Blue emission in sterically shielded multiresonant thermally activated delayed fluorescence emitters

Abstract

Multi-resonant TADF materials are a promising class of emitters capable of addressing the BT.2020 industry requirement for blue emission in electroluminescent displays as they simultaneously show narrowband emission and can harvest both singlet and triplet excitons to produce light. However, these emitters are typically planar and prone to aggregation and their moderately large singlet–triplet energy gap (ΔE_ST) leads to slow upconversion kinetics resulting in severe efficiency roll-off in the device. In this study we present a molecular design that simultaneously results in an emitter having a faster reverse intersystem crossing rate constant (k_RISC) and suppressed aggregation in the film state. Mes-tDABNA emits at λ_PL of 465 nm as 4 wt% doped films in SF3-RZ and has a short delayed lifetime of 45.2 μs. Vacuum-deposited OLEDs with Mes-tDABNA showed blue emission at CIE coordinates of (0.13, 0.15) and a maximum external quantum efficiency, EQE_max, of 18.4%. Unsurprisingly, these devices suffered from rather strong efficiency roll-off (EQE₁₀₀₀ of 5.6%). With the aim of addressing this efficiency roll-off, hyperfluorescent devices containing DMAC-DPS as a TADF sensitizer were fabricated, which showed an improved EQE_max of 23.1% at CIE coordinates of (0.13, 0.17) and milder efficiency roll-off (EQE₁₀₀₀ of 12.7%). These devices showed one of the highest EQE₁₀₀₀ based on DABNA-based emitters to date.

---

Introduction

In recent years, thermally activated delayed fluorescence (TADF) emitters have emerged as one of the most promising emitter materials for organic light-emitting diodes (OLEDs). This is because of their excellent exciton harvesting properties that rival those of phosphorescent materials. TADF emitters are capable of harvesting up to 100% of the electrically generated excitons due to their small energy gap between the lowest excited singlet (S₁) and triplet excited states (T₁), ΔE_ST.¹ A subclass of TADF compounds, multi-resonant TADF (MR-TADF) emitters, produces desirable, bright, narrowband emission associated with high photoluminescence quantum yields (Φ_PL > 90%) and narrow full-width half maxima (FWHM < 50 nm). MR-TADF emitters are typically p- and n-doped polycyclic aromatic hydrocarbons (PAHs) having low-lying excited states possessing short-range charge transfer (SRCT) character.^2–7 A key challenge in emitter design, including for MR-TADF emitters, is achieving the blue Commission Internationale de l’Éclairage (CIE) coordinates of (0.131, 0.046) specified by the BT.2020 standard for pure blue emission.

