Iridium Azocarboxamide Complexes: Variable Coordination Modes, C − H Activation, Transfer Hydrogenation Catalysis, and Mechanistic Insights

: Azocarboxamides, a special class of azo ligands, display intriguing electronic properties due to their versatile binding modes and coordination ﬂ exibility. These properties may have signi ﬁ cant implications for their use in homogeneous catalysis. In the present report, half-sandwich Ir − Cp * complexes of two di ﬀ erent azocarboxamide ligands are presented. Di ﬀ erent coordination motifs of the ligand were realized using base and chloride abstracting ligand to give N ∧ N, N ∧ O, and N ∧ C-chelated monomeric iridium complexes. For the azocarboxamide ligand having methoxy substituted at the phenyl ring, a mixture of N ∧ C-chelated mononuclear ( Ir-5 ) and N ∧ N,N ∧ C-chelated dinuclear complexes ( Ir-4 ) were obtained by activating the C − H bond of the aryl ring. No such C − H activation was observed for the ligand without the methoxy substituent. The molecular identity of the complexes was con ﬁ rmed by spectroscopic analyses, while X-ray di ﬀ raction analyses further con ﬁ rmed three-legged piano-stool structure of the complexes along with the above binding modes. All complexes were found to exhibit remarkable activity as precatalysts for the transfer hydrogenation of carbonyl groups in the presence of a base, even at low catalyst loading. Optimization of reaction conditions divulged superior catalytic activity of Ir-3 and Ir-4 complexes in transfer hydrogenation over the other catalysts. Investigation of the in ﬂ uence of binding modes on the catalytic activity along with wide range substrates, tolerance to functional groups, and mechanistic insights into the reaction pathway are also presented


INTRODUCTION
−3 The azo group (−NN−) in the azocarboxamide ligand is surrounded by a carbonyl oxygen of the amide group, which can potentially be used as a donor atom along with the nitrogen atom of the azo group (N ∧ O-chelated ring, Chart 1).Furthermore, in the presence of a base, the azocarboxamide ligand can also bind to a metal center via the deprotonated nitrogen atom of the amide together with the nitrogen atom of the azo group (N ∧ N-chelated ring). 1,3Interestingly, it can also provide an additional binding pocket to the metal center by activating the C−H bond from one of the phenyl rings.This depends on the functionalization of the phenyl ring and the electronic properties of the metal center and its ancillary ligands, leading to the formation of the stable five-membered cyclometalated complex (N ∧ C-chelated ring).While examples of the first two coordination modes are already available in the literature for ruthenium complexes, 1−3 the latter cyclometalated coordination mode is not reported for azocarboxamide ligands.
The aforementioned possible coordination mode via the amide oxygen, the nitrogen atoms of the azo group and the activated C−H bond should further allow the formation of a bimetallic complex, in which one of the metal centers coordinates via one of the nitrogen atoms from the azo group and the amide nitrogen forming the N ∧ N chelate ring, while the other metal center binds through the second nitrogen atom from the azo group and the orthometalation of the phenyl ring to form N ∧ C chelate ring.This type of cyclometalated bimetallic complex has never been observed with azocarboxamide ligands.Moreover, such a coordination mode, in which a cyclometalated bridging ligand provides two different binding modes for two metals within the same ligand framework, is generally rare.
While other azo-containing ligands such as pap (phenylazopyridine) and azobispyridine have been extensively studied in coordination chemistry as redox-active bridging ligands, 4−8 azocarboxamides, a special class of azo-containing ligands, have rarely been used in coordination chemistry and for the study of redox as well as catalytic properties.A recent report on ruthenium azocarboxamide half-sandwich complexes from our group illustrated the redox-triggered change in the chelating binding pocket and the electronic properties of the complexes. 2e have also investigated the antitumor activity of ruthenium azocarboxamide complexes. 3In addition, the effect of the position and protonation state of the amide group on the catalytic activities of their ruthenium complexes has also been described. 1−13 In past decades, considerable effort has gone into the development of new transfer hydrogenation catalysts based on transition metals that use readily available hydrogen donors such as isopropanol or formic acid, thereby avoiding the use of hazardous hydrogen gas under high pressure or the stoichiometric amount of strong reducing agents.−21 In contrast, the N ∧ O-chelated half-sandwich iridium complexes derived from different αand β-amino acids are rarely used in catalysis, 22−24 despite their importance from the synthetic and structural viewpoint.
