Test beam track reconstruction and analysis of ATLAS 3D pixel detectors

3D silicon pixel sensors are a new radiation detection technology with electrodes etched into the silicon wafer. The technology offers some advantages over the originally installed ATLAS planar pixel sensors in radiation tolerance, as well as the possibility of being sensitive all the way to the edge of the sensor. ATLAS 3D sensors are used in the ATLAS insertable B-layer, a layer of pixel detectors that is inserted closer to the interaction point than the original innermost pixel layer. Several test beam experiments have been performed to characterize ATLAS 3D silicon devices in a particle beam produced by the CERN Super Proton Synchrotron. The object of the experiments was to study the response of the devices to particles with known trajectories. The particle trajectories are reconstructed from a beam telescope, an instrument made from position sensitive detector planes. This thesis summarizes the methods that are used, and the results obtained, in the analysis of ATLAS 3D silicon sensor data from the test beam experiments. This includes the reconstruction of particle trajectories from the beam telescope, and the characterizations of the devices under test. Some theoretical background is given in the first three chapters. Chapter 1 gives a brief introduction to the physics studied at the LHC, as well as the detector systems of the ATLAS experiment. Chapter 2 gives background on semiconductor detectors and the interactions between fast particles and matter. Chapter 3 provides background on least squares estimation and track fitting with the Kalman filter. The experimental setups of the test beam characterizations are summarized in Chapter 4. This includes descriptions of the two different beam telescopes that have been used in the characterization. The methods used for reconstructing particle trajectories from the beam telescope are presented in Chapter 5. This includes the steps needed to prepare the data for track reconstruction, obtaining a description of the detector system, and the track reconstruction procedure. Results showing excellent track quality in real data are presented. Chapter 6 discusses some of the methods used for the analysis of the performance of the ATLAS 3D silicon detectors. This thesis is based on four papers. The ATLAS public note “Straight line track reconstruction for the ATLAS IBL testbeam with the EUDET telescope” is a description of the methods that have been implemented for straight line track reconstruction in the EUDET beam telescope. This includes introducing a new method for track finding, and validation of all the implemented methods on real and simulated data. The paper “Optimizing track reconstruction by simultaneous estimation of material and resolutions” introduces novel methods for simultaneous estimation of detector resolution and material distribution. A correct description of the detector system is a requirement for the Kalman filter to give optimal estimates of the particle trajectories, and for the uncertainties of the estimates to be correct. Results from the characterization of 3D devices are presented in “Tracking Efficiency and Charge Sharing of 3D Silicon Sensors at Different Angles in a 1.4 Tesla Magnetic Field” and “Test Beam Results of 3D Silicon Pixel Sensors for the ATLAS upgrade”. 3D silicon pixel sensors have a detection efficiency similar to planar pixel sensors in conditions similar to the insertable B-layer. The signal distribution in planar pixel sensors is affected by the presence of the magnetic field in the insertable B-layer. For 3D silicon pixel sensors, this effect is much smaller. My work has consisted of taking part in the mounting and operation of the test beam experiments. I have held a leading role in the group working on analysis of the test beam data, and was the initial and main developer of the analysis software framework tbmon that was used for the analysis. This framework has since been used and further developed by other test beam experiments, including ATLAS planar pixels for the insertable B-layer. I have worked on data reconstruction, providing aligned and fitted tracks for the analysis group. I have implemented and developed several methods for track reconstruction of data from the EUDET telescope, including the information formulation of the Kalman filter, the combinatorial Kalman filter, the Deterministic Annealing Filter, and a new method called the cluster track finder. I have worked on improving the alignment estimation in the framework used by the EUDET telescope by configuring it to estimate tilt angles through scale factors, and by implementing a pre-alignment step. I have performed research and development of new methods for simultaneous estimation of sensor resolution and thicknesses for optimization of tracking performance.