Dalton Transactions View Article Online
Published on 30 October 2013. Downloaded by St. Petersburg State University on 14/01/2014 10:21:52.
PAPER
Cite this: Dalton Trans., 2014, 43, 1672
View Journal | View Issue
Strongly emissive long-lived 3IL excited state of coumarins in cyclometalated Ir(III) complexes used as triplet photosensitizers and application in triplet–triplet annihilation upconversion† Xiuyu Yi, Caishun Zhang, Song Guo, Jie Ma and Jianzhang Zhao* Three different coumarin chromophores were used to prepare the Ir(III) complexes, i.e. coumarin 6 (Ir-1), 7-diethylamino coumarin (Ir-2) and x-phenyl-y-diethylamino coumarin (Ir-3). Ir( ppy)2(bpy)[PF6] was used as the reference complex (Ir-0). The coumarin units were connected to the coordination center of the complexes by using a CuC π-conjugation bond. The photophysical properties of the complexes were studied with steady state and time-resolved absorption and luminescence spectroscopy, low-temperature luminescence (77 K), as well as DFT calculations. All the three new complexes show strong absorption of visible light (molar absorption coefficient ε is up to 42 000 M−1 cm−1 at 487 nm) and a long-lived triplet excited state (τT = 65.9 μs), compared to the reference complex Ir( ppy)2(bpy)[PF6], which shows the typical weak visible light-absorption (ε < 5000 M−1 cm−1 in the region beyond 400 nm) and a short triplet excited state (τT = 0.3 μs). Interestingly the long-lived triplet excited states are strongly phosphorescent (quantum yield is up to 18.2%, with emission maxima at 607 nm), which is rare for phosphorescent transition metal complexes. With nanosecond time-resolved transient difference absorption spectroscopy we proved that a coumarin-localized triplet excited state (3IL) was produced upon photoexcitation. The complexes were used as triplet photosensitizers for triplet–triplet annihilation upconversion and upcon-
Received 22nd August 2013, Accepted 30th October 2013
version quantum yields up to 22.8% were observed. Our results are useful for the preparation of visible
DOI: 10.1039/c3dt52306c
light-harvesting transition metal complexes, the study of the triplet excited state of organic chromophores, as well as the application of these visible light-harvesting transition metal complexes as efficient
www.rsc.org/dalton
triplet photosensitizers.
Introduction Transition metal complexes such as cyclometalated Ir(III) complexes have attracted much attention due to the applications in photocatalysis,1–3 photodynamic therapy,3–5 phosphorescence and molecular probes,6–8 and more recently in the triplet–triplet annihilation (TTA) upconversion.9–13 However, most of the known Ir(III) complexes show weak absorption of visible light (molar absorption coefficients are smaller than 10 000 M−1 cm−1 in the region >400 nm), and the lifetime of the triplet excited state is short (usually less than 5 μs).14–17 These properties do not hamper the complexes from being used in conventional areas such as electroluminescence,8 but
State Key Laboratory of Fine Chemicals, School of Chemical Engineering, Dalian University of Technology, E-208 West Campus, 2 Ling-Gong Road, Dalian 116024, P. R. China. E-mail: [email protected]; http://finechem.dlut.edu.cn/photochem † Electronic supplementary information (ESI) available: 1H NMR, 13C NMR and HRMS data of the compounds, the photophysical data and DFT calculations of the Ir(III) complexes. See DOI: 10.1039/c3dt52306c
1672 | Dalton Trans., 2014, 43, 1672–1683
they are detrimental to the new applications concerning visible light-harvesting, such as photocatalysis1 and TTA 9–13 upconversion. Therefore, it is highly desired to design cyclometalated Ir(III) complexes that show strong absorption of visible light and long-lived triplet excited states.17–20 To attach a visible light-harvesting ligand to the coordination center, either by a π-conjugation linker or a non-conjugate linker, is the effective method to prepare the aforementioned transition metal complexes.3,17,21–23 A π-conjugation linker between the organic chromophore and the Ir(III) coordination center is preferred, so that the intersystem crossing (ISC) will be improved.24–26 Concerning this aspect, Bodipy,24,27 naphthaimide,28 naphthalenediimide,29 and perylenebisimide have been attached to the Ir(III) coordination center.26 The common feature of these organic chromophorecontaining Ir(III) complexes is the strong absorption of visible light and long-lived triplet excited states.3 We have used these complexes in photocatalysis and TTA upconversion.10 Generally much improved efficiencies were observed for these complexes compared to those conventional Ir(III) complexes such
This journal is © The Royal Society of Chemistry 2014
View Article Online
Dalton Transactions
Published on 30 October 2013. Downloaded by St. Petersburg State University on 14/01/2014 10:21:52.
as Ir( ppy)3, which shows weak absorption of visible light and short-lived triplet excited states.30 However, much room is left to explore the preparation of visible light-absorbing Ir(III) complexes. For example, coumarin is a versatile organic chromophore, which has been extensively used in fluorescent
Scheme 1 Some known coumarin-containing cyclometalated Ir(III) complexes.
Paper
molecular probes or light-harvesting arrays.31–33 However, its application in the triplet excited state manifold is rare.25,34–36 Previously coumarin 6 was used for the preparation of Ir(III) complexes, by the direct cyclometalation method (coumarin 6 as the C donor of the C–Ir bond, Ir–Cou–a, Scheme 1),14a,37 Ru(II) complexes.34,35 However, coumarin derivatives were never connected to the Ir(III) coordination center via a π-conjugation linker, such as a CuC bond. Furthermore, the 3IL emission of coumarin was rarely reported.37 Previously we prepared coumarin-containing Ir(III) complexes with the imidazole linker, and normally the complexes are weakly phosphorescent (Ir–Cou–b–d, Scheme 1).25,38 In these complexes, the π-conjugation framework of the coumarin is not directly connected to the Ir(III) center. In order to overcome the aforementioned challenges, herein we prepared C^N cyclometalated Ir(III) complexes in which three different coumarin chromophores were connected to the Ir(III) coordination center with the π-conjugation linker (Schemes 2 and 3). Strong absorption of visible light and long-lived triplet excited 3IL state were observed for the complexes. Interestingly, strong red phosphorescence originating from the 3IL excited state was observed. The photophysical properties were investigated by steady state and time-resolved spectroscopy, as well as DFT calculations. The phosphorescence lifetimes of Ir-1–Ir-3 are much longer than that of the reference complex Ir-0. The complexes were used as triplet photosensitizers for TTA upconversion. Our results are useful for the preparation of visible light-harvesting cyclometalated Ir(III) complexes and for the application of these complexes in photocatalysis and TTA upconversion.
Scheme 2 Synthesis of Ir-0, Ir-1, Ir-2 and Ir-3. (a) IrCl3·3H2O, 2-ethoxyethanol–water = 3 : 1 (v/v) as solvent, 110 °C, 24 h; (b) CH2Cl2–MeOH = 2 : 1 (v/v), reflux, 5–6 h.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 1672–1683 | 1673
View Article Online
Dalton Transactions
Published on 30 October 2013. Downloaded by St. Petersburg State University on 14/01/2014 10:21:52.
Paper
Scheme 3 Synthesis of Ir-0, Ir-1, Ir-2 and Ir-3. (a) CH3COOCOCH3, HCl/CH3COOH, pyridine, 1 h, room temperature; (b) Br2, CH3COOH, room temperature; (c) K2CO3, CH3OH, 10 min, room temperature; (d) 130 °C, 6 h; (e) CH3OH, 30 °C, 4 h; (f) pyridine, reflux; (g) tributyltin chloride, n-BuLi, DEE; (h) 2,5-dibromopyridine, Pd(PPh3)4, xylene; (i) trimethylsilylacetylene, Pd(PPh3)2Cl2, PPh3, CuI, THF/NEt3; ( j) TBAF, THF, Ar, room temperature; (k) Pd(PPh3)4, CuI, THF/NEt3.
Experimental General information NMR spectra were taken on a 400 MHz Varian Unity Inova spectrophotometer. Mass spectra were recorded with a Q-TOF Micro MS spectrometer. UV-Vis spectra were taken on a HP8453 UV-Visible spectrophotometer. Fluorescence spectra were recorded on a Shimadzu RF-5301PC spectrofluorometer. Luminescence quantum yields were measured with [Ru(dmb)3] [PF6]2 (Ru-1) as standard (ΦP = 0.073 in deaerated acetonitrile). Luminescence lifetimes were measured on an OB 920 luminescence lifetime spectrometer (Edinburgh, UK).
1674 | Dalton Trans., 2014, 43, 1672–1683
Compounds 1–8 were synthesized according to the literature methods.39,40 9–12 were prepared by the reported method.41 Compound L1 5 (120 mg, 0.28 mmol) and 5-ethynyl-2,2-bipyridine (12) (50.0 mg, 0.28 mmol) were added to deaerated triethylamine (20 mL). Then Pd(PPh3)4 (8.0 mg, 5 mol%) and CuI (3 mg, 5 mol%) were added under an Ar atmosphere. The mixture was stirred and refluxed for 8 h under Ar. After completion of the reaction, the mixture was cooled to RT and the light red precipitate was collected by filtration. The crude product was
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 30 October 2013. Downloaded by St. Petersburg State University on 14/01/2014 10:21:52.
