Efficient host-free delayed fluorescence organic light-emitting diodes

Abstract

Efficient non-doped bluish-green to green thermally activated delayed fluorescence (TADF) OLEDs are fabricated with two TADF emitters, namely HDT1 and 36Me-HDT1, achieving maximum external quantum efficiency (EQE_max) values of 16.5% and 21.0%, respectively. Single-stack, three-color, warm-white hyperfluorescent OLEDs are also demonstrated with an EQE_max of 12.1%.

---

Thermally activated delayed fluorescence (TADF) has been considered as the third-generation display technology due to its ability to harvest both singlet and triplet excitons without the use of heavy atoms.^1,2 TADF materials can achieve 100% internal quantum efficiency, resulting in an external quantum efficiency of over 20% in TADF OLEDs. Despite the significant progress of high-efficiency TADF OLEDs, most reported TADF OLEDs employ the host–guest strategy,³ where the TADF emitter is doped and dispersed in the host material forming the emitting layer (EML).^4–6 This approach can largely inhibit the intermolecular interactions, hence reducing the effect of aggregation-caused quenching (ACQ). However, doped TADF OLEDs always require precise control of the doping concentration to maintain their efficiency and color purity.⁷ Moreover, host materials possess a low molecular weight, which leads to a low glass transition temperature—a disadvantage for prolonged device operation. As a result, non-doped TADF OLEDs offer an alternative to doped TADF OLEDs, if a high photoluminescence quantum yield (PLQY) of TADF materials can be achieved in a neat film. One of the possible strategies to obtain a high PLQY is to build a sterically bulky TADF emitter that can inhibit ACQ. Recently, Li, Lee, Tang, and coworkers reported non-doped blue TADF OLEDs based on benzonitrile-based TADF emitters, achieving high efficiency and low driving voltage.⁸ Based on their findings, it is expected that our previously reported sky-blue TADF emitter, i.e., HDT1, can achieve high PLQYs not only in solution and doped films, but also in neat films.⁹ Moreover, chemical modification can be made to further enhance the photoluminescence properties of the TADF emitter. In this study, we designed and synthesized a new sky-blue TADF emitter, namely 36Me-HDT1. Additionally, we fabricated highly efficient non-doped TADF OLEDs. Furthermore, efficient white OLEDs based on a hyperfluorescent architecture with three red, green, and blue (RGB) fluorescent emitters in a single EML were demonstrated for the first time.

HDT1 is based on the heterodonor strategy, utilizing carbazole and diphenylcarbazole.^10,11 Compared with diphenylcarbazole, unsubstituted carbazole is less sterically bulky. Moreover, it has been found that by substituting the 3,6-positions of carbazole, the TADF properties are fairly enhanced.¹² Thus, we proposed and modified a new sky-blue TADF emitter by replacing the carbazole in HDT1 with 3,6-dimethylcarbazole (Fig. 1). 36Me-HDT1 was well characterized by nuclear magnetic resonance spectroscopy (Fig. S1–S3, ESI†). Density functional theory (DFT) was performed to estimate the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO). 3,6-Dimethylcarbazole and 3,6-diphenylcarbazole possess stronger electron-donating ability than that of the unsubstituted carbazole. In 36Me-HDT1, although the LUMO remained localized on the central benzonitrile core, the HOMO of 36Me-HDT1 was localized on one side of 3,6-dimethylcarbazole and 3,6-diphenylcarbazole, which is different from HDT1 with the HOMO localized only on two 3,6-diphenylcarbazole units. The singlet–triplet energy gap (ΔE_ST) was calculated for 36Me-HDT1, which showed a larger ΔE_ST of 0.09 eV (0.06 eV for HDT1).

Fig. 1 Chemical structures and DFT calculations of HDT1 and 36Me-HDT1.

