DOI: 10.1002/chem.201403278

Full Paper

& OLEDs

Trifluoromethylation of Tetraphenylborate Counterions in Cationic Iridium(III) Complexes: Enhanced Electrochemical Stabilities, Charge-Transport Abilities, and Device Performance Dongxin Ma, Lian Duan,* Yongge Wei, and Yong Qiu[a]

Abstract: Trifluoromethylation of tetraphenlyborate counterions was successfully used to improve the electrochemical stabilities and device performance of cationic iridium(III) complexes. Melioration of the thermal, photoluminescent, electrochemical, and electrophosphorescent characteristics was achieved. Interionic hydrogen bonds were first found between the aromatic hydrogen atoms in the ancillary ligands of cations and the fluorine atoms in the trifluoromethyl groups of the anions. The strong impact of the counter-

ions on the charge transport in the devices was investigated. A compound with two trifluoromethyl groups in the tetraphenlyborate ion shows the highest photoluminescent efficiency, the best electrochemical stability, and the greatest performance in green-blue-emitting devices, with a high current efficiency of 12.4 cd A1 and an emission peak at l = 480 nm. The efficiencies achieved are the highest reported for OLEDs with ionic complexes emitting in the blue-green region.

1. Introduction

from a quite low current density and showed no luminance, suggesting that the counterions could affect the electroluminescent properties of the cationic iridium complexes. The poor electrochemical stabilities caused by four electron-rich phenyl groups around the boron atom in [Bph4] may be the chief culprit, leading to a low charge-transport ability. As a result, we expected that introducing electron-withdrawing trifluoromethyl groups could reduce the electron densities of the tetraphenylborate counterions, thereby increase the electrochemical stabilities and the charge-transport abilities of the corresponding complexes, thus improving the performance of the devices. To test this hypothesis, we employed two other counterions, that is, tetrakis[4-(trifluoromethyl)phenyl]borate ([BArF12]) and tetrakis[3,5-bis(trifluoromethyl)-phenyl]borate ([BArF24]), trifluoromethylated at different levels, respectively. In this paper, two series of novel cationic iridium(III) complexes (complexes 1–6 in Scheme 1) with [Bph4] , [BArF12] , and [BArF24] were designed and synthesized. All of their single crystals were successfully grown and characterized. All of their structural, thermal, photophysical, and electrochemical properties were investigated in detail. Green-blue- and blue-emitting single-layer OLEDs based on them were then demonstrated. Interionic hydrogen bonds were firstly found between the aromatic hydrogen atoms in the ancillary ligands of cations and the fluorine in the trifluoromethyl groups of the anions. The strong impact of the counterions on the charge transport in the devices was first investigated. Experiments indicate that the electrochemical stabilities, the charge-transport abilities, and the device performance are sharply enhanced by trifluoromethylation, with a highest current efficiency of 12.4 cd A1 and an emission peak at l = 480 nm obtained. With our hypothesis proven, trifluoromethylation of tetraphenylborate counterions is a simple

The current strong interest in organic electroluminescent materials was connected mostly with the development of organic light-emitting diodes (OLEDs) in the past several decades.[1] Ionic transition-metal complexes (iTMCs) as triplet phosphorescent emitters have attracted much attention of researchers. Ionic, mostly cationic, iridium(III) complexes with excellent photochemical stabilities and efficient emission of virtually all colors have emerged as a promising candidate among iTMCs.[2, 3] Up to now, tremendous efforts have been made to exploit efficient cationic iridium(III) complexes with numerous results achieved, and the applications in OLEDs are ever-accelerated. However, the development of highly efficient blue phosphorescent emitters in the field of OLEDs is still in its infancy. One of the best known reasons is that the photoluminescent quantum yields (PLQYs) of blue emitters are barely satisfactory. Because cationic iridium(III) complexes in previous literatures always share hexafluorophosphate (PF6 )[4–11] or trifluoromethanesulphate (CF3SO3 )[12–14] as counterions, we proposed and proved as a new strategy that the introduction of bulky counterions such as tetrakis(phenyl)borate ([Bph4]) could sharply improve the PLQYs in solids.[15] However, OLEDs based on cationic iridium complexes with [Bph4] as counterion suffered [a] D. Ma, Dr. L. Duan, Prof. Y. Wei, Prof. Y. Qiu Key Lab of Organic Optoelectronics and Molecular Engineering of the Ministry of Education, 100084 (P.R. China) E-mail: [email protected] Supporting information (containing full details on the synthetic routes, the thermogravimetric analysis, and parts of the cyclic voltammograms) for this article is available on the WWW under http://dx.doi.org/10.1002/ chem.201403278. Chem. Eur. J. 2014, 20, 1 – 11

These are not the final page numbers! ÞÞ

1

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

&

&

Full Paper but effective way to improve the performance of devices of cationic iridium complexes. It also suggests that the impact of the counterions on the charge transport is significant and only counterions with good electrochemical stabilities and chargetransport abilities are suitable for cationic iridium(III) complexes as emitters in OLEDs.