Dalton Transactions
purified with column chromatography (silica gel, ethyl acetate–petroleum ether = 1 : 2, v/v) to give an orange solid. Yield: 101 mg (68.2%). M.p.: 201.3–201.9 °C. 1H NMR (400 MHz, CDCl3), δ = 8.91 (s, 1H); 8.81 (s, 1H); 8.72 (d, 1H, J = 2.0 Hz); 8.46 (t, 2H, J = 8.7 Hz); 8.06 (d, 1H, J = 7.6 Hz); 7.96–7.85 (m, 3H); 7.79 (s, 1H); 7.51 (t, 1H, J = 7.6 Hz); 7.39 (m, 2H); 6.74 (s, 1H); 3.71 (m, 4H); 1.34 (t, 6H, J = 5.8 Hz). 13C NMR (100 MHz, CDCl3): δ = 161.0, 160.5, 155.9, 155.3, 154.8, 152.6, 151.4, 149.3, 141.0, 140.5, 139.1, 137.4, 136.6, 131.1, 129.0, 126.4, 124.2, 121.8, 121.6, 120.7, 120.6, 118.8, 115.2, 110.6, 108.4, 102.0, 92.5, 90.5, 46.1, 13.4 ppm. EI-HRMS: calcd (C32H24N4O2S): m/z = 528.1620, found, m/z = 528.1628. Compound L2 The synthetic procedure is similar to that of L1 except that compound 7 (63 mg, 0.28 mmol), instead of 5, was used. The crude product was purified with column chromatography (silica gel, ethyl acetate–petroleum ether = 1 : 2, v/v) to give an orange solid. Yield: 81 mg (73.6%). M.p.: 183.7–184.1 °C. 1H NMR (400 MHz, CDCl3), δ = 8.83 (s, 1H); 8.70 (s, 1H); 8.42 (t, 2H, J = 4.9 Hz); 7.97 (d, 1H, J = 9.0 Hz); 7.85 (m, 3H); 7.79 (m, 2H); 7.33 (d, 1H, J = 4.1 Hz); 7.27 (d, 1H, J = 7.9 Hz); 6.61 (d, 1H, J = 8.7 Hz); 6.50 (s, 1H); 3.46–3.41 (m, 4H); 1.23 (t, 6H, J = 7.2 Hz). 13C NMR (100 MHz, CDCl3): δ = 160.9, 156.6, 154.8, 151.6, 149.4, 146.3, 139.6, 137.2, 132.3, 129.3, 128.7, 128.6, 124.1, 121.6, 120.5, 109.5, 108.5, 103.9, 97.4, 90.3, 89.1, 45.1, 12.6 ppm. EI-HRMS: calcd (C25H21N3O2): m/z = 395.1634, found, m/z = 395.1644. Compound L3 The synthetic procedure is similar to that of L1 except that compound 8 (104 mg, 0.28 mmol), instead of 5, was used. The crude product was purified with column chromatography (silica gel, ethyl acetate–petroleum ether = 1 : 2, v/v) to give an orange solid. Yield: 66 mg (50.1%). M.p.: 181.2–181.7 °C. 1H NMR (400 MHz, CDCl3), δ = 8.82 (s, 1H); 8.69 (d, 1H, J = 3.6 Hz); 8.43–8.40 (m, 2H); 7.95 (d, 1H, J = 8.3 Hz); 7.83–7.73 (m, 4H); 7.60 (d, 2H, J = 8.1 Hz); 7.29 (t, 2H, J = 9.7 Hz); 6.59 (d, 1H, J = 10.2 Hz); 6.51 (s, 1H); 3.44–3.38 (m, 4H); 1.23 (t, 6H, J = 7.0 Hz). 13C NMR (100 MHz, CDCl3): δ = 161.4, 156.4, 155.6, 154.9, 151.7, 150.9, 149.4, 140.9, 139.4, 137.1, 136.4, 131.7, 129.3, 128.2, 124.0, 121.7, 121.5, 120.4, 119.8, 109.2, 109.1, 97.1, 93.8, 87.2, 45.0, 12.6 ppm. EI-HRMS: calcd (C31H25N3O2): m/z = 471.1947, found, m/z = 471.1953. General method for synthesis of the Ir(III) complexes All procedures were carried out under an Ar atmosphere. Cyclometalated Ir(III) chloro-bridged dimers were synthesized according to a literature method.38 IrCl3·3H2O and 2.5 equiv. of ligand were heated in a 3 : 1 mixture of 2-ethoxyethanol and water (v/v). The slurry was heated at 110 °C for 24 h. After cooling to rt, the crude product was filtered. The solid residue was washed with water, followed by two portions of n-hexane and ether. Then the solid was dried under vacuum.
This journal is © The Royal Society of Chemistry 2014
Paper
Complex Ir-0 A solution of [Ir( ppy)2Cl]2 (43 mg, 0.04 mmol) and 2,2′-bipyridine (16 mg, 0.1 mmol) in CH2Cl2–MeOH (12 mL, 2 : 1 v/v) was refluxed under an Ar atmosphere. After 5–6 h, the orange solution was cooled to rt, and then a 10-fold excess of ammonium hexafluorophosphate was added. The suspension was stirred for 15 min and then filtered to remove insoluble inorganic salts. The solution was evaporated to dryness under reduced pressure to obtain a crude orange solid. The solid was dissolved in CH2Cl2 and filtered to remove the residual traces of inorganic salts, then purified with column chromatography (silica gel, CH2Cl2–MeOH = 25 : 1) to give a light yellow solid. Yield: 48 mg (76.2%). M.p. >250 °C. 1H NMR (400 MHz, CDCl3, CH3OD), δ = 8.49 (d, 4H, J = 7.8 Hz), 8.18 (t, 1H, J = 9.0 Hz), 7.94–7.87 (m, 6H), 7.78 (t, 1H, J = 7.8 Hz), 7.70 (d, 1H, J = 7.3 Hz), 7.52 (d, 1H, J = 7.8 Hz), 7.42–7.36 (m, 6H), 7.07–7.00 (m, 2H), 6.94 (t, 1H, J = 7.1 Hz), 6.31 (d, 1H, J = 7.6 Hz). ESI-HRMS: [(M − PF6)]+, calc (C32H24N4Ir): m/z = 657.1630, found, m/z = 657.1625. Complex Ir-1 The synthetic procedure is the same as that for Ir-0, except that L1 (53 mg, 0.1 mmol) was used. The solution was evaporated to dryness under reduced pressure to obtain a crude orange solid. The solid was dissolved in CH2Cl2 and filtered to remove the residual inorganic salts, then purified by column chromatography (silica gel, CH2Cl2–MeOH, 25 : 1, v/v) to give a red solid. Yield: 70 mg, (75.1%). M.p. >250 °C. 1H NMR (400 MHz, DMSO-d6), δ = 9.00 (s, 1H); 8.89 (d, 2H, J = 8.4 Hz); 8.32 (t, 4H, J = 16.4 Hz); 8.09 (d, 1H, J = 7.9 Hz); 8.03–7.93 (m, 6H); 7.89 (s, 2H); 7.83 (d, 1H, J = 5.5 Hz); 7.74 (t, 1H, J = 12.8 Hz); 7.68 (d, 1H, J = 5.5 Hz); 7.55 (t, 1H, J = 15.3 Hz); 7.42 (t, 1H, J = 15.0 Hz); 7.23–7.17 (m, 2H), 7.07–7.02 (m, 2H); 6.96–6.91 (m, 2H); 6.81(s, 1H); 6.24 (t, 2H, J = 17.1 Hz); 3.50 (m, 4H); 1.04 (t, 6H, J = 13.6 Hz). 13C NMR (100 MHz, DMSO-d6): δ 166.9, 166.8, 160.4, 159.6, 155.8, 154.7, 154.0, 153.8, 152.0, 150.7, 150.2, 149.9, 149.2, 143.9, 143.8, 141.2, 140.6, 139.8, 139.1, 135.6, 131.2, 130.4, 126.6, 125.5, 124.7, 124.2, 123.8, 122.6, 122.1, 120.2, 113.9, 110.0, 105.0, 100.8, 96.1, 87.7, 45.5, 13.1 ppm. ESI-HRMS: [(M − PF6)]+, calc (C54H40N6IrO2S): m/z = 1029.2563, found, m/z = 1029.2568. Complex Ir-2 The synthetic procedure is the same as that for Ir-0, except that L2 (40 mg, 0.1 mmol) was used. The solution was evaporated to dryness under reduced pressure to obtain a crude red solid. The solid was dissolved in CH2Cl2 and filtered to remove the residual inorganic salts, then purified by column chromatography (silica gel, CH2Cl2–MeOH, 25 : 1, v/v) to give a red solid. Yield: 65.8 mg, (79.0%). M.p. >250 °C. 1H NMR (400 MHz, CDCl3/CD3OD), δ = 8.59 (d, 2H, J = 4.4 Hz); 8.27 (d, 1H, J = 7.6 Hz); 8.17 (t, 1H, J = 15.8 Hz); 8.07 (d, 2H, J = 6.8 Hz); 7.88 (s, 2H); 7.64–7.59 (m, 4H); 7.54 (s, 3H); 7.31 (d, 2H, J = 7.9 Hz); 7.10–7.06 (m, 4H); 6.96 (d, 2H, J = 7.1 Hz); 6.71 (d, 1H, J = 9.3 Hz); 6.51(s, 1H); 6.33–6.28 (m, 2H); 3.52–3.47 (m, 4H); 1.27
Dalton Trans., 2014, 43, 1672–1683 | 1675
View Article Online
Paper
Dalton Transactions
(t, 6H, J = 13.8 Hz). 13C NMR (100 MHz, CDCl3/CD3OD): δ 167.7, 166.6, 156.7, 153.8, 152.3, 150.7, 148.3, 143.3, 139.3, 138.3, 132.4, 131.4, 130.6, 129.8, 129.2, 128.7, 124.7, 123.2, 122.8, 120.5, 120.0, 113.7, 110.0, 108.2, 96.6, 88.6, 87.3, 44.9, 12.0 ppm. ESI-HRMS: [(M − PF6)]+, calc (C47H37N5IrO2): m/z = 896.2576, found, m/z = 896.2562.
Φsam ¼ 2Φstd
Complex Ir-3 Published on 30 October 2013. Downloaded by St. Petersburg State University on 14/01/2014 10:21:52.
78.0% in CH3CH2OH) and the quantum yields were calculated with eqn (2), where Φunk, Aunk, Iunk and ηunk represents the quantum yield, absorbance, integrated photoluminescence intensity of the samples and the refractive index of the solvents.9
The synthetic procedure is the same as that for Ir-0, except that L3 (47 mg, 0.1 mmol) was used. The solution was evaporated to dryness under reduced pressure to obtain a crude red solid. The solid was dissolved in CH2Cl2 and filtered to remove the residual inorganic salts, then purified by column chromatography (silica gel, CH2Cl2–MeOH, 25 : 1, v/v) to give a red solid. Yield: 67.2 mg (75.2%). M.p. >250 °C. 1H NMR (400 MHz, CDCl3), δ = 8.65 (d, 2H, J = 8.3 Hz); 8.17 (m, 2H); 7.98–7.92 (m, 4H); 7.80–7.70 (m, 7H); 7.57 (d, 1H, J = 5.3 Hz); 7.52–7.46 (m, 3H); 7.42–7.36 (m, 2H); 7.07–7.01 (m, 4H); 6.96–6.90 (m, 2H); 6.70 (s, 1H); 6.61 (s, 1H); 6.32–6.27 (m, 2H); 3.47–3.42 (m, 4H); 1.25 (t, 6H, J = 14.2 Hz). 13C NMR (100 MHz, DMSO-d6): δ 166.7, 166.6, 160.2, 156.0, 154.7, 154.5, 150.8, 150.1, 149.8, 149.4, 149.0, 139.7, 138.9, 138.8, 137.2, 131.4, 131.1, 130.9, 130.4, 130.2, 130.0, 128.1, 125.5, 125.0, 124.9, 124.0, 123.6, 122.4, 120.2, 120.0, 119.0, 117.0, 109.3, 108.3, 96.4, 85.1, 44.1, 12.3. ESI-HRMS: [(M − PF6)]+, calc (C53H41N5IrO2): m/z = 972.2890, found, m/z = 972.2897. Singlet oxygen (1O2) quantum yields (ΦΔ) The 1O2 quantum yields (ΦΔ) of the photosensitizers were determined with 2,6-diiodo-bodipy as standard (ΦΔ = 0.83 in dichloromethane, DCM). Air saturated DCM was obtained by bubbling air for 15 min. The absorbance of the 1O2 scavenger 1,3-diphenylisobenzofuran (DPBF) was adjusted around 1.0 in air saturated DCM. Then, the photosensitizer was added to the cuvette and the photosensitizer’s absorbance was adjusted to around 0.2–0.3. The cuvette was irradiated with monochromatic light at the peak absorption wavelength for 10 seconds. Absorbance was measured several times after each irradiation. The slopes of absorbance maxima of DPBF at 414 nm versus time graph for each photosensitizer were calculated. Singlet oxygen quantum yield (ΦΔ) were calculated according to a modified equation: Φsam ¼ Φref
ksam F ref kref F sam
ð1Þ
where ‘sam’ and ‘ref’ designate the photosensitizers and ‘MB’ respectively. k is the slope of the plots of the absorbance of DPBF (414 nm) against the irradiation time, F is the absorption correction factor, which is given by F = 1–10−O.D. (O.D. at the irradiation wavelength). Triplet–triplet annihilation upconversion Diode pumped solid state laser was used for the upconversions. The samples were purged with N2 for 15 min before measurement. The upconversion quantum yields were determined with coumarin-6 as the quantum yield standard (Φ =
1676 | Dalton Trans., 2014, 43, 1672–1683
Astd Asam
I sam I std
ηsam 2 ηstd
ð2Þ
The delayed fluorescence of the upconversion was measured with nanosecond pulsed laser (Opolett™ 355 II nanosecond pulsed laser, pulse length: 7 ns; pulse repetition: 20 Hz; peak OPO energy: 4 mJ; OPOTEK, USA), which is synchronized to a FLS 920 spectrofluorometer (Edinburgh, U.K.). The pulsed laser is sufficient to sensitize the TTA upconversion. The decay kinetics of the upconverted fluorescence (delayed fluorescence) was monitored with a FLS 920 spectrofluorometer (synchronized to the OPO nanosecond pulsed laser). The prompt fluorescence lifetimes were measured with EPL picosecond pulsed laser (405 nm) which is synchronized to the FLS 920 spectrofluorometer. The photographs of the upconversion were taken with Sumsung NV 5 digital camera. DFT calculations The density functional theory (DFT) calculations were used for the optimization of the singlet states and triplet states. The spin density surfaces of the complexes were calculated based on the optimized triplet state geometries. All the calculations were performed with Gaussian 09W.42
Results and discussion Design and synthesis of the complexes Three different coumarin chromophores were selected, i.e. the coumarin 6, coumarin and the phenylcoumarin. All the chromophores were connected to the bpy ligand via the π-conjugation linker, so that the ISC can be enhanced. Coumarin 6 and diethylamino coumarin are known to be highly fluorescent. The subtle difference between the two chromophores may cause different effects on the photophysical properties of the complexes. In Ir-3, the diethylamino coumarin moiety was isolated from the Ir(III) coordination center by a phenyl unit. In order to prepare coumarin 6, 5-bromo-4-diethylsalicylaldehyde was prepared. Then the bromo coumarin 6 was prepared (Scheme 3). Sonogashira coupling between compound 5 and 5-ethynyl-2,2′-bipyridine gives ligand L-1. Ligands L-2 and L-3 were prepared by similar methods. All the compounds were obtained in satisfactory yields (Scheme 2). Complex Ir-0 was used as reference complex. UV-Vis absorption and emission spectra The UV-Vis absorption spectra of the complexes were studied (Fig. 1). The reference complex Ir-0 gives weak absorption in the visible region, with ε < 5000 M−1 cm−1 in the region beyond 400 nm. In contrast, all the coumarin-containing complexes Ir-1, Ir-2 and Ir-3 give strong absorption in the visible
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 30 October 2013. Downloaded by St. Petersburg State University on 14/01/2014 10:21:52.
