{Co4O4} and {CoxNi4âxO4} Cubane Water Oxidation Catalysts as

Article pubs.acs.org/JACS

{Co4O4} and {CoxNi4−xO4} Cubane Water Oxidation Catalysts as Surface Cut-Outs of Cobalt Oxides Fangyuan Song,† René Moré,† Mauro Schilling,† Grigory Smolentsev,‡ Nicolo Azzaroli,‡ Thomas Fox,† Sandra Luber,† and Greta R. Patzke*,† †

Department of Chemistry, University of Zurich, Winterthurerstrasse 190, CH-8057 Zurich, Switzerland Paul Scherrer Institute, CH-5232 Villigen PSI, Switzerland

‡

S Supporting Information *

ABSTRACT: The future of artificial photosynthesis depends on economic and robust water oxidation catalysts (WOCs). Cobalt-based WOCs are especially promising for knowledge transfer between homogeneous and heterogeneous catalyst design. We introduce the active and stable {CoII4O4} cubane [CoII4(dpy{OH}O)4(OAc)2(H2O)2](ClO4)2 (Co4O4-dpk) as the first molecular WOC with the characteristic {H2O-Co2(OR)2-OH2} edge-site motif representing the sine qua non moiety of the most efficient heterogeneous Co-oxide WOCs. DFT-MD modelings as well as in situ EXAFS measurements indicate the stability of the cubane cage in solution. The stability of Co4O4-dpk under photocatalytic conditions ([Ru(bpy)3]2+/S2O82−) was underscored with a wide range of further analytical methods and recycling tests. FT-IR monitoring and HR-ESI-MS spectra point to a stable coordination of the acetate ligands, and DFT-MD simulations along with 1H/2H exchange experiments highlight a favorable intramolecular base functionality of the dpy{OH}O ligands. All three ligand types enhance proton mobility at the edge site through a unique bioinspired environment with multiple hydrogen-bonding interactions. In situ XANES experiments under photocatalytic conditions show that the {CoII4O4} core undergoes oxidation to Co(III) or higher valent states, which recover rather slowly to Co(II). Complementary ex situ chemical oxidation experiments with [Ru(bpy)3]3+ furthermore indicate that the oxidation of all Co(II) centers of Co4O4-dpk to Co(III) is not a mandatory prerequisite for oxygen evolution. Moreover, we present the [CoIIxNi4−x(dpy{OH}O)4(OAc)2(H2O)2](ClO4)2 (CoxNi4−xO4-dpk) series as the first mixed Co/Ni-cubane WOCs. They newly bridge homogeneous and heterogeneous catalyst design through fine-tuned edge-site environments of the Co centers.

■

INTRODUCTION

this widely accepted sine qua non of cobalt oxide-based WOCs63−77 in a molecular environment. Detailed mechanistic insight into Co-cubane/oxide WOCs still relies widely on computational results,61,65,78−81 and an exemplary study on catalytically active {CoIII4O4} cubane models37 identified water attack on an oxygen radical associated with a CoIV state as a favorable pathway.82 However, their unified mechanistic understanding through experimental studies remains indispensable, but still controversial, as outlined briefly below. Molecular {H2O-Co2(OH)2-OH2} units have only recently been studied on a catalytically inactive {CoIII2} dipyridylethane naphthyridine complex with special emphasis on its aqua ligand and buffer anion exchange properties.66,83 Most of the experimentally substantiated mechanistic proposals for cubane WOCs are focused on oxidized forms of CoIII4O4(OAc)4(py)4, together with new ligand tuning approaches.84 Although the direct observation of such cubane intermediates remains difficult,85 two recent studies on their [CoIII3CoIV] state

The worldwide quest for clean hydrogen fuel production through photocatalytic water splitting is widely inspired by nature’s {CaMn4O5} oxygen evolving complex (OEC) as a major paradigm for the construction of low-cost and robust water oxidation catalysts (WOCs).1−12 Knowledge transfer between molecular WOC mechanisms13−19 and heterogeneous catalyst optimization is a forefront task of current artificial photosynthesis research.20−28 While major progress was made in the demanding synthesis and analysis of Mn-cubane OEC replicas,29−35 {CoIII4O4} cubane WOCs emerged as alternative and active model systems for both fundamental studies36−45 and the construction of molecular photoanodes.46−52 In our previous work, we equipped {CoII4O4} WOC types as versatile targets32,53−58 with bioinspired features for remarkably flexible ligand shells, protonation states, and electronic as well as nuclear structures.59−62 We present [CoII4(dpy{OH}O)4(OAc)2(H2O)2](ClO4)2 (Co4O4-dpk; dpk = di(2-pyridyl)ketone) as the first active homogeneous Co-cubane WOC with the {H2O-Co2(OR)2OH2} edge-site motif. Co4O4-dpk provides unique access to © 2017 American Chemical Society

Received: July 14, 2017 Published: September 27, 2017 14198

DOI: 10.1021/jacs.7b07361 J. Am. Chem. Soc. 2017, 139, 14198−14208

Journal of the American Chemical Society

■

underscore the key role of HO− addition for water oxidation, while ruling out the participation of bridging oxygen atoms.44,45 Combined kinetic and density functional theory (DFT) studies furthermore suggested the formation of a CoV site in the cubane core.45 An alternative model, developed in line with EPR data, is based on a charge-delocalized [CoIII, III, IV, IV] core undergoing oxygen release via a gem-peroxo intermediate.44 Next, the combination of X-ray absorption and 1s 3p resonant inelastic X-ray scattering on a faster time scale than EPR identified the apparent delocalization of CoIV states in oxidized [CoIII4O4] cubanes as an extremely fast hopping rate.86 UV/vis and X-ray spectroelectrochemical studies provided further strong evidence for the formation of a [CoIII2CoIV2] cubane, whose cofacial CoIV centers enable antiferromagnetic exchange coupling as a favorable prerequisite for O−O bond formation.86 Recent EPR investigations in a wider magnetic field range, however, point to the involvement of a high-spin d7 Co3IIICoIIO-O-H intermediate during photocatalytic oxygen evolution ([Ru(bpy)3]2+/S2O82− assay).87 A comparable variety of mechanistic options emerged from pioneering experimental studies of multiple edge-sites on heterogeneous Co3O4 surfaces. Rapid-scan FT-IR investigations of Co3O4 under photocatalytic conditions ([Ru(bpy)3]2+/ S2O82−) provided strong evidence for a kinetically competent superoxide-based intermediate resulting from oxidation and water nucleophilic attack (WNA) of a {(OH)-CoIII2O2-(OH)} moiety together with a parallel O2 formation pathway via a slower mononuclear CoIVO site.64,88,89 This stands in contrast with the results of first turnover analyses of cobalt oxide nanoparticles formed in the presence of methylenediphosphate. Isotopologue ratio data recorded after [Ru(bpy)3]3+ addition strongly suggest O2 evolution from active {H2OCoIII2(O)2-OH2} units via an intramolecular O−O coupling mechanism.90 The new Co4O4-dpk now places the controversially discussed edge-site functionality within a clear-cut molecular ligand environment that may serve as a rigid matrix. This is the first direct link to compare the influence of the edge-site synthetic paradigm on molecular and heterogeneous WOCs, respectively. Co4O4-dpk furthermore opens up new avenues to heavily sought-after mixed 3d−3d metal cubanes.91 Little is still known about their WOC performance.92,93 They are attractive vectors for the transfer of doping and solid solution strategies94−96 from heterogeneous WOC design to the molecular level. Nickel centers, for example, frequently promote the performance of heterogeneous Co-oxide WOCs (cf. Table S8).97,98 We here introduce the first Co/Ni mixed cubane WOC series [Co II x Ni 4−x (dpy{OH}O) 4 (OAc) 2 (H 2 O) 2 ](ClO 4 ) 2 (CoxNi4−xO4-dpk). This cubane series newly puts the wellknown synergism of Co/Ni (as well as other heterometallic combinations) in heterogeneous WOCs to the test for molecular systems, in search of overarching concepts for WOC construction. In the following, we first discuss performance, stability, ligand exchange behavior, spectroscopy, and in situ XAS monitoring of Co4O4-dpk under photocatalytic and chemical water oxidation conditions. Next, we present computational modeling of the behavior of Co4O4-dpk in solution, followed by a detailed structural and photocatalytic performance discussion of the CoxNi4−xO4-dpk series.