In view of the effect of variable binding modes of the azocarboxamide ligand on the electronic and structural properties, we present here the synthesis and catalytic behavior of N ∧ N,N ∧ C-chelated or N ∧ O-chelated half-sandwich Ir(III) complexes (Chart 2).The effect of additional functional groups on the phenyl substituents of the azocarboxamide ligands (L1 vs L2) on the C−H activation of the resulting metal complexes is investigated.In addition to the monometallic complexes, we report the first example of a bimetallic iridium complex with azocarboxamide ligand L2, in which two iridium centers are coordinated to the ligand in an unusual coordination mode.Electronic and geometric structures of these metal complexes are also discussed.Furthermore, strong effects of the variable mode of chelation on the transfer hydrogenation of carbonyl groups is demonstrated.The overall coordination diversity of azocaboxamides is explored, and the effect of the different coordination modes on the transfer hydrogenation efficiency of the resulting complexes is demonstrated.

RESULTS AND DISCUSSION
2.1.Synthesis and Characterization.Deprotonation of ligand L1 with trimethylamine NEt 3 and subsequent reaction with [Cp*IrCl 2 ] 2 at room temperature gave green-colored complex Ir-1.While the presence of a base led to the formation of a neutral complex, reaction of L1 with [Cp*IrCl 2 ] 2 in the presence of the chloride-abstracting reagent NH 4 PF 6 under reflux conditions led to the formation of cationic complex Ir-2 (Scheme 1).Both complexes are stable in air and were purified by column chromatography over neutral alumina followed by crystallization from a mixture of dichloromethane and hexane.Under these reaction conditions, no activation of the C−H bond of the phenyl ring was detected, even after a prolonged reaction time.The two complexes were characterized by various spectroscopic techniques such as NMR and mass spectrometry.
In the case of Ir-1, the Organometallics the high-field region compared to the starting materials (Figure S3).The disappearance of the signal corresponding to the −NH proton of the free ligand at 8.45 ppm indicates the deprotonation and subsequent coordination to the metal through the amide nitrogen (N ∧ N-chelated).The 1 H NMR spectrum of complex Ir-2 shows multiplets in the range of 8.41−7.19ppm due to the aromatic protons of the phenyl ring and a singlet at 1.75 ppm for the methyl groups of the coligand Cp* (Figure S4).Compared to those of complex Ir-1, all the resonances of Ir-2 are shifted significantly downfield.The presence of the amide NH signal at 9.56 ppm suggests that the ligand is coordinated to the iridium through the desired N ∧ O coordination motif.Moreover, NMR studies (in CDCl 3 ) in the presence of a base (NEt 3 ) reveal that Ir-2 can be converted to Ir-1 as evident from the gradual disappearance of the amide− NH signal at 9.56 ppm with the progress of the reaction (Figure S17). 3 Cyclometalated complex Ir  S7).The 13 C{1H} NMR spectrum shows the resonance for the cyclometalated carbon atom at 163 ppm, which is significantly shifted to the downfield region compared to the other carbon resonances (Figure S7).
On treating L2 with [Cp*IrCl 2 ] 2 in the presence of a base at room temperature under the same reaction conditions as for complex Ir-1, a mixture of two complexes Ir-3 and Ir-4 was obtained in 30 and 15% yields, respectively (Scheme 2).All attempts to exclusively synthesize one of the products failed and gave the mixture of both complexes.However, both complexes are stable in air and could be separated by column chromatography over neutral alumina.All complexes were unambiguously characterized by NMR spectroscopy.The 1 H NMR spectrum of red-colored complex Ir-4 shows two distinct resonances for the methyl groups of the coligand Cp* at 1.82 and 1.29 ppm in a 1:1 ratio and only eight instead of nine resonances for aromatic C−H protons in the aromatic region at 8.39−6.84ppm, indicating the activation of the C−H bond of the phenyl ring (Figure S6).Moreover, the disappearance of the resonance of the −NH proton of the free ligand at 8.45 ppm clearly suggests the formation of a binuclear complex in which the metal centers are bridged through N,N-donor atoms on one side of the binding pockets and N,C-donor atoms on the other side.The ESI mass analysis in a positive mode also confirmed the formation of the binuclear complex, showing a molecular peak at m/z 944.2141.
