Light-Induced H2 Evolution with a Macrocyclic Cobalt Diketo

Article Cite This: Inorg. Chem. XXXX, XXX, XXX−XXX

pubs.acs.org/IC

Light-Induced H2 Evolution with a Macrocyclic Cobalt DiketoPyrphyrin as a Proton-Reducing Catalyst Evelyne Joliat-Wick,†,§ Nicola Weder,†,§ Daniel Klose,‡ Cyril Bachmann,† Bernhard Spingler,† Benjamin Probst,† and Roger Alberto*,† †

Department of Chemistry, University of Zürich, Winterthurerstrasse 190, CH-8057 Zürich, Switzerland Laboratory of Physical Chemistry, Department of Chemistry, ETH Zürich, Vladimir-Prelog-Weg 2, CH-8093 Zürich, Switzerland

‡

S Supporting Information *

ABSTRACT: Cobalt complexes are well-known catalysts for photocatalytic proton reduction in water. Macrocyclic tetrapyridyl ligands (pyrphyrins) and their CoII complexes emerged in this context as a highly efficient class of H2 evolution catalysts. On the basis of this framework, a new macrocyclic CoII complex consisting of two keto-bridged bipyridyl units (Co diketo-pyrphyrin) is presented. The complex is synthesized along a convenient route, is well soluble in water, and shows high activity as a water reduction catalyst (WRC). In an aqueous system containing [Ru(bpy)3]Cl2 as a photosensitizer and NaAscO as a sacrificial electron donor, turnover numbers (TONs) of 2500 H2/Co were achieved. Catalysis is terminated by a limited electron supply and decomposition of the photosensitizer but not of the WRC, highlighting the distinct stability of Co diketo-pyrphyrin.

■

INTRODUCTION The replacement of fossil fuels by sustainable energy sources is essential to ensure energy supplies after the exhaustion of the “black gold” as well as to reverse the steady increase in atmospheric CO2. With a power of 120000 TW reaching the earth’s surfacewhich is more energy per hour than mankind would need in one year1the sun provides by far the largest amount of renewable and essentially inexhaustible energy.2 Therefore, conversion of sunlight into chemical energy, such as the photocatalytic splitting of water into H2 and O2, is a suitable approach to store this energy for times of reduced power supply by the sun. Current research on the reductive side has focused on molecular catalysts (WRC for water reduction catalyst) containing platinum,3 manganese,4 iron,5 nickel,6 cobalt,7−10 or molybdenum11 ions incorporated in an amine-, diimine-, cobaloxime-, or polypyridyl-based ligand framework. Macrocyclic polypyridyl complexes featuring cobalt as the center ion are known as highly efficient catalysts12−14 with a substantially higher stability in comparison to commonly used diimine/cobaloxime15,16 complexes. Moreover, they are derivatized relatively easily to tune electronic properties, as recently presented by our group.12−14,17 To extend the basic ligand framework,18−20 we present a polypyridyl macrocycle in which two bipyridyl units are bridged by keto groups. A copper analogue of our complex was already described,21 but photocatalytic properties of such a compound have not been considered so far. The CoII complex of this ligand is highly stable and shows TONs up to 2500 H2/WRC under photocatalytic conditions. Catalysis is not terminated by © XXXX American Chemical Society

WRC decomposition but by a limited electron supply and photosensitizer (PS) decomposition.

■

RESULTS AND DISCUSSION

Double lithiation of 6,6′-dibromo-2,2′-bipyridyl22 and addition of ethyl chloroformate led to the macrocyclic ligand, which was subsequently subjected as a crude mixture for complexation to cobalt(II) as its dibromide salt (Scheme 1). The final compound 2-(TFA)2 was obtained after preparative HPLC separation in modest overall yield but in excellent purity as a dark green powder (see experimental details in the Supporting Information). Scheme 1. Synthetic Route to Co Diketo-pyrphyrin 2-TFA2a

a Reagents and conditions: (i) n-BuLi, ethyl chloroformate in dry THF, −80 °C, 3 h; (ii) CoBr2 in CHCl3/MeOH, 25 °C, 5 min, 2.5% overall yield after HPLC separation.

