⚛️ nuclear experiments

Impact of Environmental Stress on Low Gain Avalanche Diode Sensors Response

This paper presents a systematic evaluation and simulation-based analysis of how environmental stressors, specifically temperature and humidity variations, impact the performance of Low Gain Avalanche Diode (LGAD) sensors fabricated at Brookhaven National Laboratory.

Original authors: Mohamed Hijas Mohamed Farook, Gaetano Barone, Diallo Boye, Gabriele D'Amen, Gabriele Giacomini, Timo Peltola, Jennifer Roloff, Enrico Rossi, Trevor Russell, Sally Seidel, Alessandro Tricoli, Lixing Wa

Published 2026-08-25

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Mohamed Hijas Mohamed Farook, Gaetano Barone, Diallo Boye, Gabriele D'Amen, Gabriele Giacomini, Timo Peltola, Jennifer Roloff, Enrico Rossi, Trevor Russell, Sally Seidel, Alessandro Tricoli, Lixing Wang

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

In the vast, invisible world of particle physics, scientists rely on silicon sensors to catch fleeting glimpses of subatomic particles. These sensors act like highly sensitive cameras, recording the exact moment a particle passes through. To do this with the precision required for modern experiments, the sensors must be incredibly fast and accurate, capable of measuring time down to the billionth of a second. A specific type of silicon sensor, known as a Low-Gain Avalanche Diode, has emerged as a leading technology for this task. Its secret lies in an internal mechanism that amplifies the tiny electrical signal created by a passing particle, making it easier to detect and time with extreme accuracy. However, these delicate devices are not just used in climate-controlled laboratories; they are destined for use in massive experimental setups where they must operate reliably despite fluctuations in the surrounding air. Just as a musician's instrument can go out of tune if the temperature shifts or the air becomes too damp, these silicon sensors can change their behavior when exposed to different environmental conditions. Understanding exactly how heat and moisture affect their performance is essential to ensuring they function correctly in the real world.

A team of researchers at Brookhaven National Laboratory and several universities set out to map these changes with rigorous precision. They took three different silicon sensors—two standard models and one with a slightly different design—and subjected them to a series of intense environmental stress tests. The goal was not to break the sensors, but to observe how their electrical signals shifted when the temperature swung from freezing cold to scorching hot, and when the humidity in the air varied from dry to damp. The sensors were placed inside a specialized chamber where scientists could control the climate with exacting detail. They ran the devices through cycles that mimicked a full day of rapid temperature changes, long nights of stable conditions, and even weekends where the temperature was shocked from mild to extreme heat. Throughout these tests, the researchers monitored the electrical current flowing through the sensors, looking for any signs of instability or degradation.

The results painted a clear picture of how these sensors behave under stress. The researchers found that the voltage required to fully activate the sensor remained stubbornly constant, regardless of whether the air was hot, cold, dry, or humid. This stability is a good sign, meaning the fundamental operating point of the device does not drift. However, the amount of electrical "noise" or leakage current did change significantly with temperature. As the sensors got hotter, the background electrical noise rose sharply, a behavior that is well understood in semiconductor physics but needed to be quantified for these specific new devices. The team also discovered that the voltage at which the sensor begins to avalanche, or amplify the signal, increases in a straight, predictable line as the temperature rises. This relationship was so consistent that the researchers could use computer simulations to model it with high accuracy, confirming that their understanding of the physics was correct.

Humidity presented a different kind of challenge. The sensors performed perfectly well in dry air, but as the moisture in the chamber increased, a critical threshold was reached. When the air became humid enough that the temperature was only about ten degrees above the point where water would condense on the sensor, the electrical leakage current began to spike dramatically. This suggests that a thin film of moisture on the surface of the sensor creates a path for electricity to leak, degrading the signal quality. Interestingly, the newer sensors, which featured an improved protective coating and better electrical contacts, showed much lower leakage currents than the older models, proving that design improvements can make the devices more robust against environmental stress.

Throughout weeks of testing, the sensors showed no signs of permanent damage or long-term deterioration. They withstood the repeated thermal shocks and humidity swings without losing their ability to function. The study confirmed that while temperature and humidity do alter the electrical characteristics of these sensors, the changes are predictable and manageable. The breakdown voltage, which marks the limit of safe operation, depends on the thickness of the silicon layer inside the sensor, with thicker layers requiring higher voltages to reach that limit. By combining real-world measurements with detailed computer models, the researchers established a reliable framework for predicting how these sensors will behave in any environment. This work ensures that when these sensors are deployed in the field, scientists can trust them to deliver precise timing data, even when the weather outside is far from perfect.

