Preface for Small Molecule Activation: From Biological Principles to

Preface for Small Molecule Activation: From Biological Principles to Energy Applications. Part 3: Small Molecules Related to (Artificial) Photosynthes...
0 downloads 0 Views 2MB Size
Download PDF
Forum Article pubs.acs.org/IC

Preface for Small Molecule Activation: From Biological Principles to Energy Applications. Part 3: Small Molecules Related to (Artificial) Photosynthesis Marcetta Y. Darensbourg*,† and Antoni Llobet*,‡ †

Department of Chemistry, Texas A&M University, College Station, Texas 77845, United States Departament de Quı ́mica, Universitat Autònoma de Barcelona, Cerdanyola del Vallès, 08193 Barcelona, Spain

‡

T

sites of hydrogenases, biocatalysts for reversible proton reduction, hydrogen evolution reaction (HER), or hydrogen oxidation, the hydrogen uptake reaction, are now well-known; see Figure 2. They occur in (at least) hundreds of organisms, many of which are of primordial origin.6 The hydrogenase catalysts, needed during the development of early life on planet Earth and performing under a reducing environment, arguably began on iron−sulfur surfaces and evolved into sophisticated molecular machines with highly efficient methods for hydrogen catalysis.7 Despite the absence of the noble metals that are in most fuel cells today, the hydrogenase active sites, comprised of iron, sulfur, and sometimes nickel, are extremely efficient. Also reflecting their early and lowly origins is the presence of ligands that are typical of organometallic-like, low-valent and low-spin iron: CO and CN−. Nevertheless, intricate arrangements permit extremely high rates of proton reduction or hydrogen oxidation. The H2O oxidation reaction that provides electrons for the reductive processes is represented in Figure 2 by the active site of PSII.1 Nature uses sunlight as the primary energy input in this reaction that drives the synthesis of ATP and NADPH, which are the source of energy and electrons required to generate carbohydrates (a solar fuel) from CO2 and H2O. The oxidation of H2O to molecular O2 in Nature is carried out by green plants, algae, and cyanobacteria. This process, responsible for today’s oxygen in the atmosphere, is one of the fundamental processes for the maintenance of life in the biosphere. In particular, the oxygen-evolving complex or system in PSII (OEC-PSII) is responsible for catalytic H2O oxidation to O2 and consists of a Mn4CaO5 cubane cluster anchored on the protein via carboxylate and histidine polypeptide residues.8,9 Artificial systems designed to capture light, oxidize H2O, and reduce protons or other organic compounds are of great interest because they can generate useful chemical fuels that can potentially replace the fossil fuels that we use today. Accordingly, a clean and renewable energy source can be obtained. In this Forum, a collection of articles by experts in various fields address challenges in understanding the active sites of the enzymes as revealed through their composition and molecular arrangements that utilize earth-abundant metals. Such challenges have inspired inorganic chemists to stalk small molecules that might reproduce key features of structure and

he ribbon diagrams of protein crystal structures of the hydrogenases and photosystem II (PSII), one of which is included in the cover art of this Forum issue, represent truly amazing advances in the latter decades of the 20th century. While the structures themselves may be interpreted as the culmination of global contributions from molecular and structural biologists, biochemists, and spectroscopists, they present a starting point for even greater challenges of the 21st century. That is to say, the need to understand the fundamental molecular features that control the activity of these evolutionarily perfected biocatalystsones that selectively make and break the toughest chemical bondsis paramount for developing earth-abundant and efficient catalysts for energy storage and conversion. Figure 1 shows one of many ways to represent the processes involved in photosynthesis, both natural and artificial. The

Figure 1. Energy and material flow in photosynthesis, applicable to both natural and artificial processes.

photosensitizer or antenna molecule acts as an electron pump, while catalysts I and II are electrocatalysts in charge of oxidation and reduction half-reactions, respectively. The antenna molecule, which has low-lying excited states, collects photons and energy from the sunlight, excites itself, passes its excited electron(s) to catalyst II, and successively grabs electron(s) from catalyst I, creating an electron flow. Such a flow, driven by the continuous energy input from light, drives the endothermal process of photosynthesis forward, converting H2O and CO2 into O2, H2, and various carbohydrates depending on the catalysts used. That the challenge of directing energy from the sun into the production of “solar” electrons is being addressed is well documented by the proliferation of solar panels in large scale as well as minor uses. The further challenge of storing such electrons as chemical energy in the form of H2, CO, or CH3OH for times when the renewable energy sources are unavailable, or for other devices, draws inorganic chemists to known biological processes that have evolved to do just that. Especially the active © 2016 American Chemical Society

