The 2020 demonstration of 5Cz-TRZ, a thermally activated delayed fluorescence (TADF) emitter in which multiple carbazole donors on a triazine acceptor form charge-resonance-type hybrid triplet states, showed that a reverse intersystem crossing (RISC) rate of about 1.5x10^7 s^-1 can coexist with high efficiency (>29.3% external quantum efficiency), low roll-off, and an estimated T90 of roughly 600 h at 1,000 cd m^-2. The originating work itself left three linked issues unresolved: the design's dependence on the particular donor-acceptor combination (the benzonitrile-core analogue 5CzBN keeps a much larger singlet-triplet gap, ~0.17 eV), the chemical and physical degradation pathways that ultimately set device lifetime, and the tentatively attributed microcavity origin of an electroluminescence/photoluminescence spectral difference in one device configuration. The 2020-2026 follow-up literature shows substantial but partial progress: several independent redesigns now reach the fast-spin-flip regime on other cores, and molecular-stability levers such as targeted deuteration extend lifetime, yet no later work establishes a general transferable design rule or a mechanistic account of long-term degradation for fast-RISC charge-transfer emitters; 2025-2026 statements from the field still describe the photophysics-stability relationship and long-lifetime deep-blue devices as unresolved. The open core is to determine the scope of validity of the multi-donor charge-resonance principle across acceptor cores and device contexts and to identify the degradation pathways and their emitter/host/architecture dependence that bound the operational lifetime of such devices.

Background

Organic light-emitting diodes (OLEDs) emit light from electron-hole recombination in a thin film of organic semiconductors; because spin statistics initially form triplet excitons three times as often as singlets, purely organic emitters must convert triplets to the emissive singlet state to reach high efficiency. Thermally activated delayed fluorescence (TADF) does this by reverse intersystem crossing (RISC), a thermally activated spin-flip from the lowest triplet state T1 to the first excited singlet state S1. TADF molecules are donor-acceptor (D-A) structures in which spatially separated frontier orbitals give S1 and T1 strong charge-transfer (CT) character, so the singlet-triplet energy gap ΔE(S1-T1) is small; however, pure CT-to-CT spin-flips have very small spin-orbit coupling (SOC, the interaction that mixes singlet and triplet wavefunctions and enables the spin flip), so typical TADF emitters have RISC rate constants k_RISC below ~10^6 s^-1. Slow RISC leaves long-lived triplets in the emission layer, where triplet-triplet and triplet-polaron annihilation (non-radiative loss processes in which two triplet excitons, or a triplet exciton and a trapped charge, collide and deactivate each other) cause efficiency roll-off at high brightness and accelerate chemical degradation. The mechanism usually invoked for fast RISC is vibronic coupling of the CT state to a locally excited (LE) triplet state that carries larger SOC. Cui, Gillett, Zhang et al. (Nat. Photonics 14, 636-642, 2020) introduced 5Cz-TRZ, in which five carbazole donors are attached at ortho, meta and para positions of a phenylene-triazine acceptor. The multiple donors form charge-resonance-type hybrid triplet states, i.e. triplet states in which the positive charge is shared among several donor fragments, producing a dense manifold of triplet levels close in energy to S1, a small ΔE(S1-T1), and enhanced SOC between the singlet and the relevant triplets. The measured k_RISC reached 1.5x10^7 s^-1, about two orders of magnitude above typical TADF emitters, and devices showed >29.3% maximum external quantum efficiency (EQE, the fraction of injected charges emitted as photons), <2.3% roll-off at 1,000 cd m^-2, and an estimated T90 (time for the luminance to fall to 90% of its initial value) of about 600 h at 1,000 cd m^-2, extrapolated from 35.4 h measured at 5,000 cd m^-2 by the standard lifetime-acceleration formula, under which lifetime scales with initial luminance as a power law with an empirical acceleration exponent; 5Cz-TRZ also served as a sensitizer in hyperfluorescent OLEDs, where a TADF molecule transfers its harvested singlets to a narrowband terminal fluorescent emitter. The authors noted three limitations that motivate this problem. First, the electronic structure is donor/acceptor dependent: the closely related benzonitrile-core molecule 5CzBN (2,3,4,5,6-penta-(9H-carbazol-9-yl)benzonitrile) also has a dense triplet manifold but a much larger ΔE(S1-T1) of about 0.17 eV, so the design did not obviously transfer to that acceptor core. Second, the T90 value is an extrapolated estimate for one architecture, and the specific chemical or physical degradation pathways by which such devices ultimately fail, and how they depend on emitter, host and device architecture, were not elucidated. Third, in one hyperfluorescent device configuration (device D), the electroluminescence (EL) spectrum differed from the photoluminescence (PL) spectrum of the co-deposited film; the authors attributed this 'probably' to the microcavity effect, the optical resonance and interference of the weakly reflective OLED stack that reshapes the outcoupled far-field spectrum, rather than to incomplete energy transfer from sensitizer to terminal emitter, but did not verify this attribution.