Dalton Transactions
Paper
Fig. 3 Photoluminescence spectra of the complexes at RT and 77 K. (a) Ir-1 (λex = 470 nm), (b) Ir-2 (λex = 470 nm). In C2H5OH–CH3OH (4 : 1, v/v), 1.0 × 10−5 M. Fig. 1 UV-Vis absorption of Ir(III) complexes. c = 1.0 × 10−5 M in CH2Cl2, 20 °C.
region. For example, Ir-1 shows a broad absorption band in the region of 350 nm–450 nm, with maximum at 439 nm (22 440 M−1 cm−1). Ir-2 gives stronger absorption at 487 nm (42 000 M−1 cm−1). In comparison Ir-3 gives slightly blueshifted absorption at 437 nm (45 800 M−1 cm−1). The luminescence spectra of the complexes were studied (Fig. 2). Ir-1 gives emission at 607 nm. Similar emission bands were observed for Ir-2 and Ir-3. Furthermore, we found that the emission of the complexes Ir-1–Ir-3 can be dramatically quenched by O2. Therefore, the luminescence can be attributed to phosphorescence. Moreover, the emissions of Ir-1 and Ir-2 are almost completely quenched by O2, indicating the longer phosphorescence lifetime of Ir-1 and Ir-2 than that of Ir-3.
measured and were compared with those at room temperature (Fig. 3).24,25,27,43 For Ir-1, the emission spectra became more structured at 77 K compared to that at RT. The thermally induced Stokes shift is very small. Therefore, the emission of Ir-1 can be attributed to the intraligand triplet excited state (3IL state) localized on the coumarin unit.25 For Ir-2, the structured emission spectrum at 77 K was observed, with a small thermally induced Stokes shift of 481 cm−1. The emissive triplet excited state of Ir-2 can also be assigned as an 3IL state. Similar results were observed for Ir-3 (see ESI† for detail). For the reference complex Ir-0, however, the emission band at 77 K is structureless and the thermally induced Stokes shift is larger. DFT calculations on the T1 state energy level support the assignment of the triplet excited states as the intraligand triplet excited state.44,45 Nanosecond time-resolved transient absorption spectra
77 K emission spectra In order to study the emissive states of the complexes, the phosphorescence spectra of the complexes at 77 K were
In order to study the triplet excited state of the complexes, the nanosecond time-resolved transient difference absorption spectra of the complexes were investigated (Fig. 4). For Ir-1,
Fig. 2 The emission spectra of the complexes under different atmosphere. (a) Ir-1 (λex = 470 nm), (b) Ir-2 (λex = 475 nm), (c) Ir-3 (λex = 438 nm), (d) Ir-0 (λex = 403 nm). c = 1.0 × 10−5 M in CH2Cl2, 20 °C.
Fig. 4 Nanosecond time-resolved transient difference absorption spectra of (a) Ir-1, (c) Ir-2 after pulsed excitation (λex = 532 nm). Decay trace of (b) Ir-1 at 470 nm and (d) Ir-2 at 490 nm. 1.0 × 10−5 M in deaerated CH2Cl2. 20 °C.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 1672–1683 | 1677
View Article Online
Paper
Dalton Transactions
Table 1
Photophysical parameters of Ir-0, Ir-1, Ir-2 and Ir-3
Published on 30 October 2013. Downloaded by St. Petersburg State University on 14/01/2014 10:21:52.
τL e/μs
Ir-0 Ir-1 Ir-2 Ir-3
λabs a
εb
λem c
ΦL d/%
77 K
298 K
τT f (μs)
ΦΔ g
405 473 487 437
0.50 2.01 4.20 4.58
578 607 645 640
4.3 18.2 6.5 1.1
4.8 9962.8 137.6 1241.0
0.3 102.2 36.7 7.37
0.3 65.9 34.8 2.7
— 0.803 0.721 0.585
In CH2Cl2 (1.0 × 10−5 M). b Molar extinction coefficient at the absorption maxima. ε: 104/cm−1 mol−1 dm3. c In CH2Cl2. d In CH2Cl2, with [Ru(dmb)3][PF6]2 (Ru-1) (Φ = 0.073 in deaerated acetonitrile) as the standards. e Luminescence lifetime, at 298 K (in CH2Cl2) and 77 K (in C2H5OH–CH3OH = 4 : 1). f Triplet state lifetime, measured by timeresolved transient absorption. 1.0 × 10−5 M in deaerated CH2Cl2. g Singlet oxygen quantum yield. With 2,6-diiodo-Bodipy as a standard (ΦΔ = 0.83 in DCM). Photoexcitation wavelengths of 494 nm, 509 nm, 490 nm were used. a
upon visible light excitation, a bleaching band at 444 nm was observed, which is due to the depletion of the ground state. Positive absorption in the region of 550–750 nm was observed. The lifetime of the transient was determined as 65.9 μs, which was substantially quenched in an aerated solution. Therefore, the observed long-lived transient species is due to the triplet excited state of the complex, which is localized on coumarin 6. These results indicated a subtle variation of the molecular structure of the moiety.24,25 A similar transient was observed for Ir-2, but the lifetime was reduced to 34.6 μs. For Ir-3, however, the triplet excited state lifetime is much shorter, 2.7 μs (Table 1). Since the main application of the triplet photosensitizers is the triplet–triplet energy transfer (TTET), the efficiency of the TTET processes will be affected by the lifetime of the triplet excited state of the complexes.10,12,13,46,47 The photophysical properties of the compounds are listed in Table 1. Notably, the complexes have long-lived triplet excited states, yet the phosphorescence quantum yields are high. Such long-lived strong emissive T1 excited states are rarely reported for cyclometalated Ir(III) complexes. Normally the long-lived triplet excited states and the strong emissive T1 state are exclusive for Ir(III) complexes.14,15,17 We noted the difference between the phosphorescence lifetimes and the triplet excited state lifetimes measured by the nanosecond time-resolved transient absorption spectroscopy. The discrepancy can be attributed to the different TTA effect for the phosphorescence lifetime and the triplet excited state lifetime measurement, for which the laser power is different, as a result, the concentrations of the excited molecules are different. The TTA effect is concentration dependent.
Fig. 5 Spin density of the complexes Ir-0, Ir-1, Ir-2 and Ir-3. Calculated by DFT at the CAM-B3LYP/6-31G/LanL2DZ level using Gaussian 09W.
Firstly the spin density surfaces of the complexes were studied (Fig. 5).25,27,53–56 This approach is helpful to reveal the localization of the triplet excited state of the transition metal complexes, especially for those containing different chromophores, such as the metal coordination center and a remote organic chromophore.25,27,29,36,41,43,54,57 The spin density surfaces of Ir-1–Ir-3 are all localized on the coumarin ligand of the complexes. Therefore, the triplet excited states of the complexes can be assigned as intraligand triplet excited states, which are supported by the nanosecond transient absorption spectroscopy (Fig. 4) in which the bleaching band at the steady state absorption of the coumarin ligand was observed. The ground state geometry of the complexes was optimized (Fig. 6). The coumarin chromophore is almost coplanar to the bpy coordination ligand, thus the π-conjugation between the coumarin moiety and the coordination center was expected. Interestingly, the frontier molecular orbital of Ir-1 indicated that the π-conjugation between the coumarin moiety and the Ir(III) coordination framework is limited. For example, the HOMO of Ir-1 is localized on the coumarin 6 moiety. Conversely, the LUMO of Ir-1 is localized on the coordination bpy ligand.
DFT calculations The photophysical properties of the transition metal complexes can be studied by DFT calculations. For example, the UV-Vis absorption and luminescence emission properties of transition metal complexes have been studied with DFT and TDDFT calculations.47–52 The photophysical properties of the complexes Ir-1–Ir-3 were studied with DFT calculations.
1678 | Dalton Trans., 2014, 43, 1672–1683
Fig. 6 Frontier molecular orbitals of Ir-1. Calculated by DFT at the CAM-B3LYP/6-31G/LanL2DZ level using Gaussian 09W.
This journal is © The Royal Society of Chemistry 2014
View Article Online
Dalton Transactions
Paper
Table 2 Electronic excitation energies (eV) and the corresponding oscillator strengths ( f ), main configurations and CI coefficients of the low-lying electronically excited states of complex Ir-1, calculated by TDDFT//B3LYP/6-31G/LanL2DZ, based on the DFT//CAM-B3LYP/6-31G/LanL2DZ optimized ground state geometries. CH2Cl2 was used as a solvent in the calculations
TDDFT//B3LYP/6-31G
Published on 30 October 2013. Downloaded by St. Petersburg State University on 14/01/2014 10:21:52.
