T-CAD analysis of electric fields in n-in-p silicon strip detectors in dependence on the p-stop pattern and doping concentration

Abstract

Detectors based on silicon have been proven to be efficient for particle tracking in high energy physics collider experiments. The Compact Muon Solenoid (CMS) Tracker at CERN near Geneva has a silicon detector surface of about 200 m2. The increasing demand on more collected data for new physics studies requires an upgrade of the Large Hadron Collider to higher luminosity which is foreseen for 2023. The increase of particles per time and area also introduces a harsh environment for the silicon sensors which should withstand an integrated luminosity of about L=3000 fb−1. Several R&D studies have been undertaken to face the high luminosity challenge and n-in-p detectors have been found to be more radiation hard than p-in-n. However, n-in-p detectors necessarily need an isolation layer of the n+ strips due to an accumulation layer of electrons caused by positive charge in the SiO2 at the sensor surface. An additional implantation of acceptors like boron between the n+ strips cuts the conducting electron layer and ensures reliable strip isolation. However, the implantation dose as well as the implant energy have to be carefully calculated as they directly affect the breakdown behavior and inter-strip resistance of the sensors. Experimentally, the inter-strip resistance (Rint) and charge collection efficiency (CCE) as well as the cluster charge formation are a direct indicator whether the isolation layer is sufficient or not. Furthermore T-CAD simulation studies have been carried out in order to reproduce the measurements and to predict the performance of sensors before and after irradiation with protons, neutrons and a mixture of both. Simulations also allow detailed studies of the formation of electric fields in the sensor which influence the sensor performance. Experimentally obtained CCE and misidentified hits in n-in-p devices with p-stop isolation pattern can be explained after analysis of the electric fields in dependence on the p-stop geometry and doping concentrations.The comparison of data and simulation results allows to limit the parameter space of the p-stop isolation technique which significantly affects the performance of irradiated n-in-p type silicon sensors.