Development and commissioning of two superconducting 217 MHz CH structures for the HELIAC project

In the context of the work presented here, two identical CH structures were developed for the HELIAC project (HELmholtz LInear ACcelerator) currently under construction and accompanied during production up to the final cold tests at 4.2 K. The two CH structures are now in use. Together with the CH structure of the demonstrator project, they enable the full commissioning and first beam test of the HELIAC's first cryomodule, which consists of four cryomodules with a total of 12 CH structures. Compared to previous CH structures, the design of the cavities was fundamentally revised and optimised as part of this dissertation. By removing the girders and the conically shaped end caps, the stability of the new CH structures could be significantly increased, so that the pressure sensitivity could be reduced by approx. 80% compared to the first CH cavity of the demonstrator project. Due to the outwardly drawn lamellae of the dynamic tuners, the mechanical stress as well as the required number of lamellae and thus the risk for the occurrence of multipacting could be reduced. The reduced risk of multipacting due to the corresponding optimisations of the cavities could be verified by the permanent overcoming of all multipacting barriers in the later measurements. Both cavities were optimised with the help of the simulation programmes CST Studio Suite and Ansys Workbench. Both cavities were manufactured by Research Instruments (RI) and monitored by various intermediate measurements during the entire construction. After each individual production step, all influences on the resonance frequency were determined so precisely that the target frequency at 4.2 K could be achieved with an accuracy of more than 1‰. Automated recording routines were used both during the intermediate measurements and during the final measurements at 4.2 K, which enabled the measurement data to be read out to the second and thus a high degree of measurement accuracy. With regard to the complexity of the CH structures, the small deviations from the target frequency are direct evidence of how successful and precise the evaluations and resulting estimates of the individual intermediate measurements were. All in all, with the exception of the mechanical eigenmodes, all results of the simulations could be verified by corresponding measurements in a good approximation. Two dynamic tuners were installed in each cavity, which can compensate for static and dynamic frequency deviations in later operation. The dynamic tuners were examined and optimised in detail with regard to their mechanical stability and the frequency change that can be generated, as well as their mechanical natural frequencies, using the simulation programmes CST Studio Suite and Ansys Workbench. In order to check the results of the simulations, a specially designed measurement set-up was used, which was manufactured in the workshop of the Institute of Applied Physics and made it possible to precisely measure all the decisive properties of the dynamic tuners. Overall, the detailed measurements with the help of the designed setup represent the most comprehensive measurements of dynamic bellows tuners within superconducting CH structures to date and show which deviations between simulations and measurements are to be expected in future cavities. The field distribution along the beam axis was also checked during the production of the cavities using the interference body measurement method. The values determined in this way agreed very well with the simulations, with a maximum discrepancy of 9%. In order to guarantee the best possible surface quality, at least 200 µm was removed from the inner surface of both structures in several steps using a mixture of hydrofluoric, nitric and phosphoric acid. By dividing the treatment into individual steps, the influence of the surface treatment on the resonance frequency could be better estimated and predicted. This, together with the measurements to determine the pressure sensitivity and the thermal contraction of the cavity during cooling, led to the high agreement of the measured final resonant frequency with the target frequency. The final cold tests of the two cavities, without helium sheath, were performed at the Institute of Applied Physics of the Johann Wolfgang Goethe University in a vertical bath cryostat. The first CH structure was successfully tested up to a maximum field gradient of 9.2 MV/m, which corresponds to an effective stress of 3.37 MV. The unloaded quality dropped from 1.08 ∙ 10${}^9$ at the beginning to 2.6 ∙ 10${}^8$. The specifications of the HELIAC project are for an acceleration gradient of 5.5 MV/m with an unloaded quality of at least 3 ∙ 10${}^8$. These values were clearly exceeded by the first cavity, so that it can be used without restriction for operation within the first cryomodule. For the second cavity, a vacuum leak occurred during cooling to 4.2 K, which was not detectable at room temperature. Due to the poor vacuum conditions inside the cavity, it was therefore not possible to carry out any measurements regarding the performance as long as the cold leak was present. A new cold test of the cavity after removal of the leak could not be carried out within the scope of this work and is therefore the subject of subsequent investigations. Overall, the developments, investigations and measurements within the framework of the dissertation presented here represent a decisive step towards the commissioning of the HELIAC's first cryomodule and the development of further CH cavities. The revised design of the CH structures has proven successful, which is why it is used as a starting point for the development of all subsequent CH structures of the HELIAC, up to the completion of the complete accelerator.