Judicious Design of Cationic, Cyclometalated Ir(III) Complexes for

Jan 16, 2018 - Biography. Stefan Bernhard began his career in the food laboratories of Chocolat Tobler before going on to earn an undergraduate degree...
0 downloads 11 Views 6MB Size
Download PDF
Article Cite This: Acc. Chem. Res. 2018, 51, 352−364

pubs.acs.org/accounts

Judicious Design of Cationic, Cyclometalated Ir(III) Complexes for Photochemical Energy Conversion and Optoelectronics Isaac N. Mills,† Jonathan A. Porras,† and Stefan Bernhard* Department of Chemistry, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213, United States CONSPECTUS: The exponential growth in published studies on phosphorescent metal complexes has been triggered by their utilization in optoelectronics, solar energy conversion, and biological labeling applications. Very recent breakthroughs in organic photoredox transformations have further increased the research efforts dedicated to discerning the inner workings and structure−property relationships of these chromophores. Initially, the principal focus was on the Ru(II)-tris-diimine complex family. However, the limited photostability and lack of luminescence tunability discovered in these complexes prompted a broadening of the research to include 5d transition metal ions. The resulting increase in ligand field splitting prevents the population of antibonding eg* orbitals and widens the energy range available for color tuning. Particular attention was given to Ir(III), and its cyclometalated, cationic complexes have now replaced Ru(II) in the vast majority of applications. At the start, this Account documents the initial efforts dedicated to the color tuning of these complexes for their application in light emitting electrochemical cells, an easy to fabricate single-layer organic light emitting device (OLED). Systematic modifications of the ligand sphere of [Ir(ppy)2bpy]+ (ppy: 2-phenylpyridine, bpy: 2,2′-bipyridine) with electron withdrawing and donating substituents allowed access to complexes with luminescence emission maxima throughout the visible spectrum exhibiting room temperature excited state lifetimes ranging from nanoseconds to dozens of microseconds and quantum yields up to 15 times that of [Ru(bpy)3]2+. The diverse photophysical properties were also beneficial when using these Ir(III) complexes for driving solar fuel-producing reactions. For instance, photocatalytic water-reduction systems were explored to gain access to efficient water splitting systems. For this purpose, a variety of water reduction catalysts were paired with libraries of Ir(III) photosensitizers in high-throughput photoreactors. This parallelized approach allowed exploration of the interplay between the diverse photophysical properties of the Ir compounds and the electron-accepting catalysts. Further work enhanced and simplified the critical electron transfer processes between these two species through the use of bridging functional groups installed on the photosensitizer. Later, a novel approach summarized in this Account explores the possibility of using Zn metal as a solar fuel. Structure−activity relationships of the light-driven reduction of Zn2+ to Zn metal are described. DFT calculations along with cyclic voltammetry were utilized to gain clear insights into the complexes’ electronic structures responsible for the effective photochemical properties observed in these dyes. While [Ir(ppy)2bpy]+ and its derivatives were found to be much more photostable than the Ru(II)-tris-diimine complex family, mass spectrometry indicated that the bpy ligand still photodissociated under extensive illumination. An interesting new approach involved the substitution of the bidentate 2,2′-bipyridine with a stronger chelating terpyridine ligand. This approach leaves room for one 2-phenylpyridine ligand and a third, anionic ligand, either Cl− or CN−. This Account reviews the effect of structural modifications on the photophysical properties of these [Ir(tpy)(ppy)X]+ complexes and corroborates the findings with the results obtained through DFT modeling. These complexes found application in photocatalytic CO2 reductions as well as a solvent tolerant light-absorber for the photogeneration of hydrogen. It was also documented that the robustness of these dyes in photoredox processes supersedes those of the commercially available [Ir(ppy)2(dtbbpy)]PF6 and [Ir(dF(CF3)ppy)2(dtbbpy)]PF6 complexes pioneered in the Bernhard laboratory.

■

INTRODUCTION The use of photoactive cationic transition metal complexes in electronics, photocatalysis, and sensory applications has long been a topic of great interest. Exemplary complexes, such as [Ru(bpy)3]2+, garnered great interest due to its modest photoreducing and strong photooxidizing ability, as well as its oxygen-sensitive and long-lived triplet excited state. Use of this complex in industrially relevant applications, such as in the © 2018 American Chemical Society

fabrication of OLEDs or as a photocatalyst/photosensitizer, was unfortunately hampered by the lack of electronic and color tunability. This lack of tunability arises from the complex’s lowlying 3MC state; the complex’s ability to thermally depopulate from the 3MLCT excited state through this metal-centered Received: July 27, 2017 Published: January 16, 2018 352

DOI: 10.1021/acs.accounts.7b00375 Acc. Chem. Res. 2018, 51, 352−364

Article

Accounts of Chemical Research

oxidizing and reducing allowing the complex to be used in a variety of light-driven redox processes.

state can result in fast nonradiative deactivation, and can lead to ligand dissociation. Nevertheless, the complex was extensively researched and this deactivation pathway was mitigated by switching to a 5d metal such as Ir(bpy)33+ and its derivatives. These Ir(III) complexes have also been shown to be potent photooxidants, though they still lack color tunability; they are also very difficult to synthesize as the temperatures required often result in a mix of cyclometalated and noncyclometalated isomers owing to ring rotation of the 2,2′-bipyridine bond. A major advancement in this field was achieved through deliberate cylcometalation by replacing the electron-poor bipyridine ligand with phenylpyridine ligands to synthesize the dichloro-bridged [Ir(ppy)2Cl]2 dimer (Figure 1).1,2 Color

■

TUNING OF BIS-CYCLOMETALATED [Ir(C∧N)2(L∧L)]+ COMPLEXES AND USE IN OLEDs One of the first groups to successfully realize the applications of this family of luminophores was Lo et al., who first used the complexes as luminescent probes for biological applications. In this work, biotin was attached to the diimine ligand (Figure 3),

Figure 3. Bis-cyclometalated Ir3+ complex with pendant biotin moiety synthesized by Lo et al.

which was not found to disturb the luminescence properties of the complexes.6 Variation of the cyclometalating ligands, however, produced a modest variation in the emission maxima of the complexes, with complexes emitting between 554−587 nm in degassed acetonitrile at room temperature. Around the same time as the work of Lo et al. on iridiumbased luminescent dyes, our group was investigating the use of cationic bis-cyclometalated iridium complexes in organic lightemitting devices (OLEDs); similar work by Thompson et al. had already shown effective tuning of the luminescent properties of neutral [Ir(ppy)2(acac)] complexes.7 In work by Slinker et al., a vastly improved variant of the parent [Ir(ppy)2(bpy)]+ complex was made by substituting the 4 and 4′ positions of the bipyridine ligand with tert-butyl groups (Figure 4). These groups led to a 3-fold increase in the

