Highly Luminescent Cyclometalated Iridium Complexes Generated by

benes (NHCs),15–17 which have been extremely popular as support- ing ligands in homogeneous catalysis,17,18 have emerged as a com- mon substructure ...
0 downloads 5 Views 452KB Size
Subscriber access provided by Kaohsiung Medical University

Article

Highly Luminescent Cyclometalated Iridium Complexes Generated by Nucleophilic Addition to Coordinated Isocyanides Hanah Na, and Thomas S Teets J. Am. Chem. Soc., Just Accepted Manuscript • Publication Date (Web): 27 Apr 2018 Downloaded from http://pubs.acs.org on April 27, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

Highly Luminescent Cyclometalated Iridium Complexes Generated  by Nucleophilic Addition to Coordinated Isocyanides  Hanah Na and Thomas S. Teets* University of Houston, Department of Chemistry, 3585 Cullen Blvd., Room 112, Houston, TX 77204-5003, USA. Email: [email protected] ABSTRACT: In this work, we report a new class of blue-emitting cyclometalated iridium complexes supported by acyclic diaminocarbene (ADC) ancillary ligands. These neutral, tris-chelate complexes are not obtainable via traditional synthetic routes and instead are generated through metal-mediated nucleophilic addition to a metal-bound isocyanide, which is followed by orthometalation of the ADC under mild conditions. Importantly, four of the variants exhibit efficient phosphorescence when immobilized in PMMA matrix, achieving quantum yields of 79% for blue emitters with 2-(2,4-difluorophenyl)pyridine (F2ppy) C^N ligand and 30–37% for orange emitters with 2-phenylbenzothiazole (bt) C^N ligand. Electrochemical studies demonstrate significantly higher lying HOMO levels in the ADC complexes relative to NHC analogues, a phenomenon that results in enhanced charge-transfer character in the excited states of the ADC complexes. This study demonstrates that ADC ancillary ligands not only give rise to new structures for Ir(III)-based phosphorescent emitters, but also are promising targets for use in light-emitting devices and other thin-film optical applications.

Introduction Luminescent transition metal complexes have been widely employed in optoelectronic applications such as organic lightemitting diodes (OLEDs)1,2 and light-emitting electrochemical cells (LECs)3,4, as well as in biological media as phosphorescent labels5,6, and sensors7. Cyclometalated iridium(III) complexes are among the most well-studied classes of compounds in these applications, in large part because of their good photo- and thermal stabilities, relatively short phosphorescent lifetimes, facile color tunability, and high photoluminescence quantum yields.8 These complexes fall into two major structure classes – homoleptic complexes of the type Ir(C^N)3 (C^N = cyclometalating ligand), usually in a facial (fac) geometry, and heteroleptic complexes with the general formula [Ir(C^N)2(LL′)]n (LL′= ancillary ligand(s), and n = –1, 0, +1). Color tuning is most often achieved by changing the structure of the cyclometalating ligand, whereas the ancillary ligand(s) in heteroleptic complexes can sometimes influence the emission color9–11 but most often has profound impacts on the redox properties and excited-state dynamics.12–14 The combination of cyclometalating and ancillary ligands also determines the nature of the emissive T1 state, which most often is a combination of ligandcentered (3LC, or 3ππ*) and metal-to-ligand charge transfer (3MLCT, or 3dπ*) mixing through configuration interaction. As successful as cyclometalated iridium complexes have been in numerous photochemical applications, and as well-understood as their structure-property relationships are, a continuing challenge in OLED research is the design of complexes and devices with efficient and stable deep-blue luminescence, a requirement for the fabrication of displays with a wide color gamut. One of the advantages of iridium(III) over most other transition metals is the very large ligandfield splitting, which prevents thermal population of nonradiative metal-centered (3MC) d–d states and allows color tuning over a