2.2. Crystal structures and hydrogen bonds

To further confirm the molecular structures, single crystals of complexes 1–6 were successfully grown and characterized by X-ray diffraction crystallography. As shown in Figure 1, consistent with previous literatures,[16] all the crystals exhibit distorted octahedral geometries around the iridium center with the two cyclometalated ligands adopting C,C-cis, N,N-trans configurations in the cations. Yet in the counterions, the crystals ex2. Results and Discussion hibit tetrahedral geometries around the boron atom with four 2.1. Synthesis of complexes 1–6 identical aromatic ligands. Crystals of the different complexes show different space groups, and especially, there is a polar As shown in Scheme 1, complexes 1–3 share the same cation, space group in the single crystal of complex 6. All the ions [Ir(ppy)2(pzpy)] + , where ppy is 2-phenylpyridine and pzpy is 2rank towards a special orientation and there is no symmetry (1 H-pyrazol-1-yl)-pyridine. Whereas complexes 4–6 share ancenter in the crystal, indicating that complex 6 may be a potenother cation, [Ir(dfppy)2(pzpy)] + , where dfppy is 2-(2,4-difluorotial ferroelectric material. As depicted in Table 1 and Figure 2, the iridiumcarbon (IrC) or iridiumnitrogen (IrN) bonds in the same cation of different compounds are always similar, not affected by different counterions. Nor are boroncarbon (BC) bonds in the same anion changed by different cations. Volumes of the counterions were obtained by quantum chemical calculations on the basis of the single-crystal data. For [Bph4] , [BArF12] , and [BArF24] , the volumes are 288.44, 334.63, and 425.02 cm3 mol1, respectively, increased in sequence, suggesting that trifluoromethylation can also increase volumes. It is worth mentioning that interionic hydrogen bonds are first Scheme 1. Chemical structures of complexes 1–6. Ppy = 2-phenylpyridine, pzpy = 2-(1 H-pyrazol-1-yl)-pyridine, found in complexes with trifluordfppy = 2-(2,4-difluorophenyl)-pyridine. omethylated counterions. As la-

phenyl)-pyridine. The counterions are [Bph4] (for complexes 1 and 4), [BArF12] (for complexes 2 and 5) and [BArF24] (for complexes 3 and 6). The synthetic route towards complexes 4–6 is quite similar to that of complexes 1–3 described in our previous work (see Scheme S1 in the Supporting Information).[15] Complexes 4–6 were synthesized from a dichlorobridged diiridium complex [Ir(dfppy)2Cl]2 and the pzpy ligand, then the chloride ion (Cl) was replaced by [Bph4] , [BArF12] , and [BArF24] with a simple ion-exchange reaction, respectively. All the six complexes were fully characterized by ESI (electrospray ionization) mass spectrometry, 1H NMR spectroscopy, and elemental analysis.

Table 1. Selected bond lengths of the counterions of complexes 1–6. For the bond labeling see Figure 2.

1 2 3 4 5 6

&

&

Chem. Eur. J. 2014, 20, 1 – 11

www.chemeurj.org

2

Bond a []

Bond b []

Bond c []

Bond d []

Bond e []

2.004(4) 2.015(4) 2.004(7) 2.051(6) 2.043(6) 2.044 (6) 2.033(8) 2.069(8) 2.051(4) 2.046(4) 1.978(11) 2.011(12)

2.050(3) 2.051(3) 2.008(7) 2.048(5) 2.009(8) 2.016(7) 1.933(12) 2.000(10) 2.000(6) 2.012(5) 2.013(8) 2.033(10)

2.138(3)

2.168(3)

2.134(6)

2.165(6)

2.142(6)

2.148(6)

2.170(6)

2.159(8)

2.129(4)

2.119(4)

2.157(9)

2.181(8)

1.632(6), 1.636(6) 1.643(5), 1.654(6) 1.634(10), 1.643(10) 1.646(11), 1.650(10) 1.630(10), 1.634(9) 1.640(10), 1.642(10) 1.648(16), 1.652(15) 1.652(15), 1.654(16) 1.633(8), 1.633(8) 1.646(8), 1.663(8) 1.634(13), 1.652(12) 1.652(13), 1.666(14)

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ÝÝ These are not the final page numbers!

Full Paper

Figure 1. Single-crystal structures of complexes a) 1, b) 2, c) 3, d) 4, e) 5, and f) 6. Thermal ellipsoids are drawn at 30 % probability. The solvent molecules and hydrogen atoms have been omitted for clarity. The unlabeled atoms are carbon atoms.

Figure 2. Bonds labeled in the compounds: a) [Ir(ppy)2(pzpy)] + , b) [Ir(dfppy)2(pzpy)] + , c) counterions.

beled in Figure 3, aromatic hydrogen atoms of the pzpy ligands in the cations are chelated to the fluorine atoms of the trifluoromethyl groups in the anions. The hydrogen-bond lengths and the carbon-hydrogen-fluorine (C-H-F) angles are shown in Table 2. Bond lengths varied from 2.4 to 3.5  and the angles range from 101 to 1778, with different strength of interactions. Similar hydrogen bonds have also been found in cationic transition-metal complexes with Cl before,[17] but not yet with PF6 or CF3SO3 .

Chem. Eur. J. 2014, 20, 1 – 11

www.chemeurj.org

These are not the final page numbers! ÞÞ

Figure 3. Hydrogen-bond labeling in the compounds.

2.3. Thermal stabilities Thermal stabilities of complexes 1–6 were investigated by thermogravimetric analysis (TGA) measured under a nitrogen stream. All the novel compounds are thermally stable up to 200 8C, and thermal stabilities were improved gradually with trifluoromethylation of the counterions (see Figure S1 in the 3

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

&

&

Full Paper Table 2. Hydrogen bonds in the complexes. HF distances and C-H-F angles (given in brackets in [8]). For the bond labeling see Figure 3.

2[a] 3 5 6

HaF []

HbF []

HcF []

HdF []

HeF []

HfF []

HgF []

2.774 (121.4)[b] none 2.559 (153.8) 2.542 (143.8)

2.606 (113.6) 2.933 (143.1) none 2.647 (123.6)

none[c] 2.853 (177.4) 2.589 (168.0) 2.651 (164.3)

none 2.710 (120.7) 2.452 (146.4) 2.774 (128.9)

none 3.236 (100.6) 3.310 (109.6) 2.702 (143.5)

none 2.413 (162.5) 3.525 (118.7) 2.857 (136.0)

2.733 2.624 2.236 2.631

(161.6) (131.2) (155.6) (135.5)

[a] Both the distances and angles were taken from the single-crystal data; [b] Angles in the bracket are C-H-F angles; [c] “None” denotes no fluorine atoms were found around the hydrogen atom.

