EP1200984A2 - Method and device for cluster fragmentation - Google Patents

Abstract

A method for cluster fragmentation comprises the production of at least one cluster which contains a carrier substance and the fragmentation of the cluster into cluster fragments, with the cluster being loaded before the fragmentation with at least one reaction partner and the reaction partner being part of at least one cluster fragment after the fragmentation. A cluster beam system for performing the method, and applications of the cluster fragmentation for analysis and purification of surfaces, for analysis of clusters, and for the operation of ion thrusters are also described.

Description

 Method and device for cluster fragmentation

The invention relates to a method for cluster fragmentation, in particular a cluster fragmentation method for generating electrically unevenly charged particles and / or for manipulating electrically neutral particles, and devices for cluster fragmentation. The invention also relates to applications of cluster fragmentation for substance analysis at interfaces, for cleaning surfaces, and in the construction of ion sources and / or ion engines, and to applications in which clusters (or aerosols), in particular those of natural origin, on their quantity or composition should be analyzed.

The influence on or the detection of electrically neutral particles is associated with a relatively high technical outlay because of their weak interaction with the environment. The Coulomb interaction of electrically charged particles, on the other hand, allows simple manipulation using electromagnetic fields and also simplified detection, for example by direct electrometric measurement. There is therefore an interest in the conversion of electrically neutral atoms, molecules or corresponding groups of atoms or molecules into corresponding charged particles (ionization). In general, the transition from the electrically neutral to the charged particle takes place by adding at least one charge carrier, for example an electron, to a neutral particle or by removing charge carriers, so that a net charge remains on the originally neutral particle. The main well-known ionization techniques include electron impact ionization, laser ionization, electron attachment, and plasma ionization. In the known methods for generating positive ions, a one-stage ionization is generally carried out, in which the energy supplied to the neutral particle practically instantaneously is large enough to completely separate at least one electron from the resulting cation. The separated electrons are normally no longer used, so that only one relevant charge carrier can be generated for each ionization energy unit used.

A general problem with conventional ionization is the quantitative conversion of neutral particles into corresponding ions. The desired degree of ionization (ratio of the number of ionized particles to the original number of neutral particles) is achieved as close to one as possible only with great technical effort. Ionization is often associated with the destruction of the original neutral particles. Conventional ionization techniques are limited to generating light ions (charged molecules or groups of molecules). In various areas of application, e.g. in surface treatment and in the operation of ion engines, however, there is an interest in the generation of particularly many and particularly heavy ions.

Not only is the influencing and detection of electrically neutral particles associated with technical difficulties, but also their transfer into the gas phase: especially with larger molecular groups, e.g. biologically relevant macromolecules or DNA fragments, the interaction with the carrier material or the surrounding solvent is so strong that intramolecular bonds are broken when an attempt is made to remove or remove it and the transfer in the gas phase is accompanied by destruction of the starting substance.

As a rule, the transfer process also heats the molecule very strongly (excitation of rotation, vibration and electronic degrees of freedom). In the gas phase, the molecule has no efficient way to dissipate this excess energy (lack of coupling to a warm bath). As a result, molecular bonds can break or denaturation can occur. Spectroscopic analysis is also hampered by the high level of excitation. The gentle transfer of large molecules into the gas phase is of technical importance, for example as the first step in a mass spectrometric analysis.

A well-known method for the transfer of large molecules into the gas phase is the MALDI process (matrix assisted laser distortion ionizaton) (e.g. U.S. Patent 58 28 063). The cost of the laser required for this, however, severely limits the application.

The generation of atomic or molecular groups in the form of so-called clusters is generally known. Clusters are e.g. because of their special material properties that differ from their solid state as well as manipulable particles. of interest when modifying or cleaning surfaces. By W. Skinner et al. ("Vacuum Solutions", March / April 1999, page 29 ff) describes, for example, applications of ionized clusters of gas atoms in surface processing.

A known method for producing ionized particles is given by the cluster fragmentation of water and sulfur dioxide clusters, which, however, has so far only been of theoretical importance for the reasons given below. For example, AA Vostrikov et al. in "Chemical Physics Letters" volume 139, 1977, page 124 ff, m "Z. Phys. D", volume 20, 1991, page 61 ff, and in "Z. Phys. D", volume 40, 1997, page 542 ff, the ionization of water clusters when hitting solid surfaces. Furthermore, from the publication by Wolfgang Christen, Karl-Ludwig Kompa, Hartmut Schröder and Heinrich Stulpnagel in "Ber. Bunsenges. Phys. Chem.", Volume 96, 1992, page 1197 ff, the ionization of S0 ₂ clusters with mechanical scattering on single crystal surfaces. The formation of ionized cluster fragments upon impact of H ₂ 0 clusters on surfaces is explained by the autoprotection of the water according to H ₂ 0 - H ⁺ + OH ^~ . The ions H or OH ^{~ present} in different particles of the cluster at the time of the impact are separated from one another by the fragmentation and carried with different cluster fragments, which are then electrically charged to the outside.

The ionization by cluster fragmentation has so far been of no practical importance, since it is restricted to H ₂ 0 or S0 ₂ and has an extremely low efficiency. Under normal conditions in water, only every 10 ^9th particle is ionized. Accordingly, the probability of generating loaded cluster fragments is extremely low. Further studies on the cluster fragmentation of H ₂ 0 (see publication by PU Andersson et al. M "Z. Phys. D", volume 41, 1997, page 57 ff) are on the influence of electron transfer from the surface hit by a cluster in the Cluster and directed to the associated ionization of the cluster fragments.

Studies of the electronic properties of metal atom-doped clusters are also known. For example, R. Takaso et al. in "J. Phys. Chem. A", Volume 101, 1997, page 3078 ff and by IV Hertel et al. in "Phys. Rev. Lett.", Volume 67, 1991, page 1767 ff, an electron delocalization for alkali atoms in molecular clusters. Furthermore, from the publications by RN Barnett et al. in "Phys. Rev. Lett.", volume 70, 1993, page 1775 ff, KS Kim et al. in "Phys. Rev. Lett.", volume 76, 1996, page 956 ff and by D. Feller et al. in "J. Chem. Phys.", Volume 100, 1994, page 4981 ff, describes the behavior of sodium in H ₂ O or NH ₃ clusters. It was found that sodium in dissolved form S in the cluster caused a reduced ionization potential. Practical applications have not yet been derived from this. The previous studies of the electronic properties, for example of sodium m clusters, have been carried out in long-lived equilibrium states, which, however, have so far not allowed any statements to be made regarding the dynamics of the behavior of charge carriers in clusters.

