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

Abstract

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 °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.

Rapid Discovery of Efficient Long-Wavelength Emission Garnet:Cr NIR Phosphors via Multi-Objective Optimization

“…Gopakumar et al reported EI centroid and EI maximin strategies to optimize the multi-objective performance of shape memory alloys and piezoelectric materials. In our previous work, we introduced an EI-based layer-by-layer screening strategy that achieved the optimization of the thermal stability of Ce 3+ -doped garnet phosphors in a specific wavelength range. Hanaoka proposed the probability of goal achievement (PA), which extends a lower confidence bound to the goal-achievement-inverse-design by automatically controlling a parameter based on the distance from the goal.…”

Bibliometric Analysis of Fourth Industrial Revolution Applied to Material Sciences Based on Web of Science and Scopus Databases from 2017 to 2021

“…As such, Zhang Yan can be considered the most influential author in the studied subjects. In his latest research, he has used machine learning for material design and microstructure evolution prediction [36,37]. It is worth noting that images from Figure 2 are not tailored to the data of Table 1 since these networks' charts are focused on searching for collaborations-total link strength-and they depend on the minimum number of articles per author and on the decision of presenting the interconnection of nodes.…”

A mini review of machine learning in inorganic phosphors

“…In order to optimize the color temperature (CCT) and color rendering index (CRI) of the (TDMP)Pb(Br x Cl 1-x ) 4 :yMn phosphor system, Yuan et al trained CCT and CRI polynomial regression ML models based on 55 sets of experimental data, with the material chemical compositions as features [38] . During the iteration, the (x, y) of randomly selected 15 CCTs at 3000-7000 K were determined by the CCT model and then the maximum CRI corresponding to each CCT was determined by the CRI model.…”

“…Optimizing luminous intensity of REOF:xEu 3+[35] 2019Optimizing luminous intensity of REOF:xCe, yTb (RE = Gd, La or Lu)[36] 2020Optimizing luminous intensity of LuOF:Yb/Er/Mn/K/Li[37] 2021Optimizing color temperature and color rendering index of white phosphors[38] 2022Optimizing thermal stability of cyan-green garnet:Ce phosphors[39] …”