Article

RESULTS AND DISCUSSION

Structure and Analytical Characterization of Co4O4dpk. Co4O4-dpk was newly synthesized in a straightforward and versatile one-pot reaction from cobalt acetate, di(2-pyridyl) ketone (dpk), and sodium perchlorate in water at room temperature (for experimental details cf. Supporting Information, SI).99 The key structural feature of Co4O4-dpk is the dicobalt {H2O-Co2(OR)2-OH2} motif, which is widely considered to represent the common mimimal requirement for WOC activity of cobalt oxides (Figure 1).66 Two additional Co centers of Co4O4-dpk complete the basic cubic {Co2(H2O)2Co2O4} building block of cobalt oxide surfaces (Scheme 1, Figure 1, and Figure S1).

Figure 1. Simulated (red) and experimental (black) PXRD patterns of Co4O4-dpk (inset: crystal structure of Co4O4-dpk, anions, and ligand hydrogen atoms are omitted for clarity; Co: dark blue, C: gray, O: red, N: green, H: white).

Co4O4-dpk (S.G. C2/c, a = 22.8068(9) Å, b = 12.1904(4) Å, c = 21.0805(7) Å; β = 115.750(5)°) is isostructural with the previously reported [Ni4(dpy{OH}O)4(OAc)2(H2O)2](ClO4)2 cubane (for a detailed description cf. SI).99 The cubic core contains two Co atom types, Co1 and Co1a, at a distance of 3.165 Å, which are coordinated by aqua and acetate ligands as Scheme 1. Structural Relation between Co4O4-dpk and the Dicobalt Sites Terminating a Typical Cobalt Oxide Catalyst

14199


Article


properties of the complex (Figures S12 and S13). None of the observed resonances could be assigned to the water ligands attached directly to cobalt centers, neither with 2H labeling experiments nor based on T1 time measurements. The absence of the water signals is most probably a consequence of exceeding paramagnetic broadening effects onto their line shape. Nevertheless, the two resonances at highest field (−62.2 and −115.6 ppm) can be attributed to two different OH sites, as they disappear in less than 1 min after addition of one drop of D2O due to proton−deuterium exchange (Figure S13). Furthermore, these two 1H signals have comparable T1 times (0.7 and 0.9 ms, respectively) and thus belong to hydrogen sites with similar distances to cobalt. We assigned the resonances at −62.2 and −115.6 ppm to the hydroxyl protons of the dpy{OH}O ligands (Figure 3).

well as the hydrolyzed dpk ligands (dpy{OH}O, Figure S1). Co2/Co2a (Co2−Co2a distance 3.072 Å) are linked to three nitrogen atoms and three bridging oxygen atoms of the dpy{OH}O ligand, respectively. The oxygen atoms of the two aqua ligands are 3.022 Å apart. Powder X-ray diffraction (PXRD) patterns of Co4O4-dpk vs calculated patterns from single-crystal X-ray diffraction indicate its high phase purity (Figure 1), together with elemental analysis data (cf. SI). Additionally, Co4O4-dpk was characterized by high-resolution electrospray-ionization mass spectrometry (HR-ESIMS), FT-IR, and UV/vis spectroscopy, as well as thermogravimetric (TG) measurements (cf. SI). The HR-ESI-MS spectrum shows the main ion peak of Co4O4-dpk at m/z 579.01252 in methanol (Figure S3), which can be assigned to the [Co4(dpy{OH}O)4(OAc)2]2+ fragment after the loss of two aqua ligands from [Co4(dpy{OH}O)4(OAc)2(H2O)2]2+. The FT-IR spectrum of Co4O4-dpk is similar to its previously reported Ni-containing analogue99 (cf. Figure S4 for a detailed band assignment). The UV/vis spectrum shows the characteristic 4T1g(P) → 4T1g(ν3) transitions of octahedrally coordinated CoII centers at λmax = 490−516 nm in borate buffer solution (Figure S5). The pKa value for the deprotonation of the entire cubane was determined as 8.5 by pH-dependent UV/vis titration (Figure S6), which is in line with pH-dependent cyclic voltammetry results (Figures S7−S9). The Pourbaix diagram in the pH range 7−9 confirms the above pKa value around 8.5 (Figure S9). In order to investigate the dissociation of the acetate ligand in aqueous solution with FT-IR spectroscopy, Co4O4-dpkcontaining deuterated acetate (Co4O4-dpk(OAc-d3)) was synthesized (details cf. SI), and its phase purity was confirmed with PXRD patterns (Figure S11). In situ FT-IR spectra did not change significantly when D2O was added into a CD3CN solution of Co4O4-dpk(OAc-d3) (Figure 2). Absence of the

Figure 3. (a) T1 relaxation times vs distance of the proton to its nearest neighbor cobalt atom, (b) chemical shift vs distance of the proton to its nearest neighbor cobalt atom (open squares: aromatic protons, dpk−OH (1) = proton of the upper −OH group, dpk-OH (2) = lower −OH groups). Figure 2. FT-IR spectra of Co4O4-dpk(OAc-d3) (40 mM) in CD3CN and CD3CN/D2O mixtures with 1, 2, 4, 6, and 8% (v/v) D2O.

Besides the acetate resonance at 47.2 ppm (T1 = 6.6 ms), the other 16 proton signals in the range between −7.9 and +313.6 ppm have been assigned to four NMR spectroscopically distinguishable pyridine sites, each of them comprising four different proton positions with respect to the {Co4O4} core (Figure 3). The four resonances with the highest paramagnetic shifts of 174.6, 200.0, 241.8, and 313.6 ppm have the shortest T1 times (0.7, 0.7, 0.5, and 0.5 ms) and arise from the aromatic hydrogen sites in ortho position to the nitrogen, since these are next to the cobalt centers. One group of four proton signals with medium T1 times of 6.2, 6.3, 6.3, and 5.2 ms (at 20.0, 21.7, 34.7, and 48.5 ppm) and another group with comparable T1 times of 10.6, 10.6, 8.4, and 7.8 ms (at 23.8, 40.8, 61.4, and 72.4

characteristic absorption of the free acetic acid at ca. 1700 cm−1 (Figure 2) indicates that the acetate ligands of Co4O4-dpk do not dissociate in aqueous solution. The main m/z 579.01252 ([Co4(dpy{OH}O)4(OAc)2]2+) peak in the ESI-MS spectrum also points to a rather stable coordination of the acetate ligands to the Co(II) centers. The 1H NMR spectrum of Co4O4-dpk in CD3CN shows 19 resonances in the range between −116 and +314 ppm, which were assigned by aid of 1H/2H exchange experiments and 1HT1 time measurements, because integration and chemical shift considerations were not possible due to the paramagnetic 14200