Generation of charge carrier pairs by alkali atoms in clusters from water, ammonia and acetonitrile has also been described (see CP Schulz et al. In "Clusters of atoms and molecules II", ed. H. Haberland, Springer 1984, p. 7 -11).

It is the object of the invention to provide an improved cluster fragmentation method for generating charged particles and / or for manipulating electrically neutral particles, which can be used in particular with an expanded range of substances and has an increased and controllable efficiency. The object of the invention is also to provide devices for implementing such a method. The object of the invention also consists in the description of novel application possibilities for charged or unloaded cluster fragments which have been generated with the improved cluster fragmentation method.

These tasks are solved by the subject matter of claims 1, 21 and 29 respectively. Advantageous embodiments and further applications of the invention result from the dependent claims.

The basic idea of the invention is to further develop conventional cluster fragmentation methods in such a way that before the actual fragmentation, e.g. B. by mechanical impact of a cluster on an interface, this is loaded with a reaction partner. The reaction & partner consists of individual atoms or molecules, atomic or molecular groups or it is itself a cluster or cluster fragment.

Clusters here are generally relatively weakly bound associations of atoms or molecules or atomic or molecular aggregates due to purely physical forces (e.g. van der Waals forces or H-bridge bonds), whose internal volume density is comparable to the density of solids, which, however, have the character of a gas phase particle. The (average) cluster size is set depending on the application and can range from a few particles (e.g. around 10) to large particle numbers (e.g. one or more 1000). The clusters can also be as large as macroscopic aerosol particles.

According to one embodiment of the invention, the reaction partner consists of electrically neutral molecules which can be absorbed into the cluster fragments through physical interaction with the carrier substance.

According to a further embodiment of the invention, the reaction partner has the ability to generate a pair of electrically unevenly charged charge carriers with the particles of the cluster material (carrier substance). During the induced fragmentation of the cluster, these generated charge carriers can come to rest on different fragments of the bursting cluster and can be spatially separated by the inertial movement of the cluster fragments. In contrast to the original cluster, in which the charge carriers mutually neutralized each other, the spatial separation of the fragments and thus the individual charge carriers means that the mutual shielding no longer exists, so that the separated, charged cluster fragments form electrically charged, free particles to the outside, which are also referred to below as ions. Instead of distributing the generated charge carriers For different fragments, if the process is carried out appropriately, the exclusive generation of positively charged fragments can also be provided, while the negative landing carriers flow off to the respective border area.

The charge carrier pair is generated spontaneously by a chemical reaction or ionization of the reaction partner or alternatively by external excitation, for example by inducing charge carrier transfer through light irradiation or mechanical shock. The probability that the charge carriers are on different fragments can be influenced by the choice of cluster size, cluster speed and fragmentation conditions. In general, the likelihood increases if charge carriers are formed as reaction products that have a high mobility within the cluster (e.g. electrons or protons in hydrogen-bonded clusters), because in this case there is a spatial distance already within the cluster.

In a preferred embodiment, the cluster fragments are simultaneously ionized according to one of the methods according to the invention.

The carrier substance by which the clusters are formed preferably consists of polar molecules, ie of molecules that have their own dipole moment, for example H ₂ 0, S0 ₂ , N0 ₂ , NH ₃ , N0 ₂ , SF _n , CH ₃ CN, CHCIF2, or isobutene. The polar molecules have the advantage of weakening the Coulomb interaction of ions in the cluster. In addition, a polar environment generally requires ionic reactions to occur. Furthermore, the stronger dipole interaction of the molecules facilitates the uptake of reactants. The carrier substance has a different chemical composition than the one or more reactants. The reaction partner is loaded into the cluster to be fragmented during cluster generation, in the gas phase or at the interface immediately before fragmentation. For this purpose, atoms or molecules or groups of atoms or molecules are applied via the gas phase to the cluster or clusters or to a surface arranged for cluster fragmentation. The reaction partner preferably consists of a substance which reacts with the carrier substance of the cluster to produce the pair of charge carriers. In the case of polar carrier molecules, a substance with low ionization energy, for example below 10 eV, in particular alkali atoms such as lithium, sodium, potassium and cesium, is preferably chosen as the reaction partner. The use of substances with such a low ionization potential has the advantage that electron donation occurs spontaneously within the cluster of polar molecules. The charge carriers created with "high" efficiency can be separated efficiently by the cluster fragmentation method according to the invention. The method according to the invention can also be implemented depending on the application, in particular depending on the average cluster mass, the average cluster speed and the strength of the dipole moment of the carrier substance molecules, with other reaction partners.

The cluster fragmentation generally takes place through an energy supply. In the case of a mechanical energy supply, one or more clusters collide with a predetermined speed or speed distribution with an interface, which represents a transition between the gas phase and a solid or the gas phase and a liquid. The boundary surface can have any shape geometrically and in many applications is preferably formed by a solid substrate surface which adjoins a space in which the clusters are generated or accelerated. This has the advantage that simply by arranging the interface m an interaction with the path traversed by the cluster 3

Surface is ensured. This means that every cluster with probability 1 will hit the surface and be fragmented. The interface generally does not have to be stationary. It can be particularly advantageous to increase or decrease the relative speed between the cluster and the interface with the aid of a moving interface, in order to influence the fragmentation behavior of the cluster. In addition, it is also possible for the interface to be formed by small droplets or by clusters in the gas phase.

Alternatively, a radiation energy supply for cluster fragmentation can be provided, for example by subjecting molecules in the cluster to excitation of electronic states or vibrational states by laser irradiation.

