Open Access

The Large Hadron Collider (LHC) is the biggest and most powerful particle accelerator in the world, designed to collide two proton beams with particle momentum of 7 TeV/c each. The stored energy of 362MJ in each beam is sufficient to melt 500 kg of copper or to evaporate about 300 litre of water. An accidental release of even a small fraction of the beam energy can cause severe damage to accelerator equipment. Reliable machine protection systems are necessary to safely operate the accelerator complex. To design a machine protection system, it is essential to know the damage potential of the stored beam and the consequences in case of a failure. One (catastrophic) failure would be, if the entire beam is lost in the aperture due to a problem with the beam dumping system. This thesis presents the simulation studies, results of a benchmarking experiment, and detailed target investigation, for this failure case. In the experiment, solid copper cylinders were irradiated with the 440GeV proton beam delivered by the Super Proton Synchrotron (SPS) at the High Radiation to Materials (HiRadMat) facility at CERN. The experiment confirmed the existence of the so-called hydrodynamic tunneling phenomenon for the first time. Detailed numerical simulations for particle-matter interaction with FLUKA, and with the two-dimensional hydrodynamic code, BIG2, were carried out. Excellent agreement was found between the experimental and the simulation results that validate predictions for the 7TeV beam of the LHC. The hydrodynamic tunneling effect is of considerable importance for the design of machine protection systems for accelerators with high stored beam energy. In addition, this thesis presents the first studies of the damage potential with beam parameters of the Future Circular Collider (FCC). To detect beam losses due to fast failures it is essential to have fast beam instrumentation. Diamond based particle detectors are able to detect beam losses within a nanosecond time scale. Specially designed diamond detectors were used in the experiment mentioned above. Their efficiency and response has been studied for the first time over 5 orders of bunch intensity with electrons at the Beam Test Facility (BTF) at INFN, Frascati, Italy. The results of these measurements are discussed in this thesis. Furthermore an overview of the applications of diamond based particle detectors in damage experiments and for LHC operation is presented.

Cancer has become one of the most fatal diseases. The Heidelberg Heavy Ion Cancer Therapy (HICAT) has the potential to become an important and efficient treatment method because of its excellent “Bragg peak” characteristics and on-line irradiation control by the PET diagnostics. The dedicated Heidelberg Heavy Ion Cancer Therapy Project includes two ECR ion sources, a RF linear injector, a synchrotron and three treatment rooms. It will deliver 4*10 high 10 protons, or 1*10 high 10 He, or 1*10 high 9 Carbons, or 5*10 high 8 Oxygens per synchrotron cycle with the beam energy 50-430AMeV for the treatments. The RF linear injector consists of a 400AkeV RFQ and of a very compact 7AMeV IH-DTL accelerator operated at 216.816MHz. The development of the IH-DTL within the HICAT project is a great challenge with respect to the present state of the DTL art because of the following reasons: • The highest operating frequency (216.816MHz) of all IH-DTL cavities; • Extremely large cavity length to diameter ratio of about 11; • IH-DTL with three internal triplets; • The highest effective voltage gain per meter (5.5MV/m); • Very short MEBT design for the beam matching. The following achievements have been reached during the development of the IH-DTL injector for HICAT : The KONUS beam dynamics design with LORASR code fulfills the beam requirement of the HICAT synchrotron at the injection point. The simulations for the IH-DTL injector have been performed not only with a homogeneous input beam, but also with the actual particle distribution from the exit of the HICAT RFQ accelerator as delivered by the PARMTEQ code. The output longitudinal normalized emittance for 95% of all particles is 2.00AkeVns, the emittance growth is less than 24%, while the X-X’ and Y-Y’ normalized emittance are 0.77mmmrad and 0.62mmmrad, respectively. The emittance growth in X-X’ is less than 18%, and the emittance growth in Y-Y’ is less than 5%. Based on the transverse envelopes of the transported particles, the redesign of the buncher drift tubes at the RFQ high energy end has been made to get a higher transit time factor for this novel RFQ internal buncher. An optimized effective buncher gap voltage of 45.4KV has been calculated to deliver a minimized longitudinal beam emittance, while the influence of the effective buncher voltage on the transverse emittance can be neglected. Six different tuning concepts were investigated in detail while tuning the 1:2 scaled HICAT IH model cavity. ‘Volume Tuning’ by a variation of the cavity cross sectional area can compensate the unbalanced capacitance distribution in case of an extreme beta-lambda-variation along an IH cavity. ‘Additional Capacitance Plates’ or copper sheets clamped on drift tube stems are a fast way for checking the tuning sensitivity, but they will be replaced by massive copper blocks mounted on the drift tube girders finally. ‘Lens Coupling’ is an important tuning to stabilize the operation mode and to increase or decrease the coupling between neighboring sections. ‘Tube Tuning’ is the fine tuning concept and also the standard tuning method to reach the needed field distributions as well as the gap voltage distributions. ‘Undercut Tuning’ is a very sensitive tuning for the end sections and with respect to the voltage distribution balance along the structure. The different types of ‘plungers’ in the 3rd and 4th sections have different effects on the resonance frequency and on the field distribution. The different triplet stems and the geometry of the cavity end have been also investigated to reach the design field and voltage distributions. Finally, the needed uniform field distribution along the IH-DTL cavity and the corresponding effective voltage distribution were realized, the remaining maximum gap voltage difference was less than 5% for the model cavity. The several important higher order modes were also measured. The RF tuning of the IH-DTL model cavity delivers the final geometry parameters of the IH-DTL power cavity. A rectangular cavity cross section was adopted for the first time for this IH-DTL cavity. This eases the realization of the volume tuning concept in the 1st and 2nd sections. Lens coupling determines the final distance between the triplet and the girder. The triplets are mounted on the lower cavity half shell. The Microwave Studio simulations have been carried out not only for the HICAT model cavity, but also for the final geometry of the IH-DTL power cavity. The field distribution for the operation mode H110 fits to the model cavity measurement as well as the Higher Order Modes. The simulations prove the IH-DTL geometrical design. On the other hand, the precision of one simulation with 2.3 million mesh points for full cross section area and the CPU time more than 15hours on a DELL PC with Intel Pentium 4 of 2.4GHz and 2.096GRAM were exploited to their limit when calculating the real parameters for the two final machining iterations during production. The shunt impedance of the IH-DTL power cavity is estimated by comparison with the existing tanks to about 195.8MOmega/m, which fits to the simulation result of 200.3MOmega/m with reducing the conductivity to the 5.0*10 high 7 Omega-1m-1. The effective shunt impedance is 153 MOmega/m. The needed RF power is 755kW. The expected quality factor of the IH-DTL cavity is about 15600. The IH-DTL power cavity tuning measurements before cavity copper plating have been performed. The results are within the specifications. There is no doubt that the needed accuracy of the voltage distribution will be reached with the foreseen fine tuning concepts in the last steps.