Supporting Information). The 5 % weight reduction temperatures (DT5 %) of complexes 4–6 are 207, 224, and 276 8C, respectively. In contrast, complexes 1–3 are quite stabilized, with DT5 % of 252, 254, and 318 8C, respectively. As a result, complexes with [BArF24] as counterion have best thermal stabilities, promising to fabricate OLEDs with long-term stabilities.

2.4. Photophysical behaviors Figures 4 and 5 show the absorption and photoluminescent (PL) spectra of complexes 1–3 and complexes 4–6, respectively, both in degassed CH3CN solutions and neat films under air. Detailed photophysical behaviors are summarized in Table 3.

Figure 5. Absorption and emission spectra of complexes 4 (&), 5 (*), and 6 (~) in degassed CH3CN solutions (top) and in neat films (bottom).

As depicted in Figure 4, the absorption spectra of complexes 1–3 are quite similar. In the ultraviolet region, the intense absorption bands between l = 250 and 350 nm are ascribed to spin-allowed 1p–p* transitions from the ligands. The weak absorption bands from l = 350 nm extending to l = 450 nm correspond to excitations to 1MLCT (metal-to-ligand charge transfer), 1LLCT (ligand-to-ligand charge transfer), 3MLCT, 3LLCT, and LC (ligand centered) 3p–p* transitions. The tiny influence of different counterions on the absorption indicates that the absorption of the cations is independent and that of the anions is negligible in both solutions and neat films. Also in Figure 4, the PL spectra of complexes 1–3 in degassed CH3CN solutions are similar, with a major peak at l = 474 nm and a shoulder peak at about l = 503 nm. Whereas in

Figure 4. Absorption and emission spectra of complexes 1 (&), 2 (*), and 3 (~) in degassed CH3CN solutions (top) and in neat films (bottom).

&

&

Chem. Eur. J. 2014, 20, 1 – 11

www.chemeurj.org

4

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ÝÝ These are not the final page numbers!

Full Paper Table 3. Photophysical behaviors of complexes 1–6.

1[a] 2 3 4 5 6

Absorption l [nm] (e [104 m1 cm1])[b]

Emission at room temperature Solution l [nm][c] Neat film l [nm][d]

253 250 254 248 246 249

474, 474, 474, 451, 451, 451,

(5.31), (5.88), (5.23), (5.53), (6.14), (4.98),

380 379 379 359 357 359

(0.58) (0.55) (0.54) (0.53) (0.53) (0.51)

502 501 503 479 479 479

(sh) (sh) (sh) (sh) (sh) (sh)

483, 508 (sh) 478 (sh), 506 473, 502 (sh) 455, 482 (sh) 452, 480 (sh) 452, 481 (sh)

Solution f (t [ms])

Neat film f (t [ms])

0.31 0.35 0.34 0.26 0.28 0.26

0.14 0.25 0.39 0.05 0.20 0.17

(1.61) (1.57) (1.56) (1.43) (1.42) [0.84]

[0.30 [0.35 [0.32 [0.02 [0.10 [0.02

(51 %), 3.00] (37 %), 2.70] (27 %), 1.20] (10 %), 0.34] (15 %), 0.97] (6 %), 1.10]

e)

Emission at 77 K[f] l [nm]

t [ms]

469, 468, 467, 453, 454, 451,

4.50 4.60 5.00 3.77 (77 %), 1.11 3.56 (89 %), 0.64 4.28 (86 %), 1.24

503 503 502 481 480 479

(sh) (sh) (sh) (sh), 502 (sh) (sh), 503 (sh) (sh), 504 (sh)

[a] The results of complexes 1–3 were reported in our previous work.[15] [b] In CH3CN solutions (1  105 m). [c] In degassed CH3CN solutions, “sh denotes the shoulder wavelength. [d] Neat films were made on round substrates by solution process in air. [e] The percentage in parentheses denotes the percentage of each lifetime. [f] In CH3CN glass at 77 K.

neat films the PL spectra are quite different. The emission of complex 1 has a major peak at l = 483 nm, and a shoulder peak at l = 508 nm. For complex 3, the major peak is 10 nm blue shifted to l = 473 nm and the shoulder peak is 6 nm blue shifted to l = 502 nm, due to the molecular interactions in complex 3 decreased by the larger counterion. However, for complex 2, locations of major and shoulder peaks are interchanged, for reasons unknown. In Figure 5, the PL spectra of complexes 4–6 in degassed CH3CN solutions are also similar, with a major peak at l = 451 nm and a shoulder peak at l = 479 nm. Whereas the PL spectra in neat films exhibit differences. Complex 4 emits a major peak at l = 455 nm and a shoulder peak at 482 nm. But for complexes 5 and 6 the major peak is centered at l = 452 nm, with a 3 nm blue shift. In addition, the PL spectra of complexes 4 and 5 show long tails extending to l = 700 nm, related to the strong intermolecular interactions in neat films, which is different from complex 6 with the largest counterion. PLQYs of complexes 1–3 and complexes 4–6 in solutions are quite similar, respectively, whereas the PLQYs in solids are sharply increased with larger volumes of counterions, as we expected. For complexes 1–3, the PLQYs are 0.31, 0.35, and 0.34 in solutions, respectively, and 0.14, 0.25, and 0.39 in neat films, respectively. For complexes 4–6, the PLQYs are 0.26, 0.28, and 0.26 in solutions, respectively, and 0.05, 0.20, and 0.17 in neat films, respectively. The harvest in the PLQYs of complex 3 in neat films is attributed to the depressed intermolecular interactions and concentration quenching, indicating an efficient electrophosphorescence in the devices. By the way, for complexes sharing the same cation, lifetimes and emission characteristics at 77 K in CH3CN glass are quite similar (see Table 3).

