In recent years, metal halide perovskite memristors have emerged as a promising technology for applications in electronic neurons and synapses, offering hardware-based brain-like computing capabilities. This potential is attributed to their simple metal-insulator-metal structure and ionic conduction mechanisms, which closely mimic biological systems. However, the widespread adoption of perovskite memristors has been hindered by the reliance on expensive gold (Au) electrodes and unstable silver (Ag) electrodes. Identifying a cost-effective, stable, and high-performance electrode material has become a pressing challenge in this field.
Recently, Associate Professor Li Deli and Associate Professor Wang Yue from Fujian Normal University proposed an innovative solution: a perovskite memristor featuring a silver/bismuth (Ag/Bi) bilayer electrode. This novel design demonstrates stable resistive switching, a high switching ratio of 10², durability over 800 cycles, and a retention time of up to 10⁴ seconds, along with excellent storage stability. Their research also introduced a new mechanism explaining how the built-in electric field influences ion migration, which contributes to the lower switching voltage and longer retention time of Ag/Bi bilayer electrode perovskite memristors.
In addition to these advancements, the memristor successfully replicates key synaptic behaviors, including potentiation, depression, long-term memory, short-term memory, and adherence to Hebbian learning rules. Furthermore, the device was applied to a Spiking Neural Network (SNN) for digit recognition tasks, achieving an impressive accuracy of 86.6% on the MNIST dataset. These findings highlight the potential of Ag/Bi bilayer electrodes to significantly enhance the performance and expand the applications of perovskite memristors, paving the way for their integration into neuromorphic computing systems.
This groundbreaking research was published in the prestigious journal Advanced Functional Materials, marking a significant step forward in the development of perovskite memristors for next-generation computing technologies.
Stable halide perovskite memristor utilizing innovative silver/bismuth electrode as an alternative to Aurum (Au) electrode
The study was led by Feng Jiuchao, a master's student from the Haixia College of Flexible Electronics at Fujian Normal University, who is the first author of the paper. Associate Professor Li Deli, also from Fujian Normal University, served as the first corresponding author, while Professor Huang Wei from Northwestern Polytechnical University and Associate Professor Wang Yue from Fujian Normal University acted as co-corresponding authors.
This research was supported by the National Natural Science Foundation of China, the Science and Technology Planning Project of Fujian Province, and the Education Research Project for Young and Middle-Aged Teachers of Fujian Province. The findings represent a major step forward in advancing the performance and accessibility of halide perovskite memristors for next-generation computing technologies.
Translated by Yan Zihan
Reviewed by Lin Bin