Received: November 29, 2017

A

DOI: 10.1021/acs.inorgchem.7b02992 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 1. ORTEP representation of 2-(Br)2 with selected labels. Thermal ellipsoids are at the 50% probability level, and hydrogen atoms and solvent molecules are omitted for clarity. Selected bond lengths (Å): Co1−Br1 2.7161(7), Co1−N1 1.922(3), Co1−N2 1.914(3), C11−O1 1.217(5), C1− C11 1.505(6), C10−C11 1.496(6), N1−C1 1.346(5), C1−C2 1.383(5), C2−C3 1.366(6), C3−C4 1.372(6), C4−C5 1.378(5), C5−C6 1.463(6).

into eight lines, seven of which are well-resolved on the highfield side (gz region). In contrast, on the low-field side the pattern of partially overlapping lines is assigned by simulation of the line shape,24 showing a slight rhombic g anisotropy and a smaller hyperfine coupling in-plane in comparison to out-ofplane (gz region) (Figure 2a). The observed g value of 2.15 indicates a significant contribution of the dz2 orbital to the spin density, and in fact DFT calculations using Orca25 predict 89% spin density for this orbital in agreement with the nearly axial symmetry (Figure 2b). Furthermore, we observed no superhyperfine splitting due to the four 14N (I = 1) directly coordinating the CoII, in agreement with a calculated value of about 2 MHz, which is below the resolution limit of the cw EPR spectrum. Therefore, the DFT calculations corroborate the spectral assignment and simulation. 2-(TFA)2 exhibits properties that are particularly favorable for photocatalytic experiments in aqueous solutions: it is well soluble in water, stable under acidic conditions, and insensitive to prolonged light irradiation. In its anhydrous form, 2-(TFA)2 is a dark green powder that can easily be dissolved in water to give a purple solution, indicating that the axial ligands are readily replaced by water molecules. Aqueous solutions are stable at least down to pH 1.5 as monitored by UV−vis spectroscopy (see Figures S1 and S2). Conversely, upon addition of a base, the UV−vis spectrum changes drastically and the effect is reversible. Since the solution decolorizes by adding a base and no precipitate is observed, we conclude that the ligand is hydroxylated at the keto bridge position rather than a coordinated water molecule being deprotonated. The inherent photostability of the WRC 2 is an important feature for photocatalysis. Irradiation of 2-(TFA)2 in water with an LED at 453 nm for 5 days as in the photocatalytic experiments did not affect the WRC at all (see the Supporting Information). Electrochemical investigations of complex 2-(TFA)2 in water at various pH values were conducted in Britton−Robinson buffer26 (0.04 M H3BO3, 0.04 M H3PO4, 0.04 M CH3COOH) at a hanging mercury drop electrode (HMDE). The cyclic voltammograms are complex, featuring many essentially irreversible peaks (see the Supporting Information). These peaks are not explained by metal-centered redox reactions alone but must include reductions of the ligand framework. Some peaks are particularly large (3−5 μA) in comparison to the peak height of the reference (0.8 μA), indicating multielectron transitions at already low anodic potentials. Photocatalysis. Photocatalysis was run with varying concentrations of 2-(TFA)2 in water to give 2-(H2O)2 as the WRC, [Ru(bpy)3]2+ as the PS, and AscOH/NaAscO as the sacrificial electron donor (SED).

Single crystals suitable for X-ray diffraction analysis could be grown from 2-(Br)2 by evaporation of a bromide-containing aqueous solution of the crude product. The complex is centrosymmetric, with the bromides occupying the axial positions (Figure 1). The octahedron around the CoII center is strongly Jahn−Teller distorted with very long Co−Br bonds of 2.7161(7) Å. The IR spectrum of 2-(PF6)2 shows a split carbonyl band νCO at 1687 and 1679 cm−1, respectively. NMR spectra of 2(TFA)2 in D2O showed no signal due to the paramagnetic nature of the complex. The cw EPR spectrum of 2-(TFA)2 in a frozen aqueous solution shows a spectrum limited to the region around g ≈ 2, typical for a low-spin CoII, such as for instance cobII-alamin,23 with nearly axial symmetry (Figure 2a). The CoII nuclear spin (I = 7/2) splits the electron spin transition

Figure 2. (a) 9.5 GHz cw EPR spectrum of 1 mM 2-(TFA)2 in H2O + 20% glycerol, at 15 K (blue) and simulation using gxx = 2.2505, gyy = 2.1964, gzz = 2.0177, Axx = 10.1, Ayy = 97.1, and Azz = 264.6 (A in MHz) (red). (b) DFT-computed spin density distribution shown as 1% isosurface (yellow) on 2-(H2O)2 in ball and stick representation. Distortion of 2-(H2O)2 during DFT geometry optimization amounted to 3° in the center (N−Co−N) and ca. 20° end to end (see the Supporting Information for further details). B

