Evolution of color-tunable TADF emitters in OLEDs: from design strategies to color modulation

Abstract

Organic thermally activated delayed fluorescence (TADF) materials have gained considerable attention in recent times, specifically in advancing organic light-emitting diodes (OLEDs) due to the possibility of achieving 100% EQE and the ability to tune their emission color. This review highlights the evolution of molecular design strategies that enable precise control of the singlet–triplet energy gap (ΔE_ST) and reverse intersystem crossing (RISC), thereby advancing device efficiency and color purity. We discuss the progression from early donor–acceptor systems to more advanced methodologies, including π-conjugation tuning and steric and substituent engineering, through-space charge transfer (TSCT), multi-resonance (MR) TADF, and circularly polarized luminescence (CPL)-active TADF emitters. In the later sections, we summarize the progression from the early conceptual development to the recent emergence of blue, green, red, and white TADF OLEDs and their operation mechanism. Thus, the objective of this review is to connect molecular design strategies with the development of next-generation TADF materials for high-performance, color-tunable OLEDs.

---

Tapashi Sarmah

Tapashi Sarmah received her integrated Master of Science Degree in Chemical Science from Tezpur University in 2020. She is pursuing her doctoral research at the Department of Chemistry, Indian Institute of Technology Guwahati, India, under the supervision of Prof. Parameswar K. Iyer. Her research mainly focuses on the design and development of aggregation-induced emission (AIE), thermally activated delayed fluorescence (TADF) and room-temperature phosphorescence (RTP) based purely organic luminescent materials and their applications in bioimaging, photodynamic therapy and optoelectronic devices.

Chakali Srinivas

Chakali Srinivas received his Master of Science in Chemistry from University College of Science Saifabad in 2020. He is currently pursuing his PhD at the Department of Chemistry, Indian Institute of Technology Guwahati, under the supervision of Prof. Parameswar K. Iyer. His research interests focus on the design and development of AIE luminogens with advanced photophysical properties, including RTP and TADF. He is actively engaged in tuning these properties for potential applications in organic light-emitting diodes (OLEDs), photodynamic therapy (PDT), and various security-related technologies.

Debika Barman

Debika Barman completed her BSc in Chemistry from St. Mary's College, Shillong in 2017 and her MSc in Chemistry from the National Institute of Technology (NIT) Meghalaya in 2019. She is currently pursuing her PhD at the Department of Chemistry, Indian Institute of Technology Guwahati, under the supervision of Prof. Parameswar K. Iyer. Her research focuses on the design and development of self-assembled and co-assembled organic luminogens with AIE characteristics. She works on tuning advanced photophysical properties such as RTP, TADF, mechanoluminescence and exploring their applications in photodynamic therapy (PDT), anti-counterfeiting, and surface-enhanced Raman scattering (SERS).

Rajdikshit Gogoi

Rajdikshit Gogoi completed his BSc in Chemistry from North Lakhimpur College in 2016 and MSc in Chemical Science from Tezpur University in 2018. Currently he is pursuing his doctoral research at the Department of Chemistry, Indian Institute of Technology Guwahati under the supervision of Prof. Parameswar Krishnan Iyer. His research mainly focuses on the development of purely organic luminescent materials with advanced properties such as AIE, RTP, TADF, and mechanoluminescence.

Retwik Parui

Retwik Parui completed his BSc in Chemistry from Midnapore College under Vidyasagar University, West Bengal, in 2014 and MSc from Indian Institute of Technology (Indian School of Mines), Dhanbad, in 2016. He received his PhD in Chemistry from the Indian Institute of Technology Guwahati in 2025 under the supervision of Prof. Parameswar Krishnan Iyer. His area of research focuses on the design and development of heteroatom embedded organic luminogens and modulating their condensed state properties for diverse applications.

Kavita Narang

Kavita Narang Kavita Narang completed her BSc in Chemistry from Govt. Nagarjuna P. G. College of Science, Raipur, in 2015 and her MSc in Chemistry from the Department of Chemistry, Banaras Hindu University, Varanasi, in 2017. She received her PhD in Chemistry from the Indian Institute of Technology Guwahati in 2024 under the supervision of Prof. Parameswar K. Iyer. She is currently working as an Assistant Professor at Govt. Sukhram Nage College, Nagri, affiliated with Pt. Ravishankar Shukla University, Raipur. Her current research focuses on low molecular weight gelators, peptide self-assembly, and supramolecular materials.

Himangshu Baishya

Himangshu Baishya received his Master of Science in Physics from Gauhati University in 2019. He is currently pursuing his PhD at the Centre for Nanotechnology, Indian Institute of Technology Guwahati, under the supervision of Prof. Parameswar K. Iyer. His research focuses on developing and optimizing organic–inorganic perovskite-based optoelectronic devices, including perovskite light-emitting diodes and solar cells. His work emphasizes enhancing the device efficiency and operational stability through interfacial and bulk engineering strategies employing small organic molecules and functional polymers.

Parameswar Krishan Iyer

Parameswar Krishnan Iyer received his PhD in Chemistry from the Central Salt and Marine Chemicals Research Institute, Bhavnagar, India, in 1999. He was a Post-Doctoral Researcher with Technion, Haifa, Israel, from 1999 to 2001; University of California Santa Barbara, CA, USA, from 2001 to 2003; and Case Western Reserve University, Cleveland, OH, USA, from 2003 to 2004. In 2004, he joined the Department of Chemistry, IIT Guwahati, India, as an Assistant Professor and has been promoted to a full-time Professor in 2013. His current research interests include the design and development of advanced functional organic small molecules, polymers, and hybrid materials for optoelectronic devices, sensors, imaging and therapeutics.

---

1. Introduction

Since the pioneering report on organic light-emitting diode (OLED) devices in 1987, they have revolutionized lighting and flat-display technology owing to their flexibility, high energy efficiency, high contrast and multicolor emission.¹ An OLED device is a sandwich-like structure comprising an anode, organic function layers and a metal cathode.² According to the number of organic function layers, OLED devices can be classified into single-layer, double-layer, three-layer, and multilayer structures.³ Among them, the three-layer device structure is the most commonly used in OLEDs and is mainly composed of an electron transport layer (ETL), a hole transport layer (HTL), and an organic emitting layer (EML), which is not only conducive to selecting materials as functional layers but also to optimize device performance (Fig. 1).⁴

Fig. 1 (A) Architecture of an OLED with multiple different layers and (B) transport mechanism of holes and electrons through a working OLED device.

The performance of OLEDs is fundamentally dependent on the nature and effectiveness of the emissive layer, which regulates the brightness and spectral characteristics of the emitted light. Emissive materials must ideally possess a combination of high photoluminescence quantum yield (ϕ_PL), good charge-transport compatibility, excellent color purity, and long operational stability.⁵ Under the action of external voltage, holes and electrons are released from the anode and cathode, respectively, pass through the respective transport layers, and finally combine to form excitons in the light-emitting layer.⁶ These excitons contain both singlet excitons (S₁) and triplet excitons (T₁) due to their different spin states before and after excitation, and the ratio of S₁ to T₁ is 1 : 3 according to the spin quantum theory.⁷ Emission originates from the radiative transition of an exciton, that is, from the singlet or triplet state to the ground state. During this process, photons are generated and emitted from the EML, resulting in different efficiencies in OLED devices. Traditional fluorescent organic emitters (1st generation) can utilize only singlet excitons for light emission, and thus, their internal (IQE) and external (EQE) quantum efficiencies are limited to 25% and 5%, respectively.⁸ To overcome this, metal-based inorganic phosphorescent complexes have been developed, and owing to the strong spin–orbit oupling (SOC) feature of these materials, they can effectively harness triplet excitons for light emission. Thus, phosphorescent emitters (2nd generation) demonstrated near unity IQE and excellent EQE of >20%.⁹ However, the usage of expensive noble-metals, their toxicity and high cost jeopardize their practical applicability.⁹ Therefore, researchers have focused on the development of pure organic emitters, which can exploit “dark” triplet excitons for light emission. Furthermore, efficient pure blue and deep blue phosphorescent materials are still in urgent demand for widespread application.

In response to this need, thermally activated delayed fluorescence (TADF) is the most feasible exciton harvesting mechanism employed in third-generation OLED devices. A comparative mechanism of the exciton pathways of the three generations of OLEDs is shown in Fig. 2. Since the first reported OLED based on an organic TADF emitter in 2011,¹⁰ tremendous attention in recent years has been devoted to improving the performance of OLEDs. Similar to phosphorescent organometallic emitters, purely organic TADF emitters can encompass both singlet and triplet excitions for light emission, hence attaining 100% IQE. One important advantage of TADF emitters is that they can be purely organic, therefore eliminating the issues associated with the use of heavy metal-based organometallic complexes.¹¹ TADF relies on a small singlet–triplet energy gap, ΔE_ST, defined as the gap between the lowest energy triplet state (T₁) and the lowest energy singlet state (S₁). When ΔE_ST is sufficiently small, taken usually as <0.1 eV, thermal upconversion from the triplet state to the singlet state by reverse intersystem crossing (RISC) becomes possible. TADF emitters typically show two types of PL, i.e., prompt fluorescence, in which the history of the singlet exciton does not involve communication with the triplet manifold, and delayed fluorescence, which is the result of an initial ISC to the triplet state, followed by repopulation of the singlet state via RISC.¹² ΔE_ST governs the rate of RISC, k_RISC, according to the Boltzmann distribution, as follows:¹³

where k_B denotes Boltzmann's constant and T is the temperature.

Fig. 2 Exciton pathways of 1st (A), 2nd (B) and 3rd (C) generation OLEDs, where PF, P, TADF, ISC, RISC and ΔE_ST denote prompt fluorescence, phosphorescence, thermally activated delayed fluorescence, intersystem crossing, reverse intersystem crossing and singlet–triplet energy gap, respectively.

In principle, ΔE_ST is directly proportional to the exchange energy of the two unpaired electrons in the excited states as follows:¹⁴

ΔE_ST = E_S − E_T = 2J

where the exchange integral J depends on the electron density overlap between the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO), assuming that the S₁ and T₁ states are dominated by the HOMO to LUMO transitions, as follows:¹⁵
ϕ_HOMO and ϕ_LUMO are the spatial distributions of the HOMO and the LUMO, respectively, and r₁ and r₂ are position vectors. Hence, a small ΔE_ST can be realized by spatial wave function separation of HOMO and LUMO via molecular engineering.

The quantum efficiency is the ratio of the photons to injected charges, which includes IQE (η_int) and EQE (η_ext).⁸ Only a fraction of photons created within the EML can escape to the air; hence, the EQE has more reference value than the IQE for practical application. The commonly accepted definition for the EQE is the ratio of the number of photons that are extracted to air per injected charges, which can be calculated as follows:¹⁶

η_ext = η_intη_out = (γη_rϕ_PL)η_out

where η_out is the light outcoupling efficiency (commonly between 20% and 30%), accounting for the portion of generated photons that escape the device, overcoming losses due to waveguiding and absorption within the OLED structure. γ is the Langevin recombination factor of electrons and holes, which is taken to be unity in efficient OLEDs but can be substantially less when charge recombination is not confined to the emissive layer. ϕ_PL is the PLQY of the emissive layer or emissive dopant contained therein, which is the ratio of photons released to photons absorbed and quantifies the efficiency of light produced upon photoexcitation. β is the fraction of electrically produced excitons that can decay radiatively, which is 1 for singlets but only 0.25 in conventional fluorescent emitters due to the non-emissive nature of triplet excitons, unless advanced emitters such as TADF and phosphorescent materials are used to harness triplet states. The combination of these three components (γβϕ_PL) is termed IQE and represents the ratio of photons generated within the OLED compared to charges injected.

The color purity of emitted light is described using the Commission Internationale de l’Éclairage (CIE) coordinate system based on the (x, y, z) values corresponding to human visual response to red, green, and blue stimuli, respectively.⁵ Typically, the z coordinate is omitted (given that x + y + z = 1), and color is represented in a two-dimensional (x, y) space.⁸ In display technology, OLEDs strive to attain saturated color values consistent with standardized gamuts such as sRGB and Rec. 2020. For example, sRGB assigns red, green, and blue primaries as (0.64, 0.33), (0.30, 0.60), and (0.15, 0.06), respectively.¹⁷ Rec. 2020, a more precise standard, updates these coordinates to (0.71, 0.29), (0.17, 0.80), and (0.13, 0.05), respectively, needing more deeply saturated colors.¹⁸ Reaching these corner coordinates is particularly challenging for blue emitters. The spectral width of emission, commonly expressed as FWHM (full width at half maximum), is essential for achieving high color purity.⁹ Narrowband emitters are highly effective at making bright, saturated colors, whereas broad spectra yield more white-like and less saturated emissions.

In the case of pure white OLED lighting, the common CIE coordinates are (0.33, 0.33), while that for warm white are (0.45, 0.41).¹⁹ Depending on the intended use, broader emission profiles are preferred for mimicking natural daylight or incandescent light. The measured CIE coordinates and the potential of a material for use in displays or lighting depend on the position of the emission peak and the width of the spectrum of the emitter. The increasing demand for precise color control has driven extensive research into color-tunable emissive materials. In this context, TADF emitters offer significant promise due to their structural tunability and capacity to span the visible spectrum through rational molecular design. Color-tunability in TADF materials is achieved by modifying their D–A combinations, tuning their π-conjugation length, introducing steric hindrance or positional isomerism to control the spatial HOMO–LUMO separation, and designing through-space charge transfer (TSCT) or MR structures, especially for narrowband deep-blue emission.^5,20,21 More recently, chiral molecular designs have enabled circularly polarized luminescence (CPL)-TADF systems, further expanding their functionality.²

This review provides an extensive overview of the design strategies and evolution of color-tunable TADF emitters. We have grouped these materials based on their molecular structures, photophysical mechanisms, and the performance of EL devices, focusing on their systematic development towards high-performance RGB and white OLEDs. Particular attention is paid to the role of molecular design in improving their color purity, efficiency, and stability. To date, although numerous reviews have addressed the TADF mechanisms or device integration, few have offered a design-strategy-driven and color-specific perspective. Thus, the objective of this review is to address these gaps by mapping how molecular-level modifications have enabled systematic color tuning across the visible spectrum and directing future advancements in efficient OLED material development.

2. Historical progression of TADF emitter design strategies for color regulation

2.1 Donor–acceptor engineering

The fundamental approach for designing a TADF molecule is by D–A engineering, wherein an electron-rich donor is bonded to an electron-deficient acceptor moiety. In this system, charge transfer occurs from D to A via a conjugated molecular backbone called through bond charge transfer (TBCT).²² This strategy was first demonstrated in 2011.¹⁰ Quantum mechanical analysis showed that this D–A framework strategically separates HOMO and LUMO on D and A, respectively. This spatial separation reduces the ΔE_ST, a crucial parameter that facilitates RISC, achieving high internal quantum efficiencies without the need for heavy metal complexes. A critical factor in this system is the dihedral angle between D and A, where the nearly orthogonal geometry (∼85–90°) achieved by steric hindrance or twisted structure can reduce the orbital overlap and effectively decouple the frontier orbitals, thus suppressing ΔE_ST, while maintaining sufficient charge transfer character in the excited state to enable radiative emission. However, substantial decoupling can weaken the oscillator strength, and hence decrease the emission intensity. Therefore, maintaining a balance between reduced ΔE_ST and ensuring sufficient emission efficiency became the primary challenge in the early development of TADF. Fig. 3 denotes a schematic of the design strategies for D–A, D–A–D, A–D–A frameworks with examples.

Fig. 3 Molecular design of TADF based D–A (A), D–A–D (B), and A–D–A (C) frameworks and their respective examples (a)–(c).^23–25

Other than molecular orientation, the strength of the donors and acceptors governs both the energy levels and the CT character of the excited state, directly impacting the emission color.²⁶ Hence, modulation of these two parameters can lead to tuning of the emission color of TADF emitters along with their quantum efficiency. Stronger donors and acceptors can elevate the HOMO and lower the LOMO levels effectively, leading to a smaller energy gap, and further longer wavelength emissions, respectively. Alternatively, a moderate and weaker D–A combination gives lower wavelength emission by increasing the energy gap. In the early reports on TADF emitters, carbazole was most commonly used as a moderating donor coupled with acceptors such as dicyanobenzene, triazine and benzophenone, and therefore their emissions were limited to shorter wavelengths.²⁷ The pioneering example 4CzIPN consists of four carbazole donors and one dicyanobenzene acceptor, resulting in green emission (510 nm) with a high PLQY of 94% and small ΔE_ST of 0.1 eV (Fig. 4a).¹² Similarly, a triazine-based emitter, DMAC-TRZ, where a sterically hindered donor group, DMAC (9,9-dimethylacridine), was incorporated with a triazine acceptor, exhibits efficient green emission with high PLQY and rapid RISC (Fig. 4b).²⁸ By replacing DMAC with a stronger and more flexible donor, phenoxazine (PXZ), PXZ-TRZ (Fig. 4c) demonstrated yellowish-green emission (∼545 nm) by elevating the HOMO level and stabilizing the CT excited state.²⁹ Again, replacing the triazine acceptor with stronger acceptors such as pyrazinophenanthroline, dicyanopyrazine, imide and benzothiadiazole further lowers the LUMO levels, resulting in red-shifted emissions.³⁰ Based on the dicyanopyrazine acceptor moiety, a series of dicyanopyrazine (DCPP)-based TADF emitters, Cz-DCPP, DPA–DCPP, and DMAC-DCPP,³¹ was developed by varying the donor strength from carbazole (Cz) to diphenylamine (DPA) and dimethylacridine (DMAC) (Fig. 4d–f), respectively. The stronger electron-donating ability and steric bulk of DMAC resulted in a smaller ΔE_ST (0.04 eV) and a red-shifted emission (618 nm) compared to Cz-DCPP (560 nm) and DPA–DCPP (606 nm). DMAC-DCPP also exhibited the highest TADF performance with ϕ_TADF of 25% and k_TADF = 13.7 × 10⁴ s⁻¹, outperforming Cz-DCPP (19%, 0.47 × 10⁴ s⁻¹) and DPA–DCPP (5%, 0.1 × 10⁴ s⁻¹).

Fig. 4 Discussed D–A-type TADF molecules (a)–(f) with different donors and acceptors.

In addition to the basic D–A design technique, symmetric frameworks such as D–A–D and A–D–A architectures have been developed to further modulate the properties of the excited state, emission color, and the device performance. In the D–A–D framework, a central acceptor core is symmetrically connected to two donor units on either side. In contrast, the A–D–A framework consists of a central donor linked to two acceptors on both sides.³² This symmetric arrangement can further separate the HOMO and LUMO effectively, lower the ΔE_ST value, and allow the fine-tuning of the emission characteristics by choosing the appropriate donors and acceptors. The combination of two donors and two acceptors in the D–A–D and A–D–A configurations, respectively, may boost the oscillator strength and enhance the radiative decay rate, therefore improving the device efficiency. Though both architectures can be used to achieve a range of colors from blue to red through proper molecular design, D–A–D systems are generally preferred for deep-blue emitters, while A–D–A architectures are commonly used for orange and red TADF emitters. In D–A–D, carefully choosing weak to moderate electron-donating donors and moderate acceptors helps maintain a large energy gap, which is necessary for blue emission. Four D–A–D-type blue-emissive molecules, I, II, III, and IV, were developed based on the 2,7-bis(9,9-dimethylacridin-10-yl)-9,9-dimethylthioxanthene-S,S-dioxide (DDMA-TXO₂) framework (Fig. 5a).³³ The effects of regioisomerism and methyl substitution on their TADF properties were systematically investigated. The 2,7-substituted regioisomers (I, III, and IV) exhibited more twisted D–A geometries, leading to efficient charge separation, smaller singlet–triplet energy gaps (ΔE_ST ≈ 0.05–0.06 eV), and faster reverse intersystem crossing (RISC). In contrast, the 3,6-regioisomer (II) displayed a less twisted geometry, resulting in a larger ΔE_ST (∼0.13 eV) and poor TADF performance. Methyl substitution on the acceptor (III) enhanced the steric hindrance, further increasing the D–A dihedral angle, which reduced ΔE_ST and led to a blue-shifted emission (461 nm) with the best device performance. Conversely, donor methylation (IV) increased the electron-donating strength, causing a red-shifted emission (538 nm) without improving the TADF efficiency. These findings highlight that both the substitution pattern and the steric/electronic nature of the substituents critically influence the D–A twisting, ΔE_ST, emission wavelength, and ultimately the TADF efficiency of organic emitters.

Fig. 5 Discussed structures of D–A–D (a) and A–D–A (b) type TADF emitters. Adapted with permission.^25,33 Copyright 2019, the American Chemical Society. Copyright 2023, Elsevier.

Alternatively, in the A–D–A system, the dual acceptor units stabilize the LUMO, narrowing the energy gap and shifting emission to longer wavelengths. For example, quinoxaline- and dicyanobenzene-based A–D–A emitters have demonstrated efficient orange-red and red TADF emission with high PLQY and EQEs. Three examples of these A–D–A-type TADF emitters include QxPz, DMQxPz, and DPQxPz,²⁵ utilizing quinoxaline-based acceptors and a common 5,10-disubstituted phenazine donor (Fig. 5b). These newly designed materials exhibit a symmetric A–D–A configuration and rigid, ladder rod-like molecular structures, promoting strong charge-transfer character and minimizing ΔE_ST, with values of 0.003, 0.004, and 0.003 eV, respectively. This is a result of the extremely small overlap between the HOMO and LUMO due to the larger twist angle between the D and A. As a result, the materials exhibited a high PLQY of up to 85% and efficient RISC processes. Among them, DMQxPz demonstrated an outstanding RISC rate constant of 1.23 × 10⁶ s⁻¹ and PLQY of 85% in a 3 wt%-doped CBP film. When incorporated into doped OLED devices, these TADF materials delivered high external quantum efficiencies (EQE) approaching 20%, with moderate EQE roll-off in the orange to red emission range, highlighting the beneficial effect of molecular rigidity and steric hindrance in the A–D–A framework. This work presents a viable approach for rationalizing excited states using structural engineering, paving the way for innovative molecular designs for future developments in OLED displays.

These foundational principles have made D–A engineering a useful and versatile method for designing TADF materials, allowing fine control of their excited-state dynamics and emission colors. This basic strategy has facilitated the development of more advanced structures, such as TSCT systems and extended conjugated frameworks, which are discussed in the next sections.

2.2 Conjugation tuning by π spacer for emission alterations

Apart from the D–A, D–A–D and A–D–A frameworks, the inclusion of π-conjugated spacer groups between the D and A has become a potential approach to alter the photophysical properties of TADF emitters, especially their emission wavelength, ΔE_ST, and oscillator strength. The earlier design strategy of direct linkage between D and A often resulted in limited control over the extent of CT and emission color. The geometry and nature of the spacer significantly impact the electronic coupling between D and A as well as the spatial overlap between the frontier molecular orbitals, thereby affecting the singlet–triplet energy gap ΔE_ST, the CT nature of the excited state, and the RISC efficiency. Planar conjugated spacers such as phenylene and thiophene promote extended π-delocalization, which enhances the charge transfer character, and typically induces a red shift in emission of ∼20–50 nm. This enhanced conjugation can help tune emissions toward green or orange wavelengths but can also increase ΔE_ST slightly due to the greater HOMO–LUMO orbital overlap, potentially lowering the RISC efficiency if not correctly balanced.³⁴ For example, in the previously discussed molecules, Cz-DCPP, DPA–DCPP and DMAC-DCPP, one phenyl ring was linked between the D and A to study the π-bridge effect on the photophysical and electroluminescent (EL) properties.³¹ The introduction of a phenyl π-bridge between the D and A in Cz-Ph-DCPP, DPA-Ph-DCPP, and DMAC-Ph-DCPP increases the D–A separation and twist angle, reducing the orbital overlap and lowering the ΔE_ST values, most significantly in the DMAC-based molecules (from 0.003 to 0.001 eV), followed by the Cz (0.232 to 0.110 eV) and DPA (0.434 to 0.198 eV) systems (Fig. 6a–c), respectively. This promotes stronger charge-transfer character and enhances the reverse intersystem crossing (RISC), improving the TADF efficiency. The phenyl bridge also causes larger Stokes shifts due to the excited-state geometric changes. Its effect on the fluorescence rate (k_F) and TADF rate (k_TADF) is donor-dependent, where in the Cz and DPA compounds, it decreases k_F due to the increased rotational relaxation, while in DMAC-based molecules, it enhances k_F, reduces the nonradiative decay, and boosts ϕ_TADF (60–75%) and k_TADF (up to 2.0 × 10⁵ s⁻¹). EL measurements show apparent emission shifts, where the Cz-based OLEDs emit yellow light at ∼560–564 nm with CIE coordinates of (0.44, 0.54) and (0.46, 0.52), the DPA-based OLEDs emission shifted from red at 616 nm (0.61, 0.38) to deep-red at 644 nm (0.64, 0.36), and the DMAC-based OLEDs emission shifted from red at 624 nm (0.60, 0.40) to orange at 596 nm (0.53, 0.46). Overall, the phenyl π-bridge effectively tunes the emission color, reduces ΔE_ST for efficient TADF, and modulates the excited-state dynamics, with DMAC-Ph-DCPP achieving the most significant improvements in TADF performance and OLED device efficiency.

Fig. 6 Examples of π group tuning D–A TADF molecules (a)–(d). Adapted with permission.³³ Copyright 2023, John Wiley and Sons.