wide range. However, in compounds that emit in the deep blue region, thermal population of 3MC states deactivates the emissive T1 state, which reduces the photoluminescence quantum yield, and can lead to degradation of the complex through ligand dissociation. The ideal supporting ligand for blue-emitting compounds is thus a very strong σ-donor, which would raise the energy the unoccupied dσ* orbitals and further destabilize the 3MC states. N-heterocyclic carbenes (NHCs),15–17 which have been extremely popular as supporting ligands in homogeneous catalysis,17,18 have emerged as a common substructure in blue-emitting cyclometalated iridium complexes. Homoleptic NHC-containing Ir(C^C)3 complexes have led to profound improvements in the performance of deep-blue lightemitting diodes,19 while other complexes with NHC ligands in the cyclometalating20–23 or ancillary24,25 ligands have been shown to perform well as deep-blue emitters in solution or in electroluminescent devices. To design next-generation deep-blue emitters with even better performance, stronger σ-donating supporting ligands should be targeted. There are few commonly available ligands which surpass NHC’s as σ donors, but closely related acyclic diaminocarbenes (ADCs) have been shown to be stronger donors than aromatic unsaturated NHCs.26–30 The challenge with ADCs is that there is no commonly available route to incorporate them into cyclometalated iridium complexes. ADCs are not readily available as free ligands and are thermally unstable, making them incompatible with the relatively high temperatures and long reaction times necessitated by the substitutional inertness of iridium(III).31 Our group has recently developed a synthetic route to install ADC ligands on cyclometalated iridium complexes, which leverages the inherent electrophilicity of coordinated isocyanides, using them as reactive synthons to perform “on-complex” reactions to install new ancillary ligand structures. A series of bis-cyclometalated iridium complexes

ACS Paragon Plus Environment

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 9

Scheme 1. Synthesis of Ir complexes containing cyclometalated ADC ancillary ligands

with electron-deficient aryl isocyanide ancillary ligands react with hydrazine to form Chugaev-type chelating bis(carbene) complexes, a conceptual advance in the synthesis of iridium cyclometalates that gave the first examples of ADC-containing luminescent Ir(III) complexes.32 Similar reactions using hydrazine or amine nucleophiles and coordinated isocyanide electrophiles have been used to form acyclic diaminocarbenes (ADCs)30,33–40, primarily at square planar metal centers and including some examples of luminescent compounds.38,41 Our work was the first to extend this strategy to cyclometalated iridium. However, the complexes prepared by this route exhibit disappointing luminescence properties, in particular quantum yields that are abnormally low (Φ < 0.1 in each case). In this work, we report a new class of cyclometalated Ir complexes supported by ADC ancillary ligands, generated by a simple reaction between mono-aryl isocyanide Ir(III) complexes and amine nucleophiles. Nucleophilic addition to the isocyanide is accompanied by cyclometalation of the ADC’s aryl substituent, generating neutral, tris-chelate complexes. In contrast to previous cyclometalated iridium ADC complexes prepared by our group, some of the complexes are exceptionally efficient emitters. The emission is color-tunable depending on the cyclometalating ligand, with quantum yields as high as 0.79 for blue-emitting versions, demonstrating that this strategy is especially promising for complexes which emit in the blue region. This work presents an advance in the synthetic chemistry of cyclometalated iridium(III) that will benefit future designs of highperforming, deep-blue-emitting materials. Results Synthesis. As depicted in Scheme 1, the syntheses begin with precursors of the type Ir(C^N)2(CNAr)(Cl)(1a–2b), which are readily prepared by reaction of the chloro-bridged Ir dimer [Ir(C^N)2(μCl)]2 with two equivalents of aryl isocyanide in dichloromethane. Two different cyclometalating ligands, 2-(2,4-difluorophenyl)pyridine (F2ppy) and 2-phenylbenzothiazole (bt), and two different isocyanide ancillary ligands, 4-trifluoromethylphenylisocyanide (CNArCF3) and 4-nitrophenylisocyanide (CNArNO2), were combined to form the four precursor compounds. The complex 1b42 and other structurally similar complexes43 have been reported previously. The electrophilic character of the isocyanide carbon increases upon coordination to the metal center, allowing precursors 1a–2b to react with amines at room temperature to produce the Ir(III)

