Bioartificial Synapses for Neuromorphic Computing

Abstract

As a promising solution for overcoming the bottlenecks of traditional von Neumann computers, the hardware implementation of neuromorphic computing has attracted increasing interest. High-performance artificial synapses are the basic units of brain-like chips and are important for achieving efficient neuromorphic calculations. This paper reports the fabrication of Al/chitosan (CS)/graphene oxide (GO)/indium tin oxide (ITO) artificial synapses. The electronic insulation and proton conduction properties of CS enable it to conduct electricity alone and accelerate the movement of oxygen vacancies in GO. In particular, an experimental device successfully simulated the Pavlov associative memory experiment and was made to exhibit both short-term and long-term memory capabilities by modifying the external stimuli. This device provides a possible avenue to realize neuromorphic engineering on the basis of biomemristors.