AI-Dandan Song-Abstract

The integration of Machine Learning (ML)and Quantitative Structure-Property Relationships (QSPR) is profoundlytransforming the paradigm of material development. For thermally activateddelayed fluorescent (TADF) materials, utilizing descriptors from quantumchemical calculations along with various molecular descriptors, we havesuccessfully established the relationship between molecular structure and boththe horizontal orientation of the transition dipole moment (TDM) and the energylevel configuration. This understanding lays the foundation for informedmolecular design. Additionally, we have developed PLQY prediction models forMR-TADF materials and conducted rapid high-throughput virtual screening acrossextensive molecular libraries. These efforts not only facilitate thedevelopment of new luminescent materials but also underscore the effectivenessof ML in material screening.