Design of Thermally Activated Delayed Fluorescence Materials
for Organic Light-Emitting Diodes

NAOYA AIZAWA¹, IN SEOB PARK^1,2 AND TAKUMA YASUDA^1,2
1 INAMORI FRONTIER RESEARCH CENTER (IFRC), KYUSHU UNIVERSITY, JAPAN
² DEPARTMENT OF APPLIED CHEMISTRY, GRADUATE SCHOOL OF ENGINEERING, KYUSHU UNIVERSITY, JAPAN

ABSTRACT

Organic light-emitting diodes (OLEDs) are ultrathin two-dimensional light sources based on soft amorphous organic semiconductors. Owing to their high color rendering, high flexibility, and low power consumption, OLEDs have gained tremendous interest as next-generation displays and lighting sources. Here, we discuss a concept of thermally activated delayed fluorescence (TADF) that has recently been introduced to realize 100% internal quantum efficiencies of electron-to-photon conversion in OLEDs. We aim to provide a rational guide for designing intramolecular donor−acceptor systems that exhibit highly efficient TADF.

INTRODUCTION

Organic light-emitting diodes (OLEDs), which consist of very thin, stacked layers of organic semiconductors, can transform electrical energy into light (Fig. 1(a)). Owing to their form factor, OLEDs can serve as two-dimensional light sources, which enables practical commercial applications in general solid-lighting and display devices, including mobile phones and flat-panel televisions. Moreover, OLEDs have many excellent characteristics, such as self-luminescence, high color rendering, high electroluminescence (EL) efficiency, high flexibility, and low power consumption. It is expected that the number of products and applications utilizing OLED technologies will increase to a large extent in the future.

Fig. 1: (a) Photograph of an OLED that transforms electricity into visible light. (b) Device architecture of typical OLEDs. (c) Schematic energy-level diagram of typical OLEDs.

The prospect of using organic semiconductor materials for OLEDs has flourished over the last three decades since the seminal work of Tang and Van Slyke in 1987 [1]. Their prototype OLED using tris(8-quinolinolato)aluminum(III) (Alq₃) as a fluorescence emitter exhibited an external EL quantum efficiency (η_ext) of 1% and a brightness of 1000 cd m^-2. In 1990, Burroughes et al. demonstrated the first OLED based on a fluorescent polymer emitter, poly(p-phenylene vinylene) [2].

Driven by an external electrical field, the holes and electrons in OLEDs are injected from the anode and cathode, respectively, and transported to an emission layer, located at the center of the device (Fig. 1(b), 1(c)). Upon recombination of holes and electrons in the emission layer, the emitter molecules are electronically excited. In general, the theoretical maximum of η_ext, which corresponds to the number of photons emitted from the OLED device per charge carriers injected into the device, is expressed by the following equation.

Here, η_int is the internal EL quantum efficiency, η_out is the light out-coupling efficiency, γ is the charge balance factor, η_ST is the fraction of radiative excitons, and Φ_PL is the photoluminescence (PL) quantum efficiency of the emitters. Ideally, γ and Φ_PL can be close to unity if holes and electrons are fully balanced and recombined to generate excitons in the emission layer, and the emitter material possesses an appropriately high luminescence property. η_out is generally assumed to be approximately 20% in a randomly oriented emitter for a device without out-coupling enhancement. Meanwhile, the electrical excitation typically leads to the formation of singlet and triplet excitons with a probability of 25% and 75%, respectively, according to spin statistics [3]. Therefore, for conventional fluorescent OLEDs, the theoretical maximum of η_int is limited to 25% (Fig. 2(a)), because only the singlet excitons enable spin-allowed emissive transitions, and the remaining triplet excitons are lost as heat.

Fig. 2: Schematic representation of three different types of electroluminescence mechanisms in OLEDs: (a) fluorescence (up to 25% internal quantum efficiency), (b) phosphorescence, and (c) thermally activated delayed fluorescence (TADF) (up to 100% internal quantum efficiency).

