Extreme Value Theory Applications to Space Radiation Damage Assessment in Satellite Microelectronics.

Area of Science:

• Materials Science
• Semiconductor Physics
• Radiation Effects

Background:

• Proton radiation poses a significant threat to silicon (Si) imaging arrays, particularly in space environments.
• Understanding displacement damage energy distributions is crucial for predicting device performance degradation.
• Pixel-to-pixel variance in damage can arise from various radiation interaction mechanisms.

Purpose of the Study:

• To model the probability of proton-induced displacement damage in silicon as a function of proton fluence and micro-volume size.
• To compare model predictions with measured leakage currents in Si imaging arrays.
• To apply extreme value statistics (EVS) for quantifying damage extremes and identifying anomalous pixel behavior (spikes).

Main Methods:

• Calculated first and second moments of displacement damage energy distributions for proton energies from 10 MeV to 300 MeV.
• Developed a model linking damage probability to proton fluence and sensitive micro-volume size.
• Utilized extreme value statistics (EVS) to analyze damage extremes and dark current spikes.

Main Results:

• The model accurately describes damage probability based on proton fluence and micro-volume.
• Poisson distribution of nuclear reaction recoils explains pixel-to-pixel damage variance in Si imaging arrays under prolonged proton exposure.
• EVS demonstrated excellent agreement with measured damage extremes and identified distinct mechanisms for high dark current pixels ('spikes') in different imager designs.

Conclusions:

• Proton-induced displacement damage in silicon is well-characterized by the developed model and EVS.
• Nuclear reaction effects significantly influence damage variance in imaging arrays, impacting long-term reliability.
• EVS is a powerful tool for predicting radiation-induced failures and understanding anomalous pixel behavior in sensitive electronic devices.