Table 4. Electrochemical characteristics of complexes 1–6.

1 2 3 4 5 6

These are not the final page numbers! ÞÞ

E ½b HOMO [eV]

ELUMO [eV]

E ½c gap [eV]

0.42 0.87 0.87 0.42 1.15 1.19

2.18 2.18 2.17 2.15 2.11 2.12

5.22 5.67 5.67 5.22 5.95 5.99

2.62 2.62 2.63 2.65 2.69 2.68

2.60 3.05 3.04 2.57 3.26 3.31

0.42 V, ascribed to the anion [Bph4] (see Figure S2 in the Supporting Information). In [Bph4] , the boron center is surrounded by four electron-donating phenyl groups, sharing a high electron density thus becoming easily oxidized. Trifluoromethylation can reduce the electron densities and improve the electrochemical stabilities. Hence, [BArF24] with two trifluoromethyl groups has a better redox stability than [BArF12] with only one trifluoromethyl group. To gain insight into the photophysical and electrochemical behaviors of these six compounds, quantum chemical calculations were performed on the basis of the geometries optimized from their single-crystal structures. The highest occupied molecular orbital energies and the lowest unoccupied molecular orbital energies of both cations and anions in these compounds were obtained, and molecular orbital surfaces are shown in Figure 7, with an isocontour value of j Y j = 0.02. The HOMO energies of [Bph4] , [BArF12] . and [BArF24] are 2.52, 3.66, and 4.36 eV, respectively. The LUMO energies are 3.19, 1.98, and 1.06 eV, respectively. In conclusion, from [Bph4] and [BArF12] to [BArF24] , the energy orbitals are quite stabilized, which is in accord with the electrochemical stabilities mentioned above. On the other hand, the HOMO energies of the cations [Ir(ppy)2(pzpy)] + and [Ir(dfppy)2(pzpy)] + are 7.83 and 8.27 eV, respectively, residing on the ppy and dfppy ligands. Likewise, the LUMO energies are 4.70 and 4.92 eV, respectively, residing on the pzpy ligand. Besides, the energy gap of [Ir(dfppy)2(pzpy)] + is 3.35 eV, which is wider than that of [Ir-

The electrochemical properties of complexes 1–6 were investigated by cyclic voltammetry in degassed CH3CN solutions, and the detailed redox potentials are listed in Table 4. As depicted in Figure 6, all these compounds exhibit reversible oxidation processes and irreversible reduction processes in solutions. The redox potentials of complexes 2 and 3 as well as 5 and 6 are quite similar. For complex 1 and 4, there is an anodic peak at www.chemeurj.org

Reduction potential Ered [V]

[a] Electrochemical data versus Fc + /Fc were collected in degassed CH3CN solution (103 m). [b] The highest occupied molecular orbital (HOMO) energies and lowest unoccupied molecular orbital (LUMO) energies of the compounds were calculated based on the electrochemical data.[18] [c] Egap = EHOMOELUMO.

2.5. Electrochemical properties and theoretical calculations

Chem. Eur. J. 2014, 20, 1 – 11

Oxidation potential Eox [V][a]

5

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

&

&

Full Paper (ppy)2(pzpy)] + (3.13 eV), leading to the bluer emission of complexes 4–6.

2.6. Green-blue- and blue-emitting OLEDs To explore the electroluminescent properties of these compounds, single-layer OLEDs were fabricated with a structure of ITO/PEDOT:PSS (60 nm)/PVK:OXD-7:Ir complex (85 nm)/Cs2CO3 (2.3 nm)/Al (150 nm) (see Figure 8, ITO = indium tin oxide, PE-

Figure 8. Structure of the OLEDs.

DOT:PSS = poly(3,4-ethylenedioxythiophene) polystyrene sulfonate, PVK:OXD-7 = poly(N-vinylcarbazole):1,3-bis(5-(4-tert-butylphenyl)-1,3,4-oxadiazol-2-yl)benzene). The energy level diagrams of the devices are shown in Figure 9. A non-doped OLED with a similar structure of ITO/PEDOT: PSS (60 nm)/PVK: 50 wt % OXD-7 (85 nm)/Cs2CO3 (2.3 nm)/Al (150 nm) was also fabricated for comparison. Both the photoluminescent and the electroluminescent properties of the light-emitting layers of the OLEDs are summarized in Table 5. As we can see, complexes 1–3 and com-

Figure 6. Cyclic voltammogram of complexes 1 (&), 2 (*), and 3 (~) (top) and complexes 4 (&), 5 (*), and 6 (~) (bottom) in CH3CN solutions (1  103 m). Potentials were recorded versus the ferrocenium/ferrocene (Fc + / Fc) couple.

Figure 7. Molecular orbital surfaces of the cations. a) HOMO of [Ir(ppy)2(pzpy)] + . b) LUMO of [Ir(ppy)2(pzpy)] + . c) HOMO of [Ir(dfppy)2(pzpy)] + . d) LUMO of [Ir(ppy)2(pzpy)] + . All the molecular orbital surfaces correspond to an isocontour value of j Y j = 0.02. Hydrogen atoms have been omitted for clarity.

&

&

Chem. Eur. J. 2014, 20, 1 – 11

www.chemeurj.org

Figure 9. Energy level diagrams of a) ITO/PEDOT: PSS/PVK:OXD-7:green-blue Emitters/Cs2CO3/Al (based on complexes 1–3) and b) ITO/PEDOT: PSS/PVK: OXD-7:blue emitters/Cs2CO3/Al (based on complexes 4–6) OLEDs, where the HOMO and LUMO energies were calculated from the redox potentials of the complexes in degassed CH3CN solutions (see Table 4).