Figure 1. Cyclometalated dimer and cationic mixed-ligand complex.

and redox tunability was subsequently observed with its conversion to the mixed-ligand [Ir(ppy)2(bpy)]PF6 complex (Figure 1) and derivatives.3 Analysis of the excited states of the mixed-ligand complex indicated that the dominant emission is the result of a mixed 3 ILCT- 3MLCT transition, with partitioning of the HSOMO on the bipyridine and LSOMO on the phenylpyridine ligands (Figure 2).4,5 The result is a complex with substantial metal character in the excited states, strong spin−orbit coupling, and readily partitioned and therefore synthetically tunable photoexcited states; additionally, the triplet state creates a complex which is both powerfully

Figure 4. Slinker complex with absorption (blue) and emission spectra (green).

luminescence quantum yield along with an approximate doubling of the excited state lifetime. The complex was found to be an efficient luminophore in a single layer OLED with a light output of 10 Lm/W as well as being a bright emitter at 300 cd/m2, both at 3 V.8 Compared to the neutral complexes pioneered by Thompson et al., the new Slinker complex was attractive as it allowed for better charge separation in the emissive state as well as increased tunability of the emission wavelength.

Figure 2. HSOMO and LSOMO of Ru(bpy)32+ (left) and [Ir(ppy)2(bpy)]+ (right) showing pure 3MLCT character in the [Ru(bpy)3]2+ orbitals and mixed 3ILCT−3MLCT character in the [Ir(ppy)2(bpy)]+ orbitals. 353

DOI: 10.1021/acs.accounts.7b00375 Acc. Chem. Res. 2018, 51, 352−364

Article

Accounts of Chemical Research

Figure 5. Ligands evaluated in the combinatorial study by Lowry et al. in the Bernhard group.

Figure 6. Emission energy (A), luminescence quantum yield (B), and log[excited state lifetime] in nanoseconds (C) for the complexes [Ir(C∧N)2(L∧L)]+. For all graphs, (C∧N) is enumerated as (1) ppy, (2) dtbppy, (3) bhq, (4) thpy, (5) Fmppy, (6) Clmppy, (7) Brmppy, (8) MeOmppy, (9) Phmppy, and (10) Hmppy; (L∧L) is indicated as (a) dppz, (b) dppe, (c) Me4-phen, (d) 5-MePhen, (e) 4-MePhen, (f) phen, (g) 5,5′-dmbpy, (h) 4,4′-dmbpy, (i) 4,4′-dtbbpy, and (j) bpy. Adapted with permission from ref 9. Copyright 2004 American Chemical Society.

Table 1. Electrochemical Data for Several [Ir(C∧N)2(N∧N)]+ Complexes photosensitizer +

[Ir(ppy)2(bpy)] [Ir(ppy)2(phen)]+ [Ir(ppy)2(dphphen)]+ [Ir(Fmppy)2(bpy)]+ [Ir(Fmppy)2(phen)]+ [Ir(Fmppy)2(dphphen)]+

E0′, Mn+/M(n+1)+b (V vs SCE)

ΔEp (mV)

E0′, L/L− (V vs SCE)

ΔEp (mV)

emission λmax(nm)

+1.25 +1.24 +1.23 +1.38 +1.36 +1.36

65 65 75 75 60 75

−1.42 −1.42 −1.38 −1.39 −1.39 −1.35

70 80 70 60 80 70

585 579 587 558 550 556

complexes, the dominant emissive pathway is from the T1 → S0 transition.10 As part of this study, two extremely electronrich complexes, [Ir(ppy) 2 (dmabpy)](PF 6 ) 5 and [Ir(dFppy)2(dmabpy)](PF6) 6, were synthesized by replacing the di-tert-butyl groups of the Slinker complex with dimethylamino groups; though the produced complexes possessed extraordinary photophysical properties, they still exhibited the T1 → S0 transition as the dominant emissive pathway.

Following this, Lowry et al. in the Bernhard group synthesized 76 new bis-cyclometalated Ir(III) complexes in a combinatorial effort to probe the limits of the motif’s tunability. Here 100 complexes were synthesized, including 24 previously studied controls, by stepwise synthesis of the dimer with one of 10 cyclometalating ligands followed by dimer cleavage with one of the 10 neutral, bidentate ligands (Figure 5). The controls were in excellent agreement with previous results, while the new complexes proved these complexes’ emission maxima, excited state lifetime, and quantum yield can be dramatically affected by variation of the ligands employed (Figure 6).9 Following the success of the Slinker complex, computational insights into the mechanism of luminescence were sought to inform future OLED luminophore design. These studies provided additional evidence that for [Ir(ppy) 2 (bpy)] +

Electrochemical Tuning

As a consequence of the partitioning of the frontier orbitals to separate ligands in these complexes, the oxidation and reduction potentials may be directly targeted via ligand modification. The basic strategy follows that with the HOMO 354

DOI: 10.1021/acs.accounts.7b00375 Acc. Chem. Res. 2018, 51, 352−364

Article

Accounts of Chemical Research (highest occupied molecular orbital) and LSOMO (lowest singly occupied molecular orbital) residing on the cyclometalating ligand, they may be stabilized by the inclusion of electron-withdrawing groups there; this would increase the oxidation potential of the complex. The LUMO (lowest unoccupied molecular orbital) and HSOMO (highest singly occupied molecular orbital) may be targeted in a similar manner by incorporation of electron-donating groups to the diimine ligand; this would be expected to produce a destabilizing effect on those orbitals and hence produce a complex with a more negative first reduction potential. In practice, modification of the cyclometalating ligand is the preferred strategy to produce dramatic tuning effects. Indeed, substitution of the diimine ligand with bipyridine, phenanthroline, or modified phenanthrolines produces only a small effect on the reduction potential of these complexes (Table 1).11

■

Figure 7. Maximum PS TON achieved using [Ir(ppy)2(dtbbpy)]+ in various solvents both neat and with 10% v/v MeCN added. Adapted with permission from ref 12. Copyright 2013 American Chemical Society.