The photophysical properties of the newly synthesized 36Me-HDT1 in toluene at a concentration of 10⁻⁵ M were investigated at room temperature (Fig. S4 (ESI†) and Table 1). Expectedly, with increasing donor strength, the charge-transfer (CT) emission wavelength of 36Me-HDT1 was redshifted to 488 nm when compared with that of HDT1 (477 nm). The singlet energy levels of HDT1 and 36Me-HDT1 were determined to be 2.90 and 2.81 eV, respectively. The phosphorescence spectrum was also recorded in toluene solution at 77 K. It has been found that 36Me-HDT1 showed structureless emission at 77 K (Fig. S5, ESI†). Besides, 36Me-HDT1 displayed a high photoluminescence quantum yield (PLQY) of 100% in degassed toluene solution (in aerated solution, the PLQY was only 12%).

Table 1 Photophysical data of HDT1 and 36Me-HDT1 in toluene

	l PL (nm)	PLQY (%)	τ p (ns)	τ d (μs)	S1 (eV)	T1 (eV)	k RISC (s^−1)
a Prompt lifetime. b Delayed lifetime. c Measured in oxygen-saturated toluene.
HDT1	477	100 (12)^c	8.1	10.5	2.90	2.88	8.3 × 10^5
36Me-HDT1	488	100 (21)^c	11	3.3	2.81	2.79	1.4 × 10^6

---

The transient decay profile of 36Me-HDT1 was measured in toluene, in which the delayed lifetime (τ_d) was found to be 3.3 μs, which is much faster than that of HDT1 (10.5 μs) (Fig. S6, ESI†). The rate constant of the reverse intersystem crossing process (k_RISC) was determined to be 1.4 × 10⁶ s⁻¹ (vs. 8.3 × 10⁵ s⁻¹ for HDT1). Similar to the toluene solution, the 36Me-HDT1-doped film displayed a shorter delayed lifetime with a redshifted emission at 492 nm (Fig. S7, ESI†).

To evaluate the electroluminescence properties of 36Me-HDT1 as a TADF emitter, we first fabricated doped TADF-OLEDs with the following configuration: indium-tin-oxide (ITO)-coated glass (100 nm)/HAT-CN (10 nm)/TrisPCz (30 nm)/mCBP (5 nm)/mCBP: 20 wt% of HDT1 (device A) or 36Me-HDT1 (device B) (30 nm)/SF3-TRZ (10 nm)/SF3-TRZ: 30 wt% Liq (20 nm)/Liq (2 nm)/Al (100 nm). The electroluminescence characteristics of the fabricated OLEDs were then measured to compare the performance of HDT1 and 36Me-HDT1 emitters. 1,4,5,8,9,11-Hexaazatriphenylene-hexacarbonitrile (HAT-CN) served as the hole-injection layer, 9-phenyl-3,6-bis(9-phenyl-9H-carbazol-3-yl)-9H-carbazole (TrisPCz) served as the hole-transport layer, mCBP was used as the exciton-blocking and host layers, 2-(9,9′-spirobi[fluoren]-3-yl)-4,6-diphenyl-1,3,5-triazine (SF3-TRZ) was used as the electron-transport layer, and 8-hydroxyquinolinolato-lithium (Liq) and Al were employed as the electron injection and cathode layers, respectively. The turn-on voltages of devices A and B were 3.2 and 3.4 V, respectively. Both devices A and B achieved maximum EQEs of 22.7% and 26.9%, respectively (Fig. 2 and Table 2). At 1000 cd m⁻², the EQEs of A and B were found to be 16.4% and 22.5%, respectively. The smaller roll-off of device B was attributed to the higher k_RISC, which significantly reduced the accumulation of triplet excitons within the device. Consistent with the PL study, the emission maxima of devices A and B at 1000 cd m⁻² were found to be 485 and 493 nm, respectively, with full-width at half-maximum values of 80 and 79 nm.

Fig. 2 Doped devices with 20 wt% of the TADF emitter. (a) EQE versus luminance; (b) emission spectrum at 1000 cd m⁻².