However, the planar structure of most MR-TADF emitter cores makes them prone to aggregation, leading to red-shifted and broadened emission spectra, as well as aggregation-caused quenching (ACQ), all of which are detracting features that adversely affect the color and Φ_PL of blue MR-TADF emitters. For example, the first reported MR-TADF emitter, DABNA-1 (Fig. 1), emits at λ_PL at 460 nm (FWHM = 28 nm) and has a Φ_PL of 88% in 1 wt% doped films in mCBP.⁸ However, Lee et al. disclosed that serious ACQ occurred when concentrations of greater than 5 wt% of DABNA-1 were used in doped films,⁹ and peripheral bulky groups like tert-butyl moieties can mitigate it. The derivative t-DABNA (Fig. 1) maintains a high Φ_PL of 85%, and blue λ_PL of 464 nm at 5 wt% doping in DPEPO film. The OLEDs with 5 wt% t-DABNA in DPEPO showed a higher EQE_max of 25.1% at λ_EL of 467 nm and CIE coordinates of (0.13, 0.15) compared to the device with 5 wt% DABNA-1 in mCBP (EQE_max = 12.5%; λ_EL of 461 nm, CIE coordinates of (0.13, 0.10)).⁸ However, both devices still suffered from severe efficiency roll-off and the devices did not even reach a luminance of 1000 cd m⁻².⁹ There have since been a number of studies focussing on addressing this issue by decorating the tDABNA core with different bulky moieties. For instance, tDABNA-dtB,¹⁰ DABNA-NP-TB (also published as 3tPAB and t-DAB-DPA),^11–13 tDPAC-BN¹⁴ and M-tDABNA¹⁵ are all examples with bulky substituents at the para position to the boron on the central phenyl ring (Fig. 1). The DABNA derivatives containing an electron-donating group show blue-shifted emission at λ_PL of 461, 458, and 454 nm for M-tDABNA (3 wt% mCBP), DABNA-NP-TB (3 wt% mCP) and tDPAC-BN (1 wt% : 30 wt% DMAc-DPS : DPEPO), respectively, compared to tDABNA (466 nm, 3 wt% mCBP). Whereas t-DABNA-dtB, contains uses di-tert-butylphenyl substituent, shows a red-shifted emission to 472 nm (5 wt% mCBP), compared to tDABNA (466 nm, 3 wt% mCBP), this red shift can be explained by the larger conjugated system in this molecule. The device with 5 wt% t-DABNA-dtB showed a rather low EQE_max of 11.4% but also low efficiency roll-off, reflected in an EQE₁₀₀₀ of 10.9%. The lower EQE_max is due to the choice of device stack that is optimised for device lifetime rather than high EQE (LT₉₅ = 205 h). The CIE coordinates of the device of (CIE 0.11, 0.14) are similar to those of devices with DABNA-1. The OLED with the donor-extended derivative tDPAC-BN showed an EQE_max of 21.6% at CIE coordinates of (0.14, 0.09). Similar to the devices with the aforementioned derivatives as emitters, that with tDPAC-BN showed a reduced efficiency roll-off and the EQE₁₀₀₀ was 5.4%. Devices reported by Kim et al. employing DABNA-NP-TB (t-DAB-DPA) as the emitter showed a higher EQE_max of 27.9% [CIE coordinates of (0.13, 0.08)] as compared to that with tDABNA [23.9%; CIE coordinates of (0.13, 0.10)] accompanied by a less severe efficiency roll-off (EQE₁₀₀₀ of 8.1% as compared to 3.4% for the device with tDABNA).¹¹ This consistently improved performance of the OLEDs employing tDABNA derivatives as emitters compared to devices with the parent can be correlated to two factors: one is the reduced ACQ and second is the faster k_RISC of 3.97 × 10⁴ s⁻¹ for DABNA-NP-TB (t-DAB-DPA) (3 wt% in mCBP:mCBPCN) versus 1.17 × 10⁴ s⁻¹ for tDABNA (3 wt% in mCBP:mCBPCN),¹¹ which reduces the likelihood of triplet quenching processes.

Fig. 1 Chemical structures, photophysical properties and device performance of Mes-tDABNA and examples of tDABNA derivatives in literature.^10–15

From these examples, a correlation between reduced efficiency roll-off and mitigated ACQ can be drawn. As previously reported by some of us, the introduction of mesityl groups onto planar MR-TADF cores can effectively reduce ACQ. One example is the decoration of three mesityl groups about the carbonyl/nitrogen based core DiKTa (QAO).¹⁶ The OLEDs with Mes3DiKTa showed an EQE_max of 21.1% and an EQE₁₀₀ of 14.5%, which represents an improved performance as compared to the device with DiKTa (EQE_max = 14.7%; EQE₁₀₀ = 8.5%).¹⁷

In this study, we present a novel blue MR-TADF emitter, Mes-tDABNA (Fig. 1) that conserves the blue emission of the tDABNA core while simultaneously mitigating aggregation and alleviating efficiency roll-off in vacuum-deposited OLEDs. This is achieved by incorporating a mesityl group at the para position to the boron atom, which in contrast to the 3,5-di-tert-butyl-phenyl group used in t-DABNA-dtB does not increase the conjugation length and only serves as a steric blocking group. Theoretical calculations predict a small ΔE_ST of 160 meV and low-lying SRCT excited states. In toluene solution, Mes-tDABNA shows narrowband blue emission at λ_PL of 460 nm (FWHM of 25 nm). The emission is red-shifted to 464 nm and broadened (FWHM of 39 nm) as a 4 wt% doped film in SF3-TRZ. In these films, Mes-tDABNA has a prompt lifetime, τ_p,avg of 13.4 ns, and a delayed emission lifetime, τ_d,avg, of 45 μs. This leads to a relatively faster k_RISC of 5.53 × 10⁴ s⁻¹ as compared to tDABNA (k_RISC = 2.44 × 10³ s⁻¹; 5 wt% doped films in DPEPO).⁹ The OLEDs showed an EQE_max of 18.4% and a maximum luminance (L_max) of 3600 cd m⁻². Unfortunately, these devices showed a strong efficiency roll-off (EQE₁₀₀₀ of 5.6%), so we explored a hyperfluorescent (HF) device structure to enhance the device performance. The HF devices showed an improved EQE_max of 23.1% and a milder efficiency roll-off (EQE₁₀₀₀ of 12.7%). The maximum luminance was also more than doubled to 8400 cd m⁻².

