Search

Commercial off-the-shelf (COTS) system-on-chip (SoC) are becoming widespread in embedded systems. Many of them include a multicore central processing unit (CPU) and a high-end graphics processing unit (GPU). They combine high computational performance with low power consumption and flexible multilevel parallelism. This kind of device is also being considered for radiation environments where large amounts of data must be processed or compute-intensive applications must be executed. In this article, we compare three different strategies to perform matrix multiplication in the GPU of a Tegra TK1 SoC. Our aim is to analyze how the different use of the resources of the GPU influences not only the computational performance of the algorithm, but also its radiation sensitivity. Radiation experiments with protons were performed to compare the behavior of the three strategies. Experimental results show that most of the errors force a reboot of the platform. The number of errors is directly related with how the algorithms use the internal memories of the GPU and increases with the matrix size. It is also related with the number of transactions with the global memory, which in our experiments is not affected by the radiation. Results show that the smallest cross section is obtained with the fastest algorithm, even if it uses the cores of the GPU more intensively. This work was supported in part by the Valencian Regional Government under Grant PROMETEO/2019/109, in part by Jaume I University under Project UJIB2019-36, and in part by the Spanish Ministry of Science and Innovation under Project PID2019-106455GB-C21 and Project PID2020-113656RB-C21. Publicado

Effects of ~35 years of aging during storage are investigated on the radiation response and 1/f noise of Oki nMOS and pMOS transistors with high oxygen vacancy densities in SiO2. Short-term interface-trap buildup during irradiation is enhanced significantly, relative to that observed in 1989, 1997, and 2008. In contrast, similar latent interface-trap buildup is observed for aged pMOS devices irradiated and annealed at room and elevated temperatures at positive bias in this and earlier studies. Significant latent interface-trap buildup is observed for nMOS devices irradiated at 0 V and annealed at room and elevated temperatures under positive bias, a condition not evaluated in prior work. Results strongly suggest that latent interface-trap buildup is due to H2 diffusion and dissociation at charged or dipolar O vacancies in SiO2, followed by proton transport to the Si/SiO2 interface and reactions with Si-H complexes. Models that attribute latent interface-trap buildup to long-term proton trapping at O vacancies in SiO2 appear to be ruled out by these results. Additional insight is provided into mechanisms of post-irradiation interface- and border-trap buildup after long-term MOS device storage.

The combined effects of proton irradiation and forward gate-bias stress on the interface traps of AlGaN/GaN heterostructure have been studied in this article. It is found that the effect of proton irradiation and forward gate-bias on the shift of flat band voltage $V_{\mathrm {FB}}$ is independent. By utilizing the frequency-dependent conductance technique, it is found that the trap density $D_{\mathrm {T}}$ at metal/AlGaN interface decreases after proton irradiation and the $D_{\mathrm {T}}$ at most energy levels increases after the following forward gate-bias. The $D_{\mathrm {T}}$ at AlGaN/GaN interface increases after proton irradiation and then decreases after the following forward gate-bias. The energy level range of metal/AlGaN interface traps reduces significantly under the forward gate-bias for the irradiated devices, however, that of AlGaN/GaN interface traps decreases little. In summary, the combined effect of proton irradiation and forward gate-bias stress on the interface traps is more complex than that on the bulk traps in AlGaN layer.

The contribution of Coulomb processes induced by incident electrons in single event upset (SEU) sensitivity is investigated. The case of 65-nm silicon on insulator (SOI) memory is studied. Several orbits (GPS, Jupiter Icy Moons Explorer’s (JUICE’s) orbits) are analyzed. The relative importance to the total SEU rate of electron, proton, and gamma environments is compared. SEU induced by electrons is obtained during Earth missions. In the environment of Jupiter, equivalent SEU sensitivities have been obtained between the studied device and the European Space Agency (ESA) SEU monitor. However, the relative importance of the physical processes involved is very different. Proton-induced SEU rates are estimated about one decade greater than electron-induced SEU rates. The Bremsstrahlung contribution has been found negligible in such an environment.