USE OF [Ir(C∧N)2(N∧N)]+ COMPLEXES FOR SOLAR FUEL GENERATION

Development of Water Reduction Photocatalysis Systems

Photocatalytic Generation of Zinc Metal as a Solar Fuel

Based on our successes tuning and utilizing bis-cyclometalated Ir(III) complexes as luminophores for OLEDs, work also began on using these complexes to drive the photogeneration of hydrogen from water. Previous systems realized photocatalytic water reduction with [Ru(bpy)3]2+ derivatives as a photosensitizer and [Co(bpy)3]2+ as the water reduction catalyst.13 This system was adapted using promising candidates from the combinatorial screening in place of [Ru(bpy)3]2+. Despite the lower molar extinction coefficients of the iridium complexes compared to the ruthenium complexes, the iridium-based photosensitizers achieved much higher turnovers, up to a maximum of 920 for [Ir(Fmppy)2(bpy)]+ compared to 580 for an optimized [Ru(N∧N)3]2+ system.11 Compared to Ru(bpy)32+ and [Ru(dmphen)3]2+, these complexes are 4−20 times more efficient when accounting for the difference in molar absorptivity. Efforts to further improve upon the above photocatalytic water reduction system were realized by our group in 2007 with the replacement of [Co(bpy)3]2+ by K2PtCl4 as a catalyst. This catalyst allowed photocatalytic hydrogen generation via reductive quenching of the excited iridium complex by a tertiary amine without the use of an electron relay (Figure 8),

As a more easily handled, transported, and stored fuel than hydrogen gas, the production of reducing metals has attracted considerable interest in the field of renewable energy storage. Unfortunately, most of the commonly studied metals (Li, Na, Mg, etc.) have reduction potentials that place them out of the reach of many visible-light driven photocatalytic processes. Zinc metal, however, has a more modest reduction potential (−0.76 V vs SCE), is air-stable, and is a common component of current battery technologies. Our group has developed a system composed of a cationic Ir(III) photocatalyst, sacrificial reductant, and Zn2+ ions as an acceptor. This three-component system functions analogously to the water reduction photolysis systems producing zinc metal for use as a solar fuel. A representative mixture of the bestperforming photosensitizers and OLED materials were assessed as zinc photoreduction catalysts (Table 2).12 The effect of Table 2. Maximum Ir Turnovers (TON) for Various Photocatalysts with ZnCl2 in MeCN Using Triethylamine (TEA) as a Sacrificial Donor photocatalyst

TON

[Ir(MeO-mppy)2(bpy)]+ [Ir(ppy)2(bpy)]+ [Ir(ppy)2(dppe)]+ [Ir(dF-mppy)2(dtbbpy)]+ [Ir(Ph-mppy)2(Me4phen)]+ [Ir(ppy)2(phen)]+ [Ir(ppy)2(Me4phen)]+ [Ir(F-mppy)2(Me4phen)]+ [Ir(F-mppy)2(dtbbpy)]+ [Ir(ppy)2(dtbbpy)]+

230 275 5 180 125 195 205 160 250 290

Figure 8. Reductive quenching mechanism observed in systems without an electron relay using Pt colloids formed in situ from K2PtCl4. Here the quenching of the excited Ir complex is depicted using TEA, though other tertiary amines function analogously.

different anions on Zn2+ reduction was also evaluated, with zinc halides, especially zinc chloride, being the most efficient. Finally, a solvent study was performed which indicated that acetonitrile is the best solvent for zinc reduction photocatalysis; even mixed solvent systems featuring 10% added acetonitrile typically saw an increase in maximum turnover numbers (Figure 7). A fully optimized system reducing zinc chloride in pure acetonitrile achieved over 400 Ir turnovers after 10 days of illumination using triethylamine as a sacrificial reductant.

simplifying future studies on efficient photosensitizer and catalyst designs. Initially, this system was only capable of 63 PS turnovers, but a photon-to-hydrogen conversion efficiency of 26% was achieved.5 Subsequent studies of the postillumination reaction mixture confirmed that loss of the bipyridine ligand was responsible for photosensitizer degradation and subsequent system failure. 355

DOI: 10.1021/acs.accounts.7b00375 Acc. Chem. Res. 2018, 51, 352−364

Article

Accounts of Chemical Research

sensitizer design. In 2009, our group studied the effect of varying the cyclometalating ligands and the bipyridine ligands on the photosensitizer on the overall evolution of H2 from water photolysis. From the results below (Figure 11), it is

Again in 2008, our group made advances in the photogeneration of hydrogen with the development of a fully homogeneous system for photoinduced water reduction. Modifications were again made to the catalyst, this time featuring four new homogeneous catalysts [Rh(bpy)3]3+ 7, [Rh(dtbbpy)3]3+ 8, [Rh(dFbpy)3]3+ 9, and [Rh(dMeObpy)3]3+ 10 which probed the effects of electronic tuning and sterics on catalyst activity.14 Replacing K2PtCl4, which is known to form colloids in situ, with a truly molecular catalyst allowed a maximum of 5000 PS turnovers using [Ir(Fmppy)2(dtbbpy)](PF6) and [Rh(dtbbpy)3]3+ 8. As before, triethylamine proved to be an effective sacrificial reductant, and solvents with weaker coordinating ability proved more effective (Figure 9).

Figure 11. Effects of varying the cyclometalating and chelating ligands on photosensitizer performance in a water reduction system using colloidal platinum as a catalyst. Adapted with permission from ref 16. Copyright 2009 American Chemical Society. Figure 9. Effects of different sacrificial donors, solvents on water photoreduction catalysis with Rh(dtbbpy)3(PF6)3 as catalyst and [Ir(Fmppy)2(dtbbpy)](PF6) as photosensitizer. Adapted with permission from ref 14. Copyright 2008 American Chemical Society.

unclear how modification of the cyclometalating ligand can improve performance; placement of electron-rich sterically blocking groups in the 4- and 4′- or 5- and 5′-positions of the bipyridine, however, drastically improves performance of the complexes.16 While this is counter to the strategy seen before in the OLED tuning efforts, it can be rationalized in terms of preventing dissociation of the bipyridine ligand; groups which make the coordinating nitrogens more basic, and hence more strongly coordinating, or groups which block the coordination of other ligands to the metal center during bipyridine dissociation will improve sensitizer longevity. This is most realized in the 5,5′-diisopropyl bipyridine ligand, which achieved the highest number of turnovers. Variation of the solvent was also found to play a strong role in the performance of the system. While keeping the catalyst the same, three selected photosensitizers demonstrated a notable difference in performance across mixtures of water with tetrahydrofuran, N,N-dimethylformamide, and acetonitrile. Based on the turnover numbers achieved, it is clear that increased ligating power of the solvent favors degradation of the photoactive species in solution (Figure 12). Since previous studies had indicated our photosensitizers can work with Pt or Pd colloids and with either triethylamine or triethanolamine, an

In 2013, we again developed a successful photocatalytic water reduction system using a new catalyst featuring a nickel thiolate hexamer (Figure 10) in lieu of K2PtCl4 or [Rh(bpy)3]3+

Figure 10. New nickel thiolate hexamer (Left) used as a water reducing catalyst with [Ir(Fmppy)2(dtbbpy)](PF6) (Right) as a photosensitizer. Adapted with permission from ref 15. Copyright 2013 American Chemical Society.

