Moiré Superlattices‐Based Artificial Optoelectronic Neuro‐Synapses for Long‐Term Information Retention

ABSTRACT The development of synaptic devices capable of integrating optical sensing and long‐term information storage remains a critical challenge in bio‐inspired neuromorphic computing. To address the rapid decay of information in photoelectric synapses following cessation of illumination, we fabricated an artificial photoelectric synaptic device based on a small‐angle (∼1.9°) twisted bilayer MoS 2 moiré superlattice structure and ReS 2 . By exploiting the cooperative effects of moiré potential wells and heterointerface charge trapping, our device achieves prolonged optical information retention (∼4.8 × 10 3 s) and demonstrates a “1 + 1 >> 2” synergistic enhancement effect. Based on techniques including Kelvin probe force microscopy (KPFM), scanning tunneling microscopy (STM), scanning tunneling spectroscopy (STS), and transient absorption, we elucidate the roles of moiré potentials and interfacial charge trapping in governing synaptic performance. The device successfully emulates essential neurobiological behaviors, including excitatory postsynaptic current, long/short‐term memory transition, paired‐pulse facilitation, Pavlovian conditioning, and handwritten digit recognition in neural network simulation. This work not only advances the understanding of moiré‐regulated charge storage mechanisms but also establishes a promising platform for high‐fidelity, high‐performance neuromorphic hardware.