DOI: 10.1021/acs.inorgchem.7b02992 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry Upon photoexcitation, the photosensitizer (PS*) is reductively quenched by the SED, as concluded from time-resolved fluorescence spectroscopic measurements (see the Supporting Information). Oxidative quenching of the PS* by electron transfer to the WRC can be excluded, since the lifetime of the PS* is not influenced by the presence of WRC. Electron transfer to the WRC occurs from the reduced photosensitizer (PS−). Transient spectroscopic measurements at the absorption maximum of the PS− at 510 nm clearly imply the decrease in the lifetime (τ) of reduced PS− with increasing WRC concentration. Evaluation of the transient spectroscopic measurements according to Stern−Volmer analysis confirm dynamic quenching of PS− by the WRC with a bimolecular quenching constant of 2.71 × 109 M−1 s−1 (see Figures S4 and S6). For photocatalytic H2-evolution measurements, a final volume of 10 mL was irradiated at pH 4.0 with an LED at 453 nm (photon flux of 0.35 ± 0.02 μE/s) (see the Supporting Information). In-line dynamic light scattering measurements as well as mercury-poisoning experiments excluded the formation of nanoparticles (see the Supporting Information). Compound 2 was confirmed as the catalytic species by blank measurements without any cobalt, showing only marginal formation of H2, stemming from PS− decomposition.27 Ligand dissociation and catalysis by hydroxylated cobalt ions were excluded as well by blank measurements with CoBr2 (5 μM) only or with CoBr2 in the presence of a 4-fold excess of bpy or py, respectively. None of these blank experiments showed higher H2 production in comparison to catalysis without any cobalt. Typical H2 formation traces as a function of time and for different concentrations of 2 are shown in Figure S7. Within the response time of the instrument setup, the H2 evolution rate increases rapidly and peaks as a function of WRC concentration after 30 min up to 200 min. Subsequently, H2 formation decreases slowly and catalysis ceases after 4−24 h, depending on the initial concentration of the WRC. Varying the concentration of the WRC and/or photon flux under otherwise unchanged conditions are key experiments to elucidate limiting factors of the catalytic system. As shown in Figure 3a, at low WRC concentrations and a photon flux of 0.35 μΕ/s, the maximum rate increases with increasing WRC concentration up to a value of 3 nmol of H2 per second at 1 μM 2, indicating that 2 becomes rate limiting below this concentration. Beyond that concentration, the maximum rate remains constant, implying that photon flux limits the rate. Increasing the light intensity while keeping the WRC concentration constant at 5 μM confirms this hypothesis (Figure 3b): the maximum H2 evolution rate increases linearly with increasing intensity of the incoming light. Therefore, the rate-limiting step of photocatalysis above a WRC concentration of 1 μM is given by the PS (light absorption and conversion efficiency). Only upon exceeding a photon flux of 6 μE/s does the maximum rate reach a plateau, where factors other than the PS confine the production rate. Components determining the stability-limiting factors of the catalytic system were established either by at-line HPLC measurements of the catalytic solution and by correlating achieved TONs to WRC concentration or photon flux, as well as addition experiments, in which new amounts of PS or WRC, respectively, were added to the catalytic solution after H2 evolution had ceased. As a most relevant observation, the WRC can be excluded as the stability-limiting factor in this photocatalytic system. HPLC

Figure 3. H2 evolution vs (a) WRC concentrations at a photon flux of 0.35 ± 0.02 μΕ/s and (b) light intensity with constant WRC concentration of 5 μM. Reaction conditions: volume 10 mL of H2O, 500 μM [Ru(bpy)3]Cl2, 0.7 M AscOH, 0.3 M NaAscO, pH 4.0, 453 nm LED. Black dots denote the TON in H2/Co and red circles the maximum H2 production rate. Error bars indicate the standard deviation of three independent experiments.