Results and discussion

Synthesis and characterisation

The synthesis was carried out as shown in Scheme 1. Nucleophilic aromatic substitution of bis(di-tert-butyl)amine with bromo-4-chloro-3,5-difluorobenzene gave 2 in 43% yield. The mesityl group was installed via a Suzuki–Miyaura cross-coupling between mesitylboronic acid and 2 in 84% yield. Compound 3 was subjected to a tandem lithiation and borylation protocol⁸ to afford Mes-tDABNA in a 21% yield, which is somewhat lower compared to those reported for DABNA-1 (32%)⁸ and tDABNA (31%)⁹ but similar to that reported for M-tDABNA (27%).¹⁵ The identity and purity of Mes-tDABNA and the intermediates were determined using melting point (Mp) analysis, ¹H and ¹³C nuclear magnetic resonance (NMR) spectroscopy, high-resolution mass spectrometry (HRMS), high performance liquid chromatography (HPLC) and elemental analysis. Thermal gravimetric analysis (TGA) and Mp determination revealed that Mes-tDABNA has a high melting point of 300–304 °C and degradation temperature, T_d (at 5% weight loss) of 363 °C (Fig. S3), the latter of which is lower compared to that of tDABNA (419 °C).⁹

Scheme 1 Synthesis of Mes-tDABNA.

Theoretical calculations

The optimized ground-state geometry and electronic structure of Mes-tDABNA were calculated using density functional theory (DFT) at the PBE0/6-31G(d,p) level in the gas phase.¹⁸ The HOMO is localized on the central benzene and one half of each of the amine donor moieties that is conjugated to it. Due to the symmetry of the compound, there is no electron density on either the boron atom or the ipso or para carbons to it of the central aryl ring. The electron density of the LUMO is distributed over the same skeleton as the HOMO but with significant density on boron and also the carbon centres ipso and para to it. The calculated HOMO/LUMO energy levels are −4.82/−0.94 eV. Both the HOMO and LUMO are stabilized compared to those of M-tDABNA (−4.79/−0.88 eV) (Fig. S1), which is due to the weakly inductively electron-withdrawing character of the mesityl group.

The spin–orbit coupling matrix element (SOCME) between S₁ and T₁ state at the T₁ optimized geometry calculated at the TDA-DFT-PBE0/6-31G(d,p) level of theory is not unusually small at 0.06 cm⁻¹.¹⁹ The excited-state energies were calculated at the spin-component scaling second-order algebraic diagrammatic construction (SCS-(ADC)2/cc-pVDZ) level, as this level of theory has been shown to predict accurately the ΔE_ST of MR-TADF compounds.²⁰ The energy for the first excited singlet (S₁) and triplet (T₁) states are 2.91 and 2.75 eV, respectively, leading to a moderately small energy gap between (ΔE_ST) of 0.16 eV. Compared to M-tDABNA (3.15/3.03 eV), both the S₁ and T₁ states are stabilized while the ΔE_ST is larger by 40 meV. The difference density plots of the S₁ and T₁ states are shown in Fig. 2 (Fig. S1 for M-tDABNA). Their patterns indicate that both of these states possess SRCT character. The S₀–S₁ transition has a calculated oscillator strength, f, of 0.24, which is lower than the corresponding transition in M-tDABNA (0.28).

Fig. 2 Calculated HOMO and LUMO energies and the electron density plots calculated of M-tDABNA at the PBE0/6-31G(d,p) level in the gas phase and calculated excited-state energies and the corresponding difference density plots for hole and electron for the S₁ and T₁ states at the SCS-ADC(2)/cc-pVDZ level in the gas phase (blue: negative, red: positive, isovalue: 0.02).

