14572
J. Phys. Chem. B 2010, 114, 14572–14581
Supramolecular Cobaloxime Assemblies for H2 Photocatalysis: An Initial Solution State Structure-Function Analysis† Karen L. Mulfort and David M. Tiede* DiVision of Chemical Sciences and Engineering, Argonne National Laboratory, 9700 South Cass AVenue, Argonne, Illinois 60439 ReceiVed: March 15, 2010; ReVised Manuscript ReceiVed: May 27, 2010
In this report, we have investigated the correlations between structure and light-induced electron transfer of one known and three new axially coordinated cobaloxime-based supramolecular photocatalysts for the reduction of protons to hydrogen. Solution-phase X-ray scattering and ultrafast transient optical spectroscopy analyses were used in tandem to correlate the self-assembled photocatalysts’ structural integrity in solution with electron transfer and charge separation between the photosensitizer and catalyst fragments. Biphasic excited state decay kinetics were observed for several of the assemblies, suggesting that configurational dispersion plays a role in limiting photoinduced electron transfer. Notably, an assembly featuring a “push-pull” donorphotosensitizer-acceptor triad motif exhibits considerable ultrafast excited state quenching and, of the assemblies examined, presents the strongest opportunity for efficient solar energy conversion. These results will assist in the design and development of next-generation supramolecular photocatalyst architectures. Introduction The production of carbon-neutral and sustainable fuel sources by solar energy conversion is a vital component of our future energy landscape.1 Biology lends us several examples for the processes necessary to transform abundant and clean solar energy to that stored in the form of chemical bonds in small, energy-dense molecules. Following biomimetic or bioinspired approaches, the initial steps of light-harvesting and chargeseparation have been successfully mimicked in synthetic donor-acceptor type complexes.2,3 Significant progress has been made in understanding and reproducing molecular features necessary for the efficient capture of solar photons and the formation of excited states as well as those necessary for efficient photoinduced electron transfer (PET) while minimizing pathways for back electron transfer and sustaining long-lived charge separation. However, efficiently coupling ultrafast, generally one-electron photoexcitation and charge separation with the slower, often multielectron, processes relevant to catalysis (i.e., diffusion and bond-making and -breaking) has been, up to now, quite challenging. Specifically toward the goal of solar H2 production, several first-row, earth-abundant transition-metal-based catalysts have been developed and structurally and mechanistically characterized for the reduction of protons to hydrogen.4-9 Of these, cobaloxime compounds have been shown as productive electrocatalysts for the reaction 2H+ + 2e- f H2 with low overpotential in the presence of a proton source in organic solvents.10-13 Cobaloximes are among the best synthetic transition metal complexes known for H2 production, are relatively easy to synthesize and oxygen tolerant, and are amenable for coupling into natural and artificial photosynthetic systems. Additionally, cobaloximes have been employed in bimolecular photocatalytic systems with relevant photosensitizers (PSs) such as the well-known Ru(bpy)32+,14 Re(I) complexes,15 Pt(II)-acetylide complexes,16,17 and even those containing no noble †
Part of the “Michael R. Wasielewski Festschrift”. * To whom correspondence should be addressed. E-mail:
[email protected].
Figure 1. Photocatalytic cycle of 2H+ + 2e- f H2 via a cobaloxime catalyst, via a homolytic or heterolytic H-H bond-forming pathway.20 In most cases, the cobaloxime ground state (top) oxidatively quenches the PS excited state (PS*). The PS ground state is regenerated by a sacrificial donor, D.
metals.18 Generally, catalysis is initiated by PET from the PS excited state to the cobaloxime catalyst by an oxidative quenching mechanism (Figure 1). The PS ground state is then regenerated by a sacrificial donor, usually triethylamine or triethanolamine. One Pt(II)-cobaloxime system in particular demonstrates the potential of cobaloximes to function as efficient catalysts for H2 production, exhibiting up to 2150 turnovers in 10 h using water as the only proton source.17 Though these are the best synthetic homogeneous H2 photocatalysts to date (containing nonnoble metal active sites), they still lag far behind biological hydrogenases which exhibit turnovers in the range of 9000 s-1.19 Recently, a handful of photocatalytic supramolecular assemblies containing a cobaloxime active site have been described.21,22 The supramolecular approach provides an attractive strategy to design optimized pathways for electron and proton
10.1021/jp1023636 2010 American Chemical Society Published on Web 07/01/2010
Supramolecular Cobaloxime Photocatalysts
J. Phys. Chem. B, Vol. 114, No. 45, 2010 14573
Figure 2. Chemical structures of axially linked cobaloxime photocatalyst assemblies.