6

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ÝÝ These are not the final page numbers!

Full Paper Table 5. Photoluminescent and electroluminescent properties of the light-emitting layers of the devices doped with the different iridium complexes. Emitting layer

PLQY f

PL emission l [nm]

Current density at 15, 18 V J [A m2]

Current efficiency hc, max [cd A1]

PVK:50 wt % OXD-7 PVK:OXD-7:5 wt % 1 PVK:OXD-7:5 wt % 2 PVK:OXD-7:5 wt % 3 PVK:OXD-7:5 wt % 4 PVK:OXD-7:5 wt % 5 PVK: OXD-7:5 wt % 6

0.01 0.22 0.21 0.21 0.06 0.05 0.06

446 478, 507 (sh) 479 (sh), 508 478, 506 (sh) 454, 481 (sh) 454, 481 (sh) 453, 482 (sh)

251.3, 0.192, 242.7, 313.8, 0.754, 39.32, 60.12,

– – 10.8 9.7 – 1.1 0.18

1010 0.688 518.0 935.2 1.819 100.1 281.5

plexes 4–6 show similar phosphorescent properties in PVK: 50 wt % OXD-7 films. However, the electroluminescent characteristics of the OLEDs are quite different. It suggests that cationic complexes with different counterions have different charge-transport abilities in the devices. The current densities of the devices with the different complexes are shown in Figure 10. As we can see, the densities of complex 1 complex 4 are sharply lower than those of the others, and OLEDs based on these two complexes show no lu-

Brightness Bmax [cd m2]

EL emission l [nm]

CIE (x, y)

– – 2600 6700 – 105 113

– – 480 (sh), 508 480, 508 (sh) – 460, 484 (sh) 460, 484 (sh)

– – (0.24, 0.48) (0.21, 0.48) – (0.20, 0.30) (0.17, 0.26)

For blue-emitting devices, the current densities are quite low. The one for a device incorporating complex 5 is even lower than the one for devices based on complex 6, in the same manner. The PLQYs of light-emitting layers are also unsatisfactory, as shown in Table 5. As a result, the performance of OLEDs based on complexes 5 and 6 is quite poor. Electroluminescent (EL) spectra of devices doped with 5 wt % of complexes 2, 3, 5, and 6 are depicted in Figure 11. The emission wavelengths of devices based on complexes 2 and 3 are l = 508 and 480 nm, respectively, similar to the PL spectra in solids. Devices based on complexes 5 and 6 both give a blue emission at l = 460 nm.

Figure 10. Current density plots of the OLEDs based on complexes 1–6. Hexagon = PVK:OXF-7, & = PVK:OXF-7:1, * = PVK:OXF-7 = PVK:OXF-7:2, ~ = PVK:OXF-7:3, ! = PVK:OXF-7:4, ^ = PVK:OXF-7:5, and 3 = PVK:OXF-7:6. Figure 11. EL spectra of OLEDs based on complexes 2 (*), 3 (~), 5 (^), and 6 (3).

minance at even 21 V, which can be explained by the low electrochemical stability and charge-transport ability of [Bph4] . For green-blue-emitting devices, incorporating complex 2, the current density is 242.7 and 518.0 A m2 at 15 and 18 V, respectively, which lower than the current density for devices incorporating complex 3 (313.8 A m2 at 15 V and 935.2 A m2 at 18 V) and the one for non-doped devices (251.3 A m2 at 15 V and 1010 A m2 at 18 V). This suggests that trifluoromethylation of the tetraphenylborate counterions is conductive to the charge transport in the devices, with the performance being improved. Seen from Table 5, the OLED based on complex 2 shows the highest current efficiency, that is, 10.8 cd A1, whereas the OLED based on complex 3 shows a little lower current efficiency of 9.7 cd A1, but a significantly better luminance of 6700 cd m2. As a whole, devices based on complex 3 exhibit the best performance. Chem. Eur. J. 2014, 20, 1 – 11

www.chemeurj.org

These are not the final page numbers! ÞÞ

As OLEDs based on complex 3 show the best performance, a dopant concentration-dependence experiment was carried out in the range between 2 and 20 wt % to optimize the device performance. Both photoluminescent and electroluminescent properties of the light-emitting layers of the OLEDs doped with different concentrations of complex 3 have been summarized as shown below in Table 6. As we can see, for the film doped with 20 wt % of complex 3, the PLQY is quite low. But films doped with 2 to 15 wt % of complex 3 show similar phosphorescent properties but different electroluminescent characteristics. Measurement of the J–V curves provides for a simple comparison of how the counterions influence the device characteristics. Figure 12 shows how the current density (J) and the brightness (B) develop over the voltage (V) with a different doping ratio, respectively. A current-density fluctua7

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

&

&

Full Paper Table 6. Photoluminescent and electroluminescent properties of the light-emitting layers of devices doped with x wt % of complex 3. Emitting layer

PLQY f

PL emission l [nm]

Current density at 15, 18 V J [A m2]

Current efficiency hc, max [cd A1]

PVK:50 wt % OXD-7 PVK:OXD-7:2 wt % 3 PVK:OXD-7:3 wt % 3 PVK:OXD-7:5 wt % 3 PVK:OXD-7:10 wt % 3 PVK:OXD-7:15 wt % 3 PVK:OXD-7:20 wt % 3

0.01 0.26 0.18 0.21 0.17 0.16 0.07

446 480, 479, 479, 479, 480, 481,

251.3, 816.3, 552.3, 313.8, 66.46, 58.89, 32.80,

– 12.4 11.2 9.7 10.8 7.9 6.8

507 505 508 507 508 509

(sh) (sh) (sh) (sh) (sh) (sh)

