Electrically erasable multi-level charge trapping memory with metal nanoparticle engineering for organic synaptic transistors

The development of wearable neuromorphic electronics is critical for advancing human-machine interfaces, personalized healthcare, and brain-inspired computing. Organic synaptic transistors (OSTs) have emerged as promising candidates due to their biocompatibility, mechanical flexibility, and tunable optoelectronic properties by molecular design. However, achieving efficient electrical erasing in charge-trapping-based OSTs remains challenging, particularly for oligomeric semiconductors with relatively large bandgaps. Here, we introduce a novel approach to enhance the vertical electric field in OSTs by incorporating metal nanoparticles (NPs) on top of a wide-bandgap organic semiconductor, significantly improving erase operations. The proposed device demonstrates an enlarged memory window and successful realization of 30 distinct potentiation and depression states, overcoming the write-once-read-many limitation observed in conventional charge-trapping devices. Furthermore, neural network simulations employing our multi-level memory states achieved an 87.3% classification accuracy on hand-written digit dataset, comparable to software-based systems. This work provides a simple yet efficient strategy for engineering neuromorphic transistors, paving the way for next-generation artificial intelligence hardware.