Cyclometalated Iridium(III) Complexes Incorporating Aromatic Phosphonate Ligands: Syntheses, Structures, and Tunable Optical Properties.

The incorporation of phosphonate ligands into the cyclometalated iridium(III) complexes can not only tune their electronic and optical properties but also provide the possibility of anchoring these molecules on the semiconductor surfaces for further applications. Herein, we report the first examples of mononuclear cyclometallated iridium(III) complexes incorporating phosphonate ligands, namely, [Ir(ppy)2(HL1)]·0.5H2O (1), [Ir(ppy)2(HL2)]·0.5H2O (2), [Ir(dfppy)2(HL1)] (3), and [Ir(dfppy)2(HL2)]·3.5H2O (4) (ppy = 2-phenylpyridine, dfppy = 2-(2,4-difluorophenyl)pyridine, H2L1 = 2-pyridylphosphonic acid, H2L2 = 2-quinolinephosphonic acid). Luminescent spectra are studied both in solution and in the solid state, and significantly red-shifted broad emission bands are observed in complexes 2 and 4. The experimental and density functional theory (DFT) time-dependent-DFT calculation results indicate that the expansion of the aromatic conjugation length in the ancillary phosphonate ligands decreases the lowest unoccupied molecular orbital energy levels of the systems, originating from the triplet state associated with the ancillary ligand such as 3MLCT, 3LC, and 3LLCT charge-transfer transitions.

Introduction

The phosphorescent iridium(III) complexes are promising d-block chromophores because of their high quantum efficiency and tunable emission wavelength.[1,2] The highly efficient emission originates from the strong spin–orbit coupling of the IrIII ion, which leads to efficient intersystem crossing of the singlet to the triplet excited state. Cyclometalated iridium(III) complexes have received particular attentions because of their relatively short triplet excited-state lifetimes, high quantum yields, flexibility in color tuning, and thermal stability[3] and have been widely explored as triplet emitters in organic light-emitting diode devices, chemosensors, photoredox catalysts, and bioimaging.[4−8] Recently, a cyclometalated iridium(III) complex was grafted to indium tin oxide and titanium dioxide (TiO2) semiconductors and served as photosensitizer (chromophore) in dye-sensitized photoelectrochemical cells,[9,10] similar to the Ru–polypyridyl complexes. Noting that the phosphonate ligands (RPO3H2), which show strong coordination capabilities toward metal ions,[11−13] were frequently introduced to the Ru complexes to anchor the chromophores strongly on the metal oxide surfaces,[14,15] cyclometalated iridium(III) complexes incorporating phosphonate ligands remain almost unexplored. The only examples, as far as we are aware, are LnIr6 clusters containing 2-pyridylphosphonate as bridging ligand, which shows magnetic and optical bifunctions.[16] It is thus necessary to synthesize cyclometalated iridium(III) complexes containing uncoordinated or partially coordinated phosphonate ligands and understand their electronic and phosphorescent properties for future device applications.

Generally, the emissions of cyclometalated iridium(III) complexes are assigned to the mixed metal to ligand charge transfer (MLCT) and the π–π* transition of ligands. Both the cyclometalating and the ancillary ligands can pose significant influences on the highest occupied molecular orbital–lowest unoccupied molecular orbital (HOMO–LUMO) gap and hence tune the electronic and optical properties of the complexes.[17,18] Noting that picolinate was frequently utilized as the ancillary ligand to adjust the HOMO–LUMO gap of cyclometalated iridium(III) complexes,[19−22] herein we use two related phosphonate ligands, namely, 2-pyridylphosphonic acid (H2L1) and 2-quinolinephosphonic acid (H2L2). Four new complexes are obtained successfully with formulae [Ir(ppy)2(HL1)]·0.5H2O (1), [Ir(ppy)2(HL2)]·0.5H2O (2), [Ir(dfppy)2(HL1)] (3), and [Ir(dfppy)2(HL2)]·3.5H2O (4) (ppy = 2-phenylpyridine, dfppy = 2-(2,4-difluorophenyl)pyridine) (Scheme ), which provide the first examples of mononuclear cyclometalated iridium(III) complexes incorporating phosphonate ligands. Their optical and electrochemical properties are studied in detail. Theoretical calculations are also performed to illustrate the experimental results.