The 1 H NMR spectrum of green-colored complex Ir-3 shows the resonances expected for the desired mononuclear complex for the ligand in the aromatic region at 8.02−6.98ppm, and the singlet for the methoxy group at 3.92 ppm.The singlet for the methyl groups of the coligand Cp* is shifted to the high field at 1.26 ppm compared to the precursor.The absence of the NH signal at 8.45 ppm indicates N,Ncoordination of the ligand to the metal center (Figure S5).
The 1 H NMR chemical shift as a measure of the C−H acidity of the respective C−H proton for ligands L1 and L2 could be considered as a simplified useful tool to distinguish their different reactivities toward C−H bond activation.Recently, Steel and co-workers reported a good correlation between the 1 H NMR chemical shift as a measure of C−H acidity and the preferential site selectivity of iridium-catalyzed borylation of C−H bonds. 25,26Thus, the resonance of the corresponding C−H proton in L2 (δ = 8.04 ppm) shifted slightly to lower fields than the C−H proton of the corresponding L1 (δ = 7.95), leading to the formation of cyclometalated complexes Ir-4 and Ir-5.Therefore, the activation of the C−H bond by the "Ir III −Cl" motif through a concerted metalation−deprotonation process similar to σbond metathesis could be considered as a plausible mechanism for the activation of the more acidic C−H protons in the case of ligand L2, with the chloride ligand acting as an internal base to deprotonate the C−H protons (Figure 1). 27,28The electron density on the iridium center, the Ir−C aryl bond strength, as well as the basicity of the chloride ion could also play a role for their different reactivity toward C−H activation. 26.2.X-ray Diffraction.Three of the complexes were structurally characterized by single-crystal X-ray diffraction.Single crystals suitable for the diffraction studies were obtained by diffusion of concentrated solutions of the sample in a dichloromethane/hexane solvent mixture at ambient temperature.The crystallographic data and relevant binding parameters for all complexes are collected in Table 1.Analysis of the structural data of all the complexes reveals the formation of five-membered iridacycle in each of the complexes with the iridium(III) center coordinated to nitrogen or oxygen donor atoms of azo/amide groups along with the ortho-metalated C − Scheme 2. Synthesis of Iridium Complexes Ir-3, Ir-4, and Ir-5 center of the aryl ring.The crystal structures of all complexes exhibit typical half-sandwiched three-legged piano-stool geometry, with the η 5 -Cp* ring occupying the seat of the piano stool, with N azo /N amide, O amide donor, C aryl donor, and Cl donor atoms imitating three legs of the stool.All complexes show similar metric parameters within a limit of experimental standard deviation.Consistent with Ir(III)−Cp* complexes reported in the literature, the Cp* ring is coordinated to the central Ir(III) atom in an η 5 fashion, with the distance between the centroid of the Cp* ring and Ir(III) falling within the range of 1.794−1.834Å in all complexes.Depending on the N N azo / N amide donor and C aryl donor atoms of the azocarboxamide ligand, the variable binding pockets are formed in the three complexes.In complex Ir-1, the chelating ligand is bound to the central metal atoms through the N donor atoms of the azo group along with the deprotonated N donor atoms of the amide moiety forming a five-membered chelate ring (Figure 2).However, a five-membered chelate ring in complex Ir-5 consists of Ir(III) attached to the azocarboxamide ligand through N azo and C aryl donor atoms, respectively.In dinuclear complex Ir-4 (Figure 3), two five-membered iridacycles are formed, in which the N azo and N amide donor atoms coordinate to the Ir on one side, while N azo and aryl carbon atom of the phenyl ring bind to the Ir on the other side.The distance between the Ir•••Ir centers in the dinuclear complex is 4.830 Å, which is relatively shorter than that in the reported dinuclear complexes of Ir(III)−Cp*.29 The N−N bond distances of the coordinated azo group are in the range of 1.276−1.294Å, deviating slightly from the free ligand and indicating an unreduced form of the ligand.3 The C−N bond distances, however, vary quite significantly (1.334−1.461Å) among the three different complexes, which could be attributed to the delocalization of charge due to the deprotonation of the NH group of the amide.In contrast, the C−O bond distances in all the complexes are in the range of 1.21−1.22Å, which is almost equal to the free CO bond length.The bite angle around the   Ir center from the N ∧ N and N ∧ C chelate lies in the range of 74.5−76°.30 While the Ir-center in complex Ir-1 is almost coplanar with the ligand, the phenyl ring deviates by a torsion angle of 32.63°.Furthermore, the phenyl group containing the carboxamide is twisted from the chelating plane by ∼44°.In the cyclometalated dinuclear Ir complex, one of the Ir-centers is twisted by 15.40°from the plane of the azocarboxamide.Analysis of the binding parameters around the coordination environment of Ir in complex Ir-4 clearly establishes its identity as an anti-isomer.Thus, the X-ray structural parameters of all complexes clearly demonstrate the molecular variable binding modes of the azocarboxamide ligand in three different cyclometalated Ir(III)−Cp* complexes.