Singlet
Triplet
Electronic transition
Energya (eV)
fb
Compositionc
CId
Character
S0→S1
3.11 eV 398 nm
1.5630
S0→S2 S0→S3 S0→S4 S0→S8 T1→S0 T2→S0 T3→S0
3.31 eV 374 nm 3.38 eV 367 nm 3.74 eV 332 nm 4.12 eV 301 nm 2.03 eV 609 nm 2.28 eV 545 nm 2.78 eV 446 nm
0.0177 0.6317 0.1644 0.4319 0.0000e 0.0000 0.0000
H→L H → L+1 H−1 → L H−1 → L H−1 → L+2 H−8 → L H → L+1 H→L H−1 → L+2
0.4704 0.4564 0.6563 0.4899 0.6478 0.3202 0.6019 0.6568 0.3716
LLCT ILCT ILCT ILCT ILCT ILCT ILCT LLCT ILCT
a
Only the selected low-lying excited states are presented. b Oscillator strength. c H stands for HOMO and L stands for LUMO. Only the main configurations are presented. d The CI coefficients are in absolute values. e No spin–orbital coupling effect was considered, thus the f values are zero.
The UV-Vis absorption of Ir-1 was calculated based on the optimized ground state geometry (Table 2). The calculated absorption band is located at 398 nm, which is close to the experimental value of 439 nm (Fig. 1). The molecular orbitals involved in the S0→S1 transition are HOMO, LUMO and LUMO+1. Inspection of these orbitals (Fig. 6) indicated that the S0→S1 transition can be identified as IL and LLCT transition. Interestingly, the Ir(III) center does not contribute to this transition. The calculated T1 state energy level is 2.03 eV (609 nm), which is close to the experimental value (1.99 eV, 622 nm, approximated by the 77 K emission wavelength of Ir-1). HOMO/LUMO+1 are involved in the T1 state electronic configuration. Therefore, the T1 state can be assigned as an IL state, which is in agreement with the nanosecond timeresolved transient absorption spectra and the long-lived triplet excited state of Ir-1. Similar calculations were carried out for Ir-2 and Ir-3 (see ESI† for details). T1 state energy levels of 1.85 eV and 1.96 eV were found for Ir-2 and Ir-3, respectively. 3 IL excited states were identified for these complexes. Interestingly, we found that using a solvent is necessary in the calculation of the triplet excited state energy levels for the applied functional and the base set. For example, without solvent CH2Cl2, the T1 state energy level of the complexes Ir-1, Ir-2 and Ir-3 were calculated as 1.93 eV, 1.76 eV and 1.85 eV, which are different from the experimental values.
Ir(III), Ru(II) and Re(I) complexes as triplet photosensitizers for TTA upconversion.17 However, more triplet photosensitizers for TTA upconversion are desired so that the tunability of the TTA upconversion can be increased. The ideal triplet photosensitizers for TTA upconversion are those compounds that show strong absorption of visible light and long-lived triplet excited states, so that the triplet–tripletenergy-transfer (TTET) can be enhanced.3,17 Complexes Ir-1, Ir-2 and Ir-3 were used as triplet photosensitizers for TTA upconversion (Fig. 7). Firstly, the emissions of the complexes were recorded upon 473 nm laser excitation. With the addition of the triplet acceptor 9,10-diphenylanthracene, DPA, strong blue emission bands in the region 400 nm–500 nm were observed (Fig. 7b). Under the same experimental conditions, a conventional complex Ru(bmb)3[PF6]2 gave very weak upconversion emission intensity.63 Ir-1 gives the strongest upconverted emission intensity, which is followed by Ir-2. Ir-3 gives the weakest upconverted emission intensity among the three complexes. Notably, Ir-1 and Ir-3 give almost similar absorbances at the excitation wavelength, but there is almost two-fold difference
Triplet–triplet annihilation upconversion Triplet–triplet annihilation upconversion has attracted much attention due to its advantage of tunable photophysical parameters (excitation and emission), strong absorption of excitation light and high upconversion quantum yields.3,9–13 TTA upconversion has been used for solar cells58–61 and photocatalytic reactions.62 One of the major challenges in the development of the TTA upconversion is the availability of the triplet photosensitizers.3,11,17 Conventionally, triplet photosensitizers are limited to the Pt(II)/Pd(II) porphyrin complexes. Recently, we prepared a series of visible light-harvesting Pt(II),
This journal is © The Royal Society of Chemistry 2014
Fig. 7 Emission and upconversion of the complexes with 473 nm (5 mW) continuous laser excitation. (a) Emission of the Ir(III) complexes. (b) The up-converted DPA fluorescence and the residual phosphorescence of the mixture of DPA and the photosensitizers [Ru(dmb)3](PF6)2, Ir-2, Ir-1, or Ir-0 respectively. c[DPA] = 6.0 × 10−5 M; c [sensitizers] = 1.0 × 10−5 M. In deaerated CH2Cl2. The asterisks in (a) and (b) indicate the scattered 473 nm excitation laser, 20 °C.
Dalton Trans., 2014, 43, 1672–1683 | 1679
View Article Online
Paper Table 3
Published on 30 October 2013. Downloaded by St. Petersburg State University on 14/01/2014 10:21:52.
Ir-0 Ir-1 Ir-2 Ir-3
Dalton Transactions Photophysical parameters of Ir-0, Ir-1, Ir-2 and Ir-3
εa
τT b (μs)
ΦUC c
KSV d/M−1 (×105)
Kq e/M−1S−1 (×109)
ηf
0.50 2.01 4.20 4.58
0.3 65.9 34.8 2.7
—g 22.8% 9.8% 9.3%
0.05 31.6 3.21 1.86
16.6 48.0 9.2 68.8
—g 4.6 3.9 4.3
Molar extinction coefficient at the absorption maxima. ε: 104/cm−1 mol−1 dm3. b Triplet state lifetime, measured by time-resolved transient absorption. 1.0 × 10−5 M in deaerated CH2Cl2. c Excited with 473 nm laser. d Quenching constants. e Bimolecular quenching constants. f Overall upconversion ability (brightness). η = ε × ΦUC. In 103 M−1. g Too low to be determined accurately. a
between the upconverted emission intensity. This result is attributed to the long-lived triplet excited state of Ir-1 (65.9 μs) vs. the short-lived triplet excited state of Ir-3 (34.8 μs). The upconversion quantum yields with Ir-1–Ir-3 were calculated as 22.8, 9.8 and 9.3%, respectively (Table 3). In order to study the triplet–triplet-energy-transfer (TTET) efficiency, the quenching of the phosphorescence of the complexes in the presence of DPA was studied (Fig. 8). In the same DPA concentration range, complex Ir-1 gave the most significant quenching effect. Ir-2 gave a slightly decreased quenching effect. For Ir-3 and Ru(dmb)3[PF6]2 (Ru-1), however, quenching was hardly observed. The quenching constants of Ir-1–Ir-3 were determined as 3.16 × 106 M−1, 3.21 × 105 M−1 and 1.86 × 105 M−1, respectively. In order to unambiguously prove that the blue emission in Fig. 7b is due to the TTA upconversion, the luminescence lifetimes of the blue emission bands were measured (Fig. 9). The fluorescence lifetime for the Ir-1/DPA upconversion was
Fig. 8 Quenching of the phosphorescence of the triplet photosensitizers with increasing the concentration of DPA, (a) Ir-1 (λex = 470 nm), (b) Ir-2 (λex = 470 nm), (c) Ir-0 (λex = 445 nm), (d) Stern–Volmer plots for phosphorescence quenching. c[ photosensitizers] = 1.0 × 10−5 M. In deaerated CH2Cl2, 20 °C.
1680 | Dalton Trans., 2014, 43, 1672–1683
Fig. 9 Delayed fluorescence observed in the TTA upconversion with (a) Ir-1, (b) Ir-2, and (c) Ir-3 as triplet photosensitizers and DPA as the triplet acceptor. Excited at 473 nm (nanosecond pulsed OPO laser synchronized with spectro-fluorometer) and monitored at 410 nm. Under this circumstance the Ir-1, Ir-2, and Ir-3 are selectively excited and the emission is due to the upconverted emission of DPA. (d) Prompt fluorescence decay of DPA determined in a different experiment (excited with picosecond 405 nm laser, the decay of the emission was monitored at 410 nm). In deaerated CH2Cl2. c [Sensitizers] = 1.0 × 10−5 M; c [DPA] = 6.0 × 10−5 M; 20 °C.
determined as 282.6 μs. This exceptionally long-lived luminescence lifetime proved the TTA upconversion feature of the blue emission band. Similar results were observed for Ir-2 and Ir-3. For the DPA alone, the fluorescence lifetime was determined as 5.5 ns (Fig. 9d). In order to unambiguously verify the delayed fluorescence feature of the upconverted emission, the time-resolved emission spectra of the complexes and the upconversion were studied. Ir-1 alone shows long-lived phosphorescence emission (τ = 79.9 μs). In the presence of the triplet acceptor DPA, an emission band at 400–500 nm was observed. The lifetime was determined as 282.6 μs (Fig. 9a). Similar long-lived upconverted fluorescence emissions were also observed with Ir-2 and Ir-3. The delayed fluorescence is characteristic of the TTA upconversion. Therefore, the TTA upconversion with the Ir(III) complexes Ir-1, Ir-2 and Ir-3 was proved. The time-resolved emission spectra of the photosensitizers alone and the TTA upconversion are presented in Fig. 10. The TTA upconversion with the triplet photosensitizers is visible to un-aided eyes (Fig. 11). For Ir(III) complexes alone, red emission was observed upon 473 nm laser excitation (Fig. 11a). In the presence of triplet acceptor DPA, a strong blue emission was observed (Fig. 11b), which is due to the TTA upconversion. For the reference complexes Ir-0 and Ru-1, however, no upconverted emission was observed. Previously Ir-0 and Ru-1 were used as triplet photosensitizers for TTA upconversion.63 These results indicated that the Ir(III) complexes (Ir-1–Ir-3) are more efficient triplet photosensitizers for TTA upconversion.
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 30 October 2013. Downloaded by St. Petersburg State University on 14/01/2014 10:21:52.
Dalton Transactions
Fig. 10 Time-resolved emission spectra (TRES) of Ir(III) complexes alone and the TTA upconversion. With DPA as the triplet acceptor. (a) Ir-1 alone (τP = 79.9 μs). (b) TRES of Ir-1 in the presence of DPA. Upconverted emission in the range of 380 nm–750 nm was observed (τDF = 282.6 μs). (c) Ir-2 alone (τP = 30.7 μs). (d) TRES of Ir-2 in the presence of DPA. Upconverted emission in the range of 380 nm–750 nm was observed (τDF = 186.5 μs). Note for Ir-1–Ir-2 the phosphorescence wavelengths are much longer than the upconverted emission wavelength. c [DPA] = 6.0 × 10−5 M; c [sensitizers] = 1.0 × 10−5 M. In deaerated CH2Cl2. Excited with nanosecond pulsed OPO laser synchronized with spectro-fluorometer (λex = 473 nm). 20 °C.
Paper
strong absorption of visible light and a long-lived triplet excited state. Three different coumarin chromophores were used to prepare the Ir(III) complexes, i.e. coumarin 6 (Ir-1), 7-diethylamino coumarin (Ir-2) and 3-phenyl-7-diethylamino coumarin (Ir-3). Ir( ppy)2(bpy)[PF6] was used as the reference complex (Ir-0). All the three new complexes show strong absorption of visible light and the long-lived triplet excited state. We found that the triplet excited state lifetime (65.9 μs) is longer than the complexes prepared with the direct metalation method (visible light-harvesting chromophore as the C donor of the C–Ir bond,
Published on 30 October 2013. Downloaded by St. Petersburg State University on 14/01/2014 10:21:52.