Scheme 1

Molecular Structures of Complexes 1–4

Experimental Section

Materials and Measurements

All chemicals and solvents were of reagent grade and used as purchased. 2-Pyridylphosphonic acid (2-C5H4NPO3H2, H2L1)[23,24] and 2-quinolinephosphonic acid (2-C9H6NPO3H2, H2L2)[25] were prepared according to the literature procedure. The chloride-bridged dimeric iridium complexes [Ir(ppy)2(μ-Cl)]2 and [Ir(dfppy)2(μ-Cl)]2 were synthesized according to the Noyama route by refluxing IrCl3·3H2O with 2.2 equiv of 2-phenylpyridine or 2-(2,4-difluorophenyl)pyridine in a 3:1 mixture of 2-ethoxyethanol and water.[26]

Elemental analyses for C, N, and H were determined with a PerkinElmer 240C elemental analyzer. Infrared spectra were measured as KBr pellets on a VECTOR 22 spectrometer in the range of 400–4000 cm–1. Thermogravimetric analysis (TGA) were performed on a Mettler-Toledo TGA/differential scanning calorimetry STARe thermal analyzer in the range of 25–600 °C under a nitrogen flow at a heating rate of 10 °C/min. Electrospray ionization mass spectra (ESI-MS) were recorded on a LCQTM MSn ESI mass spectrometer. The 1H and 31P NMR spectra were recorded at room temperature with a Bruker Advance III 400 spectrometer. The UV–vis spectra were recorded on a PerkinElmer Lambda 950 UV–vis spectrometer. The fluorescence spectra were measured using a PerkinElmer LS 55 fluorescence spectrometer. Time-resolved fluorescence and quantum yield measurements were carried out on a Fluorolog TCSPC spectrofluorometer (Horiba Scientific).

Cyclic voltammetry (CV) was carried out with a CHI 760C electrochemical workstation (CH Instruments, Shanghai Chenhua Co.) with a conventional three-electrode cell at room temperature. The cell was equipped with a glassy carbon working electrode, a Pt counter electrode, and a nonaqueous electrode (Ag/AgNO3). n-Bu4NPF6, 0.1 M, was employed as a supporting electrolyte. Ferrocene was used as an external standard. All sample solutions in the cell were purged with high-purity nitrogen for at least 10 min prior to experiments and a nitrogen environment was then kept over solutions in the cell to protect the solution from oxygen.

Synthesis of [Ir(ppy)2(HL1)]·0.5H2O (1)

H2L1 (0.35 g, 2.2 mmol) was dissolved in 10 mL of CH2Cl2 containing triethylamine (0.36 g, 3.6 mmol) and then a solution of [Ir(ppy)2(μ-Cl)]2 (1.07 g, 1 mmol) in 200 mL of CH2Cl2 was added. The resulted solution was allowed for stirring at room temperature for 24 h. The yellow precipitate was isolated, washed with a small amount of water and CH2Cl2, and dried in air. Yield: 76%. Elemental analysis calcd for C27H21N3O3PIr·0.5H2O: C, 48.57; H, 3.32; N, 6.29%. Found: C, 48.87; H, 3.68; N, 6.14%. IR (KBr, cm–1): 3057 (w), 1605 (s), 1581 (s), 1560 (w), 1478 (s), 1437 (m), 1420 (s), 1306 (w), 1268 (m), 1225 (w), 1159 (s), 1086 (m), 1061 (m), 1030 (m), 920 (m), 757 (s), 729 (s), 669 (w), 582 (m), 524 (w), 455 (w), 420 (w). ESI-MS of [Ir(ppy)2(L1)]− (calcd): m/z 658 (658). 1H NMR (400 MHz, DMSO-d6): δ 9.11 (d, J = 5.3 Hz, 1H), 8.16 (t, J = 7.3 Hz, 2H), 8.03–7.84 (m, 3H), 7.84–7.71 (m, 3H), 7.64 (d, J = 5.4 Hz, 1H), 7.55 (d, J = 5.3 Hz, 1H), 7.45–7.33 (m, 2H), 7.25–7.16 (m, 1H), 6.93–6.79 (m, 2H), 6.78–6.64 (m, 2H), 6.15 (d, J = 7.3 Hz, 1H), 5.99 (d, J = 7.2 Hz, 1H). 31P NMR (162 MHz, DMSO-d6): δ 23.18.