An effective way to circumvent this problem is to utilize phosphorescence [4], i.e., the radiative transition from the lowest excited triplet (T₁) states to the singlet ground (S₀) state (Fig. 2(b)), from heavy transition metal-containing organometallic complexes, such as tris(2-phenylpyridinato)iridium(III) (Ir(ppy)₃) [5]. Phosphorescence is a spin-forbidden transition, yet it becomes allowable due to strong spin−orbit coupling induced by the heavy transition metal center. As a result, all electro-generated excitons can be harvested by the T₁ state via intersystem crossing (ISC), leading to η_int of nearly 100% in phosphorescent OLEDs [6]. However, phosphorescent materials rely on precious metals such as iridium and platinum. The expense and rarity of such precious-metal complexes limit the cost effectiveness and long-term mass production of phosphorescent OLEDs. Moreover, it is still difficult to produce efficient pure-blue phosphorescent materials exhibiting high quantum efficiency, color purity, and long-term stability, which would hamper the widespread application of phosphorescent OLEDs in the future.

Recent studies by Adachi and coworkers and other research groups have paved an alternative way for the realization of high-efficiency, precious-metal-free OLEDs by utilizing the mechanism of thermally activated delayed fluorescence (TADF) (Fig. 2(c)) [7,8]. In the TADF mechanism, purely organic luminescent compounds with very small singlet−triplet energy splitting (ΔE_ST) enable efficient up-conversion of the excited T₁ states to the emissive lowest excited singlet (S₁) state via reverse intersystem crossing (RISC). In the last couple years, considerable research efforts have been invested for exploring efficient TADF materials and understanding their unique photophysical properties, in order to realize η_int of 100% without using precious metals in OLEDs [8]. In this Feature Article, we focus on recent progress on purely organic TADF materials, and describe the molecular design and photophysical properties of TADF materials as well as their OLED performances.

BASIC DESIGN PRINCIPLES FOR TADF MATERIALS

TADF is typically observed in luminescent organic materials having a very small ΔE_ST, since the thermal energy (k_BT ≈ 25.6 meV) at room temperature can assist RISC from the T₁ state to the S₁ state. The origin of RISC involves the mixing of the excited T₁ and S₁ states, which is caused by spin−orbit coupling (H_SO). According to first-order perturbation theory, the first-order mixing coefficient (λ) is expressed by the following equation.

Since there is finite H_SO even in purely organic materials, minimizing the ΔE_ST can lead to the enhancement of λ, and hence, an efficient RISC from the T₁ to S₁ states. The key molecular design for achieving the desired minimal ΔE_ST is to reduce the spatial overlap between the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO). Assuming S₀ energy is zero, ΔE_ST is given by the following equation [9].

Here, E_S and E_T are the energies of the S₁ and the T₁ states, respectively, and J is the electron exchange integral associated with the Pauli principle, and is expressed by the following equation.

Here, Φ_H and Φ_L are the wave functions of the HOMO and LUMO, respectively, e is the charge on an electron, and r₁₂ is the distance separating the electrons. From these equations, it is crucial to separate the HOMO and LUMO wave functions in rational TADF molecular design, in order to minimize ΔE_ST and to accelerate the RISC process.

Donor−acceptor (D−A) molecular systems, which are structurally defined by an electron-rich moiety covalently linked to an electron-deficient moiety, are commonly used for the design of TADF materials. Distorting the π-conjugations of these D−A molecules leads to well separated HOMO and LUMO distributions [10]. These D−A systems can also provide the additional advantage of precise control over the emission wavelength (i.e., the energy level of the lowest excited charge-transfer singlet (¹CT) state) by varying the strength of the composed donor and acceptor moieties. Recent research activities have been focused on the design and synthesis of new D−A TADF molecules containing aryl amines (e.g., carbazole, diphenylamine, phenoxazine, and acridan) as a donor moiety. This is likely due to the fact that these aryl amines not only have an electron-donating ability but also offer effective intramolecular steric repulsion to the adjacent acceptor moiety, which induces a large dihedral angle between the donor and acceptor moieties. This propensity leads to an effective spatial separation of the HOMO and LUMO, and thus, a small ΔE_ST of the emitter molecules. Dias et al. pointed out that the origin of TADF involves population of triplet excitons to the ³nπ* states of these aryl amines, followed by the RISC from the ³CT to ¹CT states [11]. The selection of the acceptor moiety also significantly affects the photophysical properties of the TADF molecules. Representative TADF materials are discussed below, by classifying them based on the structural features of the acceptor moieties.

