Charge collection studies of MAPS in a 65 nm CMOS imaging process

Abstract

Monolithic Active Pixel Sensors (MAPS) are candidates to serve as vertex and tracking detectors in the future lepton colliders. This is due to the many advantages they provide compared to hybrid sensors, such as a reduced material budget and lower power consumption. The TANGERINE project aims for the development of the next generation of monolithic silicon pixel sensors produced in the 65 nm imaging process and investigates their possible use as vertex detectors in future lepton colliders. To investigate the 65 nm imaging process, simulations are implemented to reproduce the sensor response and thus save time and resources during the development phase. Technology Computer-Aided Design (TCAD) simulations permit the production of complex electric field profiles by the use of generic doping concentrations, a solution to the restricted access that commercial foundries implement in their fabrication process. A combination of Monte Carlo (MC) simulations using electrostatic TCAD fields was introduced to address the intensive computational requirements of using transient TCAD simulations. This allows the utilization of realistic electric fields combined with a high particle rate, resulting in simulations with high statistics. Furthermore, to include the effects that the electronics have on the signal produced by the silicon detector, Simulation Program with Integrated Circuit Emphasis (SPICE) is integrated into the simulation chain. <br/> The sensors used for the validation of the simulation chain are the ALICE Analogue Pixel Test Structure (APTS) are MAPS produced in a 65 nm CMOS imaging process designed at CERN for the CERN EP R&D program and are part of the studies for the ALICE ITS3 upgrade. In particular, the Analogue Pixel Test Structure Source Follower (APTS-SF) allows for the study of the different layouts and their effect on the signal due to several source follower stages used a simple electronic readout. <br/> This work showcases the results of a simulation approach combining TCAD, MC, and SPICE, a comparison of charge studies of these simulations with MAPS using a Fe-55 source for calibration and measurements taken at the DESY-II test beam facility using a 4 GeV electron beam. A discussion of the highlights and discrepancies between simulation and experimental data, and the possible reasons for these disagreements, is presented.