Photophysical properties

The absorption and steady-state photoluminescence (PL) spectra in dilute toluene (×10⁻⁵ M) are shown in Fig. 3. The absorption spectrum of Mes-tDABNA shows an intense lowest energy band at 444 nm with a molar extinction coefficient (ε) of 24 × 10³ M⁻¹ cm⁻¹. This band is associated with the SRCT transition and is essentially isoenergetic and isoabsorptive as that of tDABNA (444 nm, 21 × 10³ M⁻¹ cm⁻¹).²¹ The PL spectrum is narrowband with a peak maximum, λ_PL, of 460 nm and a full-width at half maximum (FWHM) of 25 nm. This is effectively the same as that of tDABNA (λ_PL of 457 nm and FWHM of 26 nm in toluene).¹⁰ The small Stokes shift of 16 nm is associated with the rigid structure and the small degree of geometric reorganization in the excited state of this compound. The photoluminescence quantum yield, Φ_PL, is 75%, which decreases to 54% under aerated conditions. The time-resolved photoluminescence measurements revealed monoexponential decay kinetics, with a prompt lifetime, τ_p, of 7.9 ns, while no delayed emission was observed for Mes-tDABNA similar to both tDABNA and M-tDABNA, and no solution Φ_PL or lifetime are reported in the literature. The S₁ and T₁ levels were determined from the onsets of the steady-state and time-gated PL spectra in dilute 2-methyltetrahydrofuran (2-MeTHF) glass at 77 K (Fig. 3). The S₁ is 2.78 eV and the T₁ is 2.65 eV, resulting in a ΔE_ST of 130 meV, all of which are in good agreement with those predicted by the SCS-ADC(2) calculations (S₁ = 2.91 eV; T₁ = 2.75 eV; ΔE_ST = 0.16 eV), and is smaller than those reported for tDABNA (2.82/2.62 eV, ΔE_ST = 200 meV in toluene/frozen THF).¹⁰ Compared to M-tDABNA (ΔE_ST = 110 meV in toluene)¹⁵ the ΔE_ST is larger by 20 meV, which again is in good agreement with the trend predicted by the SCS-ADC(2) calculations (Fig. 2 and Fig. S1).

Fig. 3 Absorption and PL spectra of dilute solutions of Mes-tDABNA; (a) absorption (dashed) and PL spectra (solid) of Mes-tDABNA in dilute (×10⁻⁵ M) toluene solution (λ_exc = 340 nm); (b) steady-state PL spectra of Mes-tDABNA in dilute 2-MeTHF solution/glass at room temperature (black) and 77 K (blue), time gated (1–9 ms) PL spectrum of Mes-tDABNA in 2-MeTHF glass (red) (λ_exc = 340 nm).

The n-type material SF3-TRZ was chosen as the host for our emitter due to its suitably high T₁ energy (2.80 eV) and balanced carrier transporting abilities.^5,22 Excitation at 340 nm into the host of the 4 wt% doped film resulted in efficient Förster resonance energy transfer (FRET) and a Φ_PL of 58% (Fig. S6); in air the Φ_PL decreases to 20%. In this film Mes-tDABNA emits at λ_PL of 464 nm, with a slightly broader envelope (FWHM of 39 nm), which may indicate that aggregates in the solid state contribute to the emission profile (Fig. 4). The time-resolved PL measurements using time correlated single photon counting (TCSPC) revealed an average τ_p,avg of 13.4 ns (fitted using a multiexponential decay function) and a long emission tail. This delayed emission decay was measured using multichannel scaling (MCS), with an average delayed emission lifetime, τ_d,avg, of 45.2 μs. Temperature-dependent time-resolved PL measurements revealed a very weak temperature dependence of the prompt fluorescence decay and a strong temperature dependence of the delayed fluorescence decay, the latter of which confirmed the TADF character of this compound (Fig. 4 and Fig. S8). Additionally, the temperature-dependent PL spectra revealed the emergence of a second emission band at λ_PL of 491 nm as the temperature decreased below 150 K, which can be ascribed to phosphorescence (Fig. S9). A ΔE_ST of 140 meV was determined from the difference in the onsets of the steady-state PL and the phosphorescence spectra at 77 K. The corresponding S₁ (2.78 eV) and T₁ (2.64 eV) state energies coincide with those measured in 2-MeTHF glass. The rate constants for radiative decay (k_R) and non-radiative decay (k_NR) were determined to be 1.46 × 10⁷ and 1.49 × 10⁷ s⁻¹, respectively. The intersystem crossing rate constant for the transition between S₁ and T₁ (k_ISC) is 4.47 × 10⁷ s⁻¹, while the reverse intersystem crossing rate constant (k_RISC) for the transition from the T₁ level to S₁ level is 5.53 × 10⁴ s⁻¹, a value that is faster than that of t-DABNA (2.15 × 10⁴ s⁻¹ as a 3 wt% doped films in mCBP¹⁰). A summary of photophysical data is provided in Table 1.

