Search for B⁺→l⁺ν_lγ decays with hadronic tagging using the full Belle data sample

The Standard Model (SM) is the foundation of particle physics which provides a framework for all known particles and their interactions. It has been able to correctly reproduce experimental results at particle colliders for the last forty years. However, experimental findings and theoretical considerations indicate that an extension of the SM into a more complete model is necessary. The first evidence for physics beyond the SM is the discovery of neutrino masses through neutrino oscillation. Another important phenomenon which cannot be understood within the SM is dark matter since no known particle is a viable candidate to describe all of its properties. Additional important theoretical concerns which arise in the SM are: the fine tuning problem of the Higgs mass; the baryogenesis in the early universe; the missing concept of a Grand Unified Theory which also includes quantum gravity; and the hierarchy of the particle masses as well as the weak mixing angles. Many compelling arguments are given for the search of physics beyond the SM. Extensions of the SM postulate new particles to account for the unexplained effects. The Belle experiment and its successor, the Belle II experiment, are designed to investigate the B meson system to the highest precision. The most important result by Belle was the measurement of CP violation in the B system in several decay channels. The CP violation is caused by an irreducible complex phase in the Cabibbo-Kobayashi-Maskawa (CKM) matrix which contains the weak mixing angles between the quarks. This mixing mechanism is studied intensely to look for new CP-violating phases which would be a sign of New Physics. Additionally, branching fractions and polarizations in B decays provide information about the flavor and spin structure of the weak decay and ultimately about New Physics. New particles which couple to the SM particles are expected to be too massive to be produced directly at the low center-of-mass energy. However, they can contribute to the decays as virtual particles in higher-order loop processes. These contributions will result in small deviations to the SM predictions. The biggest obstacle for precise theoretical predictions is posed by strong interactions at low energies which are present in all B decays. These cannot be calculated perturbatively due to the size of the strong coupling constant which is of order one resulting in a non-convergent power series. This is solved by moving the nonperturbative parts of the calculation into parameters which are determined in the experiment. Non-perturbative calculations of these quantities are, in many cases, difficult and have large errors. With the measurement of the B⁺→l⁺ν_lγ decay, a yet unmeasured parameter of the B meson can be determined which is called λ_B. This is needed in a QCD factorization approach which provides an important ansatz to compute hadronic B decays in two mesons. Specifically, the parameter is an input in the computation of charmless hadronic decays where the B meson decays in two light mesons with a quark content of up, down, or strange. This comprises many decay channels where the parameter introduces a sizable uncertainty. Especially the calculations for color-suppressed modes B→ππ, πρ, and ρρ have large errors which are dominated by this parameter and experimental data hints to values of λ_B∝200 MeV. Non-perturbative calculations with QCD-sum-rules yield disagreeing results of λ_B∝(350-500) MeV. The measurement of B⁺→l⁺ν_lγ is the only way to resolve this tension experimentally and ultimately determine whether inconsistencies persist in the QCD factorization approach. The best limit for the decay is reported by the BaBar collaboration with a branching fraction of B(B⁺→l⁺ν_lγ)<14 x 10^-6 resulting in a limit of about λ_B>120 MeV at 90% confidence level. This limit is insufficient to provide stringent constraints for the theory. The BaBar analysis is performed with a simple cut-based method which has room for improvement to achieve a more significant result. This work presents the rst measurement of this decay by Belle. I performed an analysis of the B⁺→l⁺ν_lγ decay with the full Belle dataset of 772 x 10⁶B anti B pairs. Here, the charged lepton is either an electron or a muon and two separate analyses are performed with different energy requirements on the signal photon. The thesis is structured as follows. A discussion of the QCD factorization approach as well as a summary of the latest result of the calculation for the B⁺→l⁺ν_lγ decay is given in Section 2. Section 3 contains different analysis techniques which are essential to the analysis. The signal selection is described in Section 4. A detailed description of the fit model which is used to extract the signal as well as an examination of the fit bias is given in Section 5. Before the fit on data, sideband distributions are examined in Section 6 and the systematic error of the procedure is estimated in Section 7. For the systematic error, a control channel of B⁺→K^*(892)γ is analyzed to obtain the error on a neural network estimator which is crucial to the analysis. The measurement on data is described in Section 8, which is followed by a discussion with respect to the BaBar result mentioned above. Finally, the conclusion to the analysis is given in Section 9.