BENZONITRILE-BASED TADF MATERIALS

Benzonitriles are one of the most widely used acceptor units in the design of TADF materials [8,12-22]. Compound 1 (Spiro-CN) was developed as a yellow TADF material by Nakagawa et al. in 2012 [12]. In Spiro-CN, a 2,7-bis(di-p-tolylamino)fluorene donor unit and 2,7-dicyanofluorene acceptor unit were orthogonally connected through a spiro bridge. As can be seen from Fig. 3, the connection via spiro bridge effectively breaks the π-conjugation between the donor and acceptor moieties, resulting in well separated HOMO and LUMO distributions. Spiro-CN exhibited a yellow TADF emission with a peak at around 540 nm and a TADF emission lifetime of 14 μs. Indeed, the ΔE_ST of Spiro-CN was determined to be 57 meV, which is much smaller than that of C₇₀ (0.26 eV) [23] and tin(IV) fluorideporphyrin (0.24 eV) [7], suggesting that the spiro-based D−A structure is promising in obtaining small ΔE_ST. Because of its low Φ_PL of 27%, the η_ext of a Spiro-CN-based OLED was limited to 4.4%. A further structural modification by replacing the donor moiety of Spiro-CN with N-phenylacridan afforded compound 2 (ACRFLCN), resulting in a higher Φ_PL of 67% and η_ext of 10.1% [13,14].

Fig. 3: (a) Optimized molecular structure and (b,c) the distributions of the HOMO and LUMO of Spiro-CN calculated at the B3LYP/6-31G(d) level of theory.

As described above, the key design rule for obtaining a small ΔE_ST is to reduce the overlap between the HOMO and LUMO. However, this molecular design in turn results in the decrease of the oscillator strength between S₁ and S₀, and thus, lowers the Φ_PL. In 2012, Uoyama et al. attained both small ΔE_ST and high Φ_PL values in phthalonitrile-based TADF molecules (compounds 3−8) by carefully tuning the spatial overlap between the HOMO and LUMO [8]. Compound 3 (4CzIPN) bearing an isophthalonitrile acceptor core and four carbazole donor arms was found to show a small ΔE_ST of 83 meV and a high Φ_PL of 94% with an emission peak at 507 nm in toluene solution. Besides having a high Φ_PL, TADF emission lifetime of 4CzIPN was very short (5.1 μs). For a 4CzIPN-based OLED, the η_ext reached 19.3%, which was considerably higher than those of conventional fluorescent OLEDs (η_ext < 5%) and even comparable with those of phosphorescent OLEDs. Time-dependent density functional theory (TD-DFT) calculations predicted that the highest occupied and lowest unoccupied natural transition orbitals (NTOs) of 4CzIPN were located on the donor and acceptor moieties, respectively, while they were moderately overlapped on the central benzene ring. Additionally, the reorganization energies in the excited state geometries were calculated to be relatively small for these intramolecular D−A molecules. These structural and electronic features can contribute to the high Φ_PL in 4CzIPN.

Further device optimization improved the η_ext and operational lifetime of 4CzIPN-based OLEDs [24-27]. Sun et al. reported that the transition dipole moments of the 4CzIPN molecules tend to orient horizontally in a host matrix, which results in an enhancement of η_out [28]. As a result, a high η_ext of 29.6% was obtained, which was consistent with the theoretical maximum η_ext value predicted by the optical simulations of the device. Nakanotani et al. reported that the operational lifetime of 4CzIPN-based OLEDs was improved to 2800 hours for aging to 50% of an initial luminance (LT50) of 1000 cd m^-2 when the concentration of 4CzIPN in the emission layer was increased [29]. This improvement of the device stability can be attributed to the shift of the recombination zone to the center of the emission layer at the high concentration of 4CzIPN, which acts as an electron-trapping site in a carbazole-based host matrix.

The promising TADF properties of the phthalonitrile-based TADF materials encouraged further exploration of their analogues and derivatives. Nishimoto et al. developed compound 9 (CzTPN) incorporating an electron-accepting terephthalonitrile unit and two electron-donating carbazole units [15]. CzTPN exhibited a blue-green emission with a peak at 494 nm and a high η_ext of 15.0% when embedded in a cyclotriphosphazene-based host material with a high T₁ energy. Park et al. demonstrated that the emission colors of TADF can be tuned systematically by varying the donor and acceptor strengths in a simple D−A system (Fig. 4) [16]. Depending on the combination of the donor and acceptor units, compounds 10−14 displayed a wide variety of emission colors originating from TADF, covering the entire visible region from 450 nm for Cz-VPN to 610 nm for Px-CNP. As an example, compound 11 (AcVPN) incorporating a phthalonitrile acceptor unit and two 9,9-dimethylacridan donor units was found to show both a small ΔE_ST of 0.2 eV and a high Φ_PL of 86%. An OLED based on AcVPN exhibited a high η_ext of 18.9% for a sky-blue TADF emission with a peak at 504 nm [16].

