Modular Homogeneous Chromophore–Catalyst Assemblies

Apr 22, 2016 - McCormick , T. M.; Calitree , B. D.; Orchard , A.; Kraut , N. D.; Bright , F. V.; Detty , M. R.; Eisenberg , R. Reductive Side of Water...
0 downloads 0 Views 3MB Size
Download PDF
Article pubs.acs.org/accounts

Modular Homogeneous Chromophore−Catalyst Assemblies Karen L. Mulfort* and Lisa M. Utschig Division of Chemical Sciences and Engineering, Argonne National Laboratory, 9700 South Cass Avenue, Argonne, Illinois 60439, United States CONSPECTUS: Photosynthetic reaction center (RC) proteins convert incident solar energy to chemical energy through a network of molecular cofactors which have been evolutionarily tuned to couple efficient light-harvesting, directional electron transfer, and long-lived charge separation with secondary reaction sequences. These molecular cofactors are embedded within a complex protein environment which precisely positions each cofactor in optimal geometries along efficient electron transfer pathways with localized protein environments facilitating sequential and accumulative charge transfer. By contrast, it is difficult to approach a similar level of structural complexity in synthetic architectures for solar energy conversion. However, by using appropriate self-assembly strategies, we anticipate that molecular modules, which are independently synthesized and optimized for either light-harvesting or redox catalysis, can be organized into spatial arrangements that functionally mimic natural photosynthesis. In this Account, we describe a modular approach to new structural designs for artificial photosynthesis which is largely inspired by photosynthetic RC proteins. We focus on recent work from our lab which uses molecular modules for light-harvesting or proton reduction catalysis in different coordination geometries and different platforms, spanning from discrete supramolecular assemblies to molecule−nanoparticle hybrids to protein-based biohybrids. Molecular modules are particularly amenable to highresolution characterization of the ground and excited state of each module using a variety of physical techniques; such spectroscopic interrogation helps our understanding of primary artificial photosynthetic mechanisms. In particular, we discuss the use of transient optical spectroscopy, EPR, and X-ray scattering techniques to elucidate dynamic structural behavior and lightinduced kinetics and the impact on photocatalytic mechanism. Two different coordination geometries of supramolecular photocatalyst based on the [Ru(bpy)3]2+ (bpy = 2,2′-bipyridine) light-harvesting module with cobaloxime-based catalyst module are compared, with progress in stabilizing photoinduced charge separation identified. These same modules embedded in the small electron transfer protein ferredoxin exhibit much longer charge-separation, enabled by stepwise electron transfer through the native [2Fe-2S] cofactor. We anticipate that the use of interchangeable, molecular modules which can interact in different coordination geometries or within entirely different structural platforms will provide important fundamental insights into the effect of environment on parameters such as electron transfer and charge separation, and ultimately drive more efficient designs for artificial photosynthesis.

■

INTRODUCTION The mission to efficiently convert abundant and carbon-free solar energy to portable and storable chemical energy has drawn scientists and engineers from many disciplines and led to the development of a broad portfolio of photocatalyst architectures.1 Our approach toward artificial photosynthesis (AP) is inspired by photosynthetic reaction center (RC) designs, and largely focused on using a modular strategy for incorporating both photochemical and catalytic function within new architectures. Photosynthetic RC proteins are Nature’s solar energy converters, which effectively capture and convert sunlight with near unity quantum efficiency into a useable electrochemical potential via a series of light-induced rapid, sequential electron transfer reactions.2 These electron transfer events occur through a protein embedded chain of electron donor or acceptor molecules which can include chlorophyll, pheophytin, quinone, and Fe sites. Like natural photosynthesis, © XXXX American Chemical Society

artificial photosynthesis requires several integral steps (i.e., light absorption, charge transfer, substrate capture and small molecule activation) that span multiple length and time scales and dictate a delicate interplay of structure, oxidation state, and coordination. The development of individual modules to address each step allows for independent optimization before system integration and, importantly, the use of molecular modules allows high resolution studies of the ground and transient states using a suite of physical characterization techniques. In our schemes for AP, we aim to link the lightabsorbing and catalytic modules via relatively weak but specific interactions such as hydrogen bonding, metal coordination, or electrostatics which are the same interactions used by RC proteins to finely tune molecular cofactor environments, and, Received: December 11, 2015

A

DOI: 10.1021/acs.accounts.5b00539 Acc. Chem. Res. XXXX, XXX, XXX−XXX

Article

Accounts of Chemical Research

Figure 1. Modules of photocatalyst assemblies and linking strategies discussed in this Account. Chromophore module shown is [Ru(bpy)3]2+, and catalyst module shown is Co(dmgBF2)2, a representative cobaloxime. Linking strategies include supramolecular assembly (top), nanoparticle binding (middle), and protein labeling (bottom).

Figure 2. Progression of modular supramolecular H2 photocatalysts based on Ru(bpy)32+ chromophore and cobaloxime catalyst modules.

accumulative charge separation required for solar energy conversion3 but provide an important conceptual starting point to link molecular chromophore and catalyst modules. However, many of the shortcomings that plague discrete supramolecular photocatalysts may be addressed by linking these same molecular modules through an electron or hole reservoir to store redox equivalents transferred from the chromophore module until they can be used by the catalyst module. Through thoughtful design of the complete system, molecular modules can be integrated into photocatalyst architectures that function as greater than a sum of their parts.

likewise, can be incorporated within artificial designs to gate electron transfer and stabilize charge separation. Also, this assembly strategy enables complex arrangements of the molecular modules without dramatically altering their structure, thereby aiding in understanding how each piece contributes to the system. In this Account, we draw upon our own research and examples from the literature to examine how molecular modules that accomplish light absorption or redox catalysis interact with one another and their environments to achieve AP. We focus on three general classes of photocatalyst architectures which allow us to probe the impact of environment and coordination on parameters relevant to AP: (1) discrete supramolecular assemblies, (2) nanoparticle-based assemblies, and (3) biohybrids (Figure 1). When linked together via covalent bonds, supramolecular interactions, or a scaffold such as a nanoparticle or a protein, molecular modules may not necessarily interact as they do via purely diffusional contact. Discrete supramolecular assemblies of the type described here are rarely able to support the long-lived and