Fig. 4 Photophysical data of the 4 wt% doped film of Mes-tDABNA in SF3-TRZ; (a) steady-state PL spectra at room temperature (black) and 77 K (blue) and time-gated emission spectrum (1–9 ms) at 77 K (red) (λ_exc = 340 nm); (b) time-resolved PL decay under vacuum measured by TCSPC (λ_exc = 375 nm); (c) time-resolved PL decay under vacuum measured by MCS (λ_exc = 375 nm); (d) temperature-dependent time-resolved PL decay (λ_exc = 375 nm).

Table 1 Photophysical data of Mes-tDABNA in dilute solution and doped film

Compound	λAbs(ε)^a/nm (×10^3 M^−1 cm^−1)	λPL^a^b/nm	FWHM^a^b/nm	ES1^c^d/eV	ET1^e^f/eV	ΔEST^g/eV	ΦPL^h^i/%	τp,avg^j/ns	τd,avg^k/μs	kISC^l/×10^7 s^−1	kRISC^l/×10^4 s^−1	ks_r^l/×10^7 s^−1
a In toluene solution (10^−5 M).b Measured as spin-coated thin films consisting of 4 wt% emitter in SF3-TRZ host. λexc = 340 nm.c Onset of steady-state emission at 77 K in 2-MeTHF glass λexc = 340 nm.d Onset of the SS PL at 77 K in 4 wt% doped film in SF3-TRZ. λexc = 340 nm.e Onset of the time-gated PL spectrum (1–9 ms) at 77 K in 2-MeTHF glass. λexc = 340 nm.f Onset of the time-gated PL spectrum (1–9 ms) at 77 K in 4 wt% doped film in SF3-TRZ. λexc = 340 nm.g ΔEST = E(S1) − E(T1).h Absolute ΦPL of the thin films measured using an integrating sphere. λexc = 340 nm.i ΦPL in solution was measured by the relative method using quinine sulfate as the reference (Φr = 54.6% in 1 N H2SO4).^23j Prompt PL lifetimes were measured by TCSPC and fitting the decay to a multiexponential decay function λexc = 375 nm.k Delayed PL lifetimes were measured by MCS and fitting the decay to a multiexponential decay function λexc = 375 nm.l Intersystem and reverse intersystem crossing rate constants were calculated using the steady-state approximation method as described in the literature.^24
Mes-tDABNA	444 (24)	460/464	25/39	2.78/2.78	2.65/2.64	0.13/0.14	58/75	7.9/13.4	—/45.2	4.47	5.53	1.49

---

Electrochemistry

Cyclic voltammetry (CV) and differential pulse voltammetry (DPV) measurements in degassed dichloromethane (DCM) containing 0.1 M [ⁿBu₄N]PF₆ as a supporting electrolyte and Fc/Fc⁺ as the internal reference (0.46 V vs. saturated calomel electrode (SCE)²⁵) were used to determine the HOMO and LUMO energies (Fig. S2). The oxidation wave is irreversible, with an E_ox of 0.72 V vs. SCE. The corresponding HOMO level is −5.07 eV. This is similar to the HOMO levels reported for 3tPAB (−5.09 eV) and tDPAC-BN (−5.10 eV).^13,14 No reduction could be detected, indicating a shallow LUMO level. Therefore, the LUMO level was inferred from the E_ox and the optical bandgap, itself determined to be 2.74 eV from the intersection of the normalized absorption and emission spectra in dilute toluene solution (Fig. S13). The corresponding LUMO level is −2.33 eV, which also aligns with the reported measured values for 3tPAB (−2.36 eV) and tDPAC-BN (−2.37 eV) in DCM.^13,14 However there are varying HOMO and LUMO levels for DABNA-based emitters reported in the literature. For tDABNA, M-tDABNA and t-DABNA-dtB the reported HOMO/LUMO level are −5.72/−3.02, −5.70/−3.00 and −5.71/−3.08, respectively.^9,10,15 The origin of this difference remains unclear, given that there are only minor structural differences across these three emitters; a possible explanation for the divergence in HOMO/LUMO values could be the application of different conversion equations to derive the energy levels from experimental data.