Article


structurally related cubane [Co4(dpy{OH}O)4(OAc)3(H2O)](ClO4) (Figure S16).100 A high TOF value of 1.2 s−1 within the same order of magnitude as reported for selected Co-cubane WOCs, e.g., [Co4(hmp)4(μ-OAc)2(μ2-OAc)2(H2O)] (60 μM, 7 s−1)59 and [Co3Ho(hmp)4(OAc)5(H2O)] (42 μM, 9.5 s−1)60 can be obtained with a relatively low Co4O4-dpk concentration of 12.5 μM. This value is still higher than those reported for representative tetranuclear WOCs without cuboidal cores, namely, K10.2Na0.8[{Co4(μ-OH)(H2O)3}(Si2W19O70)] (10 μM, 0.1 s−1),101 Li10[{Ru4O4(OH)2(H2O)4}(γ-SiW10O36)2] (4.34 μM, 0.125 s−1),102 and Rb8K2[{Ru4O4(OH)2(H2O)4}(γ-SiW10O36)2] (5 μM, 0.8 s−1).103 Stability of Co4O4-dpk under Photocatalytic Conditions. In order to confirm the stability of Co4O4-dpk in borate media, it was directly synthesized from a concentrated pH 8.5 borate buffer solution (for details cf. SI), wherein the asformed crystals were kept in the mother liquid for 1 week. PXRD patterns and FT-IR spectra of Co4O4-dpk obtained from borate buffer media agreed well with data of the pure product isolated from aqueous solution (Figure S18). Furthermore, time-dependent UV/vis spectra of Co4O4-dpk in pH 8.5 borate buffer solution displayed no absorption changes over 2 h (Figure S5). Neither did FT-IR spectra show obvious absorption changes for Co4O4-dpk in pH 8.5 borate buffer over 2 h (Figure S19). Therefore, Co4O4-dpk can be considered stable in borate buffer media over the investigated catalytic time scale. Definite exclusion of possible CoOx particle formation in postphotocatalytic solutions by dynamic light scattering (DLS) is exacerbated by assay-related issues such as traces of insoluble buffer compounds or the formation of poorly soluble [Ru(bpy)3]2+ salts. As the removal of such background impurities by filtration prior to DLS tests may as well eliminate the sought-after CoOx traces, we developed a flexible filtrationDLS protocol to demonstrate that catalysis in the present Co4O4-dpk system is exclusively of molecular origin. First, a standard Co4O4-dpk postphotocatalytic solution was filtered through a 200 nm filter (i.e., among the smallest available pore sizes). The pH value of the filtrate was readjusted to pH 8.5 with solid sodium borate, followed by a recycling test after addition of 9.5 mg of Na2S2O8 (5 mM). Comparable O2 evolution to the standard first recycling procedure of unfiltered Co4O4-dpk solution was observed (Figure S21(a)). Importantly, no particles were detected by DLS (Figure S22) in the as-obtained recycled and unfiltered solution. In sharp contrast, no O2 evolution was observed from a postcatalytic filtered and recycled solution of Co(OAc)2 (Figure S21(b)). We thus conclude that Co4O4-dpk-assisted oxygen evolution is a homogeneous process and that no significant CoOx particles were formed after the first two runs. Next, the application of 40 mg of Chelex resin to Co4O4-dpk during photocatalysis did not significantly influence O2 evolution (Figure S23(a)). In comparison, an equimolar Co2+ solution lost ca. 66% of O2 evolution activity in the presence of equal Chelex amounts (Figure S23(b)). This major difference indicates that no significant Co2+ quantities leached from Co4O4-dpk during photocatalysis. Furthermore, the orange solid mixture obtained from a lyophilized postphotocatalytic solution of Co4O4-dpk, consisting of [Ru(bpy)3]2+-based substances, Na2SO4, H3BO3Na2B2O7, and the Co-containing WOC, was separated. Most [Ru(bpy)3]2+-containing substances were removed with 3-fold butanol extraction (3 mL per extraction step). The remnant

ppm) are attributed to the aromatic meta positions with respect to the nitrogen center. Finally, the four resonances at −7.9, 3.7, 10.2, and 18.7 ppm which show the longest T1 times (19.8, 19.5, 17.2, and 14.0 ms) are assigned to the para positions displaying the largest distances to cobalt. Such clear-cut 1H NMR peak assignments have rarely been reported for paramagnetic Co(II) complexes, and they represent an outstanding new example to the best of our knowledge. Photocatalytic Oxygen Evolution Performance of Co4O4-dpk. The visible light-driven photocatalytic water oxidation activity of Co4O4-dpk was evaluated with a standard assay employing [Ru(bpy)3]Cl2 as photosensitizer (PS) and Na2S2O8 as sacrificial electron acceptor (Figures 4 and S14−

Figure 4. Clark electrode kinetics of visible light-driven water oxidation catalyzed by 100 μM Co4O4-dpk in 80 mM phosphate buffer solution (pH 7.0) and in 80 mM borate buffer solution (pH 8.0, 8.5, and 9.0; conditions: 470 nm LED, 1 mM [Ru(bpy)3]Cl2, 5 mM Na2S2O8; cf. also Figure S15(b)).

S17, Tables S4−S6). O2 evolution was monitored with both gas chromatography (GC) and Clark electrode techniques. The catalytic activity of Co4O4-dpk concentrations in the range from 12.5 to 200 μM was investigated in pH 8.5 borate buffer (80 mM, Figure S14). Co4O4-dpk showed the best performance at 100 μM with a turnover number (TON) of 20, a turnover frequency (TOF) of 0.24 s−1, and an O2 evolution yield of 80% (based on persulfate). The pH-dependent photocatalytic activity of Co4O4-dpk was determined from catalytic tests in different buffer types (Figure 4). The catalytic activity of Co4O4-dpk (100 μM) decreased slightly at pH 9.0 (80 mM borate buffer; TON 16.9, TOF 0.21 s−1, and 67.5% O2 yield). Even lower activity (TON 13.8, TOF 0.16 s−1, and 55% O2 yield) was observed at pH 8.0 (80 mM borate buffer), most likely due to the comparably lower buffer capacity of borate. Indeed the postcatalytic pH value dropped to 2.70, accompanied by a color change from orange-red to green. In line with previous studies on Co-WOCs,59 Co4O4-dpk (100 μM) displayed the lowest activity at pH 7.0 in phosphate buffer (80 mM, TON 3.0, TOF 0.08 s−1, 12% O2 yield), most likely due to phosphate blocking of the active site. In order to minimize unwanted cubane−buffer interactions,69 photocatalytic optimization tests were thus performed with different concentrations of borate buffer at pH 8.5 at an optimal working condition of 80 mM (Figure S15(a)). Co4O4-dpk displayed an approximately 20% higher photocatalytic activity than the 14201


Article

Journal of the American Chemical Society solid was extracted thrice with 3 mL of methanol in order to completely dissolve the molecular Co-based WOC. The white solid phase after methanol extraction was checked with ICP-MS for leached cobalt amounts arising from any potential Co4O4dpk decomposition, and only 0.02% Co was found. Reference tests with Co2+ trace amounts gave rise to max. 2.0 μmol of O2 evolution, compared to 16 μmol emerging from Co4O4-dpk (GC data). Note that this is a very conservative estimation of possible CoOx contributions, because the remnant solid may still contain Co4O4-dpk moieties which are insoluble in methanol. X-ray Absorption Spectroscopy of Co4O4-dpk under Operational Conditions. The solution Co K-edge spectrum of Co4O4-dpk displays an edge of 7718.6 eV and a white line position of 7726.6 eV, which are in the same range as the values reported for our earlier discovered Co(II)-cubane WOCs59,60 (Figure 5). Addition of photosensitizer and Na2S2O8 to a buffered Co4O4-dpk solution in the dark did not lead to significant changes of the XANES spectra (Figure 5).

Figure 6. (a) In situ XANES spectra of 100 μM Co4O4-dpk in pH 8.5, 80 mM borate buffer solution; (b) time-dependent white line position of 100 μM Co4O4-dpk under photocatalytic conditions.

Figure 5. XANES spectra of 100 μM Co4O4-dpk in pH 8.5, 80 mM borate buffer solution before and after addition of [Ru(bpy)3]Cl2 (1 mM) and Na2S2O8 (5 mM).

Upon illumination, however, the XANES spectrum of Co4O4-dpk underwent an immediate blue shift of the edge position and of the white line (Figures 6 and S31). This clearly indicated an oxidation of the Co4O4-dpk catalyst. The blue shift reached its maximum after ca. 15 min due to the different dimensions of the open jet setup applied for in situ measurements, which may give rise to less homogeneous irradiation of the sample compared to the smaller ex situ test vials (Figure 6 and SI). The slowly proceeding shift of the white line back to lower energy values indicates a partial reduction of Co4O4-dpk. However, monitoring the full recovery was beyond the allotted time scale of the experiment. In order to obtain structural information on Co4O4-dpk after illumination, EXAFS data were recorded and fitted against the crystallographic data (protons were omitted during the procedure). Reasonable fitting results with an r-factor of 1.6% were obtained (cf. Figure 7 and Table S7 for further details). The slightly shortened intermolecular distances compared to the crystallographic data point to the presence of remnant Co(III) centers with generally shorter bond lengths than

Figure 7. Fourier transform magnitudes of the EXAFS region for Co4O4-dpk (black line) vs the EXAFS fit (red line; dashed green line = fitting window): (a) before photocatalysis, (b) after photocatalysis.