Table 2 Device performance of devices A–D

Device	EML	V on (V)	EQEmax/100/1000 (%)	λ EL (nm)
a At 1 cd m^−2. b At 1000 cd m^−2.
A	20% HDT1	2.8	22.7/19.8/16.4	486
B	20% 36Me-HDT1	3.0	26.9/25.0/22.5	493
C	100% HDT1	3.2	16.5/12.4/10.4	507
D	100% 36Me-HDT1	3.6	21.3/17.3/14.8	517

---

In addition to the doped device, reports on non-doped/host-free TADF OLEDs are limited. We first measured the PLQYs of neat films of HDT1 and 36Me-HDT1, which were found to be 70% and 80%, respectively. The emission maxima of neat films of HDT1 and 36Me-HDT1 were gradually redshifted to 507 and 516 nm, respectively (Fig. S8, ESI†). The transient decay profiles of neat films were measured with τ_ds of 1.9 and 1.5 μs, respectively (Fig. S9, ESI†). The non-doped TADF OLEDs were fabricated with a similar structure to devices A and B; only the emitting layer consisted of a neat film of HDT1 (device C) or 36Me-HDT1 (device D). The turn-on voltage of devices C and D was found to be 3.2 and 3.6 V, respectively. Device C, based on HDT1, achieved an EQE_max of 16.5%, whereas device D, based on 36Me-HDT1, resulted in a higher EQE_max of 21.3% (Fig. 3 and Table 2). Due to aggregation, the efficiencies of both non-doped devices were lower than those of doped devices A and B; however, the resulting EQEs remained one of the best values amongst non-doped blue TADF OLEDs reported in the literature (Table S1, ESI†).¹³ As a result of aggregation, the emission wavelength of devices C and D was redshifted to 507 and 517 nm, respectively. White OLEDs are attractive for lighting applications; however, white TADF OLEDs have been seldom reported in the literature.^14–16 Moreover, white TADF OLEDs have often been fabricated using a two-color system, specifically blue and yellow colors. Given the successful demonstration of highly efficient non-doped sky-blue TADF OLEDs by the HDT1 neat film, we have attempted to fabricate single-stack three-color white OLEDs by using the hyperfluorescence technology, i.e., three-color HF-OLEDs (Fig. 4). For cool white OLEDs, more blue emission is highly desired. A better spectral overlap between HDT1 and νDABNA will result in more blue emission (Fig. S10, ESI†), so we had chosen HDT1 as the sensitizer for the two terminal emitters in our white HF-OLED.⁹ In the three-color HF-OLEDs, the single-layer emitting layer consisted of three fluorescent emitters.^17–19

Fig. 3 Non-doped devices. (a) EQE versus luminance; (b) emission spectra at 1000 cd m⁻².

Fig. 4 Dual energy transfer channels for three-color white HF-OLEDs.

In the emitting layer, a deep-blue MR-TADF emitter (νDABNA) and a red emitter (DBP) were doped in very low concentrations, and HDT1 acted as the TADF host and the green emitter. Due to the good spectral matching between (1) HDT1 and νDABNA and (2) HDT1 and DBP, both νDABNA and DBP were sensitized by HDT1. Single-stack white hyperfluorescent OLEDs (W-HFOLED) were fabricated by incorporating three RGB emitters in the emitting layer with the following configuration: indium-tin-oxide (ITO)-coated glass (100 nm)/HAT-CN (10 nm)/TrisPCz (30 nm)/mCBP (5 nm)/1 wt% of νDABNA: 1 wt% of DBP: HDT1 (device E) (30 nm)/SF3-TRZ (10 nm)/SF3-TRZ: 30 wt% Liq (20 nm)/Liq (2 nm)/Al (100 nm). HDT1 acted as the host and the green emitter in the W-HFOLEDs, while νDABNA acted as the blue terminal emitter and DBP acted as the red terminal emitter. During electrical excitation, singlet and triplet excitons will be generated on HDT1 first. Due to TADF properties, triplet excitons are converted back to singlet excitons via the RISC process. Then, two FRET channels exist, in which energy from HDT1 is transferred to blue νDABNA and red DBP simultaneously. Since the doping concentrations of νDABNA and DBP are kept at 1 wt%, the FRET efficiencies are relatively low, resulting in residual emission from HDT1. Eventually, the EL spectrum with RGB color can be achieved. The resulting W-HFOLEDs displayed a decent EQE_max of 12.1%, which was slightly lower than that of the non-doped HDT1 device (Fig. 5). The lower EQE can be attributed to the possibility of direct exciton formation on νDABNA and DBP. A warm-white EL color was obtained from W-HFOLEDs with CIE_x,y of (0.37, 0.40) at 1000 cd m⁻². The electroluminescence spectra at different current densities are shown in Fig. S11 (ESI†). It has been found that when the current density increased, the ratio of the emissions of HDT1 and νDABNA to that of DBP increased, which might be due to a decreased FRET to DBP at a high current density. With a greater contribution of short-wavelength emission, the W-HFOLEDs emitted warm white light in the higher current density region. Nonetheless, the corresponding device stability was good, with an LT₅₀ (50% of the initial luminance) of only 20 hours at an initial luminance of 1000 cd m⁻². The relatively short device stability may be due to the imbalanced carrier transport properties, in which the efficiency roll-off was also severe.^20–22 Nonetheless, this W-HFOLED represents, for the first time, the construction of purely fluorescent white OLEDs using three RGB emitters in a single emitting layer via a dual energy transfer channel strategy.

