The Best of Both Worlds: ΔDFT Describes Multiresonance TADF Emitters with Wave-Function Accuracy at Density-Functional Cost

Abstract

With their narrow-band emission, high quantum yield, and good chemical stability, multiresonance thermally activated delayed fluorescence (MR-TADF) emitters are promising materials for OLED technology. However, accurately modeling key properties, such as the singlet–triplet (ST) energy gap and fluorescence energy, remains challenging. While time-dependent density functional theory (TD-DFT), the workhorse of computational materials science, suffers from fundamental issues, wave function-based coupled-cluster (CC) approaches, like approximate CC of second-order (CC2), are accurate but suffer from high computational cost and unfavorable scaling with system size. This work demonstrates that a state-specific ΔDFT approach based on unrestricted Kohn–Sham (ΔUKS) combines the best of both worlds: on a diverse benchmark set of 35 MR-TADF emitters, ΔUKS performs as good as or better than CC2, recovering experimental ST gaps with a mean absolute deviation (MAD) of 0.03 eV at a small fraction of the computational cost of CC2. When combined with a tuned range-separated LC-ωPBE functional, the excellent performance extends to fluorescence energies and ST gaps of MR- and donor–acceptor TADF emitters and even molecules with an inverted ST gap (INVEST), rendering this approach a jack of all trades for organic electronics.