Co(II) centers (Table S7). Obviously, part of the postcatalytic Co4O4-dpk was not yet completely recovered to the pristine all Co(II) state. Nonetheless, the EXAFS results in their entirety clearly support the presence of an intact cubane core in solution after the photocatalytic process. XANES Monitoring of Chemical Oxidation. The chemical oxidation process of Co4O4-dpk was monitored with different equivalents of [Ru(bpy)3]3+ both ex situ and with 14202


Article


does not contradict the above hypothesis that Co(II) centers may remain during oxygen evolution. Further Tracking of Co4O4-dpk-Assisted Photocatalytic O2 Evolution. Interferences of borate buffer anions with the catalytic edge-site of Co4O4-dpk were investigated with in situ FT-IR spectra. Pristine Co4O4-dpk was first monitored in CH3CN, followed by addition of 1 mL of borate buffer (240 mM) (Figure 9). Borate-induced blocking of the edge-site can

XANES spectroscopy in search of intermediate oxidation states. CV measurements demonstrate that [Ru(bpy)3]3+ (E([Ru(bpy)3]2+/3+)ox = ca. 1.1 V at pH 8.5) is capable of driving oxygen evolution with Co4O4-dpk (ca. 0.9 V in comparison, cf. Figure S10). UV/vis reference spectra show that some [Ru(bpy)3]3+ still remains after 1 min, thereby ensuring that the XANES spectra were recorded under operational conditions (Figure S27), despite the typical losses of [Ru(bpy)3]3+ through side reactions with the buffer.104 Ex situ experiments with corresponding stoichiometries (cf. Clark electrode data in Figures S25 and S26) demonstrate that oxygen evolution continues after 1 min, albeit with diffusion of oxygen into the headspace setting in. Most importantly, oxygen evolution with Co4O4-dpk only sets in above a threshold addition of 5 nominal equiv of [Ru(bpy)3]3+, i.e., not taking the unavoidable buffer-related losses into account. If all Co(II) centers had to be oxidized to the Co(III) state prior to O2 formation, a minimum of 8 nominal equiv of [Ru(bpy)3]3+ would have been required (i.e., 4 equiv for Co(II) oxidation and 4 equiv for O2 formation). This suggests that Co4O4-dpk can possibly evolve oxygen with some remnant Co(II) centers present. Keeping in mind that Co(IV)-containing XANES reference compounds remain hard to obtain,105 we synthesized 5nitrosalicylatotetraamminecobalt(IV) chloride nitrate monohydrate ([(NH3)4Co(IV)sal-NO2]ClNO3) from its ([(NH3)4Co(III)sal]Cl) precursor as one of the few storable Co(IV)containing reference complexes with a Co(III) analogue.106,107 1 H NMR spectra of [(NH3)4Co(IV)sal-NO2]ClNO3 agreed well with literature data (Figure S28).108 Unfortunately, XANES spectra and their first derivatives for both complexes (Figures S29 and S30) are quite closely related, so that a clear distinction between Co(IV) and Co(III) valence states was not possible. XAS data of samples quenched after 1 min at 77 K were recorded at 40 K using a He-cryostat (for further details cf. SI). All spectra show a blue shift of both absorption edge and white line (see Figure 8). Even in the presence of excess RuIII oxidant, the average valence state of Co4O4-dpk during oxygen evolution is always below the Co(III, IV) state (cf. tentative fit in Figure S32). This

Figure 9. FT-IR spectra of Co4O4-dpk (40 mM) in CH3CN (red) and CH3CN/borate buffer (black).

be excluded, because the FT-IR spectrum of Co4O4-dpk remained basically unchanged. Synthesis of pure Co4O4-dpk from a concentrated borate buffer solution (cf. above) further confirms that borate does not coordinate to the {Co2(OR)2(OH2)2} edge-site. In contrast, two new absorption bands were observed at 1548 and 1306 cm−1 in the presence of phosphate buffer (Figure S24) in addition to the characteristic cubane cluster bands. Given that the aqua ligands of Co4O4dpk are most likely to be substituted first, this newly suggests an interaction between the edge-site and phosphate anions. 18 O isotope-labeling experiments were carried out using a 10.8% H218O enriched borate buffer to check for water as exclusive oxygen source of Co4O4-dpk-catalyzed O2 evolution. This was confirmed through the presence of the 16O18O and 18 18 O O peaks (Figures S33−S35) in the 18O-labeling measurements and the good agreement of calculated and measured ratios of 16O16O, 16O18O, and 18O18O (Figure S36). Synthesis, Structure, and Analytical Characterization of Co4−xNixO4-dpk. The recently discovered mixed cubane MnCo3O4(OAc)5py3 illustrates the need for further WOC studies of these highly interesting yet elusive targets.92,93 To the best of our knowledge, only a limited number of 3d−3d heterometallic cubane compounds have been reported to date with few Co/Ni representatives among them.109 We thus synthesized a series of Co and Ni heterometallic analogues of Co4O4-dpk, namely, [CoxNi4−x(dpy{OH}O)4(OAc)2(H2O)2](ClO4)2 (CoxNi4−xO4-dpk), from a modified protocol for Co4O4-dpk with Co(OAc)2 and Ni(OAc)2 precursors (cf. experimental procedure in the SI and Figure S37). The CoxNi4−xO4-dpk series is isostructural with Co4O4-dpk (Figure 10). Co:Ni ratios were determined through ICP-MS measurements as 2.80:1.20, 2.65:1.35, 2.05:1.95, and 1.15:2.85 for the CoxNi4−xO4-dpk series, respectively (cf. SI). Single-crystal Xray diffraction data indicate a disordered distribution of Co and Ni over the M sites of the cubic M4O4 core. Almost identical

Figure 8. Cobalt K-edge XANES spectra of 250 μM Co4O4-dpk chemically oxidized by various amounts of [Ru(bpy)3](ClO4)3 at 40 K. 14203


Article


Co2.65Ni1.35O4-dpk exhibit almost identical O2 evolution activities, both 20% lower than Co4O4-dpk. Co2.05Ni1.95O4dpk and Co1.15Ni2.85O4-dpk displayed 46% and 65% lower O2 evolution than Co4O4-dpk, respectively, and Ni4O4-dpk showed almost no photocatalytic activity. 18O isotope labeling experiments were carried out using an 8.3% H218O enriched borate buffer to confirm water as exclusive oxygen source for Co2.65Ni1.35O4-dpk as a representative example of the mixed cubane series (Figures S45−S48). The pKa value of Ni4O4-dpk was determined as 9.5 through UV/vis titration (Figure S49), while the pKa values of CoxNi4−xO4-dpk cubanes could not be determined. This may arise from CoxNi4−xO4-dpk being in fact a mixture of Co1Ni3O4-dpk, Co2Ni2O4-dpk, and Co3Ni1O4-dpk with different Co−Ni distributions in their respective cubic M4O4 cores. The resulting divergent CoxNi4−xO4-dpk pKa values within a titration aliqout would then render a proper titration impossible. Computational Study. To obtain insight into the structural behavior of Co4O4-dpk in solution, density functional theory based molecular dynamics (DFT-MD) simulations of the cubane in a box of 208 and 452 water molecules, respectively, were carried out (see SI for computational details). The six Co−Co distances were monitored during the DFT-MD run, and their distributions are shown in Figure 11.

Figure 10. PXRD patterns of the CoxNi4−xO4-dpk series vs calculated pattern from single-crystal X-ray diffraction data.