Problem Statement

Under the common objective of establishing fast-spin-flip charge-transfer (TADF) emitters as a general and operationally stable OLED platform, the source paper left three linked questions open. (1) Generality of the design: the multi-donor charge-resonance strategy was demonstrated for 5Cz-TRZ (five carbazole donors on a triazine-phenylene core, k_RISC about 1.5x10^7 s^-1), but the near-analogue 5CzBN with a benzonitrile acceptor retains a much larger singlet-triplet gap ΔE(S1-T1) of about 0.17 eV and hence a much slower expected RISC; determine which features of the acceptor core, donor identity and donor geometry (number, arrangement, conformational confinement) are necessary and sufficient to produce the combination of small ΔE(S1-T1), appreciable spin-orbit coupling, and a dense near-degenerate triplet manifold, and establish the scope of validity of the resulting design rule across acceptor cores and device contexts (solution-processed as well as vacuum-deposited, sensitizer as well as emitter), specifying which cores and contexts it covers, which counterexamples bound it, and the efficiency, color and stability penalties it incurs. (2) Operational-stability mechanisms: the reported T90 of about 600 h at 1,000 cd m^-2 is an extrapolation for a single architecture; identify the specific chemical and physical degradation pathways (for example triplet- or polaron-induced bond cleavage, annihilation-driven product formation, host or interface degradation) by which such fast-RISC charge-transfer devices ultimately fail, and quantify how these pathways depend on the choice of emitter, host and device architecture. (3) Spectral-discrepancy origin: in the hyperfluorescent device D, the electroluminescence spectrum differs from the photoluminescence spectrum of the co-deposited film; decide whether the microcavity effect of the device stack, rather than incomplete energy transfer to the terminal emitter, accounts for the difference.