Fig. 5 (a) Emission spectrum at 1000 cd m⁻²; (b) EQE versus luminance (inset: photograph of the warm white emission of the HF-OLED).

Conclusions

In summary, a new sky-blue TADF emitter, namely 36Me-HDT1, has been designed and synthesized, which exhibited improved TADF properties. Highly efficient doped and non-doped TADF OLEDs have been fabricated with EQE_max values of 26.9% and 21.3%, respectively. Most importantly, single-stack warm white HF-OLEDs were first demonstrated using a hyperfluorescence strategy, which achieved an EQE_max of 12.1% and CIE_x,y of (0.37, 0.40). This finding paves a new way for the construction of highly efficient non-doped TADF OLEDs for display and lighting applications.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the ESI.†

Acknowledgements

This work was supported by the City University of Hong Kong (Project No. 9610637). The authors acknowledge JSPS KAKENHI International Leading Research (ILR) (23K20039) and Kyulux Inc. Y.-T. Lee thanks the support from the National Science and Technology Council (NSTC), Taiwan, under Grant 111-2113-M-031-008-MY3, EA0002.

References

1. H. Uoyama, K. Goushi, K. Shizu, H. Nomura and C. Adachi, Highly efficient organic light-emitting diodes from delayed fluorescence, Nature, 2012, 492, 234.
2. H. Nakanotani, T. Higuchi, T. Furukawa, K. Masui, K. Morimoto, M. Numata, H. Tanaka, Y. Sagara, T. Yasuda and C. Adachi, High-efficiency organic light-emitting diodes with fluorescent emitters, Nat. Commun., 2014, 5, 4016.
3. Y. Li, R. Weerasinghe, Y. Hu, X. Tan, B. Cai, C. Adachi and C.-Y. Chan, The magic methyl effect of thermally activated delayed fluorescent emitters on blue organic light-emitting diodes, J. Mater. Chem. C, 2025, 13, 12691.
4. D. Park, S. Kang, C. H. Ryoo, B. H. Jhun, S. Jung, T. N. Le, M. C. Suh, J. Lee, M. E. Jun, C. Chu, J. Park and S. Y. Park, High-performance blue OLED using multiresonance thermally activated delayed fluorescence host materials containing silicon atoms, Nat. Commun., 2023, 14, 5589.
5. T. Chatterjee and K.-T. Wong, Perspective on Host Materials for Thermally Activated Delayed Fluorescence Organic Light Emitting Diodes, Adv. Opt. Mater., 2018, 7, 1800565.
6. M. Shi, J. Du, B. Zhao, B. Liu, G. Yue, H. Wang, Y. Hao, Y. Miao and K. Guo, Unveiling and Refining an Interfacial-Resonant Exciplex for Ultrathin Emitters in High-Efficiency Organic Light-Emitting Diodes, ACS Mater. Lett., 2025, 7, 2397.
7. Y. Li, X. Tan, B. Cai and C.-Y. Chan, Recent Progress in Molecular Design of Boron/Nitrogen-Based Multi-Resonance Materials for Narrowband Organic Light-Emitting Diodes, Adv. Opt. Mater., 2025, 2403556.
8. S.-J. Zou, F.-M. Xie, M. Xie, Y.-Q. Li, T. Cheng, X.-H. Zhang, C.-S. Lee and J.-X. Tang, High-Performance Nondoped Blue Delayed Fluorescence Organic Light-Emitting Diodes Featuring Low Driving Voltage and High Brightness, Adv. Sci., 2020, 7, 1902508.