derivatives. As before, [Ir(Fmppy)2(dtbbpy)](PF6) was employed as the photosensitizer, triethylamine was the sacrificial reductant, and a THF/H2O mixture was the solvent. The new catalyst was shown, like [Rh(dtbbpy)3]3+, to proceed via a reductive quenching mechanism.15 Mercury poisoning tests confirmed the active species are molecular. At optimal concentrations of both the catalyst and photosensitizer, a maximum of over 3700 PS turnovers at a turnover frequency of 970 h−1 were observed; lower concentrations of the photosensitizer yielded up to 30,000 catalyst turnovers. While the [Ir(ppy)2(bpy)]+ photosensitizers have been demonstrated to work with a variety of water reduction catalysts and under diverse conditions, it is difficult to discern clear patterns from separate studies that would inform future

Figure 12. Effect of solvent on selected photosensitizers using colloidal palladium water reducing catalysts. Adapted with permission from ref 16. Copyright 2009 American Chemical Society. 356

DOI: 10.1021/acs.accounts.7b00375 Acc. Chem. Res. 2018, 51, 352−364

Article

Accounts of Chemical Research

Figure 13. [Ir(ppy)2(bpy)](PF6) derivatives with pendant pyridyl moieties.

additional sequence was run testing all of the reductant/catalyst combinations; triethylamine with Pt colloids was shown to have the greatest turnover frequency while triethanolamine with Pd colloids was shown to be a longer-lived pair. Tailoring Photosensitizers to Specific Water Reduction Catalysts

Though [Ir(ppy)2(bpy)](PF6) derivatives possess incredible versatility regarding working solvent, catalyst, and donor conditions, they all suffer the same eventual fate: displacement of the bipyridine upon photoexcitation leads to loss of activity. To counter this, we sought to design photosensitizers which could rapidly transfer photoexcited electrons to the catalyst. This was done primarily in two ways: first by tethering of the photosensitizer to a catalyst and second by formation of microemulsions which promoted PS-catalyst aggregation. In the first work on this strategy, our group synthesized iridium complexes with pendent pyridyl moieties (Figure 13). These moieties were placed at the 4- and 4′-positions of the bipyridine in lieu of the traditional tert-butyl moieties. The pyridyl moieties were intended to coordinate to colloidal platinum catalysts formed in situ, resulting in rapid transfer of electrons from the π* orbitals of the bipyridine to the tethered colloid. Noncoordinating methyl or phenyl groups on the bipyridine served as a control. Of the photosensitizer groups evaluated, those containing qpy were consistently strong performers, typically achieving 7000−8000 PS turnovers; similar results were obtained using dethqpy.17 Comparators with noncoordinating methyl and phenyl groups evolved much less hydrogen, with maximum turnovers reaching around 2000 (Figure 14). Another approach using tethered photosensitizers was again made by our group using nitriles in place of the pyridines. Two new complexes based on the most successful of the traditional photosensitizers, [Ir(ppy) 2 (dtbbpy)](PF 6 ) and [Ir(Fmppy)2(dtbbpy)](PF6) were synthesized replacing traditional tert-butyl groups with nitriles (Figure 15). These replacements served as an extension of the strategies pioneered by DiSalle and Metz, who demonstrated the use of vinyl groups on the bipyridine ligand to stabilize platinum colloids.18 These complexes were evaluated against their tert-butyl-bearing counterparts in water reduction photolysis using both colloidal

Figure 14. PS turnovers for complexes with and without pendant coordinating pyridyls. Adapted with permission from ref 17. Copyright 2011 American Chemical Society.

Figure 15. Nitrile-bearing photosensitizers.

catalysts and the molecular Rh(dtbbpy)3(PF6)3 catalyst. Both complexes performed well with colloidal catalysts, but did not outperform the previous complexes using the molecular catalyst (Figure 16). The new catalysts did possess unusual photophysical properties, namely, a direct visible light populated 3 MLCT.19 Due to the interesting photophysical properties of the dCNbpy complexes, a follow-up study was done replacing the fluorine of the traditional Fmppy complexes with nitriles (Figure 17). The new complexes were markedly better than their predecessors in terms of photophysical properties with [Ir(CNmppy)2(dtbbpy)]+ possessing an excited state lifetime of almost 8 μs and a luminescence quantum yield of 63%.20 These unusual properties can be explained by the alteration of 357

DOI: 10.1021/acs.accounts.7b00375 Acc. Chem. Res. 2018, 51, 352−364

Article

Accounts of Chemical Research

Figure 18. Photosensitizer (left) and water reduction catalyst (right) combination employed in toluene/water/1-octylamine microemulsion study. Figure 16. Comparison of nitrile-bearing photosensitizers vs previous complexes with both molecular and colloidal catalysts. Adapted with permission from ref 19. Copyright 2014 Elsevier Ltd.

Under optimal conditions, the system produced 240 PS turnovers and 400 catalyst turnovers; the system was still able to produce hydrogen from oxalic acid after 95 h of illumination.21

the HSOMO−LSOMO transition from the typical mixed-3MLCT, still present in the [Ir(CNmppy)2(dCNbpy)]+, to one with predominantly 3ILCT character, as with [Ir(CNmppy)2(dtbbpy)]+ (Figure 17). The complexes were employed as photosensitizers for water reduction photocatalysis using platinum colloids as the water reduction catalyst and triethylamine as the sacrificial donor; a maximum of 2000 and 1400 PS turnovers were obtained under ideal conditions. Unusually, the complexes demonstrated remarkable solvent tolerance, being able to function in alcohols, acetonitrile, acetone, and tetrahydrofuran. The second method employed to facilitate rapid transfer of electrons from photosensitizer to the catalyst is through the use of microemulsions. Our group developed a new photosensitizer/water reduction catalyst system (Figure 18) with oxalic acid as a sacrificial reductant. The system relied on the formation of an emulsion of toluene and water, aided by an in situ generated ion pair between 1-octylamine and the oxalic acid sacrificial donor. This system, unlike typical water reduction photolysis systems, is driven by an oxidative quenching mechanism. This new system is also advantageous in that it takes place in a biphasic system; integrated water splitting systems would require some method of separating the oxidation and reduction half-reactions and a biphasic system would be a great first step toward linking these two processes.