The effects of buried-well potential perturbation under the buried-oxide (BOX) layer are studied in both a heavy-ion single event upset (SEU) test and a high-energy proton-SEU test of a silicon-on-insulator (SOI) dual interlocked storage cell (DICE)-based flip-flop. Their dependence on incident angle and back bias is discussed. We fabricated both DICE-based flip-flop and conventional flip-flop, which are designed as 80 000-stage shift-register chains. In a heavy-ion test, a considerable number of SEUs were observed at back bias exceeding 2.4 V, and a ten-times larger SEU-cross section was finally recorded at back bias of 3.0 V compared with the total active area of a DICE-based flip-flop cell. This marks the first case where DICE topology was found to be broken by buried-well potential perturbation on an SOI DICE-based flip-flop. In a proton test, one error was observed at back bias of 2.0 V. The SEU rate in the Van Allen belt at an altitude of 2300 km and an inclination of 90° was estimated as being once every 5 years.

Single-event upsets are observed in a 72-layer 3-D NAND flash memory operated in a single-level cell mode after low-energy proton (500 keV–1.2 MeV) and heavy-ion irradiation. The layer-by-layer error count is analyzed to visualize the stopping of low-energy protons within the memory stack, and Monte Carlo simulations are correlated with the experimental data. Direct ionization by low-energy protons is identified by 3-D data analysis and the energy dependence of device-sensitive cross section. Heavy-ion data is also presented for comparison.

The correct quantification of the dose released in charged particle therapy treatments requires the knowledge of the double differential fragmentation cross section of particles composing both the beam and the target. The FragmentatiOn Of Target (FOOT) experiment aims at measuring these cross sections for ions of interest for charged particle therapy applications. This article describes the performance of the time-of-flight (TOF)-wall detector of the experiment. The detector is composed of two layers of 44 cm $\times 2$ cm $\times 3$ mm plastic scintillator bars (20 for each layer), arranged orthogonally and read-out by silicon photomultipliers. The detector is designed to identify the charge of fragments ranging from protons to oxygen ions, with a maximum energy of 700 MeV/u, by measuring the energy released in the scintillators and the TOF with respect to a start counter. In this study, the detector was scanned with carbon ions of energy between 115 and 400 MeV/u and with a 60-MeV proton beam. The measurements show an energy resolution ( $\sigma _{E}/\mu _{E}$ ) between 6% and 4% and a contribution of the detector to the TOF system time resolution between 25 and 20 ps (standard deviation) for carbon ions and between 100 and 80 ps for protons.

Space radiation damage and proton-induced transient effects were evaluated on 4.4- $\mu \text{m}$ cutoff HgCdTe avalanche photodiode (APD) arrays developed by Leonardo DRS. Device performances as a function of total dose up to 100 krad (Si) were measured with ~60-MeV protons on three types of APD samples: $4 \times 4$ pixel APD fanout arrays with and without connection to a read-out integrated circuit (ROIC) and a $2 \times 8$ pixel photon-counting APD focal plane array (FPA). A gamma-ray test was also conducted to study ionization effects. Both APD arrays exhibited a small decrease in the quantum efficiency and a linear increase in the dark current with the proton fluence. The $2 \times 8$ pixel photon-counting FPA also exhibited an increase in the dark count rate with proton dose. After the proton irradiation and an overnight room-temperature warm-up, the APD dark currents at 80 K increased significantly in both types of APD arrays. All radiation damage to these HgCdTe APD arrays annealed out after baking them at >85 °C for several hours. Transient protons through the devices were found to cause large pulses at the detector output, but recover within $1~\mu \text{s}$ .