PXRD patterns of the CoxNi4−xO4-dpk series compared to Co4O4-dpk (Figure 10) confirm the phase purity of the mixed cubanes. The FT-IR spectra of the CoxNi4−xO4-dpk series are also identical with Co4O4-dpk, pointing to the same transition metal−ligand interactions in both cubane types (Figure S4). The UV/vis spectra of the CoxNi4−xO4-dpk series (Figures S38−S41) display a significant intensity decrease of the characteristic Co absorption band (490−516 nm) from Co2.65Ni1.35O4-dpk to Co1.15Ni2.85O4-dpk, which goes hand in hand with an increase of the characteristic Ni absorptions (388 and 624 nm). The HR-ESI-MS spectrum of Co 1.15 Ni 2.85 O 4 -dpk is compared to Ni4O4-dpk and Co4O4-dpk as a representative example in the following (Figure S42). The m/z = 577.01625 peak can be exclusively assigned to [58Ni4(CH3COO)2(C11H9N2O2)4]2+, while the ESI-MS spectrum of Co4O4-dpk does not display peaks at m/z = 577.01625 or m/z = 577.51777. Note that a given bulk composition of CoxNi4−xO4dpk can arise either from the corresponding macroscopic mixture of binary Co4O4-dpk and Ni4O4-dpk cubanes or from genuine heterometallic compounds with a mixed core. In case of the mixture scenario for Co1.15Ni2.85O4-dpk, the intensity ratio between m/z = 577.01557 and m/z = 577.51777 for Co1.15Ni2.85O4-dpk is supposed to be similar to the one of Ni4O4-dpk. In contrast, an inverse ratio between the m/z = 577.01557 and m/z = 577.51777 intensities was obtained, indicating that Co and Ni form a heterometallic M4O4 core, resulting in a relative increase of the m/z = 577.51777 peak intensity. The low intensity of the m/z = 577.01557 peak demonstrates that there are only small amounts of Ni4O4-dpk in Co1.15Ni2.85O4-dpk. This is in line with the calculated spectrum of Co1.15Ni2.85O4-dpk, and calculations demonstrate that all three possible cubane core types (i.e., Co 2 N i 2 O 4 : C o N i 3 O 4 : N i 4 O 4 = 2 . 5 : 7 : 2 ) c o e x i s t i n Co1.15Ni2.85O4-dpk. Neutron diffraction experiments are the next step to determine the precise crystallographic locations of Co and Ni. All in all, ESI-MS spectra confirm the main presence of Co−Ni heterometallic cubanes in Co1.15Ni2.85O4dpk, and related results were obtained for the remaining members of the CoxNi4−xO4-dpk series. The photocatalytic water oxidation activity of the CoxNi4−xO4-dpk series was compared with Co4O4-dpk for the same catalyst concentration (100 μM) under standard conditions (Table S9 and Figure S44). Co2.80Ni1.20O4-dpk and

Figure 11. Distribution of the Co−Co distances of Co4O4-dpk in a box of 208 water molecules (normal distribution is shown for all distances as a bold line).

Co4O4-dpk is sufficiently flexible to adapt to local changes of the solvation shell or the ligand sphere. Although the Co−Co distances show fluctuations during the DFT-MD simulation, integrity of the cubane core is maintained and no evidence for an opening of the cubane cage was found. In one of our previous works on lanthanide-containing Co(II)-based cubanes,60 we observed that opening of their cubane cage might be beneficial from a thermodynamic point of view.62 In the case of Co4O4-dpk, the tridentate dpk ligand prevents this in the first place. The presence of a closed Co4O4-dpk cage is further supported by the constant average volume enclosed by the cubane core Co4O4 (an irregular cube of ca. 9.3 Å3) during the simulation time (see Figure S51). While Co4O4-dpk was modeled under neutral conditions above, the catalytic tests were performed in aqueous borate buffer (pH 8.5), where a substantial fraction of Co4O4-dpk is expected to be deprotonated (overall pKa 8.5, Figure S6). Assuming that the cobalt-coordinated water molecules are the 14204


Article


Influence of the Cobalt Nuclearity. Inspired by the seminal rapid-scan FT-IR study on the relative photocatalytic kinetics and mechanisms of mono- and dinuclear sites on Co3O4 surfaces,64 we compared the photocatalytic performance of Co4O4-dpk to the structurally related mononuclear complex [CoIII(dpy{OH}O)2]ClO4 (mono-Co-dpk). The latter was obtained from a slightly varied literature protocol (cf. SI), and it was recently reported as a stable and efficient WOC under [Ru(bpy)3]2+/S2O82− test conditions.114 The O−O bond formation step for the {HO-CoIII(μ2-O)2CoIII-OH} edge site of Co3O4 was identified to be at least 10 times faster than for the isolated Co(III) surface sites. In line with the trends for Co3O4, the normalized TOF of Co4O4-dpk is generally higher than that of mono-Co-dpk over a wide concentration range (Figures S52 and S53). A comparison of the TOFs for Co4O4dpk and mono-Co-dpk is reasonable here, because a kinetic analysis (Figure S54) showed no evidence for different reaction orders over the relevant concentration range. The above spectroscopic, computational, and kinetic information in its entirety is expanded in ongoing studies toward a first model for the catalytic cycle of Co4O4-dpk.

most acidic ligands, a deprotonated species of Co4O4-dpk was therefore simulated to provide a deeper understanding of the ground state behavior of Co4O4-dpk under experimental conditions for forthcoming mechanistic studies. During the initial phase of the trajectory we observed a proton transfer from one Odpk atom of the dpk (i.e., dpy{OH}O) ligand closest to the edge-side to the adjacent oxygen atom of the cobalt-bound hydroxyl group (OOH). As a consequence of the deprotonation, the according Odpk atom of the ligand participates in hydrogen bonding with the other cobalt-bound water molecule (OH2O), which in turn also engages in additional hydrogen bonding with the solvent (see Figure 12). It becomes evident that under ambient conditions

■

CONCLUSIONS [Co 4(dpy{OH}O)4(OAc)2(H2O)2](ClO4)2 (Co4O4-dpk) is established as a stable molecular water oxidation catalyst that newly transfers the reactive edge-site motif of cobalt oxide WOCs to the molecular level. The hydrolyzed dpk and acetate ligands surrounding the {HO-Co2(OR)2-OH} site provide a dynamic bioinspired framework for multiple hydrogen-bonding interactions. Computational studies indicate that the dpy{OH}O ligand mediates proton transfer through this network with the edgesite aqua ligands. Moreover, the deprotonated dpy{OH}O ligands can exert an advantagoeus intramolecular base function during later stages of the catalytic cycle. 1H/2H exchange experiments support the proton acceptor function of the dpy{OH}O ligands and provide clear-cut evidence for the longterm stability of Co4O4-dpk, together with a wide variety of additional stability tests (DLS, different spectroscopic methods, recycling, extraction, chelation, etc.). Both DFT-MD simulations and EXAFS fits indicate that the Co4O4-dpk cubane cage remains closed in solution, and both FT-IR and HR-ESIMS data point to a stable coordination of the acetate ligands to the cubane core. Time-dependent XANES measurements under operational photocatalytic conditions demonstrate that the Co(II) centers of Co4O4-dpk are readily oxidized to Co(III) or higher oxidation states with rather slow recovery. The related edge positions of the Co(III)- and Co(IV)-containing molecular references were too close for a clear assignment. However, chemical oxidation with [Ru(bpy)3]3+ showed that complete conversion of the Co(II) centers to Co(III) does not seem to be mandatory for O2 evolution. Furthermore, the novel mixed CoxNi4−xO4-dpk series is a unique model system for heterometallic edge-site configurations. The observed sharp activity decrease of 20% from Co4O4-dpk to Co2.65Ni1.35O4-dpk and Co2.05Ni1.95O4-dpk, respectively, may indicate that an intramolecular oxygen coupling pathway is prevailing over a single-site route here, as previously observed for Co3O4 surfaces. Kinetic comparisons of Co4O4-dpk to the related mononuclear WOC [CoIII(dpy{OH}O)2]ClO4 support this hypothesis.