Current State of Bis-Cyclometalated [Ir(C∧N)2(N∧N)]+ Photosensitizers

Since the development of bis-cyclometalated Ir(III) complexes for OLEDs and solar fuel applications, they have found application in the recently emerging field of photoredox catalysis. Chief among cited Ir(III) photoredox catalysts is [Ir(dFCF3ppy)2(dtbbpy)](PF6), which was developed in our group and has seen extensive use by MacMillan and others.22−25 Though [Ir(dFCF3ppy)2(dtbbpy)](PF6) is quite robust, solvent stability, especially in acetonitrile, remains an area for further improvement. This lower solvent stability often requires loading as much as 1 mol % photocatalyst and extended reaction times under typical photoredox conditions. Additionally, while the electronic properties of highly fluorinated Ir(III) complexes remain attractive, their molar absorptivity in the visible spectrum decreases with increasing fluorination, lowering their utility in such applications.

Figure 17. HSOMO−LSOMO transitions for [Ir(CNmppy)2(dtbbpy)]+ 14 (left) and [Ir(CNmppy)2(dCNbpy)]+ 15 (right) showing 3ILCT character vs mixed 3ILCT-3MLCT character for their respective transitions. Adapted with permission from ref 20. Copyright 2016 American Chemical Society. 358

DOI: 10.1021/acs.accounts.7b00375 Acc. Chem. Res. 2018, 51, 352−364

Article

Accounts of Chemical Research

Figure 19. Mechanism of photosensitizer degradation following excitation and subsequent dissociation and replacement of bipyridine ligand. The effect is pronounced in ligating solvents such as acetonitrile.

■

CYCLOMETALATED CATIONIC Ir(III) COMPLEXES WITH TRIDENTATE PYRIDYL LIGANDS

Complex Architectures

The versatility of [Ir(ppy)2(bpy)]+ based complexes as photosensitizers stems from the ability to easily tune and functionalize the ligand architecture. Major improvements in the TON of the photosensitizers for water reduction and their electroluminescent efficiency in OLEDs have originated from functionalization of the diimine ligand which holds the majority of the electron density of the LUMO. However, upon photoexcitation, the antibonding π* orbital of the diimine ligand is populated, resulting in ligand dissociation from the reduced Ir(III) photosensitizer. This effect is even amplified upon reductive quenching and this loss of the diimine ligand renders the complex photochemically inactive. This degradation is especially prominent in coordinating solvent environments, like water or acetonitrile, that facilitate the dissociation of the weakened diimine ligand, as shown in Figure 19. As a result, circumventing this degradation pathway is an important area of research. To improve the photostability of the complexes, one strategy would employ switching to a bis-terdentate ligand system instead of a tris-bidentate ligand arrangement. Our group synthesized a bis-tridentate analogue to 2 utilizing 6-phenyl2,2′-bipyridine, to generate complex 18, [Ir(phbpy)2]+.26 When water reduction experiments were conducted in acetonitrile, 18 exhibits an almost 3-fold improvement in TON compared to 2 (Figure 20). While the complexes had identical maximum rates

Figure 21. Cationic, bis-terdentate Ir(III) complexes 19−21.

motif was synthetically tunable, but the emission maxima were consistently orange-red with short excited-state lifetimes relative to other cationic Ir(III) photosensitizers. Work by Scandola et al. synthesized complexes with the structure 21, which had an [Ir(C∧N∧C)(N∧N∧N)]+ configuration.29 Unfortunately, while it was possible to install several functional groups with low synthetic yields, dramatic tuning of the photophysical or electrochemical properties similar to the [Ir(ppy)2(bpy)]+ complex family was not observed. Given the synthetic challenges and overall lack of tunability of the ligands, new photosensitizer structures were worth exploring as replacements for cationic Ir(III) bis-terdentate complexes. Photocatalytic CO2 Reduction. A family of cationic Ir(III) complexes featuring a new photosensitizer architecture, [Ir(tpy)(ppy)Cl]+ (22−24) became promising catalysts for the photoreduction of CO2.30 These complexes were unusual in that they were standalone catalysts, and did not require the use of separate photosensitizer and cocatalyst. The overall structure still contained one cyclometalating ligand, but the labile diimine ligand was replaced with a more substitution-inert triimine ligand. The complex structure also had a third area of tuning in the anionic ligand that could be readily exchanged. Ishitani et al. synthesized this series of complexes (Figure 22), changing the functionality of the phenylpyridine ligand while leaving the terpyridine ligand untouched. Photocatalytically, the methyl derivative 23 was found to perform the best, and the complexes were able to function in coordinating solvents including acetonitrile. Changes in the phenylpyridine ligand also tuned the photophysical properties. Simultaneously, the electrochemical oxidation was dramatically influenced by substitutions on the phenylpyridine, with a +1.66 V potential for the methyl

Figure 20. Photocatalytic H2 evolution (solid lines, right axis) and rate of H2 evolution (dotted lines, left axis) from 18 (red) and 2 (black) in 4:1 acetonitrile/water. Adapted with permission from ref 26. Copyright 2009 American Chemical Society.

of hydrogen generation, the decay of 18 was significantly delayed, owing to the enhanced resilience from the better chelated ligand structure. While not used specifically for water reduction, several other cyclometalated, cationic, bis-tridentate Ir(III) photosensitizers have been developed (Figure 21). The earliest example was published by Campagna et al.,27 who synthesized the [Ir(C∧N∧N)(C∧N∧N)]2+ complex 19. Similarly, Williams et al. synthesized several complexes with the [Ir(N∧C∧N)(N∧N∧C)]+architecture (20).28 The electrochemistry of this 359

DOI: 10.1021/acs.accounts.7b00375 Acc. Chem. Res. 2018, 51, 352−364

Article

Accounts of Chemical Research

maxima, whereas the anthracene substituted 27, had a noticeable drop in its quantum yield of emission. Ultimately, while products of catalyst deactivation were still observed, addition of an anthracene to the complex resulted in a significant improvement in the TON’s (Figure 23) within the family of complexes. Further optimization including control of the intensity of the light source resulted in TONs of 265 over 10 days which is at this time, a record performance for a cationic Ir(III)-based standalone photocatalyst for CO 2 reduction, without the use of a cocatalyst. Tunability of [Ir(tpy)(ppy)X]+ Photosenstizers. In order to develop strong structure−activity relationships, our group designed an even more structurally diverse series of [Ir(tpy)(ppy)X]+ based complexes where the terpyridine, phenylpyridine, and the anionic ancillary ligand was modified.32 The complexes containing a chloride as the ancillary ligand were easily synthesized in two steps starting from the terpyridine ligand and IrCl3 in high yields, without being dramatically affected by different functional groups on the terpyridine (Scheme 1).33,34 Cyclometalation with the corresponding phenylpyridine was also equally tolerant of different functional groups. Exchange of the chloride to a stronger field cyanide ligand was also achieved in good yields, allowing a third parameter of tunability for the complexes. DFT calculations demonstrated that tuning of the HOMO level is achieved through the phenyl ring of the ppy as well as the monodentate ancillary ligand, whereas the LUMO level is impacted by the terpyridine, although not dramatically given that the aryl substituents do not readily participate in the LUMO. Electrochemically, the oxidation potentials are influenced by the phenyl ring from the ppy as well as the ancillary ligand, corroborating results from DFT calculations (Figure 24). The first reduction shows little variation with the different aryl groups on the terpyridine. A second, irreversible reduction arises from the cyclometalating ring of the ppy in the chloride complexes. Exchange of the chloride from 22 to a cyanide in 28 not only makes the second reduction potential reversible, but also eliminates the third reduction observed in the chloride complexes, ascribed to loss of the chloride ligand. Photophysically the complexes are easily tuned, whereby decreased electron density of the ppy or a cyanide ligand results in a