Special Issue: Small Molecule Activation: From Biological Principles to Energy Applications Part 3 Published: January 19, 2016 371

DOI: 10.1021/acs.inorgchem.5b02925 Inorg. Chem. 2016, 55, 371−377

Forum Article

Inorganic Chemistry

Figure 2. Active sites of PSII1 and the hydrogenases,2−5 rendered from protein structural files (http://www.wwpdb.org).

iron complex, including a terminal CO, an acyl CO, and an iron hydride.11 It is one of the few synthetic analogues of [Fe]hydrogenase and is attractive as an example that might lead to inexpensive iron-based catalysts for hydrogenation catalysis. The field of low-molecular-weight functional and structural models for mimicking the activity of the diiron hydrogenases is currently very active. As a consequence, a large number of reports with several types of complexes that are capable of reducing protons to hydrogen have recently been reported. However, the performance of these catalysts is studied under significantly different conditions. In the present Forum Article, Gloaguen12 reports benchmarking experiments and methodologies that allow a fair comparison within proton reduction catalysts based on simple dinuclear iron(I) thiolate complexes. Further, he analyzes the interplay between the redox potential and basicity and their impact on proton-coupled electrontransfer mechanisms. Useful conclusions are then drawn for efficient catalyst design and for understanding the hydrogenase activity at a molecular level. In addition, Gloaguen reports the behavior of iron thiolate complexes holding rigid and unsaturated bridging ligands, with regard to the photodriven evolution of H2, not only in organic solvents but also in aqueous solutions using Eosin Y as a photosensitizer and triethylamine as a sacrificial electron donor.12 The intermediates in proton reduction electrocatalysis are typically presumed to be metal hydrides; i.e., the added proton is expected to park on a reduced metal and shift the electron density onto itself.13,14 Such classic oxidative addition reactions leading to metal hydrides are common in organometallic chemistry but their detection in catalytic processes is not trivial or straightforward, especially for first-row transition metals. Transient spectroscopic techniques, particularly IR spectroscopy for CO- and CN-containing model complexes of the [FeFe]hydrogenase active sites, have provided evidence for such expectations, spanning rates that differ by many orders of magnitude. The Forum Article by Pickett et al. addresses an approach to monitoring the progress of protonation in low valent biomimetic diirion complexes rendered electron rich by phosphine/CO exchange.15 While terminal hydrides are the computationally preferred position for subsequent uptake of a second proton for the H+/H− genesis of H2, isomerization into the less available, more protected bridging hydride position has been seen in all but a few special cases. Tandem studies using stopped-flow UV−vis and IR conclude that protonations at the Fe−Fe bond are rapid, some on the order of milliseconds. From IR band pattern shifts, Pickett et al. conclude that the initial protonation site in this study is between the two iron atoms; slower intramolecular isomerizations lead to apical/basal phosphine positional isomers detectable and resolved by NMR

spectroscopic signals that give clues to their extraordinary activity, perhaps persisting even outside of the protein matrix.

■

CONTRIBUTIONS TO REDUCTIVE PROCESSES One of the most direct connections between natural and artificial photosynthesis lies in the photochemical conversion of CO2 to reduced products such as CO, formate, methanol, etc. The need for protons for C−H and O−H bond formation leads to serious selectivity issues regarding the order of events: uptake of electrons by CO2 versus H+. Hence, in the pursuit of a strategy for the design of electrocatalysts that would select such simple substrates, Taheri and Berben explored the potential of iron carbonyl clusters.10 The well-known tetrairon dodecacarbonyl cluster anions have as core features a carbide, [Fe4C(CO)12]2−, or a nitride, [Fe4N(CO)12]− (Figure 3).

Figure 3. Classic tetrairon dodecacarbonyl cluster anions with carbide or nitride centers that serve as electrocatalysts for CO2 and proton reduction. Structures were derived from the Cambridge Crystallographic Data Base, from entries BAHDIX and PIQROY.

While both clusters are electrocatalysts for proton reduction, in the presence of added CO2 the carbide cluster demonstrates exclusive proton reduction and H2 production, while the nitride is selective for CO2, generating HCO2−. Differences in the hydride donor ability (hydricity) of the reduced and protonated, presumed catalytic intermediates, H[Fe 4C(CO)12]2− and H[Fe4N(CO)12]−, are based on experimental values obtained in both aqueous and nonaqueous media; they provide a solid basis for interpreting the noted selectivity of such clusters. The greater hydricity of the former carbide leads to rates of H2 generation > C−H bond formation. The latter is preferred with clusters of more modest hydricity values. The molecular construction of the monoiron hydrogenase has an active site (Figure 2) that is perhaps the one most directly connected to organometallic chemistry; its heterolytic H2-splitting ability renders it a hydride-transfer catalyst. The contribution of Rose et al. describes a strategy to obtain a synthetic analogue containing several features of the natural 372

