Interface Element Accumulation‐Induced Single Ferroelectric Domain for High‐Performance Neuromorphic Synapse

作者：Xiaoqi Li, Jiaqi Liu, Fan Xu, Sajjad Ali, Han‐Chieh Wu, Biaohong Huang, Haoyue Deng, Yizhuo Li, Yuxuan Jiang, Zhen Fan, Yunlong Tang, Yujia Wang, M. Bououdina, Teng Yang, Weijin Hu, Zhidong Zhang · 发表于：Advanced Functional Materials · 年份：2025 · DOI：10.1002/adfm.202423225 · 被引用次数：16 · 研究领域：Advanced Memory and Neural Computing、Ferroelectric and Negative Capacitance Devices、Neuroscience and Neural Engineering

Abstract Ferroelectric (FE) synapses are promising for neuromorphic computing toward enhanced artificial intelligence systems. Nonetheless, there is a significant gap in understanding how to effectively tailor self‐polarization and its implications on synaptic device performance. Here, an approach using interfacial element accumulation is reported to tailor the self‐polarization states of BaTiO 3 (BTO)/La 0.67 Sr 0.33 MnO 3 (LSMO) FE heterostructure into a single domain state. This single domain configuration results are demonstrated in a gradient distribution of oxygen vacancies across the film thickness, yielding an extraordinary on/off ratio of 10 7 in Pt/BTO/LSMO FE diodes. This giant resistive switching enables the long‐term potentiation and long‐term depression synaptic functions of excellent linearity and symmetry (with a nonsymmetry factor as low as 0.1), leading to a supervised learning ability of the associated artificial neural network with a high pattern recognition accuracy of 95%. This work provides a simple design principle for FE single domain, which is substantial in enhancing the performance of FE synapses for neuromorphic computing.