Fig. 4: (a) Photographs showing full-color TADF emissions and (b) PL spectra of the thin films of compound 10−14 embedded in a host matrix under UV irradiation. (c)η_ext−current density characteristics of the TADF OLEDs based on compounds 10−14.

TRIAZINE-BASED TADF MATERIALS

A highly electron-withdrawing 1,3,5-triazine is another useful acceptor building block for designing a wide variety of efficient TADF materials [10,30-38]. 1,3,5-Triazine was first utilized in TADF molecules by Endo et al. in 2011 [10]. Compound 15 (PIC-TRZ), where a biphenyltriazine unit functions as an acceptor and two indolocarbazole units perform as a donor, exhibited green TADF emission with a peak at 495 nm and a TADF emission lifetime of 230 μs [10]. The ΔE_ST of PIC-TRZ was determined to be 0.11 eV. Owing to its relatively large ΔE_ST, a large portion of the T₁ excitons underwent non-radiative decay rather than RISC to the S₁ state, resulting in a relatively low η_ext of 5.3%. With a suitable structural modification, compound 16 (PIC-TRZ2) was found to exhibit a smaller ΔE_ST of 20 meV and a higher η_ext of 14.0% [30]. Lee et al. reported that compound 17 (CC2TA) incorporating a phenyltriazine acceptor unit and two bicarbazole donor units exhibited a small ΔE_ST of 60 meV and a high η_ext of 11% [31]. DFT calculations for CC2TA revealed that the outermost carbazoles were more electron rich, compared to the inner carbazoles in the bicarbazole moieties. This electron-distribution gradient in the bicarbazole moieties can give rise to an effective spatial separation of the HOMO and LUMO and a small ΔE_ST.

Hirata et al. reported that for triazine-based TADF materials, extending the π-conjugation of the donor moieties increases the oscillator strength without change in ΔE_ST values [32], which originates from an increased transition dipole moment induced by delocalization of the HOMO on the extended donor moieties. As a representative example, compound 18 achieved both small ΔE_ST of 90 meV and high Φ_PL of 100%. A TADF OLED based on compound 18 exhibited a high η_ext of 20.6% with an emission peak at 487 nm. This result suggested that η_int of nearly 100% can be attained, assuming a η_out of 20% (assuming the refractive index of 1.8 for the organic layers). Compound 19 (DACT-II) possessing a stronger electron-donating 3,6-diphenylaminocarbazole unit exhibited a red-shifted TADF emission at 520 nm in an OLED [33]. The η_ext of DACT-II-based TADF OLED reached 29% because DACT-II molecules tend to orient parallel to the substrate in a host matrix, giving rise to an improved η_out to 29.6%.

ARYL KETONE-BASED TADF MATERIALS

An electron-accepting aryl ketone unit was introduced as an acceptor moiety for the design of a spiro-type TADF molecule by Nasu et al. in 2013 [39]. Compound 20 (ACRSA), consisting of an anthrone acceptor unit and an N-phenylacridan donor unit through a spiro bridge, exhibited a greenish blue TADF emission with a peak at around 480 nm with a high Φ_PL of 81%, when doped in a host matrix. The fluorescence and phosphorescence spectra of ACRSA were nearly identical with a very small ΔE_ST of 30 meV. It is known that aryl ketones intrinsically possess a small ΔE_ST; however, their Φ_PL is typically less than 1% at room temperature because of the noticeably small overlap integral for the ¹nπ* transitions. The highly emissive feature of ACRSA suggests that the S₁ state should correspond to a ¹CT excited state. Thus, ACRSA exhibited a high η_ext of 16.5% in a TADF OLED, which was four times higher than the theoretical maximum for a conventional fluorescent material with the same Φ_PL of 81% [39].