DOI: 10.1021/acs.inorgchem.5b02925 Inorg. Chem. 2016, 55, 371−377

Forum Article

Inorganic Chemistry

environments that permit the iron to maintain a 2+ oxidation state, while the nickel undergoes redox activity, NiII,I, III, without substantial changes in the coordination-sphere geometry. While the preparation of synthetic analogues with the core NiII(μSR)2FeII unit of the active site is relatively accessible, reproduction of the remainder of the coordination sphere, particularly for the redox-active nickel, has been elusive. Rauchfuss and colleagues have discovered a novel synthetic route to a new class of mimics of two unprotonated states of [NiFe]-hydrogenase, namely, the Ni−L state, which is paramagnetic, S = 1/2, and assigned to NiI(μ-SR)2FeII, as well as the S = 0 Ni−Si state, NiII(μ-SR)2FeII (Figure 5).21 Success

spectroscopy. The dependence of the protonation rates on steric features of the sulfur-to-sulfur bridge also relates to the reduction potential of (μ-SRS)[Fe(CO)2L)]2 complexes. The placement of the new transient spectroscopy techniques in the context of development over the past 15 years revitalizes insight into the roles of interesting multiiron derivatives that are now known to develop during catalysis. Applications of newer techniques also arise in this report including two-dimensional IR spectroscopy, which identifies minimal vibrational coupling of the CN and CO ligands in (μ-SRS)[Fe(CO)2CN)]22− complexes.15 While the (deceptively) simple (μ-S(CH2)3S)[Fe(CO)3]2 complex has served extremely well to report the fundamental features of the 2Fe subsite of the [FeFe]-hydrogenase active site, there are continuing surprises, as noted in the tetrairon formulations of Pickett et al. and by Heinekey et al.16,17 Still, the singular μ-S(CH2NHCH2S) cofactor, demonstrated to be a functional requirement in biology and in the laboratory, which positions the amine base within a frustrated Lewis pair distance to the open face on the “rotated” iron, was considered to be a largely rigid unit. Its structural maneuvers when attached to the diiron unit were limited to chair−boat interconversions in the cyclohexane-like six-membered FeS2C2N ring. An unusual discovery reported in 2014 from the Dalian laboratories challenged this tenet, conclusively showing that in the doubly oxidized diiron complex a twist that positioned the C−H bond α to sulfur in proximity to iron could result in C−H heterolytic cleavage, proton removal by another built-in pendant base;18 the presence of a variant of the famous Dubois P2N2 ligand was required to extract the proton as noted in Figure 4.19

Figure 5. Switching the redox activity to nickel in the [NiFe]hydrogenase active site mimics using the π ligand η5-C5H5−. Reproduced from ref 21. Copyright American Chemical Society 2015.

depended on the uncommon reagent [(η5-C5H5)3Ni2]+, a synthon for CpNi+, which could be connected to Fe(pdt)(CO)2(dppe). Reduction of the resultant S = 0 NiIIFeII bimetallic afforded the S = 1/2 derivative [CpNiI(μ-pdt)FeII(dppe)CO]. The absence of strong hyperfine coupling from 31 P attached on the iron atoms by electron paramagnetic resonance (EPR) spectra, as well as isomer shifts from Mössbauer spectra and spin densities from density functional theory (DFT) calculations, supported this redox assignment. The authors note that the geometric persistence of the (η5C5H5)Ni unit is significantly different from common NiL4 moieties, which undergo square planar−tetrahedral transformation upon reduction of NiII to NiI, and the η5-C5H5 tridentate ligands are more reliable mimics of the proteinenforced rigidity of three of the thiolate sulfur atoms in the donor environment about nickel in [NiFe]-hydrogenase.21 The possibility that a redox-active ligand can assist the metal to which it is bound in the uptake of electrons for electrocatalysis and provide multiple landscapes with low barriers for electron and proton addition has been a popular approach in catalyst design. Artero et al. have furthered the understanding of the paradigm of noninnocent ligands, dithioleneswith their unsaturated sulfur-to-sulfur linker (the S−C−C−S conjugated system)in the interaction with nickel under electrocatalytic conditions for proton reduction.22 To elucidate the effects from the ligand, Artero et al. surveyed three nickel bis(diphenyl-1,2-dithiolene) complexes with zero, one, or two methoxy groups on each dithiolene, respectively. The electron-donating methoxy groups impose a redistribution of the electrons within the conjugated system, weaken the C−C bond, and reinforce the C−S bond; these conclusions are derived from shifts in the IR and UV−vis spectra. Cyclic voltammetry finds two mainly ligand-based reduction waves ([NiII(S2)2]0/− and [NiII(S2)2]−/2−) whose reduction potentials shift to more negative values with the number of methoxy substituents. Nevertheless, electrocatalysis is concluded to follow E[ECEC] or E[ECCE] mechanisms, depending on the concentration of acid. The complex with one methoxy placed on each dithiolene in an asymmetric fashion was determined to have the strongest catalytic activity because of the asymmetry in

