SnO2-Based Optoelectronic Synapses for Artificial Visual Applications

Abstract

Optoelectronic synapse units have been drawing a strong interest lately due to their obvious crucial role in the evolution of optoelectronic integration-based neuromorphic computing. Here, optoelectronic synaptic devices are fabricated based on stannic oxide (SnO2) film and with the structure of ITO/SnO2/P++-Si. Using optical stimulation, we found that the synaptic sandwich devices are highly responsive and exhibit fast switching behavior. These optically stimulated synaptic devices accurately could imitate synaptic functions, including paired-pulse facilitation (PPF), excitatory postsynaptic current (EPSC), learning experience behavior, and the conversion from short-term memory (STM) to long-term memory (LTM). The EPSC of such devices could be tuned via light, contributing to an LTM-based analog of optical memory and enhancing learning across affective situations. The work presented here advances the research and manufacture of optoelectronic synapses based on the SnO2 thin films for neuromorphic computing applications.