Fernando Abudinén ; Christian Kiesling ; Thomas Kuhr

2018
Max-Planck-Institut für Physik Munich

Abstract: The Belle II experiment is located at the energy-asymmetric electron-positron collider SuperKEKB, the next-generation B factory at KEK in Tsukuba, Japan. With a design luminosity of 8x10^{35} cm^{-2}s^{-1}, SuperKEKB will overtake by nearly two orders of magnitude the record reached by its predecessor KEKB, leading ultimately to a data sample corresponding to a recorded integrated luminosity of about 50 ab^{-1}, which is fifty times larger than the one collected by the predecessor experiment Belle. The Belle II detector is a new state-of-the-art detector designed to cope with the challenging high-luminosity conditions and with the larger beam-backgrounds. The primary goal of the Belle II experiment is to reveal, or to impose stringent constraints on, possible new physics contributions beyond the SM by studying accurately the properties of B and D mesons as well as of tau leptons produced at SuperKEKB. A major part of the Belle II physics program is focused on measuring the violation of CP symmetry in B-meson decays. For a CP analysis of neutral B-meson decays at B factories, the flavor of one of the two exclusively produced neutral B mesons has to be determined, a task that is performed by flavor tagging algorithms, or shortly flavor taggers. In this thesis, a new flavor tagging algorithm was developed for \ac{belleii} with a specific and dedicated optimization to exploit the capabilities of the new detector and to cope with the harsher experimental conditions at SuperKEKB. The new Belle II flavor tagger developed in this thesis is based on novel multivariate methods exploiting the flavor specific signatures of B0 decays. In the absence of \ac{belleii} data, the new Belle II flavor tagger was comprehensively validated using Belle data, on which the algorithm reached an effective tagging efficiency of (33.5 +- 0.5(stat))%, outperforming the previous algorithm at Belle by about 10%. On Belle II MC events simulated with background, the algorithm reaches an effective tagging efficiency of (33.89 +- 0.04(stat))%. Thanks to the expected large size of its data sample and to its vertex reconstruction capabilities, Belle II will be able to make measurements that were not possible at its predecessors. A substantial part of this thesis is dedicated to prepare the data analysis for the measurement of the time-dependent \CP-violation parameter S_{CP} of the decay B0->pi0pi0, which is essential to reduce the 8-fold ambiguity in the determination of the unitarity angle phi2/alpha considering B->pipi decays. The time-dependent CP analysis of B0->pi0pi0 requires a precise determination of the B0-decay position, which cannot be achieved in the dominant four-photons final state. Thus, a novel analysis method was developed to make use of rare events with photons converting into e+e- pairs or neutral pions undergoing Dalitz pi0->e+e-gamma decays. The studies showed that, with the full Belle II data sample of 50 ab^{-1}, the measurement of S_{pi0pi0} will be feasible. The expected uncertainty of S_{pi0pi0} was estimated to be +- 0.28(stat) +- 0.03(syst), resulting in a reduction of the current 8-fold phi2-ambiguity to a 2-fold one. Considering the isospin analyses of both B->pipi and B->rhorho decays, the overall expected Belle II precision for phi2 at 50 ab^{-1} was estimated to be around 0.6°, which is more than five times smaller than the current world average.

Note: Presented on 19 10 2018
Note: PhD