Track Based Alignment for the Mu3e Pixel Detector

Abstract

In the hunt for new physics, the Mu3e experiment searches for the lepton flavour violating decay μ+ → e+e−e+ with a sensitivity goal of 1 in 10^{16} decays. In the Standard Model of particle physics, this process is heavily suppressed, making its observation a clear indicator of physics beyond the Standard Model. Up to 10^9 muons per second will be stopped in a target and decay at rest inside a magnetic spectrometer. To reconstruct the trajectories of the decay products, multiple concentric barrel shaped layers, consisting of about 3000 very thin (50 μm) high-voltage monolithic active pixel sensors, are used. With sensors that thin, the customary way of aligning a tracking detector - i.e. for position and orientation of detector parts only - is not sufficient. In addition, possible surface deformations will need to be taken into account. Since the assembly tolerance for the pixel sensors is too large to ensure good tracking results, a track based alignment programme, utilising the General Broken Lines fit and the Millepede-II algorithm will be used to achieve this precision in the final detector. A reliable alignment programme is an essential prerequisite for the Mu3e experiment, since inaccurate knowledge about the detector geometry will foil all attempts at data analysis. In this work, preparations for aligning the pixel detector using a detailed simulation and the track based alignment tool Mu3ePixAl - which was developed in the context of this work - will be presented. The effects of misalignment will be discussed and it will be demonstrated that Mu3ePixAl is capable of determining corrections to misplacements of pixel sensors well enough to reach tracking performances agreeing almost perfectly with the results for the nominal detector.