神经形态工程学
材料科学
突触
薄膜晶体管
光电子学
兴奋剂
晶体管
薄膜
纳米技术
电压
计算机科学
人工神经网络
电气工程
神经科学
人工智能
工程类
心理学
图层(电子)
作者
Yuling Peng,Wei Dou,P. Chen,XU Xiao-dong,G. C. Jiang,Pufan Deng,N. Zhang,Yanling Yin,Yuehua Peng,Dongsheng Tang
摘要
Inspired by the human brain's capacity as a powerful biological computer capable of simultaneously processing a vast array of cognitive tasks, many emerging artificial synapse devices have been developed in recent years. Electric-double-layer (EDL) transistors based on interfacial ion-modulation have attracted widespread attention for simulating synaptic plasticity and neural functions. Here, low-voltage EDL p-type thin-film transistors (TFTs) are fabricated on glass substrates, with Ni-doped cuprous iodide (Ni0.06Cu0.94I) as the channel and chitosan as the dielectric. The electrical performance of the Ni0.06Cu0.94I TFTs is investigated: current on/off ratio of 6.4 × 104, subthreshold swing of 33 mV/dec, threshold voltage of 1.38 V, operating voltage of 2 V, and saturation field-effect mobility of 15.75 cm2 V−1 s−1. A dual in-plane gate OR logic operation is demonstrated. Importantly, by applying single voltage pulses, dual voltage pulses, and multiple voltage pulses to the gate, the Ni0.06Cu0.94I transistors exhibited typical synaptic characteristics, including short-term potentiation, short-term depression, long-term potentiation, long-term depression, paired-pulse facilitation, and spiking-rate-dependent plasticity. Furthermore, the synaptic transistor can also simulate the learning–forgetting–relearning process of the human brain. These remarkable behaviors of voltage-stimulated synaptic transistors have potential for neuromorphic applications in future artificial systems.
科研通智能强力驱动
Strongly Powered by AbleSci AI