Owing to their low dark current, high transparency, high thermal conductivity, and potential radiation hardness, there is a special interest in silicon carbide (SiC) devices for radiation monitoring in radiation harsh environments and with elevated temperatures and, especially, for the plasma diagnostic systems in future nuclear fusion reactors. In this work, four-quadrant p-n junction diodes produced on epitaxial 4H-SiC substrates are studied. The impact of electron, neutron, and proton irradiations (up to fluences of 1 × 10 16 electrons (e)/cm 2 , 2 × 10 15 neutrons (n)/cm 2 , and 2.5 × 10 15 protons (p)/cm 2 , respectively) on the electrical characteristics is studied by means of current-voltage (I-V) and capacitance-voltage (C-V) techniques. Regardless of the particle type and applied fluences, the results show similar low reverse currents for irradiated SiC devices, which are at least about four orders of magnitude lower than comparable Si devices. The effects of irradiation on interquadrant resistance and charge build-up in the interquadrant isolation are assessed. Furthermore, device performance as a radiation detector is investigated upon exposure to a collimated 239 Pu- 241 Am- 244 Cm trialpha source. The performance at room temperature is preserved even for the highest irradiation fluences, despite the fact that the rectification character in electrical characteristics is lost. From the results, advantages of using SiC devices in alpha particle detection in harsh environments can be envisaged., This work was supported in part by the Spanish Ministry of Science, Innovation and Universities through the Nuclear and Particle Physics Program under Project FIS-FPN-RTI2018-094906-B-C22 (MCIU/FEDER UE), in part by the European Union’s Horizon 2020 Research and Innovation Program under Grant 654168 (AIDA-2020), in part by a collaborative research project at Nuclear Professional School, School of Engineering, The University of Tokyo, under Grant 20016, in part by The Japan Society for the Promotion of Science KAKNHI under Grant JP19K05337, and in part by MINECO through the use of the Spanish ICTS Network MICRONANOFABS. The work of Gemma Rius was supported by the Spanish Ministry of Science and Innovation through Ayudas Ramón y Cajal 2016 under Reference RYC-2016-21412.

The nonionizing energy loss (NIEL) concept has been used for many years for qualitative estimation of the damage induced by radiation (particles and gamma rays). Different approaches, based on the physics of radiation interaction with the nuclear and electronic subsystems of the target atoms, are used for calculating NIEL values. The simplest model used is the binary collision approximation (BCA), which is applicable for energies larger than the threshold energy for Frenkel pair creation. In this article, we analyze the dependence of the calculated NIEL values (in silicon and germanium) on the basic interaction characteristics: the differential cross section (DCS) of energy transferred to the recoiling atoms and the partition factor between ionization and ion displacement. We estimate the differences between existing approaches and the possible scatter of the NIEL data for Coulomb, elastic, and inelastic nuclear scatterings. We also present new partition factors based on fits to experimental data. The contribution of low-energy cascade zones of displaced atoms (pockets) to the total damage is estimated using results of molecular dynamics (MD) calculation. We find for silicon a total increase of NIEL by 30%–40% relative to the original (BCA) NIEL value. The role of phonon excitations in the subthreshold cascades on the damage value is discussed.

The effects of the white-light X-ray and 170-keV proton beam irradiation on the electrical and optical characteristics of the InGaN/GaN multiple quantum wells (MQWs) light-emitting diodes (LEDs) are analyzed and compared. Different from the negative effects of the proton irradiation, the X-ray irradiation shows positive effects on the LEDs’ performance. In detail, after the 100 Mrad(Si) X-ray irradiation, the p-n junction resistance decreases from the original 2.51 to $2.08~\Omega $ , the light output power increases from 170 to 203 mW (at a forward voltage of 3.2 V), and the maximal external quantum efficiency (EQE) increases from 56.7% to 61.8%. Based on the ABC model fittings, both the Shockley–Read–Hall recombination rate and the radiative recombination rate in the MQWs are improved by the X-ray irradiation. The Hall effect measurements reveal the chemical bond breaking of the Mg-H complex in the p-type GaN. Therefore, the improvement of the LED by the X-ray irradiation should be caused by the chemical bond variations of the defect-related complex in the MQWs and the Mg-H complex in the p-type GaN.