analysis after cessation of catalysis showed that the PS was decomposed to a large extent (see Figure S8); thus electron supply to the WRC was reduced and finally stopped. To support this interpretation, we added fresh PS after cessation of H2 evolution. Catalysis was reinitiated (Figure 4), albeit at a reduced rate in comparison to the first run (red trace in Figure 4). Decomposed PS is likely to compete for light absorption with the fresh PS, thereby slowing down catalysis. Nevertheless, between 60 and 90% of the TONs of the precedent run was reached both times, confirming that the catalytic potential of the WRC is essentially retained. Adding fresh WRC after cessation of the catalysis (green trace in Figure 4) did not lead to further H2 production, confirming PS degradation as the performance-limiting factor of the whole catalytic system. The maximum performance in terms of WRC stability was reached at a concentration of 0.5 μM 2, where TONs amounted to 2500 H2/WRC. (Below 0.5 μM, the total amount of H2 (3.5 μmol) was not significantly different from the blank experiment without any cobalt (see Table S3). That is why in Figure 3 the TON at 0.1 μM is excluded from the series by a cross mark.) The TONs achieved with this cyclic tetrapyridyl-based ligand/complex compare well with those of other systems.12,28 C

DOI: 10.1021/acs.inorgchem.7b02992 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

precipitate. Upon irradiation, no immediate H2 formation was observed as was the case in the absence of TCEP, suggesting deactivation of the catalyst by phosphine coordination, as has already been described previously in a similar system.21 The precipitate could be filtered and the residue crystallized to give dark green crystals of [Co(L)(TCEP)] (3). X-ray diffraction analysis of a single crystal of this precipitate revealed that TCEP is coordinated by phosphorus to the in total 5-fold coordinated CoII center (Figure 5). Two of the carboxylic acid groups are

Figure 4. H2 formation traces of PS (red) and WRC (green) addition experiments: addition of a further 1 equiv each, as indicated in the graph by the colored arrows. Starting conditions: 500 μM [Ru(bpy)3]Cl2, 5 μM WRC 2, 0.7 M AscOH, 0.3 M NaAscO, pH 4, 453 nm LED, 0.35 μE/s.

Figure 5. ORTEP plot of 3. Thermal ellipsoids are given at the 50% probability level. All solvent molecules and hydrogen atoms except for the one at the carboxylic acid group are omitted for clarity. Selected bond lengths (Å): Co1−P1 2.2802(14), Co1−N1 1.895(4), Co1−N2 1.898(4), Co1−N3 1.901(4), Co1−N4 1.890(4), C11−O2 1.241(6), C22−O1 1.239(6), C29−P1 1.825(5), C23−P1 1.835(5), C26−P1 1.829(4).

Substantially higher TONs of >20000 TONs could only be achieved in the presence of an additional SED (see below) which recovers dehydroascorbic acid (DHA), the oxidized form of ascorbic acid. When H 2 evolution proceeds, DHA concentration increases and PS− decreases due to decomposition (Scheme 2). The DHA concentration increases

deprotonated, which leads to an overall neutral complex, explaining its low solubility in water. Phosphine coordination and precipitation of the complex prevent immediate H2 formation with high micromolar concentrations of 2, whereas only at very low concentrations in the range of 50 nM TONs of several tens of thousands in H2 were observed. We conclude that the precipitate interferes with H2 formation, whereas in this minute concentration range precipitation does not take place and the WRC is still active.

Scheme 2. Illustration of the Electron Short-Cut Reaction (2) Competing with the Desired H2 Formation (3) after Excitation and Reduction of the PS by AscO− (1)

■

CONCLUSION The macrocyclic polypyridyl ligand platform forms highly stable complexes with CoII and allows the tuning of catalytic properties by altering the chemical nature of the bridgehead carbon. In this study, we synthesized a new water-soluble macrocyclic polypyridyl cobalt complex comprising two keto groups at these sites. Photocatalytic H2 evolution experiments evidenced that the WRC is not rate-limiting down to a concentration of ∼1 μM. Above this concentration range, the photon flux confines the maximum H2-evolution rate, as shown with variable photon flux experiments. TONs of 2500 H2/Co could be achieved, limited by PS but not by WRC decomposition. We therefore conclude that macrocyclic WRCs provide a highly stable photocatalyst platform. The option of introducing substituents at the keto sites will enable further functionalization with coordinating and/or anchoring groups.

