© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 2777wileyonlinelibrary.com

CO

MM

UN

ICATIO

N

Ultrapure Blue Thermally Activated Delayed Fluorescence Molecules: Effi cient HOMO–LUMO Separation by the Multiple Resonance Effect

Takuji Hatakeyama ,* Kazushi Shiren , Kiichi Nakajima , Shintaro Nomura , Soichiro Nakatsuka , Keisuke Kinoshita , Jingping Ni , Yohei Ono , and Toshiaki Ikuta

Prof. T. Hatakeyama, K. Nakajima, S Nakatsuka, K. Kinoshita Department of Chemistry School of Science and Technology Kwansei Gakuin University 2-1 Gakuen, Sanda , Hyogo 669-1337 , Japan E-mail: hatake@kwansei.ac.jp Prof. T. Hatakeyama PRESTO Japan Science and Technology Agency 5, Sanbancho, Chiyoda , Tokyo 102-0075 , Japan Dr. K. Shiren, S. Nomura, Dr. J. Ni, Y. Ono, T. Ikuta JNC Petrochemical Corporation 5-1 Goi Kaigan , Ichihara, Chiba 290-8551 , Japan

DOI: 10.1002/adma.201505491

OLED displays employ a color fi lter and/or an optical micro-cavity [ 17,18 ] in order to satisfy the Commission Internationale de l’Eclairage (CIE) coordinate requirements defi ned by broad-casting standards, the broadening of the original EL peaks sig-nifi cantly reduces the external quantum effi ciency (EQE) of the OLED displays. [ 19 ] An improved approach to overcome these issues might involve the establishment of a new strategy to minimize Δ E ST of the fl uorophores without broadening of the fl uorescence peaks.

Here, we report a novel design of organic molecules that exhibit ultrapure blue fl uorescence based on TADF (Figure 1 b). The molecule shown in the fi gure is triphenylboron, possessing two nitrogen atoms, which combine neighboring phenyl groups to construct a rigid polycyclic aromatic framework. Since the nitrogen atom exhibits the opposite resonance effect of the boron atoms, para-substitution relative to it can enhance the resonance effect and can signifi cantly separate the HOMO and LUMO without the need to introduce donor or acceptor groups. The calculated molecular orbitals of the parent mole-cule DABNA-1 are shown in Figure 2 : the LUMOs are localized on the boron atom and at the ortho and para positions relative to it, whereas the HOMOs are localized on the nitrogen atoms and at the meta position relative to the boron atom.

Because of the rigid π-conjugated framework and the large oscillator strength of the S 0 –S 1 transition ( f = 0.205), OLEDs employing DABNA-1 exhibited an emission at 459 nm with an FWHM of 28 nm, CIE coordinates of (0.13, 0.09), and a high EQE of 13.5%. The introduction of substituents improved the oscillator strength ( f = 0.415) without affecting the localization of MOs, as demonstrated by derivative DABNA-2 in Figure 2 , which enabled us to fabricate a blue OLED exhibiting an emis-sion at 467 nm with an FWHM of 28 nm, CIE coordinates of (0.12, 0.13), and an EQE of 20.2%.

Compound DABNA-1 was synthesized in two steps from a commercially available starting material, 1-bromo-2,3-dichlorobenzene, as shown in Scheme 1 , left. In the pres-ence of a catalytic amount of dichlorobis[di- tert -butyl( p -dimethyl-aminophenyl)phosphino]palladium(II) and a stoichiometric amount of sodium tert -butoxide, a cross-coupling reaction with diphenylamine smoothly took place at 80–120 °C to give intermediate 1 in 66% yield. A boron atom was introduced in one pot via a lithium–chloride exchange reaction with tert -butyllithium, electrophilic trapping with boron tribromide, and tandem electrophilic arene borylation [ 20–22 ] in the pres-ence of a stoichiometric amount of diisopropylethylamine to give DABNA-1 in 32% yield. According to the same strategy,