In contrast, twisted or sterically hindered spacers such as ortho-methylated phenyl rings disrupt the π-conjugation by imposing dihedral angles between D and A, which spatially separate the HOMO and LUMO.³³ This separation effectively reduces ΔE_ST, enhancing RISC and enabling efficient TADF even in high-energy (blue) emissive systems. The torsional conformation imposed by the π-spacer also affects the rigidity and reorganization energy of the molecules, which in turn influence the PLQY and stability of the device. Rigid and twisted spacers often reduce the non-radiative decay pathways, resulting in a higher quantum yield, while making the emission spectra sharper. Two examples include DCN-Ph-DPA and DCN-SP-DPA (Fig. 6d), where a phenyl and a rigid spiro group are attached as spacers between the donor, diphenylamine, and the acceptor, dibenzo[a,c]phenazine-11,12-dicarbonitrile, respectively.³⁵ The rigid spiro group reduces the conformational flexibility and suppresses the nonradiative channels. Compared to DCN-Ph-DPA, the photophysical properties of DCN-SP-DPA are considerably enhanced by the addition of a spiro-bridge. Nonradiative decay pathways are effectively reduced by the rigid bulky spiro structure, which inhibits intramolecular vibrations and conformational relaxation. As a result, the spiro-bridged emitter DCN-SP-DPA achieved a PLQY of 100%, which is much higher than that of 89% observed for the phenyl-bridged DCN-Ph-DPA, along with a significantly lower nonradiative decay rate (k_nr). Despite the structural change, both molecules retain similar HOMO–LUMO distributions, preserving efficient charge transfer. These features collectively improve the exciton utilization and delayed fluorescence performance of the spiro-bridged emitter.

The length of the π-spacer also plays a crucial role, where short spacers comprised of one or two aromatic units offer optimal control, whereas overly extended bridges can excessively stabilize the charge transfer state, red shift the emission beyond the desired range, and decrease radiative efficiency. Thus, designing π-spacers needs careful balancing between connection and distance to allow accurate tuning of the spectrum, while keeping the necessary structure and electrical properties for effective thermally induced delayed fluorescence.

2.3 Substituents, steric and positional isomerism driven color tuning

Over the past few years, molecular engineering through variation in substituents, altering their position, and introduction of bulky functionality has had a significant impact on the development of novel TADF materials and modulation of their luminescence wavelength.^36,37 These design strategies not only determine the molecular conformation and packing orientation, but also play a pivotal role in regulating the electronic distribution and orbital overlap coefficient. The strategic introduction of steric hindrance in the backbone helps to achieve a completely separated HOMO–LUMO electronic distribution in a D–A-incorporated system by suppressing the through-bond conjugation. In a directly connected D–A framework, the regulation of the degree of steric repulsion between these two functionalities can produce different twist angles and ensure a significant shift in the wavelength maximum of a TADF material.³⁸ Importantly, fine-tuning of the bulkiness, conjugation, and torsion angle presents a promising route to produce a low ΔE_ST with minimal HOMO–LUMO overlap, which enables adequate radiative emission involving an efficient RISC process (Fig. 7(i)).^34,39 A series of exclusively emissive carbazolyl dicyanobenzene-based TADF materials was proposed, where steric hindrance between the carbazole core caused angular distortion (Fig. 7a).¹² This distortion led to a small ΔE_ST along with a well-separated HOMO–LUMO, while the modulation in donating affinity resulted in in emission tuning from blue to orange color, with PLQY varying from 26 ± 1% to 74 ± 3%. However, the direct D–A linkage fails to produce a steric-mediated torsional angle for some specific acceptors, such as triazine, a conventional acceptor unit for the development of TADF materials.^40,41 This strong acceptor core has the ability to construct hydrogen bonds with the donor moiety, which restrict the molecular twist and promote the possibility of overlapping between the HOMO–LUMO electron clouds. Despite this tendency, it became possible to control the limited overlapping by introducing extended conjugation with the donor unit and regulating a well-balanced interplay between the D–A strength (CC2TA).⁴² Later on, the use of a π-spacer enabled a large torsional angle of 48° to be attained in a system containing diphenylaminocarbazole as the donor and triphenyltriazine as the acceptor (DACT-II) (Fig. 7b).⁴³ The bulky donor and π-spacer-driven twisted geometry exhibited a remarkably low ΔE_ST value close to zero, which facilitates achieving 100% PLQY (in a doped film).

Fig. 7 (i) Schematic for designing TADF emitters based on steric hindrance and positional linkage. (a)–(f) Chemical structures of TADF emitters regulating steric hindrance and position substitution.

Similar to spacer-assisted dihedral angle modulation, varying the substituent positions such as ortho, meta, and para is also another eminent strategy that could trigger both HOMO–LUMO separation and the RISC rate to produce effective TADF properties. Here, a triphenyl-substituted triazine core was also found to be a potential acceptor owing to its controllable substitution site. Utilizing this acceptor unit, three TADF emitters were established, where the donor benzofurocarbazole group was anchored at the three different positions of the phenyl ring, connected with a diphenyltriazine moiety (Fig. 7c).⁴⁴ In the ortho isomer (oBFCzTrz), the closer proximity of the D and A units presented a distorted geometry owing to the large dihedral angle, whereas the meta (mBFCzTrz) and para isomers (pBFCzTrz) exhibited a comparatively shorter angle due to their less steric repulsive force. Consequently, the possibility of better charge transfer in the ortho isomer was completely evidenced in its lowest ΔE_ST of 0.002 eV and highest PLQY of 97.9% compared to the other two isomers. Similar to triphenyl-substituted triazine, diphenylsulfone was also found to be an ideal acceptor to evaluate the positional isomeric effect.

The incorporation of a donor core at different positions of both the phenyl rings had a profound influence on their TADF property. To investigate this, 9,9-dimethyl-9,10-dihydroacridine donor was connected to both sides of the phenyl ring, varying its position relative to the sulfone unit (Fig. 7d).⁴⁵ Depending on the degree of charge transfer at different attachment positions, the ortho (o-ACSO2) and meta isomers (m-ACSO2) exhibited a red-shifted emission with respect to the para isomer (DMAC-DPS)⁴⁶ and both displayed low ΔE_ST values of 0.04 and 0.07 eV, respectively, reflecting a promising characteristic for TADF. Apart from these acceptors, a few boron-containing polycyclic acceptors were designed, where the placement of the donor moiety at different positions had a notable impact on their photophysical properties. Employing this strategy, three positional isomers were developed, where the dimethylacridine donor moiety was anchored at the para (p-AC-DBNA) and meta positions (m-AC-DBNA, and m′-AC-DBNA) of the wing phenyl ring of a boron-embedded rigid core (Fig. 7e).⁴⁷ The donating ability and the steric effect produced by the orthogonally linked dimethylacridin unit caused a planar distortion in the boron-containing rigid part. The comparably more planar conformation of the para isomer enabled excellent charge transfer phenomena and resulted in very small ΔE_ST with PLQY exceeding 94%.

In addition to the variation in the donor location, the adjustment in the acceptor position exerts a huge influence on the TADF properties. To investigate the steric effect of the dimethylacridine unit, three D–A based positional isomers were reported, where the cyano functionality of the benzonitrile acceptor was varied at the ortho, meta, and para positions (Fig. 7f).⁴⁸ Despite the analogous steric environment, a distinct difference in RISC rate was observed, which was attributed to the difference in the electron-withdrawing affinity of the cyano group arising from resonance or inductive effect. However, the extra intramolecular dipole interaction caused a restriction in D–A bond rotation, leading to an enhanced overall performance. Thus, the above discussion highlighted that the modulation of the D/A position could be a critical factor to regulate the HOMO–LUMO electron distribution, control the RISC rate, and optimize the TADF emission.

2.4 Through-space charge transfer TADF systems for color modulation

Given that charge transfer plays a pivotal role in developing TADF emitters, in addition to TBCT, intermolecular and intramolecular TSCT holds great promise to design a diverse range of TADF materials.^49,50 Usually, the principal criterion to construct a TADF emitter is to achieve a system where the HOMO–LUMO can be well-separated in their excited state. Thus, naturally separating donors and acceptors with through-space electronic communication is found to be a great tactic for the rational design of TADF emitters. In contrast to TBCT materials, TSCT molecules exhibit weaker electronic coupling between the D and A units, depending on the D–A distance and angle, which leads to an increase in the singlet and triplet energy levels, resulting in blue emission.⁵¹ In intermolecular TSCT, naturally, charge transfer takes place between two separate donors and acceptors when they form an exciplex or electroplex under photoexcitation in their blended or condensed form (Fig. 8(i)).^52–54

Fig. 8 (i) and (ii) Diagram illustrating the design of TADF emitters utilizing intermolecular TSCT. (a)–(c) Chemical structures of TADF emitters regulating intermolecular TSCT.

To conceptualize the exciplex-based TADF property, a separate carbazole-containing donor (CN-CZ2) and triazine-based acceptor (PO-T2T) were mixed in a 1 : 1 weight ratio (Fig. 8a).⁵⁵ It was observed that the blended film emits a significant red-shifted emission compared to the isolated component. This red-shifted emission was attributed to the charge transfer from the HOMO of the donor to the LUMO of the acceptor, exhibiting a delayed lifetime of 3.39 μs and PLQY of 55%. Further modification of the D/A affinity revealed an excellent way to tune the emission color of the exciplexes. A diphenylsulfone-containing donor (DPSTPA), which produced different exciplex emissions in combination with three different acceptors, 2CzPN (540 nm), 4CzIPN (596 nm), and CzDBA (592 nm), depending on the energy gap between the HOMO of the donor and LUMO of the acceptor (Fig. 8b).⁵⁶ The resultant exciplexes displayed delayed fluorescence lifetimes of 13.9, 6.8, and 10.1 μs with PLQY of 79%, 10% and 59%, respectively. The relatively low PLQY for 4CZIPN was further investigated and attributed to the weak exciplex formation between these two cores. In contrast to conventional intermolecular TSCT between two closely placed separate D and A moieties, a novel design strategy was introduced where intermolecular TSCT took place between the D and A of adjacent identical molecules (Fig. 8(ii)).⁵⁷ To restrict the intramolecular TBCT process, a conjugation-forbidden ether linkage was introduced between the D and A cores (DMAC-o-TRZ), which enhanced the possibility of intermolecular TSCT phenomena (Fig. 8c). This TSCT process ensured the ΔE_ST of 0.018 eV and exhibited TADF emission at 487 nm.

Along with intermolecular TSCT, intramolecular TSCT is another potential formula to avoid the probability of extended conjugation-mediated nonradiative decay in TBCT. Intramolecular TSCT materials are typically designed using the closely stacked D–A, D–A–D and A–D–A approach, where the distance, angle, and D/A affinity are regulated to govern the TADF efficiency and emission color tunability (Fig. 9(i)). A series of intramolecular TSCT-based TADF emitters was projected, positioning different donors close to the diphenyltriazine acceptor on a 9,9-dimethylxanthene scaffold (XPT, XCT, and XtBuCT) (Fig. 9a).⁵⁸ The co-facial adjacent D–A arrangements assist the TSCT process and restrict molecular motion in the condensed state, resulting in aggregation-induced delayed fluorescence (AIDF) property. Moreover, a variation in the donor strength led to a red shift in the emission wavelength from 419 nm to 560 nm, along with a delayed lifetime up to 3 μs. In another study, four orange-to-red TADF emitters were developed by anchoring a quasi-planar O-bridged triphenylamine and planar dibenzo[a,c]phenazine acceptor arranged in a D–A (DPXZ-QX, DPXZ-DFQX) and A–D–A (DPXZ-2QX, DPXZ-2DFQX) sandwich-like architecture to achieve close spatial alignment (Fig. 9b).⁵⁹ The considerable amount of π–π interaction between the D–A moiety facilitates the participation of strong TSCE, which was supported by SCXRD analysis, DFT calculations, and polarity-dependent shift in luminescence wavelength. Notably, the spatial design strategy of connecting multiple acceptors allowed efficient low-energy TSCT and demonstrated an elevated PLQY up to 91% with a delayed lifetime down to 4.9 μs (in doped film). Later, the CT efficiency was improved via modification of the donor conformation. An additional oxygen atom in the donor core (TPXZ-QX and TPXZ-2QX) induced planarity and electron-donating strength (Fig. 9c).⁶⁰ This improvement in the donor conformation not only promoted π–π interactions but also boosted the TSCT performance, reflecting a deep-red emission with considerable delayed lifetime. In summary, despite the suboptimal delayed lifetime, this unique architecture offers a promising approach for achieving new TADF materials with diverse functionalities.

Fig. 9 (i) Diagram illustrating the design of TADF emitters utilizing modulation of D/A architecture to enhance intramolecular TSCT. (a)–(c) Chemical structures of TADF emitters regulating intramolecular TSCT.

2.5 MR TADF emitters for modifications of color purity

A breakthrough in TADF emitter design was recently achieved by introducing the concept of MR-TADF.⁶¹ This innovative approach involves engineering planar fused aromatic compounds that incorporate boron and either oxygen or nitrogen atoms. These MR-TADF materials represent a distinct class of rigid, polycyclic systems in which electron-donating and electron-withdrawing atoms are positioned in a para relationship within a flat, conjugated framework. The strategic placement of these atoms allows resonance interactions, which result in an offset between the HOMO and LUMO distributions by one atomic position.⁶²

The incorporation of heavy atoms can further enhance SOC, thereby promoting RISC.⁶³ The modularity of the D–A approach offers considerable design flexibility, enabling the development of emitters that cover the entire visible spectrum. However, these D–A-type TADF emitters, which rely on intramolecular charge transfer (ICT), often display broad emission bands. This is largely due to the strong vibronic coupling between the ground state (S₀) and the first excited singlet state (S₁), as well as notable structural relaxation in the S₁ state.⁶⁴

In contrast, MR-TADF emitters exhibit significantly narrower emission bandwidths (FWHM),⁵ making them highly suitable for high-definition display applications (Fig. 10a and b). This spectral sharpness arises from localized, short-range charge transfer, which effectively minimizes the geometric distortion during excitation and emission. MR-TADF molecules are structurally distinct, consisting of rigid, fused polycyclic aromatic frameworks where electron-donating and electron-withdrawing atoms are placed in alternating positions often in para-arrangements within the conjugated system. The alternating placement of elements such as boron (B), oxygen (O), nitrogen (N), and sulfur (S), or carbonyl-containing groups contributes to complementary resonance stabilization across the framework (Fig. 10c).⁶⁵ Boron occupies a unique position in the design of MR-TADF emitters because of its electron-deficient character and ability to stabilize distinct electronic distributions within rigid π-conjugated frameworks. As a trivalent element with only six valence electrons, boron contains a vacant p_z orbital, making it an intrinsic electron acceptor within π-systems. This property allows boron atoms to serve as electron-withdrawing centres when embedded into aromatic frameworks, producing localized regions of electron deficiency that are essential for constructing the MR effect.⁶⁶

Fig. 10 (a) Schematic representation of energy levels. (b) Comparison of the EL spectra of MR-TADF and ICT-TADF materials. (c) Several types of MR cores are currently available, and HOMO–LUMO separation by the MR resonance effect. Adapted with permission.⁶⁵ Copyright 2023, the Royal Society of Chemistry.

Within these materials, electron-rich areas are typically centred on the donor atoms and their adjacent carbon atoms, whereas electron-deficient zones are localized on the acceptor atoms and their neighboring carbons. In both the singlet and triplet excited states, the electron density distribution follows a repeating electron-rich and electron-poor pattern across the aromatic core. Interestingly, a similar alternating distribution persists in the ground state, though the roles of the electron-rich and poor regions are reversed. This spatially controlled and minimal reorganization of electron density during the excited-state transitions contributes to the high radiative efficiency and sharp spectral profiles characteristic of MR-TADF emitters.

Thus far, MR-TADF-based OLEDs have successfully demonstrated high EQE along with narrowband emission across a broad spectral window spanning from 450 to 690 nm. Despite this progress, achieving efficient and pure red MR-TADF emitters remains a significant challenge. This limitation primarily stems from the rigid constraints imposed by the current molecular design frameworks, which hinder effective, red-shifted emission, while maintaining the desired narrow spectral bandwidth.⁶⁵

Fig. 11 Schematic of the structural features, excited-state decay characteristics and emission spectra of (a) traditional D–A-based and (b) MR-based TADF materials. Adapted with permission.⁶⁷ Copyright 2024, the American Chemical Society.

The efficient spatial overlap and minimal excited-state distortion result in high oscillator strength during excitation characteristic of locally excited (LE) states. Consequently, MR-TADF emitters achieve both intense delayed fluorescence and high ϕ_PL.⁶⁸ This design uniquely combines the benefits of localized and delocalized electronic transitions within a compact, resonance-stabilized framework.

2.5.1 Core mechanism of narrow-band MR-TADF. The unique mechanism of MR-TADF originates from its ability to combine efficient exciton harvesting with narrowband emission, making it fundamentally distinct from conventional D–A-type TADF. The fundamental challenge in high-purity emitters lies in suppressing vibrational coupling and structural relaxation between the excited singlet state and the ground state. According to the International Telecommunication Union BT.2020 standard, ultra-high- definition displays demand RGB emitters with exceptionally narrow FWHM values, which can only be realized if molecular reorganization upon excitation is minimized.⁶⁹ In general, according to photophysics, the emission spectral bandwidth is strongly governed by vibronic coupling between S₁ and S₀. When the geometry of S₁ and S₀ remains similar (structural displacement, ΔQ ≈ 0), the reorganization energy (ΔE) is minimal, and emission arises primarily from the v = 0 → 0 transition, resulting in sharp and narrowband emission (Fig. 12a). As the structural displacement between the excited and ground states increases, ΔQ becomes larger, which induces stronger vibronic contributions from transitions such as v = 0 → 1, 0 → 2, and higher order, broadening the emission spectrum (Fig. 12b and c). This relationship is quantitatively expressed by ΔE = ½ω²ΔQ², where ω is the vibrational frequency. Thus, minimizing ΔE is critical for designing high-color-purity emitters, and MR-TADF materials achieve this through their rigid, resonance-stabilized frameworks.⁷⁰

Fig. 12 Vibronic coupling between the S₁ and S₀ states with small (a), medium (b), and large (c) structural changes within the two states. Adapted with permission.⁷¹ Copyright 2023, the Royal Society of Chemistry.

The key structural feature of MR-TADF molecules is their alternating arrangement of electron-deficient (e.g., boron and carbonyl) and electron-rich (e.g., nitrogen, oxygen, sulfur, and selenium) atoms in a rigid polycyclic aromatic skeleton (Fig. 10c). This architecture generates complementary resonance effects, which localize the HOMO and LUMO on different atomic sites, while maintaining significant overlap. Compared to conventional D–A TADF molecules, such as DMAC–ND, where their HOMO and LUMO are spatially segregated across donor and acceptor units, MR-TADF molecules such as DABNA-1 display alternating electron-rich and electron-deficient regions distributed over their entire skeleton (Fig. 13). In the S₁ state, the electron density is delocalized across the π-framework with alternating character, whereas in S₀, the polarity of these regions is reversed. Therefore, the S₁ → S₀ transition only involves a short-range redistribution of charge density rather than large-scale reorganization, thereby minimizing the vibronic coupling and structural distortion.⁷¹

Fig. 13 HOMO–LUMO distributions for D–A-type TADF (DMAC-ND) and MR-TADF (DABNA-1) molecules. Adapted with permission.⁷¹ Copyright 2023, the Royal Society of Chemistry.

This alternating resonance effect enables MR-TADF molecules to simultaneously satisfy two critical conditions for high-efficiency emission, as follows: (i) small singlet–triplet energy splitting (ΔE_ST), which promotes efficient RISC for exciton utilization, and (ii) large oscillator strength, ensuring strong radiative transitions with high ϕ_PL. Importantly, the rigid MR framework suppresses nonradiative decay pathways by limiting molecular vibrations and bond rotations, which are often detrimental in flexible D–A systems. As a result, MR-TADF emitters achieve both efficient exciton harvesting via TADF and remarkably narrowband emission spectra.

In essence, the mechanism of MR-TADF can be summarized as a synergistic interplay among molecular rigidity, complementary resonance effects, and minimized reorganization energy. By confining electronic transitions to short-range charge redistribution, MR-TADF molecules avoid broad vibronic progressions and deliver sharp, stable emission bands. This photophysical behavior establishes MR-TADF as a breakthrough concept for high-purity RGB OLED emitters aligned with the BT.2020 standard requirements.

One of the critical performance criteria for OLED emitters is the width of their emission spectra, often measured by FWHM. As shown in Fig. 11,⁶⁷ conventional TADF materials, particularly those relying on D–A architectures, typically exhibit broad FWHMs ranging from 70 to over 100 nm, especially in the long-wavelength (red) region. This spectral broadness undermines their color purity and limits their suitability for high-resolution display technologies.⁷²

Thus, to address this, commercial OLEDs commonly use optical filters to suppress the undesired tails in their emission spectrum, thereby improving their color saturation. However, this comes at the cost of reduced external quantum efficiency (EQE), diminished brightness, and compromised image clarity. In contrast, MR-TADF materials intrinsically produce sharp, narrowband emissions, eliminating the need for this filtering. This property makes them highly desirable for display applications that demand precision in color reproduction.

The ideal RGB emitters for OLED displays are expected to emit at peak wavelengths of 467 nm (blue), 532 nm (green), and 630 nm (red), with the corresponding CIE coordinates of (0.131, 0.046), (0.170, 0.797), and (0.708, 0.292), respectively.⁷³ A key metric in evaluating color purity is the FWHM of the emission spectrum. To meet industrial standards for high color fidelity, materials should exhibit FWHM values below 0.14 eV.⁷⁴

It is important to note that FWHM can be reported in either wavelength (nm) or energy (eV) but comparing values across these units can be misleading due to the inverse relationship between energy and wavelength. A fixed FWHM in energy terms translates to broader FWHM in wavelength as the emission shifts toward the red region. Therefore, for consistency and practical relevance, FWHM is typically discussed in wavelength units in most OLED research.⁷⁵ As illustrated in Fig. 14, significant progress has been made recently in the development of highly efficient MR-TADF emitters across the primary RGB color spectrum, demonstrating strong potential for next-generation OLED applications. In the case of blue emission, a notable advancement was introduced via a design strategy combining both short-range and long-range charge transfer mechanisms within an MR framework.⁷⁶ By extending the conjugation through C–C bond formation between a carbazole-based donor and a boron-centered acceptor, two efficient emitters, DBACzPh and DBADCzPh, were developed. These materials exhibit pronounced TADF characteristics in both the solution and film states. Particularly, DBADCzPh shows nearly perfect horizontal molecular alignment in a polar host matrix, promoting balanced bipolar charge transport and effective energy transfer from the host to dopant. This molecular orientation significantly enhances the light outcoupling efficiency, resulting in sky-blue EL with an impressive EQE exceeding 40%.

Fig. 14 Development of high-efficiency RGB MR-TADF emitters (a)–(c) exhibiting impressive EQEs and narrowband emission, enabled by tailored resonance-based molecular design strategies. Adapted with permission.^76–78 Copyright 2025, Nature. Copyright 2023, John Wiley and Sons. Copyright 2022, John Wiley and Sons.

In the green emission category, a nitrogen-embedding molecular engineering (NEME) approach was employed to fine-tune the electronic structures of materials. By introducing nitrogen atoms at varying positions within a triphenylene-based core, a family of MR-TADF emitters BN-TP-Nx (x = 1–4) was created.⁷⁷ These structural variations allow precise modulation of the electronic distribution, enabling fine control over the emission profile. Among them, BN-TP-N3 stands out, delivering ultra-pure green emission centered at 524 nm, a narrow FWHM of 33 nm, and CIE coordinates of (0.23, 0.71). This emitter achieves a peak EQE of 37.3%, meeting stringent industrial standards for display-grade materials.

In the case of red emission, where design challenges are more pronounced due to the structural rigidity in traditional B/N-based MR scaffolds, a novel solution has been reported through the construction of modular MR cores. The BNO1–BNO3 emitter was developed by incorporating symmetric resonance units, N–π–N, O–π–O, and B–π–B, in a benzene core.⁷⁸ This approach allows straightforward synthesis and flexible emission tuning via peripheral modifications. BNO1 exhibits remarkable photophysical properties including near-unity PLQY, a high radiative decay rate (up to 7.4 × 10⁷ s⁻¹), and sharp emission with a FWHM of approximately 32 nm. Integrated into a phosphor-sensitized OLED structure, BNO1 delivers a state-of-the-art device performance with an EQE surpassing 36%, minimal efficiency roll-off (25.1% at 50 000 cd m⁻²), brightness exceeding 130 000 cd m⁻², and excellent operational durability. Together, these RGB MR-TADF systems presented in Fig. 14 underscore the effectiveness of tailored resonance design strategies in achieving narrowband emission and high efficiency, paving the way for high-performance OLED displays with superior color purity and brightness.

