Hongjia Song

Memristive neural network hardware shows great potential for Space applications thanks to its high efficiency, low power, high integration, and radiation tolerance. Yet no memristor model captures the influence of irradiation effects on its conductance update behavior, impeding the radiation-hardening design of memristive neural network hardware. This article presents a proton irradiation damage compact model of memristive synapses. Simulations of $I$ – $V$ curves and conductance-update linearity agree with experiment within an rms error <0.2. Using this model, we design a radiation-hardened synaptic circuit that reduces long-term potentiation/depression nonlinearity by 42% pre- and 33% postirradiation. A convolutional neural network (CNN) built with these radiation-hardened synapses retains 95%/92% recognition accuracy pre-/postirradiation, surpassing single-memristor networks (93%/88%). The model and circuit form a foundation for Space-grade memristive neural network hardware.

We synthesized a novel chain extender (HDF) featuring disulfide and imine bonds and incorporated it into PUE. The PUE cured with HDF exhibits outstanding damping and mechanical properties.

Ferroelectric artificial synaptic devices meet the intelligence of the new generation of spacecraft with their characteristics of high efficiency and low power consumption. However, there are few reports on its radiation effect. Here, a multidomain ferroelectric tunnel junction (FTJ) synaptic device model with a TiN/HZO/Si (MFS) structure was built via technology computer‐aided design (TCAD) software. The impact of total ionizing dose (TID) on the device's synaptic performance was simulated. The results indicated that the conductance ratio of the device decreased from 18.8 to 8.8 when the irradiation dose accumulated to 1 Mrad(Si). Meanwhile, the nonlinear values of conductance update increased from 0.66 to 0.73 and −4.88 to −5.73 for long‐term potentiation (LTP) and long‐term depression (LTD), respectively. The reduction of the conductance ratio and the enhancement of the nonlinear values were attributed to the synergistic effect of irradiation‐induced positive charge capture and interface defects. Using the +NeuroSim platform, the effect of irradiation on the image recognition accuracy of MFS‐FTJ‐based neural networks was predicted. These changes cause the image recognition accuracy of the corresponding artificial neural network to drop from 88.6% to 86%. These results indicate that the device makes them promising for application in brain‐like chips in space radiation environment.

Studying the atmospheric neutron irradiation effects on memristive synapses is fundamental to extending the application of memristive neuromorphic chips in aviation. However, such studies remain rarely reported. Here, atmospheric neutron irradiation effects on the conductance update linearity of Mg:HfOx (9 at. %) memristive synapses—key to chip computation—were studied, and degradation was analyzed from the perspective of conductive filaments formation/rupture. During long-term depression, the memristive synapse fails to fully return to its original high-resistance state under optimal pulse programming voltage, leading to a reduction in the conductance update linearity. This decrease becomes increasingly pronounced with the increase in atmospheric neutron irradiation fluence. Analysis by x-ray photoelectron spectroscopy and electrochemical impedance spectroscopy reveals that the atmospheric neutron irradiation generates new oxygen vacancies, forming oxygen-vacancy-enriched regions that inhibit conductive filaments rupture, which is the main reason for the decrease in the linearity of conductance update. The research can provide some supports for the development of memristive neuromorphic chips for aviation applications.

Through the three-stage oxygen vacancy distribution in TaO _x /TiO _x , the conductance update characteristics of STDP were optimized and the SNN recognition was strengthened.

Threshold switching (TS) memristors have emerged as promising candidates for artificial neurons in brain-inspired chips due to advantages of easy integration, high density, and low power consumption. However, in aerospace applications, their performances are susceptible to degradation caused by radiation, posing a threat to the reliability of brain-inspired chips. Currently, the radiation degradation laws and mechanisms of TS memristors characteristics are not yet fully understood. In this paper, we firstly investigate the radiation damage law and mechanisms of Ta₂O₅: Ag-based TS memristors based on neutron source. Then the impact of neutron irradiation on the spike firing frequency of Leaky Integrate-and-Fire (LIF) neurons based on these devices are simulated using PSpice. The results indicate that as the neutron irradiation fluence increases, holding voltage, high resistance state, and on/off ratio of the memristors remain basically unchanged while the threshold voltage increases and the switching speed decreases. When the neutron fluence reaches 4.5×10¹² n/cm², the threshold voltage and on/off time of the memristor increase to 1.10~1.58 V and 39/42 ns respectively, which results in more stringent conditions for the LIF neuron to generate spike frequencies. X-ray Photoelectron Spectroscopy analysis reveals that the main reason for the degradation of device performance is the suppression of silver ion migration by irradiation-induced silver interstitial atoms and oxygen vacancy defects. The research findings offer support for the development of radiation-hardened high-speed memristor-based neurons and brain-inspired chips.