Figure 4. New role for the unique dithiolate connector of the two iron atoms in [FeFe]-hydrogenase active site mimics. Reproduced from ref 18. Copyright American Chemical Society 2014.

Computational studies found a role for the bridgehead CH3 or C2H5 substituents in that agostic C−H interactions from the substituent can promote and stabilize the rotated form. Yet, it was not the γ-C−H agostic interaction that led to C−H activation; instead, a lower-energy path was found that involved the α carbon, as shown in Figure 4.18 In their Forum Article, Zheng et al. demonstrate the consistency of this mechanistic feature in analogues with greater and with less bridgehead steric bulk.20 The complexity of the positional isomers in analogues of [1′]+ increases with simpler sulfur-to-sulfur linkers. Nevertheless, the pendant base within the P2N2 ligand was an absolute requirement for heterolytic C−H bond cleavage. The quest for first-coordination-sphere ligand scaffolds that might reproduce the general electronic environment that is a multivaried collection of first/second/third-coordinationsphere interactions may lead to ligands quite uncommon to biological binding sites. This is particularly true for the active sites of metalloenzymes that are tied into the protein by donors from several residues, as is the case of the [NiFe]-hydrogenase, Figure 2. The challenge in these models has been to produce 373

DOI: 10.1021/acs.inorgchem.5b02925 Inorg. Chem. 2016, 55, 371−377

Forum Article

Inorganic Chemistry

[Ni2(P2W15O56)2]20− (Ni2) exhibits an unusual αααα geometry, the trinickel-containing Wells−Dawson POM [Ni3(OH)3(H2O)3P2W16O59]9− (Ni3) shows a unique structure, where the [α-P2W15O56]12− ligand is capped by a triangular Ni3O13 unit and a WO6 octahedron in a cubanelike structure. Ni3 turns out to be an effective H2O reduction catalyst upon visible-light photosensitization, using the [Ir(ppy)2(dtbbpy)]+ complex [ppy is (2-phenylpyridine (−1H) and dtbbpy is 4,4′-di-tert-butyl-2,2′-dipyridyl] as a light harvester and triethanolamine as a sacrificial electron donor. In contrast, Ni2 under comparable conditions shows negligible activity. This lack of reactivity for Ni2 is attributed primarily to the lack of solvent accessibility to the first coordination sphere of the nickel center.

the sulfur nucleophilicity. The computationally derived mechanism suggests that all proton addition is at the sulfur atoms of the ligand.23 On the other side of the reactant (H+/e−) shuttle system in the hydrogenase mimics is the delivery of protons, found in the [FeFe]-hydrogenase active site to be facilitated by the bridgehead nitrogen atom in the sulfur-to-sulfur linker (Figure 2). Pioneering work with nickel as the metal center has confirmed the efficacy of a built-in pendant base for proton reduction and dihydrogen oxidation.19 In this Forum, Raugei et al. describe how fundamental questions related to proton movement influence the reaction mechanisms in lowmolecular-weight hydrogenase models based on earth-abundant metal complexes.23 They highlight the effect of the pendant amine functionality in the main structure of the ligands that works as a proton relay in the second coordination sphere of a metal complex. This significantly improves the proton mobility, resulting in faster and more energy-efficient catalysis. These reactions are shown to be influenced by factors including pKa values, steric effects, hydrogen bonding, and solvation/desolvation of the exogenous base as well as the acid used. In addition, they also show that the presence of multiple protonation sites can lead to branching points along the catalytic cycle (Figure 6). This then can favor the accessibility

■

CONTRIBUTIONS IN THE BIOLOGICAL ARENA: HYDROGENASES Bren and co-workers25 have used this forum to overview various biohybrid approaches to hydrogen evolution catalysts that are at this time in the forefront of research in this field. The rationale is to optimize synthetic catalysts with promise for a better performance with a needed microenvironment conferred by a supramolecular binding site within a protein or other macromolecule. Examples include complexes of synthetic catalysts such as the well-studied cobaloxime catalyst placed within photosystem I (PSI), which is evolutionarily optimized for converting light energy into a charge-separated state. The performance of this biohybrid is contrasted to fully synthetic, photosensitized systems. Other analyses are directed toward applications of the Dubois−Bullock Ni[P2N2]2 system, which has also been paired with PSI.26 In addition the simple diiron carbonyl complex that serves as a model of the [FeFe]hydrogenase active site has been presumably substituted into a heme binding site of the nitrobindin protein and found to be active for HER in the presence of a synthetic photosensitizer. The impressive review and analysis of Bren et al. is concluded with an original example of a biomolecular, cobalt-containing catalyst within the heme pocket of cytochrome c552 derived from thermally robust Hydrogenobacter thermophiles (Figure 7).