2.5.2 Recent breakthroughs in MR-TADF.
2.5.2.1 Spectral purity in ultra-narrowband deep-blue MR-TADF emitters. The initial and most striking breakthrough in this field was the demonstration that carefully designed MR cores deliver intrinsically ultra-narrow emission, while maintaining large radiative rates. The seminal DABNA family introduced by Hatakeyama and co-workers showed that embedding tricoordinate boron and complementary donor heteroatoms into a rigid fused arene produces a short-range alternating electron distribution that minimizes the reorganization energy and suppresses vibronic sidebands, giving very sharp 0–0 dominated emission peaks suitable for display standards. Subsequent π-extension and systematic peripheral engineering (for example, v-DABNA/V-DABNA variants) preserved the MR character, while pushing the device external quantum efficiencies into the high-performance regime for deep-blue OLEDs, establishing MR-TADF as the leading route to high-purity blue emitters.⁶¹
2.5.2.2 Extending MR-TADF across the visible region (green → red) without loss of color purity. A pressing challenge after the blue success was shifting the MR emission to longer wavelengths without reintroducing large structural reorganization or long-range CT character (which broaden spectra). Several controlled molecular strategies have proven effective including introducing additional boron atoms in para-arrangements to strengthen the acceptor character, embedding heteroatoms (S and Se) to tune conjugation, and structural locking (e.g., spiro-junctions) to retain planarity, while enabling redder emission. These approaches have produced narrowband green and orange MR emitters, and in targeted cases, pushed toward red with acceptable PLQYs, showing that the MR motif can be generalized beyond blue when the atom placement and rigidity are judiciously chosen.⁷⁹
2.5.2.3 Device stability, degradation understanding, and lifetime improvements. Early high-efficiency MR-TADF devices sometimes suffered from limited operational lifetimes. Recent work has both diagnosed degradation pathways and proposed molecular and device remedies. Mechanistic studies have identified that radical cationic instability (oxidative degradation) and fragile bonding motifs around boron are common failure modes, and mapping these pathways has guided design changes such as steric shielding of boron centres, use of more robust heterocycles, and host/stack engineering to reduce oxidative stress in the emissive layer. Implementing these changes in conjunction with recombination-zone optimization has yielded devices with markedly improved lifetimes, while retaining narrow emission.⁸⁰
2.5.2.4 Accelerating RISC and reducing efficiency roll-off at high brightness. Although MR cores produce rigid localization, they can result in limited spin–orbit and spin-vibronic coupling that slows RISC and lengthens the delayed lifetimes, which exacerbates triplet accumulation and roll-off at high currents. Breakthrough strategies to accelerate triplet harvesting include introducing controlled heavy-atom character (S/Se substitution), peripheral groups that enhance specific vibrational modes coupled to spin mixing, and device-level host designs that reduce reorganization losses. These combined molecular and device tactics have improved the RISC rates and reduced the EQE roll-off in high-brightness operation, enabling MR-TADF emitters to perform better under practical drive conditions.²²
2.5.2.5 Solution processability and dendritic architectures for scalable manufacture. To move beyond vacuum-deposited small-area devices, researchers have designed dendronized and solubilized MR cores that keep the rigid emitting unit intact, while permitting spin-coating or printing. By attaching bulky, well-designed solubilizing groups (dendrons and branched alkyl chains) at non-perturbing sites, teams have produced solution-processed films exhibiting narrowband deep-blue emission and respectable device efficiencies. This line of work represents an important step toward scalable, lower-cost fabrication.⁶⁶
2.5.2.6 Chiroptical MR-TADF and CP-EL via intrinsic chirality and CP-hyperfluorescence. A very recent and influential breakthrough is the realization of circularly polarized electroluminescence (CP-EL) from MR systems. Two complementary strategies have emerged, as follows: (i) embedding chirality directly into MR scaffolds (helicene-type, planar-chiral motifs) to create intrinsically CPL-active MR emitters and (ii) CP-hyperfluorescence in which a chiral TADF sensitizer transfers energy and effective handedness to an achiral MR terminal emitter (e.g., v-DABNA), enabling narrowband CP-EL, while leveraging the superior spectral purity of achiral MR cores. Notably, recent hyperfluorescence reports demonstrated efficient CP-EL with good EL efficiency and measurable g_EL values, showing that CP-active MR technologies are now viable for advanced display and photonic applications.⁸¹

2.6 Designing of CPL-TADF

Circularly polarized luminescence (CPL) defines the emission of light from chiral luminescent materials with unequal intensities of left- and right-handed circularly polarized components.⁸² Conventional methods to generate CP light using polarizers and wave plates result in at least 50% energy loss, making chiral emitters highly desirable for applications such as 3D displays. Employing chiral luminophores instead of non-polarized emitters in OLEDs enables the devices to generate circularly polarized electroluminescence (CPEL) directly.⁸³ Chiral materials are promising for future optoelectronic technologies given that they can improve the efficiency by minimizing energy loss without external filters.

Circularly polarized OLEDs (CP-OLEDs) offer direct CP light emission without external filters, improving the efficiency and contrast. The performance of CPL materials is evaluated by PLQY and the luminescence dissymmetry factor (g_lum), defined as: g_lum = 2(I_L − I_R)/(I_L + I_R), where I_L and I_R are the left- and right-handed emission intensities, respectively.⁸⁴ Replacing unpolarized emitters with chiral luminophores in OLEDs enables high-performance chiral emitters for next-generation optoelectronics, potentially eliminating energy loss and the need for anti-glare filters. Fig. 15 depicts the advantages of conventional OLEDs over CP-OLEDs. The ground-state absorption and excited-state emission properties of chiral luminous materials are generally described using ECD (electronic circular dichroism) and CPL, respectively. Integrating TADF into CPL-active materials gives an effective approach to boost the EL efficiency, while preserving circular polarization. These systems offer both high-efficiency light emission and chiral optical characteristics, which are crucial for the development of CP-OLEDs.

Fig. 15 Illustration of advantages of using CP-OLEDs over normal OLEDs [redrawn from ref. 83].

In addition to enhancing their efficiency, chirality in TADF systems can also influence their color. Chirality in D–A or MR-TADF systems can alter the HOMO–LUMO distribution, modulate ΔE_ST and influence the emission bandwidths, thereby affecting the colour purity and CIE coordinates. Rigid chiral units, such as spiro and planar scaffolds, may hinder spectral broadening by limiting the conformational relaxation. This feature is necessary for narrowband emission (FWHM < 30 nm) with clear chromaticity.⁸¹ Alternatively, D–A systems that are flexible exhibit broad spectra (FWHM > 70 nm), which make colors less pure.⁸⁵ Thus, chirality in TADF molecules not only induces chiroptical functionality but also advances the color-tuning strategies in OLEDs.

Designing CP-TADF materials requires dual consideration of both efficient TADF characteristics and the induction of strong chiroptical activity. To date, two main strategies have been developed to construct CP-TADF materials, where the first involves introducing inherent chirality into the TADF chromophore (Fig. 16, left), while the second relies on the chiral perturbation of achiral TADF systems (Fig. 16, right). There are contrasting design trade-offs associated with each method, where intrinsic chirality can produce higher g_lum values because stereogenic centres directly participate in the emission process, but it frequently has limited scalability and synthetic complexity.⁸⁶ In contrast, chiral perturbation strategies generally allow easier synthesis, modular design, and emission color tunability.⁸⁷ In both cases, the type of chirality along with the donor–acceptor strength directly influences the efficiency, emission color, and spectral width of the resulting TADF systems. These two approaches together outline the basic design framework for CP-TADF materials.

Fig. 16 Design strategies for CP-TADF emitters [redrawn from ref. 84].

2.6.1 Design of inherent chirality in TADF chromophores. Inherent chirality can be achieved by integrating chiral elements (point, axial, planar and helical) into the D–A TADF core, and thus the chiral centre directly participates in the emission process such as ΔE_ST tuning and radiative decay pathways.⁸³ This approach gives higher g_lum values, typically in the range of 10⁻³–10⁻² but it requires various complex methods such as HPLC and diastereoisomeric separation for the resolution of enantiomers, which limit scalability. The most common examples of this type of CP-TADF emitters include spirofused architectures and binaphthyl-bridged D–A systems. However, these types of CP-TADF devices often suffer from broad emission, which limits the color purity in displays.
2.6.1.1 Point chirality. In point chirality, a chiral carbon center is incorporated into the TADF molecule, either in the donor, acceptor, or spacer unit. This class represents the earliest reported category of CP-TADF emitters. The incorporated chiral center governs the chiroptical properties of the molecule and significantly influence its molecular orientation, conformation and extent of charge transfer. Consequently, these structural effects modulate the key photophysical parameters such as ΔE_ST, emission wavelength, PLQY, and g_lum values. The pioneering example of a point chirality-based CP-TADF molecule, DPHN (Fig. 17a), was reported in 2015.⁸⁸ It is composed of a triarylamine donor and a naphthacene acceptor linked via a chiral carbon, which breaks the conjugation between D and A. Theoretical analysis revealed that the molecular orbitals of DPHN in the S₁ state are well separated, with the HOMO and LUMO localized on D and A, respectively. This large spatial separated molecular orbital leads to a small ΔE_ST of 0.19 eV and a broad green TADF emission at 513 nm with PLQYs of 4% in solution and 26% in a host film. The chiral centre between D and A also induces CPL activity with a moderate g_lum value of 1.1 × 10⁻³.

Fig. 17 Examples of intrinsically chiral CP-TADF compounds involving point (a), axial (b), planar (c) and helical (d) chirality. Adapted with permission.^88,90–92 Copyright 2015, the Royal Society of Chemistry. Copyright 2019, the American Chemical Society. Copyright 2018, the American Chemical Society. Copyright 2021, the Royal Society of Chemistry.

2.6.1.2 Axial chirality. Axial chirality in CP-TADF can be achieved from either a chiral C–C axis or spiro-axis. Chirality from the C–C axis relies on steric hindrance, and hence racemization can occur during sublimation, imposing restrictions in design flexibility.⁸⁹ To prevent this, these molecules use bulky groups next to the C–C axis to maintain chirality, low molecular weight and twisted conformation to reduce the sublimation temperature. Conversely, spiro-axis chirality offers greater configurational stability, and the presence of a spiro-carbon can prevent racemization even at elevated temperatures. The first axial chiral CP-TADF emitter, R/S-1 (Fig. 17b), was designed in 2019, featuring a BINOL core with chromone units fused at the [2,3] and [2′,3′] positions, each linked to a phenoxazine donor.⁹⁰ Because of its stable chiral configuration and high chiral induction, axially chiral 1,1′-binaphthol (BINOL) is one of the most important chiral molecules. Along with R/S-1, another molecule, R/S-2, was also reported, in which the rigid acceptor xanthenone is replaced by a rotatable benzophenone unit; however, it fails to exhibit CPL properties. This is due to the rigid conjugated framework between the chiral binaphthyl and the electron acceptor xanthenone, which facilitates direct chirality transfer from the chiral binaphthyl to the TADF chromophore in R/S-1, leading to CPL emission. Alternatively, the rotatable benzophenone structure in R/S-2 can reduce the chirality transfer process, which is why its CPL response is inactive. The BINOL framework enforces a cross-like arrangement between the two acceptor arms, leading to a nearly orthogonal alignment between the D and A groups. This orientation leads to a clear separation of HOMO and LUMO, resulting in a very small ΔE_ST of 0.059 eV with rapid RISC process. R/S-1 exhibits yellow emission at 568 nm with a PLQY of up to 19%. Compared with R/S-2, R/S-1 shows a 38 nm red shift and a higher PLQY, which can be attributed to its more rigid conjugated skeleton. However, only R/S-1, with its fixed conjugated structure, can emit circularly polarized luminescence signals, with g_lum values reaching 1.6 × 10⁻³ in toluene solution and 9.2 × 10⁻⁴ in neat film.
2.6.1.3 Planar chirality. Planar chirality occurs when steric hindrance prevents the rotation of planar structures, resulting in molecules that are mirror symmetric. [2.2]Paracyclophane has emerged as the predominant planar chiral scaffold in CP-TADF emitters, owing to its unique structure. However, its synthesis is challenging and it is prone to racemization, limiting its broader application. Despite this, planar chirality is vital for achieving strong CPL signals and high g factors. The first chiral [2.2]paracyclophane-based CP-TADF molecules were reported in 2019, i.e., g-BNMe₂-Cp and m-BNMe₂-Cp, where one benzene ring of [2.2]paracyclophane carries the donor unit, dimethylamine, while the other contains the acceptor, dimesitylboron.⁹¹ In these molecules, g and m denote the pseudo-gem and pseudo-meta positions of the donor unit relative to the acceptor unit, respectively. In the pseudo-gem form, the nitrogen centre is more coplanar with the amino-bonded benzene ring, which brought a big structural difference between the two molecules. The close spatial arrangement of the central benzene rings in these molecules enables TSCT, ensuring strong HOMO–LUMO separation with small ΔE_ST values of 0.17 and 0.12 eV, respectively, which are favorable for TADF. With its more favourable orientations and stronger TSCT nature, g-BNMe₂-Cp exhibited a 10 nm red-shifted strong yellowish green emission at 531 nm and higher PLQY of 46% compared to m-BNMe₂-Cp. Due to the superior optical properties of the g-BNMe₂-Cp (Fig. 17c) isomer, its chiroptical properties were solely investigated, which showed a larger g_lum value of 4.2 × 10⁻³ than the previously reported ones. Overall, the superior photophysical and CPL properties of g-BNMe₂-Cp arise from its synergistic combination of planar chirality and highly efficient through-space D–A electronic coupling, both of which are optimal in the pseudo-geminal arrangement compared to the pseudo-meta substitution pattern of m-BNMe₂-Cp.
2.6.1.4 Helical chirality. Helical chirality is the last family of CP-TADF molecules based on intrinsically chiral molecules. In contrast to point, axial and planar chirality, helical chirality in TADF emitters is less common due to its design complexity and strong intermolecular interactions, which reduce the performance of OLEDs. Despite their potential for high CPL/CPEL, these materials are difficult to synthesize, require chiral resolution, and may racemize during device fabrication. Their application is further restricted by their low g factors and limited helix frameworks. To maintain configurational stability, substantial intramolecular overlap is essential. The first helical chirality-based CP-TADF molecule was reported in 2021, featuring a rigid hetero-helicene scaffold as a pair of blue-emitting enantiomers, P/M-QAO–PhCz (Fig. 17d).⁹² These molecules employ quinolino[3,2,1-de]acridine-5,9-dione (QAO) as the acceptor, a carbonyl/nitrogen (C=O/N) system as the MR-TADF framework, and 9-phenyl-9H-carbazole (PhCz) as the donor at the ortho-position of the central nitrogen. This arrangement forms a helical structure, which serves as the origin of molecular chirality. QAO has a twisted form with a dihedral angle of 47.23°, according to the single-crystal geometries. The dihedral angle increased to 51.21° with the addition of functional groups at the ortho position of the central nitrogen. The addition of functional groups at the para position of the central nitrogen had no effect. The helical twist of QAO–PhCz is indeed caused by the increase in steric hindrance between the QAO and PhCz groups and the increased overlap between the terminal aryl rings. In QAO–PhCz, the overlap of the benzene rings between the QAO and PhCz units induces weak intramolecular π–π interactions, while the QAO group simultaneously acts as a bridge and acceptor to enforce a quasi-face-to-face configuration with PhCz; this short-range arrangement promotes TSCT with well-separated orbitals, resulting in a very small ΔE_ST of 0.11 eV. The room temperature fluorescence exhibits a high PLQY of 46.6%, a narrow FWHM of 29 nm, and a sharp blue emission peaking at 461 nm. The synergistic interaction of the rigid emission core of QAO–PhCz and its sterically hindered structure account for the narrow FWHM. The blue emission peak of QAO–PhCz differs from that of conventional D–π–A-type molecules, which exhibits broadening emission and a wide FWHM when functioning as a donor or acceptor. This discrepancy may be due to the antagonistic effect between the distorted structure with disrupted conjugation and the π-stacked configuration influenced by the TSCT effect. It was also noticed that the racemic QAO could not be separated by chiral HPLC at room temperature because the enantiomers were not stable enough. Conversely, the racemic QAO–PhCz might be separated by chiral HPLC at room temperature because its PhCz unit makes its helical overlap larger and enantiomers more stable. Also, the enantiomers of QAO–PhCz show CPL properties with |g_lum| factors of up to 1.1 × 10⁻³ in toluene solution.

These four intrinsic chirality strategies show that the type and rigidity of the chiral element are crucial for balancing the CPL intensity, emission color, and other photophysical properties relevant to the device. Increasing the molecular rigidity, as in axial, planar, and helical systems, usually makes chiral transfer more efficient and allows stronger CPL signals with higher quantum yields or narrower emission. However, it may additionally make things more challenging because of spectral broadening or synthetic complexity. To design a CP-TADF emitter in an effective manner, one needs to carefully choose and place chiral motifs to achieve the best color purity and dissymmetry factor. This will lead to the next generation of high-performance CP-OLEDs.

2.6.2 Design of CP-TADF emitters by chiral perturbation. This involves the attachment of a chiral auxiliary unit to an achiral TADF system without modification of the core.⁹³ Binaphthol and bicarbazole are the two most commonly used chiral units in these systems, which are covalently tethered to the D–A TADF core via rigid linkers. These chiral units influence the frontier molecular orbitals, i.e. HOMO and LUMO, involved in the emission of the chromophore. These emitters are synthesized using commercially available enantiopure products, which facilitates their scalability. Therefore, this approach is currently gaining significant attention owing to its simplified synthesis, tunable color and high quantum efficiencies. The first of these chiral TADF emitters were reported in 2018 in a pair of green emissive enantiomers, S,S-/R and R-CAI-Cz (Fig. 18a) by incorporating the chiral 1,2-diaminocyclohexane into aromatic imide-based TADF structures.⁹⁴ These molecules exhibited excellent TADF properties, including a small ΔE_ST of 0.06 eV and high PLQY of up to 98%. The CP-OLEDs fabricated with these emitters achieved a high EQE of 19.7% and 19.8%, along with g_EL values of 1.7 × 10⁻³ and 2.3 × 10⁻³. This pioneering work established a foundation for efficient CPEL materials and device development.

Fig. 18 Examples of discussed CP-TADF emitters by chiral perturbation. Adapted with permission.^86,94 Copyright 2018, Wiley. Copyright 2023, the Royal Society of Chemistry.

However, the structural relaxation in D–A systems results in a wide energy distribution in the excited state, which can broaden their EL spectra, hence reducing color purity. Therefore, currently, chiral perturbation in MR-TADF systems is more versatile and promising, which gives narrowband emission (FWHM < 30 nm) via meta-substitution or chiral electron-donating groups with large g_lum and rapid RISC.⁹⁵ Although meta-substitution in D–A systems increases the orbital overlap, resulting in a higher g_lum, the inclusion of an electron-donating group can enhance the TADF efficiency without compromising CPL. Thus far, most of the examples reported are based on TICT (twisted intramolecular charge transfer) emitters, and only a few are based on red/orange CP-TADF materials. To optimize the chiral perturbation efficiency, a deeper understanding of factors such as the position and orientation of the chiral unit and the nature of the D and A groups is crucial. An example of chiral perturbation in MR-TADF systems involves introducing binaphthol-derived chiral units (BAM) into boron- and nitrogen-containing MR-TADF cores (BN) via meta-substitution (Fig. 18b).⁸⁶ A centroid distance of <5 Å between the BAM unit and the B/N core maximizes the chiral perturbation in the frontier molecular orbitals. The meta-linkage preserves the MR characteristics, while allowing chiral unit participation in the orbitals, enhancing g_lum. The electron-donating strength of BAM boosts the TADF efficiency without broadening the emission (FWHM < 30 nm). With optimized BAM units, g_lum values >1.5 × 10⁻³ and rapid RISC with SOC values >0.1 cm⁻¹ are achieved.

These two design strategies advance CP-OLEDs for ultrahigh-definition microdisplays, with chiral perturbation providing superior synthetic versatility and scalability, while intrinsic designs yield greater dissymmetry. Overall, chiral perturbation is favored for high-performance, color-pure CP-TADF emitters, while intrinsic chirality is essential for boosting chiroptical effects.

3. Journey of color-tunable TADF emitters in OLEDs

3.1 Progression from early to recent blue TADF-based emitters for OLEDs

TADF emitters commonly suffer a major drawback of broad emission spectra, which significantly undermines their ability to deliver the saturated color purity required for achieving a wide color gamut in display applications.⁹⁶ This issue is especially pronounced in the case of blue emitters, where near-saturated emission is crucial for enhancing the overall display performance. However, realizing highly efficient pure blue TADF emitters remains a major challenge, given that most candidates fall short of the expected color purity. Various intrinsic and extrinsic factors account for the limited development and suboptimal performance of blue TADF materials and their integration into OLED devices. Several factors contribute to the limited availability of highly efficient blue TADF emitters and their corresponding OLED devices, as follows:

(a) Low PLQY: blue emission demands a large bandgap (>3 eV), which inherently lowers the PLQY.⁹⁷

(b) High driving voltage requirements: the wide bandgap necessary for blue emission impedes efficient charge injection from adjacent layers, resulting in elevated operating voltages for OLEDs, which in turn compromise their energy efficiency.⁹⁸

(c) Limited operational stability: blue TADF emitters often suffer from rapid degradation during operation, leading to a sharp decline in the device efficiency and noticeable shifts in the color output, which limits their practical longevity.⁹⁹

(d) Inadequate host materials: a critical bottleneck is the scarcity of host materials capable of offering both balanced charge transport and high triplet energy levels (above 3 eV).¹⁰⁰ These characteristics are essential to ensure effective exciton confinement, prevent reverse energy transfer, and maintain a stable, efficient device performance.

Considering these challenges, the development of effective design strategies for blue TADF emitters is essential. These approaches offer significant potential to enhance the PLQY, maintain high color purity, and address the persistent issue of long-term device stability.

In a groundbreaking advancement in OLED technology, the first purely organic blue TADF emitters were reported in 2012.⁶⁴ The strategy was centered on an electron D–A molecular design, where diphenyl sulfone acted as the electron-accepting unit, while diphenylamine and carbazole derivatives served as electron donors. This approach led to the development of three novel blue-emitting compounds, identified as molecules 1, 2, and 3 (Fig. 19). Density functional theory (DFT) provided valuable insights into the electronic behavior of these molecules, where the HOMO levels were predominantly localized on the donor units, while the LUMO levels were concentrated on the sulfone acceptor. This distinct spatial separation of orbitals resulted in a reduced ΔE_ST, a key parameter for enabling RISC, which triggers the enhanced population in the emissive singlet state. The ΔE_ST values for molecules 1, 2, and 3 were determined to be 0.54 eV, 0.45 eV, and 0.32 eV, respectively, highlighting the impact of donor group modifications on the exciton dynamics. Enhancing the donor strength by introducing tert-butyl groups to the diphenylamine moiety led to a lower CT energy and further reduced ΔE_ST. Replacing the donor unit with a tert-butyl-substituted carbazole offered additional benefits, increasing the energy of the singlet CT state and enhancing the triplet ππ* level, which further minimized the ΔE_ST. Among them, molecule 3 demonstrated the most impressive TADF performance, which facilitated an efficient upconversion process following the ³ππ* → ³CT → ¹CT transition sequence. In a non-doped film, molecule 3 exhibited a long-delayed fluorescence lifetime (τ_d) of 270 μs and PLQY of 0.69, emitting at 404 nm under oxygen-free conditions. When embedded in a DPEPO host, its PLQY increased to 0.80 owing to the rigid matrix provided by the host material, which effectively suppressed the non-radiative decay. The OLEDs fabricated using these TADF emitters within DPEPO hosts showcased excellent performance metrics. The device incorporating molecule 3 achieved a peak EQE of 9.9%, with CIE coordinates of (0.15, 0.07) (Fig. 19c and d). Notably, this EQE surpasses the typical ∼5% limit associated with conventional fluorescence-based emitters. However, although these devices achieve high EQE, they suffer from a pronounced efficiency roll-off, which can be attributed to the unbalanced charge capture of the wide band gap emitters in the EML. The carefully optimized molecular design preserved the necessary π-conjugation to meet the stringent energy demands for blue emission, illustrating the vast potential of metal-free TADF materials for next-generation blue OLEDs particularly in applications where traditional phosphorescent materials fall short. A detailed summary of the photophysical properties and device performance is presented in Table 1.

Fig. 19 (a) Molecular structures highlighting D and A units; (b) transient PL decay profiles of molecules 1–3 in toluene, with the inset showing emission decay of molecule 3 in various solvents; (c) PL and EL spectra, with the inset displaying the corresponding CIE coordinates; and (d) EQE versus current density plot. Adapted with permission.⁶⁴ Copyright 2012, the American Chemical Society.