Preferred applications of the method according to the invention are in the modification, cleaning or analysis of solid surfaces, in the analysis of clusters and aerosol particles with regard to quantity and composition, and in the provision of ion sources for measurement or analysis purposes or also for ion engines. According to a further application of the invention, it is provided that the cluster fragmentation method is used to manipulate molecules which are neutral per se, in that the molecules to be manipulated, like the reaction partner, are taken up by the cluster and transferred to the cluster fragments before the cluster fragmentation. Through the transfer into cluster fragments, molecules are transferred into the gas phase, possibly ionized in the course of the transfer by one of the methods according to the invention, and thus made accessible to manipulation or measurement known per se.

According to a further aspect of the invention, a device for implementing the aforementioned cluster fragmentation method in the form of a cluster beam system is described A ben. This device is characterized in particular by a

Cluster generation device and a cluster fragmentation direction as well as control, steering and measuring devices for the cluster fragments. The cluster generation device comprises a cluster source known per se. The cluster fragmentation direction is designed to allow the cluster or the clusters provided by the cluster generation device to strike an application-dependent boundary surface.

The invention has the following advantages. In contrast to conventional ionization processes, charge carrier generation takes place in two stages. First of all, the cluster loading with the reaction partner forms an externally neutral cat ion / amone pair, which is then separated by the cluster fragmentation. A large number of efficient chemical reactions are already available for the formation of the cation / anion pair. Another advantage is that the energy required to generate the charge carrier pairs is considerably smaller than the energy required to generate the corresponding emissions. The energy difference results from the mutual stabilization of the cation / anion pairs in the cluster through the Coulomb interaction. This stabilization is only released by the fragmentation of the cluster, the energy for overcoming the mutual Coulomb attraction coming from the kinetic energy of the cluster fragments. The required ionization energy is thus supplied in two stages or parts according to the invention.

This two-stage approach enables the use of various forms of energy, which differ in particular in terms of the cost or effort involved in providing the respective energy. For example, part of the ionization energy can be replaced by an "expensive" energy package (eg a laser photon) and a more part can be provided by a "cheaper" energy package (eg kinetic energy).

When the charge carrier pairs are embedded in the cluster, the Coulomb interaction is considerably reduced due to the dielectric influence of the cluster edium (carrier substance). In contrast to the gas phase, there is already the possibility of spatial charge carrier separation in the cluster, which considerably reduces the amount of energy required to generate free charge carriers.

An important advantage over conventional ionization processes is that the process involves the formation of equal amounts of positive and negative charge carriers for each cluster fragmentation. High charge carrier densities in the form of a cat ion / an ion plasma can be generated, which can lie far above the space charge limited density of charge carriers of one polarity.

Loading a cluster with a reaction partner has the advantage that in the cluster in a predetermined manner e.g. only a few, predictable in number of charge carrier pairs are generated. Since the energy for separating the charge carrier pairs is determined by the kinetic energy of the incident clusters before fragmentation, a relationship between the amount of maximum free charge carriers that can be generated and the original kinetic energy of the cluster is defined for a given mean cluster size. When the cluster is loaded with the reaction partner, the amount of charge carrier pairs generated per cluster can be adapted to the kinetic energy of the cluster.

The cluster fragmentation according to the invention provides an ionization process which is distinguished by a high efficiency and by the possibility of the masses of the ionized particles (ion masses) in a wide range depending on the application to vary. According to the current state of knowledge, around 5% of the fragments loaded on clusters hitting a solid surface are typically broken down. This represents a high value in comparison with the conventional ionization processes. Furthermore, ion masses of typically up to a few thousand atomic mass units can be provided. This is particularly important for the operation of ion engines.

In connection with the absorption of the reaction partner from an interface, the cluster fragmentation according to the invention enables the gentle transfer of even large molecules into the gas phase. The recording of the cluster and the transfer into a cluster fragment has the advantage that breakage of intramolecular bonds is avoided. Excess excitation energy can be released from the absorbed molecule to the surrounding cluster fragment, so that very cold, easily spectroscopic molecules are transferred into the gas phase. The energy contained in the cluster fragment can be sufficient to completely vaporize the weakly bound carrier gas molecules of the cluster fragment. In this case, the method has the advantage of transferring the absorbed molecules into the gas phase without a surrounding cluster shell.

A particular advantage of the method is that, simultaneously with the transfer of a reaction partner (e.g. large molecule), its electrical charge can be brought about by one of the processes according to the invention. In this case, the reaction partner can be fed directly to an electromagnetic analysis method.

The cluster fragmentation method according to the invention can also be used particularly advantageously for the quantification and analysis of clusters and aerosol particles. In particular, the particles to be examined can be aerosol particles of natural origin, such as those of about Earth atmosphere occur. The majority of them contain water and other polar molecules, so that they can be converted into ionized fragments in a particularly simple manner, for example by impact with a surface covered with alkali metal. These ions can be supplied, for example, to a charge quantity measurement in order to determine their concentration in the air volume under investigation and / or a mass spectrometric analysis to clarify their composition. An aerosol fragmentation can be examined directly on board a measuring river body (eg an airplane) using the relative speed between the missile and the aerosol.

Further details and advantages of the invention will be described with reference to the accompanying drawings. Show it:

1 shows an illustration of the charge carrier separation in a cluster fragmentation according to the invention;

2 shows an illustration of the cluster loading at an interface area occupied by reaction partners;

3 shows an application of the cluster fragmentation according to the invention for surface analysis;

4 shows an application of the cluster fragmentation according to the invention for surface cleaning;

5 shows an illustration of the recording of neutral surface adsorbates;

6 shows a first embodiment of a cluster fragmentation device according to the invention, which is designed for the analysis of surface adsorbates; FIG. 7 shows curves to illustrate measurement results obtained with a device according to FIG. 6;

8 shows a further exemplary embodiment of a cluster fragmentation device according to the invention in the form of an ion engine, and

Fig. 9 graphs to illustrate further measurement results.

The invention is explained below by way of example with reference to the collision of clusters with solid, flat substrate surfaces. The invention can be used in a corresponding manner in the event of collisions at gas phase / flux boundary surfaces and / or differently shaped boundary surfaces or with radiation-reduced fragmentation. The figures only show schematic, enlarged illustrations of clusters and cluster fragments, while dimensions and compositions are selected depending on the application and the principles explained below.