complexes containing an ADC moiety. Following nucleophilic addition of the isocyanide, subsequent base-assisted intramolecular activation of an aromatic C–H bond occurs under the same conditions, yielding the final orthometalated Ir ADC complexes (3a–4d). The Ir–C bond formation which results in cyclometalation is likely to occur through an electrophilic bond activation mechanism rather than an oxidative addition, as the iridium metal center is coordinatively saturated in the unobserved intermediate that forms prior to cyclometalation. The scope of the reactivity in Scheme 1 was examined with respect to the cyclometalating ligand, the aryl isocyanide, and the amine nucleophile. The nitro-substituted aryl isocyanide in complex 1b (C^N = F2ppy, X = NO2) reacts readily with the unhindered secondary amines NHEt2 and NHPr2, generating products containing two alkyl N-substituents on the ADC (3c and 3d), but exhibits no reactivity toward the more sterically congested amine NHiPr2. However, complex 2b, which still contains the NO2-substituted isocyanide but with C^N = bt, only reacts with primary amines. In fact, attempted addition of NHEt2 and NHPr2 to complex 2b furnished products with only one N-alkyl substituent (4c and 4d), which are formed via preferential reaction with minor primary amine impurities (NH2Et or NH2Pr). Reaction of 2b with re-distilled NHEt2 and NHPr2 gave no discernible yield of ADC product, further confirming the notion that complex 2b is completely unreactive towards secondary amines, likely due to the greater steric hindrance of bt as compared to F2ppy. Both F2ppy and bt complexes containing the weaker electron-withdrawing CF3 group on the isocyanide react only with primary amines, showing no reactivity with secondary amines. Complexes 1a and 2a were thus combined with NH2Pr and 4-methoxybenzylamine to produce four CF3-substituted ADC complexes 3a/3b and 4a/4b. Finally, analogues of 1a and 1b with the electron-rich aryl isocyanide 4-methoxyphenylisocyanide do not even react with primary amines, suggesting that electron withdrawing groups are required to activate the carbon atom of the aryl isocyanide towards nucleophilic attack. A previous study on the kinetics of nucleophilic addition to coordinate isocyanides44 in square-planar palladium(II) complexes parallels the trend reported here, with electron withdrawing groups greatly increasing the electrophilic character of the isocyanide. Taken together, all of these observations suggest that the reaction is controlled electronically by the substituent on the isocyanide and sterically by the amine and the cyclometalating C^N ligand.

ACS Paragon Plus Environment

Page 3 of 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

Figure 1. X-ray crystal structures of 3a, 3c, 3d and 4c. Ellipsoids are shown at the 50 % probability level with carbon-hydrogen atoms and solvent molecules omitted.

Several NMR features evince the ADC structure that is present in 3a–4d. The 1H NMR spectra of all complexes show diagnostic downfield (9.81–10.26 ppm) chemical shifts, corresponding to the N–H protons proximal to the aryl group of the ADC (C–NH–Ar), while the additional C–NH–R resonances in complexes prepared with primary amines (R1 = H) lie in a further upfield region of 6.05– 6.90 ppm. In the 13C{1H} NMR spectra of 3a–4d, signals corresponding to the Ccarbene appear substantially downfield from those of NHCs, around 200 ppm, providing evidence for superior donor abilities of ADC ligands.27,45 Crystal structures. The molecular structures of 3a, 3c, 3d, and 4c were determined by single-crystal X-ray diffraction and are shown in Figure 1. Selected bond length (Å) and angles (°) are summarized in Table 1 and detailed crystallographic data are reported in the Supporting Information. The iridium metal center resides in the center of a distorted octahedral coordination geometry with two C^N cyclometalated ligands and one ADC ancillary ligand. In all cases, nitrogen atoms of the C^N ligands are in a trans position relative to each other. Table 1. Selected bond lengths and angles from X-ray crystal structures