Figure 22. [Ir(tpy)(ppy)Cl]+ based photocatalysts 22−24 investigated by Ishitani et al. for photocatalytic CO2 reduction.30 Turnover numbers for CO (TONCO) after 300 min are provided for each complex.

derivative 23, +1.73 V for the parent 22, and +1.88 V for the trifluoromethyl derivative 24. It is worth noting that the first reduction remained constant despite structural changes in the phenylpyridine. This electronic disconnection suggested that the complexes had independently tunable HOMO and LUMO levels that could be synthetically adjusted for specific catalytic applications. Our group decided to take advantage of the tunability of these complexes to synthesize new, more efficient standalone photocatalysts for CO2 reduction, water reduction, and fabrication of OLEDs.31 Work by Ishitani et al.30 postulated the low TONs for the complexes for CO2 reduction were from a dimerization of the one-electron reduced species at the monodentate ligand site. To limit catalyst deactivation, we decided to modify the LUMO of the complexes by addition of aromatic groups at the 4′-position of the terpyridine (25−27), with the idea being the steric bulk of the aromatic groups should create obstructing barriers for deactivation, but also tune the LUMO energy level.31 Interestingly, modulating the size and electron density of the aromatic group did not alter the LUMO energy, based on electrochemical studies, whereby the first reduction potential of the complexes varied only by 0.02 V throughout the series. DFT calculations showed that the pendant aromatic groups are not coplanar with the central pyridine ring, resulting in little to no contribution to the structure of the LUMO (Figure 23). Photophysically the family of complexes exhibit comparable emission lifetimes at similar

Figure 23. TONCO of 22 and 25−27. Singlet, ground state frontier orbitals of complexes 22 and 27 on the left and right, respectively. Frontier orbitals obtained with Gaussian ’09 (B3LYP/LANL2DZ). Orbitals adapted with permission from ref 31 and 32. Copyright 2017 American Chemical Society and 2014 American Chemical Society. 360

DOI: 10.1021/acs.accounts.7b00375 Acc. Chem. Res. 2018, 51, 352−364

Article

Accounts of Chemical Research Scheme 1. General Synthetic Pathways of [Ir(tpy)(ppy)X]+ Complexes

lead our group to develop additional complexes that take this “push-pull” phenomena to an extreme. We subsequently synthesized a series of complexes with an electron-rich terpyridine, 4,4′,4″-tri-tert-butyl- 2,2′:6′,2′′-terpyridine, phenylpyridines with varying degrees of fluorination, and chloride or cyanide ancillary ligands.35 Interestingly, during the synthesis, we discovered that C−F insertion was taking place when ortho-fluorinated ppy ligands were reacted with 30, a process that was not observed during the synthesis of [Ir(ppy)2(bpy)]+ based complexes (Scheme 2). Depending on the ligand precursor, complexes 31 and 32 were synthesized via C−H or C−F activation almost exclusively. This allowed for the synthesis of otherwise difficult or inaccessible complexes, including those with a perfluorinated phenylpyridine. While DFT calculations demonstrated the band gap widening based on this “push−pull” design, it also demonstrated that addition of fluorine atoms gradually reduced the metal character of the excited states, thereby causing the complexes to phosphoresce from an ILCT instead of the more efficient MLCT. Photophysically, the complexes had higher quantum yields and longer excited state lifetimes than previous [Ir(tpy)(ppy)X]+. When used for water reduction in acetonitrile, however, the complexes (TON 300−950) were unable to match the performance of the previously established “push− pull” complex, with the exception of complex 33, which performed better (TON 1250) due to the addition of the cyano ligand, and an appropriate number of fluorine atoms on the phenylpyridine to minimize loss of metal character (Figure 26). Given the stability of the complexes, we also evaluated two of the complexes, 32 and 34, as photosensitizers for photoredox catalysis, specifically the decarboxylative fluorination of carboxylic acids (Scheme 3). The complexes were able to outperform the well-established [Ir(dF(CF3)ppy)2(dtbbpy)]+ complex 35 as a photosensitizer (Table 3).25 Due to the enhanced stability and stronger absorbance of the [Ir(tpy)(ppy)X]+ structures in coordinating solvents, smaller loadings of the complexes (0.05 mol % vs 1.00 mol %) could be used without impacting reaction yields. This could not be said for the reactions using 35 as the photosensitizer.

Figure 24. Cyclic voltammograms of 22 and 28. Adapted with permission from ref 32. Copyright 2014 American Chemical Society.

blue shift of the emission, increase in quantum yield, and extension of the excited state lifetime. Complex 29 was the most photochemically efficient, with an electron-rich terpyridine and electron-poor ppy, generating a “push−pull” force within the complex (Figure 25). This directly ties to the performance of the complexes for water reduction in acetonitrile, whereby the cyano complexes (TON 1350−1400) perform better than the chloro complexes (TON 800−1275) due to the less labile nature of the cyano ligand, with the exception of the “push-pull” chloro complex (TON 1350). This

■

CONCLUSION The use of cationic Ir(III) based complexes has several key advantages that allow for an expansive array of uses. The high metal character in the excited states gives strong spin−orbit coupling, making the complexes more photophysically efficient compared to notable photosensitizers like [Ru(bpy)3]2+. Greater partitioning of the frontier orbitals within the molecule gives the [Ir(C∧N) 2(N∧N)]+ and [Ir(N∧N∧N)(C∧ N)X]+ complexes a synthetic tunability that allows for optimization of the electronic structure, tailored to the desired application. This versatility has allowed our group to investigate such complexes for solar fuel generation from abundant sources including Zn, CO2, and H2O, but also in emerging fields such as photoredox catalysis. [Ir(C∧N)2(N∧N)]+ complexes have been

Figure 25. First generation “push−pull” complex 29 with its corresponding photophysical properties.32 361

DOI: 10.1021/acs.accounts.7b00375 Acc. Chem. Res. 2018, 51, 352−364

Article

Accounts of Chemical Research Scheme 2. C−F and C−H Activation of Fluorinated Phenylpyridinesa

a

Adapted with permission from ref 35. Copyright 2016 American Chemical Society.