Organic light-emitting diodes (OLEDs) [ 1 ] play an important role in new-generation fl at-panel displays and solid-state lighting. The internal quantum effi ciency (IQE) of OLEDs employing a conventional fl uorophore is up to 25% because triplet exci-tons, constituting 75% of the electrically generated excitons, do not contribute to the emission. The IQE can be increased to ≈100% using phosphorescent materials such as Ir and Pt complexes owing to the heavy-atom effect that accelerates the intersystem crossing (ISC). [ 2 ] However, phosphorescent OLEDs (PHOLEDs) present potential drawbacks such as the high cost of the heavy metals and the limited availability of pure blue phosphorescent materials. [ 3 ] Therefore, even the newest OLED displays employ blue fl uorophores together with triplet–triplet annihilation, for which the theoretical maximum IQE is only 62.5%. [ 4,5 ] Recently, another promising pathway—thermally activated delayed fl uorescence (TADF)—achieved an IQE of ≈100% without employing any precious heavy metals. [ 6–16 ] In order to achieve the optimal performance in TADF-based OLEDs through effi cient upconversion from the T 1 state to the S 1 state, a fl uorophore with a small energy gap (Δ E ST ) between S 1 and T 1 is required. A common strategy for minimizing Δ E ST involves introducing donor and acceptor groups, which can separate the highest occupied and lowest unoccupied molec-ular orbitals (HOMO and LUMO, respectively) and reduce the exchange interaction between singly occupied molecular orbitals (SOMOs) in the excited states ( Figure 1 a). However, such a strategy enhances the structural relaxation in the excited states and increases the Stokes shift, resulting in a broadening of the electroluminescence (EL) peaks, whose typical full-width at half-maximum (FWHM) is 70–100 nm. Since commercial

Adv. Mater. 2016, 28, 2777–2781

www.advmat.dewww.MaterialsViews.com

2778 wileyonlinelibrary.com © 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

CO

MM

UN

ICATI

ON

DABNA-2 was synthesized in three steps from 1-bromo-2,3-dichlorobenzene (Scheme 1 , right). A sequential cross-coupling with N 1 , N 1 , N 3 -triphenylbenzene-1,3-diamine (90 °C) and N 1 -(2,3-dichlorophenyl)- N 1 , N 3 , N 3 -triphenylbenzene-1,3-di-amine (120 °C) gave intermediate 2 in 71% yield, which was readily converted to DABNA-2 in 31% yield via the one-pot borylation reaction. Notably, the tandem electrophilic arene borylation took place at the para position relative to the diphe-nylamino group probably because of its electron-donating prop-erty and bulkiness. These processes were robust and scalable and thus enabled us to prepare 6.0 g of the target compounds.

The photophysical properties of the two compounds are listed in Table 1 . The UV–visible absorption spectra measured in CH 2 Cl 2 exhibit a strong absorption band corresponding to the HΟMΟ−LUMO transition with maxima at 437 and 444 nm

for DABNA-1 and DABNA-2 , respectively (log ε = 4.39 and 4.63, where ε is the molar excitation coeffi cient). The 298 K fl uores-cence spectra of DABNA-1 and DABNA-2 exhibit strong and sharp emission bands at 462 and 470 nm. Notably, the Stokes shifts and FWHMs are very small, 25 and 33 nm for DABNA-1 and 26 and 34 nm for DABNA-2, respectively, because of their rigid π-conjugated framework. Similar emission spectra were observed in EtOH, suggesting negligible interac-tion between the boron atoms of these compounds and the oxygen atom of EtOH. Similar photophysical properties were observed for 1 wt%-doped fi lms of DABNA-1 and DABNA-2 in 3,3′-bis( N -carbazolyl)-1,1′-biphenyl (mCBP), one of the more common host materials for blue emitters. Notably, high abso-lute PL quantum effi ciencies ( Φ = 0.84–0.90) were observed for both compounds, not only in solution but also in the solid state. To estimate Δ E ST value, we obtained the emission spectra of the solutions and fi lms with various delay times at 77 K (see the Supporting Information for details); two independent emis-sions with different lifetimes were observed and assigned to fl uorescence and phosphorescence, respectively. Δ E ST values in the range of 0.14–0.18 eV were estimated from the emission maxima of each spectrum. As shown in Table 2 , the quantum yields ( Φ F , Φ TADF ) and lifetimes ( τ F , τ TADF ) of the fl uorescence and TADF components were determined from the total Φ and

Adv. Mater. 2016, 28, 2777–2781

www.advmat.dewww.MaterialsViews.com

Figure 1. a) Conventional design and b) new design for effi cient HOMO–LUMO separation. The localized HOMO and LUMO are indicated by the blue and red colors, respectively. Typical absorption and emission spectra are indicated by the black and green lines, respectively. The blue emis-sion spectrum of a commercial OLED display is indicated by the green dotted line.