3a 3c 3d 4c

Bond length /Å Ir-CAryl Ir-Ccarbene

Bond angle/ ° N-Ccarbene-N C-Ir-C

2.039(6) 2.126(4) 2.096(3) 2.041(5)

116.4(5) 115.0(3) 113.9(3) 116.5(4)

2.099(5) 2.092(3) 2.092(3) 2.084(4)

substituent of the metalated aryl (NO2 or CF3). The C−Ir−C bond angles are all larger than 78°, and these are on the high end of reported angles of C^C cyclometalated NHC Ir complexes (76– 79°)25,46–48. Electrochemistry. The electrochemical properties of the complexes 3a–4d were investigated by cyclic voltammetry (CV) experiments (Figures S1–S8), and the data are summarized in Table 2. In complexes 3a and 3b the first oxidation waves are quasi-reversible, with peak separations (ΔEp) of 80–100 mV, and current ratios (ip,a/ip,c) ~ 0.7. The remaining complexes display first oxidation waves that are clearly irreversible. These first oxidation potentials are assigned to metal-centered IrIV/IrIII redox couple. It is notable that the oxidation potentials of all complexes described here (< 0.6 V vs. Fc+/Fc) are cathodically shifted by at least 180 mV when compared to structurally similar Ir complexes with CF3-substituted cyclometalated NHC ancillary ligands.25 This comparison demonstrates significantly higher lying HOMO levels in the ADC complexes, where stronger N→Ccarbene π-donation results in a more electron-rich carbene donor. Complexes with C^N = F2ppy (3a–3d) are more difficult to oxidize by ~180 mV when compared to compounds where C^N = bt, indicating that the HOMO is stabilized by the electronwithdrawing fluorine atoms of F2ppy. Table 2. Summary of electrochemical dataa

Eox

79.4(2) 79.15(13) 78.74(12) 78.59(19)

The Ir−Ccarbene bond distances (2.039(6)–2.126(4) Å) of all complexes are similar to those of previously reported Ir bis(ADC) complexes (2.039(4)-2.079(4) Å)32 and Ir-NHC cyclometalated complexes (2.060(5)–2.099(2) Å).25,46–48As expected, the N−Ccarbene−N bond angles (113.9(3)–116.5(4)°) are larger than those observed for NHCs (101–106°), where the carbene is located in a rigid five-membered ring.49–51 Comparison of primary-aminederived 3a/4c with secondary-amine analogues 3c/3d reveals a small influence of the N-substituents of the ADC on N−Ccarbene−N angles. The angles of 3a and 4c (116.4(5) and 116.5(4)°), which have only one N-alkyl group on the ADC are slightly larger than those of complexes with two alkyl groups (3c and 3d, 113.9(3)– 115.0(3)°). Ir−CAryl bond lengths for the orthometalated ADC span a narrow range of 2.084(4)–2.099(5) Å, showing minimal dependence on the N-substituents of ADC ligand or on the

3a 3b 3c 3d 4a 4b 4c 4d

Ered b

0.52 0.54b 0.59 0.59 0.41 0.42 0.47 0.46

–2.62 –2.62 –1.87c, –2.12c, –2.56 –1.88c, –2.14c, –2.57 –2.41, –2.62b –2.42, –2.70 –1.88c, –2.14c, –2.67 –1.89c, –2.15c, –2.68

a

Measured in 0.1 M TBAPF6 acetonitrile solution with a scan rate of 0.1 V/s. All potentials are reported relative to Fc+/Fc, and irreversible features are quoted as Ep,a or Ep,c values. bQuasi-reversible. c NO2-centered reduction. Changing the ADC substituents from CF3 to NO2 induces a slight anodic shift in the oxidation potential (50–70 mV for F2ppy complexes and 40–60 mV for bt complexes), whereas changes in Nsubstituents (R1 and R2) on the ADC ligands does not have a substantial impact on the oxidation potentials. All NO2-containing complexes (3c, 3d, 4c and 4d) exhibit first reduction potentials at comparatively positive potentials of –1.88 V and –2.14 V, independent

