Deciphering the doublet luminescence mechanism in neutral organic radicals: spin-exchange coupling, reversed-quartet mechanism, excited-state dynamics

Abstract

Neutral organic radical molecules have recently attracted considerable attention as promising luminescent and quantum-information materials. However, the presence of a radical often shortens their excited-state lifetime and results in fluorescence quenching due to enhanced intersystem crossing (EISC). Recently, an experimental report introduced an efficient luminescent radical molecule, tris(2,4,6-trichlorophenyl)methyl-carbazole-anthracene (TTM-1Cz-An). In this study, we systematically performed quantum theoretical calculations combined with the path integral approach to quantitatively calculate the excited-state dynamics processes and spectral characteristics. Our theoretical findings suggest that the sing-doublet D₁ state, originating from the anthracene excited singlet state, is quickly converted to the doublet (trip-doublet) state via EISC, facilitated by a significant nonequivalence exchange interaction, with ΔJ_ST = 0.174 cm⁻¹. The formation of the quartet state (Q₁, trip-quartet) was predominantly dependent on the exchange coupling 3/2J_TR = 0.086 cm⁻¹ between the triplet spin electrons of anthracene and the TTM-1Cz radical. Direct spin–orbit coupling ISC to the Q₁ state was minimal due to the nearly identical spatial wavefunctions of the and Q₁ levels. The effective occurrence of reverse intersystem crossing (RISC) from the Q₁ to D₁ state is a critical step in controlling the luminescence of TTM-1Cz-An. The calculated RISC rate k_RISC, including the Herzberg–Teller effect, was 3.64 × 10⁵ s⁻¹ at 298 K, significantly exceeding the phosphorescence and nonradiative rates of the Q₁ state, thus enabling the D₁ repopulation. Subsequently, a strong electronic coupling of 37.4 meV was observed between the D₁ and D₂ states, along with a dense manifold of doublet states near the D₁ state energy, resulting in a larger reverse internal conversion rate k_RIC of 9.26 × 10¹⁰ s⁻¹. Distributed to the D₂ state, the obtained emission rate of k_f = 2.98–3.18 × 10⁷ s⁻¹ was in quite good agreement with the experimental value of 1.28 × 10⁷ s⁻¹, and its temperature effect was not remarkable. Our study not only provides strong support for the experimental findings but also offers valuable insights for the molecular design of high-efficiency radical emitters.