Influences of 170-keV protons beam irradiation on the static and dynamic properties of blue light InGaN/GaN multiple quantum wells (MQWs) microlight-emitting diodes (Micro-LEDs) were investigated. It was interesting to find out that, although threshold voltage and light output power of Micro-LEDs deteriorated after proton irradiation, a 3-dB bandwidth was greatly improved. In quantitative terms, at the forward current density of 1 kA/cm2, 3-dB bandwidth increased from original 7.34 to 119.74 MHz when Micro-LED exposed to proton beam with the fluence of $5\times 10^{14}$ p/cm2. Based on the frequency response data analysis, differential carrier lifetimes of Micro-LEDs, including Shockley–Read–Hall (SRH) lifetime and differential radiation recombination lifetime, were compared. The results pointed out that both SRH and recombination lifetimes became shorter after proton irradiation, indicating that competition between the nonradiative and radiative recombination processes was enhanced by proton beam. To reveal the origination of the 3-dB bandwidth improvement, photoluminescence (PL) and time-resolved PL (TRPL) spectrums of Micro-LEDs were measured. MQWs’ PL spectrum with a peak wavelength of 450 nm was observed and its intensity decreased as proton fluence increasing. Meanwhile, a PL peak at 550 nm, which was well known as defect-related PL spectrum, was enhanced by proton irradiation, especially at a proton fluence of $5\times 10^{13}$ and $5\times 10^{14}$ p/cm2, proving the increase of defects in epitaxial thin films as proton fluence increasing. In the TRPL experimental study, the nonradiative recombination lifetime decreased with proton fluence, which was consistent with the results analyzed by frequency response. Overall, the improvement of 3-dB bandwidth could be mainly attributed to the decrease of carrier lifetime in MQWs, which were caused by the generation of defects due to the atom displacement effect of proton irradiation.

The electrical characteristics and low-frequency noise (LFN) of AlGaN/GaN high-electron-mobility transistors (HEMTs) after 3-MeV proton irradiation up to a total dose of 5.0 $\times 10^{14}$ p/cm2 have been studied in this article. The devices experienced a 14.6% decrease in the saturation current, a 0.35-V positive shift in the threshold voltage ( $V_{\mathrm {th}}$ ), and a significant decrease in the reverse gate leakage current after 3-MeV proton irradiation. Extracted by the LFN method, the flat-band voltage noise power spectral density ( $S_{{V{\textrm {fb}}}}$ ) increases from 4.112 $\times \,\,10^{-12}\,\,\cdot $ Hz−1 to $5.603\times 10^{-12}\,\,\cdot $ Hz−1, and the trap density increases from $2.79\times 10^{17}$ to $3.84\times 10^{17}$ cm $^{-3}\,\,\cdot $ eV−1 after the radiation. The main degradation mechanism is considered to be the increase of negatively charged trap density in the channel, which depletes more electrons and reduces the carrier mobility in the channel 2-D electron gas (2DEG). We also extracted the radiation-induced trapped charge density according to the theoretical calculation; the result is consistent with the LFN measurement.

Fissionable materials can be distinguished from normal materials by their emission of fast neutrons and neutrons moderated by surrounding materials. Radioisotopes can often be identified by their gamma-ray spectra. In this work, we use the nuclear physics Monte Carlo package MCNP6.2 to compare the detection efficiencies of different detector designs for the triple-mode detection of fast and moderated neutrons, and gamma-ray spectroscopy. For each design, we tabulate the detection efficiencies for neutrons from 0.1 eV to 5 MeV and for the full absorption of gamma rays from 0.1 to 5 MeV. All designs use 100 scintillators 1 cm $\times $ 1 cm $\times $ 5 cm. Of these, 36 are Cs2 LiLaBr6:Ce (90% 6Li) (CLLB) and 64 are polyvinyl toluene (PVT). The first design uses separate arrays of $6\times 6$ CLLB and $8\times 8$ PVT scintillators. Three other designs are $10\times 10$ composite arrays using all 100 scintillators combined in different ways. PVT is a fast scintillator that can be used to detect fast neutrons by time-separated multiple proton recoils and discriminate against gammas. PVT in the three composite designs serves to moderate neutrons of all energies for capture by the 6Li in the CLLB. Gamma rays interact in the PVT and CLLB by Compton scattering and pair production; full energy absorption requires photoelectric absorption in the CLLB. The energy deposited by each gamma ray can be determined from a weighted sum of the CLLB and PVT signals from separate photodetectors. We find that for a fixed number of neutrons/cm2, one of the composite arrays has much higher 6Li detection efficiencies than the design that uses separate CLLB and PVT arrays, and has similar detection efficiencies for fast neutrons and gammas. We also find that in all cases, the average energy of the neutrons at capture by 6Li is well above the 25-meV thermal energy.