OLEDs

We next fabricated vacuum-deposited OLEDs with Mes-tDABNA as the emitter using a device structure consisting of indium tin oxide (ITO)/1,4,5,8,9,11-hexaazatriphenylenehexacarbonitrile (HATCN, 5 nm)/1,1-bis[(di-4-tolylamino)phenyl]cyclohexane (TAPC, 30 nm)/tris(4-carbazoyl-9-ylphenyl)amine (TCTA, 10 nm)/mCP (5 nm)/emitting layer (EML) (20 nm)/1,3,5-tris(3-pyridyl-3-phenyl)benzene (TmPyPB, 40 nm)/lithium fluoride (LiF, 1 nm)/aluminum (Al, 100 nm). Here, HATCN was used as the hole injection layer, TAPC and TCTA as the hole transporting layers, mCP as an exciton blocking layer, TmPyPB as an electron transporting layer, and LiF was used to reduce the work function of the top Al electrode. The OLED device stack and the chemical structures of the organic layers are shown in Fig. 5d.

Fig. 5 Device data of Mes-tDABNA (green = conventional device) (blue = HF device) (a) EQE vs. luminance plot; (b) electroluminescence spectra; (c) J–V–L plot (d) device structure and chemical structures of used materials.

We first optimized the doping concentration in SF3-TRZ as a function of Φ_PL (Fig. S6). A doping concentration of 4 wt% provided charge balance and a sufficiently large exciton recombination zone, and the Φ_PL remained high at this concentration (vide supra) and an EQE_max of 18.4% was achieved at this doping concentration. The electroluminescence (EL) spectrum, current density–voltage-luminance (J–V–L) curves and EQE vs. luminescence curves are shown in Fig. 5a–c. The device with Mes-tDABNA emitted narrowband blue light at λ_EL of 468 nm (FWHM of 32 nm), which corresponds to CIE coordinates of (0.13, 0.15). These are identical to the reported device with tDABNA in DPEPO.⁹ Compared to the devices with t-DABNA-dtB (λ_EL = 471 nm), the λ_EL is blue-shifted; however, due to the broader envelope of the EL spectrum of the device with Mes-tDABNA the CIE_y coordinate is increased from 0.13 for the device with t-DABNA-dtB (FWHM = 140 meV) to 0.15 for the device with Mes-tDABNA (FWHM = 190 meV). The EL spectrum is narrower than the corresponding PL spectrum (Fig. 4), which is likely due to the more homogeneous film achieved by vacuum deposition compared to the spin-coated film used for the PL measurements. The device with Mes-tDABNA showed an EQE_max of 18.4% and a maximum brightness, L_max, of 3600 cd m⁻². Due in part to its relatively slow k_RISC, the device showed relatively severe efficiency roll-off at high luminance (EQE₁₀₀ and EQE₁₀₀₀ of 12.5 and 5.6%, respectively); however, the efficiency roll-off was nonetheless milder compared to devices with DABNA-1 and tDABNA, as neither were reported to reach 1000 cd m⁻².^8,9 The device with Mes-tDABNA performed comparably to one of the best devices with the DABNA derivative DABNA-NP-TB, which has a reported EQE_max/EQE₁₀₀₀ of 19.5/12.0% (Fig. 6 and Table S3).¹²

Fig. 6 EQE_max, EQE₁₀₀₀ and EQE roll-off vs. λ_EL of devices employing DABNA-based emitters in the literature; EQE roll-off refers to the efficiency roll-off from EQE_max to EQE₁₀₀₀. Light blue points represent conventional devices; orange points represent HF devices; dark blue points represent conventional devices that did not reach a L_max of 1000 cd m⁻²; brown points represent HF device that did not reach a L_max of 1000 cd m⁻²; data presented are listed in Table S2.

In a bid to improve the exciton utilization efficiency and reduce the efficiency roll-off, HF OLEDs were fabricated (Fig. 5d). DMAC-DPS was used as the TADF assistant dopant because of its high Φ_PL of 90%, relatively fast k_RISC of 2.9 × 10⁵ s⁻¹ in 20 wt% doped SF3-TRZ film, and the strong overlap between the absorption spectrum of Mes-tDABNA and the PL spectrum of DMAC-DPS in toluene (Fig. S13). An optimized doping ratio of 2 wt% Mes-tDABNA: 20 wt% DMAC-DPS: 78 wt% SF3-TRZ was identified to be used as the EML as this formulation minimized the probability for triplet excitons from the host to transfer to the Mes-tDABNA and permitted an efficient FRET from DMAC-DPS to Mes-tDABNA, thereby conserving the narrowband emission (Table 2 and Fig. S10). The photophysical properties of the hyperfluorescent film are compiled in Table S2. As shown in Fig. 5b, the HF devices showed similar narrowband emission at λ_EL of 469 nm (FWHMs of 34 nm) and CIE coordinates of (0.13, 0.17) to the conventional device. The turn-on voltages (V_on) were reduced from 3.9 to 3.1 V and the EQE_max was improved to 23.1%. More importantly, the efficiency roll-off was reduced, with EQE₁₀₀ and EQE₁₀₀₀ of 20.5 and 12.7%, respectively. Compared to the HF devices with the derivatives tDPAC-BN (EQE_max/EQE₁₀₀₀ = 21.0/5.4%) and M-tDABNA (EQE_max/EQE₁₀₀₀ = 8.6/8.1%), the OLED with Mes-tDABNA showed an enhanced performance evidenced by the milder efficiency roll-off (see additional device comparison data in Table S3 in the SI).