ACS Paragon Plus Environment

Journal of the American Chemical Society

Photophysical properties. The UV-Vis absorption spectra of complexes 3a–4d were obtained in CH2Cl2 solutions at room temperature, and Figure 2 shows overlaid spectra of complexes with CF3 substituents (3a, 3b, 4a, and 4b) as representative examples. Figures S9–S12 show absorption spectra for all complexes. In all cases intense absorption bands in the UV region with high molar extinction coefficients (ε = 16−77 ×103 M−1 cm−1) are observed, which can be assigned to spin-allowed ligand centered π−π* transitions (1LC) involving C^N and ADC ligands.

60

3a 3b 4a 4b

50 40 30 20 10 0 300

400 / nm

500

600

steady-state and time-resolved emission data, and Figure 3 shows overlaid room temperature and low temperature emission spectra of 3a and 4a. Emission spectra of 3b and 4b are deposited in SI (Figures S13 and S14). The excitation spectra (Figures S15–S18) of the emissive complexes are well-matched to the absorption spectra, indicating that phosphorescence originates from the corresponding Ir complexes and ruling out emission that originates from a minor impurity. At room temperature, complexes with F2ppy C^N ligands (3a and 3b) exhibit broad, blue phosphorescence with λmax = 498 and 495 nm and moderate solution quantum yields (Φ) of 0.22 and 0.18, respectively. Upon cooling of the solution to 77 K, the emission spectra show well-defined vibronic structure with rigidochromic blue shift of ~ 42 nm (ca. 1800 cm−1). Replacing the cyclometalating F2ppy ligand with bt (4a and 4b) induces a red shift of the emission maxima to λmax = 588 and 585 nm, resulting in yellow-orange emission with significantly lower solution quantum yields, Φ = 0.0049 (4a) and 0.0047 (4b). Similarly, 4a and 4b display better resolved phosphorescence spectra with rigidochromic shifts of 24 nm (ca. 820 cm−1) at 77 K. Energy / cm1 24000 22000 20000 18000 16000 14000

Normalized Emission Intensity

of the cyclometalating C^N ligands and thus likely attributed to a nitrophenyl-centered reduction.52 Every complex shows reduction waves from –2.4 V to –2.7 V, which are cathodically shifted when compared to those of Ir NHC complexes, indicating destabilized LUMOs due to the strong donor character of the ADC moiety. Alteration of the N-substituents on ancillary ADC ligands likewise has negligible influence on the reduction potentials.

x 103 / M1cm1

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 9

3a 298 K 3a PMMA 3a 77 K

4a 298 K 4a PMMA 4a 77 K

Figure 2. Overlaid UV-vis absorption spectra of 3a, 3b, 4a and 4b. 400

500

600

700

F2ppy complexes with NO2 substituents (3c and 3d) show intense absorption bands centered at 350 nm with ε ~ 3×104 M−1 cm−1, which are absent in the other complexes and attributed to the chromophoric p-nitrophenyl (Figure S10). All complexes exhibit less intense lowest-energy absorption bands tailing to 450 nm for F2ppy (3a–3d) and 500 nm for bt (4a–4d) complexes; these bands are largely enveloped by the intense nitrophenyl-centered transition in F2ppy complexes 3c and 3d. These weak and lowest-energy absorption bands can be assigned as both singlet and triplet metal-to-ligand charge transfer (1MLCT/3MLCT), as has been previously established for many other cyclometalated iridium complexes.8,53 The energies of the MLCT absorption bands are affected by altering the cyclometalating C^N ligand. The complexes with the fluorine-substituted cyclometalating C^N ligand (C^N = F2ppy, 3a and 3b) exhibit blue shifted MLCT bands relative to the complexes with bt C^N ligand (4a–4d), attributed to an increased HOMO–LUMO gap. However, N-substituent variation on the ADC ligands doesn’t cause a significant shift of the aforementioned bands and only exerts minimal influence on the extinction coefficients.