Single crystals of 1·C4H10O2 was obtained by refluxing a mixture of [Ir(ppy)2(μ-Cl)]2 (0.5 mmol) and H2L1 (1.1 mmol) in 2-ethoxyethanol/H2O (100/10 mL) under an Ar atmosphere at 140 °C for 24 h. The mixture was left to stand for one day at room temperature and yellow crystals were collected for single crystal structural determination. Yield: 47%. Elemental analysis calcd for C31H31N3O5PIr: 49.72; H, 4.17; N, 5.61%. Found: C, 49.64; H, 4.10; N, 5.59%. IR (KBr, cm–1): 3419 (s), 3040 (w), 2970 (w), 2866 (w), 1636 (w), 1607 (s), 1580 (m), 1561 (w), 1478 (s), 1418 (m), 1306 (w), 1267 (m), 1226 (w), 1170 (s), 1062 (s), 1031 (m), 925 (m), 757 (s), 729 (s), 669 (w), 629 (w), 582 (s), 525 (w), 417 (w).

Synthesis of [Ir(ppy)2(HL2)]·0.5H2O (2)

Compound 2 was synthesized following a similar procedure to that for compound 1 except that H2L2 instead of H2L1 was used as the starting material. Yield: 60%. Elemental analysis calcd for C31H23N3O3PIr·0.5H2O: C, 51.87; H, 3.37; N, 5.85%. Found: C, 51.37; H, 3.84; N, 5.61%. IR (KBr, cm–1): 3422 (br), 310 6(w), 3041 (w), 2985 (w), 1637 (w), 1596 (m), 1567 (m), 1478 (s), 1437 (m), 1422 (m), 1315 (w), 1291 (w), 1270 (w), 1217 (w), 1154 (s), 1100 (w), 1064 (s), 944 (w), 861 (w), 758 (m), 728 (m), 609 (m), 530 (w). ESI-MS of [Ir(ppy)2(L2)]− (calcd): m/z 708 (708). 1H NMR (400 MHz, DMSO-d6): δ 8.03–7.84 (m, 3H), 7.84–7.71 (m, 3H), 7.64 (d, J = 5.4 Hz, 1H), 7.55 (d, J = 5.3 Hz, 1H), 7.45–7.33 (m, 2H), 7.25–7.16 (m, 1H), 6.93–6.79 (m, 2H), 6.78–6.64 (m, 2H), 6.15 (d, J = 7.3 Hz, 1H), 5.99 (d, J = 7.2 Hz, 1H). 31P NMR (162 MHz, DMSO-d6): δ 23.18.

Synthesis of [Ir(dfppy)2(HL1)] (3)

Compound 3 was synthesized following a similar procedure to that for compound 1·C4H10O2 except that [Ir(dfppy)2(μ-Cl)]2 instead of [Ir(ppy)2(μ-Cl)]2 was used as the starting material. Yield: 40%. Elemental analysis calcd for C27H17F4N3O3PIr: C, 44.38; H, 2.34; N, 5.75%. Found: C, 44.61; H, 3.13; N, 5.27%. IR (KBr, cm–1): 3423 (br), 3066 (w), 2972 (w), 2866 (w), 1603 (s), 1574 (m), 1478 (s), 1429 (m), 1404 (s), 1294 (m), 1267 (w), 1246 (w), 1164 (s), 1105 (m), 1080 (m), 988 (m), 927 (m), 849 (m), 832 (m), 785 (m), 759 (m), 727 (w), 586 (s), 527 (w), 464 (w). Crystalline sample of 3 can also be obtained by microwave assisted reaction at 135 °C for 30 min. Yield: 62%. ESI-MS of [Ir(dfppy)2(L1)]− (calcd): m/z 730 (730). 1H NMR (400 MHz, DMSO-d6): δ 9.12 (d, J = 5.1 Hz, 1H), 8.24 (t, J = 8.6 Hz, 2H), 8.09–7.95 (m, 3H), 7.80 (t, J = 6.7 Hz, 1H), 7.71 (d, J = 5.3 Hz, 1H), 7.59 (d, J = 5.3 Hz, 1H), 7.54–7.43 (m, 2H), 7.37–7.26 (m, 1H), 6.80 (dddd, J = 27.7, 12.1, 9.5, 2.3 Hz, 2H), 5.60 (dd, J = 8.8, 2.3 Hz, 1H), 5.41 (dd, J = 8.6, 2.3 Hz, 1H). 19F NMR (376 MHz, DMSO-d6): δ −107.54 (d, J = 10.1 Hz, 1F), −108.69 (d, J = 9.8 Hz, 1F), −109.59 (d, J = 10.3 Hz, 1F), −110.59 (d, J = 9.1 Hz, 1F). 31P NMR (162 MHz, DMSO-d6): δ 23.35.

