Search for Dark Matter and other beyond the Standard Model physics with XENON1T and XENONnT

Despite numerous evidence suggesting that the majority of matter in the Universe is made of elusive Dark Matter, its nature remains a mystery and is one of the main topics of modern particle physics. The XENON Dark Matter project is aiming for the direct detection of the hypothetical Dark Matter particles with increasingly sensitive detectors. It takes advantage of the dual-phase xenon time projection chamber (TPC) technology, continuously reaching higher target masses and lower experimental backgrounds. The XENON1T experiment placed the most stringent upper limits of weakly interacting massive particle (WIMP) interaction strengths for a variety of interaction types and WIMP masses, from which several still remain to date. The unprecedented low electronic recoils background in XENON1T lead to competitive searches for beyond the Standard Model physics apart from WIMPs. An excess of events over known backgrounds at lowest electronic recoil energies was observed which received the attention of physicists around the world. However, the origin of this excess of events remained inconclusive. This work investigates the possibility that the excess of electronic recoil events originates from inelastic scattering of Dark Matter off electrons in the xenon target. Sensitivity studies for the XENON experiments are conducted with respect to a concrete theoretical model describing the inelastic exothermic down-scattering of a dark particle which interacts with the visible sector via gauge interactions involving dark photons. It is shown that the predicted annual modulation of the electronic recoil event rate and of the energy spectrum provides an additional tool to distinguish between various theoretical models and constraining the Dark Matter particle properties. Assuming a distinctive excess of events is found in XENONnT, it would be possible to determine the Dark Matter particle mass with a precision of ~1 GeV over the lifespan of the experiment. In addition, a simplified model describing the energy- and time-dependent modulations is developed. Previous analyses of the excess of events from XENON1T were only performed by investigating the electronic recoil energy spectra, neglecting most of the time information. This work describes the re-analysis of the XENON1T electronic recoil events focusing on the time information leading to various improvements over the previous analysis. The statistical analysis was performed for the first time as a function of recoil energy and time of the event, demonstrating the deep understanding of the background contributions. The time signature of the excess of events is analyzed, testing several hypothesis including a potential 3H component. The XENONnT experiment was designed as a fast upgrade of XENON1T and features an even larger TPC, as well as ~5 times lower detector background rate. The decreased materials background rate is achieved by a careful material selection and reduction. The thickness of the PTFE detector wall is optimized for light tightness and mass reduction. Measurements of the light transmission at xenon scintillation light wavelengths lead to the optimal wall thickness of 3mm. It is demonstrated that the Kubelka and Munk model is applicable to describe the transmission for all PTFE thicknesses. Optical simulations of a particle detector are crucial for the design and data analysis. This work outlines the development of the optical simulations framework for XENONnT. It was used to perform sensitivity predictions of the experiment for WIMP Dark Matter prior to commissioning. The results of this work are also used as essential part for the position reconstruction, detector emission models and data quality selections during the first science run. A novel optical simulation technique for S2 signals and a detailed description of the photosensor detection efficiencies are introduced. The author of this work was responsible for the detector characterization analysis group including the work on the position reconstruction, signal corrections and event selections. The results are part of the first electronic recoil data analysis which excludes the presence of an excess and placed limits on various beyond the Standard Model searches. In addition, a blind analysis of nuclear recoil events is performed and no significant excess found, placing upper limits for spinindependent WIMP-nucleon interactions with a minimum upper limit of 2.58 × 10−47 cm2 for WIMP masses of 28 GeV/c2 at 90% confidence level.