Christian Roca Catala ; Christian Kiesling ; Luigi Li Gioi

2015
Ludwig-Maximilians-Universität - Max-Planck-Institut für Physik Munich

Abstract: This thesis consists of two correlated but differentiated parts, both of them regarding the optimization of the vertexing procedure of the Belle II experiment. The time dependent analysis of the decay mode: B0(B0bar) -> J/psi Ks , B0bar(B0) -> generic has been used as benchmark for the Belle II vertexing performance. In order to perform this analysis the vertex of the two exclusively produced B mesons must be determined. The first neutral B meson, usually called BCP , is fully reconstructed then its decay vertex can be calculated with a kinematic fit of the two tracks corresponding to the two muons of the J/Â decay. The vertex of the other B is determined without reconstructing it. The complete reconstruction, in fact, would produce a very low efficiency. The usual procedure is to take all the tracks remaining after the signal reconstruction and fit them in one single vertex. For this purpose, a spatial constraint has been designed so that contributions of tracks pointing to secondary vertices of the tag side weight less during the fit. This prevents potential bias on the Btag vertex. This procedure is called Standard Algorithm. The first part of this study treats about the Standard Algorithm. More concretely, a deep analysis about the stability of the spatial constraint used in the algorithm is made. Shifts in different directions are applied to the center of the constraint, allowing the stability study. The results are very conclusive: a part from extreme cases, the shifts applied had little to no effect in the Tag Side vertex resolution. The inclusion of the newly developed Pixel Vertex Detector in the Belle II detector simulation has improved the full reconstucted B vertex resolution with respect to the Belle results by almost a factor 3. Nevertheless, this vertexing improvement cannot be translated to the tag side as its resolution depends heavily on the fitting algorithm. The second part of this thesis explores the idea of doing a selection of the tracks, that are more suitable for the Btag vertex fit. The track’s selection is performed by analysing different properties of the tracks than could indicate whether the track comes either directly from the Btag (primary track) or from a secondary vertex (secondary track). The main intention is to substitute or complement the Standard Algorithm. Two options are studied, called Secondary Track Rejection and Single Track Fit: .) Secondary Track Rejection: selection criteria are applied to reject secondary tracks. This method ends up being unsuccessful, leading to a worse resolution compared to the one obtained with the Standard Algorithm. .) Single Track Fit: this algorithm selects only one track to perform the fit, ideally a primary track. The results are very promising since the resolution is improved by a factor 1.4 with respect to the results obtained via Standard Algorithm. The total improvement with respect to the Belle collaboration’s results is of a factor 2.2. As the efficiency of this algorithm is about a 15%, it needs to be combined with the Standard Algorithm.

Note: Presented on 30 09 2015
Note: MSc