Synthesis of [Ir(dfppy)2(HL2)]·3.5H2O (4)

Compound 4 was synthesized following a similar procedure to that for compound 2 except that [Ir(dfppy)2(μ-Cl)]2 instead of [Ir(ppy)2(μ-Cl)]2 was used as the starting material. Yield: 35%. Elemental analysis calcd for C31H19F4N3O3PIr·3.5H2O: C, 44.13; H, 3.10; N, 4.98%. Found: C, 44.44; H, 4.07; N, 5.40%. IR (KBr, cm–1): 3415 (br), 2982 (w), 2676 (w), 1602 (s), 1572 (m), 1503 (w), 1477 (m), 1452 (w), 1428 (w), 1402 (m), 1291 (w), 1267 (w), 1246 (w), 1162 (m), 1102 (m), 985 (m), 845 (m), 829 (m), 788 (w), 755 (m), 714 (w), 648 (w), 626 (w), 553 (w), 529 (w), 495 (w). Yield: 35%. ESI-MS of [Ir(dfppy)2(L2)]− (calcd): m/z 780 (780). 1H NMR (400 MHz, DMSO-d6): δ 9.14 (d, J = 5.6 Hz, 1H), 8.63 (dd, J = 8.2, 2.5 Hz, 1H), 8.26 (d, J = 8.4 Hz, 1H), 8.13 (d, J = 8.6 Hz, 1H), 8.03 (t, J = 8.3 Hz, 2H), 7.97–7.84 (m, 3H), 7.67–7.48 (m, 2H), 7.39 (t, J = 6.5 Hz, 1H), 7.32–7.19 (m, 2H), 6.82 (tdd, J = 12.4, 9.6, 2.2 Hz, 2H), 5.69 (dd, J = 8.9, 2.3 Hz, 1H), 5.27 (dd, J = 8.6, 2.3 Hz, 1H). 19F NMR (376 MHz, DMSO-d6): δ −107.35 (d, J = 10.1 Hz, 1F), −108.25 (d, J = 9.4 Hz, 1F), −109.39 (d, J = 10.3 Hz, 1F), −110.62 (d, J = 9.6 Hz, 1F). 31P NMR (162 MHz, DMSO-d6): δ 26.00.

X-Ray Crystallographic Analyses

For X-ray diffraction analyses, single crystals of dimensions 0.42 × 0.25 × 0.10 mm3 for 1·C4H10O2, 0.20 × 0.10 × 0.06 mm3 for 3 were mounted on a glass rod. The crystal data were collected on a Bruker SMART APEX DUO diffractometer (for 1·C4H10O2) and Bruker SMART APEX II diffractometer (for 3) using monochromated Mo Kα radiation (λ = 0.71 073 Å) at 293 K. The numbers of observed and unique reflections are 19 740 and 5112 (Rint = 0.0537) for 1·C4H10O2, 9687 and 4690 (Rint = 0.027) for 3. The structures were solved by direct methods and refined on F2 by full-matrix least squares using SHELXTL.[27] All nonhydrogen atoms were refined anisotropically. All hydrogen atoms were either put in calculated positions or found from the difference Fourier maps and refined isotropically. Details of the crystal data and refinements are summarized in Table . Selected bond lengths and angles are given in Table .