Electrochemistry and Transfer Hydrogenation.
Complex Ir-2 shows irreversible reduction at −1.51 V, in addition to quasi-reversible oxidation at 0.66 V (Figure 4).In complex Ir-1, both the oxidation and reduction processes are irreversible (Figure S1).The irreversible reduction process is likely a ligand-centered process considering the free radical EPR signal without hyperfine coupling with a g value of 2.038 at 298 K (Figure S2) obtained for the one-electron-reduced form of complex Ir-1.−3 The irreversible reduction and oxidation processes indicate the involvement of an electron-transfer/chemical reaction (EC) mechanism with the elimination of chloride ion for the reduction and possible ligand rearrangement for the oxidation processes.
Whereas no oxidative process was observed in a previously studied corresponding ruthenium complex, 3 iridium complex Ir-2 shows a quasi-reversible oxidation process which could be related to the presence of the negatively charged coligand Cp*.In addition, the negatively charged coligand stabilizes higher oxidation states of iridium.The redox behavior of iridium complex Ir-3 is similar to that of complex Ir-1 with one irreversible reduction at −1.25 V with three small reoxidation waves occurring in response to the first reduction (Figure S1).In contrast to mononuclear complex Ir-3, binuclear Ir-4 iridium complex shows two irreversible reductions at −1.15 and −1.97 V, respectively (Table S2).
The catalytic properties of iridium complexes Ir-1−Ir-5 were studied in the transfer hydrogenation of carbonyl groups to investigate the effects of their different binding modes and the ligand substituents at the azocarboxamide ligands L1 and L2.Benzophenone was chosen as the model substrate, isopropanol was used as the both hydrogen source and solvent.KOH was used as the base to optimize the critical reaction parameters.Under N 2 atmosphere, a solution of benzophenone and 1 mol % of catalyst in the presence and absence of 10 mol % KOH was heated to 100 ο C for 20 h.
In the absence of a base, all complexes showed almost no catalytic activity in transfer hydrogenation.As can be seen from Table 2, in the presence of base and 1 mol % catalyst loading, benzophenone could be reduced by all complexes.Significantly different catalytic activity was observed for the different iridium complexes.
Both N ∧ O-and N ∧ N-chelated iridium complexes Ir-1 and Ir-2 showed nearly identical activity, achieving only about 80% conversion within 20 h at a catalyst loading of 1 mol %.It is known from our previous report that in similar ruthenium complexes, the N ∧ O-chelated complex could be converted to the N ∧ N-chelated complex by the addition of base. 3Therefore, the identical activity of the N ∧ O-and N ∧ N-chelated iridium complexes could be attributed to the formation of the same active species by the reaction of Ir-1 and Ir-2 with base.
While N ∧ N-chelated iridium complex Ir-1 showed only 80% conversion after 20 h, complex Ir-3 (N ∧ N), which is the analog of Ir-1, exhibited much higher activity and offered almost quantitative conversion after 20 h.The activity of Ir-3 was found to be higher than that of Ir-1.In the case of Ir-3, the presence of the electron-donating methoxy group makes the iridium center more electron-rich compared to Ir-1, therefore the difference in activity between Ir-1 and Ir-3 upon transfer hydrogenation could be explained by the higher electron density at the iridium center in Ir-3, which normally facilitates the formation of the metal hydride species. 31The bimetallic complex Ir-4, containing both N ∧ N and N ∧ C donor sets, also showed high catalytic activity and near-quantitative conversion after 20 h even at 0.5% catalyst loading.The high reactivity of complex Ir-4 is likely related to the presence of the amide chelating pocket in this complex.The lowest activity was observed for mononuclear cyclometalated iridium complex Ir-5 with a chelating carbon donor (N ∧ C donor), which gave only 60% conversion after 20 h.