1010 1822 1413 935.2 220.3 206.4 111.9

Brightness Bmax [cd m2] – 5700 6600 6700 2900 2900 1600

EL emission l [nm]

CIE (x, y)

– 480, 480, 480, 480, 480, 480,

– (0.21, 0.48) (0.19, 0.44) (0.21, 0.48) (0.21, 0.47) (0.20, 0.45) (0.21, 0.46)

508 508 508 508 508 508

(sh) (sh) (sh) (sh) (sh) (sh)

3. Conclusion In summary, two series of cationic iridium(III) complexes have been successfully designed and synthesized, with the tetraphenylborate counterion trifluoromethylated at different levels. Experiments indicate that trifluoromethylation can improve the thermal, photoluminescent, electrochemical, and electrophosphorescent properties. Hydrogen bonds have been first found between the aromatic hydrogen atoms of the pzpy ligands in cations and the fluorine atoms of the trifluoromethyl groups in the anions. The strong impact of the counterions on the charge transport in the devices was first investigated. Complex 3 with [BArF24] as counterion shows the best thermal and electrochemical stabilities and provides the highest PLQY of 0.39 in neat films. The best performance in a single-layer OLED was obtained with a maximum efficiency of 12.4 cd A1, which is the highest one reported for OLEDs with ionic complexes emitting in the blue-green region. This indicates that trifluoromethylation of tetraphenylborate counterions is a simple but effective way to improve the performance of a devices of cationic iridium complexes. It also suggests that the impact of the counterions on the charge transport is significant and only counterions with a good electrochemical stabilities and charge-transport abilities are suitable for cationic iridium(III) complexes as emitters in OLEDs. Further optimization of OLEDs based on complexes 5 and 6 will be carried out in the future. In addition, the impact of hydrogen bonds on the performance of the device, such as lifetime and stability, is also an open area of research.

Figure 12. Current density–voltage (top) and brightness–voltage characteristics (bottom) of OLEDs with different doping concentration (x wt %) of complex 3. & = 2, * = 3, ~ = 5, ! = 10, ^ = 15, and 3 = 20 wt % of complex 3.

4. Experimental Section tion is observed, reduced sharply with dopant concentration gradually increased. The J value with a dopant concentration of 20 % is 111.9 A m2, which are only 6 % of the current density at a dopant concentration of 2 %. The brightness is also decreased with the doping concentration increased, possibly due to the electron trap caused by ionic complex addition. Finally, based on complex 3, we obtained the most efficient green-blue electrophosphorescence of a single-layer OLED device when the dopant concentration is 2 wt %, with a maximum efficiency of 12.4 cd A1 and a maximum brightness of 5700 cd m2. &

&

Chem. Eur. J. 2014, 20, 1 – 11

www.chemeurj.org

Synthesis Synthesis and characterization of complex 4: 1H NMR (600 MHz, [D6]DMSO): d = 9.29 (d, J = 3.0 Hz, 1 H), 8.50 (d, J = 8.5 Hz, 1 H), 8.30 (dt, J = 15.6, 4.3 Hz, 1 H), 8.26 (d, J = 6.2 Hz, 2 H), 8.01 (ddd, J = 5.6, 5.2, 2.6 Hz, 2 H), 7.80 (d, J = 5.3 Hz, 1 H), 7.74 (d, J = 5.2 Hz, 1 H), 7.71 (d, J = 4.8 Hz, 1 H), 7.16 (d, J = 1.3 Hz, 10 H), 6.97–6.86 (m, 10 H), 6.76 (t, J = 7.2 Hz, 5 H), 5.63 (d, J = 2.3 Hz, 1 H), 5.62 (d, J = 2.3 Hz, 1 H), 5.60 (d, J = 2.3 Hz, 1 H), 5.58 ppm (d, J = 2.3 Hz, 1 H); MS (ESI): m/z: 718.37 [MBph4] + , 319.39 [MIr(dfppy)2(pzpy)] ; elemental analysis calcd (%) for C54H39BF4IrN5 : C 62.55, H 3.79, N 6.75; found: C 63.13, H 3.64, N 6.89; yield: 75 %.

8

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ÝÝ These are not the final page numbers!

Full Paper Synthesis and characterization of complex 5: 1H NMR (600 MHz, [D6]DMSO): d = 9.33 (d, J = 3.0 Hz, 1 H), 8.53 (d, J = 8.5 Hz, 1 H), 8.33 (d, J = 8.5 Hz, 1 H), 8.28 (d, J = 6.5 Hz, 2 H), 8.03 (dd, J = 11.6, 7.5 Hz, 2 H), 7.81 (t, J = 9.1 Hz, 1 H), 7.75 (d, J = 5.6 Hz, 1 H), 7.72 (d, J = 5.3 Hz, 1 H), 7.50 (dd, J = 8.6, 4.3 Hz, 2 H), 7.33 (s, 16 H), 6.92 (tt, J = 20.5, 10.1 Hz, 3 H), 5.65 (d, J = 2.1 Hz, 1 H), 5.63 (d, J = 2.2 Hz, 1 H), 5.61 (d, J = 2.1 Hz, 1 H), 5.60 ppm (d, J = 2.2 Hz, 1 H); MS (ESI): m/z: 718.39 [MBArF12] + , 591.40 [MIr(dfppy)2(pzpy)] ; elemental analysis calcd (%) for C58H35BF16IrN5 : C 53.22, H 2.70, N 5.35; found: C 52.13, H 3.37, N 5.12; yield: 67 %. Synthesis and characterization of complex 6: 1H NMR (600 MHz, [D6]DMSO): d = 9.34 (d, J = 3.1 Hz, 1 H), 8.55 (d, J = 8.5 Hz, 1 H), 8.40–8.36 (m, 1 H), 8.30 (d, J = 7.3 Hz, 2 H), 8.07 (t, J = 6.3 Hz, 2 H), 7.84 (d, J = 5.1 Hz, 1 H), 7.78 (d, J = 5.8 Hz, 1 H), 7.73 (s, 4 H), 7.63 (s, 8 H), 7.56 (t, J = 5.4 Hz, 2 H), 7.32 (t, J = 6.7 Hz, 1 H), 7.29 (t, J = 6.7 Hz, 1 H), 7.01–6.91 (m, 2 H), 5.65 (d, J = 2.3 Hz, 1 H), 5.64 (d, J = 2.3 Hz, 1 H), 5.62 (d, J = 2.3 Hz, 1 H), 5.60 ppm (d, J = 2.3 Hz, 1 H); MS (ESI): m/z: 718.39 [MBArF24] + , 863.35 [MIr(dfppy)2(pzpy)] ; elemental analysis calcd (%) for C62H31BF28IrN5 : C 47.10, H 1.98, N 4.43; found: C 48.26, H 3.08, N 4.34; yield: 61 %.