Table 1

Crystallographic Data for Compounds 1·C4H10O2 and 3

	1·C4H10O2	3
formula	C31H31IrN3O5P	C27H17F4IrN3O3P
Fw	748.78	730.60
crystal system	monoclinic	monoclinic
space group	P21/c	C2/c
a (Å)	12.232(2)	34.92(3)
b (Å)	15.846(3)	10.096(8)
c (Å)	15.553(3)	16.247(13)
β (deg)	105.279(3)	104.980(14)
V (Å3), Z	2908.2(10), 4	5533(8), 8
Dc (g cm–3)	1.710	1.754
μ (mm–1)	4.691	4.943
F (000)	1480	2816
R1a,wR2b [I > 2σ(I)]	0.0315, 0.0926	0.0270, 0.0748
R1a,wR2b (all data)	0.0415, 0.0991	0.0324, 0.0781
goodness-of-fit	1.021	0.970
(Δρ)max, (Δρ)min (e Å–3)	1.223, −0.920	0.78, −0.68
CCDC number	1856485	1856486

R1 = ∑||Fo| – |Fc||/∑|Fo|.

w2 = [∑w(Fo2 – Fc2)2/∑w(Fo2)2]1/2.

Table 2

Selected Bond Lengths (Å) and Angles (deg) of Compounds 1·C4H10O2 and 3

compound 1·C4H10O2
Ir1–C16	1.987(6)	Ir1–C27	2.006(5)
Ir1–N1	2.166(4)	Ir1–N2	2.050(4)
Ir1–N3	2.043(4)	Ir1–O1	2.208(4)
C16–Ir1–N2	80.1(2)	C27–Ir1–N3	80.8(2)
N1–Ir1–O1	80.90(15)	C16–Ir1–O1	174.5(2)
N2–Ir1–N3	172.60(17)	C27–Ir1–N1	172.3(2)

Computational Details

Molecular geometries of all the complexes studied were optimized using B3LYP functional of density functional theory (DFT). The 6-31+G(d) basis set was used for nonmetal elements. LANL2DZ effective core potential was used for Ir transition metal. Frequency calculations at the same level of theory were also performed to identify all the stationary points as minima. At the same computational level, the time-dependent-DFT (TD-DFT) calculations were carried out on the basis of the optimized ground-state geometry. Typically, the excitation energies of the lowest 3 triplet and 3 singlet states were predicted. Multiwfn[28] was employed to analyze the charge transfer between fragments of the excited states. All the calculations were performed with Gaussian 09 packages.[29]

Results and Discussion

Synthesis and Crystal Structures

Compounds 1–4 were obtained in the solution following a similar procedure, confirmed by both elemental analyses and ESI-MS (Figure S1). Single-crystal structural analyses were performed only for compounds 1·C4H10O2 and 3. The other compounds appeared as powder or polycrystalline materials.

Compound 1·C4H10O2 crystallizes in the monoclinic system with a P21/c space group. As shown in Figure a,b, the structure consists of two ppy fragments as cyclometalated ligands, one HL1 as an ancillary ligand and a 2-ethoxyethanol molecule as the solvent molecule. The iridium center is six coordinated with a distorted octahedral geometry (Figure a). The ppy ligands exhibit cis-C–C and trans-N–N chelate configuration, an arrangement that is identical to that of the chloro-bridged dimer complex [Ir(ppy)2(μ-Cl)]2 and many [Ir(C∧N)2(LX)] complexes. 2-Pyridylphosphonate serves as a chelating ligand by using its pyridyl nitrogen (N1) and one phosphonate oxygen (O1) atoms. The Ir–C bond lengths [1.987(6), 2.006(5) Å] are close to the reported values for the dimeric and mononuclear complexes containing the Ir(ppy)2 fragments. The Ir–N bond length for N from 2-pyridylphosphonate is longer than those for N or C from the cyclometalating ligands because of the trans effect of the C donor in the C∧N ligands. The C16–Ir1–N2, C27–Ir1–N3, and N1–Ir1–O1 angles all deviate from 90°, which support a distorted octahedral coordination geometry around the IrIII ion. Moderately strong hydrogen bonds are found between the molecules through the phosphonate oxygen atoms [O2···O3A: 2.546(5) Å], forming a supramolecular dimer (Figure b). The solvent molecules help stabilize the crystal lattice via hydrogen bonding interactions [O4···O2: 2.771(11) Å]. Intermolecular interactions are observed between adjacent dimers (Figure S4a): (i) π–π overlap for parallel 2-phenylpyridine rings with the centroid–centroid distance between the two pyridine rings of 3.608 Å, (ii) C–H···π interaction between carbon atom at 4-position of pyridine ring from cyclometalating ligand and pyridine ring from phosphonate ligand. Thus, a supramolecular chain structure is constructed.

