Benchmarking time-dependent density functional theory for singlet excited states of thermally activated delayed fluorescence chromophores

Thermally activated delayed fluorescence (TADF) is the internal conversion of triplet excitons into singlet excitons via reverse intersystem crossing. TADF can significantly enhance the efficiency of organic light-emitting diodes (OLEDs). In order for a chromophore to display TADF the energy difference between its lowest singlet and lowest triplet states, S1 and T1, should be as small as possible. This requirement is facilitated by spatial separation between the frontier orbitals. Computer simulations based on time-dependent density functional theory (TDDFT) have been used extensively to predict the excited state properties of TADF chromophores. However, the accuracy of TDDFT largely depends on the choice of exchange-correlation functional. Here, we present a benchmark study of the performance of TDDFT based on different classes of hybrid functionals for 16 TADF chromophores consisting of different donor and acceptor moieties. We find that only the range-separated double hybrid functionals, ωB2PLYP and ωB2GP-PLYP, provide qualitatively correct predictions of the relative singlet excitation energies of different molecules, the spectral composition of excited states, and the energy ordering of intramolecular charge-transfer versus valence excited states. Therefore, we recommend using these functionals to assess prospective TADF chromophores. Nevertheless, further development is needed to improve the quantitative performance of TDDFT. These findings are important for our ability to computationally screen and design candidate TADF chromophores and advance the development of highly efficient OLEDs.