A mini review of machine learning in inorganic phosphors

Abstract

Machine learning has promoted the rapid development of materials science. In this review, we provide an overview of recent advances in machine learning for inorganic phosphors. We take two aspects of material properties prediction and optimization based on iterative experiments as entry points to outline the applications of machine learning for inorganic phosphors in terms of Debye temperature prediction and luminescence intensity and thermal stability optimization. By analyzing the machine learning methods and their application objectives, current problems are summarized and suggestions for subsequent development are proposed.

Recent progress in trivalent europium (Eu3+)-based inorganic phosphors for solid-state lighting: an overview

“…It is still a big challenge to screen and find suitable host lattices for Eu 3+ doping or substitution, and further research direction needs to be encouraged in computational evolutionary, 236,237 data driven, 238 mineral-inspired phosphor discovery 239 and machine learning (ML) as a tool to design new phosphors. [240][241][242] Furthermore, the anxieties for developing red emitters compare to other complementary color emitting phosphors are more stringent (which includes narrowband emission spectra, apt spectral excitation/emission wavelengths, high rigidity (thermal stability), as well as good chemical stability and high brightness. Most importantly, further study on the structure-compositions-property relationship of phosphors needs to be sustained.…”

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

“…In recent years, data-driven machine learning (ML) has been widely employed in materials research. − The development of MOO provides a means to simultaneously predict multiple properties and direct the optimization to improve on the existing Pareto front. , MOO realizes efficient design of materials through a choice of utility functions that define the space within which the optimization is performed as well as selection of next recommendations. The commonly used utility functions are expected improvement (EI), upper confidence bounds (UCB), etc.…”

Developing and optimizing novel Cr3+-activated inorganic NIR phosphors by combining triple-objective optimization and crystal field engineering

“…[37][38][39][40][41] In particular, the multi-objective ML approach can achieve the simultaneous optimization of multiple properties. [42][43][44][45] For example, in our previous work, we used a dual-objective collaborative optimization approach to obtain garnet-type NIR phosphors with an excellent performance through iterative experiments. 29 In multi-objective optimization, the optimization algorithm (utility function) can provide promising material systems to guide research.…”