Figure 12. (a) Schematic representation of protonation states of the edge-site cubane (solvent, ligands, and the bottom half of the cubane are omitted for clarity); (b) distance between the edge-site oxygen atoms (Odpk, OOH, OH2O, and the two closest solvent molecules denoted as “solv1” and “solv2”) and the protons in between them. During the equilibration run (not shown here), the deprotonation of Odpk took place.

the deprotonation of the dpk ligand is favored over the one of the edge-site water ligands, because the versatile hydrogenbonding network is able to stabilize the deprotonated Odpk involving both solvent molecules as well as the adjacent water ligand. During later stages of the catalytic cycle, the acidity of the edge-site water ligands may increase, and the deprotonated dpk ligand can potentially act as an intramolecular base facilitating further steps of the catalytic cycle. Although the organic ligand framework of our previously studied {LnCo3O4} (Ln = Ho− Yb) cubanes did not display any suitable functional groups for participation in the water oxidation mechanism, we observed a proton transfer from the attacking water molecule to the hydroxyl ligands of the Ln centers.61,62 Previous works have clearly shown that the presence of an intramolecular base in close proximity to the active site of a catalyst can enhance its activity.110−112 The presence of the dpk ligand and the acetate ligands undergoing hydrogen bonding with the edge-site protons completes a unique biorelated environment providing exceptional flexibility for intramolecular proton transfer (Figure 12).113 14205

II


Article


(9) Liu, Z.-J.; Wang, X.-L.; Qin, C.; Zhang, Z.-M.; Li, Y.-G.; Chen, W.-L.; Wang, E.-B. Coord. Chem. Rev. 2016, 313, 94−110. (10) Asraf, M. A.; Younus, H. A.; Yusubov, M.; Verpoort, F. Catal. Sci. Technol. 2015, 5, 4901−4925. (11) Blakemore, J. D.; Crabtree, R. H.; Brudvig, G. W. Chem. Rev. 2015, 115, 12974−13005. (12) Karkas, M. D.; Verho, O.; Johnston, E. V.; Akermark, B. Chem. Rev. 2014, 114, 11863−12001. (13) Gimbert-Surinach, C.; Moonshiram, D.; Francas, L.; Planas, N.; Bernales, V.; Bozoglian, F.; Guda, A.; Mognon, L.; Lopez, I.; Hoque, M. A.; Gagliardi, L.; Cramer, C. J.; Llobet, A. J. Am. Chem. Soc. 2016, 138, 15291−15294. (14) Ishizuka, T.; Watanabe, A.; Kotani, H.; Hong, D.; Satonaka, K.; Wada, T.; Shiota, Y.; Yoshizawa, K.; Ohara, K.; Yamaguchi, K.; Kato, S.; Fukuzumi, S.; Kojima, T. Inorg. Chem. 2016, 55, 1154−1164. (15) Nakazono, T.; Parent, A. R.; Sakai, K. Chem. - Eur. J. 2015, 21, 6723−6726. (16) Schwarz, B.; Forster, J.; Goetz, M. K.; Yücel, D.; Berger, C.; Jacob, T.; Streb, C. Angew. Chem., Int. Ed. 2016, 55, 6329−6333. (17) Menendez Rodriguez, G.; Bucci, A.; Hutchinson, R.; Bellachioma, G.; Zuccaccia, C.; Giovagnoli, S.; Idriss, H.; Macchioni, A. ACS Energy Lett. 2017, 2, 105−110. (18) Huang, Z.; Luo, Z.; Geletii, Y. V.; Vickers, J. W.; Yin, Q.; Wu, D.; Hou, Y.; Ding, Y.; Song, J.; Musaev, D. G.; Hill, C. L.; Lian, T. J. Am. Chem. Soc. 2011, 133, 2068−2071. (19) Song, F.; Ding, Y.; Ma, B.; Wang, C.; Wang, Q.; Du, X.; Fu, S.; Song, J. Energy Environ. Sci. 2013, 6, 1170−1184. (20) Dau, H.; Limberg, C.; Reier, T.; Risch, M.; Roggan, S.; Strasser, P. ChemCatChem 2010, 2, 724−761. (21) Li, W.; He, D.; Sheehan, S. W.; He, Y.; Thorne, J. E.; Yao, X.; Brudvig, G. W.; Wang, D. Energy Environ. Sci. 2016, 9, 1794−1802. (22) Al-Oweini, R.; Sartorel, A.; Bassil, B. S.; Natali, M.; Berardi, S.; Scandola, F.; Kortz, U.; Bonchio, M. Angew. Chem., Int. Ed. 2014, 53, 11182−11185. (23) Soriano-López, J.; Musaev, D. G.; Hill, C. L.; Galán-Mascarós, J. R.; Carbó, J. J.; Poblet, J. M. J. Catal. 2017, 350, 56−63. (24) Sartorel, A.; Bonchio, M.; Campagna, S.; Scandola, F. Chem. Soc. Rev. 2013, 42, 2262−2280. (25) González-Flores, D.; Sánchez, I.; Zaharieva, I.; Klingan, K.; Heidkamp, J.; Chernev, P.; Menezes, P. W.; Driess, M.; Dau, H.; Montero, M. L. Angew. Chem. 2015, 127, 2502−2506. (26) Smith, P. F.; Kaplan, C.; Sheats, J. E.; Robinson, D. M.; McCool, N. S.; Mezle, N.; Dismukes, G. C. Inorg. Chem. 2014, 53, 2113−2121. (27) Du, P.; Kokhan, O.; Chapman, K. W.; Chupas, P. J.; Tiede, D. M. J. Am. Chem. Soc. 2012, 134, 11096−11099. (28) Klingan, K.; Ringleb, F.; Zaharieva, I.; Heidkamp, J.; Chernev, P.; Gonzalez-Flores, D.; Risch, M.; Fischer, A.; Dau, H. ChemSusChem 2014, 7, 1301−1310. (29) Zhang, C.; Chen, C.; Dong, H.; Shen, J.-R.; Dau, H.; Zhao, J. Science 2015, 348, 690−693. (30) Gerey, B.; Gouré, E.; Fortage, J.; Pécaut, J.; Collomb, M.-N. Coord. Chem. Rev. 2016, 319, 1−24. (31) Lin, P.-H.; Takase, M. K.; Agapie, T. Inorg. Chem. 2015, 54, 59− 64. (32) Scheurer, A.; Korzekwa, J.; Nakajima, T.; Hampel, F.; Buling, A.; Derks, C.; Neumann, M.; Joly, L.; Petukhov, K.; Gieb, K.; Müller, P.; Kuepper, K.; Meyer, K. Eur. J. Inorg. Chem. 2015, 2015, 1892−1901. (33) Davis, K. M.; Palenik, M. C.; Yan, L.; Smith, P. F.; Seidler, G. T.; Dismukes, G. C.; Pushkar, Y. N. J. Phys. Chem. C 2016, 120, 3326− 3333. (34) Liao, R.-Z.; Siegbahn, P. E. M. J. Photochem. Photobiol., B 2015, 152, 162−172. (35) Lin, P.-H.; Tsui, E. Y.; Habib, F.; Murugesu, M.; Agapie, T. Inorg. Chem. 2016, 55, 6095−6099. (36) Berardi, S.; La Ganga, G.; Natali, M.; Bazzan, I.; Puntoriero, F.; Sartorel, A.; Scandola, F.; Campagna, S.; Bonchio, M. J. Am. Chem. Soc. 2012, 134, 11104−11107. (37) McCool, N. S.; Robinson, D. M.; Sheats, J. E.; Dismukes, G. C. J. Am. Chem. Soc. 2011, 133, 11446−11449.

Co4O4-dpk and the CoxNi4−xO4-dpk series provide unique opportunities to explore transfer options for synthetic paradigms from heterogeneous to molecular WOC design, such as preferred short-range motifs and heterometallic synergisms. The performance trend among mixed Co/Ni cubanes differs considerably from reported Co/Ni oxide electrocatalysts. The underlying mechanistic pathways are now explored in detail, on the way to comprehensive insight into the general applicability of surface- and compositionrelated WOC construction guidelines.

■

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b07361. Detailed synthetic and catalytic protocols, stability tests, analytical investigations, and computational methods and results (PDF) Crystallographic data (CIF)

■

AUTHOR INFORMATION

Corresponding Author

*[email protected] ORCID

Greta R. Patzke: 0000-0003-4616-7183 Notes

The authors declare no competing financial interest. Atomic coordinates have been deposited in the Cambridge Structural Database, Cambridge Crystallographic Data Centre, http://www.ccdc.cam.ac.uk, CCDC 1574034.

■

ACKNOWLEDGMENTS The work has been supported by the University Research Priority Program “Solar Light to Chemical Energy Conversion” (LightChEC) and by the Swiss National Science Foundation (Sinergia Grant No. CRSII2_160801/1). S.L. thanks the Swiss National Science Foundation (grant no. PP00P2_170667) for financial support and the Swiss National Supercomputing Center (project ID: s502 and s745) for computing resources. We thank Prof. Dr. Anthony Linden and Dr. Olivier Blacque (Department of Chemistry, University of Zurich) for crystallographic discussions. We are grateful to PD Dr. Laurent Bigler (Department of Chemistry, University of Zurich) for support and discussions concerning HR-ESI-MS measurements. We thank the PSI Swiss Light Source for granting beamtime. Support from NCCR MARVEL, NCCR MUST, and Energy System Integration (ESI) platform at PSI is also acknowledged.