Table 1 Photophysical and computational parameters of blue TADF emitters and related OLED device configurations with EL device performance

TADF emitter name	HOMO energy (eV)	LUMO energy (eV)	ΔEST Exp. (eV)	ΔEST Theo. (eV)	λ max, abs (nm)	λ max, PL (nm)	τ P (ns)	τ d (μs)	ϕ [%]	OLED structure	MaxEQE (%)	CIE	Ref.
a Solution and non-doped film. b Doped film.
3	−5.81	−2.52	0.32	0.34	300–400	404^a/423^b	5.3^a/7.6^b	270^a/540^b	69^a/80^b	ITO/α-NPD/TCTA/CzSi/EML-3/DPEPO/TPBI/LiF/Al	9.9%	(0.15, 0.07)	64
2a	—	—	0.09	0.073	362	457^a/480^b	6.7^a	—	100^b	(ITO)/m-CBP/6 wt% 2a:DPEPO/DPEPO/TPBi/LiF/Al	20.6% ± 1.8%	—	34
Ac-OSO	−5.7	−2.6	0.06	0.10	283, 356	451^a/458^b	29^b	2.7^b	48^a/98^b	ITO/TAPC/mCBP/6 wt%-Ac-OSO:DPEPO/DPEPO/B3PyPB/LiF/Al	20.5%	—	101
3b	−5.75	−2.08	—	0.42	351	450^a/399^b	—	70^a/210^b(half-life)	79^a/80^b	ITO/PEDOT:PSS/3b:PYD2/TPBi/LiF/Al	8.5% ± 0.4%	(0.16, 0.08)	102
SpiroAC-TRZ	−5.70	−3.12	0.072	—	350–450	480^b	17^b	2.1^b	100^b	Glass substrate/ITO/MoO3/TAPC/mCP/mCPCN doped with SpiroAC-TRZ (12 wt%)/3TPYMB/LiF/Al	36.7%	(0.18, 0.43)	103
MXAc-BF	−5.44	−2.48	0.08	0.04	350–400	472^b	24^b	4.3^b	92^b	ITO/HAT-CN/α-NPD/CCP/50 wt%-MXAc-BF:PPF/PPF/TPBi/Liq/Al	16.2%	(0.17, 0.29)	105
1	−5.62	−2.67	0.25	0.15	309, 386	458^a/464^b	12^b	38^b	87^b	ITO/HAT-CN/a-NPD/CCP/1 (18 wt%):PPF/PPF/TPBi/Liq/Al	20.4%	(0.16, 0.23)	104
4BPy-mDTC	−5.65	−2.83	0.01	0.13	342, 396	495^a	28.1^b	18.0^b	7.5^a/97^b	ITO/NPB/TAPC/mCBP:4BPy-mDTC (7 wt%)/DPEPO or PPT/TmPyPb/LiF/Al	28.1%	(0.17, 0.37)	92
Cz-TRZ3	5.23	1.93	0.17	0.13	300–400	435^a	—	13^b	60^a/92^b	ITO/HAT-CN/a-NPD/TCTA/mCP/DPEPO:6 wt% Cz-TRZ3/DPEPO/TPBi/LiF/Al	19.2%	(0.148, 0.098)	107
3DPyM-pDTC	−5.76	−2.76	0.02	—	384	464^a	26.4^a	0.27^a	27^a/98^b	ITO/NPB/TAPC/mCBP/mCBP:3DPyM-pDTC (7 wt%)/DPEPO/TmPyPb/LiF/Al	31.9%	(0.14, 0.18)	108
DMAC-DMT	−6.10	−3.08	0.01	—	—	447^a	24.5^b	2.49^b	90^b	ITO/HATCN/TAPC/DCDPA/DPFPO:DMAC-DMT:BD/TSPO1/TPBi/LiF/Al	19.0%	(0.14, 0.15))	109
DABNA-2	−4.69	−1.21	0.14	0.28	444	470^a/469^b	7^a/6^b	65.3^b	90^b	(ITO)/NPD/TCTA/mCP/1 wt% DABNA-2:mCBP/TSPO1/LiF/Al	20.2%	(0.12, 0.13)	61
ADBNA-Me-Tip	−5.35	−1.90	0.19	0.35	457	479^b	6^b	147^b	88^b	ITO/HATCN/NPD/TCTA/mCP/ADBNA-Me-Tip:DOBNA-OAr/TSPO1/LiF/Al	21.4%	(0.11, 0.29)	110
ν-DABNA-O-Me	−4.88	−1.51	0.119	0.30	449	464^b	5.08^b	7.74^b	90^b	ITO/NPD/TCTA/mCP; 1 wt% ν-DABNA-O-Me emitter and 99 wt% DOBNA-Tol/3,4-2CzBN/BPy-TP2/LiF/Al	29.1%	(0.13, 0.10)	111
BSBS-N1	—	—	0.14	0.31	458	473^a/478^b	0.9^b	5.6^b	59^a/89^b	(ITO)/HAT-CN/TAPC/mCBP/2 wt%-BSBS-N1:mCBP emission layer/PPF/B3PyPB/Liq/Al	21.0%	(0.11, 0.22)	112
Endo-D2	−5.38	−2.74	0.12	—	446^a	470^a	3.6^b	106.8^b	93^b	(ITO)/PEDOT:PSS/PVK/(mCP):10 wt% dendrimer/mSiTRZ/TmPPPyTz/LiF/Al	22.6%	(0.13, 0.22)	115
AC-BO	—	—	0.13	—	300, 380^a	446^b	56^b	11.7^b	76.9^b	ITO/HAT-CN/TAPC/mCP/20% emitters: DPEPO/DPEPO/TmPyPB/LiF/Al	19.6%	(0.148, 0.122)	117
DBADCzPh	−5.20	−1.51	0.13^a/0.21^b	0.39	335, 408^a/298^b	460^a/482^b	8.15^a/7.67^b	17.94^a/97.91^b	50^a/91^b	(ITO)/HAT-CN/TAPC/TCTA/mCBP/PPF: x wt% DBADCzPh/PPF/TmPyPB/Liq/Al	42.5%	(0.16, 0.30)	76

---

Building on the earlier success, further advancement in blue TADF emitters was achieved by introducing a new series of molecules designated as 1a, 2a, 2b, and 2c, which incorporated triazine-based acceptors paired with carbazole- and indolocarbazole-derived donors (Fig. 20a).³⁴ This series showcased an advanced molecular design approach focused on achieving a delicate balance between a low ΔE_ST and high PLQY. The performance of these emitters depended on the strategic delocalization of the frontier molecular orbitals (FMOs) across the D–A charge-transfer system. The spatial separation between the HOMO and LUMO allowed both a high transition dipole moment and remarkably low ΔE_ST of ∼0.1 eV. In compounds 1a and 2a, their low ΔE_ST was primarily attributed to their sterically hindered and twisted D–A geometries. Notably, in 2a, its HOMO was significantly delocalized over its entire 3,6-bis(3,6-diphenylcarbazolyl)carbazole (BDPCC) donor unit, while a moderate twist between the carbazole and BDPCC moieties helped maintain a high T₁ in its donor component. The experimentally determined ΔE_ST values were 0.09–0.12 eV for 1a, 0.09 eV for 2a, 0.28 eV for 2b, and 0.32 eV for 2c, indicating how structural variations influenced the excitonic behavior. Owing to their minimized ΔE_ST, both 1a and 2a displayed strong blue TADF emission when doped in a DPEPO host, with the peak emission wavelengths (λ_max) at 495 nm for 1a and 457 nm for 2a. Their TADF quantum yields (ϕ_TADF) were also notable at 20% for 1a and 21% for 2a. Most impressively, the OLED devices using 2a as the emitting layer achieved an exceptional performance, delivering blue EL with an internal quantum efficiency (IQE) approaching 100% and a groundbreaking EQE of 20.6% ± 1.8% (Fig. 20a(iii)). These results, as summarized in Table 1, underscore the profound impact of finely tuned ΔE_ST values on both the TADF efficiency and overall device performance, surpassing previous benchmarks set by earlier blue TADF emitters.

Fig. 20 (a) (i) Molecular conformations of the emitters; (ii) EL performance characteristics; (iii) EQE versus current density (J) plots; and (iv) schematic of the PL mechanism, top: emitter 1a and bottom: emitter 2a. Adapted with permission.³⁴ Copyright 2014, Springer Nature Limited. (b) (i) Chemical structures of the emitters along with their FMO representations; (ii) EQE versus current density characteristics, with the inset showing the EL spectra; and (iii) CIE chromaticity coordinates of the emitters. Adapted with permission.¹⁰¹ Copyright 2016, WILEY-VCH, Weinheim.

Expanding on the proven twisted D–A design strategy, two highly efficient blue TADF emitters were reported utilizing dimethylacridine (Ac) as the electron-donating unit and either phenoxaphosphine oxide or phenoxathiin dioxide as the electron-accepting moieties, resulting in the creation of Ac-OPO and Ac-OSO, respectively (Fig. 20b).¹⁰¹ The spatial separation of FMOs is a direct consequence of their twisted D–A structures, facilitating extremely small ΔE_ST of 0.03 eV for Ac-OPO and 0.06 eV for Ac-OSO. Furthermore, replacing the phosphine oxide group in Ac-OPO with a more electron-withdrawing sulfone unit in Ac-OSO led to a substantial decrease in the LUMO energy level. As a result, Ac-OSO displayed a lower S₁ excitation energy (2.99 eV) compared to Ac-OPO (3.04 eV), consistent with the stronger electron-withdrawing nature of the sulfone group. Both compounds exhibited efficient blue TADF emissions in solution, with the emission maximum at 433 nm for Ac-OPO and 451 nm for Ac-OSO. Their PL performance improved significantly in DPEPO-doped films, owing to the rigid host matrix, which suppresses non-radiative decay by limiting the vibrational relaxation. The delayed fluorescence lifetimes were measured to be 3.8 μs for Ac-OPO and 2.7 μs for Ac-OSO. The OLEDs fabricated using these emitters in a DPEPO host matrix delivered an outstanding performance. The Ac-OPO-based device achieved an external EQE of 12.3%, with CIE coordinates of (0.15, 0.14). The Ac-OSO-based device exhibited an even higher EQE of 20.5% and CIE coordinates of (0.16, 0.26) (Fig. 20b(ii)). Impressively, the Ac-OSO device demonstrated a reduced efficiency roll-off, with EQE values remaining as high as 17% at 150 cd m⁻² for display use and 13% at 1000 cd m⁻² for lighting use, attributed to its shorter delayed fluorescence lifetime, which facilitates efficient triplet harvesting. A summary of their photophysical and device performance is presented in Table 1.

Another breakthrough was achieved towards the advancement of blue TADF emitters when a series of solution-processable TADF emitters (3a–e) based on a para-biphenylsulfonebenzene (p-BPSB) core was reported for OLEDs (Fig. 21).¹⁰² The key innovation was enhancing the solubility through the n-alkylation of carbazole units or by introducing sterically bulky ortho-substituted diphenylamine groups. By varying the ortho-substituents, specifically methyl and tert-butyl groups, fine control over the steric hindrance was achieved, which altered the dihedral angle between the D and A units. This subtle geometric tuning modulated the ΔE_ST, thereby influencing the τ_d and RISC efficiency. Under an argon atmosphere in toluene solution, compounds 3a–d exhibited high PLQYs of 70%, 79%, 82%, and 80%, respectively. However, in the presence of oxygen (non-degassed conditions), the PLQYs dropped sharply to 38%, 43%, 41%, and 41%, indicating substantial quenching of the triplet excitons by molecular oxygen, which is a hallmark of TADF behavior. This was further supported by the τ_d of 50, 70, 15, and 20 μs for compounds 3a–d, respectively. The OLEDs incorporating these emitters delivered deep-blue emissions across the range of 436–466 nm, with the luminance reaching up to 10 000 cd m⁻². PYD2 was used as the host for fabricating OLEDs owing to its wide band gap, high triplet energy, and suitable energy alignment, which enable efficient exciton confinement and energy transfer to compounds 3a–d. In comparison to PMMA, PYD2 achieved a slight red shift in PL, while retaining comparable transient decay behavior. The devices also achieved efficiencies of up to 8.2 cd A⁻¹ at a brightness of 1000 cd m⁻². Compound 3a displayed the maximum current and power efficiencies due to its relatively low current density, while compound 3b, despite realizing higher brightness and EQE, exhibited reduced efficiencies, which could be attributed to its higher current density and blue-shifted emission. These performance metrics, along with the detailed photophysical data, are summarized in Table 1. This study underscores the effectiveness of steric modification via bulky alkyl groups in creating highly soluble, solution-processable blue TADF emitters. The simplicity and scalability of this approach present a compelling route toward the cost-effective fabrication of next-generation OLEDs.

Fig. 21 (a) Molecular structures of the emitters featuring various D units; (b) emission decay profiles in the μs range; (c) EL characteristics alongside their CIE chromaticity coordinates; (d) EQE as a function of current density; (e) micrographs of the fabricated OLED pixels. Adapted with permission.¹⁰² Copyright 2017, the American Chemical Society.

The development of D–A-based blue emitters saw a significant leap forward with a novel molecular design combining a spiroacridine (donor) and triazine (acceptor) hybrid (Fig. 22a).¹⁰³ This innovative architecture resulted in highly efficient sky-blue TADF emission, characterized by an outstanding PLQY of 100%, excellent thermal stability, and a remarkable horizontal dipole orientation of 83%. Three new blue TADF emitters, SpiroAC-TRZ, DPAC-TRZ, and DMAC- TRZS, were synthesized and evaluated. These emitters exhibited sky-blue emission with peak wavelengths between 478 and 495 nm, alongside impressively low singlet–triplet energy gaps of 72 meV, 133 meV, and 62 meV, respectively.

Fig. 22 (a) (i) Chemical structures of the TADF emitters; (ii) transient PL decay profiles; (iii) normalized EL spectra; (iv) plot of EQE versus doping concentration. Adapted with permission.¹⁰³ Copyright 2016, WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim. (b) (i) Chemical structures of the emitters; (ii) EL characteristics along with corresponding CIE coordinates; (iii) EQEs plotted as a function of luminance. Adapted with permission.⁹³ Copyright 2017, the American Chemical Society. (c) (i) Molecular conformations of the TADF emitters; (ii) PL spectra in doped films along with corresponding CIE coordinates; (iii) (EQE) versus luminance plot, with the inset showing EL characteristics. Adapted with permission.¹⁰⁴ Copyright 2017, the American Chemical Society.

The inclusion of a phenylene linker between the D and A units played a critical role in ensuring sufficient spatial separation of the FMOs, the key to minimizing ΔE_ST, thereby promoting efficient RISC and enabling robust TADF behavior. The materials also exhibited pronounced delayed fluorescence, with lifetimes ranging from 1.9 to 2.9 μs. Notably, the OLEDs incorporating SpiroAC-TRZ as the emitter achieved an exceptional EQE of 37% (Fig. 22a(iv)), positioning them among the highest-performing blue TADF-based OLEDs reported to date. The OLEDs composed of SpiroAC-TRZ, DPAC-TRZ, and DMAC-TRZ (12 wt% doping) establish low turn-on voltages (∼2 V), low operating voltages (∼3 V at 100 cd m⁻²), and efficient sky-blue to green emission. Their EQEs correlated with PLQY and horizontal dipole ratios (Θ_‖), with SpiroAC-TRZ delivering the best performance owing to its highest PLQY and Θ_‖. Remarkably, SpiroAC-TRZ achieved a record EQE of ∼37% in planar devices without outcoupling, retaining ∼30.5% even at 1000 cd m⁻². Doping concentration-dependent studies (8–100 wt%) disclosed the optimal EQEs of 35–37% at 8–16 wt%, while higher concentrations resulted efficiency loss from carrier imbalance or quenching. A detailed summary of their photophysical and EL properties is provided in Table 1.

A novel class of blue TADF emitters was demonstrated by integrating chromone (CM) or isobenzofuranone (BF) as unique electron acceptors with spiro-structured acridine-based donors, specifically acridine-9,9′-xanthene, retaining a twisted D–A π-conjugated architecture (Fig. 22b).¹⁰⁵ Although conventional arylamine-based donors are widely used in TADF systems, the introduction of spiroacridine derivatives in these emitters resulted in an enhanced TADF efficiency. The choice of BF and CM as acceptor units was strategic, given that their limited π-conjugation helped increase the HOMO–LUMO energy gap and elevate the T₁ energy level, both of which are critical for efficient blue emission. The three synthesized emitters, MXAc-BF, MXAc-CM, and XAc-CM, exhibited small ΔE_ST values in the range of 0.08 to 0.11 eV, largely attributed to their highly twisted D–A geometries, with near-orthogonal dihedral angles (∼88°), effectively decoupling the HOMO and LUMO. The key photophysical properties of these emitters include blue emission peaks ranging from 460 to 485 nm with short delayed lifetimes between 2.8 and 4.3 μs, observed in both doped and non-doped film states (PLQY spanning from 53% to 93%). A standout result came from the OLED device using MXAc-BF as the emitter, which achieved an impressive EQE of up to 16.2% (Fig. 22b(iii)). The devices exhibited pure emitter-origin EL spectra, with XAc-CM exhibiting the deepest-blue emission (462 nm, CIE 0.15, 0.19) and MXAc-BF peaking at 478 nm. Among them, the MXAc-BF-based OLEDs displayed the best performance, presenting a low turn-on voltage of 3.0 V, high EQE of 16.2%, and power efficiencies of 31–33 l m W⁻¹, while maintaining an EQE of >12% even at 1000 cd m⁻² with nominal roll-off. This stability is primarily attributed to the short-delayed fluorescence lifetime, which minimizes exciton losses. The MXAc-CM- and XAc-CM-based devices performed optimally at 50 wt% doping concentration, achieving EQE values of 15.0% and 12.1%, respectively; nonetheless, their efficiencies were restricted by their lower PLQY compared with MXAc-BF. The comprehensive photophysical data, OLED architectures, and efficiency metrics are summarized in Table 1.

Another series of deep-blue emitters was developed utilizing pre-twisted D–A architectures, in which a pyrimidine-based acceptor is linked to either spiroacridine or acridine donor units through a phenylene spacer (Fig. 22c).¹⁰⁴ The inclusion of a phenylene linker plays a critical role by introducing steric hindrance between the hydrogen atoms of the donor and the spacer. This steric interaction enforces a near-orthogonal D–A geometry, effectively separating the HOMO and LUMO and minimizing the ΔE_ST, a prerequisite for efficient RISC. Pyrimidine was strategically selected as the acceptor due to its relatively weaker electron-withdrawing nature compared to triazine. This leads to an increased HOMO–LUMO bandgap and elevated S₁ and T₁ energy levels, both favorable for achieving deep-blue emission. The five deep-blue TADF emitters (designated Molecules 1–5) exhibited emission maxima ranging from 448 to 464 nm in both solution and doped thin films. Among the series, Molecule 3 achieved the highest PLQY of 91%, with a delayed fluorescence contribution (Φ_d) of 41% and a delayed lifetime (τ_d) of 45 μs. Molecule 4 (PLQY = 90%, Φ_d = 30%, τ_d = 70 μs) and molecule 1 (PLQY = 87%, Φ_d = 46%, τ_d = 38 μs) exhibited close performances. Notably, due to its highest Φ_d, molecule 1 delivered an outstanding OLED performance, reaching a peak EQE of 20.4%, (Fig. 22c(iii)) with CIE color coordinates of (0.16, 0.23) (Fig. 22c(ii)), indicative of deep-blue emission. In the case of the TADF-OLEDs based on emitters 1–5, their maximum EQE values followed the order of 1 (20.4%) > 3 (17.1%) > 4 (14.3%) > 2 (12.2%) > 5 (11.4%). The lower efficiencies of 2 and 5 originated from their reduced PLQY and ϕ_d, but their EQE values still surpassed that of conventional fluorescent emitters. At higher current densities, a pronounced efficiency roll-off was observed, primarily due to triplet-triplet annihilation (TTA) involving long-lived T₁ excitons. Device 5 exhibited more severe roll-off (J₀ = 0.9 mA cm⁻²) than device 1 (2.1 mA cm⁻²), consistent with its longer τ_d. Achieving deep-blue TADF emitters with shorter τ_d (<1 μs) is expected to retain EQE above 20% even under high current operation. These results are comprehensively summarized in Table 1.

A streamlined D–A molecular strategy was employed to design four efficient blue TADF emitters, 4BPy-mDTC, 3BPy-mDTC, 2BPy-mDTC, and BP-mDTC (Fig. 23a).¹⁰⁶ These emitters incorporate benzoylpyridine (BPy) or benzophenone (BP) cores as electron-accepting units and di(t-butyl)carbazole moieties as donating units. The tert-butyl groups, strategically placed at the C₃ and C₆ positions of the carbazole units, play a pivotal role in enhancing both the chemical and electrochemical stability, which significantly boosts the PLQY. Particularly, the three BPy-based emitters demonstrated remarkedly superior PLQYs (ranging from 92% to 97%) in an mCBP host compared to BP-mDTC, which showed a PLQY of only 75%. This contrast suggests that the nitrogen atom in the BPy acceptor contributes crucially to enhancing the emission efficiency. Among them, 4BPy-mDTC emerged as the top-performing emitter, exhibiting an exceptionally small ΔE_ST of 0.01 eV, delivering a PLQY of 97%, a short τ_d of 18 μs, and excellent OLED performance, with a peak EQE of 28.1% in an mCBP host (Fig. 23a(iv)). Although BP-mDTC had a smaller ΔE_ST (0.03 eV) than 3BPy-mDTC (0.05 eV), the device based on 3BPy-mDTC demonstrated a much higher EQE than the device with BP-mDTC. This improvement is mainly ascribed to the superior PLQY of 3BPy-mDTC in the mCBP thin film, which is likely promoted by the pyridine group. This work demonstrates the effectiveness of a minimalistic molecular design for high-performance blue TADF. The ability to fine-tune ΔE_ST by manipulating the position of the nitrogen heteroatom within the acceptor unit, combined with the use of alkylated carbazole donors with low-lying HOMOs, enables the generation of deep-blue emission with high efficiency.

Fig. 23 (a) (i) Molecular structures of the emitting materials, (ii) and (iii) temperature-dependent transient decay plots of 4BPy-mDTC and BP-mDTC in the mCBP host; (iv) EQE of the devices as a function of luminance; (v) EL spectra of the fabricated devices. Adapted with permission.⁹² Copyright, 2017, the Royal Society of Chemistry. (b) (i) Chemical structures of the molecules; (ii) transient PL decay plots in the μs range; (iii) EQE versus current density plots; (iv) normalized EL spectra of the OLEDs; (v) CIE chromaticity coordinates of Cz-TRZ3. Adapted with permission.¹⁰⁷ Copyright 2017, WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

A straightforward yet effective molecular strategy was introduced for achieving deep-blue TADF emission by incorporating methyl groups at specific positions in D–A molecules (Fig. 23b).¹⁰⁷ This modification enabled precise tuning of the ΔE_ST without significantly affecting the optical bandgap. Four D–A-type emitters were synthesized, Cz-TRZ1 to Cz-TRZ4, featuring cyaphenine as the electron acceptor and carbazole-based donors. Cz-TRZ1, with minimal steric hindrance, exhibited a smaller D–A dihedral angle, resulting in a larger ΔE_ST than its counterparts. The inclusion of two methyl groups at the 1 and 8 positions of the 3,6-dimethylcarbazole donor in Cz-TRZ2 significantly lowered its ionization potential (IP). Despite its large dihedral angle (86.78°), this alteration positioned the charge transfer triplet (³CT) energy below the locally excited triplet (³LE) state, giving Cz-TRZ2 a ³CT-dominated triplet. In Cz-TRZ1, Cz-TRZ3, and Cz-TRZ4, the triplet state retained LE character, with ΔE_ST values of 0.43 eV, 0.17 eV, and 0.15 eV, respectively, while Cz-TRZ2 stood out with a much smaller gap of 0.08 eV. Cz-TRZ2 to Cz-TRZ4 displayed efficient blue TADF emission in the range of 430–470 nm. Cz-TRZ2 exhibited a 33 nm red-shift relative to the others, which is attributed to its reduced donor IP. Further, Cz-TRZ2 demonstrated exceptional PLQY values of 86% in toluene and 98% in DPEPO-hosted films, with a delayed lifetime of 3.5 μs. The devices incorporating these emitters showed strong EL properties with the following efficiencies: Cz-TRZ2:EQE of 22%, Cz-TRZ3:EQE of 19.2%, and Cz-TRZ4:EQE of 18.3% respectively (Fig. 23b(iii)). Additionally, Cz-TRZ3 and Cz-TRZ4 achieved near deep-blue CIE coordinates, making them promising candidates for next-generation blue OLED applications. Further details on the device architecture and performance metrics can be found in Table 1.

Although most blue TADF emitters utilize twisted D–A architectures, these designs frequently encounter issues with structural relaxation, resulting in broadened and red-shifted emission spectra, a major drawback when aiming for high color purity, especially in pure blue emission. To overcome this limitation, an alternative strategy was proposed by adopting linear, rigid, rod-like molecular frameworks to enhance the spectral stability and purity. Two isomeric TADF emitters were demonstrated, 2DPyM-mDTC and 3DPyM-pDTC, both constructed with a central keto group and two pyridine rings serving as electron-accepting units and two di(t-butyl)carbazolyl groups functioning as electron donors (Fig. 24).¹⁰⁸ In 2DPyM-mDTC, the donor units are connected at the meta positions relative to the central carbonyl group, whereas in 3DPyM-pDTC, the donor–acceptor linkage occurs at the para positions. Though subtle, this positional change had a clear impact on their crystal structures, which in turn influenced their photophysical and EL behavior. 3DPyM-pDTC exhibited a rigid, planar conformation, which minimized the non-radiative relaxation pathways, resulting in a remarkable PLQY of 98%, with effective TADF characteristics that delivered an exceptional EQE of 31% in a flexible OLED (Fig. 24b). It also produced narrow-band blue emission with an FWHM of just 62 nm, alongside desirable CIE coordinates of (0.14, 0.18), demonstrating excellent color purity. These findings make 3DPyM-pDTC a highly promising candidate for next-generation pure blue TADF emitters in commercial OLED applications. The detailed photophysical and device performance data are provided in Table 1.

Fig. 24 (a) Chemical structures of the two TADF emitters; (b) EQE versus luminance plots, with the inset showing EL characteristics; (c) crystal structure of 3DPyM-pDTC; (d) photographs of the corresponding devices. Adapted with permission.¹⁰⁸ Copyright 2017, the American Chemical Society.

To enhance the blue emission efficiency, a TADF host-sensitization (THS) strategy that utilizes conventional fluorescent emitters was explored. In this method, both singlet and triplet excitons generated within a TADF host are transferred to a fluorescent emitter, provided there is strong spectral overlap between the emission of the TADF host and absorption of the fluorescent emitter. As part of this approach, a new TADF sensitizer was developed, SPAC-DMT, by structurally combining two established blue TADF materials, DMAC-DMT and DMAC-DPS (Fig. 25).¹⁰⁹ This hybrid emitter was employed as the energy donor, while a newly synthesized fluorescent molecule, BPPyA, served as the acceptor, emitting light with a peak wavelength at 458 nm. PL studies of doped films revealed a significantly enhanced emission intensity when the TADF sensitizer was present, clearly indicating efficient fluorescence resonance energy transfer (FRET) from the TADF donor to the fluorescent acceptor. An OLED device was fabricated using DMAC-DMT as the sensitizer, paired with the fluorescent emitter in a DPEPO host, which achieved an impressive EQE of 19% with CIE coordinates of (0.14, 0.15), underscoring the effectiveness of the THS approach for achieving high-performance blue emission (Fig. 25e). In films, TADF materials often realize reduced PLQY due to singlet CT stabilization, self-quenching, and aggregation. To alleviate these effects, a three-component THS system consisting of a host with a wide band gap and high-triplet-energy (DPEPO), the proposed TADF hosts, and BPPyA fluorescent emitter is projected. In this configuration, DPEPO disperses the TADF material, minimizing aggregation and quenching, while the fluorescent emitter can maintain deep-blue, narrow-band emission with nearly 100% IQE.

Fig. 25 (a) Molecular structures of SPAC-DMT and BPPyA; (b) spectral overlap between the emission of the sensitizers and the absorption of the fluorescent receptor probe; (c) transient emission decay curves; (d) schematic of the sensitization mechanism; (e) EQE versus current density plot; (f) normalized EL spectra. Adapted with permission.¹⁰⁹ Copyright 2018, the American Chemical Society.