Proton-induced fission of high- $Z$ materials can produce fluxes of high linear energy transfer (LET) ions that greatly exceed those from galactic cosmic ray (GCR) or even solar particle event (SPE) environments. These ions can pose significant threats to microelectronics. We develop methods to evaluate the risks for a range of destructive and nondestructive single-event effects (SEE) modes caused by this threat by evaluating the role of the proton environment, the amount and placement of high- $Z$ materials and overburden in the part, and the sensitive volume geometry for the SEE modes under consideration.

The radiation hardness of a fully depleted pinned photodiode (PPD) CMOS image sensor (CIS) has been evaluated with gamma and proton irradiations. The sensors employ an additional n-type implant in pixel which allows full depletion to be achieved by reverse biasing the substrate and have been irradiated alongside reference devices in which the implant has been omitted. The results show no visible degradation of the sensor’s radiation hardness by the additional implant. The main degraded parameters caused by ionizing radiation, from both gamma and proton irradiation, are the image lag and the surface dark current. High dark current in the sense node (SN) is identified and is found to be caused by the shallow trench isolation (STI) around the SN. The dark current increase after proton irradiation has been shown to be dominated by displacement damage-induced bulk defects and the value is consistent with those from other PPD CISs in the literature. The charge collection efficiency is not significantly affected by the bulk traps even after 10-MeV-equivalent proton fluence of up to 1012 p/cm2. However, the ionizing damage from the highest proton fluence reached causes a considerable drop in the charge-to-voltage conversion factor.

The design of suitable electromagnetic bandgap (EBG) cavities has been performed by means of a hybrid numerical/analytical approach implemented via a homemade code with the aim of optimizing a novel accelerating structure for proton linear accelerators (linacs). In particular, a 3-GHz proton linac tank with on-axis coupled EBG cavities closed with full-end cells has been optimized. The proton beam input energy is 27 MeV. The performances of the proton linac EBG accelerating cavities have been compared with the performances of 27-MeV 3-GHz side-coupled proton linac accelerating cavities in terms of typical linac figures of merit. The use of EBG cavities allows to increase the transit time factor (by about 8%). Moreover, the peak surface electric field is strongly reduced (by about 65%), paving the way to the design of very high accelerating gradient microwave proton linacs. Furthermore, the wakefields of the EBG structure have been compared with those of the SCL structure, showing that the EBG structure provides effective damping of the transverse wakefields.

Plastic organic scintillators have been tailored in composition to achieve ultra-fast temporal response, thereby enabling the design and development of fast neutron detection systems with high timing resolution. Eljen Technology's plastic organic scintillators -- EJ-230, EJ-232, and EJ-232Q -- are prospective candidates for use in emerging neutron imaging systems, where fast timing is paramount. To support the neutron response characterization of these materials, the relative proton light yields of EJ-230, EJ-232, and EJ-232Q were measured at the 88-Inch Cyclotron at Lawrence Berkeley National Laboratory. Using a broad-spectrum neutron source and a double time-of-flight technique, the proton light yield relations were obtained over a proton recoil energy range of approximately 300 keV to 4 MeV. The EJ-230, EJ-232, and EJ-232Q scintillators exhibited similar proton light yield relations to each other as well as to other plastic scintillators with the same polymer base material. A comparison of the relative proton light yield of different sized cylindrical EJ-232 and EJ-232Q scintillators also revealed consistent results. This work provides key input data for the realistic computational modeling of neutron detection technologies employing these materials, thereby supporting new capabilities in near-field radionuclide detection for national security applications., Comment: 10 pages, 8 figures