The correlation derived from the NMR experiments between the different coordination mode for iridium complexes and the catalytic activity clearly suggests that the  a Reaction conditions: benzophenone (0.25 mmol) in 3 mL of isopropanol at 100 ο C, catalyst precursor (0.5−1 mol %), base (5−10 mol %).Yield measured by 1 H NMR spectroscopy using hexamethylbenzene as an internal standard.
catalytic activity could be modified by improving the ligand donor properties and also by modifying the second coordination site. 32It is known that the reactivity of catalysts for the reduction of ketones can be improved by introducing an N−H unit in the ligand in close proximity to the active metal center, which can lower the barrier of the transition state through a hydrogen bonding network. 33n the case of Ir-5, the amide group is not directly connected to the metal center, which may be the reason for the lower catalytic activity. 24Moreover, conversion between the two different coordination modes in the presence of a base, as suggested for Ir-1 and Ir-2, is probably not possible here because of the strength of the Ir−C bond to the cyclometalated phenyl group.
Further optimization of base, reaction time, and catalyst loading was carried out with Ir-3 and Ir-4, which showed the highest activity among all iridium complexes.The higher activity of Ir-3 and Ir-4 could possibly be ascribed to the bifunctional feature of the irdium azocarboxamide catalyst resulting in the facile activation of the isopropanol which is reported previously by Ikariya and co-workers. 34Identical reactivity was observed for Ir-3 and Ir-4 when using 0.5 mol % loading.Both complexes showed almost complete conversion after 20 h.In the case of Ir-4, the catalyst performance decreased when the amount of catalyst was further reduced to 0.25 mol %, resulting in a conversion of only 70% after 20 h.This observation likely points to the fact that the N,N binding pocket containing the amide donor as in Ir-3 is the most important structural motif for achieving high catalytic activity.The amount of base had a significant effect on the catalytic activity.Decreasing the amount of base from 10 to 5 mol % significantly decreased the catalyst performance.Ir-3 was chosen as the precatalyst for investigating the scope of the transfer hydrogenation process over Ir-4, due to the significantly lower synthesis yield of the latter, although both complexes showed comparable catalytic activity (Table 3).
Subsequently, the scope and limitations of our protocol were evaluated by using a series of ketones and aldehydes with the above optimized reaction conditions.Aromatic ketones could be reduced in excellent yields.The catalyst showed a relatively low activity toward acetophenones with electron-donating substituents.The catalytic activity increased with increasing electron-withdrawing nature of the para substituent.Halogenated aromatic ketones with −F, −Br, and −CF 3 consistently afforded the corresponding alcohols in excellent yields.Sterically encumbered ketone also afforded the desired alcohol in good yields.Of interest was the effective transfer hydrogenation of heteroaromatic ketones such as furyl-based ketones, which present a greater challenge due to the potential coordination of heteroatoms to the metal center.In addition to the reduction of saturated aryl ketones, the transfer hydrogenation of α,β-unsaturated ketone and sterically hindered ketones demonstrated excellent functional group compatibility.
Complex Ir-3 also shows high catalytic activity for the transfer hydrogenation of aldehydes.Normally, aldehydes tend to undergo the Cannizzaro reaction in the presence of many transfer hydrogenation catalysts, but no such products were detected in our case.
