Carbon Nanotube‐Based Optoelectronic Synaptic Thin‐Film Transistors at 940 Nm for Covert Neuromorphic Visual Surveillance and Image Recognition

ABSTRACT In this work, we report single‐walled carbon nanotubes (SWCNTs) synaptic transistors with high photoresponsivity at 940 nm, achieved by functionalizing the channel with the narrow‐bandgap photosensitive molecule‐IEICO‐4F. Under 940‐nm pulsed illumination, the devices exhibit clear synaptic behaviors, including excitatory postsynaptic current (EPSC), paired‐pulse facilitation, and a tunable transition from short‐term to long‐term plasticity. The optoelectronic synaptic transistors operate at V DS of −10 µV with an ultra‐low energy consumption of 54.8 aJ per synaptic event and maintain stable EPSCs after 30 days of ambient storage, demonstrating outstanding power efficiency and long‐term reliability. Leveraging our device's performance, we demonstrate a covert, event‐driven 940‐nm Near‐infrared (NIR) optical tripwire for monitoring indoor trajectories under near‐dark conditions. Furthermore, we exploit the device's conductance states to emulate network weights, achieving 96.63% accuracy on MNIST with a spiking neural network and demonstrating its utility in covert imaging by attaining 91.37% aircraft recognition accuracy with a convolutional neural network under 940‐nm NIR illumination.