Artificial synapses based on CIPS/Te vdW heterojunction ferroelectric transistor for traffic light recognition

Abstract

Neuromorphic computing is a key technology for simulating brain function and plays a crucial role in the next-generation computing, offering a potential solution to the challenges posed by the von Neumann bottleneck. Tellurium (Te) and CuInP2S6 (CIPS), as two-dimensional (2D) materials with excellent properties, have been widely used in advanced electronics and optoelectronics. However, the combination of the stable ferroelectricity of CIPS and the high current characteristics of Te, which both electrical and optical stimuli can modulate, offers great potential for addressing complex application scenarios, yet this type of 2D van der Waals (vdW) device has been largely unexplored. In this study, we developed an optoelectronic neuromorphic device based on CIPS and Te, which exhibits fundamental synaptic behaviors in response to electrical stimulation and demonstrates different current responses under light of varying wavelengths. Additionally, we constructed an RC system based on this device to address the problem of traffic light recognition. In this system, the ferroelectric modulation of CIPS by voltage enables short-term depression (STD) to simulate human braking behavior in response to dangerous signals. This approach effectively enhances the response capabilities of intelligent traffic systems to traffic signals, offering significant application potential.