Speaker
Description
Low-Gain Avalanche Diode (LGAD) sensors for the CMS MTD Endcap Timing Layer must retain charge collection (> 8 fC after irradiation) and timing performance at <50 ps level after fluences up to $2.5×10^{15}n_{\mathrm{eq}}/\mathrm{cm}^{2}$, while operating cold ($-25 ^{\circ}\mathrm{C}$) with sufficient bias-voltage margin. We present radiation-tolerance studies of LGAD sensors proton-irradiated at KOMAC. Stepwise irradiation was followed by I-V/C-V and MIP-like signal measurements to quantify leakage current, gain-layer depletion voltage, breakdown behavior, collected charge, and timing resolution versus bias. The study evaluates the bias-voltage operating window required to recover radiation-degraded gain while maintaining charge and timing performance within ETL sensor-qualification requirements.
Summary (500 words)
Low-Gain Avalanche Diode (LGAD) sensors are the baseline precision-timing sensors for the CMS MTD Endcap Timing Layer (ETL). In the high-fluence region of the ETL, the sensors must retain sufficient charge collection and timing performance after irradiation while operating cold with adequate bias-voltage margin. The key engineering challenge is radiation-induced acceptor removal in the p⁺ gain layer, which reduces the internal electric field and avalanche gain. This loss can be partially compensated by increasing the bias voltage, but the usable range is constrained by leakage current, noise, breakdown behavior, thermal stability, and single-event-burnout limits. The relevant qualification metric is therefore the operating-voltage window satisfying ETL charge and timing requirements after irradiation.
This contribution presents radiation-tolerance characterization of LGAD sensors proton-irradiated at the KOMAC facility in Gyeongju. The campaign used the 20 MeV proton target room, with stepwise fluence levels corresponding to partial and full ETL end-of-life targets after conversion to 1 MeV neutron-equivalent fluence using a hardness-factor approach. Pixelated LGAD sensors were characterized before and after irradiation. Probe-station I–V measurements quantified leakage-current evolution and breakdown behavior, while C–V measurements extracted the gain-layer depletion voltage, V_GL, from linear-fit intersections to quantify gain-layer degradation.
The current electrical results show two clear radiation-induced trends. First, leakage-related currents increase significantly with irradiation and rise more rapidly at high bias voltage. At 300 V, the 100% ETL-equivalent irradiated device shows currents at the 10–50 μA level, compared with approximately 0.05–0.25 μA before irradiation, corresponding to an increase of roughly two orders of magnitude. This directly constrains thermal load, noise level, and stable high-voltage operation. Second, for a representative device, |V_GL| decreases from approximately 20.13 V to 12.97 V with irradiation. This shift is a direct electrical signature of acceptor removal and defines the bias increase required to compensate radiation-degraded gain.
To connect electrical degradation to detector performance, MIP-like signal measurements are performed using a Sr90 beta-source setup to mimic minimum-ionizing particles. The readout chain uses a GHz-scale broadband amplifier and high-speed oscilloscope to measure collected charge, signal shape, rise time, slew rate, time-walk behavior, and timing resolution versus bias voltage. Complementary SPS beam-test results, if available, will be used as a cross-check. The final performance metric is the voltage interval over which irradiated sensors provide at least 8 fC of collected charge and timing performance at the <50 ps level, while maintaining sufficient margin from breakdown and excessive-noise operation.
By combining KOMAC proton irradiation, I–V/C–V characterization, and MIP-like charge/timing measurements, this work evaluates LGAD radiation tolerance in terms of an electronics-relevant operating margin. The results constrain the ETL front-end operating range, including the signal amplitude available to ETROC, threshold stability, noise tolerance, and time-walk correction performance under HL-LHC radiation conditions.