Moving beyond traditional D–A frameworks, a revolutionary strategy was introduced to achieve spatial separation of the HOMO and LUMO energy levels by leveraging the opposing resonance effects of boron (B) and nitrogen (N) atoms (Fig. 26a). Two compounds, DABNA-1 and DABNA-2, were synthesized, which are based on a rigid polycyclic aromatic hydrocarbon (PAH) skeleton (Fig. 26b).⁶¹ In these structures, triphenylboron units are intricately fused with phenyl rings containing two N atoms. The design concept centres around the contrasting electronic characteristics of B and N, with N acting as an electron donor due to the presence of lone pair of electrons, whereas B, with an empty p-orbital, serves as an electron acceptor. This opposite resonance effect results in a unique distribution of frontier molecular orbitals with the HOMO localized around the nitrogen atoms and their adjacent ortho/para positions and the LUMO is primarily centred on the boron atom and its surrounding ortho/para positions. This distribution creates a small ΔE_ST of approximately 0.2 eV, making these molecules highly effective as ultrapure blue TADF emitters. This novel design paradigm is classified as multiple resonance TADF (MR-TADF). The photophysical properties of these emitters are particularly striking, where DABNA-1 and DABNA-2 exhibit sharp blue emission peaks at 459 nm and 467 nm, respectively. They display an extremely narrow full width at half maximum (FWHM) value of just 28 nm, ensuring exceptional color purity. Their low Stokes shifts and high oscillator strengths (f = 0.205 for DABNA-1 and f = 0.415 for DABNA-2) further reflect the efficiency of the S₀ → S₁ transition. This approach offers a compelling alternative to conventional TADF strategies, with strong potential for next-generation high-color-purity OLED applications. Building upon the promising results of DABNA-1 and DABNA-2, which exhibited delayed fluorescence lifetimes of 93.7 μs and 65.3 μs, along with impressive OLED EQEs of 13.5% and 20.2% (Fig. 26d(i)), respectively, further research has continued to push the boundaries of multiple resonance TADF (MR-TADF) emitter design. Using the same MR approach, two sky-blue MR-TADF emitters, ADBNA-Me-Mes and ADBNA-Me-Tip, were introduced, which were successfully integrated into OLED devices (Fig. 26b).¹¹⁰ These emitters delivered EL maxima at 480 nm and 481 nm, with the corresponding EQEs of 16.2% and 21.4% (Fig. 26d(ii)), respectively, showcasing their high efficiency in sky-blue emission regions. More recently, a new set of pure blue MR-TADF emitters, ν-DABNA, ν-DABNA-O, and ν-DABNA-O-Me, was demonstrated with a pivotal structural modification in this series by replacing the N atom with an O atom in ν-DABNA-O (Fig. 26b).¹¹¹ This substitution caused a blue-shift in emission due to the lower HOMO energy level of O and reduced π-conjugation, which effectively widened the HOMO–LUMO energy gap. Among them, ν-DABNA-O-Me stood out by achieving a record-breaking EQE of 29.5% (Fig. 26d(iii)), the highest reported to date for deep-blue TADF OLEDs.

Fig. 26 (a) (i) and (ii) Schematic representation of the MR-TADF mechanism and design strategy, highlighting HOMO–LUMO separation and comparing the resulting properties with that of conventional TADF emitters; (b) molecular structures of selected MR-TADF emitters; (c) EL characteristics and (d) EQE versus luminance plots for devices based on: (i) DABNA-1 (blue) and DABNA-2 (red), (ii) ADBNA-Me-Tip (red) and ADBNA-Me-Mes (blue), (iii) ν-DABNA-O-Me, and (iv) BSBS-N1. Adapted with permission.⁶¹ Copyright 2016, WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim. Adapted with permission.¹¹⁰ Copyright 2018, the American Chemical Society. Adapted with permission.¹¹¹ Copyright 2021, Wiley-VCH GmbH. Adapted with permission.¹¹² Copyright 2021 Wiley-VCH GmbH.

In a more recent advancement, ternary-doped MR systems were explored incorporating B, N, and S (sulfur) atoms, and a fused nonacyclic MR emitter was demonstrated, BSBS-N1, which significantly broadens the structural landscape of MR-TADF emitters (Fig. 26b).¹¹² The inclusion of two S atoms not only enhances the electron-donating characteristics (similar to nitrogen) but also introduces a moderate heavy atom effect and SOC and promotes RISC. As a result, BSBS-N1 achieved a sharp sky-blue TADF emission with a remarkably high k_RISC of 1.9 × 10⁶ s⁻¹, the fastest among known MR-TADF system at the time it was reported. OLEDs using BSBS-N1 demonstrated an excellent performance, with a maximum EQE of 21.0% (Fig. 26d(iv)), a narrow FWHM of just 25 nm, and suppressed efficiency roll-off.

Unlike conventional D–A TADF materials, which exhibit broad emission spectra (FWHM: 50–100 nm), due to ICT and extensive vibrational relaxation in excited-state MR-TADF emitters, they offer narrowband emission and superior color purity. This makes MR-TADF a highly promising design paradigm for next-generation blue-emitting OLEDs, especially in display technologies demanding high precision in color coordinates.

Despite the significant progress, multiple challenges persist in the development of MR-TADF emitters. Firstly, their rigid planar structures often assist molecular aggregation at high doping concentrations, leading to aggregation-caused emission quenching (ACQ) and undesirable spectral broadening, which is detrimental to color purity.¹¹³ Further, many MR-TADF emitters suffer from relatively slow RISC rates (k_RISC ≈ 10³–10⁴ s⁻¹), which hinder the effective utilization of triplet excitons.^114,115 Even though high-efficiency OLEDs based on MR-TADF materials have been realized through vacuum deposition methods, solution-processed devices, which are more compatible with low-cost, large-area fabrication, still exhibit a considerably lower performance. A recent study revealed the development of solution-processable MR dendrimers by integrating second- and third-generation carbazole dendrons into B, N, and O-doped polycyclic MR emitters. These materials enabled the fabrication of blue OLEDs with a maximum EQE of 13.4%.¹¹⁶ However, the reported dendrimers employed conventional exo-tethered conformations and the fluorophore resided outside the dendritic framework. This incomplete encapsulation triggered molecular aggregation at high doping levels (>10 wt%), resulting in spectral broadening and emission redshift. Thus, to address these drawbacks, the researchers proposed an endo-encapsulated dendrimer architecture, where a B- and S-doped MR fluorophore is embedded within carbazole dendrons.¹¹⁵ The structure was realized by attaching first- and second-generation dendrons to a central carbazole core through 1,8-linkages, forming a cavity that efficiently encloses the MR fluorophore (Fig. 27). This novel design strategy suppresses intermolecular aggregation, resulting in stable and narrowband blue emission (469–471 nm, FWHM: 32–34 nm) even at doping concentrations up to 100 wt%. Additionally, the adjacent intramolecular π-stacking between the dendrons and the encapsulated MR fluorophore induce through-space electronic interactions, facilitating TSCT, which lower the S₁ energy and reduce the ΔE_ST. This leads to a 4.2-fold augmentation in k_RISC ≈ 4.7 × 10⁵ s⁻¹ compared to its exo-tethered counterpart. The OLEDs fabricated using this dendrimer exhibited highly efficient, narrowband blue emission with a peak EQE of 22.6%, representing the highest efficiency achieved for solution-processable MR dendrimer-based devices to date. This work presents a new topological strategy for MR luminescent dendrimers and highlights the auspicious potential of MR-dendritic frameworks in realizing high-performance, solution-processable, narrowband OLEDs. Blue TSCT-TADF emitters have achieved remarkable efficiencies, but their color purity remains inadequate to meet the ultrapure blue standards due to the significant Stokes shifts and broad emission profiles of TSCT emitters, which lead to reduced PLQY. Integrating TSCT with MR-TADF in the molecular design can help mitigate spectral broadening and minimize the Stokes shift, thereby enhancing the overall PLQY of the emitters.

Fig. 27 (a) Dendrimer-based endo and exo-encapsulated MR fluorophores, (b) crystal structure of Endo-D1 and reduced density gradient iso-surfaces (RDG) of Endo-D1 and Endo-D2, (c)–(f) EL spectra of Exo-D1, Endo-D1, Exo-D2, and Endo-D2 at various doping concentrations, (g)–(j) EQE as a function of luminance for Exo-D1, Endo-D1, Exo-D2, and Endo-D2, respectively, at various doping concentrations of the emitters. Adapted with permission.¹¹⁵ Copyright 2025, WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

Recently, three deep-blue TSCT emitters, AC-BO, QAC-BO, and Cz-BO, were designed (Fig. 28a) by meticulously monitoring the intramolecular interactions between rigid acridine-based donors and B/O-containing MR-based acceptors.¹¹⁷ Steric locking, comprehended by rigidifying the D units with acridine-based frameworks, plays a vital role in shaping the molecular conformation. The incorporation of covalent connections and sterically hindering substituents effectively limits intramolecular rotations, imposing either a fixed coplanar alignment or a definite twisted orientation between D and A frameworks, which augmented the radiative transitions and repressed vibrational relaxation. This orientation permitted remarkably close D–A stacking (2.64 Å in QAC-BO), enabling efficient and controlled TSCT. Subsequently, the emitters displayed pronouncedly reduced Stokes shifts, narrower FWHM as low as 35 nm for Cz-BO, and high PLQYs ranging from 77% to 89%. The OLEDs fabricated from AC-BO achieved a maximum EQE of 19.3% with CIE coordinates of (0.147, 0.122), whereas QAC-BO exhibited ultrapure-blue EL with the maximum EQE of 15.8% and CIE coordinates of (0.145, 0.076), meeting the desired blue standards (Fig. 28d and e). However, Cz-BO lacked TADF features due to its larger ΔE_ST, but delivered deep-blue fluorescence with a maximum EQE of 5.5% and ultranarrow FWHM of 43 nm, which is the narrowest reported among TSCT emitters.

Fig. 28 (a) and (b) Molecular structures and basic design strategy, (c) normalized EL spectra of the respective emitters, (d) CIE co-ordinates and FWHM plots of the emitters, (e) EQE as a function of current density, and inset: EQE as a function of luminance. Adapted with permission.¹¹⁷ Copyright 2022, WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

In a recent study, two novel sky-blue multi-resonance charge transfer (MRCT) TADF emitters were introduced, DBACzPh and DBADCzPh, by inventively linking carbazole-based donor groups to an oxygen-bridged boron acceptor (DBA) through C–C bonds, departing from the traditional C–N connections (Fig. 29).⁷⁶ This pivotal structural adjustment reduced the dihedral angles between the D and A units to 37–38°, thereby enhancing the conjugation and molecular planarity, and nurturing preferential horizontal orientation in thin films. The use of C–C bonds significantly improved the chemical and thermal stability, owing to their much higher bond dissociation energy (∼188 kcal mol⁻¹) compared to C–N bonds (∼89 kcal mol⁻¹), a crucial factor for prolonging the OLED device lifetime. Both emitters exhibited strong MRCT behavior by integrating short- and long-range charge transfer states, resulting a small ΔE_ST ≈ 0.20–0.24 eV, which triggered efficient RISC and effective triplet harvesting.

Fig. 29 (a) Schematic of the TADF mechanism, (b) chemical structures of the designed C–C connected D–A and D–A–D-type MRCT emitters, DBACzPh and DBADCzPh, along with their photophysical properties and device performance metrics, (c) OLED device architecture used for fabrication, and (d) EQE versus brightness curves; the inset depicts the orientation of the host and emitter molecules within the emissive layer (EML). Adapted with permission.⁷⁶ Copyright 2025, Springer Nature Limited.

Photophysical studies revealed the PLQYs in doped films (68% for DBACzPh and 91% for DBADCzPh) in the polar PPF host with short prompt decay (∼8 ns) and long delayed lifetimes (17–20 μs), suggesting pronounced TADF behavior. Both emitters exhibited narrow sky-blue emission peaks (460–477 nm) with minimal solvatochromic shifts (35–40 nm), confirming their stable MRCT-based PL. DFT calculations aligned with the experimental results, revealing that the HOMOs are localized on the donor moieties and LUMOs centered on the acceptor, a distribution vital for efficient TADF. Additionally, both emitters exhibited a preferential horizontal transition dipole orientation, particularly in the polar PPF host, remarkably triggering light outcoupling. This molecular design with optimized orientation is the key to realizing the record-high EQE of up to 42.5% in TADF-based sky-blue OLEDs. By leveraging MRCT design strategies, this specific state of the art molecular engineering greatly boosted TADF performance, while overcoming the persisting challenges in blue TADF emitters, such as limited stability and poor horizontal alignment.

3.2 Breakthrough towards green and red TADF emitters

In the early phase of TADF research, blue and red emitters attracted significant focus due to the complexity involved in their molecular design. In parallel, various structural strategies have also been actively applied to develop efficient green TADF emitters, aiming to achieve strong TADF characteristics and high PLQYs. To meet these criteria, different D–A configurations have been explored, including strong donor–weak acceptor, weak donor–strong acceptor, and balanced D–A combinations.¹¹⁸ These arrangements help optimize green TADF materials by reducing ΔE_ST and promoting effective intramolecular charge transfer.¹¹⁹ Initial breakthroughs in green TADF emitters often involved cyano- and triazine-based acceptors due to their strong electron-withdrawing properties. Over time, the scope expanded to include other acceptor units such as pyrimidine, aromatic ketones, boron-containing groups, sulfones, oxadiazoles, triazoles, thiazoles, oxazoles, and imidazoles. More recently, refinements in both D and A structures have been pursued to further enhance the EL efficiency.¹² As a result of these efforts, the EQE of pure green TADF OLEDs has approached 40%, which is possible by carefully selecting and combining suitable D and A fragments that maintain the desired emission wavelength without compromising the efficiency.¹²⁰

Green OLEDs designed using D–A and D–A–D molecular frameworks have demonstrated excellent EL performances with appropriate emission characteristics. However, these emitters typically exhibit broad emission profiles, with FWHM values exceeding 45 nm. This spectral broadening is primarily attributed to the pronounced ICT nature inherent in the D–A structure. In the case of advanced display technologies, particularly ultra-high-definition displays, the strict standards set by the International Telecommunication Union require emitters with high color purity, which necessitates materials with narrow FWHM emission.^33,121,122

Many green TADF emitters were constructed using strong electron-donating units such as phenoxazine, phenothiazine, and dihydrophenazine, along with extended π-conjugated donors such as carbazole and diphenylamine derivatives.¹²³ On the acceptor side, the commonly employed groups include cyano-functionalized aromatics, triazine, and benzophenone, owing to their effective electron-withdrawing capabilities. In contrast, acceptors such as diphenylsulfone and pyrimidine are generally considered less favorable due to their relatively weak electron-accepting nature. Notably, cyano-based acceptors are among the most frequently used in green to yellow TADF materials.

3.2.1 Green TADF emitters. Green TADF emitters utilizing cyano- and triazole-based acceptors are among the earliest and most extensively explored classes of TADF materials, with a well-established foundation in both research and application. Structurally, cyano-containing green TADF compounds can be broadly categorized into two main types: (i) monomeric systems incorporating ortho steric hindrance, which helps spatially separate the HOMO and LUMO; and (ii) homo conjugated systems, where through-space electronic interactions are leveraged to achieve the same orbital separation.

One of the most notable breakthroughs in this field came in 2012, with the introduction of 4CzIPN,¹² a benchmark TADF emitter (Fig. 30a). This molecule features carbazole units substituted at the 2, 4, 5, and 6 positions and cyano groups at the 1 and 3 positions on a central benzene ring. The spatial crowding between the carbazole and dicyanobenzene fragments induces a significant twist of about 60° between their planes. This molecular geometry results in effective spatial separation between the HOMO and LUMO, yielding a small ΔE_ST of 83 meV. At the same time, OLEDs based on 4CzIPN achieved a remarkably high EQE of 19.3% (Fig. 30b). To date, this compound continues to be a benchmark green TADF emitter, widely recognized not only for its use in optoelectronic applications but also for its utility in photocatalytic systems. A particularly notable feature is its tuneable EQE, which varies depending on the host matrix employed. When doped into different host materials, the EQE has been reported to reach values of 19.3% with CBP, 26.5% with mCPSOB, 26.7% with DCzDCN, and 31.2% with 3CzPFP, demonstrating its versatility and performance across multiple device configurations.¹²⁴

Fig. 30 (a) Molecular structures of CDCB emitters. (b) EQE trends of OLEDs incorporating 4CzIPN, 4CzTPN-Ph, and 2CzPN as emitters with an increase in current density, maintaining low measurement errors (within 1.5%, 1.0%, and 1.0%, respectively). The inset displays their corresponding EL spectra recorded at 10 mA⁻². (c) PL decay of a 6 wt% 4CzIPN:CBP film was monitored at 300 K, 200 K, and 100 K under 337 nm excitation, showing variation in emission lifetimes with temperature. (d) PL profile further separated into prompt and delayed components to analyze the emission dynamics. Adapted with permission.¹² Copyright 2012, Nature.

Relying on RISC, TADF offers an effective method to enhance the light output in OLEDs. However, one of the main bottlenecks is the generally slow RISC rate (k_RISC), which negatively impacts both performance and stability. In 2020, Adachi and group developed a TADF molecule featuring multiple donor groups, which has shown considerable promise (Fig. 31a).¹²⁵ In 5Cz-TRZ, the multi-carbazole orientation around the triazine core enables strong charge-resonance interactions, a dense triplet manifold, and fast multi-channel RISC, unlike the other emitters. This unique orientation also favors horizontal alignment in films, reducing ΔE_ST and boosting the exciton utilization, which together result in a much higher device efficiency. These donor units create charge-resonance hybrid triplet states, resulting in a small singlet–triplet energy gap, enhanced spin–orbit coupling, and a dense network of triplet states closely spaced with the singlet states. This configuration enables a rapid RISC rate of 1.5 × 10⁷ s⁻¹, which is approximately two orders of magnitude faster than that of the typical TADF materials. The OLEDs incorporating this molecule demonstrate strong operational stability (T₉₀ ≈ 600 h at 1000 cd m⁻²), outstanding peak EQE (>29.3%) (Fig. 31b), and minimal efficiency loss under high brightness conditions (<2.3% roll-off at 1000 cd m⁻²).

Fig. 31 (a) Chemical structures of TADF emitters. (b) Absorption spectra (300 K) of 5Cz-TRZ in toluene and PL spectra (300 K) of 5Cz-TRZ in different solvents. (c) Transient PL spectra (300 K) of 5Cz-TRZ in oxygen-free toluene, chloroform and dimethyl formamide. The concentration of all solution samples is 1 × 10⁻⁶ mol L⁻¹. (d) Phosphorescence and PL spectra of 5Cz-TRZ in toluene (1 × 10⁻⁶ mol L⁻¹) at 77 K. Adapted with permission.¹²⁵ Copyright 2020, Nature.

Aromatic ketones and sulfones are also frequently used as acceptor units in the design of green TADF emitters. In 2014, the first TADF materials exhibiting aggregation-induced emission (AIE) characteristics, employing 9H-thioxanthen-9-one-10,10-dioxide (TXO) as the electron-accepting core, were introduced.¹²⁶ Two new thioxanthone (TX)-based emitters, namely TXO-TPA and TXO-PhCz, have recently been reported as efficient TADF materials. These compounds adopt a classic donor–acceptor (D–A) architecture, in which the oxidized thioxanthone-10,10-dioxide (TXO) fragment serves as the electron-accepting unit, while triphenylamine (TPA) or N-phenylcarbazole (PhCz) act as the electron donor (Fig. 32a). The oxidation of sulfur in the TX unit significantly enhances its electron-withdrawing capacity, and the incorporation of TPA or PhCz contributes favorable hole-transporting characteristics. The angular linkage of donor and acceptor units effectively reduces the orbital overlap, thereby separating the HOMO and LUMO distributions. This molecular design yields A small ΔE_ST of 52 meV for TXO-TPA and 73 meV for TXO-PhCz, which facilitates RISC and results in high PL quantum efficiency. Correspondingly, the OLEDs fabricated with these emitters achieve an excellent device performance, exhibiting maximum external quantum efficiencies of 18.5% and 21.5%, respectively, along with stable emission spectra (Fig. 32b and d). TXO-PhCz shows higher efficiency than TXO-TPA because its carbazole donor enforces a more rigid and planar orientation with TXO, enhancing the orbital overlap and horizontal dipole alignment. In contrast, the flexible geometry of TPA reduces the exciton utilization and light outcoupling efficiency. These results highlight the effectiveness of sulfur oxidation and donor selection as structural tuning strategies for developing high-performance TADF emitters in OLED applications.

Fig. 32 (a) Molecular structures of TXO-PhCz, TPBI, and mCP. (b) Absorption spectra of TXO-PhCz and TPBI films, along with fluorescence spectra of TXO-PhCz:TPBI-doped films and pure TPBI, highlight their optical properties. (c) Schematic of the OLED device structure (d) EQE versus luminance curve illustrating the performance of the fabricated OLEDs. Adapted with permission.¹²⁷ Copyright 2014, John Wiley and Sons.

A series of four regioisomeric D–A–D compounds, namely 2,3-TXO-PhCz, 2,6-TXO-PhCz, 2,7-TXO-PhCz, and 3,6-TXO-PhCz, was developed in 2019¹²⁷ with the variation in the attachment positions of the PhCz donor groups significantly influencing their photophysical behavior (Fig. 33a). These positional differences notably affect both the ΔE_ST and oscillator strength. Among them, 2,3-TXO-PhCz shows weak luminescence due to the substantial steric hindrance between its donor groups. In contrast, 2,6-, 2,7-, and 3,6-TXO-PhCz display intense emission.

Fig. 33 (a) Energy level diagram of OLEDs based on 2,3-TXO-PhCz, 2,6-TXO-PhCz, 2,7-TXO-PhCz, and 3,6-TXO-PhCz; the EL properties of 2,6-TXO-PhCz and 3,6-TXO-PhCz were tested using device structure (a) with a doping concentration of 8 wt% in the CBP host. The EL properties of 2,3-TXO-PhCz and 2,7-TXO-PhCz were tested using the same device structure with a doping concentration of 15 wt% in the CBP host. (b) Molecular structure; (c) J–V–L characteristics; (d) CE–L–PE characteristics; (e) EQE–L characteristics of the device; and (f) EL spectra. Adapted with permission.¹²⁷ Copyright 2019, John Wiley and Sons.

All four isomers feature small ΔE_ST values ranging between 0.01 and 0.24 eV, with the oscillator strengths measured at 0.064, 0.107, 0.026, and 0.134, respectively. When incorporated into doped films using CBP as the host, their PLQYs are as follows: 62.1% for 2,3-TXO-PhCz, 83.8% for 2,6-TXO-PhCz, 89.0% for 2,7-TXO-PhCz, and 85.4% for 3,6-TXO-PhCz. Although 2,6- and 2,7-TXO-PhCz differ significantly in ΔE_ST and f values, both yield comparable PLQYs and high device performance, achieving EQEs of 23.2% and 24.4%, respectively (Fig. 33f). 2,7-TXO-PhCz achieves higher efficiency because its symmetric donor substitution enforces a more planar and rigid molecular orientation, minimizing steric hindrance and enhancing the orbital overlap. This alignment promotes stronger intermolecular interactions and higher horizontal dipole ratios, improving the exciton utilization and device efficiency. This performance similarity is attributed to the cooperative balance between their oscillator strengths and energy gaps. These results emphasize the importance of fine-tuning the substitution patterns on the acceptor unit to optimize ΔE_ST and f, ultimately enhancing the OLED efficiency.

Boron-containing acceptor units have recently emerged as highly effective components in the design of blue and green TADF emitters, particularly with the rapid advancement of MR-TADF systems. The inclusion of boron typically enhances the LUMO localization, which helps achieve a small singlet–triplet energy gap ΔE_ST, a key requirement for efficient TADF behavior. In 2018, two green-emitting TADF molecules, CzDBA and tBuCzDBA, were constructed using 9,10-dihydro-9,10-diboraanthracene as the diboron-based acceptor (Fig. 34a).¹²⁸ These materials demonstrated excellent photophysical properties and enabled the fabrication of high-performance OLEDs with minimal efficiency roll-off.

Fig. 34 (a) Chemical structure of CzDBA and tBuCzDBA. (b) EQE versus luminance. Inset, EL spectra. (c) and (d) Absorption spectra in toluene, fluorescence spectra at room temperature and phosphorescence spectra at 77 K of CzDBA (c) and tBuCzDBA (d) in thin films. (e) and (f) Temperature-dependent transient PL decay curves of CzDBA (e) and tBuCzDBA (f) in doped films at temperatures ranging from 77 K to 300 K. Prompt and delayed (10 μs) emission spectra are shown as insets. Adapted with permission.¹²⁸ Copyright 2018, Nature.

Both emitters adopt a D–A–D configuration and possess a rod-like molecular shape. They exhibit near unity PLQY and a high degree of horizontal dipole orientation (∼84%) in thin films. This molecular alignment is favorable, given that horizontally oriented emitters significantly boost the light out-coupling efficiency (η_out) of OLEDs. This orientation is typically achieved because rod-shaped or disc-like molecules naturally align parallel to the substrate surface.

CzDBA shows higher efficiency than tBuCzDBA because its less bulky structure enables tighter molecular packing and higher horizontal dipole orientation, enhancing exciton harvesting. In contrast, the bulky tert-butyl groups in tBuCzDBA disrupt the molecular planarity, reducing the charge transport and overall device efficiency. The devices incorporating CzDBA as the emitter achieved an impressive EQE of 37.8% ± 0.6%, with a current efficiency of 139.6 ± 2.8 cd A⁻¹ (Fig. 34e) and power efficiency (PE) of 121.6 ± 3.1 lm W⁻¹. Notably, the efficiency roll-off was as low as 0.3% at a brightness level of 1000 cd m⁻². The device showed green emission centered at 528 nm with CIE coordinates of (0.31, 0.61), highlighting its potential for display technologies.

Recently, a promising green TADF emitter has been reported from the MR family, which is increasingly recognized as strong candidates to meet the stringent BT.2020 green color purity standard.¹³³ Despite the straightforward synthetic accessibility of C/N- and C=O/N-based MR frameworks, they remain relatively underexplored. To overcome this limitation, a hybrid strategy was proposed by combining C/N and C=O/N MR cores, leading to the development of a pure-green emitter, DPQAO-ICz, whose molecular structure is shown in Fig. 35e. This design preserved the MR character of the parent fragments, while effectively red shifting the emission into the target green region. In toluene, DPQAO-ICz displayed a sharp emission peak at 504 nm with a narrow FWHM of 26 nm, and the extension of its π-conjugated framework enhanced the oscillator strength, yielding a ϕ_PL as high as 92% in doped films (Fig. 35f). DPQAO-ICz exhibits high efficiency due to its rigid and planar orientation with a very small dihedral angle, ensuring strong conjugation, while steric protection from the diphenyl groups suppresses π–π stacking and excimer formation. This optimized molecular orientation preserves the short-range charge transfer character, narrows the FWHM, and enhances the PL efficiency.

Fig. 35 (a) Device configuration of mCP-hosted TADF OLEDs and Trz-Py-NCS-sensitized hyper-OLEDs. (b) EL spectra, (c) current density–voltage–luminance characteristics, and (d) EQE, current efficiency, and power efficiency versus luminance for DPQAO-ICz devices. (e) Chemical structure of DPQAO-ICz emitter and (f) UV-vis absorption and PL spectra in toluene. Adapted with permission.¹³³ Copyright 2025, John Wiley and Sons.