FIG. 1 illustrates the principles of a cluster fragmentation according to the invention in accordance with a first embodiment of the invention. The left part of FIG. 1 shows the initial situation of a cluster 2 moving towards the target 1 at a predetermined average speed. The target 1 forms the interface for fragmentation compared to the reaction space that the cluster is moving. Cluster 2 consists of a specific carrier substance, which preferably contains at least partially molecules with a permanent molecular dipole moment. Cluster 2 is loaded with a reaction partner (not shown), which has entered into a chemical reaction with the carrier substance, the result of which has been generated by a pair of charge carriers with different signs (anions 3, cations 4). According to the invention, the cluster to be fragmented is loaded with the reaction partner before fragmentation. Depending on the application, this can already be done when the cluster is formed. The reaction partner can in particular consist of the same material as the carrier substance of the cluster, ie the starting materials involved in the reaction can be components of the cluster itself. Alternatively, the loading takes place during the movement of the cluster to the interface. Finally, it is also possible that the loading only takes place at the interface itself (see FIG. 2).

Cluster 2 consists, for example, of S0 molecules and is loaded with a Na atom. The loading was carried out by collision of a cluster beam with a sodium atom beam or a sodium vapor. The reaction between the carrier substance sulfur dioxide and the reactant sodium consists in the spontaneous release of an electron from the sodium to the surrounding SO ₂ molecules with the formation of the sulfur dioxide anion 3 and the sodium cation 4. Sulfur dioxide is the carrier substance for the cluster of the following Reasons preferred. It is chemically stable, shows no hydrogen bonds or auto-dissociation, and has a relatively high electron affinity (EA) of around 1 eV. This high EA value facilitates the formation of stable anion clusters. Another advantage of sulfur dioxide is that clusters can be easily generated from this carrier substance at room temperature (see below). In the left part of FIG. 1, the cluster 2 still represents an externally neutral particle after the charge separation, since the inner charges are of the same size in opposite directions.

The movement (to the right in FIG. 1) of the cluster 2 leads to a collision (not shown) with the target 1, as a result of which the cluster disintegrates into fragments 5, 6 and 7, which collide against the rigid boundary A 6 Move area to the left. The right part of FIG. 1 shows the situation after the collision between cluster 2 and target 1. The cluster fragments 5, 6 and 7 move away from the interface, the cation 4 and the anion 3 being on different fragments 5 and 6, respectively. The inertial movement of cluster fragments 5 and 6 overcomes the mutual Coulomb attraction. After the fragmentation, the mutual shielding of the charge carriers 3, 4 ceases, so that the fragments 5, 6 form two free particles which are charged to the outside and are available for further use (see below).

FIG. 2 illustrates a modified embodiment of the invention, in which the cluster is loaded with the reaction partner only when it collides with the boundary surface. According to the left part of FIG. 2, the cluster 10 moves towards the surface of the target 1, which, for example, consists of gold and carries adsorbates 11 on its surface, which represent the reaction partners for charge carrier separation in the cluster. The substrate surface is covered by a reactant supply device 12, which is formed, for example, by an evaporation furnace. According to the invention, it can be provided that adsorbates are continuously fed onto the substrate surface by the reactant supply device 12 in order to replace adsorbates worn away in previous clusters of the surface and thus to maintain a surface coverage that is constant over time. This provides an ion source that works continuously during cluster bombardment.

During the collision, not shown, of the cluster 10 with the adsorbate-covered surface of the target 1, the cluster 10 picks up at least one adsorbate atom or molecule from the target 1. The atom or molecule is dissolved as a reaction partner in the carrier substance of the cluster. It happens in the cluster directly to the chemical reaction between the reactant taken up and at least one cluster component, which leads to ionic products (charge carrier separation). After the cluster 10 collides with the adsorbate-covered surface of the substrate 1 (see FIG. 2, right part), the cluster fragments 13, 14 and 15 formed by the interaction of the cluster at the interface are removed from the surface of the substrate 1 The absorption of the adsorbate 11 in the cluster 10 resulting ionic products are on different fragments 13, 14 and are separated from one another. Again, the energy required to overcome the mutual Coulomb attraction is applied by the inertial movement of the cluster fragments. The cluster fragments 13, 14 form externally charged, free particles which are available for further use.

The process illustrated in FIG. 2 is the basis for various applications of the cluster fragmentation according to the invention. By irradiating the target surface with a cluster beam while continuously supplying adsorbate, a continuously working ion source is formed, for example. Instead of a flat target 1, it is also possible to provide a substrate which is formed in a manner dependent on the application with predetermined margins in the manner of a mask and which forms a local ion source m t of certain geometric properties when irradiated with clusters. Alternatively, adsorbates can be picked up from the surface and subjected to an analysis using the method. For this purpose, the charged fragments are transferred using electrical fields, for example a mass spectrometer.

2 simultaneously illustrates the use of the cluster fragmentation according to the invention for the quantification and analysis of clusters and aerosol particles, in particular of natural origin. In this case, the incident particle 10 constitutes a cluster or an aerosol particle with possibly unknown ter composition, which was transferred by suitable devices from a sample room into the cluster fragmentation room. Before the fragmentation, the reaction partner is fed to the cluster or aerosol particle 10, which leads to the formation of charge carrier pairs in the cluster or aerosol particle 10. As shown, the supply can take place in a bump with a surface covered with reaction partners. Since aerosol particles of natural origin contain a high proportion of water and other polar molecules, an alkali metal atom is particularly advantageous as a reaction partner, since the alkali atom of a polar environment spontaneously releases its valence electron to form an alkali metal cation. Likewise, all atoms with a low ionization energy below 10 eV are suitable, especially representatives of the 3rd main group. The charged fragments released by means of the cluster fragmentation can be fed to a charge quantity determination in order to determine the concentration of the original clusters / aerosols in the sample volume and / or a mass spectrometric analysis to determine the composition of the starting clusters or aerosols.