■

COBALOXIMES AS CATALYST MODULES The recognition that cobalt bis(glyoximato) macrocycles, or cobaloximes, can catalyze the reduction of acidic aqueous protons to H2 in the presence of Cr(II) as a sacrificial electron donor was a critical step toward the development of earthabundant, molecular-scale complexes for sustainable energy considerations.4 The chemically driven redox catalysis was accompanied by seminal work which demonstrated that visible B

DOI: 10.1021/acs.accounts.5b00539 Acc. Chem. Res. XXXX, XXX, XXX−XXX

Article

Accounts of Chemical Research

Figure 3. Chemical structures of axially coordinated cobaloxime photocatalyst dyads.

that a notable advantage of linked chromophore and catalyst modules over a multimolecular system is removing the requirement of a very long chromophore excited state lifetime to permit diffusionally controlled quenching (either reductive quenching by the sacrificial electron donor or oxidative quenching by the catalyst module or electron relay).

light can be used to drive cobaloxime-based H2 catalysis using the archetypal [Ru(bpy)3]2+ (bpy = 2,2′-bipyridine) chromophore with a sacrificial electron donor triethanolamine in dimethylformamide (Figure 2, left panel).5 Although this work reached only 38 turnovers (TON) and required excess dimethylglyoxime ligand to offset the macrocycle instability, it provided an important foundation on which to explore new chromophores and modifications to the cobaloxime catalyst module. More recently, cobaloximes were shown to be highly active as H2 electrocatalysts with low overpotential6−9 and are active in multimolecular photocatalytic schemes initiated by excitation of a diverse set of chromophores.10−16 Despite their relative instability, cobaloximes are the most widely used molecular compounds based on earth-abundant elements as H2 photocatalysts, likely because the mechanism is quite well understood.10,11,17−19 Briefly, H2 catalysis by cobaloximes is thought to proceed via protonation of the Co(I) oxidation state to form a Co(III) hydride species that undergoes a second reduction and protonation step to release H2. Recent isolation of cobalt dimers under reducing conditions has called this mechanism into question,20,21 and pathways for H2 catalysis compatible with these species should be investigated. However, in this Account we will focus on the interactions between the cobaloxime catalyst module and the [Ru(bpy)3]2+ chromophore module to compare coordination and environment on the photocatalytic performance parameters. By largely restricting our focus to these molecular modules, this will allow us to emphasize the kinetics of photoinduced electron transfer (PET) and charge recombination which are supported or hindered by the linking chemistry or environment. Certainly the environment plays an important role in the redox properties of each module, but in general the excited state properties of Ru(bpy)32+ chromophore modules (and close analogs) provide sufficient thermodynamic driving force for electron transfer to most cobaloxime modules by either an oxidative or reductive quenching mechanism: Ru3+/2+* ∼ −0.8 V vs SCE, Ru2+/+ ∼ −1.3 V vs SCE,22 and the Co(II/I) reduction potential varies from −0.5 to −1.1 V vs SCE depending on solvent and specific macrocycle.17 We propose

■

MODULAR SUPRAMOLECULAR PHOTOCATALYST ASSEMBLIES A significant step in cobaloxime-based photocatalyst design was the development of supramolecular cobaloxime H2 photocatalysts by Fihri et al.23 These first-of-their-kind photocatalysts were achieved by the self-assembly of pyridyl-functionalized [Ru(bpy)3]2+-based chromophores via coordination directly to the cobalt site of the cobaloxime macrocycle [Co(dmgBF2)2] (dmgBF2− = (difluoroboryl)dimethylglyoximato anion) (Figure 2, center). Following illumination for 4 h in acetone, an increase in H2 TON from 20 to 32 was observed over the analogous multimolecular system using [Et3NH]+ as a proton source. Based upon a slight decrease in the luminescence lifetime of the [Ru(bpy)3]2+ chromophore module following self-assembly, this work proposed that photocatalysis was initiated by oxidative quenching of the chromophore excited state by the cobaloxime catalyst module. A follow up study from the same group used slightly different chromophore modules, based on [Ru(1,10-phenanthroline)3]2+ and [Ir(2phenylpyridine)2(bpy)]+, and a UV cutoff filter so that excitation was limited to the visible region (λ > 380 nm).24 In this work, total H2 TON was higher for the supramolecular photocatalyst but initial turnover frequency was identical to the multimolecular system, which suggests the operable mechanism for photocatalysis is initiated by reductive quenching of the chromophore excited state by the sacrificial electron donor. Related work from Li et al. concluded that photocatalysis was initiated by oxidative quenching of the [Ru(bpy)3]2+ excited state in a study that compared the effect of an aliphatic link between the chromophore and catalyst modules.25 Despite identical chromophore−catalyst coordination and similar driving force for electron transfer, these studies came to C

DOI: 10.1021/acs.accounts.5b00539 Acc. Chem. Res. XXXX, XXX, XXX−XXX

Article

Accounts of Chemical Research

Figure 4. Ultrafast transient optical spectroscopy of modular photocatalysts 2 and 3 in CH3CN following 528 nm excitation; kinetic trace follows decay of chromophore excited state at 600 nm. Kinetic decay is strikingly similar despite addition of covalently bound phenothiazine electron donor to chromophore module.

