Graphene enhanced charge transfer in ITO optoelectronic synapses for artificial vision systems

Abstract

The persistent photoconductivity effect is one of the main working mechanisms of indium tin oxide (ITO) based artificial photoelectric synapses. But this effect results in a longer relaxation time for ITO based devices, making it difficult to simulate the short-term plasticity of biological synapses. In this study, we proposed Au/ITO-graphene/Au structure optoelectronic synapses, in which the conductivity was controllable by introduced graphene to overcome the relaxation effect. Under purple light stimulation, the conductivity of the device decreases, while gate electrical stimulation increases the conductivity of the device and shortens the relaxation time, achieving simulation of synaptic short-term plasticity. By utilizing the electron exchange between the two under photoelectric stimulation to bi-directionally regulate the conductivity of graphene, ITO graphene synapses can simulate various biological synaptic plasticity characteristics. A 2 × 2-pixel imaging chip was constructed using ITO/graphene synapses to simulate artificial vision systems, verifying the transition of photoelectric synapses from short-term memory to long-term memory and their simple image memory function. The simple structure and high light response amplitude of artificial optoelectronic synapses are of great significance for the development of artificial vision systems.