Dual‐Functional Optoelectronic Devices Composed of Cs ₃ Cu ₂ I ₅ /Graphene/Ge Multi‐Heterostructures Enabling Deep‐Ultraviolet Synaptic Behavior and Near‐Infrared Photodetection

Multi-functional optoelectronic devices that integrate the capacities of neuromorphic processing and photodetection are of pivotal importance for advancing optoelectronic techniques. Here, we present a dual-functional optoelectronic device based on a multi-heterostructure of Cs₃Cu₂I₅/graphene/Ge, which realizes high-performance deep-ultraviolet (DUV) synaptic behavior and self-driven near-infrared (NIR) photodetection simultaneously. On one hand, the device leverages the photogating effect in the Cs₃Cu₂I₅/graphene heterostructure to produce persistent photo-conductivity, emulating synaptic behavior upon DUV light stimulation. Electron trapping by introducing B₄PyMPM as capture sites significantly promotes spatial divorcement of photocarriers, guaranteeing remarkably improved synaptic characteristics. The utilizations in DUV fingerprint recognition with an accuracy of 97.8%, and optoelectronic reservoir computing with recognition accuracy reaching 100% for three 4-bit numbers are demonstrated. On the other hand, the device utilizes the photovoltaic effect in the graphene/Ge heterostructure to generate photoresponse at zero working bias, enabling self-driven photodetection upon NIR light illumination. The wide-bandgap Cs₃Cu₂I₅ functions as an effective NIR anti-reflection layer to induce a strong light trapping effect, giving rise to boosted photoresponse properties. The excellent photoresponse supports the device in performing single-pixel NIR optical imaging. It is believed that this work is inspiring to the design of high-performance multi-functional optoelectronic devices toward integrated and intelligence applications.