神经形态工程学
材料科学
晶体管
计算机科学
光电子学
兴奋性突触后电位
物理
电气工程
神经科学
人工神经网络
电压
人工智能
工程类
生物
抑制性突触后电位
作者
Yarong Wang,Yafen Yang,Zhenyu He,Hao Zhu,Lin Chen,Qingqing Sun,David Wei Zhang
标识
DOI:10.1109/led.2020.3008728
摘要
The dynamically active synaptic elements are the fundamental building blocks in neuromorphic systems towards artificial intelligence with computing and sensing capabilities. Here, a two-dimensional (2D) MoS2synaptic transistor is fabricated by using poly(ethylene oxide) (PEO) and lithium perchlorate (LiClO4) as a laterally coupled ion-conducting electrolyte. Due to the strong electric double layer (EDL) effect, a low operating voltage of 1 V and a high current on/off ratio of 105have been obtained. In addition, short-term and long-term plasticity of typical synaptic behaviors have been successfully simulated, such as excitatory postsynaptic current, paired pulse facilitation long-term potentiation, long-term depression, and dynamic filtering. These results can provide new opportunities and strategies in building hybrid and low-dimensional neuromorphic systems for future artificial intelligence applications.
科研通智能强力驱动
Strongly Powered by AbleSci AI