Figure 2. Orbital localization and energies. Highest occupied and lowest unoccupied Kohn–Sham orbitals of molecules DABNA-1 (left) and DABNA-2 (right) in the S 0 state calculated at the B3LYP/6–31G(d) level of theory. The oscillator strength ( f ) of the S 0 –S 1 transition calculated by TDDFT is shown above the corresponding arrow. The orbital energies are shown in parentheses.

2779wileyonlinelibrary.com© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

CO

MM

UN

ICATIO

N

the proportion of the integrated area of each of the components in the transient spectra to the total integrated area (see the Sup-porting Information for details). From the quantum yields and

lifetimes, Δ E ST value was estimated to be 0.20 eV according to Adachi’s method, [ 11,16 ] which is comparable to those estimated from the emission maxima and suitable for effi cient upconver-sion from T 1 to S 1 at room temperature. [ 9,11,12 ] The radiative rate constants of DABNA-2 ( k F = 14.1 × 10 7 s −1 ) calculated from Φ F and τ F are larger than those of DABNA-1 ( k F = 9.6 × 10 7 s −1 ), consistent with its larger molar excitation coeffi cient and oscil-lator strength.

To demonstrate the potential of the present pure blue emit-ters, devices with the following structure were fabricated: indium tin oxide (ITO) (50 nm); N,N ′-di(1-naphthyl)- N,N ′-diphenyl-(1,1′-biphenyl)-4,4′-diamine (NPD) (40 nm); tris(4-carbazolyl-9-ylphenyl)amine (TCTA) (15 nm); 1,3-bis( N -carb-azolyl)benzene (mCP) (15 nm); 1 wt% DABNA-1 or DABNA-2 emitter; mCBP (20 nm); diphenyl-4-triphenylsilylphenylphos-phine oxide (TSPO1) (40 nm); LiF (1 nm); and Al (100 nm). The EL characteristics, the ionization potentials, and the electron affi nities of these devices are shown in Figures 3 and 4 : the device employing DABNA-1 as an emitter exhibited a pure blue emission at 459 nm with an FWHM of 28 nm, which is even smaller than the FWHM of devices employing the latest blue-fl uorescent emitters (40–60 nm) [ 23–26 ] and comparable to the FWHM of a commercial OLED display. [ 19 ] The corresponding CIE coordinates are (0.13, 0.09), which are very close to the (0.14, 0.08) requirements defi ned by the National Television System Committee. These results suggest that the device will not suffer from an effi ciency loss caused by a color fi lter or an optical microcavity for display applications. Although a serious effi ciency roll-off was observed at a high luminance, the max-imum EQE (13.5%) and the current and power effi ciencies (10.6 cd A −1 and 8.3 lm W −1 , respectively, at 0.6 cd m −2 ) are suffi ciently high for a blue OLED. The device performance was dramatically improved upon the introduction of substituents: the device employing derivative DABNA-2 exhibited excellent EL performance (with the EQE, current effi ciency, and power effi ciency being equal to 20.2%, 21.1 cd A −1 , and 15.1 lm W −1 , respectively, at 3.2 cd m −2 and equal to 13.4%, 14.2 cd A −1 , and

Adv. Mater. 2016, 28, 2777–2781

www.advmat.dewww.MaterialsViews.com

Scheme 1. Synthesis of DABNA-1 and DABNA-2 a . a a) HNPh 2 (2.2 equiv), t -BuOK (2.5 equiv), (AMPHOS) 2 PdCl 2 (1.0 mol%), o -xylene, 80 °C, 2 h then 120 °C, 3 h. b) t -BuLi (1.2 equiv), t -butylbenzene, 60 °C, 2 h; BBr 3 (1.2 equiv), rt, 0.5 h; EtN( i -Pr) 2 (2.0 equiv), 120 °C, 3 h. c) HNPh( m -Ph 2 N-C 6 H 4 ) (1.0 equiv), t -BuOK (1.5 equiv), (AMPHOS) 2 PdCl 2 (0.5 mol%), o -xylene, 90 °C, 2.5 h. d) HN( m -Ph-C 6 H 4 ) 2 (1.0 equiv), t -BuOK (1.5 equiv), (AMPHOS) 2 PdCl 2 (1.0 mol%), o -xylene, 120 °C, 1 h. e) t -BuLi (2.0 equiv), t -butylbenzene, 60 °C, 3 h; BBr 3 (2.0 equiv), rt, 0.5 h; EtN( i -Pr) 2 (2.0 equiv), 120 °C, 1.5 h. AMPHOS: di- t -butyl( p -dimethylaminophenyl)-phosphine.