Ultrafast transient optical spectroscopy was used to probe the very first steps following photoexcitation in the axially coordinated cobaloxime assemblies. Excitation of the chromophore metal-to-ligand charge-transfer (MLCT) band yielded excited state lifetimes typical for similar compounds; pyridyl functionality did not significantly impact the excited state properties. Interestingly, following visible photoexcitation of each assembly, we observed biexponential decay of the chromophore excited state which was remarkably similar across the series (Figure 4). The kinetics of 1−4 vary to some extent, but all are marked by a picosecond decay component and a component with a time constant similar to that of the chromophore module. Since the spectral signature of the reduced cobaloxime module was not observed following visible excitation of any of the assemblies, we interpret the kinetics as PET to the cobaloxime module followed by very fast back electron transfer, leading to no detectable accumulation of the charge separated state. This interpretation is reinforced by a multifrequency EPR study by Niklas et al. that clearly demonstrates the importance of solvent on the electronic structure of the cobalt center and that axial coordination occurs through the dz2 orbital of [Co(II) (dmgBF2)2] which likely facilitates fast back electron transfer.33 These conclusions are further supported by spectroscopic analysis of supramolecular cobaloxime photocatalysts with axially linked porphyrins and corrole chromophore modules.34 Notably, photoinduced Co(I) formation was reported in an axially linked chromophorebridge-cobaloxime triad via stepwise electron transfer, although a detailed kinetic analysis revealed approximately 20% assembly dissociation following oxidative quenching, even in the noncoordinating solvent dichloromethane.35 In the context of H2 photocatalysis, back electron transfer that occurs on the picosecond time scale will likely preclude diffusional interaction between the active catalyst Co(I) state and protons in solution. Therefore, we conclude that the axially linked cobaloxime photocatalyst dyads explored here operate via reductive quenching of the chromophore module and axial ligation provides no mechanistic advantage as it inhibits stabilization of the Co(I) state by providing a direct pathway for charge recombination. To avoid some of the shortcomings of axial coordination in cobaloxime-based photocatalyst design, we synthesized two new [Ru(bpy)3]2+-based chromophore modules with glyoxime functionality to enable Co(II)-templated catalyst module formation (Figure 5).36 Like 1−4, we used synchrotron X-ray

opposing conclusions regarding the mechanism responsible for initiating H2 photocatalysis. These early studies demonstrated the considerable potential of the self-assembly strategy to generate libraries of linked cobaloxime H2 photocatalyst systems, but left significant mechanistic discrepancies regarding the chromophore−catalyst interactions. Therefore, detailed studies using high-resolution characterization techniques to correlate solution structure with the primary photoinduced mechanisms would greatly help optimize designs for efficiency. In order to shed light on the functionally important features of the first generation supramolecular assemblies, we performed detailed structural and spectroscopic studies of the original axially linked photocatalyst (1) and three new photocatalysts comprised of [Ru(terpyridyl)2]2+ and perylene diimide chromophore modules (Figure 3).26 Traditional structure determination techniques such as NMR and mass spectrometry failed to provide reliable information on the assembly structure under solvent conditions relevant to photocatalysis, and therefore we turned to synchrotron-based solution phase Xray scattering to quantitatively measure the structure of the photocatalysts.27 Wide-angle (WAXS) and small-angle (SAXS) data provided information about the intra-assembly interactions as well as the global structure. The Ru−Co distance in 2 obtained from a fragment analysis of the WAXS data matched the distance expected from energy-minimized coordinate models of the assembly within the spatial resolution of the experiment, 11.3 Å (experiment) vs 11.6 Å (model). Therefore, analysis of the WAXS data confirmed assembly formation in solution conditions similar to those used for photocatalysis, but a quantitative measure of assembly formation was obtained by comparison of the Guinier analysis of experimental SAXS data to model data. Guinier analysis is a well-established method28 to quantify the electron-density weighted radius of gyration (Rg) of a particle in solution, and has been used to estimate size and shape of related metal-coordination driven assemblies.29−31 By this analysis, we calculated that 1, 2, and 4 are 66, 90, and 18% assembled, the remainder composed of uncoordinated chromophore and catalyst modules in solution. Given the coordinative lability of the Co(II) oxidation state,32 this is not entirely surprising, although the range of assembly equilibration despite identical coordination, is unexpected. Nevertheless, this structural study using synchrotron X-ray scattering techniques indicates that incomplete formation of the complexes in coordinating solvents such as acetonitrile is likely an important factor in photocatalyst performance. D

DOI: 10.1021/acs.accounts.5b00539 Acc. Chem. Res. XXXX, XXX, XXX−XXX

Article

Accounts of Chemical Research

Figure 5. Chemical structures of glyoxime-linked [Ru(bpy)3]2+-based chromophores (5, 7) and modular photocatalyst assemblies (6, 8) enabled by Co(II) coordination in acetone. Yellow boxes highlight difference in link between modules.

Figure 6. Ultrafast transient spectra of 5 (left) and 6 (right) in CH3CN following 420 nm excitation. Feature centered at 570 nm in transient spectra of 6 correlates with Co(I) state of model cobaloxime compounds generated electrochemically.

catalyst modules. Also, X-band EPR measurements revealed that Co(II) in both equatorial assemblies is high-spin, whereas molecular cobaloxime modules and the axially linked photocatalyst assemblies are Co(II) low-spin.33,38 The design principles uncovered by the structure−activity studies of 6 and 8, as well as related cobaloxime stabilization strategies found in the literature,39 will help us organize chromophore and catalyst modules into new geometries and generate photocatalyst assemblies that uncover structural factors for further stabilization of the catalytically crucial Co(I) intermediate state and increase H2 photocatalytic efficiency.

based characterization techniques to fully characterize the solution structures. X-ray absorption spectroscopy at the cobalt K-edge confirmed the Co(II) oxidation state and local cobalt coordination environment of the equatorial assemblies. To complement analysis of the local structure, Guinier analysis of the SAXS response revealed an increase in Rg between the chromophore module 5 (Rg = 6.57 Å) and assembly 6 (Rg = 10.65 Å) and confirmed assembly formation in acetonitrile. Ultrafast transient optical spectroscopy was used to probe the very early events following excitation of the chromophore module (Figure 6). Visible excitation of 6 in acetonitrile yielded instantaneous formation of the Co(I) state of the catalyst module and this was attributed to intra-assembly PET. The Co(I) feature decays quite rapidly (τ = 26 ps) to yield a transient spectra that resembles that of the chromophore module alone and complete ground state recombination occurs with biexponential kinetics, likely through multiple MLCT states as noted for related Ru(II)poly(pyridyl) compounds.37 This is the first report of Co(I) formation by oxidative quenching in a linked photocatalyst dyad assembly, and despite the short lifetime, we propose that Co(I) formation in 6 was enabled by both structural and electronic factors. Charge separation was not observed following excitation of assembly 8 which only has one amine bond linking the chromophore and

