TY  - THES
A1  - Malygina, Hanna
T1  - Hit reconstruction for the Silicon Tracking System of the CBM experiment
N2  - The mission of the Compressed Baryonic Matter (CBM) experiment is to investigate the phase diagram of strongly interacting matter in the region of high net-baryon densities and moderate temperatures. According to various transport models, matter densities of more than 5 times saturation density can be reached in collisions between gold nuclei at beam energies  between 5 and 11 GeV per nucleon, which will be available at FAIR. The core detector of the CBM experiment is the Silicon Tracking System (STS), which is used to measure the tracks of up to 700 particles per collision with high efficiency (>95%) and good  momentum resolution (<1.5%). The technological and experimental challenge is to realize a detector system with very low material budget, in order to reduce multiple scattering of the particles, and a free-streaming data readout chain, in order to achieve reaction rates up to 10 MHz together with an online event reconstruction and selection. 
The STS comprises 8 tracking stations positioned between 30 cm and 100 cm downstream the target inside a magnetic field, covering polar emission angles up to 25 degrees. A station consists of vertical structures  with increasing number (between 8 and 16, depending on station number), each structure carrying  between 2 and 10 double-sided microstrip silicon sensors, which are connected through low-mass microcables to the readout electronics placed  at the detector periphery outside the active detector area. 
The work presented in this thesis focuses on  the detector performance simulation and local hit pattern reconstruction in the STS. For efficient  detector design and reconstruction performance, a reliable detector response model is of utmost importance. Within this work, a realistic detector response model was designed and implemented in the CBM software framework. The model includes non-uniform energy loss of an incident particle within a sensor, electric field of a planar p-n junction, Lorentz shift of the charge carriers, their diffusion, and the influence of parasitic capacitances. The developed model has been verified with experimental data from detector tests  in a relativistic proton beam. Cluster size distributions at different beam incident angles are sensitive to charge sharing effects and were chosen as an observable for the verification. Taking into account parasitic capacitances further improves the agreement with measured data. 
Using the developed detector response model, the cluster position finding algorithm was improved. For two-strip clusters, a new, unbiased algorithm has been developed, which gives smaller residuals than the Centre-Of-Gravity algorithm. For larger clusters, the head-tail algorithm is used as the default one. For an estimate of the track parameters, the Kalman Filter based track fit requires not only hit positions but their uncertainties as an input. A new analytic method to estimate the hit position errors has been  designed in this work. It requires as input neither measured spatial resolution nor information about an incident particle track. The method includes all the sources of uncertainties independently, namely: the cluster position finding algorithm itself, the non-uniform energy loss of incident particles, the electronics noise, and the discretisation of charge in the readout chip. 
The verification with simulations shows improvements in hit and track pull distributions as well as x²-distributions in comparison to the previous simple approach. The analytic method improves the track parameters reconstruction by 5-10%. 
Several STS module prototypes have been  tested in a relativistic proton beam. A signal to-noise ratio was obtained at the level of 10-15 for  modules made of 30 cm long microcable and of either one or two 6.2 x 6.2 cm² CiS sensors. 
First simulations have shown that this signal-to-noise ratio is sufficient to reach the required efficiency and momentum resolution. The high-radiation environment of CBM operation will deteriorate the sensor performance. Radiation hardness of sensors has been studied in the beam with sensors irradiated to 2 x 10[hoch 14] 1MeV [neq/cm²], twice the lifetime dose  expected for CBM operation. Charge collection efficiency drops by 17-25%, and simultaneously noise levels increase 1.5-1.75 times. The simulations show that if all sensors in the STS setup are exposed to such a fluence uniformly, the track reconstruction efficiency drops from 95.5% to 93.2% and the momentum resolution degrades from 1.6% to 1.7%.
Y1  - 2018
UR  - http://publikationen.ub.uni-frankfurt.de/frontdoor/index/index/docId/46401
UR  - https://nbn-resolving.org/urn:nbn:de:hebis:30:3-464018
N1  - Mathematische Zeichen können technischbedingt in der Zusammenfassung teilweise nicht korrekt dargestellt werden
CY  - Frankfurt am Main
ER  -