This study investigates the $\gamma $ -ray irradiation dose-dependent radiation damage effects in Ti/Au- $\beta $ -Ga₂O₃-Ti/Au metal-semiconductor-metal (MSM) solar-blind photodetectors. By analyzing the photoresponse under varying cumulative irradiation doses, we systematically examine the influence of total ionizing dose (TID) on the photodetector, aiming to evaluate the degradation behavior and underlying mechanisms. At low irradiation doses, the majority carrier removal effect in Ga₂O₃, induced by Compton scattering-generated defects that trap or compensate carriers, leads to a reduction in dark current. In contrast, high-dose irradiation promotes the formation of deep-level complex defects through the combination of ${V}_{\text {O}}$ and ${V}_{\text {Ga}}$ , triggering nonradiative recombination and leading to a sharp increase in dark current. Numerical simulations based on technology computer-aided design (TCAD) were employed to model the irradiation-induced defects, and the results show good agreement with experimental observations, further confirming the defect evolution and carrier dynamics under different TID levels. Furthermore, X-ray diffraction (XRD), photoluminescence (PL) spectroscopy, and X-ray photoelectron spectroscopy (XPS) were employed to analyze the microscopic structural changes and degradation mechanisms induced by TID irradiation. This work provides a theoretical foundation for the application of $\beta $ -Ga₂O₃ solar-blind photodetectors in extreme radiation environments.

This study investigates the displacement damage in β-Ga₂O₃ Schottky barrier diodes (SBDs) under different neutron fluences using a 14-MeV neutron irradiation source. The electrical performance of the devices before and after irradiation was evaluated through direct current–voltage (I–V) and capacitance–voltage (C–V) measurements. The results show that as the irradiation fluence increases, the forward current density of the device decreases, while the reverse current density and leakage current density increase, indicating that the degradation of the device intensifies with higher irradiation fluence. By employing frequency-dependent conductance techniques and deep-level transient spectroscopy (DLTS), changes in the bulk defect concentrations before and after irradiation were analyzed. It was found that the primary cause of device performance degradation is the increase in bulk defect concentrations related to oxygen vacancies induced by neutron irradiation. These findings were further validated through 2-D numerical simulations using technology computer-aided design (TCAD) tools, providing significant theoretical insights and experimental data to enhance the reliability and optimize the design of such devices.

The influence of 300 MeV high-energy proton irradiation on β-Ga₂O₃ solar-blind ultraviolet photodetectors was investigated for space optoelectronics applications. The photoelectric performance of the photodetectors was measured before and after proton irradiation with a fluence of up to 5 × 10¹² p/cm². Both photocurrent and dark current increased significantly, but the photo-to-dark current ratio (PDCR) remained above 10³ due to a larger rise in photocurrent. The response time increased after proton irradiation, likely due to radiation-induced defects that affected carrier transport. The decrease in the power law exponent of the current–voltage characteristics suggested an increase in internal defects within the β-Ga₂O₃ film. Simulation using stopping and range of ions in matter revealed that protons passed through the β-Ga₂O₃ film without being trapped. An X-ray diffraction analysis showed no detectable structural changes in the β-Ga₂O₃ film. An X-ray photoemission spectroscopy indicated an increased concentration of oxygen vacancies. These oxygen vacancies were likely responsible for the observed increase in both photocurrent and dark current, as they could act as trap states, enhancing carrier generation and reducing the overall barrier height. The excellent radiation hardness of β-Ga₂O₃ solar-blind ultraviolet photodetectors demonstrated the potential for space optoelectronics applications.