■

NANOPARTICLE-BASED MODULAR PHOTOCATALYSTS The use of semiconductor nanoparticles (NPs) as a link between molecular chromophore and catalyst modules can address two of the major limitations encountered by supramolecular photocatalyst assemblies: (1) long-term photostability,40 and (2) the ability to accumulate multiple photogenerated electrons or holes until they are needed by the slower, diffusion-limited steps of AP.41 In a series of papers, the Reisner group has elegantly demonstrated these concepts by using TiO2 NPs as both a structural and functional link E

DOI: 10.1021/acs.accounts.5b00539 Acc. Chem. Res. XXXX, XXX, XXX−XXX

Article

Accounts of Chemical Research between Ru(bpy)32+-based chromophores and cobaloximebased catalysts (Figure 7). By functionalizing both modules

demonstrated that visible excitation of CoP bound to CdSe/ ZnS core−shell QDs yielded approximately 150 H2 TON from trimethylamine hydrochloride in toluene.48 Importantly, transient optical spectroscopy revealed PET from the excited QD to CoP within ∼100 ps followed by slow charge recombination, >3 ns. A more detailed mechanistic study was somewhat limited by the difficulty in optical detection of the various oxidation states of CoP as they have relatively low extinction coefficients and overlap with the absorbance of the QDs. More recent studies using time-resolved X-ray absorption spectroscopy at the Co K-edge exclusively monitored the cobalt oxidation state to confirm that changes observed in the TA of the QD response did arise from PET to CoP.49 These two schemes are among many examples 40,41 that integrate molecular light-harvesting and catalyst modules with NPs and contribute to a better understanding of the essential steps of AP.

■

PROTEINS AS SCAFFOLDS FOR MOLECULAR MODULES An important design attribute of RC proteins for stabilizing charge separated states is the intermediary protein environment between electron donor and acceptor molecules. Thus, in complementary work to the modular supramolecular photocatalyst assemblies, we have designed a new set of biohybrids that use small soluble proteins as scaffolds for both directed binding and spectroscopic characterization of chromophore and catalyst modules. Our first reported system is based on the ferredoxin (Fd) protein (10.5 kDa) wherein a [Ru(bpy)3]2+based chromophore module was covalently attached to a free cysteine residue and a cobaloxime catalyst module, [Co(dmgBF2)2], was covalently bound to a histidine residue via axial coordination to the Co(II) center (Figure 8). Under visible excitation and in the presence of ascorbic acid as a sacrificial electron donor, the Ru-Fd-Co biohybrid produced 320 TON of H2 in 6 h, which bests photocatalytic efficiency of analogous linked [Ru(bpy)3]2+-molecular catalyst peptide and freely diffusing [Ru(bpy)3]2+ and protein-bound catalyst systems.50 Importantly, the relative structural simplicity of Fd enabled the first detailed EPR and transient optical spectroscopic analysis of the light-induced species in a biohybrid, and suggests that H2 photocatalysis proceeds by oxidative quenching of the chromophore excited state by stepwise electron transfer via the 2Fe-2S cluster to the [Co(dmgBF2)2] module. Charge separation to form the reduced catalyst module [Co(I) (dmgBF2)2]− occurs in less than one microsecond and decays with a time constant of

Figure 7. Chemical structures of molecular modules (top) and scheme of nanoparticle-based modular H2 photocatalysis (bottom). Adapted with permission from ref 42. Copyright 2011 Royal Society of Chemistry.

with phosphonate groups which can coordinate to the surface of TiO2 NPs, they have demonstrated high activity for H2 photocatalysis in near-neutral aqueous conditions,42 even in the presence of oxygen.43 An advanced structural design incorporates a covalently linked pyridyl group on a cobalt-diiminedioxime44 catalyst module which enhances H2 photocatalytic activity when bound to Ru(bpy)32+-sensitized TiO2 NPs.45 Detailed kinetic studies of the RuP-TiO2 NP-CoP systems in water have shown that complete charge separation from the photoexcited Ru(II) chromophore module to the cobaloxime catalyst module through the nanoparticle outcompetes charge recombination to the oxidized chromophore module by approximately 2 orders of magnitude46 and that charge separation to the cobaloxime module has a direct correlation with the distance of the cobalt center to the NP surface.47 A related strategy from our group to generate molecular-NP hybrid photocatalysts is to tether CoP (see Figure 7) to robust, visible-light absorbing quantum dots (QDs). Initial studies

Figure 8. Scheme of Ru−Fd−Co biohybrid H2 photocatalyst, with molecular chromophore and catalyst module binding sites indicated. F

DOI: 10.1021/acs.accounts.5b00539 Acc. Chem. Res. XXXX, XXX, XXX−XXX

Article

Accounts of Chemical Research

within the native flavin cofactor pocket provided by Fld, and this catalyst−Fld hybrid is combined in solution with PSI, enabling the Fld to dock with PSI and precisely position the catalyst to receive the light-driven electrons from PSI. Thus, the PSI protein environment enables molecular catalysts to function in completely aqueous conditions, yielding substantial H2 photocatalytic activity.