PAPER
Cite this: Dalton Trans., 2014, 43, 1672
View Journal | View Issue
Strongly emissive long-lived 3IL excited state of coumarins in cyclometalated Ir(III) complexes used as triplet photosensitizers and application in triplet–triplet annihilation upconversion† Xiuyu Yi, Caishun Zhang, Song Guo, Jie Ma and Jianzhang Zhao* Three different coumarin chromophores were used to prepare the Ir(III) complexes, i.e. coumarin 6 (Ir-1), 7-diethylamino coumarin (Ir-2) and x-phenyl-y-diethylamino coumarin (Ir-3). Ir( ppy)2(bpy)[PF6] was used as the reference complex (Ir-0). The coumarin units were connected to the coordination center of the complexes by using a CuC π-conjugation bond. The photophysical properties of the complexes were studied with steady state and time-resolved absorption and luminescence spectroscopy, low-temperature luminescence (77 K), as well as DFT calculations. All the three new complexes show strong absorption of visible light (molar absorption coefficient ε is up to 42 000 M−1 cm−1 at 487 nm) and a long-lived triplet excited state (τT = 65.9 μs), compared to the reference complex Ir( ppy)2(bpy)[PF6], which shows the typical weak visible light-absorption (ε < 5000 M−1 cm−1 in the region beyond 400 nm) and a short triplet excited state (τT = 0.3 μs). Interestingly the long-lived triplet excited states are strongly phosphorescent (quantum yield is up to 18.2%, with emission maxima at 607 nm), which is rare for phosphorescent transition metal complexes. With nanosecond time-resolved transient difference absorption spectroscopy we proved that a coumarin-localized triplet excited state (3IL) was produced upon photoexcitation. The complexes were used as triplet photosensitizers for triplet–triplet annihilation upconversion and upcon-
Received 22nd August 2013, Accepted 30th October 2013
version quantum yields up to 22.8% were observed. Our results are useful for the preparation of visible
DOI: 10.1039/c3dt52306c
light-harvesting transition metal complexes, the study of the triplet excited state of organic chromophores, as well as the application of these visible light-harvesting transition metal complexes as efficient
www.rsc.org/dalton
triplet photosensitizers.
Introduction Transition metal complexes such as cyclometalated Ir(III) complexes have attracted much attention due to the applications in photocatalysis,1–3 photodynamic therapy,3–5 phosphorescence and molecular probes,6–8 and more recently in the triplet–triplet annihilation (TTA) upconversion.9–13 However, most of the known Ir(III) complexes show weak absorption of visible light (molar absorption coefficients are smaller than 10 000 M−1 cm−1 in the region >400 nm), and the lifetime of the triplet excited state is short (usually less than 5 μs).14–17 These properties do not hamper the complexes from being used in conventional areas such as electroluminescence,8 but
State Key Laboratory of Fine Chemicals, School of Chemical Engineering, Dalian University of Technology, E-208 West Campus, 2 Ling-Gong Road, Dalian 116024, P. R. China. E-mail: [email protected]; http://finechem.dlut.edu.cn/photochem † Electronic supplementary information (ESI) available: 1H NMR, 13C NMR and HRMS data of the compounds, the photophysical data and DFT calculations of the Ir(III) complexes. See DOI: 10.1039/c3dt52306c
1672 | Dalton Trans., 2014, 43, 1672–1683
they are detrimental to the new applications concerning visible light-harvesting, such as photocatalysis1 and TTA 9–13 upconversion. Therefore, it is highly desired to design cyclometalated Ir(III) complexes that show strong absorption of visible light and long-lived triplet excited states.17–20 To attach a visible light-harvesting ligand to the coordination center, either by a π-conjugation linker or a non-conjugate linker, is the effective method to prepare the aforementioned transition metal complexes.3,17,21–23 A π-conjugation linker between the organic chromophore and the Ir(III) coordination center is preferred, so that the intersystem crossing (ISC) will be improved.24–26 Concerning this aspect, Bodipy,24,27 naphthaimide,28 naphthalenediimide,29 and perylenebisimide have been attached to the Ir(III) coordination center.26 The common feature of these organic chromophorecontaining Ir(III) complexes is the strong absorption of visible light and long-lived triplet excited states.3 We have used these complexes in photocatalysis and TTA upconversion.10 Generally much improved efficiencies were observed for these complexes compared to those conventional Ir(III) complexes such
This journal is © The Royal Society of Chemistry 2014
View Article Online
Dalton Transactions
Published on 30 October 2013. Downloaded by St. Petersburg State University on 14/01/2014 10:21:52.
as Ir( ppy)3, which shows weak absorption of visible light and short-lived triplet excited states.30 However, much room is left to explore the preparation of visible light-absorbing Ir(III) complexes. For example, coumarin is a versatile organic chromophore, which has been extensively used in fluorescent
Scheme 1 Some known coumarin-containing cyclometalated Ir(III) complexes.
Paper
molecular probes or light-harvesting arrays.31–33 However, its application in the triplet excited state manifold is rare.25,34–36 Previously coumarin 6 was used for the preparation of Ir(III) complexes, by the direct cyclometalation method (coumarin 6 as the C donor of the C–Ir bond, Ir–Cou–a, Scheme 1),14a,37 Ru(II) complexes.34,35 However, coumarin derivatives were never connected to the Ir(III) coordination center via a π-conjugation linker, such as a CuC bond. Furthermore, the 3IL emission of coumarin was rarely reported.37 Previously we prepared coumarin-containing Ir(III) complexes with the imidazole linker, and normally the complexes are weakly phosphorescent (Ir–Cou–b–d, Scheme 1).25,38 In these complexes, the π-conjugation framework of the coumarin is not directly connected to the Ir(III) center. In order to overcome the aforementioned challenges, herein we prepared C^N cyclometalated Ir(III) complexes in which three different coumarin chromophores were connected to the Ir(III) coordination center with the π-conjugation linker (Schemes 2 and 3). Strong absorption of visible light and long-lived triplet excited 3IL state were observed for the complexes. Interestingly, strong red phosphorescence originating from the 3IL excited state was observed. The photophysical properties were investigated by steady state and time-resolved spectroscopy, as well as DFT calculations. The phosphorescence lifetimes of Ir-1–Ir-3 are much longer than that of the reference complex Ir-0. The complexes were used as triplet photosensitizers for TTA upconversion. Our results are useful for the preparation of visible light-harvesting cyclometalated Ir(III) complexes and for the application of these complexes in photocatalysis and TTA upconversion.
Scheme 2 Synthesis of Ir-0, Ir-1, Ir-2 and Ir-3. (a) IrCl3·3H2O, 2-ethoxyethanol–water = 3 : 1 (v/v) as solvent, 110 °C, 24 h; (b) CH2Cl2–MeOH = 2 : 1 (v/v), reflux, 5–6 h.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 1672–1683 | 1673
View Article Online
Dalton Transactions
Published on 30 October 2013. Downloaded by St. Petersburg State University on 14/01/2014 10:21:52.
Paper
Scheme 3 Synthesis of Ir-0, Ir-1, Ir-2 and Ir-3. (a) CH3COOCOCH3, HCl/CH3COOH, pyridine, 1 h, room temperature; (b) Br2, CH3COOH, room temperature; (c) K2CO3, CH3OH, 10 min, room temperature; (d) 130 °C, 6 h; (e) CH3OH, 30 °C, 4 h; (f) pyridine, reflux; (g) tributyltin chloride, n-BuLi, DEE; (h) 2,5-dibromopyridine, Pd(PPh3)4, xylene; (i) trimethylsilylacetylene, Pd(PPh3)2Cl2, PPh3, CuI, THF/NEt3; ( j) TBAF, THF, Ar, room temperature; (k) Pd(PPh3)4, CuI, THF/NEt3.
Experimental General information NMR spectra were taken on a 400 MHz Varian Unity Inova spectrophotometer. Mass spectra were recorded with a Q-TOF Micro MS spectrometer. UV-Vis spectra were taken on a HP8453 UV-Visible spectrophotometer. Fluorescence spectra were recorded on a Shimadzu RF-5301PC spectrofluorometer. Luminescence quantum yields were measured with [Ru(dmb)3] [PF6]2 (Ru-1) as standard (ΦP = 0.073 in deaerated acetonitrile). Luminescence lifetimes were measured on an OB 920 luminescence lifetime spectrometer (Edinburgh, UK).
1674 | Dalton Trans., 2014, 43, 1672–1683
Compounds 1–8 were synthesized according to the literature methods.39,40 9–12 were prepared by the reported method.41 Compound L1 5 (120 mg, 0.28 mmol) and 5-ethynyl-2,2-bipyridine (12) (50.0 mg, 0.28 mmol) were added to deaerated triethylamine (20 mL). Then Pd(PPh3)4 (8.0 mg, 5 mol%) and CuI (3 mg, 5 mol%) were added under an Ar atmosphere. The mixture was stirred and refluxed for 8 h under Ar. After completion of the reaction, the mixture was cooled to RT and the light red precipitate was collected by filtration. The crude product was
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 30 October 2013. Downloaded by St. Petersburg State University on 14/01/2014 10:21:52.