Figure 6. Strategies for tuning the pendant amine, P4, nickel catalysts for hydrogen oxidation or production. Reproduced from ref 23. Copyright American Chemical Society 2015.

of less productive pathways or even the generation of stable offcycle species. Thus, they propose the use of ligands with only one pendant amine so that these problems can be minimized, leading to catalysts with high rates of hydrogen formation. They also report catalysts for hydrogen oxidation based on iron complexes that display high hydrogen binding affinity. However, the improvement of the hydrogen binding enthalpy resulted in a pKa mismatch between the protonated metal center and the protonated pendant amine and, consequently, in rate-limiting intramolecular proton movement. Finally, a general procedure based on thermodynamic arguments is also presented; it permits the simultaneous minimization of the freeenergy change of each catalytic step, yielding a nearly flat freeenergy surface. The molecular catalysts modeled after enzyme active sites are expected to be less robust than those with more mineral-like compositions. In this regard, Hill et al. report the synthesis and structure of two novel polynuclear nickel complexes containing bulky polyoxometalates (POMs) as ligands.24 The activity of these new complexes is evaluated with regard to their capacity to catalytically carry out the reduction of protons to hydrogen. Whereas the dinickel-containing sandwich-type POM

Figure 7. (A) Acetylated cobalt microperoxidase-11 (CoMp11-Ac) and (B) H. thermophilus cytochrome 2-552 (PDF:1YNR). Reproduced from ref 25. Copyright American Chemical Society 2015.

While many questions arise regarding the identity of the biohybrid approach, many avenues are available for matching the properties of the biomolecule, and for engineering new properties into the biomolecule, with those of the synthetic catalysts. Suess et al. introduce their exploration of the biosynthesis of the 2Fe component, the [2Fe]H, of the hydrogen-producing “H-cluster” of [FeFe]-hydrogenase by first describing the comparatively simpler natural photooxidation chemistry that 374

DOI: 10.1021/acs.inorgchem.5b02925 Inorg. Chem. 2016, 55, 371−377

Forum Article

Inorganic Chemistry

on alternate bacterial sources of HydG. The discussion offers as well inspiration for biomimetic studies of additional organoiron fragments in nature, i.e., those that might provide wellcharacterized small-molecule models with easily assigned spectroscopic signals. The article provides an excellent example of how biophysical studies continue to further our insight, make the link between biochemical results and organometallic chemistry, and translate this into the language of inorganic chemists.

produces the Mn3CaMn′−oxo cluster (Figure 2) of the H2Osplitting PSII. This oxygen-rich cluster with its dangling manganese as the active site bears a structural resemblance to a sulfur-rich multiiron cluster in one of the precursors to the [FeFe]-hydrogenase proteins, HydG. A crystal structure of HydG also finds a “dangler”, in this case a labile ferrous iron, not yet in the organometallic CO and CN− ligation required for the active site of [FeFe]-hydrogenase. The authors maintain that this latter iron receives the diatomics produced via HydGmediated radical S-adenosyl-l-methionine (SAM) production of the glycyl radical. Intense scrutiny of the spectral features and biochemical behavior of the pentairon cluster as key to the [2Fe]H site is further stimulated by an alternate and appealing conclusion/premise from the Broderick and Peters groups that a [2Fe−2S] cluster located on HydG is an inorganic precursor to the [2Fe]H site (Figure 8).27,28 The Suess et al. contribution

