Self‐Powered Photonic Synaptic Device Based on ZnO/P3HT Microwires for Energy‐Efficient Neuromorphic Computing

ABSTRACT Artificial synapses are core components of neuromorphic systems emulating neural functions, yet many existing devices suffer from high power consumption and limited mechanical flexibility. Here, we report a self‐powered flexible photonic synaptic device based on a ZnO/poly(3‐hexylthiophene) (P3HT) composite microwire. The type‐II ZnO/P3HT heterojunction generates an intrinsic electric field, enabling efficient photo‐generated charge separation at 0 V bias and supporting the emulation of key synaptic plasticity, including short‐term memory, long‐term memory, and spike‐timing‐dependent plasticity. It achieves recognition accuracies of 92.0%, 88.0%, and 81.4% at noise levels of 0%, 30%, and 60%, respectively, after 3000 training epochs at 0 V bias. The device maintains stable performance after 2000 bending cycles. With the integration of self‐powered operation, mechanical flexibility, and robust noise tolerance, this ZnO/P3HT microwire photonic synaptic device holds great promise for low‐power neuromorphic systems, particularly in wearable electronics, flexible robotics, and vision‐based intelligent devices.