The device performance further validated this molecular design. The OLEDs employing DPQAO-ICz produced pure green EL at 515 nm with an FWHM of 34 nm, as shown in Fig. 35b, delivering excellent color purity with a CIEy value of 0.68 and maximum EQE of 25.9%, among the highest for C/N- and C=O/N-based MR systems (Fig. 35d). Moreover, the hyperfluorescent OLEDs that integrated DPQAO-ICz as the terminal emitter and Trz-Py-NCS as the TADF sensitizer (Fig. 35a) achieved an even higher EQE_max of 31.6%, while maintaining narrowband emission and suppressed efficiency roll-off, as confirmed by the J–V–L characteristics (Fig. 35c). Together, these results highlight π-plane extension as an effective route to redshift the emission without compromising the oscillator strength or spectral sharpness, establishing DPQAO-ICz as a benchmark for next-generation green MR-TADF emitters.

A recent breakthrough in TSCT-TADF design has demonstrated that both radiative decay and RISC can be enhanced simultaneously by leveraging cooperative intra- and intermolecular charge-transfer channels.¹³⁴ Indolo phenoxazine (IPXZ) was used as the donor, a spiro-carbon-locked benzophenone as the acceptor, and either fluorene (emitter 1) or xanthene (emitter 2) as the bridging unit (Fig. 36a). Both molecules adopt rigid “face-to-face” donor–acceptor orientations, but the xanthene-bridged emitter (2) shows a more parallel and closer alignment compared to its fluorene-bridged counterpart (1). Interestingly, in emitter 1, intermolecular donor–acceptor interactions are as significant as the intramolecular ones, leading to highly efficient intermolecular TSCT. This effect enhances both the singlet radiative decay rate (k_r,s), raising it to the same level as emitter 2 (∼10⁷ s⁻¹), and the reverse intersystem crossing rate (k_RISC), thereby boosting the overall emission efficiency. In 20 wt% doped PPF films, both emitters exhibited high green TADF efficiencies with k_r,s/k_RISC values of 1.1 × 10⁷ s⁻¹/1.3 × 10⁶ s⁻¹ for emitter 1 and 1.2 × 10⁷ s⁻¹/7.7 × 10⁵ s⁻¹ for emitter 2, as confirmed by steady-state PL spectra and transient PL decay analyses (Fig. 36d and e), respectively. The device studies further validated these results, where the OLEDs adopting a conventional multilayer structure (Fig. 36b) showed excellent EL with peaks in the green region (Fig. 36c), achieving maximum EQEs up to 27.5% and maintaining low efficiency roll-off. Moreover, the current density–luminance–voltage characteristics revealed stable charge transport and balanced emission behavior (Fig. 2f). Importantly, when applied as sensitizers in hyperfluorescent OLEDs, both emitters enabled narrowband blue-green emission with EQEs reaching 30.6% (Fig. 36g). Collectively, this work demonstrates that intermolecular TSCT represents a powerful design principle to concurrently accelerate radiative and RISC processes, thereby unlocking higher efficiencies in TADF systems. A comprehensive summary of the photophysical properties and device performance of green TADF emitters is provided in Table 2.

Fig. 36 (a) Chemical structures of emitters 1 and 2; (b) schematic of the OLED device architecture; (c) EL spectra; (d) PL spectra; (e) transient PL decay profiles for 20 wt%-doped PPF films; (f) current density–luminance–voltage characteristics; and (g) EQE–luminance plots of OLEDs incorporating emitters 1 and 2 at 20 wt% doping. Adapted with permission.¹³⁴ Copyright 2025, John Wiley and Sons.

Table 2 Photophysical and computational parameters of green and red TADF emitters and related OLED device configurations with EL device performance

TADF emitter name	HOMO energy (eV)	LUMO energy (eV)	ΔEST Exp. (eV)	ΔEST Theo. (eV)	λ max, abs (nm)	λ max, PL (nm)	τ P (ns)	τ d (μs)	ϕ [%]	OLED structure	MaxEQE (%)	CIE	Ref.
a Solution and non-doped film. b Doped film.
4CzIPN	—	—	83 meV	—	375	507^a	17.8^a	5.1^a	93.8^a	—	19.3%	—	12
5CzTRZ				0.14	340	490^a	5.7^a	1.9^a	92^a	(ITO/HAT-CN (10 nm)/α-NPD (30 nm)/Tris-PCz (10 nm)/mCBP (6 nm)/15 wt% TADF:mCBP (20 nm)/Tris-PCz (20 nm)/Al (100 nm)	24.9%	—	125
TXO-PhCz	−5.78	−3.58	—	0.02	305, 345	570^a	—	—	99^b	ITO/PEDOT (30 nm)/TAPC (20 nm)/EML (35 nm)/TmPyPB (55 nm)/LiF(0.9 nm)/Al, where poly(3,4-ethylenedioxythiophene) (PEDOT)	21.5%	0.31,0.56	126
2,7 TXO-PhCz	−5.78	−3.49	0.01	0.01	350, 450	550^a/530^b	25.9^b	63.1^b	61^b/89.02^a	(ITO) (95 nm)/(TAPC) (35 nm)/CBP: x wt% emitters(35 nm)/(TmPyPB) (45 nm)/LiF(1 nm)/Al (100 nm)	24.4%	0.39, 0.55	127
CzDBA	−5.93	−3.45	0.03	0.03	342	532^a	34^b	3.2^b	100^b/90.6^a	ITO/HAT-CN (10 nm)/Tris-PCz (30 nm)/mCBP:CzDBA (10%) (30 nm)/T2T (10 nm)/BPy-TP2 (40 nm)/LiF (1 nm)/Al (100 nm)	37.8%	0.31, 0.61	128
tBuCzDBA	−5.88	−3.49	0.02	0.02	348	553^a	36^b	2.1^b	86^b/84^a	ITO/HAT-CN (10 nm)/Tris-PCz (30 nm)/mCBP:tBuCzDBA (10%) (30 nm)/T2T (10 nm)/BPy-TP2 (40 nm)/LiF (1 nm)/Al (100 nm)	37.8%	0.37, 0.60	129
SQ-omeTPA	−5.25	−3.06	—	0.17	516	653^a	6.5^b	299.5^b	50.2^a/66.8^b	ITO/HAT-CN(5 nm)/TAPC(40 nm)/TCTA (5 nm)/DMFL-CBP: emitter (3%) (30 nm)/TmPyPB (40 nm)/LiF (1 nm)/Al (80 nm)	19.1%	0.60, 0.40	130
tBuTPA-QX4CN	−5.77	−4.24	—	0.10	350/565	678^a	11.7^b	0.3^b	65^b	ITO (50 nm)/PEDOT:PSS (60 nm)/TAPC (20 nm)/TCTA:TPBi:tBuTPAQx4CN (25 nm: 50 nm: X%)/TSPO1 (5 nm)/TPBi (40 nm)/LiF (1.5 nm)/Al (200 nm)	16.3%	0.60, 0.36	129
POZ-DBPHZ	−5.36	−3.38	—	0.08	463	521^a/595^b	—	—		ITO/NPB (40 nm)/10 wt% DBPHZ in CBP (20 nm)/TPBi (20 nm)/BCP (20 nm)/LiF (1 nm)/Al (100 nm)	16%	—	131
BPPz-PXZ	−5.39	−3.22	—	0.03	396/450	607^a	18.4^b	3.6^b	100^b	ITO/TAPC (35 nm)/TCTA (10 nm)/mCP (10 nm)/emitters (20 nm)/TmPyPB (45 nm)/LiF (1 nm)	25.2%	0.57, 0.43	132
DPQAO-Icz	−5.9	—	0.29	—	488	504/508	7.1^b	266^b	92^b	ITO/TAPC (40 nm)/TCTA (10 nm)/mCP (10 nm)/emitters (20 nm)/TmPyPB (40 nm)/LiF (1 nm)/Al (100nm)	31.6%	0.23, 0.68	133
IPXZ-2	−5.18	−2.78	—	0.05	400	500/508	51^b/65^b	2^b/1.7^b		ITO/HAT-CN (7 nm)/TAPC (30 nm)/mCP (10 nm)/emitters (20 nm)/TmPyPB (30 nm)/LiF (1 nm)/Al (100nm)	30.6%	0.13, 0.45	134
Compound 1	−6.26	−3.68	—	—	470	635	3.02^b	1.21^b		fITO/PEDOT:PSS/2PACz/mCPCN: compound 1/CN-T2T (50 nm) LiF (1 nm)/Al (150 nm)	6.4%	0.53, 0.46	135
Ac-CNBPzBr	—	—	—	—	450	658	42.3^b	2.05^b		ITO (100 nm)/HAT-CN (5 nm)/TAPC (70 nm)/CBP:TADF emitters (20 nm)/T2T (3 nm)/TPBi (0.8 nm) LiF (1 nm)/Al (70 nm)	10.5%	—	136

---

3.2.2 Red TADF emitters. In comparison to blue and green emitters, red TADF emitters remain a relatively less developed area due to the intrinsic challenges in achieving high ϕ_PL in the red spectral region. This difficulty is largely attributed to the energy gap law, which indicates that as the energy difference between the excited and ground states narrows, the number of accessible vibrational states in both levels increases.¹³⁷ This elevated vibronic density facilitates faster internal conversion from S₁ to S₀, enhancing the nonradiative decay pathways. Additionally, given that the radiative decay rate is proportional to the cube of the transition frequency, a reduced S₁–S₀ gap results in a lowered radiative rate constant. Consequently, designing red and especially deep red to near-infrared (NIR) emitters with high ϕ_PL becomes inherently more complex. Moreover, achieving these low-energy excited states generally requires larger π-conjugated frameworks, which heightens the likelihood of π–π stacking interactions, often causing ACQ.¹³⁸ Due to these compounded factors, progress in developing efficient red TADF emitters has lagged behind that of blue and green systems.

The successful triplet exciton up-conversion in TADF systems has been confirmed by multiple studies, with several green and blue TADF OLEDs now achieving EQEs exceeding 40%, following earlier milestones that demonstrated nearly 30% EQE in green devices. Since then, the performance of green and blue TADF emitters has advanced significantly due to the rational design of both emitters and host materials.¹³⁹ Through continuous material innovation and device optimization, not only has the EQE been enhanced, but the operational lifetimes of TADF OLEDs have also been extended. At present, the EQEs of state-of-the-art green and blue TADF OLEDs are on par with that of phosphorescent OLEDs, although their device lifetimes remain comparatively shorter.

In addition, NIR TADF emitters are being explored using design strategies such as those employed for red TADF systems; however, their device efficiencies remain significantly lower than that of TADF OLEDs operating in the visible spectrum.¹⁴⁰ Unlike blue and green TADF systems, red and NIR emitters necessitate lower singlet excited-state energies to achieve long-wavelength emission.¹⁴¹

One of the primary challenges lies in fine-tuning the energy level alignment between the singlet charge-transfer (¹CT) state and the triplet local excited (³LE) state to boost the emission efficiency. In 2024, this issue was addressed by designing three novel D–A TADF molecules (Fig. 37a), TQ-oMeOTPA, TsQ-oMeOTPA, and SQ-oMeOTPA,¹³⁰ based on the previously reported green-emitting compound 67dTPA-FQ.¹⁴² These new emitters incorporated 4,4′-dimethoxytriphenylamine (MeOTPA) as the donor and featured sulfur atoms within the acceptor moieties to enhance both the SOC and CT characteristics. Computational studies confirmed that the inclusion of MeOTPA and sulfur significantly strengthened the CT interaction, which not only shifted the emission towards the deep-red spectral region but also facilitated more favorable excited-state energy alignment. This tuning promoted a more efficient RISC process. Among the three compounds, SQ-oMeOTPA exhibited the best device performance, delivering an EQE of 19.1% with the emission maximum at 619 nm (Fig. 37c and e). SQ-oMeOTPA shows the highest efficiency because the sulfone group introduces strong electron-withdrawing ability and rigid molecular orientation.

Fig. 37 (a) Chemical structure of the three red TADF emitters. (b) Device structure. (c) EL spectra of the devices; (d) current density–voltage–luminance (J–V–L) curves of the devices; (e) EQE vs. luminance relationships of the devices. Adapted with permission.¹³⁰ Copyright 2024, the Royal Society of Chemistry.

The comprehensive photophysical analyses of the emitters were followed by the evaluation of their EL characteristics using DMFL-CBP (9,9′-(9,9-dimethyl-9H-fluorene-2,7-diyl)bis(9H-carbazole)) as the host matrix. The OLED architecture used for these studies consisted of ITO/HAT-CN (5 nm)/TAPC (40 nm)/TCTA (5 nm)/DMFL-CBP: emitter (3%) (30 nm)/TmPyPB (40 nm)/LiF (1 nm)/Al (80 nm) (Fig. 37b). In this structure, HAT-CN and lithium fluoride (LiF) served as the hole and electron injection layers, respectively. The layers responsible for charge transport were 1,1-bis[(di-4-tolylamino)phenyl]cyclohexane (TAPC) for holes and 1,3,5-tri[(3-pyridyl)-phen-3-yl]benzene (TmPyPB) for electrons, while 4,4′,4′′-tri(N-carbazolyl)triphenylamine (TCTA) functioned as the electron-blocking layer. To counteract the pronounced ACQ often observed in red-emitting materials arising from intensified dipole–dipole interactions and π–π stacking that enhance the non-radiative decay pathways, the doping concentration of each emitter was kept low at 3%.

In 2024, a new molecular architecture for a TADF emitter named tBuTPA-Qx4CN was introduced, utilizing the 2,3-bis(4-cyanophenyl)quinoxaline-6,7-dicarbonitrile (Qx4CN) core as a potent acceptor (Fig. 38a).¹²⁹ This design strategy focused on achieving deep-red to NIR emission by connecting donor groups in a para-orientation relative to the Qx4CN unit. Rather than the commonly used peripheral donor attachment, they connected two identical donor fragments, 4-(tert-butyl)-N-(4-(tert-butyl)phenyl)-N-phenylaniline, at the 5,8-positions of the central acceptor. The intramolecular hydrogen bonding interactions between the D and A fragments enhanced the electronic coupling, leading to an increased oscillator strength and a high radiative decay rate. The strong electron-withdrawing nature of the Qx4CN acceptor enabled a significant bathochromic shift in emission, positioning the emission peak of tBuTPA-Qx4CN in the NIR region. The OLEDs employing this emitter displayed an impressive EQE of up to 16.3%, with EL peaking at 674 nm (Fig. 38b and d). This is because the additional 4-cyano groups strengthen the electron-withdrawing ability for deeper CT character, while the bulky tert-butyl groups enforce rigid molecular orientation and suppress aggregation quenching. Together, these features enhance the exciton utilization and lead to superior DR/NIR TADF efficiency compared to simpler analogues. Additionally, a novel synthetic method was developed for the construction of tBuTPA-Qx4CN, employing 4,5-diamino-3,6-dibromophthalonitrile as the starting material.

Fig. 38 (a) Chemical structure of the red TADF emitters. (b) UV-Vis absorption spectra and PL spectra in toluene solvent at room temperature. (c) Transient PL decay profile of the 1 wt% tBuTPA-Qx4CN emitter-doped thin film measured. (d) EQE–luminance characteristics. Adapted with permission.¹²⁹ Copyright 2024, the Royal Society of Chemistry.

In 2016, a new class of TADF emitters was reported featuring a U-shaped D–A–D molecular framework incorporating a novel acceptor core. These compounds exhibited small ΔE_ST ranging from 0.02 to 0.20 eV, contributing to their efficient TADF behavior. The key factor driving TADF in these systems was identified as the lowest triplet state localized on the acceptor unit. The OLEDs fabricated using these materials demonstrated an excellent device performance, achieving EQEs of up to 16%.¹³¹

To evaluate the triplet harvesting efficiency of these TADF emitters, OLEDs labeled DEV1–DEV3 were constructed with the device configuration of ITO/NPB (40 nm)/10 wt% DBPHZ-based emitter in CBP (20 nm)/TPBi (20 nm)/BCP (20 nm)/LiF (1 nm)/Al (100 nm), as shown in Fig. 39. Among the devices, the one incorporating POZ-DBPHZ (DEV3) showed the highest performance, with an EQE reaching 16% (Fig. 39c). POZ-DBPHZ shows a higher efficiency because its phenoxazine donor enforces stronger HOMO–LUMO separation and a smaller ΔE_ST, enabling faster RISC, while also minimizing the long-lived delayed fluorescence and exciton quenching. In contrast, the devices based on t-BuCZ-DBPHZ and MeODP-DBPHZ yielded significantly lower efficiencies, roughly half that of DEV3. This reduction in performance was attributed to two main factors, the larger ΔE_ST in comparison to POZ-DBPHZ, and prolonged delayed fluorescence lifetimes, which may lead to increased exciton quenching through charge interactions. Additionally, in the carbazole-containing emitter, triplet–triplet annihilation (TTA) may further suppress the device efficiency. Notably, the POZ-DBPHZ-based device also exhibited a much higher luminance, exceeding 35 000 cd m⁻², which was more than twice that of the other two devices.

Fig. 39 (a) Structures of the investigated DBPHZs. (b) and (c) Device characteristic. (d) Temperature-dependent spectra of DF in 10 wt% POZ-DBPHZ:CBP blended film. Adapted with permission.¹³¹ Copyright 2016, John Wiley and Sons.

In 2019, two new red-emitting TADF materials, 10-(dibenzo[a,c]dipyrido[3,2-h:2′,3′-j]phenazin-12-yl)-10H-phenoxazine (BPPZ-PXZ) and 10-(11,12-di(pyridin-3-yl)dibenzo[a,c]phenazin-3-yl)-10H-phenoxazine (mDPBPZ-PXZ), were designed by carefully balancing molecular rigidity and intermolecular packing (Fig. 40a).¹³² Both compounds exhibited distinct TADF characteristics, featuring the very small ΔE_ST of 0.03 eV and 0.04 eV, respectively. BPPZ-PXZ employs a highly rigid BPPZ acceptor framework with only one rotatable C–N bond, which effectively suppresses the vibrational relaxation and minimizes non-radiative losses. As a result, it achieved a near-unity ϕ_PL = 100% ± 0.8% and an EQE of 25.2%, emitting at 604 nm, setting a record for red TADF OLEDs with emission above 600 nm (Fig. 40b and c). In contrast, mDPBPZ-PXZ incorporates pyridine substituents on its acceptor, introducing additional rotational freedom that reduces rigidity and increases non-radiative decay, thereby lowering EQE to 21.7% in the doped devices (Fig. 40d). Thus, the superior molecular orientation and restricted conformation of BPPZ-PXZ ensure more efficient exciton harvesting and account for its higher EQE compared to mDPBPZ-PXZ. However, the increased steric bulk in mDPBPZ-PXZ proved advantageous in the non-doped device architectures, enabling deep red/NIR with an EQE of 5.2%, which is relatively high for this type of system. This study not only introduces two highly efficient red emitters for both doped and non-doped OLED configurations but also highlights an effective design strategy to tune molecular structures for enhanced red/NIR TADF performances.

Fig. 40 (a) Chemical structures of red TADF emitters. (b) EL spectra of the devices. (c) and (d) EQE characteristics. Adapted with permission.¹³² Copyright 2019, John Wiley and Sons.

Meanwhile, there is growing interest in deep-red and NIR OLEDs for specialized applications such as bioimaging,¹⁴³ phototherapy, and optical communication. In the case of devices emitting beyond 650 nm, achieving high efficiency remains difficult, with most reported TADF OLEDs exhibiting EQEs below 5%, a value even lower than the theoretical maximum for traditional fluorescent OLEDs. Therefore, it remains a significant challenge to develop efficient red TADF emitters that can perform well in non-doped device formats for deep red/NIR emission applications.

Exploring unconventional emissive channels has emerged as a promising way to unlock new functionalities in organic systems. A recent work demonstrated that the rarely investigated sulfur lone pair (n) orbital can act as an efficient emissive contributor, where correlated π*→ n transitions enable strong TADF suitable for OLEDs.¹³⁵ The molecular design employs a persulfide aromatic spirocycle to enhance the spin–orbit coupling and promote intersystem crossing, while a twisted donor–acceptor configuration bridged by spiro[4.4]nonane, together with the orthogonal spatial orientation of sulfur lone pairs and π* orbitals, reduces ΔE_ST (Fig. 41a). The rigid spirocyclic backbone further suppresses nonradiative deactivation, ensuring efficient excited-state utilization.

Fig. 41 (a) Structures of compound 1 and reference compounds 2–4. (b) PL spectra in DCM (10 μM). (c) Energy level diagram of the OLED with mCPCN: compound 1 as the emitter. (d) Current–voltage–luminance plots. (e) PL lifetime decays in degassed DCM (10 μM). (f) EQE and power efficiency versus current density. (g) EL spectra of optimized devices at different voltages. Adapted with permission.¹³⁵ Copyright 2025, the American Chemical Society.

The proof-of-concept emitter, compound 1, showed pronounced n-π* TADF with an emission maximum at 635 nm and a ϕ_PL of 52% in CH₂Cl₂, supported by steady-state and time-resolved PL spectra (Fig. 41b and e). When incorporated into OLEDs with an mCPCN: compound 1 emitting layer (Fig. 41c), the devices produced EL near 600 nm and delivered a maximum EQE of 6.4% at 189 cd m⁻². The spirocyclic backbone enforces an orthogonal donor–acceptor orientation, effectively suppressing HOMO–LUMO overlap and nonradiative decay. At the same time, the twisted, rigid framework enhances SOC from multiple S atoms, enabling faster ISC/RISC and higher TADF efficiency. Device measurements confirmed the stable current–voltage–luminance characteristics (Fig. 41d) as well as favorable EQE and power efficiency behavior as a function of current density (Fig. 41f). Moreover, the EL spectra at varying operating voltages demonstrated consistent red emission output (Fig. 41g). Collectively, these results highlight sulfur lone pair–assisted π*-n transitions as a distinctive molecular strategy, providing a new design avenue for practical n-π* TADF-based OLEDs.

A recent study reported how π–π stacking interactions can modulate the heavy-atom effect and SOC through dimer formation in solid-state organic emitters. In the red TADF emitter Ac-CNBPz, dimerization leads to distinct packing arrangements (Fig. 42a). In its brominated analogue Ac-CNBPzBr, the proximity of the Br atom to the electronic density of the dimer strongly enhances the SOC, yielding up to a 200-fold increase due to the external heavy-atom effect (EHAE). This effect accelerates RISC, enabling TADF within a few microseconds, which is nearly 20 times faster than that in the non-brominated system. The OLED devices incorporating Ac-CNBPzBr as the emitter, with an assistant dopant, exhibited an improved performance, including reduced efficiency roll-off by factors of 4 and 1.5, respectively (Fig. 42b and c). The EL spectra confirmed the stable red emission (Fig. 42d), while the optimized device stack design further contributed to enhanced stability (Fig. 42e). These findings highlight how favorable dimer geometries and intermolecular interactions with halogen atoms provide key design principles for developing robust all-organic TADF emitters.¹³⁶Table 2 presents a detailed summary of the photophysical properties and device performance of red TADF emitters.

Fig. 42 (a) Molecular structures of the studied emitters. (b) Current density–voltage–luminance (J–V–L) characteristics. (c) EQE versus luminance plots. (d) EL spectra. (e) Schematic of the OLED device stack. Adapted with permission.¹³⁶ Copyright 2025, the American Chemical Society.

3.3 Development of white-light and full-color emitters

Recently, the OLED field has advanced significantly, leading to its commercial use in cell phones, televisions, and lighting applications. Therefore, to meet the demands for energy-efficient lighting and high-quality displays, increasing attention has been directed towards white OLED (WOLED) technology. In 1994, the foundational research on WOLEDs was initiated, marking a significant milestone in the development of OLED technology.¹⁴⁴ Over the past few decades, the PE of WOLEDs has seen substantial progress, rising from 0.83 lm W⁻¹ to over 100 lm W⁻¹.^145,146 This advancement underscores the promise of remarkable next-generation WOLED for displays and lighting. White emission can be achieved either by combining complementary colors (such as orange and blue) or by mixing primary colors such as red, green, and blue to cover the entire visible spectrum, ranging from 380 nm to 780 nm. The limited spectral bandwidth of individual emitters often necessitates the use of multiple emitters within a single device. Thus, careful selection of well-matched materials and optimization of a judicious device configuration are highly essential to achieve efficient and balanced WOLED devices.¹⁴⁷ Thus far, various types of WOLEDs have been developed, including all-phosphorescent,¹⁴⁸ all-fluorescent,¹⁴⁹ TADF based,¹⁵⁰ and hybrid white light.^151,152 Unlike traditional emitting layers in WOLEDs, which can harness only singlet excitons, doping with TADF materials significantly boosts the EL efficiency compared to conventional host systems. In addition, fully fluorescent WOLEDs incorporating TADF materials demonstrate excellent competitiveness and advantages as hosts and sensitizers, offering high color purity and enhanced stability for fluorescent emitters.⁹ This progress highlights the promise and advantages of full-fluorescent TADF-based WOLEDs over their traditional counterparts.¹⁹ Owing to their excellent properties, such as metal-free characteristics, high EQE, bright emission, potentially long lifetime, and low power consumption, organic TADF emitters have been actively explored for the development of WOLEDs.¹⁵³ Although TADF-based WOLEDs have only been reported in recent years, their performance has steadily improved over time.¹⁵⁴ Thus far, TADF-based WOLEDs have delivered external quantum efficiencies approaching 30%, which are comparable to that of state-of-the-art phosphorescent WOLEDs and fluorescence/phosphorescence hybrid WOLEDs.¹⁵⁵ Therefore, TADF-based WOLEDs hold great potential for applications in lighting and display technologies. As our understanding of TADF has steadily improved, several approaches have been proposed for developing TADF-based WOLEDs. They include combining TADF with traditional fluorescent emitters, mixing TADF with phosphorescent emitters, introducing TADF exciplex hosts paired with phosphorescent dopants, all-TADF emitters, and single molecular TADF emitters.