A special and unexpected aspect of the invention is to be seen in the fact that for the charge separation illustrated in FIG. 2 after loading with the reaction partner, only a very short time window m of the order of 1 ps or less is available during the collision with the interface. This short time is sufficient to achieve sufficient separation of the delocalized charge carriers.

An alternative application of the principle shown in FIG. 2 is explained below with reference to the analysis of surfaces illustrated in FIG. 3. The target is formed by the substrate 21. The substrate 21 is made of silicon, for example. 3 (beekeeping part) there are adsorbates, for example in the form of sub-monolayers of electrically neutral alkali metal, on the surface of the substrate 21 to be analyzed. lime metal atoms (e.g. Li, K, Na or Cs). The movement of the cluster 20 leads to a collision with the surface, the cluster 20 receiving an alkali metal atom 22. Again, the alkali metal spontaneously releases a valence electron to the cluster environment through the interaction with the polar SO ₂ molecules, an alkali cation 23 and a sulfur dioxide anion 24 being formed. The situation after the collision, not shown, is shown in the right part of FIG. 3. The cluster fragments 25, 26 and 27 formed as a result of the cluster fragmentation move away from the surface of the substrate 21, the charge carriers 23, 24 separated in the original cluster 20 being located on different fragments 26, 27 and moving away from one another. As with the examples explained above, the Coulomb attraction is compensated for by the inertial movement of the ionized cluster fragments.

The free ions 26, 27 obtained after the spatial separation can be analyzed in a mass spectrometer in order to determine the composition of the recorded surface adsorbates.

A particular advantage of the invention is that the analysis of surface absorbates can be extended to a large number of elements. In general, all elements can be detected that have a sufficiently low ionization energy. Elements with ionization energies below 6.5 eV are preferably detected. In addition to the alkali metals mentioned, these also include the elements In, Y, Gd, U, Er, T, Tu, Sn, Ce, Pr, Ba, Rb, Yb, Tl, Th, Sr, La, Nd, Ra, Pu, Fr, AI and Ga. There is particular interest, for example, in the trace analysis of radioactive substances, such as. B. Plutonium. The sensitivity achieved with the analysis method according to the invention is approximately 1000 atoms / cm ² . This corresponds to a coverage of 10 ¹⁰ monolayers. In addition, large areas (e.g. 1 cm ² ) of the ana- lysing substrate are detected, so that a raster-like scanning of large surfaces is effectively possible. This is a decisive advantage over other highly sensitive methods for trace analysis, such as. B. the SIMS method, in which only small measuring spots in the sub-mm range can be detected. With the SIMS method, for example, a larger surface, for example the surface of a container for radioactive material, cannot be scanned within measurement times that are actually of interest.

The method illustrated in FIG. 3 can accordingly also be used for cleaning substrate surfaces. 4 (left part), a cluster 30 of polar molecules (e.g. sulfur dioxide) moves towards the substrate 31 to be cleaned. The substrate 31 is covered with electrically neutral adsorbates, e.g. Alkali metal adsorbates 32 contaminated. When the cluster 30, not shown, collides with the substrate 31, the adsorbate 32 is taken up and removed with the cluster fragments. In the right part of Fig. 4 the situation after the collision is illustrated. The amount of adsorbate on the substrate 31 is reduced. The contaminants are removed from the interface and at the same time converted into ionic particles, which can be extracted particularly easily with electromagnetic means. Again, the free ions can be subjected to an analysis to determine the composition of the surface contamination. If the type of contamination is known, it is also possible to dispense with the mass-spectroscopic analysis and instead carry out a charge measurement. With the charge measurement, the total charge of one of the two polarities is determined, and the degree of contamination or the progress of the cleaning process can be directly deduced from this.

An extension of the principle of cluster loading at the boundary surface (so-called "pickup" loading) illustrated in FIG. 2 according to a further embodiment of the invention is FIG. is shown in Fig. 5. According to the left part of FIG. 5, a large cluster 40 of molecules with low electron affinity, e.g. B. from ammonia molecules to the target 41. The interface is formed by the transition between the gas phase and the target, for example made of gold. The interface is covered with electrically neutral alkali metal adsorbents 42, for example Li, K, Na or Cs, and other neutral molecules 43. The molecules 43 comprise, for example, organic molecules or macromolecules, such as a DNA section. During the collision (not shown) between the cluster 40 and the target 41, the cluster 40, as described above, can absorb the alkali metal adsorbate 42 and / or the neutral molecule 43 and detach it from the interface 41.

If the cluster 40 picks up all of the molecule 43 during the collision and there is no reaction between the cluster and the molecule, everything is transferred into the gas phase. After the cluster envelope around the molecule 43 has been reduced in size by the shock-reduced fragmentation, individual components of the respective cluster fragment can be evaporated, so that thermal energy can be withdrawn, so that in the end the neutral molecule is brought into the gas phase with only minimal internal energetic excitation. The number of cluster building systems that surround the molecule can be reduced to 0. This procedure represents an extremely gentle transfer of neutral molecules into the gas phase, which is of particular interest for sensitive, biologically active macromolecules

If the cluster 40 absorbs all of the alkali metal adsorbate 42 when it collides with the interface, this spontaneously releases a valence electron to the environment in the cluster through the interaction with the polar ammonia molecules of the cluster 40, with an alkali cation 44 and em delocalized electron 45 are formed. Because of the lack of electron affinity However, the affinity of molecular ammonia does not lead to the formation of ammonia anions. The delocalized electron can either be stabilized by dipole cages in the cluster or can pass into the gold solid during the collision or form a free electron outside the cluster.

If the cluster 40 receives both the alkali adsorbate 42 and the neutral molecule 43 during the collision, the processes described above again arise, with the delocalized electron additionally being able to be stabilized by the molecule 43. Furthermore, the alkali cation 44 can also lie on the same cluster fragment as the molecule 43, so that the ionization is simultaneously achieved with the transfer of the molecule 43 into the gas phase. After this non-destructive ionization, the molecular ion, which is also characterized by a low kinetic energy, can be subjected directly to a mass spectroscopic analysis.

