面向类脑计算的低电压忆阻器研究进展

Recent progress of low-voltage memristor for neuromorphic computing

忆阻器是非易失性存储器和神经形态计算的优秀候选者.电压调制作为其关键性能策略,是获得纳瓦超低功耗、飞焦超低能耗工作的基础,有助于打破功耗墙、突破后摩尔时代算力瓶颈.然而基于高密度集成忆阻器阵列的类脑计算架构还需重点考虑开/关比、高速响应、保留时间和耐久性等器件稳定性参数.因此如何在低电场下实现离子/电子的高效、稳定驱动,构筑电压低于1 V的低电压、高性能忆阻器成为了当前实现类脑计算能效系统的关键问题.本文综述了近年来面向类脑计算的低电压忆阻器的研究进展.首先,探讨了低电压忆阻器的机制,包括电化学金属化机制和价态变化机制.在此基础上,系统总结了各材料体系在低电压忆阻器中的优势,涵盖了过渡金属氧化物、二维材料和有机材料等.进一步围绕材料工程、掺杂工程、界面工程提出了相应的低电压忆阻器实现策略,最后,展望了基于低电压忆阻器的类脑功能模拟及神经形态计算应用,并对现存问题和未来研究方向进行了讨论.

Memristors stand out as the most promising candidates for non-volatile memory and neuromorphic computing due to their unique properties.A crucial strategy for optimizing memristor performance lies in voltage modulation,which is essential for achieving ultra-low power consumption in the nanowatt range and ultra-low energy operation below the femtojoule level.This capability is pivotal in overcoming the power consumption barrier and addressing the computational bottlenecks anticipated in the post-Moore era.However,for brain-inspired computing architectures utilizing high-density integrated memristor arrays,key device stability parameters must be considered,including the on/off ratio,high-speed response,retention time,and durability.Achieving efficient and stable ion/electron transport under low electric fields to develop low-voltage,high-performance memristors operating below 1 V is critical for advancing energy-efficient neuromorphic computing systems.This review provides a comprehensive overview of recent advancements in low-voltage memristors for neuromorphic computing.Firstly,it elucidates the mechanisms that control the operation of lowvoltage memristor,such as electrochemical metallization and anion migration.These mechanisms play a pivotal role in determining the overall performance and reliability of memristors under low-voltage conditions.Secondly,the review then systematically examines the advantages of various material systems employed in lowvoltage memristors,including transition metal oxides,two-dimensional materials,and organic materials.Each material system has distinct benefits,such as low ion activation energy,and appropriate defect density,which are critical for optimizing memristor performance at low operating voltages.Thirdly,the review consolidates the strategies for implementing low-voltage memristors through advanced materials engineering,doping engineering,and interface engineering.Moreover,the potential applications of low-voltage memristors in neuromorphic function simulation and neuromorphic computing are discussed.Finally,the current problems of low-voltage memristors are discussed,especially the stability issues and limited application scenarios.Future research directions are proposed,focusing on exploring new material systems and physical mechanisms that could be integrated into device design to achieve higher-performance low-voltage memristors.