【Epitaxy Papers】α-Ga₂O₃ Molecular Beam Epitaxy for Photodetection and Authentication

      Researchers from the King Abdullah University of Science and Technology (KAUST) have published a dissertation titled "α-Ga₂O₃ Molecular Beam Epitaxy for Photodetection and Authentication " in IEEE Photonics Technology Letters.

Abstract

      With the growing demand for secure and tamper-resistant solutions in optical communication and authentication, optoelectronic devices that can simultaneously perform high-performance photodetection and security functions are of particular significance. In this context, physical unclonable functions (PUFs) implemented using wide- and ultrawide-bandgap semiconductors are especially attractive, as they combine the ability to detect optical signals with the intrinsic randomness needed for robust anti-cloning capabilities. Exploring the inherent variation in UV-visible light optoelectronic devices for PUF implementation can be promising for applications that potentially need authentication and encryption. Herein, we present the first proof-of-concept demonstration of a machine learning assisted authentication scheme utilizing α-Ga₂O₃ based photodetector (PD), demonstrating its dual functionality, including photodetection and its potential toward PUF implementation. Transmission electron microscope analysis confirmed that the epitaxially grown α-Ga₂O₃ exhibits both single-crystalline and amorphous regions. This, together with the fabrication variance, serves to enhance the device authentication capability. The fabricated PD shows a distinct light-to-dark current response and a pronounced persistent photoconductivity effect, with a paired-pulse facilitation index of ~130%, attributed to the intrinsic defects in the α-Ga₂O₃. Furthermore, the photo-response of the fabricated PD under chopped 260-nm illumination at different voltage biases serves as digital fingerprints for authentication. By employing a convolutional neural network, the system achieves a high authentication accuracy of 99.58%. The seminal work lays the foundation for developing a full-scale device identification and authentication framework based on multiple challenge parameters, such as wavelength, pulse width, and beyond.

DOI：

https://doi.org/10.1109/LPT.2025.3639414