FIG. 6 shows an exemplary embodiment of a device according to the invention for examining and / or modifying boundary surfaces in the form of a cluster beam system. The cluster beam system is located in a multi-part reaction chamber (not shown), which is constructed, for example, like a conventional two-chamber molecular beam apparatus (background pressure without cluster beam 10 ^~ mbar ... 10 ^~ mbar). The cluster beam system comprises a cluster generating device 60, 61, optionally with a beam limiter 63, a cluster fragmentation device 62 and a measuring device 64. Furthermore, control and steering devices can be provided for the ionized cluster fragments, which, however, are known per se as manipulators for charged particles are therefore not shown separately. The cluster generating device comprises a nozzle 60 and a supply system 61. The nozzle is preferably a pulsed nozzle with parameters selected depending on the application, but can also A3 parameters depending on the application, but can also be operated continuously.

Typical parameters for pulsed operation include a nozzle diameter of 0.5 mm, a pulse width of 400 μs and a stagnation pressure of up to 20 bar. A working gas is supplied to the nozzle via the supply system 61, which consists of the carrier substance of the clusters to be generated or of a gas mixture of the carrier substance and an inert additive or of a gas mixture of the carrier substance and the reaction partner. The working gas is, for example, a mixture of sulfur tetrafluoride and helium. At the nozzle 60, the working gas is expanded with a specific expansion ratio chosen for the application (for example 1:30). A pressure of around 10 ^-3 mbar prevails in the part of the reaction chamber located downstream of the nozzle 60. After the expansion, the clustering takes place in a manner known per se by condensation et al. in "J. Chem. Phys. ", Volume 56, 1972, page 1793 ff., Can be measured using 30 eV electron impact ionization.

The addition of the inert gas during cluster generation serves to influence the cluster speed during cluster generation. For example, Ne, He or H ₂ are used as inert gases. The cluster sizes and speeds depend on the amount of inert gas and the gas pressures during expansion. With the above parameters, the cluster speed results in values in the range from 750 ms ^" to 2.5 • 10 ³ ms ^-1 and an average cluster size in the range from 1 to 750 atoms or molecules.

The radial extension of the cluster jet emerging from the nozzle opening is restricted by the jet limiter 63 (so-called skimmer) and strikes the cluster fragments. Vi tionsemπchtung, which in the example shown is formed by a solid surface 62 arranged in the beam direction (target).

The skimmer serves to reduce the pressure and introduce a spatial resolution during target irradiation (irradiation of certain sample areas). Carrying out the cluster fragmentation at a pressure which is lower than the atmospheric pressure has the advantage that it provides a greater free path for the moving clusters and ionized cluster fragments. The radial restriction of the cluster beam makes it possible to obtain spatially resolved ion signals from the interface and thus to carry out a spatially resolved surface analysis (down to the mm ... μm range). The solid surface 62 forms the interface to the cluster fragmentation and consists, for example, of a dielectric, silicon, gold or steel. The distance of the target (solid surface 62) from the nozzle is, for example, around 30 cm in a test setup. The cluster beam diameter on the target is around 8 mm. It can be provided that the target with a temperature control device (not shown) to a certain operating temperature, e.g. is tempered in the range from 400 K to 600 K in order to set conditions under which weakly bound molecular adsorbates have already been desorbed. After the above-described cluster fragmentation processes on the solid surface 62, the cluster fragments move in the opposite direction to the original beam direction, being deflected into the measuring device 64.

The measuring device 64 is a mass spectrometer, preferably a time-of-flight mass spectrometer, which is provided for the mass analysis of the ionized cluster fragments. A time-of-flight mass spectrometer has the advantage over an alternative quadrupole mass spectrometer that it is also possible to analyze larger masses, for example above mass 200. % $

FIG. 7 shows the positive and negative mass spectra of the cations or anions in the cluster fragmentation according to the invention. on a gold surface. The selected reactive system consists of a cluster of polar SO ₂ molecules and alkali atoms on the collision surface. The reaction in the cluster consists of the spontaneous release of the alkali valence electron onto an S0 ₂ molecule, mediated by the polar environment. Alkaline cations and SO anions are formed, which come to rest on cluster fragments due to the cluster fragmentation and are spatially separated from one another. The mass scale (abscissa) is plotted in units of S0 ₂ masses. The ordinate represents the measured ion number or ion intensity (arbitrary units). As expected, the two anion spectra (below) show maxima of the form (S0 ₂ ) _n S0 ₂ . In an experiment with an additional Cs coating on the surface (lower anion spectrum), this result is reproduced, but the number of anion fragments is increased.

The two cation spectra (above) show only maxima of the form (S0 ₂ ) _n ^{M + mιt M =} Na, K, Cs. As expected, all positive cluster fragments have alkali application. If the surface is additionally coated with cesium, the Cs ⁺ (S0 ₂ ) _n -Maxιma marked with the arrows are significantly strengthened (uppermost cation spectrum). Analog fragment mass spectra were also found in other polar molecules, with H ₂ O, NH_ and SF clusters, it being confirmed that the positively charged cluster fragments each contain an alkali metal atom which has been taken up by the irradiated interface.

The cluster beam system 6 according to FIG. 6 can be modified such that a charge measuring device (not shown) is provided instead of or in addition to the measuring device 64. This consists, for example, of a m annular distance in front of the solid surface 62 arranged grid, which is applied to the mass potential with a predetermined voltage. Depending on the polarity of the voltage, one type of ion fragment is subtracted from the grid, while the other type settles on the solid surface 62 so that it is charged. This charge is measured with a charge measuring device. The number of ionized fragments can be derived directly from the measured amount of charge.