Backvall, Andersson, and co-workers have described both the monohydride and dihydride pathways in catalytic transfer hydrogenation reactions. 10To distinguish between the monohydride and dihydride pathways, the transfer hydrogenation of acetophenone was carried out in the presence of isopropanol-OD as the hydride source.For the dihydride pathway, one would expect the incorporation of deuterium in both C−H and O−H positions (Figure 5).However, no noticeable deuterium incorporation was observed in the benzylic C−H position (Figure S16), suggesting that transfer hydrogenation occurs via the monohydride pathway. 35he mechanism of transfer hydrogenation is thought to proceed through the formation of metal hydride species. 36hen a solution of precatalyst Ir-3 in isopropanol was heated in the presence of KOH at 100 ο C, a hydride resonance was detected at −13.7 ppm by NMR, demonstrating the formation of metal hydrides during the catalytic cycle (Figures S14 and  S15).A plausible mechanism is suggested based on one proposed for Noyori type catalysts with N−H ligands (Scheme 3). 16As suggested by Noyori and co-workers, treatment of precursor complex Ir-3 with a base such as KOH first leads to the loss of the chloride ligand, resulting in the formation of Table 3. Transfer Hydrogenation of Various Ketones and Aldehydes with Catalyst Ir-3 a a Unless otherwise stated, yields refer to isolated products.Reaction conditions: substrate (0.25 mmol) in 3 mL of isopropanol at 100 ο C, catalyst precursor (0.5 mol %), base (10 mol %).Yield measured by 1 H NMR spectroscopy using hexamethylbenzene as an internal standard.b Using 1 mol % catalyst.Although different mechanisms based on different transition states have been discussed, the involvement of A and C as the key catalytic intermediates is common to all these scenarios reported so far. 37,38pon heating all iridium complexes in the presence of isopropanol and KOH, metal hydride resonances were detected by NMR in all cases (Figures S14 and S15).Therefore, the steps leading to the formation of iridium hydride from Ir-3 (N ∧ N-chelated iridium complex) would be similar to those for the other N ∧ O-and N ∧ C-chelated iridium complexes.The reactivity differences between N ∧ N-, N ∧ O-, and N ∧ C-chelated iridium complexes could be due to the efficiency of hydride transfer leading to the facile formation of the hydrido complex in the presence of 2-propanol. 39

CONCLUSION
In summary, we have demonstrated that an N ∧ N-or N ∧ O-or N ∧ C-chelated half-sandwich iridium complex can be selectively synthesized from the reaction of an azocarboxamide ligand with [Cp*IrCl 2 ] 2 by simply changing the reaction conditions.
2-(4-Methoxyphenyl)-N-phenyldiazenecarboxamide ligand L2 afforded both N ∧ N-chelated mononuclear (Ir-3) and N ∧ N-and N ∧ C-chelated diiridium complexes (Ir-4) by reactions with [Cp*IrCl 2 ] 2 and base.However, the same ligand gave only the N ∧ C-chelated mononuclear iridium complex (Ir-5) in the presence of NH 4 PF 6 .C−H activation was observed for both Ir-4 and Ir-5 complexes at the phenyl ring.While Ir-5 exhibits only one N ∧ C donor set, binuclear iridium complex Ir-4 shows an unprecedented coordination motif of the azocarboxamide with both N ∧ N and N ∧ C donor sets.
However, N,2-diphenyldiazenecarboxamide ligand L1 afforded only the mononuclear N ∧ N-chelated complex (Ir-1) upon reaction with [Cp*IrCl 2 ] 2 and base and mononuclear N ∧ O-chelated iridium complex Ir-2 in the presence of NH 4 PF 6 .No C−H activation was observed at the phenyl.All these complexes were investigated electrochemically and by spectroscopic methods.
All complexes under investigation were tested in transfer hydrogenation catalysis.When comparing their reactivity in transfer hydrogenation, a prominent effect of the variable mode of chelation on their reactivity was observed.While the N ∧ N (amide) chelated iridium complexes showed the highest reactivity, the lowest activity was observed for the mononuclear N ∧ C-chelated iridium complex, in which the amide group is not directly connected to the metal center.Therefore, the incorporation of the amide N−H group in close proximity to the metal center is highly recommended for the transfer hydrogenation catalyst.We have shown here how the introduction of a remote methoxy group on the ligand periphery can have a decisive effect on the C−H activation at the ligand, the formation of various coordination modes and mononuclear versus dinuclear complexes, as well as on the activity of the resulting complexes in catalytic transfer hydrogenation reactions.It is intriguing that such a small and synthetically simple modification of the ligand can have a dramatic effect on the coordination modes and catalytic properties of the resulting metal complexes.Future work will aim to exploit these small modifications to create metal complexes with tailored properties.