In the framework of this thesis the intense low energy ion beam transport was investigated. Especially, the beam transport in toroidal magnetic field configurations was discussed, as it may allow the accumulation of high intensive beams in the future. One of the specific tasks is to design an injection system that can be used for the proposed low energy accumulator ring. This thesis regarding beam transport investigations is related to the larger research fields, storage rings used in accelerator physics and non-neutral plasmas. The proposal of building a storage ring with longitudinal guiding magnetic fields was made. Due to natural transversal focussing in magnetic fields it is possible to accumulate very intense charged particle beams, a subject of interest within the physics community. A simulation code (TBT) was written to describe the particle motion in curved segments. Particle in Cell techniques were utilized to simulate a multi particle dynamics. This code allows the user to generate different particle distributions as input parameter. A possibility of reading an external data file was made available so that a measured distribution can be used to compare simulation results with measured ones. A second order cloud in cell method was used to calculate charge density and in turn to solve Poisson’s equation. The circular toroidal coordinate system was used. The drift motion and gyrating motion was proved to be consistent with analytical values. Further simulations were performed to study the self field effects on beam transport. The experiments with single toroidal segments find niche in the work. The experiments were performed to compare the simulation results and gain practical experience. The toroidal segment has similar dimensions (major axis R = 1:3 m, minor axis r = 0:1 m, arc angle 30°) as for a full scale ring design. The main difference lies in the magnetic field strength. The available segments can be operated at room temperature producing 0:6T on axis maximum magnetic field, while for the storage ring design this value is in the range of 5T. The preparatory experiments consisted of building and characterization of the ion source in a first step. Along with the momentum spectrometer and emittance scanner the beam properties were studied. Low mass ion beams He+ and mixed p, H2+, H3+ beams were analyzed. The proton beam consisting of a 48% H+ fraction was extracted regularly and used for further experiments. A moderate beam energy of 10 keV was chosen as operational energy for which 3.08 mA proton beam current was measured. In the second stage, beams were transported through a solenoid and the phase space distribution was measured as a function of the magnetic field for different beam energies. The phase-space as distributions measured in a first stage were simulated backward and then again forward transported through the solenoid. The simulated results were then compared with the measured distribution. The LINTRA transport program was used. The phase-space distribution was further simulated for transport experiments in a toroidal magnetic field. The experiments with a single toroidal segment give basic results necessary to compare the results between transport code (TBT) and measurements. The optical diagnostic provides measurements which can be well compared with the simulated results. A digital camera with a magnetic shield was used to record images in jpeg file format. A subroutine was written to analyze an image file to give the intensity distribution of a given image file. The integrated profile in vertical and horizontal direction was used to calculate the vertical drift and the beam size. The simulated values were in good agreement with the measured ones. The injection system needs most care. The transport program that was used to simulate the beam in the toroid was also used to design the injection system. The injection system with its special field configurations was designed to perform experiments with room temperature segments. The main point to tackle was to smoothly bring the charged particles generated outside the trap into the acceptance of the ring. The designed system consists of two sources, one representing a ring beam and the other one the injection beam. While simulations showed a clear way, how to inject the particle beam via a well positioned solenoid and in combination with a transverse electric field element causing an ExB drift into the main ring acceptance. After construction of these injection elements it will be very important to measure the robustness of such a system with respect to the beam stability- especially of the injection channel.