3.3.1 TADF with phosphorescent emitters. Given that both TADF and phosphorescent emitters are capable of harvesting singlet and triplet excitons, combining them offers an effective strategy for constructing WOLEDs.¹⁵⁶ Notably, there has been growing interest in mixing blue TADF emitters with green, red, or complementary phosphorescent emitters to enhance the device performance. Blue TADF materials are particularly favorable because they naturally exhibit high triplet energies, owing to their narrow singlet–triplet energy gaps, and can deliver high efficiency.¹⁵⁷ However, phosphorescent materials typically contain heavy metal atoms, such as Os, Pd, Pt, and Ir, in their organic metal complexes.¹⁵⁸ Compared to traditional luminescent materials, the singlet S₁ and triplet T₁ excitons can be effectively utilized by the phosphor through the heavy atom effect, resulting in a potential IQE of 100% and enabling the realization of high-performance OLEDs.¹⁵⁹ Furthermore, due to their high cost, fluorescence-based materials are less feasible for mass production, making phosphorescent materials the primary focus of experimental research. Building on this concept, the first hybrid WOLED was demonstrated in 2014 by combining a blue TADF emitter (2CzPN) with an orange phosphorescent dopant (PO-01) (Fig. 43a).¹⁶⁰ The resulting device delivered warm white emission, achieving an EQE of 22.6% and a forward-viewing PE as high as 47.1 lm W⁻¹. The WOLED structures were carefully engineered to exploit the electron-trapping capability of 2CzPN, ensuring stable warm white emission. The outstanding performance of the device can be attributed to several key factors. Firstly, the triplet excitons generated on 2CzPN could either transfer energy to the lower-lying triplet states of the phosphorescent emitter PO-01 or contribute to emission through recombination. Secondly, mCP was chosen as the host for 2CzPN because its wide energy gap and high triplet energy (T₁) of 3.0 eV are crucial, considering that the host material plays a pivotal role in governing the efficiency of TADF systems.

Fig. 43 (a) (i) Energy level diagram of the materials used in the hybrid WOLED. The grey rectangle marks the main exciton zone. Solid and dashed lines indicate the HOMO and LUMO levels, while circles and diamonds represent their exciton energies, respectively. PF and DF denote prompt and delayed fluorescence, respectively. (ii) Chemical structure of materials used for device fabrication. (b) (i) Chemical structure of TADF and fluorescent emitters used in the device. (ii) Schematic representation of the proposed energy transfer mechanism from TADF emitter to TTPA and DBP under electrical excitation. (iii) EL spectra of fabricated WOLEDs. Adapted with permission.^160,161 Copyright 2013, the Royal Society of Chemistry. Copyright 2015, John Wiley and Sons.

3.3.2 TADF with traditional fluorescent emitters. Given that TADF materials can utilize triplet excitons and hybrid WOLEDs that pair blue fluorescence with green and red phosphorescent emitters have already shown excellent performances, it is expected that WOLEDs featuring a TADF-based exciton alongside complementary color fluorescent emitters are anticipated to achieve an IQE of 100%.¹⁶² In simpler terms, a TADF molecule functions as a triplet exciton harvester for other-color fluorescent emitters to produce white emission. By employing this strategy, in 2015, a TADF WOLED with an impressive η_EQE of over 12% and CIE coordinates of (0.25, 0.31) was obtained (Fig. 43b).¹⁶¹ The device incorporates a blue TADF molecule (DMACDPS) as a common exciton donor, along with red (TTPA) and green (DBP) conventional fluorescent molecules as exciton acceptors, demonstrating that the appropriate device architecture can ease the limitations on fluorescent molecule selection (Fig. 43b(i)). To achieve white-light emission, a TAF (TADF-assisted fluorescence) process was employed, wherein the singlet excitons on DMACDPS molecules generated either directly or via RISC are resonantly transferred to the TTPA and DBP acceptors through a 2 nm thick mCP spacer layer (Fig. 43b(ii)). This approach eliminates direct carrier recombination on fluorescent molecules, while enabling their delayed emission through the TAF process.

In 2022, a single-layer WOLED was fabricated, achieving an EQE of 30.7% and a PE of 120.2 lm W⁻¹, marking the highest performance recorded among pure organic WOLEDs.¹⁶³ Three blue TADF emitters, ptBCzPO2TPTZ, 2CzPN, and DMACDPS, were respectively employed as blue-emitting sensitizers to fabricate hyperfluorescent white OLEDs in combination with the conventional yellow fluorescent emitter TBRb (Fig. 44a(i)). The ptBCzPO2TPTZ-based device outperformed those using 2CzPN and DMAC-DPS, with the remarkable performance attributed to its promotion of radiative decay and suppression of non-radiative losses, which together underpin the advanced efficiency of its white fluorescent OLED. This study demonstrated that hyperfluorescent white-emitting systems can realize complete exciton utilization, provided that triplet DET and diffusion-induced quenching are efficiently inhibited (Fig. 44a(ii) and (iii)). Likewise, in 2025, a highly efficient and color-stable WOLED was developed, achieving an EQE of 23.8% along with a purer white emission characterized by CIE coordinates of (0.33, 0.43) and a CCT of 5863 K.¹⁶⁴ Here, the device incorporates a fluorescent emitter, p-DTAACN, producing orange emission, alongside the previously reported sky-blue TADF emitter 3BPy-mDTC (Fig. 44b). The incorporation of these two complementary emitters enabled the development of single-emissive-layer WOLEDs, with the ability to shift the white emission from a cooler to warmer tone by carefully adjusting the p-DTAACN doping concentration between 0.5 wt% and 1.0 wt%.

Fig. 44 (a) (i) Molecular structure of TADF sensitiser and fluorescent emitter utilized in device fabrication. (ii) Primary transitions and energy transfer pathways between the blue TADF sensitizer and the yellow fluorescent emitter. (iii) Device structure of WOLED. (b) Schematic depicting the energy transfer mechanism for white light generation through concurrent emissions from the TADF sensitizer and fluorescent emitter. Adapted with permission.^163,164 Copyright 2022, Science and Technology Review Publishing House. Copyright 2013, the Royal Society of Chemistry.

3.3.3 TADF exciplexes. High-performance WOLEDs can be achieved by employing high-T₁ exciplexes as hosts for phosphorescent dopants, give that their T₁ is higher than that of the dopants. These high-T₁ exciplexes can be obtained by combining high-T₁ p-type and n-type materials. Typically, TADF exciplexes are well-suited to fulfil the high T₁ requirement.¹⁴⁶ As a result, employing a TADF exciplex host not only minimizes the charge injection barriers but also ensures efficient TADF up conversion, owing to its intermolecular donor–acceptor architecture and small S₁–T₁ energy gap. Hence, TADF exciplex hosts offer a potential compromise between achieving 100% IQE, effective exciton confinement within the EML and low operating voltage.¹⁶⁵

In 2017, an ultra-efficient WOLED was obtained by utilizing a novel blue TADF exciplex host for the phosphorescent dopant (Fig. 45a).¹⁴⁶ The fabricated WOLED exhibited a notably high forward-viewing LE of 105.0 lm W⁻¹ along with an EQE of roughly 30%. These impressive efficiencies and superior device performance are primarily due to the benefits of employing exciplex materials as sole hosts. Additionally, the well-confined excitons effectively eliminate leakage from the emission zone, thereby enhancing the efficacy.

Fig. 45 (a) (i) Chemical structure of mCP and B4PyMPM. (ii) Formation mechanism of the exciplex in the mCP:B4PyMPM blend film. (iii) Device architecture of the WOLED device. (b) (i) Structural representations of mCP and PO-T2T. (ii) Device structure diagram of the WOLED. Adopted with permission.^146,166 Copyright 2017, John Wiley and Sons. Copyright 2023, Elsevier.

Recently, in 2023, a simple yet efficient monochrome/WOLED was developed by integrating a TADF interfacial exciplex with phosphorescent ultrathin emitting layers (Fig. 45b(i)).¹⁶⁶ In particular, the device was constructed by directly inserting a single or complementary phosphorescent ultrathin emitting layer (Ph-UEML) at the interface of the TADF interfacial exciplex (mCP/PO-T2T) through a straightforward, doping-free technique (Fig. 45b(ii)). The resulting two-, three-, and four-color WOLEDs exhibited an excellent performance, featuring a low turn-on voltage of 2.5 V and high EQEs of 23.47%, 22.70%, and 23.88%, respectively. Moreover, all the white devices maintained outstanding color stability across a practical luminance range of approximately 5000 to 12 000 cd m⁻². This study reveals that the primary factor behind the high device performance is the sensitization of Ph-UEMLs by a TADF interfacial exciplex, which enables complete exciton utilization through multi-channel energy transfer from the exciplex to the Ph-UEMLs.

3.3.4 All-TADF emitters. To develop WOLEDs with all-TADF emitters, the simplest strategy involves using TADF exciplex hosts and all-TADF materials as the blue, green, and red emitters. For instance, in 2014, this strategy was reported to achieve first metal-free TADF-based WOLED with an EQE of 17.1% with CIE coordinates of (0.30, 0.38) operating at 3.6 V.¹⁵⁴ In the device architecture, the TADF emitters 3CzTRZ, 4CzPN, and 4CzTPN-Ph were incorporated as blue, green, and red emitters, respectively, and arranged in three adjacent stacked layers (Fig. 46a). It was observed that the distinct triplet exciton lifetimes of the TADF emitters caused each layer to undergo different degrees of exciton annihilation, including singlet–triplet annihilation. This variation primarily arises from the differences in their k_RISC values, which are determined by their inherent ΔE_ST. As a consequence, a gradual shift in the EL spectrum was observed with an increase in current density.

Fig. 46 (a) Structures of the organic materials used in the OLED fabrication (i)–(iii) and energy level diagram of fabricated WOLEDs (iv). (b) Schematic representation of the energy level alignment in the devices (i) and chemical structures (ii) of the charge-transporting materials along with the green and yellow TADF emitters explored in WOLED fabrication. Adapted with permission.^15,154 Copyright 2014, AIP Publishing. Copyright 2013, the Royal Society of Chemistry.

In a similar approach, two all-TADF WOLEDs were fabricated in 2018, using a light-blue TADF emitter (5CzOXD) combined with high-performance green (4CzCNPy) and yellow (4CzPNPh) TADF dopants in the emitting layer for devices A and B, respectively (Fig. 46b).¹⁶⁷ Both devices demonstrated a high maximum luminance exceeding 47 000 cd m⁻². The maximum current and power efficiencies reached 55.5 cd A⁻¹ and 42.7 lm W⁻¹ for device A, and 50.6 cd A⁻¹ and 64.7 lm W⁻¹ for device B, corresponding to maximum external quantum efficiencies (EQEs) of 16.2% and 17.1%, respectively.

3.3.5 Single-molecular TADF. To date, WOLEDs have demonstrated numerous advantages, but their cost remains a major obstacle in their commercial adoption.¹⁶⁹ The reasons for this include: (i) more complex structures of WOLEDs compared to monochrome OLEDs; (ii) the common requirement for multiple molecular emitters, and (iii) the dependency on rare and expensive noble metal complexes such as Ir and Pt to achieve high performance. An effective way to ease this barrier is by simplifying WOLED designs.^19,170 This can be achieved by minimizing the number of molecular emitters, making the device structures less complex. In this context, doping-free TADF WOLEDs present a promising solution. For the first time, in 2018, an extremely simplified yet high-performance TADF WOLED was reported.¹⁶⁸ Unlike conventional devices that typically require at least two molecular emitters to achieve high-quality white emission, this device employed a single molecular emitter. An intrinsic TADF emitter (DDCzTrz) was positioned between p-type (TAPC) and n-type (TmPyPB) layers, resulting in a doping-free p–i–n WOLED (Fig. 47). The WOLED delivered high-quality white emission, achieving a maximum CRI of 91, along with an impressive EQE of 28.4%, a current efficiency of 65.4 cd A⁻¹, and a PE of 68.5 lm W⁻¹. The achieved results are comparable to that of phosphorescent and multi-emitter-based WOLED devices. A comprehensive summary of the photophysical properties and device performance of white-light TADF emitters is provided in Table 3.

Fig. 47 (a) Schematic structure of p–i–n WOLEDs. (b) Chemical structures of TAPC, TmPyPB, and DDCzTrz. (c) Illustration of the emission mechanism in p–i–n WOLEDs, where red and black arrows indicate holes and electron transport, respectively. The synergy of exciplex and TADF emissions contributes to the generation of white light. Adapted with permission.¹⁶⁸ Copyright 2018, the American Chemical Society.

Table 3 Photophysical and computational parameters of white and full color TADF emitters and related OLED device configurations with EL device performance

Emitter	HOMO (eV)	LUMO (eV)	Theo. ΔEST (eV)	Exp. ΔEST (eV)	λ max, abs (nm)	λ max, PL (nm)	τ P (ns)	τ d (μs)	ϕ [%]	Device structure	EQE (%)	CIE	Ref.
2CzPN	—	—	—	—	—	480	—	—	—	ITO/HATCN (5 nm)/NPB (40 nm)/TCTA (10 nm)/mCP:2CzPN (11 nm)/TAZ: 4 wt% PO-01 (4 nm)/TAZ (40 nm)/LiF (0.5 nm)/Al (150 nm)	22.6%	(0.45, 0.48)	160
DMACDPS	−5.9	−2.9	—	—	337	475	—	4	74	ITO (100 nm)/α-NPD (30 nm)/[1 wt% DBP:10 wt% TTPA:mCP	12%	(0.25, 0.31)	161
ptBCzPO2TPTZ	—	—	—	—	—	—	—	—	94	ITO|mCP|DBFDPO:ptBCzPO2TPTZ:TBRb|pTPOTZ|LiF|Al	30%	(0.30, 0.42)	163
3BPy-mDTC	−5.6	−2.7	0.05	—	—	478	—	—	92	ITO/NPB (30 nm)/TAPC (20 nm)/mCBP:7 wt% 3BPy-mDTC:0.7 wt% p-DTAACN (10 nm)/PPT (10 nm) TPBi (50 nm)/Liq (2 nm)/Al (100 nm)	23.8%	(0.33, 0.43)	164
mCP:B4PyMPM	−7.1	−3.5	—	—	—	—	45	0.55		ITO (110 nm)/TAPC (40 nm)/TCTA (10 nm)/mCP (10 nm)/mCP:50 wt% B4PyMPM:15 wt% FIrpic:0.2 wt% PO-01 (20 nm)/B4PyMPM (50 nm)/Liq (0.8 nm)/Al (120 nm)	28.1%	(0.40, 0.48)	146
mCP/PO-T2T	−6.10	−3.50	0.03	—	—	466	0.47	2.81		ITO(180 nm)/MoO3(3 nm)/TAPC(40 nm)/mCP (10 nm)/FIrpic(0.40 nm)/Ir(BT)2(acac)(x nm)/PO-T2T(50 nm)/LiF(1 nm)/Al(100 nm)	23.47%, 22.70%, 23.88%	(0.348,0.431), (0.426, 0.491), (0.437,0.424)	166
4CzTPN-Ph, 4CzPN, 3CzTRZ	—	—	0.2	—	—	580, 538, 452	—	—	26, 74, 68	ITO/HATCN (10 nm)/Tris-PCz (35 nm)/10 wt% 4CzPN:mCBP (G-EML) (x nm)/6 wt% 4CzPN:2 wt% 4CzTPN-Ph:mCBP (R-EML) (y nm)/10 wt% 3CzTRZ:PPT (B-EML) (z nm)/PPT (50 nm)/LiF (0.8 nm)/Al (100 nm)	17.1%	(0.30, 0.38)	154
5CzOXD	−5.64	−2.74	0.22	—	290	496	—	13	58.2	ITO/MoO3 (8 nm)/TAPC (40 nm)/5CzOXD:4CzPNPh (0.6 wt%, 10 nm)/BmPyPB (40 nm)/LiF (1 nm)/Al	7.2%	(0.35, 0.44)	167
DDCzTrz	−6.01	−2.9	—	—	—	462, 553	—	—	66	ITO/MoO3 (5 nm)/1,3-bis(9H-carbazol-9-yl)-benzene (mCP, 30 nm)/mCP:DDCzTrz (1:10%, 10 nm)/TmPyPB (55 nm)/LiF (1 nm)/Al (120 nm)	28.4%	(0.34, 0.35)	168

---

Despite the remarkable progress in the efficiency of TADF-based OLEDs, their stability is still a major issue for their application in the practical world. The intrinsic design requirements of TADF emitters, such as a small singlet–triplet energy gap and strong charge-transfer character, often compromise their molecular rigidity and render them more prone to exciton-induced degradation. Blue TADF emitters are especially unstable due to their high excited-state energies, which accelerate bond cleavage and chemical decomposition. In devices, exciton–polaron and exciton–exciton annihilation are two more processes that shorten their operational lifetimes considerably more than phosphorescent systems.¹⁷¹ The excited-state lifetime is extended by the repetitive RISC process feature of TADF, which raises the possibility of exciton-induced degradation pathways in the host matrix as well as the emitter. Additionally, charge carrier imbalance at elevated current densities leads to localised Joule heating, which further exacerbates interfacial degradation and exciton quenching.¹⁷² Recent strategies, such as constructing MR-TADF frameworks, sterically protected D–A architectures, and optimized host–guest systems, have shown promise in making devices more stable, but their overall lifetimes still do not meet commercial needs. Thus, achieving a balance between high efficiency and long-term operational stability remains a critical challenge in the evolution of color-tunable TADF OLEDs.

4. Conclusion

The evolution of color-tunable TADF emitters has brought a paradigm shift in OLED technology, offering a highly efficient, metal-free platform for next-generation display and lighting applications. In the last ten years, the significant development in molecular design and enhanced comprehension of excited-state dynamics have contributed to the development of TADF materials that can precisely adjust the emission colors, improve the color purity, and increase the device efficiencies. The strategy started with D–A frameworks that made it easier to cross the energy gap between the singlet and triplet states, which made the RISC process more efficient. Later improvements, such as the strategic addition of π-spacers, made the emission range wider and made it possible to change the properties of the excited state in a systematic way. Fine-tuning through substituent effects, steric modifications, and positional isomerism provided further control over the emission wavelengths and device performance. Breakthroughs in TSCT and MR-TADF systems have enhanced the color purity and emission control, especially in the blue and deep-blue areas. Because of their remarkable color purity and narrowband emission, MR-TADF emitters in particular are ideal for high-definition display technologies. Moreover, the incorporation of chiral elements has resulted in the development of CPL-active TADF emitters at the molecular level and their successful integration into device-level CPEL, extending the range of TADF applications to spintronic devices, optical data storage, and 3D displays.

As a result of these molecular breakthroughs, TADF-based OLEDs have progressed from early prototypes to high-performance devices that cover the entire visible spectrum. Blue emitters, which used to be a big challenge, have improved considerably in terms of both efficiency and stability. Alternatively, green and red emitters have reached new performance benchmarks, making it possible to make full-color displays and white-light OLEDs with high color rendering indices. However, although efficient green emitters are now well-known, the development of deep-red TADF materials still lags behind with problems such as stability, emission broadening, and lower efficiencies still remaining unresolved. Furthermore, the development of multi-color and color-tunable TADF systems has opened pathways for dynamic lighting and smart display technologies. Emerging areas such as hyperfluorescence, AIE–TADF hybrids, room-temperature phosphorescence–TADF integration, and triplet-harvesting bio-applications are also under rapid development and promise to extend the applicability of TADF systems beyond OLEDs. Despite this remarkable progress, challenges such as long-term operational stability, especially for blue and deep-blue emitters, alongside the need for more efficient and stable red emitters, scalable synthesis, and integration into flexible and large-area devices remain unresolved. Addressing these hurdles will require continuous interdisciplinary collaboration across chemistry, materials science, and engineering.

In summary, the development of color-tunable TADF emitters shows how crucial molecular innovation is to the advancement of OLED technology. TADF materials have transformed OLEDs from their original application in basic donor-acceptor systems to complex full-color, high-efficiency devices across the visible spectrum by enhancing the device performance and color purity. Despite the significant progress, there are still challenges that need to be resolved. Improving the long-term operational stability, especially for blue and deep-blue emitters, and addressing the relatively slower progress in efficient, stable red TADF systems remain major barriers to their commercial use. Moreover, it is crucial to develop scalable and economical synthesis methods, along with the seamless incorporation of these materials into flexible, transparent, and wide-ranging device designs to fully realize their full potential. With sustained research, TADF-based OLEDs have the potential to significantly influence the advancement of high-performance, energy-efficient lighting and display applications in the future.

Conflicts of interest

The authors declare no conflict of interest.

Data availability

Since this is an Invited Review Article, there is no additional data supporting this article as all of the works are already published elsewhere, which have been included as references.

Acknowledgements

Financial grants from the Department of Science and Technology, New Delhi, India, project no. DST/CRG/2019/002614; Deity, India, no. 5(1)/2022-NANO; ICMR, Grant no. 5/3/8/20/2019-ITR; and Max-Planck-Gesellschaft, IGSTC/MPG/PG(PKI)/2011A/48, are acknowledged. T. S. thanks CSIR for the fellowship.