At room temperature, the excited state decays in fluid solution are biexponential, with weighted-average lifetimes (τ) in the range of 0.90−1.1 μs for F2ppy complexes 3a and 3b and smaller values of 0.58−0.60 μs for bt complexes 4a and 4b. These observed biexponential decays might arise from two different emitting excited states56 or originate from different triplet sublevels within same excited state57,58. These relatively short lifetimes in emissive complexes 3a, 3b, 4a and 4b, combined with poorly resolved vibronic structure of the room temperature emission profile and large rigidochromic shifts, suggest that the emitting triplet states have pronounced MLCT character, consistent with the electrochemical data (see above) that indicates destabilized Ir-dπ HOMOs.

All complexes, except NO2-containing complexes (3c, 3d, 4c, 4d), are luminescent in fluid solutions of deaerated CH2Cl2 at room temperature, and become more strongly luminescent at 77 K and when doped into PMMA (poly(methyl methacrylate) films. Quenching of excited states in transition metal complexes with nitro-substituted ligands is often observed in and is known to occur through excitedstate photoinduced electron transfer.54,55 Table 3 summarizes the

From the obtained quantum yields and lifetimes, the radiative (kr) and nonradiative (knr) decay constants were calculated. Both a decrease in kr values and increase in knr values are responsible for the significantly smaller quantum yields in bt complexes (4a and 4b) when compared to F2ppy complexes (3a and 3b). The smaller kr values in bt complexes 4a and 4b are consistent with the quantum mechanical expression for kr, which has a cubic dependence on the en

/ nm Figure 3. Overlaid emission spectra of 3a (top) and 4a (bottom), recorded in CH2Cl2 solution (λex = 310 nm) and as a 2 wt% PMMA film at 298 K (λex = 365 nm), and in CH2Cl2/toluene at 77 K (λex = 310 nm).

ACS Paragon Plus Environment

Page 5 of 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society Table 3. Photophysical properties UV-vs absorptiona λ / nm, (ε×10–3 / M-1 cm-1) 3a 3b 4a 4b

249(45), 351(5.6), 386(4.4) 251(59), 351(7.3), 386(5.8) 269(33), 321(26), 398(6.5), 441(5.5) 268(43), 321(34), 397(7.5), 441(5.9)

λmaxb / nm CH2Cl2, 298 K

λmaxb / nm 2wt% in PMMA, 298 K

CH2Cl2/ Tolueneb, 77 K

498

492

456, 486

495

492

553(sh), 588 552(sh), 585

Emission Quantum yield (Φem) CH2Cl2b/ PMMAc

τd/ μs CH2Cl2/ PMMA

(kr e×10–5 / s-1) CH2Cl2/ PMMA

(knr e×10–5 / s-1) CH2Cl2/ PMMA

0.22/0.79

0.90/1.5

2.4/5.3

8.7/1.4

455, 484

0.18/0.79

1.1/1.5

1.6/5.3

7.5/1.4

548(sh), 580

529, 572, 621

0.047/0.30

0.60/1.4

0.78/2.1

16/5.0

549(sh), 585

528, 570, 618

0.049/0.38

0.58/1.4

0.84/2.7

16/4.4

a

Recorded in CH2Cl2. bλex = 310 nm. cMeasured using an integrating sphere under air with λex = 365 nm. dAll are bi-exponential. Reported lifetime is a weighted average of the two time constant. ekr = Φ/τ and knr = (1 − Φ)/τ. ergy of the excited state,59 and the much larger knr values likely represent an increase in vibrational overlap between excited and ground states, as dictated by the energy gap law.8

samples and to characterize the performance and stability of these compounds in light-emitting devices.