greater than 1.5 milliseconds, unprecedented for molecular and supramolecular systems, and only possible on the Fd protein matrix. Perhaps the most obvious way to build upon Nature’s designs for AP is to do exactly that by decorating photosynthetic proteins with abiotic catalyst modules to generate modular biohybrid photocatalysts. Photosystem I (PSI) is a massive 350 kDa RC protein encircled by its own antenna system comprised of 90 chlorophyll and 22 carotenoids.51 Following light capture, energy is transferred to the central core electron transport chain of cofactors where charge separation occurs. The RC protein matrix surrounding the cofactors is essential to specifically position the active sites relative to each other for efficient interactions yet can dynamically adjust to promote and control directional electron transfer and stabilize long-lived charge separation. PSI is poised to photochemically drive H2 production with a long-lived charge-separated state P700+FB− of 60 ms (primary donor P700 is a dimer of chlorophyll, FB is the terminal [4Fe-4S] electron acceptor) and an electrochemical potential of −580 mV (vs NHE) for the FB cluster.52 Catalytic modules have been linked to the PSI framework, including the creation of photosynthetic hybrids with platinum catalysts53,54 and hydrogenase enzymes.55,56 Successful coupling schemes include photoprecipitation reactions to form metal clusters on the surface of PSI,54,57 molecular wire attachment,52 and protein fusion via genetic modification.55 Our group has focused on using self-assembly strategies to build PSI biohybrids for H2 photocatalysis, which enables modular versatility and the potential for self-repairing biohybrid structures. A PSI-PtNP biohybrid assembled by electrostatic interaction between the positively charged acceptor end of PSI and anionically charged 3 nm PtNPs generated 80 000 H2 TON (5.8 TON s−1) in pH 6.3 phosphate buffer,58 which places this system among the highest H2 photocatalytic efficiencies reported in near-neutral aqueous conditions.59 Toward using earth-abundant, molecular catalyst modules, self-assembly of PSI with the well-known cobaloxime module, [Co(dmgH)2pyCl] (dmgH− = dimethylglyoximato anion, py = pyridine), yielded 5900 TON from pH 6.3 water following visible excitation with ascorbic acid as a sacrificial electron donor.60 Although this system only works for 1.5 h, presumably due to cobaloxime degradation and resultant disassembly from the protein, the sheer rapidity of initial H2 production (2.8 TON s−1) establishes the considerable potential of using Nature’s optimized RC photochemistry as a photochemical module when linked to a synthetic molecular catalyst. Looking toward increasing biohybrid stability and activity, we targeted the venerable DuBois-type Ni(II)(P2N2)2 molecular H2 electrocatalysts61,62 as catalyst modules for biohybrids. Like the cobaloxime module, [Ni(II)(PPh2NPh2)2](BF4)2 also selfassembles with PSI with near perfect stoichiometry, presumably via association within a hydrophobic pocket provided by the extrinsic subunits of PSI. Following visible excitation in pH 6.3 water with ascorbic acid as a sacrificial electron donor, the PSINi biohybrid yielded a respectable 1870 H2 TON in 3 h with initial rates 2 orders of magnitude faster than comparable multimolecular systems that use Eosin Y or [Ru(bpy)3]2+ as the chromophore.62 Greater productivity, 2825 H2 TON in 4 h, was achieved by using the native electrostatic interactions between PSI and its acceptor protein flavodoxin (Fld) in a novel protein-directed catalyst delivery approach. For these experiments, [Ni(PPh2NPh2)2](BF4)2 is noncovalently bound

■

CONCLUSION AND OUTLOOK In this Account, we have drawn upon our research and examples from the literature to compare modular photocatalysts in molecular, supramolecular, nanoparticle, and biohybrid systems. Each platform has its distinct advantages and challenges. Multimolecular photocatalyst systems are amenable to high-throughput screening and analysis, particularly useful when considering numerous variables including solvent, concentration, electron transfer relays, and sacrificial reagents. Supramolecular photocatalyst systems may provide the basis for future designs of highly complex photocatalyst arrangements: they circumvent the necessity for long excited state lifetimes and can avoid some of the solvent- and diffusionrelated pitfalls of multimolecular systems. Indeed, the supramolecular systems based on [Ru(bpy)3]2+-cobaloxime assembly discussed here do facilitate fast PET but cannot support longlived charge separation to initiate catalysis. Therefore, upcoming modular supramolecular architectures will incorporate gated electron transfer and molecular rectification schemes to address this shortcoming. The use of semiconductor NPs as functional scaffolds for chromophore and cobaloxime catalyst modules demonstrate the ability of molecular-NP hybrids to accomplish multiple, accumulative PET and support efficient H2 photocatalysis. Building from Nature’s exquisite photosynthetic machinery, the modular PSI biohybrids described here are exceptionally active H2 photocatalysts. The use of the small electron transfer protein ferredoxin has demonstrated that the [2Fe−2S] cofactor acts as an electron transfer relay crucial to charge separation and H2 catalysis and inspires future work in engineering modular binding sites in proteins using RC-like motifs. Emerging modular architectures such as metal−organic frameworks, which organize molecular modules into extended crystalline arrays, share some of the same favorable attributes as semiconductor NPs or proteins with respect to the ability to direct and sustain long-lived charge separation and are poised to make an enormous impact on the field of AP.63,64 We propose that the modular approach described here will be particularly useful as we look to the design of next-generation architectures which avoid sacrificial reagents, enable sequential and accumulative charge transfer, and couple oxidative and reductive multielectron catalysis.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest. Biographies Karen L. Mulfort is a Chemist in the Solar Energy Conversion group at Argonne National Laboratory. She earned her Ph.D. in 2008 at Northwestern University and was a Director’s Postdoctoral Fellow at ANL. Her research is focused on the design and discovery of new molecular, supramolecular, and MOF-based photocatalysts. G

