The finally achievable performance of the PANDA EMC, which was aiming for an extremely compact and radiation hard calorimeter, covering for the first time an energy regime from 15 GeV down to a few MeV, strongly relies on the quality parameters of the PbWO4 crystals. Therefore a very complex test procedure was elaborated, which basically consists of three stages of quality control at the locations BTCP, CERN and Giessen. The obtained data of 9; 336 crystals was analysed with respect to the specification limits. Furthermore, correlations and discrepancies between the different facilities, mainly caused due to the different treatment of crystal geometries, are discussed in detail. Finally, an outlook of the still mission fraction of crystals for the PANDA EMC is given. The SICCAS company at China is a promising manufacturer of PbWO4 experienced by the production of a significant fraction of crystals for the CMS experiment. Here, the quality of 50 test samples were tested and compared, in particular with respect to the different growing technology. The obtained results arepromising, but all parameters scatter over too wide distribution and a more homogeneous quality is expected from a pre-production run.A sufficient time resolution for the EMC is necessary to provide an accurate time stamp for the detected physics events synchronously with the time distribution system SODA of PANDA and for the rejection of background events. The time-walk corrected resolution under different conditions and digitisation procedures was determined at three separate beamtimes and compared to results from the CMS experiment. The achievements presented in this work represent an upper limit for the final time resolution, since the foreseen APFEL ASIC with two dynamically adjustable and independent gain branches will further improve the timing performance.As additional aspect of this thesis, a possible energy sum correction algorithm is introduced. Due to the presence of passive material between the crystals, the energy reconstruction is significantly reduced, if the point of impact is located close to the edge of a crystal. The so called ln(E2/E1)-method considers the shape of the lateral shower distribution and was successfully exploited for showers initiated by positrons and photons, as well. Especially for hits in between two crystals a significant improvement has been achieved.
Verknüpfung zu Publikationen oder weiteren Datensätzen