9. C.-Y. Chan, M. Tanaka, Y.-T. Lee, Y.-W. Wong, H. Nakanotani, T. Hatakeyama and C. Adachi, Stable pure-blue hyperfluorescence organic light-emitting diodes with high-efficiency and narrow emission, Nat. Photonics, 2021, 15, 203.
10. H. Noda, H. Nakanotani and C. Adachi, Excited state engineering for efficient reverse intersystem crossing, Sci. Adv., 2018, 4, eaao6910.
11. C.-Y. Chan, M. Tanaka, H. Nakanotani and C. Adachi, Efficient and stable sky-blue delayed fluorescence organic light-emitting diodes with CIEy below 0.4, Nat. Commun., 2018, 9, 5036.
12. C.-Y. Chan, L.-S. Cui, J. U. Kim, H. Nakanotani and C. Adachi, Rational molecular design for deep-blue thermally activated delayed fluorescence emitters, Adv. Funct. Mater., 2018, 28, 1706023.
13. Y.-Z. Shi, H. Wu, K. Wang, J. Yu, X.-M. Ou and X.-H. Zhang, Recent progress in thermally activated delayed fluorescence emitters for nondoped organic light-emitting diodes, Chem. Sci., 2022, 13, 3625.
14. G. W. Kim, H. W. Bae, R. Lampande, I. J. Ko, J. H. Park, C. Y. Lee and J. H. Kwon, Highly efficient single-stack hybrid cool white OLED utilizing blue thermally activated delayed fluorescent and yellow phosphorescent emitters, Sci. Rep., 2018, 8, 16263.
15. H. Liu, Y. Fu, B. Z. Tang and Z. Zhao, All-fluorescence white organic light-emitting diodes with record-beating power efficiencies over 130 lm W⁻¹ and small roll-offs, Nat. Commun., 2022, 13, 5154.
16. F.-C. Kong, Y.-L. Zhang, C. Quinton, N. McIntosh, S.-Y. Yang, J. Rault-Berthelot, F. Lucas, C. Brouillac, O. Jeannin, J. Cornil, Z.-Q. Jiang, L.-S. Liao and C. Poriel, Pure Hydrocarbon Materials as Highly Efficient Host for White Phosphorescent Organic Light-Emitting Diodes: A New Molecular Design Approach, Angew. Chem., Int. Ed., 2022, 61, e202207204.
17. L. Lu, Y. Guo, B. Zhao, H. Wang and Y. Miao, A new strategy to develop simple and efficient monochrome and white organic light-emitting diodes, Mater. Today, 2024, 74, 109.
18. Y. Miao, G. Wang, M. Yin, Y. Guo, B. Zhao and H. Wang, Phosphorescent ultrathin emitting layers sensitized by TADF interfacial exciplex enables simple and efficient monochrome/white organic light-emitting diodes, Chem. Eng. J., 2023, 461, 141921.
19. M. Alam, M. Nagar, S. Nayak, A. Choudhury, J. Jou and S. Vaidyanathan, Acceptor Interlocked Molecular Design for Solution-Processed Stable Deep-Blue TADF and Hyper FluorescenceOrganic LED Enabling High-Efficiency, Adv. Opt. Mater., 2022, 10, 2200376.
20. Y. Zhang, Z. Li, C. Li and Y. Wang, Suppressing Efficiency Roll-Off of TADF Based OLEDs by Constructing Emitting Layer With Dual Delayed Fluorescence, Front. Chem., 2019, 7, 302.
21. L. He, H. Wang, X. Wei, W. Sun, D. Zhang, J. Wang, H. Wang, W. Liu and L. Zhou, Achieving slow efficiency roll-off and tunable spectrum in high-performance tandem WOLEDs with mixed hosts, J. Lumin., 2024, 268, 120405.
22. S. Diesing, L. Zhang, E. Zysman-Colman and I. D. W. Samuel, A figure of merit for efficiency roll-off in TADF-based organic LEDs, Nature, 2024, 627, 747.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5tc02141c

‡ Equal contribution.

---