DOI: 10.1021/acs.accounts.5b00539 Acc. Chem. Res. XXXX, XXX, XXX−XXX

Article

Accounts of Chemical Research

systems of Great Activity and Mechanistic Insight. J. Am. Chem. Soc. 2010, 132, 15480−15483. (17) Dempsey, J. L.; Brunschwig, B. S.; Winkler, J. R.; Gray, H. B. Hydrogen Evolution Catalyzed by Cobaloximes. Acc. Chem. Res. 2009, 42, 1995−2004. (18) Muckerman, J. T.; Fujita, E. Theoretical studies of the mechanism of catalytic hydrogen production by a cobaloxime. Chem. Commun. 2011, 47, 12456−12458. (19) Solis, B. H.; Hammes-Schiffer, S. Theoretical Analysis of Mechanistic Pathways for Hydrogen Evolution Catalyzed by Cobaloximes. Inorg. Chem. 2011, 50, 11252−11262. (20) Kahnt, A.; Peuntinger, K.; Dammann, C.; Drewello, T.; Hermann, R.; Naumov, S.; Abel, B.; Guldi, D. M. Kinetic Studies of the Reduction of [Co(dmgH)2(py) (Cl)] Revisited: Mechanisms, Products, and Implications. J. Phys. Chem. A 2014, 118, 4382−4391. (21) Lacy, D. C.; Roberts, G. M.; Peters, J. C. The Cobalt Hydride that Never Was: Revisiting Schrauzer’s “Hydridocobaloxime. J. Am. Chem. Soc. 2015, 137, 4860−4864. (22) Photosensitization and Photocatalysis Using Inorganic and Organometallic Compounds; Kalyanasundaram, K., Gratzel, M., Eds.; Kluwer Academic Publishers: Dordrecht, The Netherlands, 1993; Vol. 14. (23) Fihri, A.; Artero, V.; Razavet, M.; Baffert, C.; Leibl, W.; Fontecave, M. Cobaloxime-Based Photocatalytic Devices for Hydrogen Production. Angew. Chem., Int. Ed. 2008, 47, 564−567. (24) Fihri, A.; Artero, V.; Pereira, A.; Fontecave, M. Efficient H2producing photocatalytic systems based on cyclometalated iridiumand tricarbonylrhenium-diimine photosensitizers and cobaloxime catalysts. Dalton Trans. 2008, 5567−5569. (25) Li, C.; Wang, M.; Pan, J.; Zhang, P.; Zhang, R.; Sun, L. Photochemical hydrogen production catalyzed by polypyridyl ruthenium−cobaloxime heterobinuclear complexes with different bridges. J. Organomet. Chem. 2009, 694, 2814−2819. (26) Mulfort, K. L.; Tiede, D. M. Supramolecular Cobaloxime Assemblies for H2 Photocatalysis: An Initial Solution State StructureFunction Analysis. J. Phys. Chem. B 2010, 114, 14572−14581. (27) Mulfort, K. L.; Mukherjee, A.; Kokhan, O.; Du, P.; Tiede, D. M. Structure-function analyses of solar fuels catalysts using in situ X-ray scattering. Chem. Soc. Rev. 2013, 42, 2215−2227. (28) Guinier, A.; Fournet, G. Small-Angle Scattering of X-rays; Wiley: New York, 1955. (29) Tiede, D. M.; Zhang, R.; Chen, L. X.; Yu, L.; Lindsey, J. S. Structural Characterization of Modular Supramolecular Architectures in Solution. J. Am. Chem. Soc. 2004, 126, 14054−14062. (30) O’Donnell, J. L.; Zuo, X. B.; Goshe, A. J.; Sarkisov, L.; Snurr, R. Q.; Hupp, J. T.; Tiede, D. M. Solution-phase structural characterization of supramolecular assemblies by molecular diffraction. J. Am. Chem. Soc. 2007, 129, 1578−1585. (31) Leininger, S.; Olenyuk, B.; Stang, P. J. Self-Assembly of Discrete Cyclic Nanostructures Mediated by Transition Metals. Chem. Rev. 2000, 100, 853−908. (32) McCormick, T. M.; Han, Z.; Weinberg, D. J.; Brennessel, W. W.; Holland, P. L.; Eisenberg, R. Impact of Ligand Exchange in Hydrogen Production from Cobaloxime-Containing Photocatalytic Systems. Inorg. Chem. 2011, 50, 10660−10666. (33) Niklas, J.; Mardis, K. L.; Rakhimov, R. R.; Mulfort, K. L.; Tiede, D. M.; Poluektov, O. G. The Hydrogen Catalyst Cobaloxime: A Multifrequency EPR and DFT Study of Cobaloxime’s Electronic Structure. J. Phys. Chem. B 2012, 116, 2943−2957. (34) Peuntinger, K.; Lazarides, T.; Dafnomili, D.; Charalambidis, G.; Landrou, G.; Kahnt, A.; Sabatini, R. P.; McCamant, D. W.; Gryko, D. T.; Coutsolelos, A. G.; Guldi, D. M. Photoinduced Charge Transfer in Porphyrin−Cobaloxime and Corrole−Cobaloxime Hybrids. J. Phys. Chem. C 2013, 117, 1647−1655. (35) Veldkamp, B. S.; Han, W.-S.; Dyar, S. M.; Eaton, S. W.; Ratner, M. A.; Wasielewski, M. R. Photoinitiated multi-step charge separation and ultrafast charge transfer induced dissociation in a pyridyl-linked photosensitizer-cobaloxime assembly. Energy Environ. Sci. 2013, 6, 1917−1928.

Lisa M. Utschig is a Chemist in the Solar Energy Conversion group at ANL. She earned her Ph.D. in 1995 at Northwestern University and was a Fermi Postdoctoral Scholar at ANL. Her research is focused on the investigation of photochemical energy conversion in natural photosynthetic systems and development of novel biohybrid complexes for solar fuel production.

■

ACKNOWLEDGMENTS The authors gratefully acknowledge support from the Division of Chemical Sciences, Geosciences, and Biosciences, Office of Basic Energy Sciences of the U.S. Department of Energy through Grant DE-AC02-06CH11357.

