PubCard - An artificial neural network chip based on two-dimensional semiconductor

(Peer-Reviewed) An artificial neural network chip based on two-dimensional semiconductor

Shunli Ma 马顺利 ¹, Tianxiang Wu 吴天祥 ¹, Xinyu Chen 陈新宇 ¹, Yin Wang 王印 ¹, Hongwei Tang 唐宏伟 ¹, Yuting Yao ¹, Yan Wang ¹, Ziyang Zhu ², Jianan Deng 邓嘉男 ², Jing Wan 万景 ², Ye Lu 陆叶 ², Zhengzong Sun 孙正宗 ¹, Zihan Xu 许子寒 ³, Antoine Riaud ¹, Chenjian Wu 吴晨健 ⁴, David Wei Zhang 张卫 ¹, Yang Chai 柴扬 ⁵, Peng Zhou 周鹏 ¹, Junyan Ren 任俊彦 ¹, Wenzhong Bao 包文中 ¹

¹ State Key Laboratory of ASIC and System, School of Microelectronics, Fudan University, Shanghai 200433, China
中国上海复旦大学微电子学院专用集成电路与系统国家重点实验室
² State Key Laboratory of ASIC and System, School of Information Science and Technology, Fudan University, Shanghai 200433, China
中国上海复旦大学信息科学与工程学院专用集成电路与系统国家重点实验室
³ Shenzhen Sixcarbon Technology, Shenzhen 518106, China
中国深圳深圳六碳科技有限公司
⁴ School of Electronic and Information Engineering, Soochow University, Suzhou 215006, China
中国苏州苏州大学电子信息学院
⁵ Department of Applied Physics, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong, China
中国香港九龙红磡香港理工大学应用物理系

Science Bulletin, 2021-10-05

https://doi.org/10.1016/j.scib.2021.10.005

https://www.sciencedirect.com/science/article/abs/pii/S2095927321006459

Abstract

Recently, research on two-dimensional (2D) semiconductors has begun to translate from the fundamental investigation into rudimentary functional circuits. In this work, we unveil the first functional MoS₂ artificial neural network (ANN) chip, including multiply-and-accumulate (MAC), memory and activation function circuits.

Such MoS₂ ANN chip is realized through fabricating 818 field-effect transistors (FETs) on a wafer-scale and high-homogeneity MoS₂ film, with a gate-last process to realize top gate structured FETs. A 62-level simulation program with integrated circuit emphasis (SPICE) model is utilized to design and optimize our analog ANN circuit.

To demonstrate a practical application, a tactile digit sensing recognition was demonstrated based on our ANN circuits. After training, the digit recognition rate exceeds 97%. Our work not only demonstrates the protentional of 2D semiconductors in wafer-scale integrated circuits, but also paves the way for its future application in AI computation.