Quantum chemical calculations Calculations on the ground electronic states of complexes 1–6 were carried out with density functional theory (DFT) at the B3LYP level.[19] Double-x quality basis sets were used for C, H, N, F (631G*), and Ir (LANL2DZ). In the iridium(III) atom, an effective core potential (ECP) replaces the inner-core electrons, leaving the outercore (5s)2(5p)6 electrons and the (5d)6 valence electrons. The initial ground-state geometries were optimized based on the X-ray single-crystal structures. All the calculations were performed with the Gaussian 09 software package by using a spin-restricted formalism.

OLED fabrication and characterization Devices were grown on a 100 nm indium tin oxide (ITO)-coated glass with a sheet resistance of about 20 W &1. Substrates were degreased with solvents, cleaned by exposure to an oxygen plasma and UV-ozone ambient, and then passivated with a 60 nm layer of poly(3,4-ethylenedioxythiophene):poly(styrene sulfonate) (PEDOT:PSS). Then a light-emitting layer was deposited from 1, 2dichloroethane in a glovebox filled with nitrogen, and baked at 80 8C for 30 min to obtain an 85 nm film. In the light-emitting layer, poly(N-vinylcarbazole) (PVK) was used as host, 1,3-bis(5-(4tert-butylphenyl)-1,3,4-oxadiazol-2-yl)benzene (OXD-7) served for electron transfer, and the cationic iridium(III) complexes were doped as blue emitters. Then the substrates were transferred into a vacuum chamber. Subsequently, a Cs2CO3/Al bilayer was evaporated at a pressure of 2  104 Pa to serve as the cathode. The current–voltage–brightness characteristics of the devices were measured by using a Keithley 4200 semiconductor characterization system. Electroluminescent spectra were collected by using a Photo Research PR705 spectrophotometer. All the measurements were performed in ambient atmosphere without further encapsulations.

X-ray crystallography Single crystals of complexes 1–6 were grown from solutions. All the good solvent is acetone. The poor solvent is ethanol for complex 5, methanol for complexes 1 and 6, and deionized water for complex 3 and 4. The low-temperature single-crystal X-ray experiments were performed on a Bruker APEX charge-coupled device (CCD) diffractometer equipped with graphite monochromatized MoKa radiation. Direct phase determination yielded the positions of all non-hydrogen atoms. All non-hydrogen atoms were subjected to anisotropic refinement. All hydrogen atoms were generated theoretically and rode on their parent atoms in the final refinement. Selected crystal data of complex 1: Space group of P21/n with a = 9.5427(19), b = 13.247(3), c = 35.237(7) ; a = 90.00, b = 93.01(3), g = 90.008; V = 4448.2(15) 3 ; Z = 4, dcalcd = 1.472 g cm3 ; R1 = 0.0321; wR2 = 0.0621 for 7674 observed reflections [I  2s(I)]. Selected crystal data of complex 2: Space group of P¯1 with a = 12.880(3), b = 13.731(3), c = 17.199(3) ; a = 77.20(3), b = 85.04(3), g = 63.44(3)8; V = 2652.8(9) 3 ; Z = 2; dcalcd = 1.600 g cm3 ; R1 = 0.0532; wR2 = 0.1426 for 10 095 observed reflections [I  2s(I)].

Acknowledgements We would like to thank the National Natural Science Foundation of China (Grant Nos. 51173096, 61177023, and 21161160447) for supporting our foundation. We also thank Haoyuan Li for the help in our calculations and discussions.

Selected crystal data of complex 3: Space group of P21/n with a = 13.363(3), b = 53.158(11), c = 17.967(4) ; a = 90.00, b = 108.99(3), g = 90.008; V = 12 068(4) 3 ; Z = 4; dcalcd = 1.698 g cm3 ; R1 = 0.0696; wR2 = 0.1214 for 20 705 observed reflections [I  2s(I)]. Selected crystal data of complex 4: Space group of P¯1 with a = 12.694(3), b = 15.159(3), c = 17.350(4) ; a = 79.17(3), b = 69.47(3), g = 72.48(3)8; V = 2968.8(10) 3 ; Z = 2; dcalcd = 1.420 g cm3 ; R1 = 0.0843; wR2 = 0.1965 for 10 956 observed reflections [I  2s(I).