In 2014, Lee et al. reported a series of TADF molecules based on a benzophenone as an acceptor moiety [40]. Depending on the strengths of the donor and acceptor moieties, compounds 21−25 (Cz2BP, CC2BP, Px2BP, m-Px2BBP, and p-Px2BBP) displayed tunable TADF emission colors with the emission maxima at 438, 462, 509, 566, and 600 nm, respectively, covering the entire visible region (Fig. 5). Moreover, the combination of light-blue-emitting CC2BP and yellow-emitting m-Px2BBP gave a white TADF emission in OLEDs with a η_ext of 6.7%. One particular feature of these benzophenone-based TADF molecules is the enhanced Φ_PL in solid states. In fact, the Φ_PL of CCz2BP was 73% in a host matrix, while it was 38% in a toluene solution. Similar emission behavior was also observed in compound 26 (AcPmBPX) and 27 (PxPmBPX) incorporating a benzoylbenzophenone acceptor unit [41]. The decreased Φ_PL in a solution can be explained by the presence of a non-radiative decay process associated with intramolecular motions, which can be suppressed in a solid state.

Fig. 5: EL emission color coordinates in the CIE chromaticity diagram and photographs of full-color TADF OLEDs based on compounds 21−25.

Zhang et al. developed anthraquinone-based TADF materials [42], and reported that the introduction of a phenylene (Ph) spacer suppressed the twisting and stretching motions of the N−C bond, resulting in enhanced Φ_PL. D−Ph−A−Ph−D structured compound 28 exhibited a Φ_PL of 80% in a host matrix, which was 1.6 times higher than that observed for the corresponding D−A−D structured compound 29.

OXADIAZOLE- AND TARIAZOLE-BASED TADF MATERIALS

Five-membered heteroaromatic rings, such as 1,3,4-oxadiazole and 1,2,4-triazole can function as acceptor moieties, and are used in the design of TADF materials [43-45]. Lee et al. developed compound 30 (PXZ-OXD), where a 2,5-diphenyl-1,3,4-oxadiazole unit served as an acceptor moiety and two phenoxazine units served as a donor moiety [43]. The PXZ-OXD-based OLED showed a green TADF with a peak at 508 nm and a η_ext of 14.9%. Replacement of the acceptor moiety of PXZ-OXD with 3,4,5-triphenyl-1,2,4-triazole gave compound 31 (PXZ-TAZ), leading to a blue-shifted TADF emission with a peak at 456 nm. The ΔE_ST of PXZ-TAZ determined from the onsets of the fluorescent and phosphorescent spectra was 0.23 eV. Because of its large ΔE_ST, PXZ-TAZ exhibited a long TADF emission lifetime of 2.09 ms. Consequently, the η_ext of a PXZ-TAZ-based OLED was limited to 5.3%. Similar millisecond-order decay times were observed for compound 32 (PPZ-3TPT) and compound 33 (PPZ-4TPT) as reported by Zhang et al [45]. TD-DFT calculations revealed that the locally excited triplet (³LE) states on the triazole acceptor moieties were lower in energy than the ³CT states in PPZ-3TPT and PPZ-4TPT. Although the ¹CT and ³CT states were energetically almost equivalent owing to the minimal overlap between the HOMO and LUMO distributions, these lower-lying ³LE states led to a large ΔE_ST, and underwent non-radiative decay for the triplet excitons.

SULFONE-BASED TADF MATERIALS

Given the role of the interplay between ³CT and ³LE states in governing ΔE_ST, one of the viable candidates for use as an acceptor moiety of blue TADF molecules is diphenylsulfone, which possess a ³LE of 3.02 eV [45-50]. In 2012, Zhang et al. reported a series of blue TADF molecules based on diphenylsulfone as an acceptor moiety [46]. Compound 34 (DTC-DPS) with 3,6-di-t-butylcarbazole donor moieties exhibited a η_ext of 10% for pure blue TADF emission with a peak at 423 nm and Commission Internationale de L'Eclairage (CIE) coordinates of (0.15, 0.07) in a TADF OLED. The low-lying ³LE (³ππ*) states of the donor moieties resulted in a deleterious large ΔE_ST of 0.32 eV and a millisecond-order lifetime of the triplet excitons in DTC-DPS. These conditions aided bimolecular singlet−triplet and triplet−triplet annihilations at a high luminance in OLEDs, causing a significant decrease in η_ext value (known as efficiency roll-off). A chemical modification of DTC-DPS by replacing the t-butyl substituents on the carbazole donor moieties with stronger electron-donating methoxy substituents, gave compound 35 (DMOC-DPS) with a smaller ΔE_ST of 0.21 eV by stabilizing the ¹CT states while maintaining the ³LE state [47]. This smaller ΔE_ST of DMOC-DPS resulted in an improved η_ext of 14.5% and reduced efficiency roll-off for deep blue emission with a peak at 460 nm in a TADF OLED. Compound 36 (DMAC-DPS) was designed to possess a small ΔE_ST of 80 meV by introducing 9,9-dimethylacridan as a donor moiety [45]. A DMAC-DPS-based TADF OLED exhibited a blue emission with a peak at 470 nm and a high η_ext of 19.5%, which was comparable to those observed in the state-of-art blue phosphorescent OLEDs.