Table 1. Photophysical data of DABNA-1 and DABNA-2.

Absorption Fluorescence Phosphorescence

Compound Conditions T [K]

λ ab a) [nm]

log ε λ em a) [nm]

E S b) [eV]

FWHM c) [nm]

Φ d) τ F e) [ns]

λ em a) [nm]

E T b) [eV]

ΔE ST b) [eV]

DABNA-1 2.0 × 10 −5 M in CH 2 Cl 2 298 437 4.39 462 2.68 33 0.89 9.5 – – –

Saturated in EtOH 298 – – 458 2.71 36 0.84 11.1 – – –

Saturated in EtOH 77 – – 453 2.74 48 – – 478 2.59 0.15

1 wt% in mCBP f) 298 – – 460 2.70 30 0.88 g) 8.8 g) – – –

1 wt% in mCBP f) 77 – – 464 2.67 27 – – 497 2.49 0.18

DABNA-2 2.0 × 10 −5 M in CH 2 Cl 2 298 444 4.63 470 2.64 34 0.85 7.0 – – –

Saturated in EtOH 298 – – 463 2.68 34 0.90 7.3 – – –

Saturated in EtOH 77 – – 458 2.71 24 – – 485 2.56 0.15

1 wt% in mCBP f) 298 469 2.64 28 0.90 g) 6.0 g) – – –

1 wt% in mCBP f) 77 – – 475 2.61 40 – – 503 2.47 0.14

a) Maximum wavelength of ultraviolet–visible absorption and emission; b) singlet and triplet energies estimated from the emission maxima. Δ E ST = E S – E T ; c) full-width at half-maximum; d) absolute photoluminescence quantum yield; e) lifetime calculated from the fl uorescence decay; f) deposited thin fi lms of 3,3′-bis( N -carbazolyl)-1,1′-biphenyl (mCBP) doped with DABNA-1 or DABNA-2; g) measured at 300 K.

2780 wileyonlinelibrary.com © 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

CO

MM

UN

ICATI

ON

7.9 lm W −1 , respectively, at 100 cd m −2 ). The maximum IQE was estimated to be ≈100% by assuming that the light out-coupling effi ciency was 20%. Since derivative DABNA-2 exhibited com-parable values of Φ and Δ E ST to those of DABNA-1 , its higher k F ( DABNA-1 : 9.6 × 10 7 s −1 , DABNA-2 : 14.1 × 10 7 s −1 ) and k RISC ( DABNA-1 : 9.9 × 10 3 s −1 , DABNA-2 : 14.8 × 10 3 s −1 ) may account for the signifi cant improvement. [ 9,11,12 ] Unfortunately, the maximum luminance cannot reach 1000 cd m −2 owing to the serious effi ciency roll-off. Since the roll-off was slightly improved by using 5 wt% of DABNA s, [ 27 ] we currently assume that the relatively small k RISC and charge carrier imbalance in the devices cause bimolecular quenching process such as triplet–triplet and exiton–polaron annihilation at high current densities. [ 28 ] Notably, the EL spectrum of DABNA-2 was slightly redshifted ( λ em = 467 nm, CIE = (0.12, 0.13)) but not broadened at all (FWHM = 28 nm) compared to that of DABNA-1 , prob-ably because of the localization of the HOMO and LUMO at the rigid-core structure rather than at the substituents (Figure 2 ). These results suggest that the device performance can be fur-ther improved by an appropriate choice of substituents without a signifi cant loss in the color purity.

In conclusion, we have designed and synthesized organoboron-based ultrapure blue emitters having small Δ E ST , high Φ , and excellent color purity. The device employing emitter DABNA-2 exhibited pure blue emission at 467 nm with a narrow FWHM of 28 nm, CIE coordinates of (0.12, 0.13), and an IQE of ≈100%, which represent record-setting performance for blue OLEDs. We believe that the novel molecular design for

Adv. Mater. 2016, 28, 2777–2781

www.advmat.dewww.MaterialsViews.com

Figure 3. Characteristics of OLED devices using DABNA-1 (blue line) and DABNA-2 (red line) as the emitter. a) Electroluminescence spectra, b) current-density–driving-voltage characteristics, c) luminance–driving-voltage characteristics, d) external-quantum-effi ciency–luminance characteristics, e) current-effi ciency–luminance characteristics, and f) power-effi ciency–luminance characteristics.

Figure 4. Energy diagram of OLED materials fabricated in this study. The ionization potential and electron affi nity (in eV) for each material are displayed.

