Deep learning for development of organic optoelectronic devices: efficient prescreening of hosts and emitters in deep-blue fluorescent OLEDs

Abstract The highest occupied molecular orbital (HOMO) and lowest unoccupied (LUMO) energies, which are key factors in optoelectronic devices, must be accurately estimated for newly designed materials. Here, we developed a deep learning (DL) model that was trained with an experimental database containing the HOMO LUMO energies of 3026 organic molecules solvents or solids capable predicting mean absolute errors 0.058 eV. Additionally, demonstrated our DL efficiently used to virtually screen optimal host emitter light-emitting diodes (OLEDs). Deep-blue fluorescent OLEDs, were fabricated selected via prediction, exhibited narrow emission (bandwidth = 36 nm) at 412 nm external quantum efficiency 6.58%. Our DL-assisted virtual screening method can further applied development component materials optoelectronics.