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 molecular orbital (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 and LUMO energies of 3026 organic molecules in solvents or solids and was capable of predicting the HOMO and LUMO energies of molecules with the mean absolute errors of 0.058 eV. Additionally, we demonstrated that our DL model was efficiently used to virtually screen optimal host and emitter molecules for organic light-emitting diodes (OLEDs). Deep-blue fluorescent OLEDs, which were fabricated with emitter and host molecules selected via DL prediction, exhibited narrow emission (bandwidth = 36 nm) at 412 nm and an external quantum efficiency of 6.58%. Our DL-assisted virtual screening method can be further applied to the development of component materials in optoelectronics.

Novel π-Extended Indolocarbazole-Based Deep-Blue Fluorescent Emitter with Remarkably Narrow Bandwidth for Solution-Processed Organic Light-Emitting Diodes

“…First, we designed 1,4-bis(indolo[3,2,1-jk]carbazol-2-yl)benzene (BICB) based on ICz and theoretical calculations and deep learning predictions were used to estimate its optical properties to ensure deep-blue emission (Figure S1 and Table S1). 32,33 Both theoretical calculations and deep learning predictions indicate that BICB has a relatively large Stokes shift and a wide emission bandwidth. This can be attributed to the fact that BICB has intramolecular torsional degrees of freedom for the phenyl linker located at the center of the ICz molecule.…”

Artificial Intelligence for Conjugated Polymers

“…These innovative devices utilize conjugated polymers to convert electrical energy into light and require intricate material design and optimization for enhanced performance. ML techniques offer a powerful toolset to navigate this complexity. , A particular category of PLEDs deserving special attention is the class known as thermally activated delayed fluorescence (TADF)-type PLEDs. This intriguing category capitalizes on nonemissive triplet states that can be harnessed through reverse intersystem crossing, resulting in heightened quantum yields of fluorescence.…”

Deep Learning-Based Chemical Similarity for Accelerated Organic Light-Emitting Diode Materials Discovery

“…Despite its high impact, yet, finding new materials has traditionally been slow, labor-intensive, and frequently based on serendipity . Computer-aided materials design (CAMD) has emerged in recent decades, facilitated by the rapid development of computing power. − It aims to accelerate materials discovery by screening a large number of virtual materials via computation. However, this also poses a challenge in dealing with a vast chemical space in terms of both accuracy and efficiency since only a limited number of candidates can undergo experimental validation.…”