Visible-light-stimulated synaptic InGaZnO phototransistors enabled by wavelength-tunable perovskite quantum dots

Abstract

Neuromorphic vision sensors are designed to mimic the human visual system, which allows image recognition with low power computational requirements. Photonic synaptic devices are one of the most viable building blocks for constructing neuromorphic vision sensors. Herein, a photonic synaptic sensor based on an inorganic perovskite quantum dot (QD) embedded InGaZnO (IGZO) thin-film phototransistor is demonstrated. The photodetection wavelength ranges of the transistor can be adjusted by changing the halogen ions (Cl, Br) of the perovskite QDs. Under low intensity 450 and 550 nm illumination, the CsPbBr₃ QD embedded phototransistor sensor shows a responsivity of 6.7 × 10² and 4.2 × 10⁻² A W⁻¹, respectively. The perovskite QD embedded transistor not only presents high responsivity to visible light, but also features excellent synaptic behavior, including an excitatory postsynaptic current (EPSC), pair-pulse facilitation (PPF), long-term memory, and memory erasure through gate voltage regulation. Moreover, the sensor fabrication process in this work is compatible with conventional photolithography processes. Taking these merits into account, the proposed QD embedded IGZO transistor presents a promising route by which to construct artificial visual sensors with color-distinguishable optical signal sensing and processing.