Table 2. Photophysical data of DABNA-1 and DABNA-2 in mCBP fi lms (1 wt%) at 300 K.

Compound Φ a) Φ F b) Φ TADF b) τ F c) [ns] τ TADF c) [µs] k F d) × 10 7 [s −1 ] k IC d) × 10 7 [s −1 ] k ISC d) × 10 6 [s −1 ] k RISC d) × 10 3 [s −1 ] ΔE ST d) [eV]

DABNA-1 0.880 0.845 0.035 8.8 93.7 9.6 1.3 4.5 9.9 0.20

DABNA-2 0.897 0.843 0.054 6.0 65.3 14.1 1.6 10.1 14.8 0.20

a) Absolute photoluminescence quantum yield; b) the fl uorescent and TADF components are determined from the total Φ and the proportion of the integrated area of each of the components in the transient spectra to the total integrated area; c) lifetimes calculated from the fl uorescence decay; d) fl uorescence decay rate ( k F ), internal conversion decay rate from S 1 to S 0 ( k IC ), intersystem crossing decay rate from S 1 to T 1 ( k ISC ), reverse intersystem crossing decay rate from T 1 to S 1 ( k RISC ), and Δ E ST are calculated from Φ , Φ F , Φ TADF , τ F , and τ TADF according to Adachi’s method reported in refs. [ 11 ] and [ 16 ] .

2781wileyonlinelibrary.com© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

CO

MM

UN

ICATIO

N

Adv. Mater. 2016, 28, 2777–2781

www.advmat.dewww.MaterialsViews.com

separating the HOMO and LUMO and minimizing Δ E ST can be used to prepare not only excellent TADF emitters but also assistant dopants [ 29 ] for fl uorescent emitters and host materials for phosphorescent emitters.

Supporting Information Supporting Information is available from the Wiley Online Library or from the author.

Acknowledgements This work was supported by the Precursory Research for Embryonic Science and Technology (PRESTO) from the Japan Science and Technology Agency (JST), a Grant-in-Aid for Scientifi c Research on Innovative Areas “π-System Figuration: Control of Electron and Structural Dynamism for Innovative Functions” (15H01004), and a Grant-in-Aid for Scientifi c Research (26288095) from the Ministry of Education, Culture, Sports, Science & Technology in Japan (MEXT). Preliminary measurements for the X-ray crystal-structure analysis were performed at the BL40XU in the SPring-8 with the approval of JASRI (2014B1815, 2015A1320). The authors are grateful to Dr. Nobuhiro Yasuda (JASRI) and Prof. Hikaru Takaya (Kyoto University) for their support with the X-ray diffraction measurements. The authors are also grateful to Dr. Takeshi Matsushita (JNC Petrochemical Corporation) for valuable discussions and Dr. Kenichiro Ikemura (Hamamatsu Photonics) for PL measurements.

Received: November 6, 2015 Revised: December 24, 2015

Published online: February 11, 2016

[1] C. W. Tang , S. A. VanSlyke , Appl. Phys. Lett. 1987 , 51 , 913 . [2] M. A. Baldo , D. F. O’Brien , Y. You , A. Shoustikov , S. Sibley ,

M. E. Thompson , S. R. Forrest , Nature 1998 , 395 , 151 . [3] K. S. Yook , J. Y. Lee , Adv. Mater. 2012 , 24 , 3169 . [4] K. Okumoto , H. Kanno , Y. Hamaa , H. Takahashi , K. Shibata , Appl.

Phys. Lett. 2006 , 89 , 063504 . [5] C.-J. Chiang , A. Kimyonok , M. K. Etherington , G. C. Griffi ths , V. Jankus ,

F. Turksoy , A. P. Monkman , Adv. Funct. Mater. 2013 , 23 , 739 . [6] H. Uoyama , K. Goushi , K. Shizu , H. Nomura , C. Adachi , Nature

2012 , 492 , 234 . [7] Q. Zhang , J. Li , K. Shizu , S. Huang , S. Hirata , H. Miyazaki ,

C. Adachi , J. Am. Chem. Soc. 2012 , 134 , 14706 . [8] J. Li , T. Nakagawa , J. MacDonald , Q. Zhang , H. Nomura ,

H. Miyazaki , C. Adachi , Adv. Mater. 2013 , 25 , 3319 .

[9] Q. Zhang , B. Li , S. Huang , H. Nomura , H. Tanaka , C. Adachi , Nat. Photonics 2014 , 8 , 326 .

