Proton direct ionization (PDI) from low-energy protons has been shown to have a potentially significant impact on the accuracy of prediction methods used to calculate the upset rates (URs) of memory devices in space applications for state-of-the-art deep submicron technologies. The general approach nowadays is to consider a safety margin to apply over the UR computed from high-energy proton and heavy-ion experimental data. The data reported here present a challenge to this approach. Different UR prediction methods are used and compared in order to establish the impact of PDI on the total UR. Regardless of the method employed, the findings suggest that PDI can contribute to up to 90% of the total UR, on average, for a general selection of space orbits, with peaks of up to 99%. Such results suggest that an approach based on the characterization of the low-energy portion of the proton spectrum would be more convenient for similar technologies than the application of a general safety margin. Based on data presented here, the previously proposed margin of 5 is exceeded, by large amounts in some cases.

Convolutional Neural Networks (CNNs) are currently one of the most widely used predictive models in machine learning. Recent studies have demonstrated that hardware faults induced by radiation fields, including cosmic rays, may significantly impact the CNN inference leading to wrong predictions. Therefore, ensuring the reliability of CNNs is crucial, especially for safety-critical systems. In the literature, several works propose reliability assessments of CNNs mainly based on statistically injected faults. This work presents a software emulator capable of injecting real faults retrieved from radiation tests. Specifically, from the device characterisation of a DRAM memory, we extracted event rates and fault models. The software emulator can reproduce their incidence and access their effect on CNN applications with a reliability assessment precision close to the physical one. Radiation-based physical injections and emulator-based injections are performed on three CNNs (LeNet-5) exploiting different data representations. Their outcomes are compared, and the software results evidence that the emulator is able to reproduce the faulty behaviours observed during the radiation tests for the targeted CNNs. This approach leads to a more concise use of radiation experiments since the extracted fault models can be reused to explore different scenarios (e.g., impact on a different application).

Accelerated terrestrial neutron irradiations were performed on different commercial SiC power MOSFETs with planar, trench, and double-trench architectures. The results were used to calculate the failure cross sections and the failure-in-time (FIT) rates at sea level. Enhanced gate and drain leakage were observed in some devices which did not exhibit a destructive failure during the exposure. In particular, a different mechanism was observed for planar and trench gate MOSFETs, the first showing a partial gate rupture with a leakage path mostly between the drain and the gate, similar to what was previously observed with heavy ions, while the second exhibiting a complete gate rupture. The observed failure mechanisms and the postirradiation gate stress (PIGS) tests are discussed for different technologies.

The very limited availability of neutron facilities, particularly in Europe, for testing of microelectronics has motivated the construction of ChipIr, a new beamline at the ISIS neutron and muon source in the UK. ChipIr has been designed for Single Event Effect testing at the device-level and board-level, which requires a beam of uniform intensity over a selectable area in the order of hundreds of cm2. Measurements of the beam uniformity are presented in this paper. A memory chip of interest for space applications, an SRAM developed for monitoring of radiation fields by ESA, has been used for a comparative characterization of the beam. Consistent results have been found, giving confidence on the intensity and shape of the fast neutron spectrum of ChipIr.