Figure 1

(a) Building unit of 1·C4H10O2. (b) Hydrogen-bond interactions between two adjacent building units in compounds 1·C4H10O2. (c) Building unit of 3. (d) Hydrogen-bond interactions between two adjacent building units in 3.

Compound 3 crystallizes in monoclinic system, space group C2/c. The molecular structure is quite similar to that of 1·C4H10O2 except that the dfppy ligands replace ppy in the latter and no lattice solvent molecule is present. The Ir–C, Ir–N, and Ir–O bond lengths become 1.975(6)–1.985(5), 2.039(4)–2.153(4), and 2.169(4) Å, respectively (Figure c), all slightly shorter than those in compound 1·C4H10O2. Intermolecular hydrogen bonds are again found through the phosphonate oxygen atoms [O2···O3A: 2.506(6) Å] forming a supramolecular dimer (Figure d). Weak intermolecular slipped π–π stacking exists between neighboring dimers via 2,4-difluorophenyl rings with the distance of 4.504 Å (Figure S4b).

Photophysical Properties

Photophysical data are given in Table . The UV–vis absorption spectra of cyclometallating ligands, ancillary ligand, and complexes 1–4 in CH3CN are shown in Figures S7 and S8. The complexes show strong absorption bands in the range of 230–300 nm, which are assigned to the ligand-centered (LC) π–π* electronic transitions of the cyclometalated ligand and the ancillary phosphonate ligand. The bands are located in the same region of absorption spectra for the free ligands of ppy, dfppy, and H2L.[1,2] The complexes also show absorption bands with lower extinction in the range 300–500 nm, which are ascribed to the singlet and triplet metal-to-ligand charge transfer (1MLCT and 3MLCT) states and ligand-centered 3LC transitions, which are not seen in the absorption spectra of free ligands.

Table 3

Absorption and Emission Data of Compounds 1–4

		emission/excitation
	absorptiona	solution (λex = 365 nm)a			solid state (λex = 365 nm)
complex	λabs/nm	λem/nm	Φ	τ/μs	λem/nm	Φ	τ/μs (average)
1	261, 395, 432	506	0.46	0.56	522, 537	0.16	0.39, 0.74
2	239, 262, 395, 445	594	0.09	0.66	569	0.03	0.63
3	255, 380	476	0.58	0.63	495, 522	0.04	0.20, 0.48
4	238, 320, 379	568	0.02	0.97	557	0.02	0.78

Measured in CH3CN solution at a concentration of 5 × 10–5 mol/L.

The luminescence spectra of compounds 1–4 were measured in CH3CN solution (5 × 10–5 M, prepurged with N2 for 10 min) and solid state (in air) at room temperature. In solution, compound 1 shows yellow light emission peaking at 506 nm and a shoulder at ca. 533 nm. The profile shows hypsochromic shift of less than 5 nm compared with the analogous heteroleptic Ir complex with 2-pyridylcarboxylate (pic) ancillary ligand, Ir(ppy)2(pic).[30,31] Significant red-shift is observed for compound 2, which shows a broad emission band centered at 594 nm and a small peak at 503 nm with the intensity reduced by about 10 times compared with that of 1. Similar phenomenon is observed for 3 and 4 containing fluorine-substituted ppy ligands. The emission peaks are at 476 and 496 nm for complex 3 and at 474 (very weak) and 568 nm for complex 4. The electron-withdrawing effect of the fluorine atom is also evident, as shown by blue-shifts of the emission spectra of 3–4 compared to those of 1–2.