Table 3. Percent Yields of Decarboxylative Fluorination with Different Ir(III) Photosensitizersa

Figure 26. Second generation “push−pull” complex 33 with its corresponding photophysical properties.35

Scheme 3. Decarboxylative Fluorination of Carboxylic Acids with Different Ir(III) Photosensitizer Structures

a Yields were determined by 19F NMR Analysis. Adapted with permission from ref 35. Copyright 2016 American Chemical Society.

degradation into photochemically inert species in coordinating solvents. A new and promising solution is the use of [Ir(N∧N∧N)(C∧N)X]+ complexes. This new structure still features the same synthetic tunability of the frontier orbitals with greatly enhanced solvent tolerance. Meticulous molecular engineering of the complexes and further optimization of the reaction conditions is required to find an economically viable path to solar fuels, single layer OLEDs, and other important applications. The unprecedented robustness and attractive photophysical properties of cationic, cyclometalated Ir(III) complexes will make them a key player in this quest.

more heavily explored due to their ease of synthesis, readily and widely tunable properties, and prior success in other photochemical applications. While they are competent photosensitizers and catalysts, no clear trend between photophysical and electrochemical properties can be made with relation to their photosystem turnovers. This is largely due to their 362

DOI: 10.1021/acs.accounts.7b00375 Acc. Chem. Res. 2018, 51, 352−364

Article

Accounts of Chemical Research

■

phenylpyridinato-C2,N’)(2,2’-bipyridine)iridium(III). Inorg. Chem. 1993, 32, 3088−3092. (5) Tinker, L. L.; McDaniel, N. D.; Curtin, P. N.; Smith, C. K.; Ireland, M. J.; Bernhard, S. Visible light induced catalytic water reduction without an electron relay. Chem. - Eur. J. 2007, 13, 8726− 8732. (6) Lo, K. K.; Chan, J. S.; Lui, L.; Chung, C. Novel cyclometalated ir(III) diimine complexes that contain a biotin moiety. Organometallics 2004, 23, 3108−3116. (7) Lamansky, S.; Djurovich, P.; Murphy, D.; Abdel-Razzaq, F.; Lee, H. E.; Adachi, C.; Burrows, P. E.; Forrest, S. F.; Thompson, M. E. Highly phosphorescent bis-cyclometalated iridium complexes: synthesis, photophysical characterization, and use in organic light-emitting diodes. J. Am. Chem. Soc. 2001, 123, 4304−4312. (8) Slinker, J. D.; Gorodetsky, A. A.; Lowry, M. S.; Wang, J.; Parker, S.; Rohl, R.; Bernhard, S.; Malliaras, G. G. Efficient yellow electroluminescence from a single layer of a cyclometalated iridium complex. J. Am. Chem. Soc. 2004, 126, 2763−2767. (9) Lowry, M. S.; Hudson, W. R.; Pascal, R. A.; Bernhard, S. Accelerated luminophore discovery through combinatorial synthesis. J. Am. Chem. Soc. 2004, 126, 14129−14135. (10) Minaev, B.; Agren, H.; DeAngelis, F. Theoretical design of phosphorescence parameters for organic electro-luminescence devices based on iridium complexes. Chem. Phys. 2009, 358, 245−257. (11) Goldsmith, J. I.; Hudson, W. R.; Lowry, M. S.; Anderson, T. H.; Bernhard, S. Discovery and high-throughput screening of heteroleptic iridium complexes for photoinduced hydrogen production. J. Am. Chem. Soc. 2005, 127, 7502−7510. (12) Brooks, A. C.; Basore, K.; Bernhard, S. Photon-driven reduction of Zn2+ to Zn metal. Inorg. Chem. 2013, 52, 5794−5800. (13) Krishnan, C. V.; Brunschwig, B. S.; Creutz, C.; Sutin, N. Homogeneous catalysis of the photoreduction of water. 6. Mediation by polypyridine complexes of ruthenium(II) and cobalt(II) in alkaline media. J. Am. Chem. Soc. 1985, 107, 2005−2014. (14) Cline, E. D.; Adamson, S. E.; Bernhard, S. Homogeneous catalytic system for photoinduced hydrogen production utilizing iridium and rhodium complexes. Inorg. Chem. 2008, 47, 10378−10388. (15) Kagalwala, H. N.; Gottlieb, E.; Li, G.; Li, T.; Jin, R.; Bernhard, S. Photocatalytic hydrogen generation system using a nickel-thiolate hexameric cluster. Inorg. Chem. 2013, 52, 9094−9101. (16) Curtin, P. N.; Tinker, L. L.; Burgess, C. M.; Cline, E. D.; Bernhard, S. Structure-activity correlations among iridium(III) photosensitizers in a robust water-reducing system. Inorg. Chem. 2009, 48, 10498−10506. (17) Disalle, B. F.; Bernhard, S. Orchestrated photocatalytic water reduction using surface-adsorbing iridium photosensitizers. J. Am. Chem. Soc. 2011, 133, 11819−11821. (18) Metz, S.; Bernhard, S. Robust photocatalytic water reduction with cyclometalated Ir(III) 4-vinyl-2,2′-bipyridine complexes. Chem. Commun. 2010, 46, 7551−7553. (19) Mills, I. N.; Kagalwala, H. N.; Chirdon, D. N.; Brooks, A. C.; Bernhard, S. New Ir(III) 4,4’-dicyano-2,2’-bipyridine photosensitizers for solar fuel generation. Polyhedron 2014, 82, 104−108. (20) Mills, I. N.; Kagalwala, H. N.; Bernhard, S. New Ir(III) 4,4’dicyano-2,2’-bipyridine photosensitizers for solar fuel generation. Dalton Trans. 2016, 45, 10411−10419. (21) Kagalwala, H. N.; Chirdon, D. N.; Mills, I. N.; Budwal, N.; Bernhard, S. Light-driven hydrogen generation from microemulsions using metallosurfactant catalysts and oxalic acid. Inorg. Chem. 2017, 56, 10162−10171. (22) Lowry, M. S.; Goldsmith, J. I.; Slinker, J. D.; Pascal, R. A.; Malliaras, G. G.; Bernhard, S.; Rohl, R. Single-layer electroluminescent devices and photoinduced hydrogen production from an ionic iridium (III) complex. Chem. Mater. 2005, 17, 5712−5719. (23) Zuo, Z.; Ahneman, D. T.; Chu, L.; Terrett, J. A.; Doyle, A. G.; MacMillan, D. W. C. Merging photoredox with nickel catalysis: Coupling of α-carboxyl sp3-carbons with aryl halides. Science 2014, 345, 437−440.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Stefan Bernhard: 0000-0002-8033-1453 Author Contributions †

I.N.M. and J.A.P. contributed equally to this manuscript.

