Optical synaptic devices with ultra-low power consumption for neuromorphic computing

Chenguang Zhu; Huawei Liu; Wenqiang Wang; Li Xiang; Jie Jiang; Qin Shuai; Xin Yang; Tian Zhang; Biyuan Zheng; Hui Wang; Dong Li; Anlian Pan

doi:10.1038/s41377-022-01031-z

Optical synaptic devices with ultra-low power consumption for neuromorphic computing

Light Sci Appl. 2022 Nov 29;11(1):337. doi: 10.1038/s41377-022-01031-z.

Authors

Chenguang Zhu^#^{1

2}, Huawei Liu^#^{1

2}, Wenqiang Wang^#^{1

2}, Li Xiang^{3

4}, Jie Jiang⁵, Qin Shuai^{1

2}, Xin Yang^{1

2}, Tian Zhang^{1

2}, Biyuan Zheng^{1

2}, Hui Wang^{1

2}, Dong Li^{6

7}, Anlian Pan^{8

9}

Affiliations

¹ Key Laboratory for Micro-Nano Physics and Technology of Hunan Province, State Key Laboratory of Chemo/Biosensing and Chemometrics, College of Materials Science and Engineering, Hunan University, 410082, Changsha, China.
² Hunan Institute of Optoelectronic Integration, Hunan University, 410082, Changsha, China.
³ Key Laboratory for Micro-Nano Physics and Technology of Hunan Province, State Key Laboratory of Chemo/Biosensing and Chemometrics, College of Materials Science and Engineering, Hunan University, 410082, Changsha, China. xiangli93@hnu.edu.cn.
⁴ Hunan Institute of Optoelectronic Integration, Hunan University, 410082, Changsha, China. xiangli93@hnu.edu.cn.
⁵ School of Physics and Electronics, Central South University, 410083, Changsha, China.
⁶ Key Laboratory for Micro-Nano Physics and Technology of Hunan Province, State Key Laboratory of Chemo/Biosensing and Chemometrics, College of Materials Science and Engineering, Hunan University, 410082, Changsha, China. liidong@hnu.edu.cn.
⁷ Hunan Institute of Optoelectronic Integration, Hunan University, 410082, Changsha, China. liidong@hnu.edu.cn.
⁸ Key Laboratory for Micro-Nano Physics and Technology of Hunan Province, State Key Laboratory of Chemo/Biosensing and Chemometrics, College of Materials Science and Engineering, Hunan University, 410082, Changsha, China. anlian.pan@hnu.edu.cn.
⁹ Hunan Institute of Optoelectronic Integration, Hunan University, 410082, Changsha, China. anlian.pan@hnu.edu.cn.

^# Contributed equally.

Abstract

Brain-inspired neuromorphic computing, featured by parallel computing, is considered as one of the most energy-efficient and time-saving architectures for massive data computing. However, photonic synapse, one of the key components, is still suffering high power consumption, potentially limiting its applications in artificial neural system. In this study, we present a BP/CdS heterostructure-based artificial photonic synapse with ultra-low power consumption. The device shows remarkable negative light response with maximum responsivity up to 4.1 × 10⁸ A W^-1 at V_D = 0.5 V and light power intensity of 0.16 μW cm^-2 (1.78 × 10⁸ A W^-1 on average), which further enables artificial synaptic applications with average power consumption as low as 4.78 fJ for each training process, representing the lowest among the reported results. Finally, a fully-connected optoelectronic neural network (FONN) is simulated with maximum image recognition accuracy up to 94.1%. This study provides new concept towards the designing of energy-efficient artificial photonic synapse and shows great potential in high-performance neuromorphic vision systems.

Abstract

Grants and funding