Table 2 Device data of Mes-tDABNA

Conventional device^a	Von/V	λEL^c (FWHM)/nm	Lmax/cd m^−2	EQEmax/100/1000/%	CIE (x, y)^c
a Using an EML consisting of 4 wt% Mes-tDABNA in SF3-TRZ.b Using an EML consisting of Mes-tDABNA:DMAC-DPS:SF3-TRZ = 2 : 20 : 78 wt%.
Mes-tDABNA (4 wt%)	3.9	468 (33)	3600	18.4/12.5/5.6	0.13, 0.15
HF device^b
Mes-tDABNA (2 wt%)	3.3	469 (34)	8400	23.1/20.5/12.7	0.13, 0.17

---

Conclusions

Here, we demonstrated an effective substitution strategy to mitigate aggregation-caused quenching in thin films while preserving narrowband blue emission in a DABNA derivative MR-TADF emitter. This approach also resulted in devices showing milder efficiency roll-off and higher EQE_max, attributed to a smaller ΔE_ST, shorter delayed lifetime, and a faster k_RISC, compared to devices with tDABNA. Mes-tDABNA emits at λ_PL at 460 nm and has a narrow emission envelope with a FWHM of 25 nm. The devices showed an EQE_max of 18.4% at blue CIE coordinates (0.13, 0.15); however though improved over devices with tDABNA there remained a significantly efficiency roll-off (EQE₁₀₀₀ of 5.6%, compared to 0.4% for the device with tDABNA). To further improve the device performance, the D–A type TADF emitter, DMAC-DPS, was employed as a sensitizer within a HF device configuration, resulting in a higher EQE_max of 23.1% and lower efficiency roll-off (EQE₁₀₀₀ of 12.7%) compared to the traditional device. In contrast, most tDABNA-based emitters in HF devices exhibit EQE₁₀₀₀ values below 10%. These results underscore the critical role of molecular design in tailoring photophysical properties to enhance device performance.

Author contributions

M. F.: conceptualization, data curation, formal analysis, investigation, methodology, visualization, writing – original draft preparation, writing – review & editing. D. C.: data curation, formal analysis, investigation, visualization, writing – original draft preparation, writing – review & editing. E. Z.-C.: conceptualization, funding acquisition, methodology, project administration, resources, supervision, writing – review & editing.

Conflicts of interest

There are no conflicts to declare.

Data availability

The research data supporting this publication can be accessed at https://doi.org/10.17630/e4521ae2-e1ea-45c2-8f60-880c6d12e8a6.

Supplementary information: ¹H NMR and ¹³C NMR spectra, HRMS, elemental analysis and reverse phase HPLC; supplementary computational data and coordinates; additional photophysical and OLED data. See DOI: https://doi.org/10.1039/d5ma00586h

Acknowledgements

This project has been partly funded by the European Union Horizon 2021 research and innovation programme under grant agreement no. 101073045 (TADF solutions) and the EPSRC grant EP/X026175/1. In the UK, this project has received funding from the Engineering and Physical Sciences Research Council for support (EP/R035164/1; EP/W007517/1; EP/Z535291/1). D. Chen thanks the support from China Postdoctoral Science Foundation (Grant no. 2022TQ0227) and Natural Science Foundation of Jiangsu Province, China (Grant no. BK20230508). We thank Prof. Xiao-Hong Zhang for providing access to the device fabrication facility at Soochow University.