Im Rahmen dieser Diplomarbeit wurde eine Detektor-Sonde entwickelt, um Strahlprofile eines Ionenstrahls entlang der gekrümmten, geometrischen Achse eines Toroidsektormagneten zu messen. Bei der Konstruktion der Sonde musste die zuverlässige Messwerterfassung im Vakuum und innerhalb magnetischer Felder von bis zu 0,6 T berücksichtigt werden. Im theoretischen Teil werden die Theorie zum Strahltransport in den verwendeten Bauteilen, sowie die Funktionsweise eines Phosphor-Schirms (P20) dargelegt. Im experimentellen Teil wird die bewegliche Sonde, der verwendete Versuchsaufbau, sowie die Messungen und deren Auswertung näher beschrieben. Im abschließenden Fazit wird auch auf Alternativen zu der verwendeten Messmethode und deren Vor- und Nachteile eingegangen.

Das Funneling-Prinzip ist für Großprojekte wie SNS und IFMIF zur Erzeugung hoher Strahlströme bei hoher Brillanz von großem Interesse und bietet die Möglichkeit der Strahlstromerhöhung bei gleichbleibender Emittanz. Das Frankfurter Funneling-Experiment ist ein skalierter Aufbau einer ersten Funneling-Stufe von HIDIF. Hauptbestandteile des Experimentes sind zwei Multicusp-Ionenquellen, ein Zwei-Strahl-RFQ-Beschleuniger, ein Einzellen- und ein Mehrzellen-Deflektor sowie eine Emittanzmeßanlage. Das Zusammenführen zweier Ionenstrahlen nach dem Funneling-Prinzip konnte am IAP im Jahr 2000 erstmalig realisiert werden. Allerdings war aufgrund der unmodulierten End-Elektroden des RFQ-Beschleunigers der Strahlradius und die Emittanz bereits bei Eintritt in den Deflektor viel zu groß. Die dadurch aufgetretenen Strahlverluste an den Elektroden führten also nicht zu der gewünschten Strahlstromverdoppelung. Daraufhin wurden die letzten Elektrodenstücke der beiden Beschleuniger gegen modulierte Elektroden ausgetauscht. Der Fokus der Ionenstrahlen wird nun mittels eines sogenannten 3D-Matchings in den Strahlkreuzungspunkt gelegt. Experimente mit den neuen RFQ-End-Elektroden und dem überarbeiteten Mehrzellen- Deflektor stehen noch aus. Die vorliegende Arbeit entstand als theoretischer Teil im Rahmen des Frankfurter Funneling-Experimentes. Es sind zahlreiche Simulationsrechnungen zum bestehenden experimentellen Aufbau durchgeführt worden, die in Auszügen in Kapitel 7 dargestellt wurden. Weiterhin wurde die Teilchendynamik und die Raumladung in Deflektoren, das Emittanzwachstums während des Funnelings, der Einfiuß der inhomogenen Felder bei verschiedenen Deflektorgeometrien ausführlich untersucht und ausgewertet (Kapitel 8). Für diese Aufgaben sind einerseits neue Programme für eine dreidimensionale Deflektorsimulation und andererseits Software zur Auswertung mit graphischer Darstellung geschrieben worden. Diese wurden in Kapitel 6 vorgestellt. Die für diese Arbeit entwickelten Programme ermöglichen die Berechnung der Potential- und Feldverteilungen in elektrischen Hochfrequenz-Funneling-Deflektoren sowie die Simulation des Funnelingprozesses zweier Ionenstrahlen. Ferner sind diverse Auswertemethoden in tabellarischer oder graphischer Form wie Strahlverlauf, Emittanzebenen, Dichteverteilungen und Verlustgraphen verfügbar. Damit sind umfangreiche Simulationen und Auswertungen bezüglich des Deflektordesigns und der Strahldynamik sowie Optimierungen solcher Systeme möglich. Der Einfluß der Raumladungskräfte und der inhomogenen Felder auf den Funnelingprozeß konnten in Kapitel 8 gezeigt werden. Für den im Experiment verwendeten Mehrzellen-Deflektor sollten folgende Strahlparameter eingehalten werden: der Strahlradius in der x-y-Ebene sollte vor Eintritt in den Deflektor kleiner als 0.5 cm sein, die Energiebreite deltaW < +- 2% und die Phasenbreite deltaPhi < +- 30° betragen. Ansonsten treten Teilchenverluste durch Elektrodenkontakt auf oder der Bunch wird in longitudinaler Richtung zu groß, so daß die Möglichkeit besteht, das eine Überlappung der Bunche stattfindet. Mit der vorliegenden Arbeit sind Programme zur detaillierten Berechnung und Analyse von Funneling-Systemen entwickelt worden. Zukünftige Aufgaben sind neben der Untersuchung der Randfelder in Deflektoren die Minimierung des Emittanzwachstums durch die inhomogenen Felder. Nach ersten Strahltests und Funnelingergebnissen ist zu entscheiden, ob eine Matching-Sektion zwischen RFQ-Beschleuniger und Funneling-Deflektor zur weiteren Strahlanpassung eingebracht werden muss.