References

1. C. W. Tang and S. A. VanSlyke, Appl. Phys. Lett, 1987, 51, 913–915.
2. D. Thakur and S. Vaidyanathan, J. Mater. Chem. C, 2024, 12, 13168–13229.
3. G. Z. Dai, H. J. Li, Y. Z. Pan, X. Y. Dai and Q. Xie, Chin. Phys., 2005, 14, 2590–2594.
4. X. Yin, Y. He, X. Wang, Z. Wu, E. Pang, J. Xu and J.-A. Wang, Front. Chem., 2020, 8, 725.
5. P. Keerthika and R. K. Konidena, Adv. Opt. Mater., 2023, 11, 2301732.
6. L. S. Hung and C. H. Chen, Mater. Sci. Eng., R, 2002, 39, 143–222.
7. L. J. Rothberg and A. J. Lovinger, J. Mater. Res., 1996, 11, 3174–3187.
8. H. Wang, Y. Yuan, Z. Wang and Y. Wang, ACS Appl. Eng. Mater., 2024, 2, 781–810.
9. M. Y. Wong and E. Zysman-Colman, Adv. Mater., 2017, 29, 1605444.
10. A. Endo, K. Sato, K. Yoshimura, T. Kai, A. Kawada, H. Miyazaki and C. Adachi, Appl. Phys. Lett, 2011, 98, 083302.
11. D. Barman, K. Narang, R. Gogoi, D. Barman and P. K. Iyer, J. Mater. Chem. C, 2022, 10, 8536–8583.
12. H. Uoyama, K. Goushi, K. Shizu, H. Nomura and C. Adachi, Nature, 2012, 492, 234–238.
13. Y. Tao, K. Yuan, T. Chen, P. Xu, H. Li, R. Chen, C. Zheng, L. Zhang and W. Huang, Adv. Mater., 2014, 26, 7931–7958.
14. K. Shizu, Y. Sakai, H. Tanaka, S. Hirata, C. Adachi and H. Kaji, ITE Trans. Media Technol. Appl., 2015, 3, 108–113.
15. M. Moral, L. Muccioli, W. J. Son, Y. Olivier and J. C. Sancho-García, J. Chem. Theory Comput., 2015, 11, 168–177.
16. S. R. Forrest, D. D. C. Bradley and M. E. Thompson, Adv. Mater., 2003, 15, 1043–1048.
17. M. Fröbel, F. Fries, T. Schwab, S. Lenk, K. Leo, M. C. Gather and S. Reineke, Sci. Rep., 2018, 8, 9684.
18. R. Zhu, Z. Luo, H. Chen, Y. Dong and S.-T. Wu, Opt. Express, 2015, 23, 23680–23693.
19. S. Reineke, M. Thomschke, B. Lüssem and K. Leo, Rev. Mod. Phys., 2013, 85, 1245–1293.
20. C. M. Cole and S. D. Yambem, Adv. Opt. Mater., 2025, 13, 2402019.
21. T. Sarmah, S. Pal, D. Barman, D. Barman, B. G. Jaganathan and P. K. Iyer, Adv. Opt. Mater., 2025, 13, 2403060.
22. J. M. Dos Santos, D. Hall and B. Basumatary, et al. , Chem. Rev., 2024, 124, 13736–14110.
23. G. Tang, A. A. Sukhanov, J. Zhao, W. Yang, Z. Wang, Q. Liu, V. K. Voronkova, M. Di Donato, D. Escudero and D. Jacquemin, J. Phys. Chem. C, 2019, 123, 30171–30186.
24. Q. Zhang, B. Li, S. Huang, H. Nomura, H. Tanaka and C. Adachi, Nat. Photonics, 2014, 8, 326–332.
25. T. Zhao, S. Jiang, X.-D. Tao, M. Yang, L. Meng, X.-L. Chen and C.-Z. Lu, Dyes Pigm., 2023, 211, 111065.
26. M. Y. Wong, S. Krotkus, G. Copley, W. Li, C. Murawski, D. Hall, G. J. Hedley, M. Jaricot, D. B. Cordes, A. M. Z. Slawin, Y. Olivier, D. Beljonne, L. Muccioli, M. Moral, J.-C. Sancho-Garcia, M. C. Gather, I. D. W. Samuel and E. Zysman-Colman, ACS Appl. Mater. Interfaces, 2018, 10, 33360–33372.
27. S. K. Pathak, Y. Xiang, M. Huang, T. Huang, X. Cao, H. Liu, G. Xie and C. Yang, RSC Adv., 2020, 10, 15523–15529.
28. W.-L. Tsai, M.-H. Huang, W.-K. Lee, Y.-J. Hsu, K.-C. Pan, Y.-H. Huang, H.-C. Ting, M. Sarma, Y.-Y. Ho, H.-C. Hu, C.-C. Chen, M.-T. Lee, K.-T. Wong and C.-C. Wu, Chem. Commun., 2015, 51, 13662–13665.
29. H. Tanaka, K. Shizu, H. Miyazaki and C. Adachi, Chem. Commun., 2012, 48, 11392–11394.
30. W. Rodphon, C. Thongsornkleeb, J. Tummatorn and S. Ruchirawat, Chem. – Asian J., 2020, 15, 3475–3486.
31. S. Wang, Z. Cheng, X. Song, X. Yan, K. Ye, Y. Liu, G. Yang and Y. Wang, ACS Appl. Mater. Interfaces, 2017, 9, 9892–9901.
32. C. P. Keshavananda Prabhu, K. R. Naveen and J. Hur, Mater. Chem. Front., 2024, 8, 769–784.
33. P. Stachelek, J. S. Ward, P. L. dos Santos, A. Danos, M. Colella, N. Haase, S. J. Raynes, A. S. Batsanov, M. R. Bryce and A. P. Monkman, ACS Appl. Mater. Interfaces, 2019, 11, 27125–27133.
34. S. Hirata, Y. Sakai, K. Masui, H. Tanaka, S. Y. Lee, H. Nomura, N. Nakamura, M. Yasumatsu, H. Nakanotani, Q. Zhang, K. Shizu, H. Miyazaki and C. Adachi, Nat. Mater., 2015, 14, 330–336.
35. H. Wang, J.-X. Chen, X. Zhang, Y.-C. Cheng, X.-C. Fan, L. Zhou, J. Yu, K. Wang and X.-H. Zhang, Adv. Opt. Mater., 2023, 11, 2300368.
36. X. Liang, Z.-L. Tu and Y.-X. Zheng, Chem. – Eur. J., 2019, 25, 5623–5642.
37. F.-M. Xie, J.-X. Zhou, Y.-Q. Li and J.-X. Tang, J. Mater. Chem. C, 2020, 8, 9476–9494.
38. Y. Im, M. Kim, Y. J. Cho, J.-A. Seo, K. S. Yook and J. Y. Lee, Chem. Mater., 2017, 29, 1946–1963.
39. C. S. Oh, D. d S. Pereira, S. H. Han, H.-J. Park, H. F. Higginbotham, A. P. Monkman and J. Y. Lee, ACS Appl. Mater. Interfaces, 2018, 10, 35420–35429.
40. Y. Xiang, S. Gong, Y. Zhao, X. Yin, J. Luo, K. Wu, Z.-H. Lu and C. Yang, J. Mater. Chem. C, 2016, 4, 9998–10004.
41. W.-Y. Hung, P.-Y. Chiang, S.-W. Lin, W.-C. Tang, Y.-T. Chen, S.-H. Liu, P.-T. Chou, Y.-T. Hung and K.-T. Wong, ACS Appl. Mater. Interfaces, 2016, 8, 4811–4818.
42. S. Youn Lee, T. Yasuda, H. Nomura and C. Adachi, Appl. Phys. Lett., 2012, 101, 093306.
43. H. Kaji, H. Suzuki, T. Fukushima, K. Shizu, K. Suzuki, S. Kubo, T. Komino, H. Oiwa, F. Suzuki, A. Wakamiya, Y. Murata and C. Adachi, Nat. Commun., 2015, 6, 8476.
44. D. R. Lee, J. M. Choi, C. W. Lee and J. Y. Lee, ACS Appl. Mater. Interfaces, 2016, 8, 23190–23196.
45. K. Wu, Z. Wang, L. Zhan, C. Zhong, S. Gong, G. Xie and C. Yang, J. Phys. Chem. Lett., 2018, 9, 1547–1553.
46. Q. Zhang, D. Tsang, H. Kuwabara, Y. Hatae, B. Li, T. Takahashi, S. Y. Lee, T. Yasuda and C. Adachi, Adv. Mater., 2015, 27, 2096–2100.
47. G. Meng, X. Chen, X. Wang, N. Wang, T. Peng and S. Wang, Adv. Opt. Mater., 2019, 7, 1900130.
48. N. A. Kukhta, H. F. Higginbotham, T. Matulaitis, A. Danos, A. N. Bismillah, N. Haase, M. K. Etherington, D. S. Yufit, P. R. McGonigal, J. V. Gražulevičius and A. P. Monkman, J. Mater. Chem. C, 2019, 7, 9184–9194.
49. A. Batra, G. Kladnik, H. Vázquez, J. S. Meisner, L. Floreano, C. Nuckolls, D. Cvetko, A. Morgante and L. Venkataraman, Nat. Commun., 2012, 3, 1086.
50. D. Graves, V. Jankus, F. B. Dias and A. Monkman, Adv. Funct. Mater., 2014, 24, 2343–2351.
51. S. Shao and L. Wang, Aggregate, 2020, 1, 45–56.
52. Y. Zhang, C. Liu, X.-C. Hang, Q. Xue, G. Xie, C. Zhang, T. Qin, Z. Sun, Z.-K. Chen and W. Huang, J. Mater. Chem. C, 2018, 6, 8035–8041.
53. J.-H. Lee, S.-H. Cheng, S.-J. Yoo, H. Shin, J.-H. Chang, C.-I. Wu, K.-T. Wong and J.-J. Kim, Adv. Funct. Mater., 2015, 25, 361–366.
54. M. Jung, K. H. Lee, J. Y. Lee and T. Kim, Mater. Horiz., 2020, 7, 559–565.
55. T.-C. Lin, M. Sarma, Y.-T. Chen, S.-H. Liu, K.-T. Lin, P.-Y. Chiang, W.-T. Chuang, Y.-C. Liu, H.-F. Hsu, W.-Y. Hung, W.-C. Tang, K.-T. Wong and P.-T. Chou, Nat. Commun., 2018, 9, 3111.
56. T.-L. Wu, S.-Y. Liao, P.-Y. Huang, Z.-S. Hong, M.-P. Huang, C.-C. Lin, M.-J. Cheng and C.-H. Cheng, ACS Appl. Mater. Interfaces, 2019, 11, 19294–19300.
57. Y.-Z. Shi, K. Wang, X. Li, G.-L. Dai, W. Liu, K. Ke, M. Zhang, S.-L. Tao, C.-J. Zheng, X.-M. Ou and X.-H. Zhang, Angew. Chem., Int. Ed., 2018, 57, 9480–9484.
58. H. Tsujimoto, D.-G. Ha, G. Markopoulos, H. S. Chae, M. A. Baldo and T. M. Swager, J. Am. Chem. Soc., 2017, 139, 4894–4900.
59. C. Jiang, J. Miao, D. Zhang, Z. Wen, C. Yang and K. Li, Research, 2022, 2022, 9892802.
60. X.-F. Song, C. Jiang, N. Li, J. Miao, K. Li and C. Yang, Chem. Sci., 2023, 14, 12246–12254.
61. 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.
62. X. Wu, S. Ni, C.-H. Wang, W. Zhu and P.-T. Chou, Chem. Rev., 2025, 14, 6685–6752.
63. P. Karak, R. Adhikary and S. Chakrabarti, J. Phys. Chem. C, 2024, 128, 16879–16891.
64. Q. Zhang, J. Li, K. Shizu, S. Huang, S. Hirata, H. Miyazaki and C. Adachi, J. Am. Chem. Soc., 2012, 134, 14706–14709.
65. H.-Z. Li, F.-M. Xie, Y.-Q. Li and J.-X. Tang, J. Mater. Chem. C, 2023, 11, 6471–6511.
66. M. Mamada, M. Hayakawa, J. Ochi and T. Hatakeyama, Chem. Soc. Rev., 2024, 53, 1624–1692.
67. P. Li, W. Li, Y. Zhang, P. Zhang, X. Wang, C. Yin and R. Chen, ACS Mater. Lett., 2024, 6, 1746–1768.
68. D. Chen, H. Wang, D. Sun, S. Wu, K. Wang, X.-H. Zhang and E. Zysman-Colman, Adv. Mater., 2024, 36, 2412761.
69. M. de Jong, L. Seijo, A. Meijerink and F. T. Rabouw, Phys. Chem. Chem. Phys., 2015, 17, 16959–16969.
70. Q. He, M. Li and S.-J. Su, ChemPhysChem, 2025, 26, e202400955.
71. C. Lv, X. Wang, Q. Zhang and Y. Zhang, Mater. Chem. Front., 2023, 7, 2809–2827.
72. S. Oda, W. Kumano, T. Hama, R. Kawasumi, K. Yoshiura and T. Hatakeyama, Angew. Chem., Int. Ed., 2021, 60, 2882–2886.
73. Y. Xu, Q. Wang, J. Wei, X. Peng, J. Xue, Z. Wang, S.-J. Su and Y. Wang, Angew. Chem., Int. Ed., 2022, 61, e202204652.
74. F. Liu, Z. Cheng, L. Wan, Z. Feng, H. Liu, H. Jin, L. Gao, P. Lu and W. Yang, Small, 2022, 18, 2106462.
75. M. Yang, S. Shikita, H. Min, I. S. Park, H. Shibata, N. Amanokura and T. Yasuda, Angew. Chem., Int. Ed., 2021, 60, 23142–23147.
76. D. Barman, Y. Tsuchiya and C. Adachi, Nat. Commun., 2025, 16, 5023.
77. Q. Wang, Y. Xu, T. Huang, Y. Qu, J. Xue, B. Liang and Y. Wang, Angew. Chem., Int. Ed., 2023, 62, e202301930.
78. Y. Zou, J. Hu, M. Yu, J. Miao, Z. Xie, Y. Qiu, X. Cao and C. Yang, Adv. Mater., 2022, 34, 2201442.
79. Y.-J. Yu, Z.-Q. Feng, X.-Y. Meng, L. Chen, F.-M. Liu, S.-Y. Yang, D.-Y. Zhou, L.-S. Liao and Z.-Q. Jiang, Angew. Chem., Int. Ed., 2023, 62, e202310047.
80. B. H. Jhun, Y. Park, H. S. Kim, J. H. Baek, J. Kim, E. Lee, H. Moon, C. Oh, Y. Jung, S. Choi, M.-H. Baik and Y. You, Nat. Commun., 2025, 16, 392.
81. L. Chen, P. Zou, J. Chen, L. Xu, B. Z. Tang and Z. Zhao, Nat. Commun., 2025, 16, 1656.
82. D.-W. Zhang, M. Li and C.-F. Chen, Chem. Soc. Rev., 2020, 49, 1331–1343.
83. M. Li and C.-F. Chen, Org. Chem. Front., 2022, 9, 6441–6452.
84. L. Frédéric, A. Desmarchelier, L. Favereau and G. Pieters, Adv. Funct. Mater., 2021, 31, 2010281.
85. J. Wang, D. Chen, J. M. Moreno-Naranjo, F. Zinna, L. Frédéric, D. B. Cordes, A. P. McKay, M. J. Fuchter, X. Zhang and E. Zysman-Colman, Chem. Sci., 2024, 15, 16917–16927.
86. P. Li, W. Li, Q. Lv, R. Chen and C. Zheng, J. Mater. Chem. C, 2023, 11, 4033–4041.
87. C. Dutta, R. V. Nataraajan and J. Kumar, Chem. Sci., 2025, 16, 1722–1729.
88. T. Imagawa, S. Hirata, K. Totani, T. Watanabe and M. Vacha, Chem. Commun., 2015, 51, 13268–13271.
89. L. Yuan, Y.-P. Zhang and Y.-X. Zheng, Sci. China: Chem., 2024, 67, 1097–1116.
90. Y. Wang, Y. Zhang, W. Hu, Y. Quan, Y. Li and Y. Cheng, ACS Appl. Mater. Interfaces, 2019, 11, 26165–26173.
91. M.-Y. Zhang, Z.-Y. Li, B. Lu, Y. Wang, Y.-D. Ma and C.-H. Zhao, Org. Lett., 2018, 20, 6868–6871.
92. S.-Y. Yang, S.-N. Zou, F.-C. Kong, X.-J. Liao, Y.-K. Qu, Z.-Q. Feng, Y.-X. Zheng, Z.-Q. Jiang and L.-S. Liao, Chem. Commun., 2021, 57, 11041–11044.
93. L. Frédéric, A. Desmarchelier, R. Plais, L. Lavnevich, G. Muller, C. Villafuerte, G. Clavier, E. Quesnel, B. Racine, S. Meunier-Della-Gatta, J.-P. Dognon, P. Thuéry, J. Crassous, L. Favereau and G. Pieters, Adv. Funct. Mater., 2020, 30, 2004838.
94. M. Li, S.-H. Li, D. Zhang, M. Cai, L. Duan, M.-K. Fung and C.-F. Chen, Angew. Chem., Int. Ed., 2018, 57, 2889–2893.
95. Y. Xu, Q. Wang, X. Cai, C. Li and Y. Wang, Adv. Mater., 2021, 33, 2100652.
96. A. Monkman, ACS Appl. Mater. Interfaces, 2022, 14, 20463–20467.
97. T.-T. Bui, F. Goubard, M. Ibrahim-Ouali, D. Gigmes and F. Dumur, Beilstein J. Org. Chem., 2018, 14, 282–308.
98. M. Zhu and C. Yang, Chem. Soc. Rev., 2013, 42, 4963–4976.
99. S. Scholz, D. Kondakov, B. Lüssem and K. Leo, Chem. Rev., 2015, 115, 8449–8503.
100. K. Matsuo and T. Yasuda, Chem. Sci., 2019, 10, 10687–10697.
101. S. Y. Lee, C. Adachi and T. Yasuda, Adv. Mater., 2016, 28, 4626–4631.
102. T.-A. Lin, T. Chatterjee, W.-L. Tsai, W.-K. Lee, M.-J. Wu, M. Jiao, K.-C. Pan, C.-L. Yi, C.-L. Chung, K.-T. Wong and C.-C. Wu, Adv. Mater., 2016, 28, 6976–6983.
103. N. Jürgensen, A. Kretzschmar, S. Höfle, J. Freudenberg, U. H. F. Bunz and G. Hernandez-Sosa, Chem. Mater., 2017, 29, 9154–9161.
104. I. S. Park, H. Komiyama and T. Yasuda, Chem. Sci., 2017, 8, 953–960.
105. J. Lee, N. Aizawa and T. Yasuda, Chem. Mater., 2017, 29, 8012–8020.
106. P. Rajamalli, V. Thangaraji, N. Senthilkumar, C.-C. Ren-Wu, H.-W. Lin and C.-H. Cheng, J. Mater. Chem. C, 2017, 5, 2919–2926.
107. L.-S. Cui, H. Nomura, Y. Geng, J. U. Kim, H. Nakanotani and C. Adachi, Angew. Chem., Int. Ed., 2017, 56, 1571–1575.
108. P. Rajamalli, N. Senthilkumar, P. Y. Huang, C. C. Ren-Wu, H. W. Lin and C. H. Cheng, J. Am. Chem. Soc., 2017, 139, 10948–10951.
109. D. H. Ahn, J. H. Jeong, J. Song, J. Y. Lee and J. H. Kwon, ACS Appl. Mater. Interfaces, 2018, 10, 10246–10253.
110. S. Oda, B. Kawakami, R. Kawasumi, R. Okita and T. Hatakeyama, Org. Lett., 2019, 21, 9311–9314.
111. H. Tanaka, S. Oda, G. Ricci, H. Gotoh, K. Tabata, R. Kawasumi, D. Beljonne, Y. Olivier and T. Hatakeyama, Angew. Chem., Int. Ed., 2021, 60, 17910–17914.
112. M. Nagata, H. Min, E. Watanabe, H. Fukumoto, Y. Mizuhata, N. Tokitoh, T. Agou and T. Yasuda, Angew. Chem., Int. Ed., 2021, 60, 20280–20285.
113. Y. Zhang, J. Wei, D. Zhang, C. Yin, G. Li, Z. Liu, X. Jia, J. Qiao and L. Duan, Angew. Chem., Int. Ed., 2022, 61, e202113206.
114. H. L. Lee, S. O. Jeon, I. Kim, S. C. Kim, J. Lim, J. Kim, S. Park, J. Chwae, W.-J. Son, H. Choi and J. Y. Lee, Adv. Mater., 2022, 34, 2202464.
115. Y. Chang, K. Zhang, L. Zhao, X. Wang, S. Wang, S. Shao and L. Wang, Angew. Chem., Int. Ed., 2025, 64, e202415607.
116. J. Liu, L. Chen, X. Wang, Q. Yang, L. Zhao, C. Tong, S. Wang, S. Shao and L. Wang, Macromol. Rapid Commun., 2022, 43, 2200079.
117. Z. Zhao, C. Zeng, X. Peng, Y. Liu, H. Zhao, L. Hua, S.-J. Su, S. Yan and Z. Ren, Angew. Chem., Int. Ed., 2022, 61, e202210864.
118. Y. Im and J. Y. Lee, J. Inf. Disp., 2017, 18, 101–117.
119. R. Braveenth and K. Y. Chai, Materials, 2019, 12, 2646.
120. R. Braveenth, H. Lee, S. Kim, K. Raagulan, S. Kim, J. H. Kwon and K. Y. Chai, J. Mater. Chem. C, 2019, 7, 7672–7680.
121. R. Ansari, W. Shao, S.-J. Yoon, J. Kim and J. Kieffer, ACS Appl. Mater. Interfaces, 2021, 13, 28529–28537.
122. H. Cho, C. W. Joo, S. Choi, C.-M. Kang, B.-H. Kwon, J.-W. Shin, K. Kim, D. H. Ahn and N. S. Cho, Org. Electron., 2022, 101, 106419.
123. Y. Liu, C. Li, Z. Ren, S. Yan and M. R. Bryce, Nat. Rev. Mater., 2018, 3, 18020.
124. M. P. Gaj, A. Wei, C. Fuentes-Hernandez, Y. Zhang, R. Reit, W. Voit, S. R. Marder and B. Kippelen, Org. Electron., 2015, 25, 151–155.
125. L.-S. Cui, A. J. Gillett, S.-F. Zhang, H. Ye, Y. Liu, X.-K. Chen, Z.-S. Lin, E. W. Evans, W. K. Myers, T. K. Ronson, H. Nakanotani, S. Reineke, J.-L. Bredas, C. Adachi and R. H. Friend, Nat. Photonics, 2020, 14, 636–642.
126. H. Wang, L. Xie, Q. Peng, L. Meng, Y. Wang, Y. Yi and P. Wang, Adv. Mater., 2014, 26, 5198–5204.
127. X. Wei, Z. Li, T. Hu, R. Duan, J. Liu, R. Wang, Y. Liu, X. Hu, Y. Yi, P. Wang and Y. Wang, Adv. Opt. Mater., 2019, 7, 1801767.
128. T.-L. Wu, M.-J. Huang, C.-C. Lin, P.-Y. Huang, T.-Y. Chou, R.-W. Chen-Cheng, H.-W. Lin, R.-S. Liu and C.-H. Cheng, Nat. Photonics, 2018, 12, 235–240.
129. S. Kothavale, K. Cheong, S. C. Kim, S.-J. Yoon and J. Y. Lee, J. Mater. Chem. C, 2024, 12, 14037–14044.
130. B. Zhang, S. Liu, J. Pei, M. Luo, Y. Chen, Q. Jia, Z. Wu and D. Wang, Chem. Sci., 2024, 15, 5746–5756.
131. P. Data, P. Pander, M. Okazaki, Y. Takeda, S. Minakata and A. P. Monkman, Angew. Chem., Int. Ed., 2016, 55, 5739–5744.
132. J.-X. Chen, W.-W. Tao, W.-C. Chen, Y.-F. Xiao, K. Wang, C. Cao, J. Yu, S. Li, F.-X. Geng, C. Adachi, C.-S. Lee and X.-H. Zhang, Angew. Chem., Int. Ed., 2019, 58, 14660–14665.
133. H.-J. Tan, J.-L. Liu, J.-R. Yu, G.-X. Yang, X.-Q. Gao, B. Wang, Z.-Q. Long, Q.-X. Tong, J.-X. Jian, S.-J. Su, Z.-L. Zhu and C.-S. Lee, Adv. Funct. Mater., 2025, e11296,  DOI:10.1002/adfm.202511296.
134. Y. Song, K. Zhang, P. Wang, Y. Cao and L. He, Adv. Funct. Mater., 2025, 35, 2411957.
135. H. Sun, X. Li, C.-H. Hsu, C.-M. Hung, B. Wu, Z.-H. Su, G. V. Baryshnikov, C. Li, H. Ågren, Z. Zhang, W. Huang, D. Wu, P.-T. Chou and L. Zhu, J. Am. Chem. Soc., 2025, 147, 5432–5439.
136. M. Mońka, P. Pander, D. Grzywacz, A. Sikorski, R. Rogowski, P. Bojarski, A. P. Monkman and I. E. Serdiuk, ACS Appl. Mater. Interfaces, 2025, 17, 9635–9645.
137. R. Englman and J. Jortner, Mol. Phys., 1970, 18, 145–164.
138. M. Fakis, D. Anestopoulos, V. Giannetas and P. Persephonis, J. Phys. Chem. B, 2006, 110, 24897–24902.
139. J. H. Kim, J. H. Yun and J. Y. Lee, Adv. Opt. Mater., 2018, 6, 1800255.
140. P. L. dos Santos, P. Stachelek, Y. Takeda and P. Pander, Mater. Chem. Front., 2024, 8, 1731–1766.
141. L. Tu, Y. Xie, Z. Li and B. Tang, SmartMat, 2021, 2, 326–346.
142. X. Wang, Y. Zhang, Z. Yu, Y. Wu, D. Wang, C. Wu, H. Ma, S. Ning, H. Dong and Z. Wu, J. Mater. Chem. C, 2022, 10, 1681–1689.
143. E. H. Cho, H.-R. Choi, Y. Park, S. Y. Jeong, Y. J. Song, Y. H. Hwang, J. Lee, Y. Chi, S.-F. Wang, Y. Jeon, C.-H. Huh and K. C. Choi, ACS Appl. Mater. Interfaces, 2023, 15, 57415–57426.
144. J. Kido, K. Hongawa, K. Okuyama and K. Nagai, Appl. Phys. Lett, 1994, 64, 815–817.
145. K. Yamae, H. Tsuji, V. Kittichungchit, N. Ide and T. Komoda, SID Symp. Dig. Tech. Pap., 2013, 44, 916–919.
146. S.-F. Wu, S.-H. Li, Y.-K. Wang, C.-C. Huang, Q. Sun, J.-J. Liang, L.-S. Liao and M.-K. Fung, Adv. Funct. Mater., 2017, 27, 1701314.
147. B. Liu, M. Xu, L. Wang, X. Yan, H. Tao, Y. Su, D. Gao, L. Lan, J. Zou and J. Peng, Org. Electron., 2014, 15, 926–936.
148. X. Li, W. Liu, Q. Zhu, R. Wu, G. Liu and L. Zhou, Opt. Mater., 2022, 124, 112005.
149. K. Chen, R. Wu, X. Li, W. Liu, Z. Wei, J. Wang and L. Zhou, Opt. Mater., 2022, 123, 111917.
150. M. Zhang, C.-J. Zheng, K. Wang, Y.-Z. Shi, H.-Y. Yang, H. Lin, S.-L. Tao and X.-H. Zhang, Org. Electron., 2022, 101, 106415.
151. Q. Wang, Y.-X. Zhang, Y. Yuan, Y. Hu, Q.-S. Tian, Z.-Q. Jiang and L.-S. Liao, ACS Appl. Mater. Interfaces, 2019, 11, 2197–2204.
152. X. Du, J. Zhao, S. Yuan, C. Zheng, H. Lin, S. Tao and C.-S. Lee, J. Mater. Chem. C, 2016, 4, 5907–5913.
153. L. Zhang, X.-L. Li, D. Luo, P. Xiao, W. Xiao, Y. Song, Q. Ang and B. Liu, Materials, 2017, 10, 1378.
154. J.-i Nishide, H. Nakanotani, Y. Hiraga and C. Adachi, Appl. Phys. Lett., 2014, 104, 233304.
155. W. Zhang, Y. Li, G. Zhang, X. Yang, X. Chang, G. Xing, H. Dong, J. Wang, D. Wang, Z. Mai and X. Jiang, Micromachines, 2024, 15, 703.
156. M. Kim, S. K. Jeon, S.-H. Hwang and J. Y. Lee, Adv. Mater., 2015, 27, 2515–2520.
157. G. Schwartz, S. Reineke, T. C. Rosenow, K. Walzer and K. Leo, Adv. Funct. Mater., 2009, 19, 1319–1333.
158. K. Sato, K. Shizu, K. Yoshimura, A. Kawada, H. Miyazaki and C. Adachi, Phys. Rev. Lett., 2013, 110, 247401.
159. D. Zhang, M. Cai, Y. Zhang, D. Zhang and L. Duan, ACS Appl. Mater. Interfaces, 2015, 7, 28693–28700.
160. D. Zhang, L. Duan, Y. Li, D. Zhang and Y. Qiu, J. Mater. Chem. C, 2014, 2, 8191–8197.
161. T. Higuchi, H. Nakanotani and C. Adachi, Adv. Mater., 2015, 27, 2019–2023.
162. G. Schwartz, M. Pfeiffer, S. Reineke, K. Walzer and K. Leo, Adv. Mater., 2007, 19, 3672–3676.
163. D. Ding, Z. Wang, C. Duan, C. Han, J. Zhang, S. Chen, Y. Wei and H. Xu, Research, 2022, 0009.
164. U. Deori, T. Viswanathan, N. Yadav and P. Rajamalli, J. Mater. Chem. C, 2025, 13, 1449–1456.
165. P. Xiao, J. Huang, Y. Yu, J. Yuan, D. Luo, B. Liu and D. Liang, Materials, 2018, 8, 1376.
166. Y. Miao, G. Wang, M. Yin, Y. Guo, B. Zhao and H. Wang, Chem. Eng. J., 2023, 461, 141921.
167. D. Zhang, X. Cao, Q. Wu, M. Zhang, N. Sun, X. Zhang and Y. Tao, J. Mater. Chem. C, 2018, 6, 3675–3682.
168. D. Luo, Q. Chen, Y. Gao, M. Zhang and B. Liu, ACS Energy Lett., 2018, 3, 1531–1538.
169. S. Reineke, F. Lindner, G. Schwartz, N. Seidler, K. Walzer, B. Lüssem and K. Leo, Nature, 2009, 459, 234–238.
170. D. Zhang, L. Duan, Y. Zhang, M. Cai, D. Zhang and Y. Qiu, Light: Sci. Appl., 2015, 4, e232.
171. E. Tankelevičiūtė, I. D. W. Samuel and E. Zysman-Colman, J. Phys. Chem. Lett., 2024, 15, 1034–1047.
172. M. Mamada, T. B. Nguyen, H. Nakanotani, T. Hatakeyama and C. Adachi, Adv. Opt. Mater., 2025, e00971,  DOI:10.1002/adom.202500971.

---