References

  1. Lin-Song Cui, Alexander J. Gillett, Shou-Feng Zhang, Hao Ye, Yuan Liu, Xian-Kai Chen, Ze-Sen Lin, Emrys W. Evans, William K. Myers, Tanya K. Ronson, Hajime Nakanotani, Sebastian Reineke, Jean-Luc Bredas, Chihaya Adachi and Richard H. Friend, 'Fast spin-flip enables efficient and stable organic electroluminescence from charge-transfer states', Nature Photonics 14, 636-642 (2020). DOI: 10.1038/s41566-020-0668-z, https://doi.org/10.1038/s41566-020-0668-z
  2. Malek Mahmoudi, Dalius Gudeika, Dmytro Volyniuk, Karolis Leitonas, Rita Butkute, Iryna Danyliv and Juozas V. Grazulevicius, 'Tuning of spin-flip efficiency of blue emitting multicarbazolyl-substituted benzonitriles by exploitation of the different additional electron accepting moieties', Chemical Engineering Journal 423, 130236 (2021). DOI: 10.1016/j.cej.2021.130236, https://doi.org/10.1016/j.cej.2021.130236
  3. Bhagya Madushani, Masashi Mamada, Kenichi Goushi, Thanh Ba Nguyen, Hajime Nakanotani, Hironori Kaji and Chihaya Adachi, 'Multiple donor-acceptor design for highly luminescent and stable thermally activated delayed fluorescence emitters', Scientific Reports 13, 7644 (2023). DOI: 10.1038/s41598-023-34623-9, https://doi.org/10.1038/s41598-023-34623-9
  4. Weidong Qiu, Denghui Liu, Mengke Li, Xinyi Cai, Zijian Chen, Yanmei He, Baoyan Liang, Xiaomei Peng, Zhenyang Qiao, Jiting Chen, Wei Li, Junrong Pu, Wentao Xie, Zhiheng Wang, Deli Li, Yiyang Gan, Yihang Jiao, Qing Gu and Shi-Jian Su, 'Confining donor conformation distributions for efficient thermally activated delayed fluorescence with fast spin-flipping', Nature Communications 14, 2564 (2023). DOI: 10.1038/s41467-023-38197-y, https://doi.org/10.1038/s41467-023-38197-y
  5. Dong-Hai Zhang, Shanshan Jiang, Xiaodong Tao, Fulin Lin, Lingyi Meng, Xu-Lin Chen and Can-Zhong Lu, 'Efficient Spin-Flip between Charge-Transfer States for High-Performance Electroluminescence, without an Intermediate Locally Excited State', Research 6, 0155 (2023). DOI: 10.34133/research.0155, https://doi.org/10.34133/research.0155
  6. Sinyeong Jung, Wai-Lung Cheung, Si-jie Li, Min Wang, Wansi Li, Cangyu Wang, Xiaoge Song, Guodan Wei, Qinghua Song, Season Si Chen, Wanqing Cai, Maggie Ng, Wai Kit Tang and Man-Chung Tang, 'Enhancing operational stability of OLEDs based on subatomic modified thermally activated delayed fluorescence compounds', Nature Communications 14, 6481 (2023). DOI: 10.1038/s41467-023-42019-6, https://doi.org/10.1038/s41467-023-42019-6
  7. Jixin Jiang and Jun Yeob Lee, 'Degradation mechanisms and lifetime extending strategy of phosphorescent and thermally activated delayed-fluorescence organic light-emitting diodes', Materials Today 68, 204-233 (2023). DOI: 10.1016/j.mattod.2023.06.016, https://doi.org/10.1016/j.mattod.2023.06.016
  8. Hengwei Tian, Xuming Zhuang, Zhi-Ping Yan, Hai Bi, Zhiqiang Li, Yue Wang and Yanping Huo, 'Manipulating the photophysical properties of multi-donor molecules for fast reverse intersystem crossing in solution-processed OLED devices', Organic Electronics 131, 107085 (2024). DOI: 10.1016/j.orgel.2024.107085, https://doi.org/10.1016/j.orgel.2024.107085
  9. Tianyu Huang, Qi Wang, Hai Zhang, Yangyang Xin, Yuewei Zhang, Xiankai Chen, Dongdong Zhang and Lian Duan, 'Delocalizing electron distribution in thermally activated delayed fluorophors for high-efficiency and long-lifetime blue electroluminescence', Nature Materials 23, 1523-1530 (2024). DOI: 10.1038/s41563-024-02004-w, https://doi.org/10.1038/s41563-024-02004-w
  10. Qing-Yu Meng, Xue-Liang Wen and Juan Qiao, '"Impossible Trinity" between Efficiency, Stability, and Color Purity for Blue OLEDs: Challenges and Opportunities', Journal of Physical Chemistry Letters 15, 12571-12583 (2024). DOI: 10.1021/acs.jpclett.4c03097, https://doi.org/10.1021/acs.jpclett.4c03097
  11. Masashi Mamada, Thanh Ba Nguyen, Hajime Nakanotani, Takuji Hatakeyama and Chihaya Adachi, 'Dominant Parameters Affecting Device Durability for Blue TADF Molecules', Advanced Optical Materials 13, e00971 (2025). DOI: 10.1002/adom.202500971, https://doi.org/10.1002/adom.202500971
  12. Hai Zhang, Tianyu Huang, Jianping Zhou, Chunlin Xu, Dongdong Zhang and Lian Duan, 'Fast reverse intersystem crossing over 10^7 s^-1 via near-enantiomeric charge-transfer transitions', Chem 12, 102685 (2026). DOI: 10.1016/j.chempr.2025.102685, https://doi.org/10.1016/j.chempr.2025.102685
  13. Dongdong Zhang, Hengyi Dai, Hai Zhang and Lian Duan, 'Stable deep-blue organic light-emitting diodes based on sensitized fluorescence', Nature Photonics 20, 136-150 (2026). DOI: 10.1038/s41566-025-01810-1, https://doi.org/10.1038/s41566-025-01810-1
  14. Mauro Furno, Rico Meerheim, Simone Hofmann, Bjorn Lussem and Karl Leo, 'Efficiency and rate of spontaneous emission in organic electroluminescent devices', Physical Review B 85, 115205 (2012). DOI: 10.1103/PhysRevB.85.115205, https://doi.org/10.1103/PhysRevB.85.115205