Machine Vision Based on an Ultra-Wide Bandgap 2D Semiconductor AsSbO3

Facing the future development trend of miniaturization and intelligence of electronic devices, solar-blind photodetectors based on ultrawide-bandgap 2D semiconductors have the advantages of low dark current, and high signal-to-noise ratio, as well as the features of micro-nanometer miniaturization and multi-functionalization of 2D material devices, which have potential applications in the photoelectric sensor part of high-performance machine vision systems. This study reports a 2D oxide semiconductor, AsSbO3, with an ultrawide bandgap (4.997 eV for monolayer and 4.4 eV for multilayer) to be used to fabricate highly selective solar-blind UV photodetectors, of which the dark current as low as 100 fA and rejection ratio of UV-C and UV-A reaches 7.6 x 10(3). Under 239 nm incident light, the responsivity is 105 mA W-1 and the detectivity is 7.58 x 10(12) Jones. Owing to the remarkable anisotropic crystal structure, AsSbO3 also shows significant linear dichroism and nonlinear optical properties. Finally, a simple machine vision system is simulated by combining the real-time imaging function in solar-blind UV with a convolutional neural network. This study enriches the material system of ultrawide-bandgap 2D semiconductors and provides insight into the future development of high-performance solar-blind UV optoelectronic devices.

Automated summary

This study presents the fabrication of highly selective solar-blind UV photodetectors using a 2D oxide semiconductor material, AsSbO3, with an ultrawide bandgap. The material exhibits low dark current and high signal-to-noise ratio, making it suitable for high-performance machine vision systems.