Technical Summary: Impact of Environmental Stress on Low Gain Avalanche Diode Sensors' Response

Problem Statement
Low Gain Avalanche Diodes (LGADs) and their AC-coupled variants (AC-LGADs) are critical silicon sensors for High Energy and Nuclear Physics, offering timing resolutions on the order of 30 ps or less and precise particle tracking. Their performance relies on an internal gain layer that facilitates impact ionization. However, the operational robustness of these sensors under varying environmental conditions—specifically temperature fluctuations, humidity changes, and storage stress—remains a critical concern for their deployment in diverse experimental setups. While silicon device physics suggests that leakage current is sensitive to temperature and surface conditions, a systematic evaluation of how rapid and repeated environmental stress affects the electrical performance of BNL-fabricated LGADs and AC-LGADs is essential for quality assurance and long-term reliability.

Methodology
The study employed a systematic characterization of three sensors fabricated at Brookhaven National Laboratory (BNL): two LGADs (from wafers W3045 and W3058) and one AC-LGAD (from wafer W3081). The sensors were subjected to rigorous environmental stress testing within a climate chamber capable of controlling temperature (-68°C to +180°C) and humidity.

The experimental protocol included three distinct thermal cycling programs:

Day Program: Temperature ramped from -60°C to +120°C in 20°C steps with IV scans performed at each step to characterize electrical response during thermal transitions.
Night Program: Sensors held at a stable +21°C and -150 V bias for approximately 14 hours to test stability under constant conditions.
Weekend Program: Rapid thermal shock cycling between +10°C and +120°C over two days with the sensors unbiased.

Humidity dependence was characterized by varying relative humidity (5% to 70%) at fixed temperatures (0°C to 25°C). Leakage currents (pad, guard ring, and DC-contact) were monitored using high-precision instruments (Keithley 2410, HP 4145B, Keithley 6482) while maintaining the sensor temperature above the dew point to prevent condensation.

Complementing the experimental data, two-dimensional TCAD simulations were performed using Synopsys Sentaurus. A survey of impact ionization models led to the selection of the Massey model, which best reproduced the measured temperature effects on avalanche onset. The simulations utilized doping profiles derived from SIMS data and were tuned to match the specific geometries and electrical properties of the 20 and 50 thick sensors.

Key Results

Depletion Voltage ( $V_{FD}$ ): The full depletion voltage was found to be independent of both temperature and humidity for all tested sensors. It remained constant across all thermal cycles, depending solely on fabrication parameters such as doping concentrations.
Leakage Current:
- Temperature Dependence: Leakage current increased exponentially with temperature for all sensors. The guard ring current showed a monotonic increase of up to five orders of magnitude across the tested temperature range.
- Humidity Dependence: A critical threshold was identified for humidity effects. For the LGAD sensors, leakage current began to increase significantly when the relative humidity exceeded approximately 50%, or when the operating temperature was within 10°C of the dew point. Below this threshold, humidity had negligible impact on leakage current.
- Design Comparison: The LGAD from wafer W3058 (featuring improved $p^+$ termination and $Si_3N_4/SiO_2$ passivation) exhibited lower leakage current compared to the earlier W3045 design. The AC-LGAD (20) showed higher leakage current at elevated temperatures and a steeper temperature dependence compared to the 50 LGADs, attributed to process differences.
Breakdown Voltage ( $V_{BD}$ ):
- $V_{BD}$ increased linearly with temperature for all sensors.
- The rate of change (slope) of $V_{BD}$ with respect to temperature was dependent on the epitaxial layer thickness. The 50 LGADs exhibited a slope of approximately 1.1 V/°C, while the 20 AC-LGAD showed a lower slope of approximately 0.41 V/°C.
- No dependence of $V_{BD}$ on the number of thermal cycles or humidity was observed.
Simulation vs. Experiment: TCAD simulations using the Massey model successfully reproduced the experimental trends. The simulated slopes for $V_{BD}$ versus temperature (1.14 V/°C for 50 and 0.42 V/°C for 20) closely matched the experimental data, validating the analytical doping profiles used in the models.

Significance and Conclusions
The paper establishes that LGAD and AC-LGAD sensors fabricated at BNL demonstrate robust performance under extreme and rapidly changing environmental conditions. The study confirms that while leakage current is highly sensitive to temperature and high humidity (specifically near the dew point), the fundamental operational parameters—full depletion voltage and breakdown voltage—remain stable.

The research highlights that improved passivation and termination designs (as seen in the W3058 wafer) effectively mitigate leakage current. Furthermore, the linear dependence of breakdown voltage on temperature, and its variation with epitaxial thickness, provides a predictable framework for sensor operation. The strong agreement between experimental data and TCAD simulations confirms that the physical models used accurately describe the device behavior, allowing for reliable prediction of sensor performance in varying environmental scenarios. The authors conclude that these sensors maintain their integrity and performance metrics even after weeks of thermal cycling, supporting their viability for long-term deployment in diverse experimental environments.

More like this