Optimization of the Gain Layer Design of Ultra-Fast Silicon Detectors

Abstract

In the past few years, the need of measuring accurately the spatial and temporal coordinates of the particles generated in high-energy physics experiments has spurred a strong R\&D in the field of silicon sensors. Within these research activities, the so-called Ultra-Fast Silicon Detectors (UFSDs), silicon sensors optimized for timing based on the Low-Gain Avalanche Diode (LGAD) design, have been proposed and adopted by the CMS and ATLAS collaborations for their respective timing layers. The defining feature of the Ultra-Fast Silicon Detectors (UFSDs) is the internal multiplication mechanism, determined by the gain layer design. In this paper, the performances of several types of gain layers, measured with a telescope instrumented with a $^{90}$Sr $\beta$-source, are reported and compared. The measured sensors are produced by Fondazione Bruno Kessler (FBK) and Hamamatsu Photonics (HPK). The sensor yielding the best performance, both when new and irradiated, is an FBK 45\mum-thick sensor with a carbonated deep gain implant, where the carbon and the boron implants are annealed concurrently with a low thermal load. This sensor is able to achieve a time resolution of 40~ps up to a radiation fluence of~\fluence{2.5}{15}, delivering at least 5~fC of charge.