Research Projects

Organic light-emitting diodes (OLEDs) have emerged as display and lighting technologies due to their flexibility, cost-effectiveness, and environmental sustainability. However, a critical challenge in the advancement of OLEDs is creating affordable, metal-free emitters to harness both singlet and triplet excitons to achieve high internal quantum efficiency (IQE). To address this limitation, Thermally Activated Delayed Fluorescence (TADF) materials have emerged as promising candidates. By facilitating reverse intersystem crossing (RISC) from triplet (T1) to singlet-excited states (S1) without relying on heavy metals like iridium or platinum, TADF can achieve up to 100% IQE. The efficiency of RISC depends critically on minimizing the singlet-triplet energy gap (ΔEST), which is essential for efficient exciton upconversion. Recent strategies to achieve efficient TADF materials, Multi-Resonance TADF (MR-TADF), and Through-Space Charge Transfer (TSCT) have received significant attention. MR-TADF materials feature rigid molecular frameworks incorporating boron and nitrogen atoms, offering small ΔEST and narrowband emission. Meanwhile, TSCT moieties employ spatially separated donor and acceptor molecules, typically connecting linkers like spiro carbons, π-π stacking, or macrocyclic geometries to facilitate charge transfer. Despite recent progress, the rational design of MR-TADF and TSCT-based emitters remains highly challenging, due to the complex relationship between molecular geometry, excited-state dynamics, and charge transfer behavior. In this context, we aim to design novel MR-TADF and TSCT-based emitters and investigate their properties through computational methods. Here, we will employ computational techniques such as density functional theory (DFT), time-dependent DFT (TD-DFT), and spin-orbit coupling (SOC) analysis to predict key properties such as emission wavelengths, ΔEST, and RISC rates. To enhance quantum chemical calculations, the project will also incorporate machine learning (ML) models trained on high-quality computational datasets for accelerating the discovery and optimization of new TADF materials. These models will facilitate rapid identification and rational design of photophysical properties. As a result, they drastically reduce both development time and the costs of materials discovery. We will utilize ML algorithms, including neural networks to establish correlations between molecular descriptors and key TADF performance metrics.