When complexes are doped into PMMA thin films (2 wt%,), the emission maxima are similar to those in solution but the quantum yields drastically increase, ranging from Φ = 0.30 to Φ = 0.79. Approximately fourfold increases are observed for F2ppy complexes 3a and 3b, resulting in quantum yields of 0.79 for both of these complexes, and six to eightfold increases are noted for the bt complexes giving quantum yields of 0.30 (4a) and 0.38 (4b). These high quantum yields in PMMA matrix bespeak the potential of these complexes for OLEDs or other thin-film lighting devices. The lifetimes of all complexes also increase to 1.4–1.5 μs as the environment becomes more rigid. Although a slight increase in kr is observed, an even greater decrease in knr results in a much larger kr/knr ratio, suggesting that enhanced quantum yields in a rigid matrix are partially attributed to the suppression of nonradiative relaxation pathways.59 Both in CH2Cl2 solution and PMMA matrix, replacing the N-substituents of the ADC from n-propyl (3a and 4a) to 4-methoxybenzyl (3b and 4b) induces minimal changes in the emission spectral profile, quantum yield, and lifetime, indicating that these substituents have little effect on the nature of the lowest excited state.

The work described here presents an advance in the design of cyclometalated iridium complexes. The design of blue-emitting complexes remains one of the most significant challenges in the chemistry of cyclometalated iridium, and it is known that strongly σ-donating supporting ligands are required for high-performing blue emitters.8,19,22 Acyclic diaminocarbene (ADC) ligands are among the strongest σ-donating ligands in coordination chemistry,26 but until our group’s recent work32 there were no prior reports of cyclometalated iridium complexes with ADC ligands.

Finally, we have investigated the chemical stability and photostability of the ADC complexes described here. Free ADC ligands are known to decompose rapidly in ambient conditions,30 but complexes 3 and 4 exhibit indefinite stability in air. They are readily purified on the benchtop by silica gel chromatography, and samples of 3a and 4a stored on the bench for eight months showed no signs of degradation. The photostability of complex 3a was investigated in comparison with the archetypal blue emitter Ir(F2ppy)2(picolinato) (FIrpic).11 Solutions of each compound were prepared in CH2Cl2, and then irradiated with a 34-W blue LED for 2.5 h, with emission spectra recorded every 20–30 min. As the data in Figure S19 shows, both complex 3a and FIrpic show a decrease in emission intensity during irradiation, although FIrpic initially diminishes more quickly than 3a. In addition, the spectral profile of FIrpic changes dramatically, as shown in Figure S20, indicating significant decomposition. In contrast, the emission intensity of 3a diminishes slower, especially at earlier times, and the final normalized emission profile (Figure S21) overlays perfectly with the initial spectrum, indicating that any decomposition is leading to products that are not luminescent and have no effect on the emission color. This data indicates some photobleaching of 3a when continuously irradiated in solution, but further studies are needed to determine the photostability in thin-film

Discussion

The challenge associated with preparing cyclometalated iridium complexes with carbene ligands beyond NHCs stems primarily from the substitutional inertness of iridium(III). Whereas high-temperature routes involving deprotonation-metalation of imidazolium precursors or transmetalation of silver-NHC complexes are appropriate for NHC-containing cyclometalated iridium complexes, analogous routes are not amenable to other carbene structures like ADCs. Free acyclic diaminocarbenes are not thermally stable,26 and ADC complexes of silver(I) are unknown, suggesting that the routes used to make NHC-containing cyclometalated iridium complexes are not viable strategies to prepare ADC analogues. A promising strategy to circumvent these limitations and potentially install a variety of carbene ligands is to use the inherent reactivity of coordinated isocyanides, an approach that has not been previously applied to the design of cyclometalated iridium complexes. This ligand-based synthetic strategy has been applied to the synthesis of color-tunable luminescent rhenium(I) complexes,60 and our recent work showed that hydrazine reacts with cis-oriented isocyanides to furnish chelating bis(ADC) ancillary ligands in cyclometalated iridium complexes,32 but all of these previously characterized ADC-containing phosphors have phosphorescence quantum yields