■

REFERENCES

(1) Concepcion, J. J.; House, R. L.; Papanikolas, J. M.; Meyer, T. J. Chemical approaches to artificial photosynthesis. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 15560−15564. (2) Blankenship, R. E. Molecular Mechanisms of Photosynthesis, 2nd ed.; Blackwell Science Ltd: Malden, MA, 2002. (3) Ashford, D. L.; Gish, M. K.; Vannucci, A. K.; Brennaman, M. K.; Templeton, J. L.; Papanikolas, J. M.; Meyer, T. J. Molecular Chromophore−Catalyst Assemblies for Solar Fuel Applications. Chem. Rev. 2015, 115, 13006−13049. (4) Connolly, P.; Espenson, J. H. Cobalt-catalyzed evolution of molecular hydrogen. Inorg. Chem. 1986, 25, 2684−2688. (5) Hawecker, J.; Lehn, J. M.; Ziessel, R. Efficient Homogeneous Photochemical Hydrogen Generation and Water Reduction Mediated by Cobaloxime or Macrocyclic Cobalt Complexes. Nouv. J. Chim. 1983, 7, 271−277. (6) Razavet, M.; Artero, V.; Fontecave, M. Proton Electroreduction Catalyzed by Cobaloximes: Functional Models for Hydrogenases. Inorg. Chem. 2005, 44, 4786−4795. (7) Baffert, C.; Artero, V.; Fontecave, M. Cobaloximes as Functional Models for Hydrogenases. 2. Proton Electroreduction Catalyzed by Difluoroborylbis(dimethylglyoximato)cobalt(II) Complexes in Organic Media. Inorg. Chem. 2007, 46, 1817−1824. (8) Hu, X.; Cossairt, B. M.; Brunschwig, B. S.; Lewis, N. S.; Peters, J. C. Electrocatalytic hydrogen evolution by cobalt difluoroboryldiglyoximate complexes. Chem. Commun. 2005, 4723−4725. (9) Hu, X.; Brunschwig, B. S.; Peters, J. C. Electrocatalytic Hydrogen Evolution at Low Overpotentials by Cobalt Macrocyclic Glyoxime and Tetraimine Complexes. J. Am. Chem. Soc. 2007, 129, 8988−8998. (10) Du, P.; Knowles, K.; Eisenberg, R. A Homogeneous System for the Photogeneration of Hydrogen from Water Based on a Platinum(II) Terpyridyl Acetylide Chromophore and a Molecular Cobalt Catalyst. J. Am. Chem. Soc. 2008, 130, 12576−12577. (11) Du, P.; Schneider, J.; Luo, G.; Brennessel, W. W.; Eisenberg, R. Visible Light-Driven Hydrogen Production from Aqueous Protons Catalyzed by Molecular Cobaloxime Catalysts. Inorg. Chem. 2009, 48, 4952−4962. (12) Probst, B.; Kolano, C.; Hamm, P.; Alberto, R. An Efficient Homogeneous Intermolecular Rhenium-Based Photocatalytic System for the Production of H2. Inorg. Chem. 2009, 48, 1836−1843. (13) Probst, B.; Rodenberg, A.; Guttentag, M.; Hamm, P.; Alberto, R. A Highly Stable Rhenium-Cobalt System for Photocatalytic H2 Production: Unraveling the Performance-Limiting Steps. Inorg. Chem. 2010, 49, 6453−6460. (14) Khnayzer, R. S.; McCusker, C. E.; Olaiya, B. S.; Castellano, F. N. Robust Cuprous Phenanthroline Sensitizer for Solar Hydrogen Photocatalysis. J. Am. Chem. Soc. 2013, 135, 14068−14070. (15) Lazarides, T.; McCormick, T.; Du, P.; Luo, G.; Lindley, B.; Eisenberg, R. Making Hydrogen from Water Using a Homogeneous System Without Noble Metals. J. Am. Chem. Soc. 2009, 131, 9192− 9194. (16) McCormick, T. M.; Calitree, B. D.; Orchard, A.; Kraut, N. D.; Bright, F. V.; Detty, M. R.; Eisenberg, R. Reductive Side of Water Splitting in Artificial Photosynthesis: New Homogeneous PhotoH