EXPERIMENTAL SECTION
4.1.General Procedures and Materials.Commercially available chemicals were used without further purification.All solvents were dried with appropriate drying agents, distilled, and degassed by standard techniques prior to use. H NMR spectra were recorded with Bruker AV 250, Jeol ECS 400, and JEOL ECZ 400R spectrometers at 25 °C.Some 1 H and 13 C spectra were recorded with a Bruker Avance III 500 MHz instrument at 23 Scheme 3. Plausible Mechanism for the Transfer Hydrogenation of Ketones with N ∧ N-Chelated Iridium Complex Ir-3 Organometallics °C.Chemical shifts are reported in ppm relative to tetramethylsilane or with reference to the residual solvent peaks.Mass spectrometry was performed on an Agilent 6210 ESI-TOF.Cyclic voltammograms were recorded with a PAR VersaStat 4 potentiostat (Ametek) by working in freshly distilled and degassed dichloromethane (DCM; anhydrous, VWR) and acetonitrile with 0.1 M NBu 4 PF 6 (dried, >99.0%, electrochemical-grade, Fluka) as electrolyte.A three-electrode setup was used with glassy carbon as the working electrode, coiled platinum wire as the counter electrode, and coiled silver wire as the pseudoreference electrode.The ferrocene/ferrocenium couple was used as internal reference.
UV/vis−NIR spectra were recorded with an Avantes spectrometer consisting of a light source (AvaLightDH-S-Bal), a UV/vis detector (AvaSpec-ULS2048), and an NIR detector (AvaSpec-NIR256-TEC). Spectroelectrochemical measurements were performed in an optically transparent thin-layer electrochemical (OTTLE) cell 41 (CaF 2 windows) with a platinum mesh working electrode, a platinum mesh counter electrode, and a silver foil pseudoreference electrode.EPR spectra at X-band frequency (ca.9.5 GHz) were obtained with a Magnettech MS-5000 benchtop EPR spectrometer equipped with a rectangular TE 102 cavity and TC HO4 temperature controller.The measurements were performed in synthetic quartz glass tubes.For EPR spectroelectrochemistry, a three-electrode setup was employed using two Teflon-coated platinum wires (0.005 in.bare, 0.008 in.coated) as the working and counter electrodes and a Teflon-coated silver wire (0.005 in.bare, 0.007 in.coated) as the pseudoreference electrode.
For the substrate scope, metal complex Ir-3 (0.5 mol %), respective aldehydes or ketones (0.5 mmol), and KOH (10 mol %) were placed in a Schlenk flask.The flask was evacuated and flushed with nitrogen.iPrOH (3 mL) was added, and the closed flask was heated to 100 °C for 20 h.The solvent was evaporated under reduced pressure, and the crude product was purified by silica column chromatography.
4.8.Metal Hydride Formation.The respective metal complex (5 mg) was dissolved in dry, degassed iPrOH (5 mL) under an inert atmosphere of nitrogen.The resulting solution was heated to 100 °C in a closed flask.After 10 h, the solvent was evaporated under reduced pressure at ambient temperature.The resulting solid was dissolved in dry degassed CD 2 Cl 2 (0.8 mL), and the 1 H NMR spectrum was recorded under a nitrogen atmosphere.

■ ASSOCIATED CONTENT
Scheme 1. Synthesis of Iridium Complexes Ir-1 and Ir-2 -5 was obtained by treating of L2 with [Cp*IrCl 2 ] 2 in the presence of NH 4 PF 6 at elevated temperature (Scheme 2).The 1 H NMR spectrum of Ir-5 features one resonance for the NH proton at 9.57 ppm and three distinct resonances for the C-bonded aryl C−H protons, in addition to five aryl C−H resonances for the phenyl ring, indicating N,C-coordination instead of the usual N,Ocoordination of the ligand (Figure

Figure 1 .
Figure 1.Schematic for the concerted metalation−deprotonation mechanism for C−H bond activation.

Figure 2 .
Figure 2. ORTEP representation of complexes Ir-1 and Ir-5.Ellipsoids are drawn at 50% probability level.Hydrogen atoms and solvent molecules are omitted for clarity.

Figure 3 .
Figure 3. ORTEP representation of dinuclear complex Ir-4.Ellipsoids are drawn at 50% probability level.Hydrogen atoms and solvent molecules are omitted for clarity.