■

CONTRIBUTIONS IN THE BIOLOGICAL ARENA: OEC The latest developments in the biological front related to the OEC-PSII are presented by Cox et al.29 and Isobe et al.30 While an X-ray crystal structure of the OEC-PSII at 2.6 Å shows most likely the molecular structure of the OEC-PSII at its resting state, the structure of the other four intermediates is more speculative and is still a topic of vigorous discussion. Cox et al. offer a comprehensive overview of the potential structures of the different S states of PSII based on the available information from different spectroscopic techniques and guided by the spin state (Figure 9).29 They describe that with the early S states, S0 and S1, the cofactor retains an open-cubane structure with a vacant coordination site within the cuboidal subunit defined by Mn1, Mn2, and Mn3. In the S3 state, the cofactor instead adopts a higher-spin ground-state configuration of SGS = 3, concomitant with a H2O binding event that renders all manganese ions six-coordinate. On the other hand, S2 displays both low-spin and higher-spin forms (SGS = 1/2 for the open-cubane form S2A and SGS = 5/2 for the closed-cubane form S2B) and thus represents the functional switching point in the catalytic cycle. Finally, they propose that the spin-state interconversion within S2 is associated with a low-barrier O5 translocation, ensuring the accessibility of the S2B form that progresses to the S3 state with an important role played by Ca2+.29

Figure 8. Retroanalysis of the biosynthesis of the H-cluster modified from Suess et al.26 displaying structures from the two major hypotheses regarding the [2Fe]H origin.

to the Forum updates the current understanding of these maturases by review of what is known about the potential for partially decomposed Fe−S clusters in radical SAM enzymes, addressing differences in conclusions from EPR spectroscopy

Figure 9. Assignment of the structures in the PSII mechanistic cycle according to spin states. Reproduced from ref 30. Copyright American Chemical Society 2015. 375

DOI: 10.1021/acs.inorgchem.5b02925 Inorg. Chem. 2016, 55, 371−377

Forum Article

Inorganic Chemistry Isobe et al. concentrate on the description of potential equilibria in the S3 state of the OEC-PSII based on hybrid DFT calculations.30 They found 11 metastable intermediates within a 13 kcal/mol energy range that differ in the degree of protonation, oxidation state, and conformational configurations that imply significant structural changes at the Mn−O bonds. From these calculations, a tentative interpretation of kinetic data on substrate H2O exchange on the order of seconds at room temperature is proposed in agreement with what is observed by time-resolved mass spectrometry.30

■

CONTRIBUTIONS IN TECHNICAL APPLICATION DEVELOPMENT: H2O OXIDATION AND HYDROGEN PRODUCTION Contributions by Meyer et al.31 and Bernhard et al.32 are related to the performance of low-molecular-weight functional models of the OEC-PSII. Meyer et al. present a new technique for the determination of Faradaic efficiencies for the electrocatalytic oxidation of H2O to O2, and in addition, they show its application to extract mechanistic information, especially with regard to deactivation pathways.31 A collector−generator (C−G) device (Figure 10)

Figure 11. H2O oxidation by new iridium(III) complexes. Reproduced from ref 32. Copyright American Chemical Society 2015.

en)Cl2]− displays C2 symmetry. Their electronic structure was explored with DFT calculations and cyclic voltammetry in nonaqueous environments, which unveiled highly reversible IrIII/IrIV redox processes and more complex, irreversible reduction chemistry. The addition of H2O to the electrolyte revealed the ability of these complexes to catalyze H2O oxidation reaction efficiently. Variation of the catalyst concentrations helped to illuminate the kinetics of these H2O oxidation processes and highlighted the robustness of these systems. A stable performance for over 10 days with thousands of catalyst turnovers was observed with the C1-symmetric catalysts. Finally, several techniques were used to ascertain the absence of active IrOx. In the area of photosensitized proton reduction, the collaborative effort of Schröder et al. presents a study directly related to the design expressed in Figure 1, with a rhenium bipyridine “antenna” molecule, and the biomimetic approach to a synthetic catalyst (Figure 12).33 A careful protocol was established for determining the mechanisms of electron transfer from the light catcher to the proton reduction catalyst that has features of both the diiron and nickel−iron hydrogenase enzyme active sites. The efficacy of the separate versus covalently linked molecular arrangement of dyad to catalyst

Figure 10. Schematic for Meyer’s C−G cell for O2 production. Reproduced from ref 31. Copyright American Chemical Society 2015.

based on two fluorine-doped tin oxide working electrodes separated by 1 mm is used for determination of the Faradaic efficiencies for electrocatalytic O2 production. As catalysts, the homogeneous mononuclear ruthenium complexes such as Ru(bda)(isoq)2 (bda = 2,2′-bipyridine and isoq = isoquinoline) and [Ru(tpy)(bpz)(OH2)]2+ (tpy = 2,2′:6′,2″-terpyridine and bpz = 2,2′-bipyrazine) were probed at acidic and basic pH’s and using different types of buffers. This C−G procedure provides a basis for identifying limitations in homogeneous H2O oxidation catalysis arising from (a) coordinative deactivation with added organic buffer bases, (b) potential competitive oxidation reactions of the added base, and (c) potential competitive reactions of the coordinated ligand. Bernhard et al. present a family of tetradentate bis(pyridine2-sulfonamide) (bpsa) compounds as a ligand platform for designing resilient and electronically tunable catalysts capable of performing H2O oxidation catalysis and other processes in highly oxidizing environments (Figure 11).32 These wraparound ligands coordinate to iridium(III) octahedrally, forming an anionic complex with chloride ions bound within the two remaining coordination sites. NMR spectroscopy documented that the more rigid ligand frameworks[Ir(bpsa-Cy)Cl2]− and [Ir(bpsa-Ph)Cl2]produced C1-symmetric complexes, while the complex with the more flexible ethylene linker in [Ir(bpsa-