A further application of the cluster fragmentation method according to the invention is illustrated in FIG. 8 using the example of an ion engine. The ion engine 7 comprises a cluster generation device 70, 71, a cluster fragmentation device 72, 73, control and steering devices 74, 75 and acceleration devices 76, 77. The entire ion engine is designed for operation in an evacuated reaction space in the laboratory or in space. The cluster generating device in turn comprises a pulsed nozzle 70 and a supply system 71. A gas mixture, for example of sulfur dioxide and helium or H ₂ , is fed from the supply system 71 to the nozzle 70 and, after passing through it, is expanded. The expansion ratio is 1:10, for example. The cluster jet emerging from the nozzle opening strikes the target 72 of the cluster fragmentation device, which furthermore includes the adsorber feed devices 73. The target 72 is at earth potential and is continuously coated with adsorbates by the adsorbate supply devices 73, for example in the form of an evaporation furnace, during the operation of the ion engine. Clusters of polar carrier molecules and adsorbates of alkali metal atoms, for example cesium, are preferred. The positive and negative cluster fragments which arise in the course of the collision of the clusters with the adsorbate-covered target 72 are spatially separated with the aid of the extraction grids 74 and deflected in the desired direction by means of magnetic and / or electrical steering devices 75. Then the separated fragments m the acceleration device 76, 77 e, which includes electrode tubes 76 and outlet grille 77. The electrode tubes 76 are made of metal and are subjected to a time-varying electrical potential which is adapted to the cluster pulses (impact, for example, every 100 ms) with respect to the ground potential. The outlet grid 77 is at ground potential. The electrode tubes 76 are actuated in such a way that after the cluster fragments have entered, a polarity-dependent acceleration h to the outlet grid takes place. To set the desired potentials, voltages of typically a few 10 kV are applied to the electrode tubes 76.

A particular advantage of the ion engine 7 over conventional ion engines is that the cluster fragmentation generates two charged fragments simultaneously, both of which can be used for thrust generation. In addition, particularly heavy ions can be provided with the cluster fragmentation, so that the thrust of the ion engine is increased.

The cluster fragmentation method according to the invention can be set up by suitable selection of the carrier material of the clusters and the geometry of the impact of the clusters on an interface so that particularly large, positively charged cluster fragments predominate as a result of the cluster generation. The generation of particularly large fragments, which are essentially the same size as the starting clusters, has particular advantages in the operation of the ion engine. The large cluster fragments have a large mass and therefore a high momentum. The material and geometry adjustment is based on the following concept.

A substance without or with a negligibly low molecular electron affinity is used as the carrier material. Examples of this are given by NH ₃ or H ₂ 0. In the 1% difference to the use of S0 ₂ with a high molecular electron afmitate (see above), the electron of the charge carrier pair present in the cluster is not taken up by the carrier material. Instead, it goes to the target or the free space. The result is only positive (and possibly neutral) cluster fragments. To request the transition of the electron to the target, the latter preferably consists of a metal with a high work function (e.g. tungsten).

In order to make the remaining positively charged cluster fragment as large as possible, it hits the interface at an angle not equal to 0 ° (based on the surface normal). It will grapple with z. B. 70 to almost 90 ° (based on the surface normal), in which relatively little kinetic energy is transferred to the cluster and used for its fragmentation. The fragmentation results in relatively large fragments. For example. can with clusters of z. B. 100 atoms with grazing incidence after fragmentation still e positively charged fragment with z. B. 80 to 90 atoms.

The generation of predominantly positive cluster fragments is illustrated in FIG. 9. FIG. 9 shows the result of the mass spectrometric analysis of fragments in grazing impact on a target coated with Na atoms with NH ₃ clusters (curves A, B) or SO ₂ -doped NH ₃ clusters (curve C). In the left part of FIG. 9, the clusters arriving over time are shown with different masses. The right part of FIG. 9 illustrates the mass distribution of the clusters over a narrow time range. The analysis of the positive clusters (curve A) in the case of pure NH ₃ clusters gives an image analogous to FIG. 7. The maxima corresponding to the multiples of the solvated nations can be seen. There are no maxima when measuring negatively charged clusters (curve B). No negatively charged clusters are detectable. The negative charge carriers (electrons) flowed to the target or the free space. If the clusters are doped with S0 ₂ , the image known from FIG. 7 is also measured in the negative channel of the mass spectrometer. In this case, the electrons are taken from S0 ₂ . Correspondingly negatively charged cluster fragments are detectable.

Compared to the examples explained, the invention can be modified as follows. To load the clusters with the reaction partner, the carrier substance and the reaction partner can be involved as two reaction partners (eg H ₂ 0 and NH ₃ ) during the cluster generation. The cluster is then built up during the adiabatic expansion of a mixture of both reactants. This has the advantage of a high density of reactive particles in the cluster that can be adjusted via the gas composition. To load the clusters in the event of a collision with the interface, instead of covering the interface with adsorbates as described, it can also be provided that the reactant itself is part of the interface or forms it. This has the advantage that the amount of charge carrier pairs in the cluster can be controlled via the surface density of the reactants. Compared to gas phase loading, there is the advantage that each cluster interacts with the surface and thus potentially with reactants, so that low efficiencies corresponding to the smaller cross sections of the gas phase are avoided. Depending on the application, it is possible to fragment individual clusters or cluster beams.

In the cluster generation for the adiabatic expansion, special precautions can be provided to control the gas composition, the temperature of the expansion nozzle and the expansion pressure to influence the cluster speed and average cluster size in the jet. This has the advantage that the generation of charge carriers in cluster fragments tion is influenced by setting the cluster size and the kinetic energy of the clusters. By mixing light gas components with heavier gas components, the speed of the heavier components can be increased (so-called "seeded beam" technology). The available energy range per particle is in the range of around 0.1 to 1 eV.

During the cluster generation, a step for ionizing the clusters with a subsequent acceleration of the cluster ions in electromagnetic fields can be provided. The ionization can take place according to the cluster fragmentation method according to the invention or according to a conventional ionization method. The use of ionized clusters for further cluster fragmentation has the advantage that the kinetic energy relevant for the cluster fragmentation can be set freely over a large range. Accordingly, a multiple run of the cluster fragmentation method according to the invention can be carried out sequentially, for example. A first run is directed to the generation of loaded cluster fragments, which are then e.g. m electromagnetic fields are accelerated in order to generate recharged cluster fragments by means of a further pass, which, however, have properties in another area of the parameter space of the kinetic energy.