ARYL BORON-BASED TADF MATERIALS

Aryl borons are another successful acceptor scaffold for designing blue TADF molecules [51-55]. Numata et al. reported a series of blue TADF molecules featuring a phenoxaborin unit [51]. Compound 37, incorporating a phenoxaborin acceptor unit and a 9,9-dimethylacridan donor unit, exhibited a small ΔE_ST of 0.10 eV and a high Φ_PL of 100% for a blue TADF emission with a peak at 475 nm. A blue TADF OLED using compound 37 exhibited a notably high η_ext of 21.7% (Fig. 6). Moreover, the emission peak was shifted to a shorter wavelength by about 20 nm when replacing the donor moiety with spiroacridan units, such as 10H-spiro(acridine-9,9'-fluorene) and 10H-spiro(acridine-9,9'-xanthene). The resulting compounds 38 and 39 exhibited η_ext of 19.0% and 20.1% for pure blue emissions with peaks at 456 and 451 nm, respectively [51].

Currently, deep-blue phosphorescent emitters with sufficient performance, especially in terms of device stability, are not available and blue OLEDs in commercial devices are still built with fluorescent emitter having intrinsically low η_ext. Therefore, the exploration for deep-blue TADF materials should be very attractive.

Fig. 6: (a) η_ext−current density characteristics, (b) EL spectra at 1 mA cm^-2, and (c) photographs of EL emissions of blue TADF OLEDs based on compounds 37−39.

SUMMARY

This Feature Article has reviewed recent progress on OLEDs based on purely organic D−A molecules that exhibit efficient TADF. These D−A electronic systems are advantageous for achieving a small ΔE_ST and high Φ_PL by precisely tuning the HOMO and LUMO distributions, which ultimately contribute to realizing a high η_int approaching 100% without having to use precious metals. The high EL efficiency as well as the lower cost of purely organic TADF materials make them a promising alternative to conventional phosphorescent organometallic complexes, as an emitter in a variety of OLED applications. Although there are further challenges to be addressed regarding the long-term device stability, given the current speed of the development of TADF technology, their mass production for general solid-lighting and display devices is expected to commence in the near future.