Notes

The authors declare no competing financial interest. Biographies Isaac N. Mills obtained his M.S. in Chemistry from Youngstown State University in 2010 and his Ph.D. from Carnegie Mellon University in 2017. He recently started as an Assistant Professor at Mount St. Mary’s University in Emmitsburg, MD where he teaches Introductory and Inorganic Chemistry. His current research interests center on the use of transition metal complexes in photoredox catalysis and the development of photoredox lab exercises and equipment to bring this exciting field to less-affluent areas. Jonathan A. Porras obtained his B.S. in Biochemistry from Marist College in 2012 and his Ph.D. from Carnegie Mellon University in 2017. He recently started as an Assistant Professor at Drew University in Madison, NJ, where he teaches introductory chemistry lectures and laboratories. His professional interests include underrepresented populations in STEM fields, specifically Hispanic/Chicano and LGBTQ+ populations and development of inorganic and photochemical undergraduate laboratory experiments. Stefan Bernhard began his career in the food laboratories of Chocolat Tobler before going on to earn an undergraduate degree from the School of Engineering in Burgdorf, Switzerland. Following undergraduate studies, Bernhard earned a Ph.D. in Chemistry from the University of Fribourg. As a postdoctoral researcher, he worked at Los Alamos National Laboratory on ultrafast vibrational spectroscopy and in the Abruña group at Cornell where he studied electrochemistry and optoelectronics of metal complexes. Bernhard launched his independent career at Princeton University and then moved to Carnegie Mellon University earning a promotion to full professor in 2014. The current research interests of Prof. Bernhard’s group center on luminescent metal complexes and their applications in solar energy conversion and optoelectronics.

■

ACKNOWLEDGMENTS The authors acknowledge support from the National Science Foundation through CHE-1362629.

■

REFERENCES

(1) Garces, F. O.; King, K. A.; Watts, R. J. Synthesis, structure, electrochemistry, and photophysics of methyl-substituted phenylpyridine ortho-metalated iridium(III) complexes. Inorg. Chem. 1988, 27, 3464−3471. (2) Sprouse, S.; King, K. A.; Spellane, P. J.; Watts, R. J. Photophysical effects of metal-carbon sigma bonds in ortho-metalated complexes of iridium(III) and rhodium(III). J. Am. Chem. Soc. 1984, 106, 6647− 6653. (3) Ohsawa, Y.; Sprouse, S.; King, K. A.; DeArmond, M. K.; Hanck, K. W.; Watts, R. J. Electrochemistry and spectroscopy of orthometalated complexes of ir(III) and rh(III). J. Phys. Chem. 1987, 91, 1047−1054. (4) Colombo, M. G.; Hauser, A.; Guedel, H. U. Evidence for strong mixing between the LC and MLCT excited states in bis(2363

DOI: 10.1021/acs.accounts.7b00375 Acc. Chem. Res. 2018, 51, 352−364

Article

Accounts of Chemical Research (24) Noble, A.; McCarver, S. J.; Macmillan, D. W. C. Merging photoredox and nickel catalysis: Decarboxylative cross-coupling of carboxylic acids with vinyl halides. J. Am. Chem. Soc. 2015, 137, 624− 627. (25) Ventre, S.; Petronijevic, F. R.; Macmillan, D. W. C. Decarboxylative fluorination of aliphatic carboxylic acids via photoredox catalysis. J. Am. Chem. Soc. 2015, 137, 5654−5657. (26) Tinker, L. L.; Bernhard, S. Photon-driven catalytic proton reduction with a robust homoleptic iridium(III) 6-phenyl-2,2′bipyridine complex ([Ir(C∧N∧N)2]+). Inorg. Chem. 2009, 48, 10507−10511. (27) Mamo, A.; Stefio, I.; Parisi, M. F.; Credi, A.; Venturi, M.; Di Pietro, C.; Campagna, S. Luminescent and redox-active iridium(III)cyclometalated compounds with terdentate ligands. Inorg. Chem. 1997, 36, 5947−5950. (28) Whittle, V. L.; Williams, J. A. G. A new class of iridium complexes suitable for stepwise incorporation into linear assemblies: Synthesis, electrochemistry, and luminescence. Inorg. Chem. 2008, 47, 6596−6607. (29) Polson, M.; Fracasso, S.; Bertolasi, V.; Ravaglia, M.; Scandola, F. Iridium cyclometalated complexes with axial symmetry. Synthesis and photophysical properties of a trans -biscyclometalated complex containing the terdentate ligand 2,6-diphenylpyridine. Inorg. Chem. 2004, 43, 1950−1956. (30) Sato, S.; Morikawa, T.; Kajino, T.; Ishitani, O. A highly efficient mononuclear iridium complex photocatalyst for CO2 reduction under visible light. Angew. Chem., Int. Ed. 2013, 52, 988−992. (31) Genoni, A.; Chirdon, D. N.; Boniolo, M.; Sartorel, A.; Bernhard, S.; Bonchio, M. Tuning iridium photocatalysts and light irradiation for enhanced CO2 reduction. ACS Catal. 2017, 7, 154−160. (32) Chirdon, D. N.; Transue, W. J.; Kagalwala, H. N.; Kaur, A.; Maurer, A. B.; Pintauer, T.; Bernhard, S. [Ir(N∧N∧N)(C∧N)L]+: A new family of luminophores combining tunability and enhanced photostability. Inorg. Chem. 2014, 53, 1487−1499. (33) Collin, J.-P.; Dixon, I. M.; Sauvage, J.-P.; Williams, J. A. G.; Barigelletti, F.; Flamigni, L. Synthesis and photophysical properties of Ir(III) bisterpyridine and its homologues: a family of complexes with a long-lived excited-state. J. Am. Chem. Soc. 1999, 121, 5009−5016. (34) Yoshikawa, N.; Matsumura-Inoue, T. Electrochemical and phosphorescent properties of new mixed-ligand Ir(III) complexes coordinated with both terpyridine and various bipyridine derivatives. Anal. Sci. 2003, 19, 761−765. (35) Porras, J. A.; Mills, I. N.; Transue, W. J.; Bernhard, S. Highly fluorinated Ir(III)-2,2’:6’,2″-terpyridine-phenylpyridine-X complexes via selective C-F activation: robust photocatalysts for solar fuel generation and photoredox catalysis. J. Am. Chem. Soc. 2016, 138, 9460−9472.

364

DOI: 10.1021/acs.accounts.7b00375 Acc. Chem. Res. 2018, 51, 352−364