Multiobjective Machine Learning-Assisted Discovery of a Novel Cyan-Green Garnet: Ce Phosphors with Excellent Thermal Stability

Ce-doped garnet phosphors play an important role in the white light-emitting diode (LED) family. In the past years, a lot of trial-and-error experiments guided by experience to discover phosphors suitable for white LEDs have been presented. The working temperature of phosphors may reach 200 degrees C in white LEDs, and so, the exploration of phosphors with excellent thermal stability at the desired wavelength continues to be a challenge. In the present study, to discover novel cyan-green garnet:Ce phosphors, wavelength and thermal stability machine learning models were built by constructing reasonable features. Among the 171,636 compounds with garnet structures predicted by our models, 25 samples were selected for preparation and characterization by multiobjective optimization based on active learning. Lu1.5Sr1.5Al3.5Si1.5O12:Ce performed the best with excellent thermal stability (>= 60% emission intensity was retained at 640 K) and exhibited emission peaks of about 505 nm, and it is a very promising phosphor for future applications, especially in high-temperature operating environments.

Automated summary

In this study, machine learning models for wavelength and thermal stability were built to discover novel cyan-green garnet:Ce phosphors. Through multiobjective optimization based on active learning, Lu1.5Sr1.5Al3.5Si1.5O12:Ce was selected as the best-performing sample with excellent thermal stability (>60% emission intensity retained at 640 K) and emission peaks around 505 nm, making it a very promising phosphor for future applications, especially in high-temperature operating environments.