材料科学
神经形态工程学
铁电性
光电子学
极化(电化学)
薄膜
钙钛矿(结构)
外延
突触可塑性
纳米技术
图层(电子)
计算机科学
化学
人工神经网络
生物化学
受体
物理化学
机器学习
电介质
结晶学
作者
Xiaoxiao Wu,Zeyang Li,Yongguang Xiao,Caihong Jia,Weifeng Zhang
摘要
SrIrO3 (SIO) with a perovskite structure exhibits broad application prospects in electronic synapses due to its high carrier mobility, scalability, and compatibility with silicon technology. On the other hand, ferroelectric polarization can efficiently and directionally regulate the movement of carriers, which provides the possibility to improve the synaptic properties of such materials. In this study, electronic synapses based on the SIO epitaxial film with introducing a ferroelectric PbZr0.2Ti0.8O3 (PZT) layer, multiple important synaptic plasticity functions including paired-pulse depression, spike-rate-dependent plasticity, spike-time-dependent plasticity, and abnormal Bienenstock–Cooper–Munro learning rule have been simulated by electrical stimulation. Furthermore, the PZT/SIO/Nb:SrTiO3 (NSTO) device exhibits larger synaptic weight and dynamic range of conductance and requires less training pulses compared to the SIO/NSTO device, which may be attributed to the enhanced modulation of the variation in the SIO/NSTO interface barrier by the polarization field. This research makes a great contribution to the efficient and fast information transmission of neuromorphic computing.
科研通智能强力驱动
Strongly Powered by AbleSci AI