For cyclometalated iridium(III) complexes, it is well-recognized that the highest occupied molecular orbital (HOMO) primarily localizes on the phenyl part of cyclometalated ligands and the iridium(III) ion and the LUMO mainly distributes on pyridine part of the cyclometalated ligand and the ancillary ligands.[17] Thus, the remarkable red-shift observed for complexes 2 and 4 can be ascribed to the expansion of the aromatic conjugation length of the phosphonate ligand H2L2, which decreases the LUMO energy level. Moreover, the emission peaks for 2 and 4 are significantly broadened, indicating that the emission in these complexes could originate from the triplet state associated with the ancillary ligand such as 3MLCT, 3LC, and 3LLCT (ligand-to-ligand charge transfer) transitions.[32] This is also supported by the TD-DFT calculations as discussed below.

Notably, the emission quantum yields of 1 (Φ = 46%) and 3 (Φ = 58%) are much higher than those of 2 (Φ = 9%) and 4 (Φ = 2%), but lower than that of Ir(dfppy)2(pic) (Φ = 95%).[33] Decay measurements in acetonitrile solution reveal that the excited state lifetimes of compounds 1–4 fall in the range of 0.56–0.97 μs (Figure S9), indicating the triplet nature of the excited state.[30]

Emission spectra of 1–4 were also studied in the solid state, as shown in Figure b and Table . The peak wavelengths for 1 (498, 524 nm) and 3 (496, 522 nm) are quite similar. While for complexes 2 and 4, which contain quinolinephosphonate ligand, broad red-shifted emission peaks are observed. These peaks are even broader in the solid state than their counterparts in acetonitrile solution, which could be due to the excimers arising from the intermolecular π–π interactions.[27] Compared with those in the solution; however, the quantum yields of complexes 1 and 2 in the solid state are significantly reduced and the lifetimes are elongated (Table ). The weak intermolecular interactions could pose nontrivial influences on stabilizing the excited state, which results in long-lived phosphorescence in these complexes.[34]

Figure 2

(a) Emission spectra of 1–4 in CH3CN solution (5 × 10–5 M). (b) Normalized emission spectra of 1–4 in the solid state at room temperature (λex = 365 nm).

Electrochemical Properties

To obtain the HOMO–LUMO gaps, the electrochemical properties of compounds 1–4 in CH3CN solutions were studied by cyclic voltammetry (CV) using ferrocene as the internal standard in the potential ranging from 0 to 1.8 V. As shown in Table and Figure , compounds 1–4 show reversible oxidation waves in the region of 0.6–1.7 V, which can be attributed to the metal-centered Ir(III)/Ir(IV) oxidation, in accordance with the reported cyclometallated Ir(III) systems. The oxidation potentials (Eox vs Fc/Fc+) were 1.083, 1.095, 1.413, and 1.432 V for compounds 1–4, respectively. Their HOMO energy levels estimated from the oxidation potentials follow the sequence 1 (−5.43 eV) > 2 (−5.44 eV) > 3 (−5.76 eV) > 4 (−5.78 eV). The results indicate that the fluoride substitution on the cyclometalating ligands and the expansion of the aromatic conjugation length of the ancillary phosphonate ligand decrease the HOMO energy level. Unfortunately, we were not able to observe the reduction peaks for these compounds within the potential window (−2.0 to 0 V); hence, the LUMO energy levels cannot be estimated through the CV measurements.

Table 4

Electrochemical Data of Compounds 1–4

complex	Eox (pa)/Va	Eox (pc)/Va	EHOMO/eVb	Ebandgap/eVc	ELUMO/eVd
1	1.083	1.017	–5.43	2.52	–2.91
2	1.095	0.988	–5.44	2.50	–2.94
3	1.413	1.363	–5.76	2.67	–3.09
4	1.432	1.307	–5.78	2.64	–3.14

Oxidation potentials measured as CH3CN solution at 100 mV s–1 with 0.1 M n-Bu4NPF6 as supporting electrolyte calibrated with ferrocene.

The HOMO energy levels were calculated using the equation EHOMO (eV) = −(Eox – E + 4.8), where E is 0.451 V.

Ebandgap energies were determined from the absorption edge of the complexes.

