Dual-Bias Optimization of the Analog Front-End and the Pinned Depleted Diode in the CPV-4 Sensor
Abstract
Future electron–positron colliders require vertex detectors with high spatial resolution, low noise, and stringent power constraints. The compact pixel for vertex (CPV-4) prototype combines a pinned depleted diode (PDD) sensor with a silicon-on-insulator (SOI) analog front-end, whose performance is jointly determined by sensor bias and circuit operating conditions. This work presents a systematic experimental study of this bias dependence through a dual-bias optimization framework. The buried p-well (BPW) bias and front-end bias parameters are jointly tuned to investigate the balance between PDD capacitance reduction and back-gate-induced threshold shifts. Front-end-only pixels are used to characterize the circuit response; PDD-integrated pixels are used to evaluate sensor-front-end coupling; and near-infrared laser measurements are used to validate the response to sensor-generated signals. The results show that a sufficiently negative BPW bias is essential for stable operation of PDD-integrated pixels. Under optimized conditions, the gain of PDD-integrated pixels increases by approximately a factor of 4, while the equivalent noise charge (ENC) decreases by more than a factor of 2. Near-infrared laser measurements confirm sensor-generated signal detection at the operating point selected by the electrical bias scans, supporting the proposed optimization approach.