CMOS-compatible electrochemical synaptic transistor arrays for deep learning accelerators
CMOS-compatible electrochemical synaptic transistor arrays for deep learning accelerators
复制标题
DOI：
10.1038/s41928-023-00939-7
复制
发表时间：
2023-03-27
影响因子：
34.3
通讯作者：
Cao, Qing
中科院分区：
文献类型：
--
作者：
Cui, Jinsong;An, Fufei;Cao, Qing
基于忆阻交叉阵列的内存计算架构可以在深度学习应用中提供比传统硬件更高的计算效率。然而，核心存储器设备必须能够执行高速且对称的模拟编程，且变化性很小。它们还应该与硅技术兼容，并可扩展到纳米尺寸的足迹。在这里，我们报告了一种电化学突触晶体管，其工作原理是通过二氧化锆质子电解质在氢化氧化钨通道和栅极之间移动质子。这些器件通过栅极电压脉冲控制和小的周期到周期变化提供多状态和对称的沟道电导编程。它们可以在接近兆赫范围的频率下编程，并表现出超过1亿个读写周期的重复性。它们还与互补金属氧化物半导体技术兼容，并且可以扩展到150 x 150 nm的横向尺寸。通过与硅晶体管的单片集成，我们证明了伪交叉阵列可以用于面积和能量高效的深度学习加速器应用。基于氧化物的固态质子电化学晶体管具有对称操作并且与CMOS技术兼容，可以用于创建深度学习应用的交叉阵列。
In-memory computing architectures based on memristive crossbar arrays could offer higher computing efficiency than traditional hardware in deep learning applications. However, the core memory devices must be capable of performing high-speed and symmetric analogue programming with small variability. They should also be compatible with silicon technology and scalable to nanometre-sized footprints. Here we report an electrochemical synaptic transistor that operates by shuffling protons between a hydrogenated tungsten oxide channel and gate through a zirconium dioxide protonic electrolyte. These devices offer multistate and symmetric programming of channel conductance via gate-voltage pulse control and small cycle-to-cycle variation. They can be programmed at frequencies approaching the megahertz range and exhibit endurances of over 100 million read-write cycles. They are also compatible with complementary metal-oxide-semiconductor technology and can be scaled to lateral dimensions of 150 x 150 nm(2). Through monolithic integration with silicon transistors, we show that pseudo-crossbar arrays can be created for area- and energy-efficient deep learning accelerator applications.Oxide-based solid-state protonic electrochemical transistors that have symmetric operation and are compatible with CMOS technology can be used to create crossbar arrays for deep learning applications.