The SC200 cyclotron is the critical component of the proton therapy system, which is being jointly developed by the Institute of Plasma Physics Chinese Academy of Sciences (ASIPP), Hefei, China, and the Joint Institute for Nuclear Research (JINR), Dubna, Russia. It is an isochronous cyclotron to accelerate H+ ions and extract proton beam energies of 75–200 MeV with the maximum beam intensity of $1~\mu \text{A}$ . The maximum magnetic field intensity of the SC200 cyclotron is 4.6 T, and the accuracy of measurements for the average isochronous field is required to be better than $10^{-4}$ . The magnetic gap decreases from 37 mm at cyclotron center to 9 mm at the pole edge. Hence, the difficulty of designing a magnetic field measurement system is increased because of the required magnetic field accuracy and the narrow gap between the poles. A measurement system has been developed to measure the magnetic field distribution in the median plane of the cyclotron by the SENIS Hall probe. After the Hall probe is calibrated by NMR, the measurement accuracy can reach $\pm 4\times 10^{-4}$ under the magnetic field intensity of 5 T. The position of the Hall probe will be monitored by grating and optical encoder, respectively, so as to ensure its correct positions. To verify the performance of the mapping system, it has been tested in the SC200 cyclotron with the designed vacuum degree. The test results showed that the mapping system can run smoothly and meet the measurement requirements of the SC200 cyclotron.

Real-time soft error rate of 65-nm QDRII+ static random access memory was measured on the Tibetan-plateau with an altitude of 4300 m. Error types including single-bit upset (SBU), multiple-cell upset (MCU), and burst errors were observed. The MCU ratio is 23%, with a maximum size of nine cells. Majority of the MCU events are vertically oriented along the well stripes. Monte Carlo simulations were further performed to study the characteristics of the neutron-induced secondary ions. Diverse secondary ion species, from the proton to Re, were found, which were significantly affected by the presence or absence of the tungsten layer. The impact of neutron energy on the linear energy transfer (LET), range, and deposited charge of secondary ions in the sensitive volume was revealed.

This paper introduces a Monte Carlo (MC) approach based on the binary collision approximation to estimate defect types and densities in proton- and heavy-ion-irradiated graphene layers. The types and concentrations of defects that appear in graphene due to irradiation with various ions with energies ranging from ~100 keV to ~100 MeV are identified. This simple method enables defect introduction rates in graphene to be predicted to within better than a factor of two accuracy, relative to experimental measurements in a beam environment. The results demonstrate that all defects generated by ions with high incident energy are formed via head-on collisions and in-plane recoils, which is unique for 2-D materials.

The impact of the total ionizing dose (TID) on a low energy proton single event upset (SEU) in a 65-nm static random access memory (SRAM) is investigated based on a low energy proton accelerator. The body potential of the P-well is elevated due to charge depletion along the sidewall of shallow trench isolation (STI) induced by TID irradiation. The radiation-induced narrow channel effect causes a decrease in the area and the thickness of the SEU sensitive volume (SV), further resulting in the low energy proton SEU cross section being lowered in 65-nm SRAM with an external independent power supply of the memory array. The impact of the TID on the low energy proton SEU cross section is dependent on test patterns written to the memory array during TID irradiation and SEU testing. The low energy proton SEU cross section is the highest when the complement test pattern is written to the SRAM memory array during TID irradiation and SEU testing compared with the same and half complement patterns.

Irradiation of 2N5339 n-p-n bipolar junction transistors (BJTs) with either 250-MeV protons or 10-keV X-rays shifts the input characteristics to lower values of $V_{\mathrm {BE}}$ while degrading the current gain and drive current. The degradation is consistent with increased recombination in the neutral base and emitter–base (E–B) depletion region. A decrease in collector current following proton irradiation suggests that recombination in the neutral base is significant. At a given ionizing dose, degradation is worse for proton irradiation than for X-ray irradiation due to the presence of displacement damage and a higher charge yield. A comparison of degradation produced by the two radiation sources suggests that ~40% of the excess base current resulting from 250-MeV proton irradiation is due to ionization damage. When compared with previous results for other devices, these results suggest that ionization-to-displacement damage ratios in bipolar devices may increase with proton energy in a way that is consistent with trends in charge yield and non-ionizing energy loss (NIEL).