Dalton Transactions
purified with column chromatography (silica gel, ethyl acetate–petroleum ether = 1 : 2, v/v) to give an orange solid. Yield: 101 mg (68.2%). M.p.: 201.3–201.9 °C. 1H NMR (400 MHz, CDCl3), δ = 8.91 (s, 1H); 8.81 (s, 1H); 8.72 (d, 1H, J = 2.0 Hz); 8.46 (t, 2H, J = 8.7 Hz); 8.06 (d, 1H, J = 7.6 Hz); 7.96–7.85 (m, 3H); 7.79 (s, 1H); 7.51 (t, 1H, J = 7.6 Hz); 7.39 (m, 2H); 6.74 (s, 1H); 3.71 (m, 4H); 1.34 (t, 6H, J = 5.8 Hz). 13C NMR (100 MHz, CDCl3): δ = 161.0, 160.5, 155.9, 155.3, 154.8, 152.6, 151.4, 149.3, 141.0, 140.5, 139.1, 137.4, 136.6, 131.1, 129.0, 126.4, 124.2, 121.8, 121.6, 120.7, 120.6, 118.8, 115.2, 110.6, 108.4, 102.0, 92.5, 90.5, 46.1, 13.4 ppm. EI-HRMS: calcd (C32H24N4O2S): m/z = 528.1620, found, m/z = 528.1628. Compound L2 The synthetic procedure is similar to that of L1 except that compound 7 (63 mg, 0.28 mmol), instead of 5, was used. The crude product was purified with column chromatography (silica gel, ethyl acetate–petroleum ether = 1 : 2, v/v) to give an orange solid. Yield: 81 mg (73.6%). M.p.: 183.7–184.1 °C. 1H NMR (400 MHz, CDCl3), δ = 8.83 (s, 1H); 8.70 (s, 1H); 8.42 (t, 2H, J = 4.9 Hz); 7.97 (d, 1H, J = 9.0 Hz); 7.85 (m, 3H); 7.79 (m, 2H); 7.33 (d, 1H, J = 4.1 Hz); 7.27 (d, 1H, J = 7.9 Hz); 6.61 (d, 1H, J = 8.7 Hz); 6.50 (s, 1H); 3.46–3.41 (m, 4H); 1.23 (t, 6H, J = 7.2 Hz). 13C NMR (100 MHz, CDCl3): δ = 160.9, 156.6, 154.8, 151.6, 149.4, 146.3, 139.6, 137.2, 132.3, 129.3, 128.7, 128.6, 124.1, 121.6, 120.5, 109.5, 108.5, 103.9, 97.4, 90.3, 89.1, 45.1, 12.6 ppm. EI-HRMS: calcd (C25H21N3O2): m/z = 395.1634, found, m/z = 395.1644. Compound L3 The synthetic procedure is similar to that of L1 except that compound 8 (104 mg, 0.28 mmol), instead of 5, was used. The crude product was purified with column chromatography (silica gel, ethyl acetate–petroleum ether = 1 : 2, v/v) to give an orange solid. Yield: 66 mg (50.1%). M.p.: 181.2–181.7 °C. 1H NMR (400 MHz, CDCl3), δ = 8.82 (s, 1H); 8.69 (d, 1H, J = 3.6 Hz); 8.43–8.40 (m, 2H); 7.95 (d, 1H, J = 8.3 Hz); 7.83–7.73 (m, 4H); 7.60 (d, 2H, J = 8.1 Hz); 7.29 (t, 2H, J = 9.7 Hz); 6.59 (d, 1H, J = 10.2 Hz); 6.51 (s, 1H); 3.44–3.38 (m, 4H); 1.23 (t, 6H, J = 7.0 Hz). 13C NMR (100 MHz, CDCl3): δ = 161.4, 156.4, 155.6, 154.9, 151.7, 150.9, 149.4, 140.9, 139.4, 137.1, 136.4, 131.7, 129.3, 128.2, 124.0, 121.7, 121.5, 120.4, 119.8, 109.2, 109.1, 97.1, 93.8, 87.2, 45.0, 12.6 ppm. EI-HRMS: calcd (C31H25N3O2): m/z = 471.1947, found, m/z = 471.1953. General method for synthesis of the Ir(III) complexes All procedures were carried out under an Ar atmosphere. Cyclometalated Ir(III) chloro-bridged dimers were synthesized according to a literature method.38 IrCl3·3H2O and 2.5 equiv. of ligand were heated in a 3 : 1 mixture of 2-ethoxyethanol and water (v/v). The slurry was heated at 110 °C for 24 h. After cooling to rt, the crude product was filtered. The solid residue was washed with water, followed by two portions of n-hexane and ether. Then the solid was dried under vacuum.
This journal is © The Royal Society of Chemistry 2014
Paper
Complex Ir-0 A solution of [Ir( ppy)2Cl]2 (43 mg, 0.04 mmol) and 2,2′-bipyridine (16 mg, 0.1 mmol) in CH2Cl2–MeOH (12 mL, 2 : 1 v/v) was refluxed under an Ar atmosphere. After 5–6 h, the orange solution was cooled to rt, and then a 10-fold excess of ammonium hexafluorophosphate was added. The suspension was stirred for 15 min and then filtered to remove insoluble inorganic salts. The solution was evaporated to dryness under reduced pressure to obtain a crude orange solid. The solid was dissolved in CH2Cl2 and filtered to remove the residual traces of inorganic salts, then purified with column chromatography (silica gel, CH2Cl2–MeOH = 25 : 1) to give a light yellow solid. Yield: 48 mg (76.2%). M.p. >250 °C. 1H NMR (400 MHz, CDCl3, CH3OD), δ = 8.49 (d, 4H, J = 7.8 Hz), 8.18 (t, 1H, J = 9.0 Hz), 7.94–7.87 (m, 6H), 7.78 (t, 1H, J = 7.8 Hz), 7.70 (d, 1H, J = 7.3 Hz), 7.52 (d, 1H, J = 7.8 Hz), 7.42–7.36 (m, 6H), 7.07–7.00 (m, 2H), 6.94 (t, 1H, J = 7.1 Hz), 6.31 (d, 1H, J = 7.6 Hz). ESI-HRMS: [(M − PF6)]+, calc (C32H24N4Ir): m/z = 657.1630, found, m/z = 657.1625. Complex Ir-1 The synthetic procedure is the same as that for Ir-0, except that L1 (53 mg, 0.1 mmol) was used. The solution was evaporated to dryness under reduced pressure to obtain a crude orange solid. The solid was dissolved in CH2Cl2 and filtered to remove the residual inorganic salts, then purified by column chromatography (silica gel, CH2Cl2–MeOH, 25 : 1, v/v) to give a red solid. Yield: 70 mg, (75.1%). M.p. >250 °C. 1H NMR (400 MHz, DMSO-d6), δ = 9.00 (s, 1H); 8.89 (d, 2H, J = 8.4 Hz); 8.32 (t, 4H, J = 16.4 Hz); 8.09 (d, 1H, J = 7.9 Hz); 8.03–7.93 (m, 6H); 7.89 (s, 2H); 7.83 (d, 1H, J = 5.5 Hz); 7.74 (t, 1H, J = 12.8 Hz); 7.68 (d, 1H, J = 5.5 Hz); 7.55 (t, 1H, J = 15.3 Hz); 7.42 (t, 1H, J = 15.0 Hz); 7.23–7.17 (m, 2H), 7.07–7.02 (m, 2H); 6.96–6.91 (m, 2H); 6.81(s, 1H); 6.24 (t, 2H, J = 17.1 Hz); 3.50 (m, 4H); 1.04 (t, 6H, J = 13.6 Hz). 13C NMR (100 MHz, DMSO-d6): δ 166.9, 166.8, 160.4, 159.6, 155.8, 154.7, 154.0, 153.8, 152.0, 150.7, 150.2, 149.9, 149.2, 143.9, 143.8, 141.2, 140.6, 139.8, 139.1, 135.6, 131.2, 130.4, 126.6, 125.5, 124.7, 124.2, 123.8, 122.6, 122.1, 120.2, 113.9, 110.0, 105.0, 100.8, 96.1, 87.7, 45.5, 13.1 ppm. ESI-HRMS: [(M − PF6)]+, calc (C54H40N6IrO2S): m/z = 1029.2563, found, m/z = 1029.2568. Complex Ir-2 The synthetic procedure is the same as that for Ir-0, except that L2 (40 mg, 0.1 mmol) was used. The solution was evaporated to dryness under reduced pressure to obtain a crude red solid. The solid was dissolved in CH2Cl2 and filtered to remove the residual inorganic salts, then purified by column chromatography (silica gel, CH2Cl2–MeOH, 25 : 1, v/v) to give a red solid. Yield: 65.8 mg, (79.0%). M.p. >250 °C. 1H NMR (400 MHz, CDCl3/CD3OD), δ = 8.59 (d, 2H, J = 4.4 Hz); 8.27 (d, 1H, J = 7.6 Hz); 8.17 (t, 1H, J = 15.8 Hz); 8.07 (d, 2H, J = 6.8 Hz); 7.88 (s, 2H); 7.64–7.59 (m, 4H); 7.54 (s, 3H); 7.31 (d, 2H, J = 7.9 Hz); 7.10–7.06 (m, 4H); 6.96 (d, 2H, J = 7.1 Hz); 6.71 (d, 1H, J = 9.3 Hz); 6.51(s, 1H); 6.33–6.28 (m, 2H); 3.52–3.47 (m, 4H); 1.27
Dalton Trans., 2014, 43, 1672–1683 | 1675
View Article Online
Paper
Dalton Transactions
(t, 6H, J = 13.8 Hz). 13C NMR (100 MHz, CDCl3/CD3OD): δ 167.7, 166.6, 156.7, 153.8, 152.3, 150.7, 148.3, 143.3, 139.3, 138.3, 132.4, 131.4, 130.6, 129.8, 129.2, 128.7, 124.7, 123.2, 122.8, 120.5, 120.0, 113.7, 110.0, 108.2, 96.6, 88.6, 87.3, 44.9, 12.0 ppm. ESI-HRMS: [(M − PF6)]+, calc (C47H37N5IrO2): m/z = 896.2576, found, m/z = 896.2562.
Φsam ¼ 2Φstd
Complex Ir-3 Published on 30 October 2013. Downloaded by St. Petersburg State University on 14/01/2014 10:21:52.
78.0% in CH3CH2OH) and the quantum yields were calculated with eqn (2), where Φunk, Aunk, Iunk and ηunk represents the quantum yield, absorbance, integrated photoluminescence intensity of the samples and the refractive index of the solvents.9
The synthetic procedure is the same as that for Ir-0, except that L3 (47 mg, 0.1 mmol) was used. The solution was evaporated to dryness under reduced pressure to obtain a crude red solid. The solid was dissolved in CH2Cl2 and filtered to remove the residual inorganic salts, then purified by column chromatography (silica gel, CH2Cl2–MeOH, 25 : 1, v/v) to give a red solid. Yield: 67.2 mg (75.2%). M.p. >250 °C. 1H NMR (400 MHz, CDCl3), δ = 8.65 (d, 2H, J = 8.3 Hz); 8.17 (m, 2H); 7.98–7.92 (m, 4H); 7.80–7.70 (m, 7H); 7.57 (d, 1H, J = 5.3 Hz); 7.52–7.46 (m, 3H); 7.42–7.36 (m, 2H); 7.07–7.01 (m, 4H); 6.96–6.90 (m, 2H); 6.70 (s, 1H); 6.61 (s, 1H); 6.32–6.27 (m, 2H); 3.47–3.42 (m, 4H); 1.25 (t, 6H, J = 14.2 Hz). 13C NMR (100 MHz, DMSO-d6): δ 166.7, 166.6, 160.2, 156.0, 154.7, 154.5, 150.8, 150.1, 149.8, 149.4, 149.0, 139.7, 138.9, 138.8, 137.2, 131.4, 131.1, 130.9, 130.4, 130.2, 130.0, 128.1, 125.5, 125.0, 124.9, 124.0, 123.6, 122.4, 120.2, 120.0, 119.0, 117.0, 109.3, 108.3, 96.4, 85.1, 44.1, 12.3. ESI-HRMS: [(M − PF6)]+, calc (C53H41N5IrO2): m/z = 972.2890, found, m/z = 972.2897. Singlet oxygen (1O2) quantum yields (ΦΔ) The 1O2 quantum yields (ΦΔ) of the photosensitizers were determined with 2,6-diiodo-bodipy as standard (ΦΔ = 0.83 in dichloromethane, DCM). Air saturated DCM was obtained by bubbling air for 15 min. The absorbance of the 1O2 scavenger 1,3-diphenylisobenzofuran (DPBF) was adjusted around 1.0 in air saturated DCM. Then, the photosensitizer was added to the cuvette and the photosensitizer’s absorbance was adjusted to around 0.2–0.3. The cuvette was irradiated with monochromatic light at the peak absorption wavelength for 10 seconds. Absorbance was measured several times after each irradiation. The slopes of absorbance maxima of DPBF at 414 nm versus time graph for each photosensitizer were calculated. Singlet oxygen quantum yield (ΦΔ) were calculated according to a modified equation: Φsam ¼ Φref
ksam F ref kref F sam
ð1Þ
where ‘sam’ and ‘ref’ designate the photosensitizers and ‘MB’ respectively. k is the slope of the plots of the absorbance of DPBF (414 nm) against the irradiation time, F is the absorption correction factor, which is given by F = 1–10−O.D. (O.D. at the irradiation wavelength). Triplet–triplet annihilation upconversion Diode pumped solid state laser was used for the upconversions. The samples were purged with N2 for 15 min before measurement. The upconversion quantum yields were determined with coumarin-6 as the quantum yield standard (Φ =
1676 | Dalton Trans., 2014, 43, 1672–1683
Astd Asam
I sam I std
ηsam 2 ηstd
ð2Þ
The delayed fluorescence of the upconversion was measured with nanosecond pulsed laser (Opolett™ 355 II nanosecond pulsed laser, pulse length: 7 ns; pulse repetition: 20 Hz; peak OPO energy: 4 mJ; OPOTEK, USA), which is synchronized to a FLS 920 spectrofluorometer (Edinburgh, U.K.). The pulsed laser is sufficient to sensitize the TTA upconversion. The decay kinetics of the upconverted fluorescence (delayed fluorescence) was monitored with a FLS 920 spectrofluorometer (synchronized to the OPO nanosecond pulsed laser). The prompt fluorescence lifetimes were measured with EPL picosecond pulsed laser (405 nm) which is synchronized to the FLS 920 spectrofluorometer. The photographs of the upconversion were taken with Sumsung NV 5 digital camera. DFT calculations The density functional theory (DFT) calculations were used for the optimization of the singlet states and triplet states. The spin density surfaces of the complexes were calculated based on the optimized triplet state geometries. All the calculations were performed with Gaussian 09W.42
Results and discussion Design and synthesis of the complexes Three different coumarin chromophores were selected, i.e. the coumarin 6, coumarin and the phenylcoumarin. All the chromophores were connected to the bpy ligand via the π-conjugation linker, so that the ISC can be enhanced. Coumarin 6 and diethylamino coumarin are known to be highly fluorescent. The subtle difference between the two chromophores may cause different effects on the photophysical properties of the complexes. In Ir-3, the diethylamino coumarin moiety was isolated from the Ir(III) coordination center by a phenyl unit. In order to prepare coumarin 6, 5-bromo-4-diethylsalicylaldehyde was prepared. Then the bromo coumarin 6 was prepared (Scheme 3). Sonogashira coupling between compound 5 and 5-ethynyl-2,2′-bipyridine gives ligand L-1. Ligands L-2 and L-3 were prepared by similar methods. All the compounds were obtained in satisfactory yields (Scheme 2). Complex Ir-0 was used as reference complex. UV-Vis absorption and emission spectra The UV-Vis absorption spectra of the complexes were studied (Fig. 1). The reference complex Ir-0 gives weak absorption in the visible region, with ε < 5000 M−1 cm−1 in the region beyond 400 nm. In contrast, all the coumarin-containing complexes Ir-1, Ir-2 and Ir-3 give strong absorption in the visible
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 30 October 2013. Downloaded by St. Petersburg State University on 14/01/2014 10:21:52.