If the interface to the cluster fragmentation is formed by gold, this has the advantage that the adsorption energies on gold surfaces are relatively low. This requires the cluster to be loaded with the reaction partner in the form of an adsorbate on the interface because of the low energy consumption. Furthermore, gold is conductive as a metal, so that with appropriate electrical wiring the interface does not charge up even after a long process operation. Any gold potential can be impressed on the gold surface, so that the potential of the charge carriers obtained is determined and used to manipulate the charge. Mung carrier, especially when accelerating, can be used. A cluster beam system according to the invention can be equipped with a device for setting the electrical potential of the interface in order to set a certain potential for formation of the cluster fragments.

If the cluster fragmentation is carried out on semiconductor surfaces, this has the advantage that these are commercially available, in particular with high purity. In addition, the surface properties of semiconductors are well known. Semiconductor surfaces can be produced with a particularly low roughness, which could adversely affect the charge carrier yield through an increased charge carrier capture through the surface. Finally, semiconductors can be changed in their conductivity and also in the electrical and dielectric properties of the interface via the doping. With a suitable doping, electrical charging of the interface can be avoided even during long process operation. Again, the generation potential of the fragment ions generated can be adjusted.

The cluster generation by supersonic expansion of a gas or gas mixture has the advantage that the clusters arise in high density in the form of a directed beam. The cluster jet has already developed after around 10 nozzle diameters. Furthermore, the clusters receive enough kinetic energy during generation, so that the cluster does not have to be re-accelerated. After all, relatively easy gaseous reactants can be built into the clusters during expansion. The jet diameter at the target is proportional to the nozzle-target distance and is e.g. at a distance of 30 cm and using a skimmer around 8 mm.

The real-time capability of analyzing and measuring the cluster fragments enables the cluster fragmentation process to be integrated into one Integrate control processes so that process parameters can be readjusted depending on the process success or the progress of the surface modification.

Claims

claims

1. A method for cluster fragmentation with the steps: generation of at least one cluster which contains a carrier substance, and

Fragmentation of the cluster m cluster fragments, characterized in that the cluster is loaded with at least one reaction partner which is chemically different from the carrier substance and which is part of at least one cluster fragment after the fragmentation.

2. The method according to claim 1, in which the cluster is loaded with at least one reaction partner which spontaneously or externally excitedly forms a pair of electrically unevenly charged charge carriers with the carrier substance in the cluster, and at least one electrically charged cluster fragment is formed during the fragmentation.

3. The method according to claim 2, wherein the cluster is additionally loaded with an electrically neutral molecule.

4. The method according to claim 3, in which, for the manipulation of the neutral molecules, these are applied as adsorbate coating to a solid surface and transferred from the solid surface m em charged cluster fragment.

5. The method according to any one of the preceding claims, wherein the cluster fragmentation takes place by collision of the cluster with a moving or stationary interface or by radiation excitation.

6. The method of claim 5, wherein the interface is a gas phase / liquid or gas phase / solid interface.

7. The method according to claim 6, wherein the interface is formed by a solid surface made of a metal, a semiconductor or a dielectric.

8. The method according to claim 6, wherein the interface with the reactant adsorbates is covered with a surface density which has a predetermined value on average over time.

9. The method according to any one of the preceding claims, wherein the loading with the reaction partner during the cluster generation, during the cluster movement to the interface by interaction with at least one gas phase particle of the reaction partner and / or during the collision with the interface by receiving reaction partner adsorbates done in the cluster.

10. The method according to any one of the preceding claims, in which polar molecules or molecular groups are used as the carrier substance and atoms and / or molecules or atomic or molecular groups with low ionization energy are used as reaction partners.

11. The method according to claim 10, in which alkali metal atoms are used as reactants.

12. The method according to any one of the preceding claims, wherein a plurality of clusters to be fragmented by Ultrasonic expansion of a gas and / or a gas mixture is generated by means of a nozzle arrangement.

13. The method according to claim 12, in which the generated clusters are subjected to a geometric beam limitation for irradiating an interface in accordance with a predetermined pattern.

14. The method according to any one of the preceding claims, wherein the cluster ionizes before the collision and the ionized clusters, in particular their kinetic energy, are influenced by electrical and / or magnetic fields.

15. The method according to claim 14, wherein the ionization of the clusters takes place according to a method according to one of claims 1 to 13.

16. The method according to any one of the preceding claims, in which the cluster fragments are subjected to a payment, a mass spectroscopic examination or a substance analysis.

17. The method according to any one of the preceding claims, wherein the fragmentation of the cluster takes place by grazing incidence of the cluster on an interface.

18. The method according to any one of the preceding claims, wherein the carrier substance consists of a chemical compound which has such a low electron affinity that electrons are not bound to a cluster fragment.

19. Use of a method according to one of the preceding claims: for taking up surface adsorbates from a surface that are to be subjected to an analysis, for taking up contaminants from solid surfaces for their cleaning, or for generating charged cluster fragments from clusters and aereosols that are to be subjected to a charge measurement or mass spectrometric analysis.

20. A method for operating an ion engine, in which the charged particles for forming the engine advance are formed by cluster fragments which have been generated by a method according to one of claims 1 to 16.

21. A cluster beam system comprising: a cluster generating device (60, 61), a cluster fragmentation device (62), and a measuring device and / or a manipulation device (64) for cluster fragments.

22. Cluster beam system according to claim 21, wherein the measuring device (64) is a mass spectrometer.

23. Cluster beam system according to claim 21 or 22, wherein the measuring device (64) is a charge measuring device.

24. Cluster beam system according to claim 21 to 23, wherein the manipulation device (64) comprises an electrode and / or coil device for generating time-constant or time-varying electromagnetic fields.

25. Cluster beam system according to one of claims 21 to 24, in which between the cluster generating device (60, 61) and the cluster fragmentation device (62) em beam limiter (63) is provided to shape the beam according to a predetermined pattern.

26. Cluster beam system according to one of claims 21 to 25, in which at least one reaction partner supply device is provided.

27. A cluster beam system according to claim 26, wherein the reactant supply means is formed by an evaporation furnace.

28. Cluster beam system according to one of claims 21 to 27, in which a device for adjusting the electrical potential of the cluster fragmentation device (62) is vorese¬

29. An ion engine (7) comprising: a cluster generator (70,71), a cluster fragmentator (72,73), steering and steering means (74,75), and an accelerator (76,77).