Performance of Radiation Hard Pixel Sensors for the CMS Experiment

Position sensitive detectors in particle physics experiments are used for the detection of the particles trajectory produced in high energy collisions. To study physics phenomena at high energies the high particle interaction rate is unavoidable, as the number of interesting events falls with the energy and the total number of events is dominated by the soft processes. The position resolution of vertex detectors has to be of few microns in order to distinguish between particle tracks produced in b-quark or tau-decays, because of the short flight path before the decay. The high spatial position resolution and the ability to detect a large number of superimposed track are the key features for tracking detectors. Modern silicon microstrip and pixel detectors with high resolution are currently most suitable devices for the tracking systems of high energy physics experiments. In this work the performance of the sensors designed for the CMS pixel detector are studied and the position resolution is estimated. In the first chapter an introduction to the LHC and the CMS experiment is given. In addition, the CMS pixel detector and its sensors designs are described. In the second chapter the physical processes in semiconductor position sensitive detectors are discussed and the effects due to irradiation are described. In the third chapter the beam test setup used for the sensors study is presented and the data reconstruction is described. The fourth chapter presents the data analysis and the results. The charge collection e±ciency and the Lorentz angle are measured. The charge collection e±ciency is about 60% of the unirradiated sensors after a fluence up of 1x10^15 neq/cm2 . This value drops to 25% after a fluence of 2.6x10^15 neq/cm2 . The p-spray design and p-stop design with two openings exposed to a fluence of 6x10^14 neq/cm2 have a particle detection efficiency of 99% with a threshold of 2000 electrons. The respective value for the p-stop design with one opening is 95%. The Lorentz angle does not depend on irradiation or sensor design but strongly depends on the bias voltage. The Lorentz angle with a magnetic field of 4 T is about 26degrees for the unirradiated devices and a bias voltage of 100 V. It drops to 8.3degrees for the sensors irradiated at 1x10^15 neq/cm2 and a bias voltage of 600 V. A new method for the extraction of the electric field in the silicon sensor bulk is developed and applied to the data. The method is based on the measurement of the Lorentz deflection of the charge carriers in the sensor bulk. The measured electric field is implemented in the sensors simulation and the performances of the CMS pixel detector are estimated in the fifth chapter. The simulation is validated with the test beam data and is used to estimate the position resolution of the CMS pixel detector. The resolution of the pixel barrel sensors along the azimuthal a ngle is in the range between 10 mum and 20 mum. It strongly depends on irradiation and weakly depends on the polar angle of the particle track. The position resolution along the beam direction averaged over the azimuthal angle is in the range between 15 mum and 40 mum. In this case the resolution along the beam direction weakly depends on irradiation and strongly depends on the polar angle of the particle track.