Silicon Nanomembrane-Based Synaptic Photodetectors Activated by Phosphorescent Stacks.

Neuromorphic computing that mimics the human brain to realize efficient parallel information processing is considered an important path to break the von Neumann bottleneck. Optoelectronic synaptic devices are of particular interest because of their critical role in the development of neuromorphic computing. This work presents a synaptic photodetector based on the hybrid structure from silicon nanomembranes and a phosphorescent film. The bright and lasting green afterglow of phosphorescent film can be absorbed by the underneath silicon nanomembranes, thus leading to persistent photoconduction. Consequently, synaptic functionalities including excitatory postsynaptic current (EPSC) and paired-pulse facilitation (PPF) are realized via the optical stimulations. Moreover, synaptic short-term and long-term plasticity can be selectively defined within the devices, which are further utilized to simulate age-related cognitive states and memory processes. These results add to the portfolio of optoelectronic synapse options in neuromorphic computing, artificial intelligence, and visual perception systems.