Preview

PDF, English - main document Download (22MB) | Terms of use

Abstract

The Mu3e experiment searches for the charged lepton flavor violating decay µ -> eee with an aimed single event sensitivity of 2*10^-15 in phase I. To achieve this goal, the highest momentum resolution possible is approached, which is mainly affected by multiple Coulomb scattering. This results in extremely tight constraints on the material budget to the level of 0.1 % radiation length per layer for the tracking detector.

The Mu3e pixel detector is based on High-Voltage Monolithic Active Pixel Sensors (HV-MAPS) developed for this project, the MuPix, which can be thinned to 50 µm. The electrical services and the support are realized by ultra-thin (around 80 µm) structured aluminum-polyimide laminates. Furthermore, the detector is cooled by gaseous helium, a novelty for particle detectors.

This thesis covers the construction of detector mock-ups in preparation for the final production of the Mu3e vertex detector, the studies of the helium cooling system, and the operation of a first functional prototype of the Mu3e vertex detector.

The mock-up construction demonstrated that an excellent longitudinal chip placement precision of σ = 4 µm and a lateral precision of σ = 3 µm is achieved with a manual approach. An average glue thickness of only (5.3 ± 1.7) µm for the joint between chips and laminates is realized.

The helium cooling system is verified to be sufficient to cool the Mu3e vertex detector. To prove this, the thermal-mechanical mock-ups were heated actively and cooled by the helium cooling system. For a heat dissipation of 350 mW/cm², all chip temperatures are found to be below 70°C for an inlet helium temperature of 0°C. The temperature change during heating up and cooling down can be described by a time constant of about 2 s to 3 s.

A first functional vertex detector prototype has been operated under Mu3e-like conditions. The communication with the vertex detector and the readout of its hits is performed by the Mu3e DAQ, which is proven to be functional. The hits from different portions of the detector are correlated, and the results are in agreement with the expectation for all the target geometries that have been used.

The conceptual detector design and the functionality of the novel detector cooling system are validated, which paves the way for the detector production in 2022.