transfer. This first generation of supramolecular cobaloximebased catalysts relies on the axial coordination of a pyridylfunctionalized PS to the Co(II) site of the cobaloxime macrocycle. Biomimetic self-assembly of the PS and catalyst is an especially attractive concept: this motif may have the potential to overcome diffusional constraints present in bimolecular systems and also provide a platform by which to investigate coupling many different potential PSs with little additional synthetic effort. This close integration of the PS fragment with the catalyst fragment though has been somewhat disappointing with respect to H2 production efficiency, particularly given the promising results of bimolecular catalyst systems and the putative thermodynamic parameters of the PS and catalyst. Despite this significant advance in photocatalyst design, there have been no detailed studies regarding the solution-phase structural features of these assemblies and the impact on the photophysical properties and successive catalytic efficacy. Therefore, we have initiated a program that investigates the structural and photophysical characteristics of linked photosensitizer-catalyst supramolecular assemblies. In this first report, we have synthesized and characterized a series of axially coordinated cobaloxime photocatalyst assemblies. In the interest of mapping the structural features of these assemblies with the electronic coupling between the PS and the catalytic fragment, we have chosen several different “prototype” PSs which exhibit varied photophysical characteristics and allow us to comment on the general feasibility of this assembly motif for the construction of solar H2 photocatalysts (Figure 2, Table 1). The assemblies have been characterized by cyclic voltammetry, solution-phase X-ray scattering, fluorescence quenching, and ultrafast transient optical spectroscopy. By solution-phase X-ray scattering, at concentrations in the millimolar range, we can unequivocally confirm that the assemblies are in tact, even in a potentially competitive solvent such as acetonitrile. Additionally, electrocatalysis measurements verify that the catalytic efficiency of the cobaloxime fragment is not hindered by PS coordination. Ground state cyclic voltammetry and steady state emission measurements indicate a substantial degree of electrostatic interaction between the PS and the cobaloxime and verify that there is a significant
TABLE 1: Photosensitizers Used in Photocatalyst Assemblies assembly
photosensitizera
1 2 3 4
Ru(bpy)2(L-pyr) · 2PF6 Ru(trpy)(pyr-trpy) · 2PF6 Ru(ttp-PTZ)(pyr-trpy) · 2PF6 pyridyl-PDI
a bpy ) 2,2′-bipyridyl; L-pyr ) (4-pyridine)oxazolo[4,5-f]phenanthroline; trpy ) 2,2′:6′,2′′-terpyridine; pyr-trpy ) 2,2′:4,4′′: 6,2′′-quaterpyridine; ttp-PTZ ) 4′-(p-phenothiazine-N-ylmethylphenyl)-2,2′:6′,2′′-terpyridine; pyridyl-PDI ) N-cyclohexyl-N′-4-pyridyl-1,7-dipyrrolidinylperylene-3,4:9,10-tetracarboxylic acid bisimide.