Keywords: complex chemistry · hydrogen bonds · iridium · organic light-emitting diodes · tetraphenylborates [1] C. W. Tang, S. A. Vanslyke, Appl. Phys. 1987, 51, 913. [2] R. D. Costa, E. Orti, H. J. Bolink, F. Monti, G. Accorsi, N. Armorali, Angew. Chem. 2012, 124, 8300; Angew. Chem. Int. Ed. 2012, 51, 8178. [3] T. Hu, L. He, L. Duan, Y. Qiu, J. Mater. Chem. 2012, 22, 4206. [4] L. He, J. Qiao, L. Duan, G. Dong, D. Zhang, L. Wang, Y. Qiu, Adv. Funct. Mater. 2009, 19, 2950. [5] L. He, L. Duan, J. Qiao, G. Dong, L. Wang, Y. Qiu, Chem. Mater. 2010, 22, 3535. [6] E. A. Plummer, A. van Dijken, J. W. Hofstraat, L. De Cola, K. Brunner, Adv. Funct. Mater. 2005, 15, 281. [7] T. H. Kwon, Y. H. Oh, I. S. Shin, J. I. Hong, Adv. Funct. Mater. 2009, 19, 711. [8] W. Y. Wong, G. J. Zhou, X. M. Yu, H. S. Kwok, Z. Liu, Adv. Funct. Mater. 2007, 17, 315. [9] E. A. Plummer, J. W. Hofstraat, L. De Cola, Dalton Trans. 2003, 2080 – 2084. [10] B. Park, Y. H. Huh, H. G. Jeon, C. H. Park, K. T. Kang, J. Appl. Phys. 2010, 108, 094506.

Selected crystal data of complex 5: Space group of P21/c with a = 10.4453(10), b = 24.136(2), c = 20.4163(19) ; a = 90.00, b = 93.3710(10), g = 90.008; V = 5138.1(8) 3 ; Z = 4; dcalcd = 1.692 g cm3 ; R1 = 0.0597; wR2 = 0.0916 for 12 056 observed reflections [I  2s(I)]. Selected crystal data of complex 6: Space group of Pn with a = 14.156(3), b = 25.208(5), c = 17.797(4) ; a = 90.00, b = 109.97(3), g = 90.008; V = 5969(2) 3 ; Z = 4; dcalcd = 1.753 g cm3 ; R1 = 0.0646; wR2 = 0.1573 for 18 026 observed reflections [I  2s(I)]. CCDC-963326 (1), 963327 (2), 992098 (3), 995213 (4), 992100 (5), and 992101 (6) contain the supplementary crystallographic data for this paper. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/ data_request/cif. Chem. Eur. J. 2014, 20, 1 – 11

www.chemeurj.org

These are not the final page numbers! ÞÞ

9

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

&

&

Full Paper [11] Y. Zhou, S. Han, G. Zhou, W. Y. Wong, V. A. L. Roy, Appl. Phys. Lett. 2013, 102, 083301. [12] N. M. Shavaleev, F. Monti, R. Scopelliti, A. Baschieri, L. Sambri, N. Armaroli, M. Gratzel, M. K. Nazeeruddin, Organometallics 2013, 32, 460. [13] N. M. Shavaleev, F. Monti, R. D. Costa, R. Scopelliti, H. J. Bolink, E. Orti, G. Accorsi, N. Armaroli, E. Baranoff, M. Gratzel, M. K. Nazeeruddin, Inorg. Chem. 2012, 51, 2263. [14] N. M. Shavaleev, F. Monti, R. Scopelliti, N. Armaroli, M. Gratzel, M. K. Nazeeruddin, Organometallics 2012, 31, 6288. [15] D. Ma, L. Duan, Y. Wei, L. He, L. Wang, Y. Qiu, Chem. Commun. 2014, 50, 530. [16] a) J. D. Slinker, A. A. Gorodetsky, M. S. Lowry, J. J. Wang, S. Parker, R. Rohl, S. Bernhard, G. G. Malliaras, J. Am. Chem. Soc. 2004, 126, 2763; b) M. S. Lowry, W. R. Hudson, R. A. Pascal, Jr., S. Bernhard, J. Am. Chem. Soc. 2004, 126, 14129. [17] G. E. Schneider, H. J. Bolink, E. C. Constable, C. D. Ertl, C. E. Housecroft, A. Pertegas, J. Zampese, A. Kanitz, F. Kessler, S. B. Meier, Dalton Trans. 2014, 43, 1961. [18] W. Song, X. Chen, F. Wu, W. Tian, Y. Ma, H. Xu, Chem. J. Chinese Universities 2000, 21, 1422.

&

&

Chem. Eur. J. 2014, 20, 1 – 11

www.chemeurj.org

[19] Gaussian 09, Revision A.02, M. J. Frisch, G. W. Trucks, H. B. Schelegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian, Inc., Wallingford CT, 2009.

Received: April 27, 2014 Published online on && &&, 0000

10

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ÝÝ These are not the final page numbers!

Full Paper

FULL PAPER & OLEDs

Deep blue something: Trifluoromethylation of tetraphenylborate counterions was successfully used to improve the thermal, electrochemical and electroluminescent behaviors of cationic iridium(III) complexes. Interionic hydrogen bonds were found between the cations and anions, and the strong impact of the counterions on the charge transport in devices was investigated.

Chem. Eur. J. 2014, 20, 1 – 11

www.chemeurj.org

These are not the final page numbers! ÞÞ

D. Ma, L. Duan,* Y. Wei, Y. Qiu && – && Trifluoromethylation of Tetraphenylborate Counterions in Cationic Iridium(III) Complexes: Enhanced Electrochemical Stabilities, Charge-Transport Abilities, and Device Performance

11

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

&

&

Trifluoromethylation of tetraphenylborate counterions in cationic iridium(III) complexes: enhanced electrochemical stabilities, charge-transport abilities, and device performance.

Trifluoromethylation of tetraphenlyborate counterions was successfully used to improve the electrochemical stabilities and device performance of catio...
949KB Sizes 3 Downloads 6 Views