Answers are grounded in this paper and cite their sources.
Observational analysis demonstrated reduced conductance from irradiation dose in artificial synapses, suggesting implications for neural networks in space radiation environments.
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.
No takes yet. Share an insight, caveat, or question.
Xie et al. (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: