Seunghun Jang1,Gyoung S. Na1,Hyunju Chang1
Korea Research Institute of Chemical Technology1
Seunghun Jang1,Gyoung S. Na1,Hyunju Chang1
Korea Research Institute of Chemical Technology1
Developing inorganic phosphor with desired properties has relied on time-consuming and labor-intensive material development processes. Moreover, the results of material development experiments depend significantly on the intuitions and experiences of each researcher. For efficient and reliable materials discovery, machine learning has been widely applied to various scientific applications in materials science. However, the prediction capabilities of machine learning methods fundamentally depend on the quality of the training datasets. In this work, we constructed a high-quality and reliable database that contains experimentally validated inorganic phosphors and their optical properties for data-driven research on inorganic phosphors. Our database includes 3,432 combinations of 27 dopant elements in 2,231 host materials. The database provides material information, optical properties, measurement conditions for inorganic phosphors, and metainformation. For the validation of the collected database, we preliminarily performed machine learning on the database and evaluated the prediction results.