References

[1] Tang, C. W. & Van Slyke, S. A. Appl. Phys. Lett. 51, 913−915 (1987).
[2] Burroughes, J., Bradley, D., Brown, A., Marks, R., Mackay, K., Friend, R., Burns, P. & Holmes, A. Nature 347, 539−541 (1990).
[3] Segal, M., Baldo, M., Holmes, R., Forrest, S. & Soos, Z. Phys. Rev. B 68, 075211 (2003).
[4] Forrest, S. R. Nature 428, 911−918 (2004).
[5] Baldo, M., Lamansky, S., Burrows, P., Thompson, M. & Forrest, S. Appl. Phys. Lett. 75, 4−6 (1999).
[6] Adachi, C., Baldo, M. A., Thompson, M. E. & Forrest, S. R. J. Appl. Phys. 90, 5048−5051 (2001).
[7] Endo, A., Ogasawara, M., Takahashi, A., Yokoyama, D., Kato, Y. & Adachi, C. Adv. Mater. 21, 4802−4806 (2009).
[8] Uoyama, H., Goushi, K., Shizu, K., Nomura, H. & Adachi, C. Nature 492, 234−238 (2012).
[9] Turro, N. J., Ramamurthy, V. & Scaiano, J. C. in Modern molecular photochemistry of organic molecules (Wiley Online Library, 2012).
[10] Endo, A., Sato, K., Yoshimura, K., Kai, T., Kawada, A., Miyazaki, H. & Adachi, C. Appl. Phys. Lett. 98, 083302−083304 (2011).
[11] Dias, F. B., Bourdakos, K. N., Jankus, V., Moss, K. C., Kamtekar, K. T., Bhalla, V., Santos, J., Bryce, M. R. & Monkman, A. P. Adv. Mater. 25, 3707−3714 (2013).
[12] Nakagawa, T., Ku, S.-Y., Wong, K.-T. & Adachi, C. Chem. Commun. 48, 9580−9582 (2012).
[13] Méhes, G., Nomura, H., Zhang, Q., Nakagawa, T. & Adachi, C. Angew. Chem. Int. Ed. 51, 11311−11315 (2012).
[14] Méhes, G., Goushi, K., Potscavage, W. J. & Adachi, C. Org. Electron. 15, 2027−2037 (2014).
[15] Nishimoto, T., Yasuda, T., Lee, S. Y., Kondo, R. & Adachi, C. Mater. Horiz. 1, 264−269 (2014).
[16] Park, I. S., Lee, S. Y., Adachi, C. & Yasuda, T. Adv. Funct. Mater. 26, 1813-1821 (2016).
[17] Li, B., Nomura, H., Miyazaki, H., Zhang, Q., Yoshida, K., Suzuma, Y., Orita, A., Otera, J. & Adachi, C. Chem. Lett. 43, 319−321 (2014).
[18] Taneda, M., Shizu, K., Tanaka, H. & Adachi, C. Chem. Commun. 51, 5028−5031 (2015).
[19] Lee, J., Shizu, K., Tanaka, H., Nakanotani, H., Yasuda, T., Kaji, H. & Adachi, C. J. Mater. Chem. C 3, 2175−2181 (2015).
[20] Wang, S., Yan, X., Cheng, Z., Zhang, H., Liu, Y. & Wang, Y. Angew. Chem. Int. Ed. 54, 13068−13072 (2015).
[21] Cho, Y. J., Yook, K. S. & Lee, J. Y. Adv. Mater. 26, 6642−6646 (2014).
[22] Cho, Y. J., Jeon, S. K., Chin, B. D., Yu, E. & Lee, J. Y. Angew. Chem. Int. Ed. 54, 5201−5204 (2015).
[23] Berberan-Santos, M. N. & Garcia, J. M. J. Am. Chem. Soc. 118, 9391−9394 (1996).
[24] Kim, B. S. & Lee, J. Y. Adv. Funct. Mater. 24, 3970−3977 (2014).
[25] Cho, Y. J., Yook, K. S. & Lee, J. Y. Adv. Mater. 26, 4050−4055 (2014).
[26] Kim, O. Y., Kim, B. S. & Lee, J. Y. Synth. Met. 201, 49−53 (2015).
[27] Cui, L. S., Xie, Y. M., Wang, Y. K., Zhong, C., Deng, Y. L., Liu, X. Y., Jiang, Z. Q. & Liao, L. S. Adv. Mater. 27, 4213−4217 (2015).
[28] Sun, J. W., Lee, J. H., Moon, C. K., Kim, K. H., Shin, H. & Kim, J. J. Adv. Mater. 26, 5684−5688 (2014).
[29] Nakanotani, H., Masui, K., Nishide, J., Shibata, T. & Adachi, C. Sci. Rep. 3, 2127 (2013).
[30] Sato, K., Shizu, K., Yoshimura, K., Kawada, A., Miyazaki, H. & Adachi, C. Phys. Rev. Lett. 110, 247401 (2013).
