

Figure 1 shows the principle of the experiment for the Coulomb explosion of a triatomic system. Collimated beams of fast ions interact with a molecular gas target or a fullerene beam provided by an oven. The slow ions and electrons generated in the collision process are separated by a homogeneous electric field of 100-200V/cm. Electrons are detected in a channeltron (CEM) at one side of the interaction region; positive ions are accelerated towards the time- and position sensitive multi-particle detector at the other side. After passing a field-free time-of-flight region the ions are post-accelerated to a few keV to increase the detection efficiency.

The time- and position sensitive detector [1] is based on microchannel plates in combination with an etched crossed wire structure consisting of independent x- and y-wires. If an electron cloud from the plates hits at a crossing of two wires, coincident pulses on the wires will be generated and registered by the time-to-digital converter (TDC) which is the central part of the detector electronics. We use a special multi-hit TDC-module which was developed in our group. The system is located on a VMEbus-card and has 32 channels with a time-range of 17us and a typical resolution of 270ps. The TDC is triggered by an electron pulse from the channeltron and the individual channels are stopped by the ion-signals. Thereby, for each positive fragment the position on the detector and the time-of-flight relative to the start electron are recorded. Although there are position sensitive detectors with higher positional resolution our system has one major advantage: as a consequence of the crossed-wire structure the detector is capable to resolve particles which arrive `at the same time' on different wires. This `zero deadtime' feature is particularly useful for the study of the fragmentation of more complex molecules like CHtex2html_wrap_inline11 or even Ctex2html_wrap_inline13, where several correlated fragments with equal masses occur.

The present experimental setup is sensitive to all reaction channels resulting in at least one electron and one or more positive ions.

Coulomb fragmentation of water molecules

As an example we consider the fragmentation of water. This system is fairly simple in the following sense: it consists only of two kinds of atoms which can be easily distinguished due to their large mass difference, and only one atom, namely O, may occur in different charge states. The coincidence map (Fig. 2) gives an overview about the two-particle events detected in collisions of fast higly charged O-ions (provided from the ECR ion source at the KVI in Groningen) with water. In the case of water most channels can be separated and analyzed in great detail; in particular, cross sections for the correlated production of selected ions can be derived. Similar methods may be used for coincidences between three and more fragments, although there is no intuitively understandable graphic representation in higher dimensions.

Among the numerous reaction channels occuring in the collision processes we will concentrate on complete fragmentations of the type
displaymath7 .
In the experiment these events appear as 4-fold coincidences between an electron and the three positive fragment ions. If the time-of-flight and the position on the detector are recorded for each fragment from a particular process the conditions for a kinematically complete experiment are fulfilled. From the derived momentum vectors various parameters may be calculated which allow to analyze the dissociation dynamics. In this case besides the total kinetic energy release two independent angular correlations can be determined. These correlations give first insight into the fragmentation dynamics: e.g. they may be used as an indicator whether the molecular bonds break simultaneously or in a step-wise fashion. Our analysis shows that both OH-bonds break in a time short on a time scale defined by the rotation and vibration periods of the system [3,4].

A simultaneous break-up into positive fragment ions suggests the application of the simple Coulomb explosion (CE) model: as a first approximation the kinetic energies and emission angles may be computed by assuming Coulomb forces acting between point charges. In this picture (at least for the short collision times under consideration) the result of the calculation is independent from the details of the ionization process. Figure 3 shows the result of a simulation based on this model in comparison to measured kinetic energy distributions. The CE-model overestimates the energy release and the width of the distribution caused by the initial distribution of the ions as given by the water vibrational groundstate is smaller than that of the experimental spectra. Furthermore, the experimental data clearly depend on the projectile type. Several competing processes which all result in three positive fragment ions must be involved to explain this behavior.

To account for the most dominant reaction channels we used the MOLPRO code for an ab initio multi-configuration self-consistent field computation (MCSCF) of the lowest molecular states of the intermediate triply charged water ions. Figure 3 shows the weighted sum of of the nine energy distributions convoluted with the response function of the detector. A comparison of the measured energy spectra to the MCSCF-prediction shows reasonable agreement. The best agreement is achieved in collisions with highly charged ions: according to the classical over-barrier model excited states are expected to be less important in such `gentle' collisions which is in agreement with the experimental finding.

Figure 2: Coincidence map of correlated two-fragment events from collisions of 742keV tex2html_wrap_inline8 on water. TR and TL are the flight times of the fragments which hit the detector at the right- and leftmost position. The abundance of a certain coincidence is encoded in the colour of the corresponding point (increasing from blue to red). For example, a coincidence between an tex2html_wrap_inline12 on the right of the detector and a tex2html_wrap_inline14 on the right results in an event at TR~200ns and TL~560ns.

Figure 3: Total kinetic energy release of coincident tex2html_wrap_inline8 fragments from collisions of water molecules with 250 keV tex2html_wrap_inline12 and 92 keV tex2html_wrap_inline14[4]. The data are compared to a MCSCF- calculation (taking into account the nine lowest states of the intermediate triply charged water-ion) and to the prediction of a point charge Coulomb explosion model (CE).

Related Recent Publications

Udo Werner

Mon Jan 30 13:00:11 CEST 2006