The LUMO energy levels were calculated using the equation ELUMO (eV) = EHOMO + Ebandgap.

Figure 3

Cyclic voltammograms of compounds 1–4 measured in CH3CN solution containing n-Bu4NPF6 (0.1 M). The scan rate was 100 mV/s.

The HOMO–LUMO gap can be obtained from the UV–vis absorbance spectra using the equation Eg (eV) = hc/λonset, where Eg is the gap, λonset is the lowest absorbance position, h is 6.626 × 10–34 J·s, and c is 3 × 1017 nm/s. The obtained Eg values follow the sequence 3 (2.67 eV) > 4 (2.64 eV) > 1 (2.52 eV) > 2 (2.50 eV). Obviously, the gaps of complexes 3 and 4 are larger than those of 1–2, which is in accordance with the fact that electron-withdrawing F on the phenyl group of ppy decreases the HOMO energy level and hence increasing the HOMO–LUMO gaps. Further, the gaps of complexes containing quinolone-phosphonate ligands (2, 4) are smaller than those of the corresponding ones containing pyridine groups (1, 3), confirming that the expansion of aromatic conjugation length of the ancillary ligand decreases the HOMO–LUMO gap. As already mentioned, the ancillary ligand mainly contribute to the LUMO energy levels. Based on the experimental EHOMO and Eg values, the ELUMO can be estimated, as given in Table .

Theoretical Calculations

The frontier molecular orbitals of compounds 1–4 were computed using B3LYP functional, as shown in Figure and Table S2. For all the studied systems, Figure displays similar population picture of HOMO, which was mainly composed of 40–44% d orbital of Ir metal and 53–55% π-orbital of ligands. Theoretical calculations of energy levels are qualitatively consistent with our experimental results. The order of calculated HOMO energy level was 1 > 2 > 3 > 4, in agreement with the experimental value obtained by electrochemical method. However, there are some differences in the distribution of LUMO. The LUMO orbitals are mainly concentrated on the ring metal ligand for compounds 1 and 3 but on the phosphonate ligands for the others. It is hence not surprising that the compounds with similar phosphonate ligands have quite similar LUMO energy levels with the difference of only 0.08–0.09 eV. The introduction of the electron-withdrawing substituent F group in ring metal ligand reduced the HOMO energy level. The increasing conjugation length also caused the decrease of the LUMO energy level. The modulation of frontier molecular orbitals, especially the LUMO ones, provides a feasible strategy for rational design and synthesis of metal Ir compounds based on organophosphate ligands. By modifying the ring metal ligands and organic phosphonate ligands, the HOMO and LUMO energy levels can be changed accordingly. To explain the broad red-shifted emission peaks observed in complexes 2 and 4, we also performed the TD-DFT calculations. As shown in Table S3, Figures S16, and S17, the lowest triplet state T1 have substantial amount of mixing transition character of 3MLCT, 3LC, and 3LLCT. However, compared with other compounds, only compound 2 has the charge transfer between ligand ppy and ancillary ligand H2L2.

Figure 4

Contour plots of HOMOs and LUMOs of Ir complex 1–4, the energy (eV) were calculated at DFT/B3LYP/6-31+G(d)/lanl2dz level.

Conclusions

We report mononuclear cyclometalated iridium(III) complexes incorporating aromatic phosphonate ligands, namely, [Ir(ppy)2(HL1)]·0.5H2O (1), [Ir(ppy)2(HL2)]·0.5H2O (2), [Ir(dfppy)2(HL1)] (3), and [Ir(dfppy)2(HL2)]·3.5H2O (4) (ppy = 2-phenylpyridine, dfppy = 2-(2,4-difluorophenyl)pyridine). Their luminescent and electrochemical properties have been studied. Both the experimental and theoretical results demonstrate that the expansion of the aromatic conjugation length of the ancillary phosphonate ligand can significantly decrease the LUMO energy level and hence the HOMO–LUMO gap. This work provides not only a new route to manipulate the electronic and optical properties of cyclometalated iridium(III) complexes by introducing phosphonate ligands but also possibilities to anchor cyclometalated iridium(III) complexes on metal oxide surfaces through phosphonate ligands for device applications.