[31] Lee, S. Y., Yasuda, T., Nomura, H. & Adachi, C. Appl. Phys. Lett. 101, 093306 (2012).
[32] Hirata, S., Sakai, Y., Masui, K., Tanaka, H., Lee, S. Y., Nomura, H., Nakamura, N., Yasumatsu, M., Nakanotani, H. & Zhang, Q. Nat. Mater. 14, 330−336 (2015).
[33] Kaji, H., Suzuki, H., Fukushima, T., Shizu, K., Suzuki, K., Kubo, S., Komino, T., Oiwa, H., Suzuki, F. & Wakamiya, A. Nat. Commun. 6, 8476 (2015).
[34] Tanaka, H., Shizu, K., Nakanotani, H. & Adachi, C. Chem. Mater. 25, 3766−3771 (2013).
[35] Tanaka, H., Shizu, K., Nakanotani, H. & Adachi, C. J. Phys. Chem. C 118, 15985−15994 (2014).
[36] Albrecht, K., Matsuoka, K., Fujita, K. & Yamamoto, K. Angew. Chem. Int. Ed. 54, 5677−5682 (2015).
[37] Shizu, K., Noda, H., Tanaka, H., Taneda, M., Uejima, M., Sato, T., Tanaka, K., Kaji, H. & Adachi, C. J. Phys. Chem. C 119, 26283−26289 (2015).
[38] Sun, J. W., Baek, J. Y., Kim, K.-H., Moon, C.-K., Lee, J.-H., Kwon, S.-K., Kim, Y.-H. & Kim, J.-J. Chem. Mater. 27, 6675−6681 (2015).
[39] Nasu, K., Nakagawa, T., Nomura, H., Lin, C.-J., Cheng, C.-H., Tseng, M.-R., Yasuda, T. & Adachi, C. Chem. Commun. 49, 10385−10387 (2013).
[40] Lee, S. Y., Yasuda, T., Yang, Y. S., Zhang, Q. & Adachi, C. Angew. Chem. Int. Ed. 53, 6402−6406 (2014).
[41] Lee, S. Y., Yasuda, T., Park, I. S. & Adachi, C. Dalton Trans. 44, 8356−8359 (2015).
[42] Zhang, Q., Kuwabara, H., Potscavage Jr, W. J., Huang, S., Hatae, Y., Shibata, T. & Adachi, C. J. Am. Chem. Soc. 136, 18070−18081 (2014).
[43] Lee, J., Shizu, K., Tanaka, H., Nomura, H., Yasuda, T. & Adachi, C. J. Mater. Chem. C 1, 4599−4604 (2013).
[44] Tanaka, H., Shizu, K., Lee, J. & Adachi, C. J. Phys. Chem. C 119, 2948−2955 (2015).
[45] Zhang, Q., Li, B., Huang, S., Nomura, H., Tanaka, H. & Adachi, C. Nat. Photon. 8, 326−332 (2014).
[46] Zhang, Q., Li, J., Shizu, K., Huang, S., Hirata, S., Miyazaki, H. & Adachi, C. J. Am. Chem. Soc. 134, 14706−14709 (2012).
[47] Wu, S., Aonuma, M., Zhang, Q., Huang, S., Nakagawa, T., Kuwabara, K. & Adachi, C. J. Mater. Chem. C 2, 421−424 (2014).
[48] Zhang, Q., Tsang, D., Kuwabara, H., Hatae, Y., Li, B., Takahashi, T., Lee, S. Y., Yasuda, T. & Adachi, C. Adv. Mater. 27, 2096−2100 (2015).
[49] Xu, S., Liu, T., Mu, Y., Wang, Y. F., Chi, Z., Lo, C. C., Liu, S., Zhang, Y., Lien, A. & Xu, J. Angew. Chem. Int. Ed. 54, 874−878 (2015).
[50] Xie, Z., Chen, C., Xu, S., Li, J., Zhang, Y., Liu, S., Xu, J. & Chi, Z. Angew. Chem. Int. Ed. 54, 7181−7184 (2015).
[51] Numata, M., Yasuda, T. & Adachi, C. Chem. Commun. 51, 9443−9446 (2015).
[52] Hirai, H., Nakajima, K., Nakatsuka, S., Shiren, K., Ni, J., Nomura, S., Ikuta, T. & Hatakeyama, T. Angew. Chem. Int. Ed. 54, 13581−13585 (2015).
[53] Kitamoto, Y., Namikawa, T., Ikemizu, D., Miyata, Y., Suzuki, T., Kita, H., Sato, T. & Oi, S. J. Mater. Chem. C 3, 9122−9130 (2015).
[54] Suzuki, K., Kubo, S., Shizu, K., Fukushima, T., Wakamiya, A., Murata, Y., Adachi, C. & Kaji, H. Angew. Chem. Int. Ed. 54, 15231−15235 (2015).
[55] Park, I. S., Numata, M., Adachi, C. & Yasuda, T. Bull. Chem. Soc. Jpn. 89, 375-377 (2016).