Figure 12. Linked dyad for photochemical proton reduction. Reproduced from ref 33. Copyright American Chemical Society 2015. 376

DOI: 10.1021/acs.inorgchem.5b02925 Inorg. Chem. 2016, 55, 371−377

Forum Article

Inorganic Chemistry was compared, finding both inter- and intramolecular electrontransfer pathways for the latter, but, as expected, only the intermolecular path is operative for the former. The unexpected result and conclusion of this study is that the linked dyad, the [Re]/[NiFe2] is detrimental to the hydrogen production ability and the photostability of the catalyst as contrasted with that of the separated pair, [Re]/[NiFe2]. By time-resolved infrared spectroscopy, the degradation of the catalyst could be monitored.33 Such thorough analyses highlight the challenges of designing both active and robust systems for artificial photosynthesis.

(6) Vignais, P. M.; Billoud, B.; Meyer, J. FEMS Microbiology Reviews 2001, 25, 455−501. (7) Bethel, R. D.; Darensbourg, M. Y. The Bioorganometallic Chemistry of Hydrogenase. Bioorganometallic Chemistry; Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim, Germany, 2014; pp 239−272. (8) Cox, N.; Pantazis, D. A.; Neese, F.; Lubitz, W. Acc. Chem. Res. 2013, 46, 1588−1596. (9) Yano, J.; Yachandra, V. Chem. Rev. 2014, 114, 4175−4205. (10) Taheri, A.; Berben, L. A. Inorg. Chem. 2015, 10.1021/ acs.inorgchem.5b02293. (11) Durgaprasad, G.; Xie, Z.-L.; Rose, M. J. Inorg. Chem. 2015, DOI: 10.1021/acs.inorgchem.5b01733. (12) Gloaguen, F. Inorg. Chem. 2015, DOI: 10.1021/acs.inorgchem.5b02245. (13) Siegbahn, P. E. M.; Tye, J. W.; Hall, M. B. Chem. Rev. 2007, 107, 4414−4435. (14) Bullock, R. M.; Helm, M. L. Acc. Chem. Res. 2015, 48, 2017− 2026. (15) Hunt, N. T.; Wright, J. A.; Pickett, C. Inorg. Chem. 2015, 10.1021/acs.inorgchem.5b02477. (16) Tard, C.; Pickett, C. J. Chem. Rev. 2009, 109, 2245−2274. (17) Aguirre de Carcer, I.; DiPasquale, A.; Rheingold, A. L.; Heinekey, D. M. Inorg. Chem. 2006, 45, 8000−8002. (18) Zheng, D.; Wang, N.; Wang, M.; Ding, S.; Ma, C.; Darensbourg, M. Y.; Hall, M. B.; Sun, L. J. Am. Chem. Soc. 2014, 136, 16817−16823. (19) DuBois, D. L. Inorg. Chem. 2014, 53, 3935−3960. (20) Zheng, D.; Wang, M.; Wang, N.; Cheng, M.; Sun, L. Inorg. Chem. 2015, DOI: 10.1021/acs.inorgchem.5b00923. (21) Chambers, G. M.; Huynh, M. T.; Li, Y.; Hammes-Schiffer, S.; Rauchfuss, T. B.; Reijerse, E.; Lubitz, W. Inorg. Chem. 2015, DOI: 10.1021/acs.inorgchem.5b01662. (22) Zarkadoulas, A.; Field, M. J.; Papatriantafyllopoulou, C.; Fize, J.; Artero, V.; Mitsopoulou, C. A. Inorg. Chem. 2015, DOI: 10.1021/ acs.inorgchem.5b02000. (23) Raugei, S.; Helm, M. L.; Hammes-Schiffer, S.; Appel, A. M.; O’Hagan, M.; Wiedner, E. S.; Bullock, R. M. Inorg. Chem. 2015, DOI: 10.1021/acs.inorgchem.5b02262. (24) Guo, W.; Lv, H.; Bacsa, J.; Gao, Y.; Lee, J. S.; Hill, C. L. Inorg. Chem. 2015, DOI: 10.1021/acs.inorgchem.5b01935. (25) Kandemir, B.; Chakraborty, S.; Guo, Y.; Bren, K. L. Inorg. Chem. 2015, DOI: 10.1021/acs.inorgchem.5b02054. (26) Suess, D. L. M.; Kuchenreuther, J. M.; De La Paz, L.; Swartz, J. R.; Britt, R. D. Inorg. Chem. 2015, 10.1021/acs.inorgchem.5b02274. (27) Joshi, N.; Shepard, E. M.; Byer, A. S.; Swanson, K. D.; Broderick, J. B.; Peters, J. W. FEBS Lett. 2012, 586, 3939−3943. (28) Broderick, J. B.; Duffus, B. R.; Duschene, K. S.; Shepard, E. M. Chem. Rev. 2014, 114, 4229−4317. (29) Krewald, V.; Retegan, M.; Neese, F.; Lubitz, W.; Pantazis, A. D.; Cox, N. Inorg. Chem. 2015, 10.1021/acs.inorgchem.5b02578. (30) Isobe, H.; Shoji, M.; Shen, J.-R.; Yamaguchi, K. Inorg. Chem. 2015, 10.1021/acs.inorgchem.5b02471. (31) Sherman, B. D.; Sheridan, M. V.; Wee, K.-R.; Song, N.; Dares, C. J.; Fang, Z.; Tamaki, Y.; Nayak, A.; Meyer, T. J. Inorg. Chem. 2015, DOI: 10.1021/acs.inorgchem.5b02182. (32) Li, M.; Takada, K.; Goldsmith, J. I.; Bernhard, S. Inorg. Chem. 2015, DOI: 10.1021/acs.inorgchem.5b01709. (33) Summers, P. A.; Calladine, J. A.; Ghiotto, F.; Dawson, J.; Sun, X.-Z.; Hamilton, M. L.; Towrie, M.; Davies, E. S.; McMaster, J.; George, M. W.; Schröder, M. Inorg. Chem. 2015, DOI: 10.1021/ acs.inorgchem.5b01744.