Dalton Transactions
Paper
Fig. 3 Photoluminescence spectra of the complexes at RT and 77 K. (a) Ir-1 (λex = 470 nm), (b) Ir-2 (λex = 470 nm). In C2H5OH–CH3OH (4 : 1, v/v), 1.0 × 10−5 M. Fig. 1 UV-Vis absorption of Ir(III) complexes. c = 1.0 × 10−5 M in CH2Cl2, 20 °C.
region. For example, Ir-1 shows a broad absorption band in the region of 350 nm–450 nm, with maximum at 439 nm (22 440 M−1 cm−1). Ir-2 gives stronger absorption at 487 nm (42 000 M−1 cm−1). In comparison Ir-3 gives slightly blueshifted absorption at 437 nm (45 800 M−1 cm−1). The luminescence spectra of the complexes were studied (Fig. 2). Ir-1 gives emission at 607 nm. Similar emission bands were observed for Ir-2 and Ir-3. Furthermore, we found that the emission of the complexes Ir-1–Ir-3 can be dramatically quenched by O2. Therefore, the luminescence can be attributed to phosphorescence. Moreover, the emissions of Ir-1 and Ir-2 are almost completely quenched by O2, indicating the longer phosphorescence lifetime of Ir-1 and Ir-2 than that of Ir-3.
measured and were compared with those at room temperature (Fig. 3).24,25,27,43 For Ir-1, the emission spectra became more structured at 77 K compared to that at RT. The thermally induced Stokes shift is very small. Therefore, the emission of Ir-1 can be attributed to the intraligand triplet excited state (3IL state) localized on the coumarin unit.25 For Ir-2, the structured emission spectrum at 77 K was observed, with a small thermally induced Stokes shift of 481 cm−1. The emissive triplet excited state of Ir-2 can also be assigned as an 3IL state. Similar results were observed for Ir-3 (see ESI† for detail). For the reference complex Ir-0, however, the emission band at 77 K is structureless and the thermally induced Stokes shift is larger. DFT calculations on the T1 state energy level support the assignment of the triplet excited states as the intraligand triplet excited state.44,45 Nanosecond time-resolved transient absorption spectra
77 K emission spectra In order to study the emissive states of the complexes, the phosphorescence spectra of the complexes at 77 K were
In order to study the triplet excited state of the complexes, the nanosecond time-resolved transient difference absorption spectra of the complexes were investigated (Fig. 4). For Ir-1,
Fig. 2 The emission spectra of the complexes under different atmosphere. (a) Ir-1 (λex = 470 nm), (b) Ir-2 (λex = 475 nm), (c) Ir-3 (λex = 438 nm), (d) Ir-0 (λex = 403 nm). c = 1.0 × 10−5 M in CH2Cl2, 20 °C.
Fig. 4 Nanosecond time-resolved transient difference absorption spectra of (a) Ir-1, (c) Ir-2 after pulsed excitation (λex = 532 nm). Decay trace of (b) Ir-1 at 470 nm and (d) Ir-2 at 490 nm. 1.0 × 10−5 M in deaerated CH2Cl2. 20 °C.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 1672–1683 | 1677
View Article Online
Paper
Dalton Transactions
Table 1
Photophysical parameters of Ir-0, Ir-1, Ir-2 and Ir-3
Published on 30 October 2013. Downloaded by St. Petersburg State University on 14/01/2014 10:21:52.
τL e/μs
Ir-0 Ir-1 Ir-2 Ir-3
λabs a
εb
λem c
ΦL d/%
77 K
298 K
τT f (μs)
ΦΔ g
405 473 487 437
0.50 2.01 4.20 4.58
578 607 645 640
4.3 18.2 6.5 1.1
4.8 9962.8 137.6 1241.0
0.3 102.2 36.7 7.37
0.3 65.9 34.8 2.7
— 0.803 0.721 0.585
In CH2Cl2 (1.0 × 10−5 M). b Molar extinction coefficient at the absorption maxima. ε: 104/cm−1 mol−1 dm3. c In CH2Cl2. d In CH2Cl2, with [Ru(dmb)3][PF6]2 (Ru-1) (Φ = 0.073 in deaerated acetonitrile) as the standards. e Luminescence lifetime, at 298 K (in CH2Cl2) and 77 K (in C2H5OH–CH3OH = 4 : 1). f Triplet state lifetime, measured by timeresolved transient absorption. 1.0 × 10−5 M in deaerated CH2Cl2. g Singlet oxygen quantum yield. With 2,6-diiodo-Bodipy as a standard (ΦΔ = 0.83 in DCM). Photoexcitation wavelengths of 494 nm, 509 nm, 490 nm were used. a
upon visible light excitation, a bleaching band at 444 nm was observed, which is due to the depletion of the ground state. Positive absorption in the region of 550–750 nm was observed. The lifetime of the transient was determined as 65.9 μs, which was substantially quenched in an aerated solution. Therefore, the observed long-lived transient species is due to the triplet excited state of the complex, which is localized on coumarin 6. These results indicated a subtle variation of the molecular structure of the moiety.24,25 A similar transient was observed for Ir-2, but the lifetime was reduced to 34.6 μs. For Ir-3, however, the triplet excited state lifetime is much shorter, 2.7 μs (Table 1). Since the main application of the triplet photosensitizers is the triplet–triplet energy transfer (TTET), the efficiency of the TTET processes will be affected by the lifetime of the triplet excited state of the complexes.10,12,13,46,47 The photophysical properties of the compounds are listed in Table 1. Notably, the complexes have long-lived triplet excited states, yet the phosphorescence quantum yields are high. Such long-lived strong emissive T1 excited states are rarely reported for cyclometalated Ir(III) complexes. Normally the long-lived triplet excited states and the strong emissive T1 state are exclusive for Ir(III) complexes.14,15,17 We noted the difference between the phosphorescence lifetimes and the triplet excited state lifetimes measured by the nanosecond time-resolved transient absorption spectroscopy. The discrepancy can be attributed to the different TTA effect for the phosphorescence lifetime and the triplet excited state lifetime measurement, for which the laser power is different, as a result, the concentrations of the excited molecules are different. The TTA effect is concentration dependent.
Fig. 5 Spin density of the complexes Ir-0, Ir-1, Ir-2 and Ir-3. Calculated by DFT at the CAM-B3LYP/6-31G/LanL2DZ level using Gaussian 09W.
Firstly the spin density surfaces of the complexes were studied (Fig. 5).25,27,53–56 This approach is helpful to reveal the localization of the triplet excited state of the transition metal complexes, especially for those containing different chromophores, such as the metal coordination center and a remote organic chromophore.25,27,29,36,41,43,54,57 The spin density surfaces of Ir-1–Ir-3 are all localized on the coumarin ligand of the complexes. Therefore, the triplet excited states of the complexes can be assigned as intraligand triplet excited states, which are supported by the nanosecond transient absorption spectroscopy (Fig. 4) in which the bleaching band at the steady state absorption of the coumarin ligand was observed. The ground state geometry of the complexes was optimized (Fig. 6). The coumarin chromophore is almost coplanar to the bpy coordination ligand, thus the π-conjugation between the coumarin moiety and the coordination center was expected. Interestingly, the frontier molecular orbital of Ir-1 indicated that the π-conjugation between the coumarin moiety and the Ir(III) coordination framework is limited. For example, the HOMO of Ir-1 is localized on the coumarin 6 moiety. Conversely, the LUMO of Ir-1 is localized on the coordination bpy ligand.
DFT calculations The photophysical properties of the transition metal complexes can be studied by DFT calculations. For example, the UV-Vis absorption and luminescence emission properties of transition metal complexes have been studied with DFT and TDDFT calculations.47–52 The photophysical properties of the complexes Ir-1–Ir-3 were studied with DFT calculations.
1678 | Dalton Trans., 2014, 43, 1672–1683
Fig. 6 Frontier molecular orbitals of Ir-1. Calculated by DFT at the CAM-B3LYP/6-31G/LanL2DZ level using Gaussian 09W.
This journal is © The Royal Society of Chemistry 2014
View Article Online
Dalton Transactions
Paper
Table 2 Electronic excitation energies (eV) and the corresponding oscillator strengths ( f ), main configurations and CI coefficients of the low-lying electronically excited states of complex Ir-1, calculated by TDDFT//B3LYP/6-31G/LanL2DZ, based on the DFT//CAM-B3LYP/6-31G/LanL2DZ optimized ground state geometries. CH2Cl2 was used as a solvent in the calculations
TDDFT//B3LYP/6-31G
Published on 30 October 2013. Downloaded by St. Petersburg State University on 14/01/2014 10:21:52.