steadily and competes with electron forward transfer by quenching reduced PS− according to reaction 2. Whereas this is not a destructive process as such, PS− still decomposes when it is not quenched either by the WRC or by DHA, which therefore limits the performance. The influence of this process is mirrored by the low quantum yield (Φ = 2vmax/photon flux) of photocatalysis, which reaches a maximum of 2−3% at the highest TOFs (see Table S3). As mentioned earlier, HPLC traces of the solution after termination of the catalysis confirmed little remaining PS, which is constantly quenched by the large excess of DHA and therefore no longer leads to any detectable H2 formation (see Figure S8). Only addition of fresh PS in turn interferes with this steady-state situation and drives reaction 1 forward (see Figure 4), coinciding with the observation that catalysis resumed after PS addition. As was reported earlier, the shortcut by DHA can be intercepted by introducing an additional sacrificial electron donor, e.g. the phosphine TCEP (tris(2-carboxyethyl) phosphine hydrochloride).17 TCEP reduces DHA back to AscOH, the latter becoming an electron relay rather than a sacrificial electron donor. Application of this concept to the present system, i.e. adding TCEP to solutions containing catalyst in micromolar concentrations, resulted in a dark green

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.7b02992. Materials and methods, experimental procedures, syntheses, photostability of 2, pH stability, UV/vis titration experiments, electrochemistry of 2, fluorescence emission D

DOI: 10.1021/acs.inorgchem.7b02992 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

(12) Guttentag, M.; Rodenberg, A.; Bachmann, C.; Senn, A.; Hamm, P.; Alberto, R. A highly stable polypyridyl-based cobalt catalyst for homo- and heterogeneous photocatalytic water reduction. Dalton Trans. 2013, 42 (2), 334−337. (13) Bachmann, C.; Guttentag, M.; Spingler, B.; Alberto, R. 3d Element Complexes of Pentadentate Bipyridine-Pyridine-Based Ligand Scaffolds: Structures and Photocatalytic Activities. Inorg. Chem. 2013, 52 (10), 6055−6061. (14) Joliat, E.; Schnidrig, S.; Probst, B.; Bachmann, C.; Spingler, B.; Baldridge, K. K.; von Rohr, F.; Schilling, A.; Alberto, R. Cobalt complexes of tetradentate, bipyridine-based macrocycles: their structures, properties and photocatalytic proton reduction. Dalton Trans. 2016, 45 (4), 1737−1745. (15) Probst, B.; Guttentag, M.; Rodenberg, A.; Hamm, P.; Alberto, R. Photocatalytic H2 Production from Water with Rhenium and Cobalt Complexes. Inorg. Chem. 2011, 50 (8), 3404−3412. (16) Guttentag, M.; Rodenberg, A.; Kopelent, R.; Probst, B.; Buchwalder, C.; Brandstätter, M.; Hamm, P.; Alberto, R. Photocatalytic H2 Production with a Rhenium/Cobalt System in Water under Acidic Conditions. Eur. J. Inorg. Chem. 2012, 2012 (1), 59−64. (17) Bachmann, C.; Probst, B.; Guttentag, M.; Alberto, R. Ascorbate as an electron relay between an irreversible electron donor and Ru(ii) or Re(i) photosensitizers. Chem. Commun. 2014, 50 (51), 6737. (18) Ibrahim, R.; Tsuchiya, S.; Ogawa, S. A Color-Switching Molecule: Specific Properties of New Tetraaza Macrocycle Zinc Complex with a Facile Hydrogen Atom. J. Am. Chem. Soc. 2000, 122 (49), 12174−12185. (19) Ogawa, S.; Narushima, R.; Arai, Y. Aza Macrocycle that selectively binds lithium ion with color change. J. Am. Chem. Soc. 1984, 106 (19), 5760−5762. (20) Ogawa, S.; Uchida, T.; Uchiya, T.; Hirano, T.; Saburi, M.; Uchidac, Y. Lithium complexation of configurational isomers of tetraaza macrocycle containing 2,2′-bipyridine. X-Ray molecular structure of the trans-isomer of a dibutyl dicyano macrocycle. J. Chem. Soc., Perkin Trans. 1 1990, 6, 1649−1653. (21) Burkholder, E.; Heirtzler, F.; Orian, L.; Ouellette, W.; Zubieta, J. Unusual hydrothermal synthesis of a heteroaromatic macrocyclic complex. Polyhedron 2008, 27 (18), 3700−3702. (22) Bai, X.-L.; Liu, X.-D.; Wang, M.; Kang, C.-Q.; Gao, L.-X. Synthesis of New Bis-BINOLs Linked by a 2,2′-Bipyridine Bridge. Synthesis 2005, 2005 (03), 458−464. (23) Pilbrow, J. R. Transition Ion Electron Paramagnetic Resonance; Clarendon Press: Oxford, U.K., 1990; p 1307. (24) Stoll, S.; Schweiger, A. EasySpin, a comprehensive software package for spectral simulation and analysis in EPR. J. Magn. Reson. 2006, 178 (1), 42−55. (25) Neese, F. The ORCA program system. WIREs Comput. Mol. Sci. 2012, 2 (1), 73−78. (26) Britton, H. T. S.; Robinson, R. A. CXCVIII.-Universal buffer solutions and the dissociation constant of veronal. J. Chem. Soc. 1931, 0, 1456−1462. (27) Creutz, C.; Sutin, N.; Brunschwig, B. S. Excited-State Photochemistry in the Tris(2,2’-Bipyridine)Ruthenium(Ii)-Sulfite System. J. Am. Chem. Soc. 1979, 101 (5), 1297−1298. (28) Nippe, M.; Khnayzer, R. S.; Panetier, J. A.; Zee, D. Z.; Olaiya, B. S.; Head-Gordon, M.; Chang, C. J.; Castellano, F. N.; Long, J. R. Catalytic proton reduction with transition metal complexes of the redox-active ligand bpy2PYMe. Chem. Sci. 2013, 4 (10), 3934−3945.