■

CONCLUDING REMARKS The early work on low-molecular-weight H2O oxidation and proton reduction catalysis gave poor performances based on turnover numbers and turnover frequencies. However, over the last 10 years, this field has evolved enormously, and today we have catalysts that are very fast and give a large number of turnovers. This impressive progress in the field has been mainly achieved thanks to the deep mechanistic knowledge generated from the synergetic relationships of synthesis, spectroscopy, biochemistry, and theory. A better understanding of deactivation pathways stresses the importance of oxidatively robust ligands that, in addition, do not detach for the metal center during catalysis. The challenges lying ahead today in the field are the integration of molecular H2O oxidation catalysts into functional materials that can carry out light-induced H2O oxidation catalysis in the heterogeneous phase and coupling it to light-induced proton reduction photoanodes for overall H2O splitting with sunlight. The recent progress in this direction permits an optimistic vision of future practical applications of solar electrons to the electrochemical generation of fuelsin a green and sustainable fashion. Along the way, the roles of inorganic chemistry in synthesis, mechanism, and understanding seem to ever expand.

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. Notes

Views expressed in this editorial are those of the author and not necessarily the views of the ACS.

■

ACKNOWLEDGMENTS M.Y.D. thanks Shengda Deng for the design and rendering of Figure 1; similarly, recognition goes to Drs. Tianbiao Liu and Xianggao Meng for Figure 2 and Dr. Jason Denny for Figure 3. A.L. and M.Y.D. express much appreciation to all contributors to this Forum and their research and inspiration toward a brighter future for planet Earth.

■

REFERENCES

(1) McEvoy, J. P.; Brudvig, G. W. Chem. Rev. 2006, 106, 4455−4483. (2) Nicolet, Y.; Lemon, B. J.; Fontecilla-Camps, J. C.; Peters, J. W. Trends Biochem. Sci. 2000, 25, 138−143. (3) Volbeda, A.; Charon, M.-H.; Piras, C.; Hatchikian, E. C.; Frey, M.; Fontecilla-Camps, J. C. Nature 1995, 373, 580−587. (4) Ogata, H.; Hirota, S.; Nakahara, A.; Komori, H.; Shibata, N.; Kato, T.; Kano, K.; Higuchi, Y. Structure 2005, 13, 1635−1642. (5) Hiromoto, T.; Warkentin, E.; Moll, J.; Ermler, U.; Shima, S. Angew. Chem., Int. Ed. 2009, 48, 6457−6460. 377

DOI: 10.1021/acs.inorgchem.5b02925 Inorg. Chem. 2016, 55, 371−377