Singlet
Triplet
Electronic transition
Energya (eV)
fb
Compositionc
CId
Character
S0→S1
3.11 eV 398 nm
1.5630
S0→S2 S0→S3 S0→S4 S0→S8 T1→S0 T2→S0 T3→S0
3.31 eV 374 nm 3.38 eV 367 nm 3.74 eV 332 nm 4.12 eV 301 nm 2.03 eV 609 nm 2.28 eV 545 nm 2.78 eV 446 nm
0.0177 0.6317 0.1644 0.4319 0.0000e 0.0000 0.0000
H→L H → L+1 H−1 → L H−1 → L H−1 → L+2 H−8 → L H → L+1 H→L H−1 → L+2
0.4704 0.4564 0.6563 0.4899 0.6478 0.3202 0.6019 0.6568 0.3716
LLCT ILCT ILCT ILCT ILCT ILCT ILCT LLCT ILCT
a
Only the selected low-lying excited states are presented. b Oscillator strength. c H stands for HOMO and L stands for LUMO. Only the main configurations are presented. d The CI coefficients are in absolute values. e No spin–orbital coupling effect was considered, thus the f values are zero.
The UV-Vis absorption of Ir-1 was calculated based on the optimized ground state geometry (Table 2). The calculated absorption band is located at 398 nm, which is close to the experimental value of 439 nm (Fig. 1). The molecular orbitals involved in the S0→S1 transition are HOMO, LUMO and LUMO+1. Inspection of these orbitals (Fig. 6) indicated that the S0→S1 transition can be identified as IL and LLCT transition. Interestingly, the Ir(III) center does not contribute to this transition. The calculated T1 state energy level is 2.03 eV (609 nm), which is close to the experimental value (1.99 eV, 622 nm, approximated by the 77 K emission wavelength of Ir-1). HOMO/LUMO+1 are involved in the T1 state electronic configuration. Therefore, the T1 state can be assigned as an IL state, which is in agreement with the nanosecond timeresolved transient absorption spectra and the long-lived triplet excited state of Ir-1. Similar calculations were carried out for Ir-2 and Ir-3 (see ESI† for details). T1 state energy levels of 1.85 eV and 1.96 eV were found for Ir-2 and Ir-3, respectively. 3 IL excited states were identified for these complexes. Interestingly, we found that using a solvent is necessary in the calculation of the triplet excited state energy levels for the applied functional and the base set. For example, without solvent CH2Cl2, the T1 state energy level of the complexes Ir-1, Ir-2 and Ir-3 were calculated as 1.93 eV, 1.76 eV and 1.85 eV, which are different from the experimental values.
Ir(III), Ru(II) and Re(I) complexes as triplet photosensitizers for TTA upconversion.17 However, more triplet photosensitizers for TTA upconversion are desired so that the tunability of the TTA upconversion can be increased. The ideal triplet photosensitizers for TTA upconversion are those compounds that show strong absorption of visible light and long-lived triplet excited states, so that the triplet–tripletenergy-transfer (TTET) can be enhanced.3,17 Complexes Ir-1, Ir-2 and Ir-3 were used as triplet photosensitizers for TTA upconversion (Fig. 7). Firstly, the emissions of the complexes were recorded upon 473 nm laser excitation. With the addition of the triplet acceptor 9,10-diphenylanthracene, DPA, strong blue emission bands in the region 400 nm–500 nm were observed (Fig. 7b). Under the same experimental conditions, a conventional complex Ru(bmb)3[PF6]2 gave very weak upconversion emission intensity.63 Ir-1 gives the strongest upconverted emission intensity, which is followed by Ir-2. Ir-3 gives the weakest upconverted emission intensity among the three complexes. Notably, Ir-1 and Ir-3 give almost similar absorbances at the excitation wavelength, but there is almost two-fold difference
Triplet–triplet annihilation upconversion Triplet–triplet annihilation upconversion has attracted much attention due to its advantage of tunable photophysical parameters (excitation and emission), strong absorption of excitation light and high upconversion quantum yields.3,9–13 TTA upconversion has been used for solar cells58–61 and photocatalytic reactions.62 One of the major challenges in the development of the TTA upconversion is the availability of the triplet photosensitizers.3,11,17 Conventionally, triplet photosensitizers are limited to the Pt(II)/Pd(II) porphyrin complexes. Recently, we prepared a series of visible light-harvesting Pt(II),
This journal is © The Royal Society of Chemistry 2014
Fig. 7 Emission and upconversion of the complexes with 473 nm (5 mW) continuous laser excitation. (a) Emission of the Ir(III) complexes. (b) The up-converted DPA fluorescence and the residual phosphorescence of the mixture of DPA and the photosensitizers [Ru(dmb)3](PF6)2, Ir-2, Ir-1, or Ir-0 respectively. c[DPA] = 6.0 × 10−5 M; c [sensitizers] = 1.0 × 10−5 M. In deaerated CH2Cl2. The asterisks in (a) and (b) indicate the scattered 473 nm excitation laser, 20 °C.
Dalton Trans., 2014, 43, 1672–1683 | 1679
View Article Online
Paper Table 3
Published on 30 October 2013. Downloaded by St. Petersburg State University on 14/01/2014 10:21:52.
Ir-0 Ir-1 Ir-2 Ir-3
Dalton Transactions Photophysical parameters of Ir-0, Ir-1, Ir-2 and Ir-3
εa
τT b (μs)
ΦUC c
KSV d/M−1 (×105)
Kq e/M−1S−1 (×109)
ηf
0.50 2.01 4.20 4.58
0.3 65.9 34.8 2.7
—g 22.8% 9.8% 9.3%
0.05 31.6 3.21 1.86
16.6 48.0 9.2 68.8
—g 4.6 3.9 4.3
Molar extinction coefficient at the absorption maxima. ε: 104/cm−1 mol−1 dm3. b Triplet state lifetime, measured by time-resolved transient absorption. 1.0 × 10−5 M in deaerated CH2Cl2. c Excited with 473 nm laser. d Quenching constants. e Bimolecular quenching constants. f Overall upconversion ability (brightness). η = ε × ΦUC. In 103 M−1. g Too low to be determined accurately. a
between the upconverted emission intensity. This result is attributed to the long-lived triplet excited state of Ir-1 (65.9 μs) vs. the short-lived triplet excited state of Ir-3 (34.8 μs). The upconversion quantum yields with Ir-1–Ir-3 were calculated as 22.8, 9.8 and 9.3%, respectively (Table 3). In order to study the triplet–triplet-energy-transfer (TTET) efficiency, the quenching of the phosphorescence of the complexes in the presence of DPA was studied (Fig. 8). In the same DPA concentration range, complex Ir-1 gave the most significant quenching effect. Ir-2 gave a slightly decreased quenching effect. For Ir-3 and Ru(dmb)3[PF6]2 (Ru-1), however, quenching was hardly observed. The quenching constants of Ir-1–Ir-3 were determined as 3.16 × 106 M−1, 3.21 × 105 M−1 and 1.86 × 105 M−1, respectively. In order to unambiguously prove that the blue emission in Fig. 7b is due to the TTA upconversion, the luminescence lifetimes of the blue emission bands were measured (Fig. 9). The fluorescence lifetime for the Ir-1/DPA upconversion was
Fig. 8 Quenching of the phosphorescence of the triplet photosensitizers with increasing the concentration of DPA, (a) Ir-1 (λex = 470 nm), (b) Ir-2 (λex = 470 nm), (c) Ir-0 (λex = 445 nm), (d) Stern–Volmer plots for phosphorescence quenching. c[ photosensitizers] = 1.0 × 10−5 M. In deaerated CH2Cl2, 20 °C.
1680 | Dalton Trans., 2014, 43, 1672–1683
Fig. 9 Delayed fluorescence observed in the TTA upconversion with (a) Ir-1, (b) Ir-2, and (c) Ir-3 as triplet photosensitizers and DPA as the triplet acceptor. Excited at 473 nm (nanosecond pulsed OPO laser synchronized with spectro-fluorometer) and monitored at 410 nm. Under this circumstance the Ir-1, Ir-2, and Ir-3 are selectively excited and the emission is due to the upconverted emission of DPA. (d) Prompt fluorescence decay of DPA determined in a different experiment (excited with picosecond 405 nm laser, the decay of the emission was monitored at 410 nm). In deaerated CH2Cl2. c [Sensitizers] = 1.0 × 10−5 M; c [DPA] = 6.0 × 10−5 M; 20 °C.
determined as 282.6 μs. This exceptionally long-lived luminescence lifetime proved the TTA upconversion feature of the blue emission band. Similar results were observed for Ir-2 and Ir-3. For the DPA alone, the fluorescence lifetime was determined as 5.5 ns (Fig. 9d). In order to unambiguously verify the delayed fluorescence feature of the upconverted emission, the time-resolved emission spectra of the complexes and the upconversion were studied. Ir-1 alone shows long-lived phosphorescence emission (τ = 79.9 μs). In the presence of the triplet acceptor DPA, an emission band at 400–500 nm was observed. The lifetime was determined as 282.6 μs (Fig. 9a). Similar long-lived upconverted fluorescence emissions were also observed with Ir-2 and Ir-3. The delayed fluorescence is characteristic of the TTA upconversion. Therefore, the TTA upconversion with the Ir(III) complexes Ir-1, Ir-2 and Ir-3 was proved. The time-resolved emission spectra of the photosensitizers alone and the TTA upconversion are presented in Fig. 10. The TTA upconversion with the triplet photosensitizers is visible to un-aided eyes (Fig. 11). For Ir(III) complexes alone, red emission was observed upon 473 nm laser excitation (Fig. 11a). In the presence of triplet acceptor DPA, a strong blue emission was observed (Fig. 11b), which is due to the TTA upconversion. For the reference complexes Ir-0 and Ru-1, however, no upconverted emission was observed. Previously Ir-0 and Ru-1 were used as triplet photosensitizers for TTA upconversion.63 These results indicated that the Ir(III) complexes (Ir-1–Ir-3) are more efficient triplet photosensitizers for TTA upconversion.
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 30 October 2013. Downloaded by St. Petersburg State University on 14/01/2014 10:21:52.
Dalton Transactions
Fig. 10 Time-resolved emission spectra (TRES) of Ir(III) complexes alone and the TTA upconversion. With DPA as the triplet acceptor. (a) Ir-1 alone (τP = 79.9 μs). (b) TRES of Ir-1 in the presence of DPA. Upconverted emission in the range of 380 nm–750 nm was observed (τDF = 282.6 μs). (c) Ir-2 alone (τP = 30.7 μs). (d) TRES of Ir-2 in the presence of DPA. Upconverted emission in the range of 380 nm–750 nm was observed (τDF = 186.5 μs). Note for Ir-1–Ir-2 the phosphorescence wavelengths are much longer than the upconverted emission wavelength. c [DPA] = 6.0 × 10−5 M; c [sensitizers] = 1.0 × 10−5 M. In deaerated CH2Cl2. Excited with nanosecond pulsed OPO laser synchronized with spectro-fluorometer (λex = 473 nm). 20 °C.
Paper
strong absorption of visible light and a long-lived triplet excited state. Three different coumarin chromophores were used to prepare the Ir(III) complexes, i.e. coumarin 6 (Ir-1), 7-diethylamino coumarin (Ir-2) and 3-phenyl-7-diethylamino coumarin (Ir-3). Ir( ppy)2(bpy)[PF6] was used as the reference complex (Ir-0). All the three new complexes show strong absorption of visible light and the long-lived triplet excited state. We found that the triplet excited state lifetime (65.9 μs) is longer than the complexes prepared with the direct metalation method (visible light-harvesting chromophore as the C donor of the C–Ir bond,