Lateral Pd/n-AlN Schottky barrier diodes (SBDs) were fabricated and subjected to 3-MeV proton irradiation at various fluences. Electrical and material characterization analysis was performed before and after each radiation fluence to quantify the change in device characteristics. It was found that the SBDs performed reliably up to a proton irradiation fluence of $5 \times 10^{13}~\text{cm}^{-2}$ , with little or no change in the key device performance, such as current, turn-on voltage, ideality factor, breakdown voltage, and so on. The electrical characteristics of the SBDs were well-predicted using a standard thermionic emission theory. The performance of the SBDs shows a significant degradation after a high-fluence irritation of $5 \times 10^{15}~\text{cm}^{-2}$ , where the current of the SBDs dropped two orders in magnitude. Material and surface characterizations, including atomic force microscopy and X-ray diffraction, indicated a consistent degradation in the AlN bulk crystal quality and a drastic increase in surface roughness. These results provide valuable information on the radiation properties of AlN electronics and can serve as important references for the future development of high-performance AlN devices for extreme environment applications.

We present heavy ion and proton data on AlGaN high-voltage HEMTs showing single event burnout (SEB), total ionizing dose, and displacement damage responses. These are the first such data for materials of this type. Two different designs of the epitaxial structure were tested for SEB. The default layout design showed burnout voltages that decreased rapidly with increasing LET, falling to about 25% of nominal breakdown voltage for ions with LET of about 34 MeV $\cdot $ cm2/mg for both structures. Samples of the device structure with lower AlN content were tested with varying gate–drain spacing and revealed an improved robustness to heavy ions, resulting in burnout voltages that did not decrease up to at least 33.9 MeV $\cdot $ cm2/mg. Failure analysis showed that there was consistently a point, location random, where gate and drain had been shorted. Oscilloscope traces of terminal voltages and currents during burnout events lend support to the hypothesis that burnout events begin with a heavy ion strike in the vulnerable region between gate and drain. This subsequently initiates a cascade of events resulting in damage that is largely manifested elsewhere in the device. This hypothesis also suggests a path for greatly improving the susceptibility to SEB as development of this technology goes forward. Testing with 2.5-MeV protons showed only minor changes in device characteristics.

$3 \times 10^{13}/\text{cm}^{2}$ 1.2-MeV proton radiation will significantly increase heavy ion single-event transient (SET) cross sections and pulsewidths in 65-nm bulk CMOS technology. After proton radiation, heavy ion SET sensitivity increase in N-hit blocks is more than that in P-hit blocks. The analysis suggests that pMOS transistor on current ( $I_{\mathrm{\scriptscriptstyle ON}}$ ) reduction due to total ionizing dose effects is the main reason for this effect. It is suggested that alternate heavy ion and proton experiments should be performed for radiation hardness assurance of candidate electronic devices used in deep space missions.

Tunneling-dominated charge transport is demonstrated in vertically stacked MoS2/interlayer/MoS2 heterostructures with Al2O3, hexagonal boron nitride (h-BN), and HfO2 dielectrics. All devices are highly resistant to 10-keV X-ray irradiation. Only small transient changes in X-ray-induced photocurrent are observed as a result of trap creation in the thin interlayer dielectric, with rapid passivation. Samples with Al2O3 and h-BN interlayer dielectrics show significant increases in conduction current during proton irradiation, due to displacement-damage-induced defects that lower the effective tunnel-barrier height. Density-functional-theory calculations provide insights into the pertinent defects. Devices with HfO2 interlayer dielectrics show great promise for use in radiation tolerant, ultimately scaled tunnel FETs.

An array of 30 16-Mbit static random access memories manufactured using 65-nm technology has been designed to characterize the proton and neutron beams used for radiation testing at TRIUMF. Measurements include single-event upset cross sections, beam profiles, collimator design, and shielding effects. The sensitivity of this dosimeter allows for quick measurements of these parameters, which are of interest in radiation test measurements.