Regensburg 2025 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
DS: Fachverband Dünne Schichten
DS 6: Thin Film Application
DS 6.1: Talk
Wednesday, March 19, 2025, 09:30–09:45, H3
Fluorescent SiON-Doped Si Thin Films in Miniature Temperature Sensor Fabrication Using Machine and Deep Learning with Low Root Mean Square Error — Ali Karatutlu1, Timuçin Emre Tabaru2, •Zehra Gizem Mutlay1, Esra Kendir Tekgül1, Nurhan Güneş2, and Bülend Ortaç1 — 1Institute of Materials Science Nanotechnology and National Nanotechnology Research Center (UNAM) Bilkent University, Ankara, Türkiye — 2Department of Electrical Electronics Engineering, Sivas University of Science and Technology, Sivas, Turkey
SiON-doped Si, discovered by our project team last year (DOI: Advanced Optical Materials, 2023, DOI: 10.1002/adom.20230009), has a special molecular matrix with a refractive index that can be controlled depending on the amount of SiON. The counterpart structures, such as SixNy, SiOxNy, or Si-rich Si3N4, are conventional structures with general properties such as mechanical durability, thermal stability, chemical resistance, electrical insulation, and optical transparency. We report that SiON-doped Si can have a relatively even higher refractive index from 2.07 to 2.56 near the telecom wavelength of 1310 nm, depending on the SiON content. In terms of application, SiON-doped Si thin films were tested at room temperature to 200°C using different packaging materials such as borosilicate glass and aluminum, and their use as miniature temperature sensors will be demonstrated. The preliminary temperature-fluorescence spectrum correlation was investigated using machine learning and deep learning methods that yield the root mean square error of this system to be as low as 2°C.
Keywords: Miniature Temperature Sensors; SiON-Doped Si Thin Films; Multi-spectral Fluorescence; Deep Learning; Machine learning