References

1. H. Uoyama, K. Goushi, K. Shizu, H. Nomura and C. Adachi, Nature, 2012, 492, 234–238.
2. S. Madayanad Suresh, D. Hall, D. Beljonne, Y. Olivier and E. Zysman-Colman, Adv. Funct. Mater., 2020, 30, 1908677.
3. R. K. Konidena and K. R. Naveen, Adv. Photon. Res., 2022, 3, 2200201.
4. K. R. Naveen, P. Palanisamy, M. Y. Chae and J. H. Kwon, Chem. Commun., 2023, 59, 3685–3702.
5. X.-C. Fan, K. Wang, Y.-Z. Shi, Y.-C. Cheng, Y.-T. Lee, J. Yu, X.-K. Chen, C. Adachi and X.-H. Zhang, Nat. Photonics, 2023, 17, 280–285.
6. Y. C. Cheng, X. Tang, K. Wang, X. Xiong, X. C. Fan, S. Luo, R. Walia, Y. Xie, T. Zhang, D. Zhang, J. Yu, X. K. Chen, C. Adachi and X. H. Zhang, Nat. Commun., 2024, 15, 731.
7. H. Wang, Y. C. Cheng, X. C. Fan, D. Y. Chen, X. Xiong, X. Y. Hao, Y. Z. Shi, J. Yu, D. Huang, J. X. Chen, K. Wang and X. H. Zhang, Sci. Bull., 2024, 69, 2983–2986.
8. T. Hatakeyama, K. Shiren, K. Nakajima, S. Nomura, S. Nakatsuka, K. Kinoshita, J. Ni, Y. Ono and T. Ikuta, Adv. Mater., 2016, 28, 2777–2781.
9. S. H. Han, J. H. Jeong, J. W. Yoo and J. Y. Lee, J. Mater. Chem. C, 2019, 7, 3082–3089.
10. J. Park, K. J. Kim, J. Lim, T. Kim and J. Y. Lee, Adv. Mater., 2022, 34, e2108581.
11. J. H. Kim, W. J. Chung, J. Kim and J. Y. Lee, Mater. Today Energy, 2021, 21, 100792.
12. S. Oda, W. Kumano, T. Hama, R. Kawasumi, K. Yoshiura and T. Hatakeyama, Angew. Chem., Int. Ed., 2021, 60, 2882–2886.
13. Y. Wang, Y. Duan, R. Guo, S. Ye, K. Di, W. Zhang, S. Zhuang and L. Wang, Org. Electron., 2021, 97, 106275.
14. Y. Wang, K. Di, Y. Duan, R. Guo, L. Lian, W. Zhang and L. Wang, Chem. Eng. J., 2022, 431, 133221.
15. H. Lim, S. J. Woo, Y. H. Ha, Y. H. Kim and J. J. Kim, Adv. Mater., 2022, 34, e2100161.
16. Y. Yuan, X. Tang, X. Y. Du, Y. Hu, Y. J. Yu, Z. Q. Jiang, L. S. Liao and S. T. Lee, Adv. Opt. Mater., 2019, 7, 1801536.
17. D. Hall, S. M. Suresh, P. L. dos Santos, E. Duda, S. Bagnich, A. Pershin, P. Rajamalli, D. B. Cordes, A. M. Z. Slawin, D. Beljonne, A. Köhler, I. D. W. Samuel, Y. Olivier and E. Zysman-Colman, Adv. Opt. Mater., 2019, 8, 1901627.
18. D. Hall, J. C. Sancho-Garcia, A. Pershin, D. Beljonne, E. Zysman-Colman and Y. Olivier, J. Phys. Chem. A, 2023, 127, 4743–4757.
19. P. Ulukan, A. Monari and S. Catak, ChemPhotoChem, 2023, 7, e202300147.
20. D. Hall, J. C. Sancho-Garcia, A. Pershin, G. Ricci, D. Beljonne, E. Zysman-Colman and Y. Olivier, J. Chem. Theory Comput., 2022, 18, 4903–4918.
21. L. Hämmerling and E. Zysman-Colman, Chem. Catal., 2024, 4, 101061.
22. L. S. Cui, S. B. Ruan, F. Bencheikh, R. Nagata, L. Zhang, K. Inada, H. Nakanotani, L. S. Liao and C. Adachi, Nat. Commun., 2017, 8, 2250.
23. W. H. Melhuish, J. Phys. Chem., 1961, 65, 229.
24. Y. Tsuchiya, S. Diesing, F. Bencheikh, Y. Wada, P. L. dos Santos, H. Kaji, E. Zysman-Colman, I. D. W. Samuel and C. Adachi, J. Phys. Chem. A, 2021, 125, 8074.
25. V. V. Pavlishchuk and A. W. Addison, Inorg. Chim. Acta, 2000, 298, 97–102.

---

Footnote

† These authors contributed equally.

---