■

REFERENCES

(1) Su, J.; Vayssieres, L. ACS Energy Lett. 2016, 1, 121−135. (2) El-Khouly, M. E.; El-Mohsnawy, E.; Fukuzumi, S. J. Photochem. Photobiol., C 2017, 31, 36−83. (3) Umena, Y.; Kawakami, K.; Shen, J.-R.; Kamiya, N. Nature 2011, 473, 55−60. (4) Vinyard, D. J.; Brudvig, G. W. Annu. Rev. Phys. Chem. 2017, 68, 101−116. (5) Tanaka, A.; Fukushima, Y.; Kamiya, N. J. Am. Chem. Soc. 2017, 139, 1718−1721. (6) Barber, J. Biochemistry 2016, 55, 5901−5906. (7) Han, X.-B.; Zhang, Z.-M.; Zhang, T.; Li, Y.-G.; Lin, W.; You, W.; Su, Z.-M.; Wang, E.-B. J. Am. Chem. Soc. 2014, 136, 5359−5366. (8) Wei, J.; Feng, Y.; Zhou, P.; Liu, Y.; Xu, J.; Xiang, R.; Ding, Y.; Zhao, C.; Fan, L.; Hu, C. ChemSusChem 2015, 8, 2630−2634. 14206


Article

Journal of the American Chemical Society (38) La Ganga, G.; Nardo, V. M.; Cordaro, M.; Natali, M.; Vitale, S.; Licciardello, A.; Nastasi, F.; Campagna, S. Dalton Trans. 2014, 43, 14926−14930. (39) Ullman, A. M.; Liu, Y.; Huynh, M.; Bediako, D. K.; Wang, H.; Anderson, B. L.; Powers, D. C.; Breen, J. J.; Abruna, H. D.; Nocera, D. G. J. Am. Chem. Soc. 2014, 136, 17681−17688. (40) Zhou, X.; Li, F.; Li, H.; Zhang, B.; Yu, F.; Sun, L. ChemSusChem 2014, 7, 2453−2456. (41) La Ganga, G.; Puntoriero, F.; Campagna, S.; Bazzan, I.; Berardi, S.; Bonchio, M.; Sartorel, A.; Natali, M.; Scandola, F. Faraday Discuss. 2012, 155, 177−190. (42) Genoni, A.; La Ganga, G.; Volpe, A.; Puntoriero, F.; Di Valentin, M.; Bonchio, M.; Natali, M.; Sartorel, A. Faraday Discuss. 2015, 185, 121−141. (43) Chen, H.-C.; Reek, J. N. H.; Williams, R. M.; Brouwer, A. M. Phys. Chem. Chem. Phys. 2016, 18, 15191−15198. (44) Smith, P. F.; Hunt, L.; Laursen, A. B.; Sagar, V.; Kaushik, S.; Calvinho, K. U. D.; Marotta, G.; Mosconi, E.; de Angelis, F.; Dismukes, G. C. J. Am. Chem. Soc. 2015, 137, 15460−15468. (45) Nguyen, A. I.; Ziegler, M. S.; Ona-Burgos, P.; SturzbecherHohne, M.; Kim, W.; Bellone, D. E.; Tilley, T. D. J. Am. Chem. Soc. 2015, 137, 12865−12872. (46) Wang, Y.; Li, F.; Zhou, X.; Yu, F.; Du, J.; Bai, L.; Sun, L. Angew. Chem., Int. Ed. 2017, 56, 6911−6915. (47) Swierk, J. R.; Mallouk, T. E. Chem. Soc. Rev. 2013, 42, 2357− 2387. (48) Zhang, B.; Li, F.; Yu, F.; Wang, X.; Zhou, X.; Li, H.; Jiang, Y.; Sun, L. ACS Catal. 2014, 4, 804−809. (49) Yuan, H.; Newton, D. L.; Seymour, L. A.; Metz, A.; Cropek, D.; Holder, A. A.; Ofoli, R. Y. Catal. Commun. 2014, 56, 76−80. (50) Li, X.; Clatworthy, E. B.; Masters, A. F.; Maschmeyer, T. Chem. Eur. J. 2015, 21, 16578−16584. (51) Wang, Y.; Li, F.; Li, H.; Bai, L.; Sun, L. Chem. Commun. 2016, 52, 3050−3053. (52) Ye, S.; Chen, R.; Xu, Y.; Fan, F.; Du, P.; Zhang, F.; Zong, X.; Chen, T.; Qi, Y.; Chen, P.; Chen, Z.; Li, C. J. Catal. 2016, 338, 168− 173. (53) Guedes, G. P.; Soriano, S.; Comerlato, N. M.; Speziali, N. L.; Lahti, P. M.; Novak, M. A.; Vaz, M. G. F. Eur. J. Inorg. Chem. 2012, 2012, 5642−5648. (54) Xie, W.-F.; Guo, L.-Y.; Xu, J.-H.; Jagodič, M.; Jagličić, Z.; Wang, W.-G.; Zhuang, G.-L.; Wang, Z.; Tung, C.-H.; Sun, D. Eur. J. Inorg. Chem. 2016, 2016, 3253−3261. (55) Lee, S. Y.; Kim, N.; Lee, M. M.; Jo, Y. D.; Bae, J. M.; Hyun, M. Y.; Yoon, S.; Kim, C. Dalton Trans. 2016, 45, 1727−1736. (56) Klinke, F. J.; Das, A.; Demeshko, S.; Dechert, S.; Meyer, F. Inorg. Chem. 2014, 53, 2976−2982. (57) Gao, Q.; Qin, Y.; Chen, Y.; Liu, W.; Li, H.; Wu, B.; Li, Y.; Li, W. RSC Adv. 2015, 5, 43195−43201. (58) Kuriyama, Y.; Kikukawa, Y.; Suzuki, K.; Yamaguchi, K.; Mizuno, N. Chem. - Eur. J. 2016, 22, 3962−3966. (59) Evangelisti, F.; Güttinger, R.; Moré, R.; Luber, S.; Patzke, G. R. J. Am. Chem. Soc. 2013, 135, 18734−18737. (60) Evangelisti, F.; Moré, R.; Hodel, F.; Luber, S.; Patzke, G. R. J. Am. Chem. Soc. 2015, 137, 11076−11084. (61) Hodel, F. H.; Luber, S. ACS Catal. 2016, 6, 1505−1517. (62) Hodel, F. H.; Luber, S. ACS Catal. 2016, 6, 6750−6761. (63) Wang, L.-P.; van Voorhis, T. J. Phys. Chem. Lett. 2011, 2, 2200− 2204. (64) Zhang, M.; de Respinis, M.; Frei, H. Nat. Chem. 2014, 6, 362− 367. (65) Plaisance, C. P.; van Santen, R. A. J. Am. Chem. Soc. 2015, 137, 14660−14672. (66) Ullman, A. M.; Brodsky, C. N.; Li, N.; Zheng, S.-L.; Nocera, D. G. J. Am. Chem. Soc. 2016, 138, 4229−4236. (67) Gardner, G.; Al-Sharab, J.; Danilovic, N.; Go, Y. B.; Ayers, K.; Greenblatt, M.; Dismukes, G. C. Energy Environ. Sci. 2016, 9, 184− 192.