[10] S. Wu , M. Aonuma , Q. Zhang , S. Huang , T. Nakagawa , K. Kuwabara , C. Adachi , J. Mater. Chem. C 2014 , 2 , 421 .

[11] Q. Zhang , H. Kuwabara , W. J. Potscavage Jr ., S. Huang , Y. Hatae , T. Shibata , C. Adachi , J. Am. Chem. Soc. 2014 , 136 , 18070 .

[12] 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 , C. Adachi , Nat. Mater. 2015 , 14 , 330 .

[13] H. Wang , L. Xie , Q. Peng , L. Meng , Y. Wang , Y. Yi , P. Wang , Adv. Mater. 2014 , 26 , 5198 .

[14] J. W. Sun , J.-H. Lee , C.-K. Moon , K.-H. Kim , H. Shin , J.-J. Kim , Adv. Mater. 2014 , 26 , 5684 .

[15] K. Kawasumi , T. Wu , T. Zhu , H. S. Chae , T. Van Voorhis , M. A. Baldo , T. M. Swager , J. Am. Chem. Soc. 2015 , 137 , 11908 .

[16] H. Kaji , H. Suzuki , T. Fukushima , K. Shizu , K. Katsuaki , S. Kubo , T. Komino , H. Oiwa , F. Suzuki , A. Wakamiya , Y. Murata , C. Adachi , Nat. Commun. 2015 , 6 , 8476 .

[17] T.-Y. Cho , C.-L. Lin , C.-C. Wu , Appl. Phys. Lett. 2006 , 88 , 111106 . [18] D. Poitras , C.-C. Kuo , C. Py , Opt. Express 2008 , 16 , 8003 . [19] The FWHMs of the RGB emission spectra of the Galaxy S5

(Samsung) were 23, 33, and 43 nm (see the Supporting Informa-tion for details) .

[20] T. Hatakeyama , S. Hashimoto , S. Seki , M. Nakamura , J. Am. Chem. Soc. 2011 , 133 , 18614 .

[21] T. Hatakeyama , S. Hashimoto , T. Oba , M. Nakamura , J. Am. Chem. Soc. 2012 , 134 , 19600 .

[22] H. Hirai , K. Nakajima , S. Nakatsuka , K. Shiren , J. Ni , S. Nomura , T. Ikuta , T. Hatakeyama , Angew. Chem. Int. Ed. Engl. 2015 , 54 , 13581 .

[23] K.-C. Wu , P.-J. Ku , C.-S. Lin , H.-T. Shih , F.-I. Wu , M.-J. Huang , J.-J. Lin , I.-C. Chen , C.-H. Cheng , Adv. Funct. Mater. 2008 , 18 , 67 .

[24] S. J. Lee , J. S. Park , K.-J. Yoon , Y.-I. Kim , S.-H. Jin , S. K. Kang , Y.-S. Gal , S. Kang , J. Y. Lee , J.-W. Kang , S.-H. Lee , H.-D. Park , J.-J. Kim , Adv. Funct. Mater. 2008 , 18 , 3922 .

[25] J.-Y. Hu , Y.-J. Pu , F. Satoh , S. Kawata , H. Katagiri , H. Sasabe , J. Kido , Adv. Funct. Mater. 2014 , 24 , 2064 .

[26] S. Zhang , L. Yao , Q. Peng , W. Li , Y. Pan , R. Xiao , Y. Gao , C. Gu , Z. Wang , P. Lu , F. Li , S. Su , B. Yang , Y. Ma , Adv. Funct. Mater. 2015 , 25 , 1755 .

[27] The use of 5 wt% content of DABNA-2 improved the device per-formance (EQE, current effi ciency, and power effi ciency of 20.0%, 23.8 cd A −1 , and 17.0 lm W −1 at 4.1 cd m −2 and of 14.8%, 18.0 cd A −1 , and 10.1 lm W −1 at 100 cd m −2 , respectively), but the EL spectrum was redshifted ( λ em = 471 nm, CIE = (0.12, 0.17)) because of con-centration quenching (see the Supporting Information for details) .

[28] C. Murawski , M. C. Gather , Adv. Mater. 2013 , 25 , 6801 . [29] H. Nakanotani , T. Higuchi , T. Furukawa , K. Masui , K. Morimoto ,

M. Numata , H. Tanaka , Y. Sagara , T. Yasuda , C. Adachi , Nat. Commun. 2014 , 5 , 4016 .