DOI: 10.1021/acs.accounts.5b00539 Acc. Chem. Res. XXXX, XXX, XXX−XXX

Article

Accounts of Chemical Research (36) Mukherjee, A.; Kokhan, O.; Huang, J.; Niklas, J.; Chen, L. X.; Tiede, D. M.; Mulfort, K. L. Detection of a charge-separated catalyst precursor state in a linked photosensitizer-catalyst assembly. Phys. Chem. Chem. Phys. 2013, 15, 21070−21076. (37) Amouyal, E.; Homsi, A.; Chambron, J.-C.; Sauvage, J.-P. Synthesis and study of a mixed-ligand ruthenium(II) complex in its ground and excited states: bis(2,2′-bipyridine) (dipyrido[3,2-a: 2′,3′c]phenazine-N4N5)ruthenium(II). J. Chem. Soc., Dalton Trans. 1990, 1841−1845. (38) Bakac, A.; Brynildson, M. E.; Espenson, J. H. Characterization of the structure, properties, and reactivity of a cobalt(II) macrocyclic complex. Inorg. Chem. 1986, 25, 4108−4114. (39) Bhattacharjee, A.; Chavarot-Kerlidou, M.; Andreiadis, E. S.; Fontecave, M.; Field, M. J.; Artero, V. Combined Experimental− Theoretical Characterization of the Hydrido-Cobaloxime [HCo(dmgH)2(PnBu3)]. Inorg. Chem. 2012, 51, 7087−7093. (40) Wang, M.; Han, K.; Zhang, S.; Sun, L. Integration of organometallic complexes with semiconductors and other nanomaterials for photocatalytic H2 production. Coord. Chem. Rev. 2015, 287, 1−14. (41) Hammarström, L. Accumulative Charge Separation for Solar Fuels Production: Coupling Light-Induced Single Electron Transfer to Multielectron Catalysis. Acc. Chem. Res. 2015, 48, 840−850. (42) Lakadamyali, F.; Reisner, E. Photocatalytic H2 evolution from neutral water with a molecular cobalt catalyst on a dye-sensitised TiO2 nanoparticle. Chem. Commun. 2011, 47, 1695−1697. (43) Lakadamyali, F.; Kato, M.; Muresan, N. M.; Reisner, E. Selective Reduction of Aqueous Protons to Hydrogen with a Synthetic Cobaloxime Catalyst in the Presence of Atmospheric Oxygen. Angew. Chem., Int. Ed. 2012, 51, 9381−9384. (44) Kaeffer, N.; Chavarot-Kerlidou, M.; Artero, V. Hydrogen Evolution Catalyzed by Cobalt Diimine−Dioxime Complexes. Acc. Chem. Res. 2015, 48, 1286−1295. (45) Willkomm, J.; Muresan, N. M.; Reisner, E. Enhancing H2 evolution performance of an immobilised cobalt catalyst by rational ligand design. Chemical Science 2015, 6, 2727−2736. (46) Lakadamyali, F.; Reynal, A.; Kato, M.; Durrant, J. R.; Reisner, E. Electron Transfer in Dye-Sensitised Semiconductors Modified with Molecular Cobalt Catalysts: Photoreduction of Aqueous Protons. Chem. - Eur. J. 2012, 18, 15464−15475. (47) Reynal, A.; Willkomm, J.; Muresan, N. M.; Lakadamyali, F.; Planells, M.; Reisner, E.; Durrant, J. R. Distance dependent charge separation and recombination in semiconductor/molecular catalyst systems for water splitting. Chem. Commun. 2014, 50, 12768−12771. (48) Huang, J.; Mulfort, K. L.; Du, P.; Chen, L. X. Photodriven Charge Separation Dynamics in CdSe/ZnS Core/Shell Quantum Dot/Cobaloxime Hybrid for Efficient Hydrogen Production. J. Am. Chem. Soc. 2012, 134, 16472−16475. (49) Huang, J.; Tang, Y.; Mulfort, K. L.; Zhang, X. The direct observation of charge separation dynamics in CdSe quantum dots/ cobaloxime hybrids. Phys. Chem. Chem. Phys. 2016, 18, 4300−4303. (50) Soltau, S. R.; Niklas, J.; Dahlberg, P. D.; Poluektov, O. G.; Tiede, D. M.; Mulfort, K. L.; Utschig, L. M. Aqueous light driven hydrogen production by a Ru-ferredoxin-Co biohybrid. Chem. Commun. 2015, 51, 10628−10631. (51) Jordan, P.; Fromme, P.; Witt, H. T.; Klukas, O.; Saenger, W.; Krauss, N. Three-dimensional structure of cyanobacterial photosystem I at 2.5[thinsp][angst] resolution. Nature 2001, 411, 909−917. (52) Lubner, C. E.; Grimme, R.; Bryant, D. A.; Golbeck, J. H. Wiring Photosystem I for Direct Solar Hydrogen Production. Biochemistry 2010, 49, 404−414. (53) Grimme, R. A.; Lubner, C. E.; Bryant, D. A.; Golbeck, J. H. Photosystem I/Molecular Wire/Metal Nanoparticle Bioconjugates for the Photocatalytic Production of H2. J. Am. Chem. Soc. 2008, 130, 6308−6309. (54) Iwuchukwu, I. J.; Vaughn, M.; Myers, N.; O’Neill, H.; Frymier, P.; Bruce, B. D. Self-organized photosynthetic nanoparticle for cell-free hydrogen production. Nat. Nanotechnol. 2010, 5, 73−79.

(55) Ihara, M.; Nishihara, H.; Yoon, K.-S.; Lenz, O.; Friedrich, B.; Nakamoto, H.; Kojima, K.; Honma, D.; Kamachi, T.; Okura, I. Lightdriven Hydrogen Production by a Hybrid Complex of a [NiFe]Hydrogenase and the Cyanobacterial Photosystem I. Photochem. Photobiol. 2006, 82, 676−682. (56) Lubner, C. E.; Applegate, A. M.; Knörzer, P.; Ganago, A.; Bryant, D. A.; Happe, T.; Golbeck, J. H. Solar hydrogen-producing bionanodevice outperforms natural photosynthesis. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 20988−20991. (57) Greenbaum, E. Platinized Chloroplasts: A Novel Photocatalytic Material. Science 1985, 230, 1373−1375. (58) Utschig, L. M.; Dimitrijevic, N. M.; Poluektov, O. G.; Chemerisov, S. D.; Mulfort, K. L.; Tiede, D. M. Photocatalytic Hydrogen Production from Noncovalent Biohybrid Photosystem I/Pt Nanoparticle Complexes. J. Phys. Chem. Lett. 2011, 2, 236−241. (59) Utschig, L. M.; Soltau, S. R.; Tiede, D. M. Light-driven hydrogen production from Photosystem I-catalyst hybrids. Curr. Opin. Chem. Biol. 2015, 25, 1−8. (60) Utschig, L. M.; Silver, S. C.; Mulfort, K. L.; Tiede, D. M. NatureDriven Photochemistry for Catalytic Solar Hydrogen Production: A Photosystem I−Transition Metal Catalyst Hybrid. J. Am. Chem. Soc. 2011, 133, 16334−16337. (61) Rakowski Dubois, M.; Dubois, D. L. Development of Molecular Electrocatalysts for CO2 Reduction and H2 Production/Oxidation. Acc. Chem. Res. 2009, 42, 1974−1982. (62) McLaughlin, M. P.; McCormick, T. M.; Eisenberg, R.; Holland, P. L. A stable molecular nickel catalyst for the homogeneous photogeneration of hydrogen in aqueous solution. Chem. Commun. 2011, 47, 7989−7991. (63) Zhang, T.; Lin, W. Metal-organic frameworks for artificial photosynthesis and photocatalysis. Chem. Soc. Rev. 2014, 43, 5982− 5993. (64) Nasalevich, M. A.; Becker, R.; Ramos-Fernandez, E. V.; Castellanos, S.; Veber, S. L.; Fedin, M. V.; Kapteijn, F.; Reek, J. N. H.; van der Vlugt, J. I.; Gascon, J. Co@NH2-MIL-125(Ti): cobaloxime-derived metal-organic framework-based composite for light-driven H2 production. Energy Environ. Sci. 2015, 8, 364−375.

I

DOI: 10.1021/acs.accounts.5b00539 Acc. Chem. Res. XXXX, XXX, XXX−XXX