Fully transparent, flexible and waterproof synapses with pattern recognition in organic environments

• Published in: Nanoscale horizons Vol. 4; no. 6; pp. 1293 - 131
• Main Authors: Wang, Tian-Yu, Meng, Jia-Lin, He, Zhen-Yu, Chen, Lin, Zhu, Hao, Sun, Qing-Qing, Ding, Shi-Jin, Zhou, Peng, Zhang, David Wei
• Format: Journal Article
• Language: English
• Published: Cambridge Royal Society of Chemistry 01.11.2019
• Subjects: Artificial intelligence; Bend radius; Biocompatibility; Computer simulation; Learning; Neural networks; Pattern recognition; Synapses
• DOI: 10.1039/c9nh00341j
• ISSN: 2055-6756; 2055-6764

Artificial intelligence applications require bio-inspired neuromorphic systems that consist of electronic synapses (e-synapses) able to perform learning and memory functions. However, all transparent and flexible organic e-synapses have the disadvantage of being easily dissolvable in water or organic solutions. In the present work, a stable waterproof artificial synapse based on a fully transparent electronic device, suitable for wearable applications in organic environments is for the first time demonstrated. Essential synaptic behaviors, including paired-pulse facilitation (PPF), long-term potentiation/depression (LTP/LTD), and learning-forgetting-relearning, were successfully emulated. The artificial synaptic device could achieve an optical transmittance of ∼87.5% in the visible light range, which demonstrated reliable long-term potentiation/depression under bent states with a bending radius of 5 mm. After being immersed in water and 5 types of common organic solvents for over 12 hours, the e-synapse could function with 6000 spikes without noticeable degradation in the organic environment. The neural network was constructed from e-synapses with controllable weights update and a device-to-system level simulation framework was developed with a recognition rate of 92.4%, which demonstrated the feasibility of highly transparent, biocompatible, flexible, and waterproof e-synapses used in artificial intelligence systems. Artificial intelligence applications require bio-inspired neuromorphic systems that consist of electronic synapses (e-synapses) able to perform learning and memory functions.