Ultraviolet Light Protection in Human Pupillary Response Emulation by Using Crystalline Self-Assembled Monolayer in Synaptic Transistors

Photonic synapses possess a variety of promising applications such as biorobotics and artificial intelligence (AI), in which the pursuit of lightweight, miniaturized, and low-energy consumption designs is a highly sought-after goal for improving efficiency and adaptability in evolving technological landscapes. To meet this challenge, we design a series of conjugated self-assembled molecules featuring photoactive cores such as pyrene, benzo-naphthol-thiophene (BNT), perylene, and benzothieno-benzothiophene, which form ultrathin (<3 nm) charge-trapping self-assembled monolayers (SAMs). The crystalline BNT aligns in an orderly arrangement with the semiconductor channel, demonstrating outstanding current contrast stability (~10⁸) and synaptic properties such as paired-pulse facilitation (153%), ultralow energy consumption (28.9 aJ), and both short-term and long-term plasticity. The device based on crystalline BNT successfully emulates human learning behaviors and exhibits ultraviolet light protection, using different types of charge traps within the crystalline and conjugated SAMs. Additionally, its image denoising capability is shown by achieving a high recognition accuracy (~90%) for preprocessed images. This study reveals the versatile functionality of self-assembled monolayers in optoelectronic devices, highlighting the significant potential of miniaturization for applications in AI and biomimetic field.