thermodynamic driving force for PET from the PS excited state to the cobaloxime. However, ultrafast transient optical spectroscopy reveals only weak, if any, evidence of PET. From these studies, we can conclude that though this architecture is amenable to the creation and study of many different PS-catalyst permutations, coordination of the PS through the Co(II) center is complicated by configurational heterogeneity and possibly a limited electronic connectivity. This work suggests strategies for designing assemblies with enhanced activity. Experimental Details 1. General Methods. All commercial reagents were of ACS grade and purchased from Sigma-Aldrich unless otherwise noted. 1H NMR (500 MHz) was performed on a Bruker DMX 500 and referenced to TMS or residual solvent peak as an internal standard. MS-MALDI-TOF was performed on a Bruker Autoflex III using a dithranol matrix. ESI-MS was conducted on a ThermoFisher LCQ Fleet from dilute acetonitrile or methanol solutions. UV-vis absorbance measurements were performed on a Shimadzu UV-1601 spectrophotometer. Steady state emission was measured on a Photon Technology International spectrofluorimeter in a right angle configuration. Samples were dissolved in spectrophotometric grade acetonitrile and fully purged with N2 before measurements. 2. Synthesis. The syntheses of compounds Co(dmgBF2)2 · 2CH3CN,10 Ru(bpy)2(L-pyr) · 2PF6,23 Ru(trpy)(pyr-trpy) · 2PF6,24
14574
J. Phys. Chem. B, Vol. 114, No. 45, 2010
and p-cyanoanilinium tetrafluoroborate25 have previously been reported. Synthetic details for compounds Ru(ttp-PTZ)(pyrtrpy) · 2PF6 and pyridyl-PDI are in the Supporting Information. 3. Electrochemistry. Cyclic voltammetry was conducted using a standard three-electrode cell on a BAS 100B potentiostat and cell stand with a 3 mm diameter glassy carbon working electrode, a Pt wire auxiliary electrode, and a silver wire coated with AgCl as a pseudoreference electrode.26 Each solution in anhydrous acetonitrile or dimethylformamide was purged with N2 prior to measurement and subsequently maintained under a blanket of N2. Tetrabutylammonium hexafluorophosphate (0.1 M) was used as the supporting electrolyte. Ferrocene (purified by sublimation) was added as an internal standard, and redox potentials were referenced to the ferrocene/ferrocenium couple (0.40 V vs SCE (CH3CN), 0.45 V vs SCE (DMF)).27 All scans were performed at 100 mV/s. 4. Electrocatalysis. Cyclic voltammetry was used to measure the electrocatalytic activity of Co(dmgBF2)2 and assemblies 1 and 2 for proton reduction. The electrocatalysis measurements were conducted in the standard three-electrode cell described above. The CV of a solution of 1 mM electrocatalyst in electrolyte (0.1 M TBAPF6/CH3CN) was measured immediately before addition of acid and used as a baseline. Aliquots of 0.20 M acid (p-cyanoanilinium tetrafluoroborate) in the electrolyte solution were added to the cell containing 5 mL of 1.0 mM catalyst in electrolyte solution to acquire proton concentrations up to 18 mM.25 Cyclic voltammograms were recorded at 100 mV/s. 5. Solution-Phase X-ray Scattering. Complete experimental details can be found elsewhere.28-30 Briefly, solution samples (2.5-5 mM) were placed in a 1.5 mm round capillary in the beam path at beamline 12-BM of the Advanced Photon Source at Argonne National Laboratory. The photon energy was set at 20 keV and the solution scattering collected for 1-5 min, depending on scattering intensity, and several scans were averaged to improve signal-to-noise. The blank solvent scattering was subtracted from the sample scattering to obtain the scattering from only the solute. Fragment and assembly scattering was compared to model scattering generated from an energyminimized coordinate model31 and transformed into real space using GNOM.32 6. Ultrafast Transient Optical Spectroscopy. Ultrafast transient absorption (TA) spectra and kinetics were carried out at the Center for Nanoscale Materials (CNM) at Argonne National Laboratory using an amplified Ti:sapphire laser system (Spectra Physics, Spitfire-Pro) and an automated data acquisition system (Ultrafast Systems, Helios). The amplifier was seeded with the 120 fs output from the oscillator (Spectra Physics, Tsunami) and was operated at 1.66 kHz. The output from the amplifier was split 90/10 to pump an optical parametric amplifier (Topas) and generate a continuum (450-750 nm) probe. The TA system enables three-dimensional data collection (spectra/ kinetics/∆OD) for probe wavelengths of 450-750 nm. The excitation wavelength was 527 nm (assemblies 1-3) or 610 nm (assembly 4). The optical density of each sample was
