Software and hardware development for the next-generation liquid scintillator detectors JUNO and OSIRIS

Large liquid scintillator~(LS) detectors are acknowledged instruments in the field of neutrino physics. Based on various successful experiments, reporting the currently best limits on several parameters of neutrino flavor oscillations, a new generation of detectors with several tens of kilotons of LS are under consideration. The Jiangmen Underground Neutrino Observatory~(JUNO) is a 20 kiloton LS detector, that is fully funded and under construction in China. Its main goal is the determination of the neutrino mass ordering~(MO) through a precision measurement of the reactor electron anti-neutrino spectrum. The first part of this thesis discusses the underlying theory of neutrino flavor oscillations, the JUNO detector design and how neutrinos of various sources can be detected with this instrument. The focus is laid on a correlated background for the inverse beta decay~(IBD) measurement of reactor anti-neutrinos, which stems from cosmic muons. When they traverse the detector, the muons can create unstable radioisotopes, which decay after a short time in a (beta + n) channel. In order to identify and reject this background, it is paramount to know the track of the muon precisely. For this purpose, a novel muon reconstruction algorithm is developed and tested in this work. It is based on the geometric model of the intersection of the first-light front with the PMT array. The track parameters are optimized in a likelihood fit based on probability density functions produced with a detailed detector simulation. In addition, a simulation of the full readout electronics is performed to yield the best estimate of the performance on real data. Excluding the edge of the CD, the muon track's distance from the detector center DeltaD can be determined with an uncertainty of 5 cm and its direction with 0.3°. The impact on the detector's exposure by a muon veto based on this reconstruction was also studied. Compared to a perfect knowledge of each muon track, the developed method only creates an additional 4 % of loss in exposure. In the second part, a pre-detector for JUNO is investigated. OSIRIS is a standalone, 20 ton LS detector, that will be used to monitor the radiopurity of the cleaned LS before it is filled into JUNO.In the scope of this work, a detailed detector simulation based on C++11 and Geant4 is developed. It is then used to determine the sensitivity of the detector to its main physics goal: the identification of Bi-Po coincidences from the decay chains of U-238 and Th-232 in the LS. Furthermore, a calibration campaign for OSIRIS is studied. Under consideration of the available hardware, the decision is made to utilize an automated calibration unit~(ACU) from the Daya Bay collaboration. The energy range of 0.5 - 3 MeV will be calibrated by exposing the detector simultaneously to Cs-137, Zn-65, and Co-60 in a single capsule. With different vertical positions on a fixed radial distance r = 120 cm from the detector's center, its non-uniformity can be properly sampled. Timing calibration of the PMTs with an accuracy of~0.1 ns is realized with a 430 nm LED, that can be deployed along the same vertical axis