(68) Hutchings, G. S.; Zhang, Y.; Li, J.; Yonemoto, B. T.; Zhou, X.; Zhu, K.; Jiao, F. J. Am. Chem. Soc. 2015, 137, 4223−4229. (69) Li, X.; Clatworthy, E. B.; Bartlett, S.; Masters, A. F.; Maschmeyer, T. J. Phys. Chem. C 2017, 121, 11021−11026. (70) Aiso, K.; Takeuchi, R.; Masaki, T.; Chandra, D.; Saito, K.; Yui, T.; Yagi, M. ChemSusChem 2017, 10, 687−692. (71) Bergmann, A.; Martinez-Moreno, E.; Teschner, D.; Chernev, P.; Gliech, M.; de Araujo, J. F.; Reier, T.; Dau, H.; Strasser, P. Nat. Commun. 2015, 6, 8625. (72) Risch, M.; Ringleb, F.; Kohlhoff, M.; Bogdanoff, P.; Chernev, P.; Zaharieva, I.; Dau, H. Energy Environ. Sci. 2015, 8, 661−674. (73) Shevchenko, D.; Anderlund, M. F.; Thapper, A.; Styring, S. Energy Environ. Sci. 2011, 4, 1284−1287. (74) Morales-Guio, C. G.; Liardet, L.; Hu, X. J. Am. Chem. Soc. 2016, 138, 8946−8957. (75) Yan, X.; Li, K.; Lyu, L.; Song, F.; He, J.; Niu, D.; Liu, L.; Hu, X.; Chen, X. ACS Appl. Mater. Interfaces 2016, 8, 3208−3214. (76) Grzelczak, M.; Zhang, J.; Pfrommer, J.; Hartmann, J.; Driess, M.; Antonietti, M.; Wang, X. ACS Catal. 2013, 3, 383−388. (77) Chen, Z.; Duan, Z.; Wang, Z.; Liu, X.; Gu, L.; Zhang, F.; Dupuis, M.; Li, C. ChemCatChem. 2017, DOI: 10.1002/ cctc.201700882. (78) Fernando, A.; Aikens, C. M. J. Phys. Chem. C 2015, 119, 11072− 11085. (79) Crandell, D. W.; Ghosh, S.; Berlinguette, C. P.; Baik, M.-H. ChemSusChem 2015, 8, 844−852. (80) Pham, H. H.; Cheng, M.-J.; Frei, H.; Wang, L.-W. ACS Catal. 2016, 6, 5610−5617. (81) Sarkar, S.; Kabir, M.; Greenblatt, M.; Saha-Dasgupta, T. J. Mater. Chem. A 2013, 1, 10422. (82) Li, X.; Siegbahn, P. E. M. J. Am. Chem. Soc. 2013, 135, 13804− 13813. (83) Davenport, T. C.; Ahn, H. S.; Ziegler, M. S.; Tilley, T. D. Chem. Commun. 2014, 50, 6326−6329. (84) Nguyen, A. I.; Wang, J.; Levine, D. S.; Ziegler, M. S.; Tilley, T. D. Chem. Sci. 2017, 6, 1676. (85) Stich, T. A.; Krzystek, J.; Mercado, B. Q.; McAlpin, J. G.; Ohlin, C. A.; Olmstead, M. M.; Casey, W. H.; Britt, R. D. Polyhedron 2013, 64, 304−307. (86) Brodsky, C. N.; Hadt, R. G.; Hayes, D.; Reinhart, B. J.; Li, N.; Chen, L. X.; Nocera, D. G. Proc. Natl. Acad. Sci. U. S. A. 2017, 114, 3855−3860. (87) Zahariou, G. Inorg. Chem. 2017, 56, 6105−6113. (88) Zhang, M.; Frei, H. Annu. Rev. Phys. Chem. 2017, 68, 209−231. (89) Zhang, M.; Frei, H. Catal. Lett. 2015, 145, 420−435. (90) Koroidov, S.; Anderlund, M. F.; Styring, S.; Thapper, A.; Messinger, J. Energy Environ. Sci. 2015, 8, 2492−2503. (91) Guedes, G. P.; Soriano, S.; Mercante, L. A.; Speziali, N. L.; Novak, M. A.; Andruh, M.; Vaz, M. G. F. Inorg. Chem. 2013, 52, 8309−8311. (92) Suseno, S.; McCrory, C. C. L.; Tran, R.; Gul, S.; Yano, J.; Agapie, T. Chem. - Eur. J. 2015, 21, 13420−13430. (93) Nguyen, A. I.; Suess, D. L. M.; Darago, L. E.; Oyala, P. H.; Levine, D. S.; Ziegler, M. S.; Britt, R. D.; Tilley, T. D. J. Am. Chem. Soc. 2017, 139, 5579−5587. (94) Dionigi, F.; Strasser, P. Adv. Energy Mater. 2016, 6, 1600621. (95) Ahn, H. S.; Bard, A. J. J. Am. Chem. Soc. 2016, 138, 313−318. (96) Burke, M. S.; Kast, M. G.; Trotochaud, L.; Smith, A. M.; Boettcher, S. W. J. Am. Chem. Soc. 2015, 137, 3638−3648. (97) Antony, R. P.; Satpati, A. K.; Bhattacharyya, K.; Jagatap, B. N. Adv. Mater. Interfaces 2016, 3, 1600632. (98) Zhao, Z.; Wu, H.; He, H.; Xu, X.; Jin, Y. Adv. Funct. Mater. 2014, 24, 4698−4705. (99) Efthymiou, C. G.; Raptopoulou, C. P.; Terzis, A.; Boča, R.; Korabic, M.; Mrozinski, J.; Perlepes, S. P.; Bakalbassis, E. G. Eur. J. Inorg. Chem. 2006, 2006, 2236−2252. (100) Tong, M.-L.; Zheng, S.-L.; Shi, J.-X.; Tong, Y.-X.; Lee, H. K.; Chen, X.-M. J. Chem. Soc., Dalton Trans. 2002, 1727−1734. 14207


Article

Journal of the American Chemical Society (101) Zhu, G.; Geletii, Y. V.; Kögerler, P.; Schilder, H.; Song, J.; Lense, S.; Zhao, C.; Hardcastle, K. I.; Musaev, D. G.; Hill, C. L. Dalton Trans. 2012, 41, 2084−2090. (102) Sartorel, A.; Carraro, M.; Scorrano, G.; Zorzi, R.; Geremia, S.; McDaniel, N. D.; Bernhard, S.; Bonchio, M. J. Am. Chem. Soc. 2008, 130, 5006−5007. (103) Geletii, Y. V.; Huang, Z.; Hou, Y.; Musaev, D. G.; Lian, T.; Hill, C. L. J. Am. Chem. Soc. 2009, 131, 7522−7523. (104) Vickers, J. W.; Sumliner, J. M.; Lv, H.; Morris, M.; Geletii, Y. V.; Hill, C. L. Phys. Chem. Chem. Phys. 2014, 16, 11942−11949. (105) Hadt, R. G.; Hayes, D.; Brodsky, C. N.; Ullman, A. M.; Casa, D. M.; Upton, M. H.; Nocera, D. G.; Chen, L. X. J. Am. Chem. Soc. 2016, 138, 11017−11030. (106) Yamamoto, Y.; Ito, K.; Yoneda, H.; Mori, M. Bull. Chem. Soc. Jpn. 1967, 40, 2580−2853. (107) Morgan, G. T.; Smith, J. D.; Main. J. Chem. Soc., Trans. 1922, 121, 1956−1971. (108) Yamamoto, Y. Bull. Chem. Soc. Jpn. 1978, 51, 2894−2897. (109) Elmehdawi, R.; EL-Kaheli, M.; Abuhmaiera, R.; Treish, F.; Ben Younes, M.; Bazzicalupi, C.; Guerri, A.; Caneschi, A.; Amjad, A. Materials 2017, 10, 178. (110) Ertem, M. Z.; Cramer, C. J. Dalton Trans. 2012, 41, 12213− 12219. (111) Wang, D.; Groves, J. T. Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 15579−15584. (112) Dogutan, D. K.; McGuire, R.; Nocera, D. G. J. Am. Chem. Soc. 2011, 133, 9178−9180. (113) Han, Z.; Horak, K. T.; Lee, H. B.; Agapie, T. J. Am. Chem. Soc. 2017, 139, 9108−9111. (114) Zhao, Y.; Lin, J.; Liu, Y.; Ma, B.; Ding, Y.; Chen, M. Chem. Commun. 2015, 51, 17309−17312.

14208


{Co4O4} and {CoxNi4âxO4} Cubane Water Oxidation Catalysts as

Recommend Documents

{Co4O4} and {CoxNi4âxO4} Cubane Water Oxidation Catalysts as