and transient spectroscopy, photocatalysis, crystal data of 2-(Br)2 and 3, and setup for photocatalysis with in-line gas detection to give rates of hydrogen production (PDF) Accession Codes

CCDC 1563592 and 1577935 contain the supplementary crystallographic data for this paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by emailing [email protected], or by contacting The Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail for R.A.: [email protected]. ORCID

Daniel Klose: 0000-0002-3597-0889 Roger Alberto: 0000-0001-5978-3394 Author Contributions §

E.J.-W. and N.W. contributed equally.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Financial support from the University Research Priority Program LightChEC is gratefully acknowledged. Prof. G. Jeschke (ETHZ) is acknowledged for providing access to the EPR spectrometer.

■

REFERENCES

(1) Morton, O.; Dennis, C. A New Day Dawning. Nature 2006, 443, 19. (2) Lewis, N. S.; Nocera, D. G. Powering the planet: Chemical Challanges in solar Energy Utilization. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 15729−15735. (3) Imanbaew, D.; Lang, J.; Gelin, M. F.; Kaufhold, S.; Pfeffer, M. G.; Rau, S.; Riehn, C. Pump-Probe Fragmentation Action Spectroscopy: A Powerful Tool to Unravel Light-Induced Processes in Molecular Photocatalysts. Angew. Chem., Int. Ed. 2017, 56 (20), 5471−5474. (4) Schonweiz, S.; Rommel, S. A.; Kubel, J.; Micheel, M.; Dietzek, B.; Rau, S.; Streb, C. Covalent Photosensitizer-Polyoxometalate-Catalyst Dyads for Visible-Light-Driven Hydrogen Evolution. Chem. - Eur. J. 2016, 22 (34), 12002−12005. (5) Gloaguen, F.; Rauchfuss, T. B. Small molecule mimics of hydrogenases: hydrides and redox. Chem. Soc. Rev. 2009, 38 (1), 100− 108. (6) Han, Z.; Shen, L.; Brennessel, W. W.; Holland, P. L.; Eisenberg, R. Nickel Pyridinethiolate Complexes as Catalysts for the Light-Driven Production of Hydrogen from Aqueous Solutions in Noble-Metal-Free Systems. J. Am. Chem. Soc. 2013, 135 (39), 14659−14669. (7) Losse, S.; Vos, J. G.; Rau, S. Catalytic hydrogen production at cobalt centres. Coord. Chem. Rev. 2010, 254 (21−22), 2492−2504. (8) Artero, V.; Chavarot-Kerlidou, M.; Fontecave, M. Splitting Water with Cobalt. Angew. Chem., Int. Ed. 2011, 50 (32), 7238−7266. (9) Eckenhoff, W. T.; Eisenberg, R. Molecular systems for light driven hydrogen production. Dalton Trans. 2012, 41 (42), 13004− 13021. (10) Dempsey, J. L.; Brunschwig, B. S.; Winkler, J. R.; Gray, H. B. Hydrogen Evolution Catalyzed by Cobaloximes. Acc. Chem. Res. 2009, 42 (12), 1995−2004. (11) Karunadasa, H. I.; Chang, C. J.; Long, J. R. A molecular molybdenum-oxo catalyst for generating hydrogen from water. Nature 2010, 464 (7293), 1329−1333. E

DOI: 10.1021/acs.inorgchem.7b02992 Inorg. Chem. XXXX, XXX, XXX−XXX