pubs.acs.org/JACS Article Surface-Promoted Evolution of Ru-bda Coordination Oligomers Boosts the Efficiency of Water Oxidation Molecular Anodes Marcos Gil-Sepulcre, Joachim O. Lindner, Dorothee Schindler, Lucía Velasco, Dooshaye Moonshiram, Olaf Rüdiger, Serena DeBeer, Vladimir Stepanenko, Eduardo Solano, Frank Würthner,* and Antoni Llobet* Cite This: J. Am. Chem. Soc. 2021, 143, 11651−11661 Read Online ACCESS * See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles. Metrics & More Article Recommendations sı Supporting Information ABSTRACT: A new Ru oligomer of formula {[RuII(bda-κ-N2O2)- (4,4′-bpy)]10(4,4′-bpy)}, 10 (bda is [2,2′-bipyridine]-6,6′-dicarbox- Downloaded via 158.109.58.130 on June 29, 2022 at 10:16:46 (UTC). ylate and 4,4′-bpy is 4,4′-bipyridine), was synthesized and thoroughly characterized with spectroscopic, X-ray, and electrochemical techni- ques. This oligomer exhibits strong affinity for graphitic materials through CH−π interactions and thus easily anchors on multiwalled carbon nanotubes (CNT), generating the molecular hybrid material 10@CNT. The latter acts as a water oxidation catalyst and converts to a new species, 10′(H2O)2@CNT, during the electrochemical oxygen evolution process involving solvation and ligand reorganization facilitated by the interactions of molecular Ru catalyst and the surface. This heterogeneous system has been shown to be a powerful and robust molecular hybrid anode for electrocatalytic water oxidation into molecular oxygen, achieving current densities in the range of 200 mA/cm2 at pH 7 under an applied potential of 1.45 V vs NHE. The remarkable long-term stability of this hybrid material during turnover is rationalized based on the supramolecular interaction of the catalyst with the graphitic surface. ■ INTRODUCTION The design of an efficient artificial photosynthetic device for their stability is highly dependent on the anchoring strategy. However, their performance is normally limited by the amount water splitting has been one of the hot topics during the past of active catalyst that can be deposited per surface area.18−20 The best molecular WOCs reported today are based on Ru decade for the scientific community because such devices can complexes containing equatorially coordinating pyridyl-carbox- provide access to clean and renewable energy sources as an ylate ligands and monodentate pyridyl groups completing the alternative to fossil fuels. This in turn could contribute to axial positions. In particular, [Ru(bda)(isoq)2] (bda is [2,2′- solving the problems of global warming and climate change bipyridine]-6,6′-dicarboxylate; isoq is 8-isoquinoline) and faced by our modern society today.1−4 Water splitting with [Ru(tda)(py)2] (tda is [2,2′:6′,2″-terpyridine]-6,6″-dicarbox- sunlight can be achieved using photoelectrochemical cells that ylate; py is pyridine), abbreviated from now on as Ru-bda and accomplish water oxidation at the anode and proton reduction Ru-tda, achieve turnover frequency (TOF) values in the range at the cathode. From the anode perspective, a key issue is the of 1000−50 000 s−1, respectively, in the homogeneous use of robust and efficient water oxidation catalysts (WOCs) phase.21−25 Interestingly, homologues of these molecular anchored on a conductive or semiconductive surface. This is a catalysts have been successfully anchored on surfaces, although challenging issue because of the considerably unfavorable their performances as WOCs are markedly different.26−28 thermodynamics and complex mechanism, which generally While Ru-tda shows a robust performance with current result in large activation energies associated with the water densities in the range of 5 mA/cm2,28 Ru-bda decomposes oxidation reaction.5 to form RuOx mainly due to the absence of a bimolecular O− Two main strategies have been employed up to now toward the development of efficient electro(photo) anodes based on the use of metal oxides (MOx) or molecular complexes as Received: May 7, 2021 WOCs deposited on the surface of a (photo)electrode.6−8 Published: July 22, 2021 Efficient electro(photo) anodes based on metal oxides as WOCs is mainly limited to pH 13 or above, and in general a large amount of MOx is needed.9,10 On the other hand, molecular catalysts can be used at a broader pH range11−17 and © 2021 The Authors. Published by American Chemical Society https://doi.org/10.1021/jacs.1c04738 11651 J. Am. Chem. Soc. 2021, 143, 11651−11661 Journal of the American Chemical Society pubs.acs.org/JACS Article O bond formation pathway (I2M mechanism; interaction of two M−O units) under restricted mobility conditions, that is, its low-energy pathway.29 Recently, a family of Ru-tda-based oligomers of general formula {[RuII(tda-κ-N3O)(4,4′-bpy)]n(4,4′-bpy)} (4,4′-bpy is 4,4′-bipyridine; n = 1, 2, 4, 5, and 15), abbreviated as (Ru- tda)15, has been prepared, isolated, and thoroughly charac- terized in homogeneous solution. These oligomers display an outstanding affinity for graphitic surfaces such as multiwalled carbon nanotubes (CNTs), thanks to multiple ligand-based CH−π interactions with the graphitic surfaces.30 The new molecular hybrid materials display an impressive water oxidation electrocatalytic activity with unprecedented current densities, for anchored molecular catalysts, in the range of 200−300 mA/cm2 at 1.45 V vs NHE at pH 7. The outstanding performance displayed by the Ru-tda oligomers anchored on graphitic surfaces inspired us to explore the performance of a new decameric Ru-bda oligomer, {[RuII(bda-κ-N2O2)(4,4′-bpy)]10(4,4′-bpy)}, 10, as an electro- catalyst for water oxidation as well as to explore its mechanism of action under restricted mobility conditions. It is important to keep in mind here that the monomeric Ru-bda complexes anchored to solid supports undergo decomposition to RuO2 under restricted mobility conditions as mentioned above.29 Toward this goal, we were encouraged by our recent successful research on anodes31 prepared from macrocyclic trimer Figure 1. Chemical structure and 1H NMR spectrum (500 MHz, 298 {[Ru(bda)]3(μ-bpb)3}, MC3 (bpb is 1,4-bi(pyridin-3-yl)- K, [d3]-TFE) of {[Ru(bda)(4,4′-bpy)]10(4,4′-bpy)} (10) oligomer. benzene), that changed the O−O bond formation mechanism The number of Ru units in the oligomer, N(Ru), was determined from I2M to water nucleophilic attack (WNA) by a from the integrals over H9 and H5 signals, I(H9) and I(H5). supramolecular effect and thereby circumvented the handicaps associated with the restricted mobility.32 We reasoned that the (see the SI). Furthermore, a DOSY experiment shows the combination of these two phenomena, i.e., on one hand the presence of only one species in solution and thus indicates that enhanced performance of the oligomers and on the other hand there is no oligodispersity (see Figure S6). An optimized the change of the O−O bond formation mechanism by structure of oligomer 10 in the gas phase was calculated using environmental perturbations, should lead to new interesting the PM6-D3H4 method to visualize its molecular structure and molecular hybrid materials. dimensions (see Figure 3). Indeed, our detailed experimental and theoretical studies Oligomer 10, as also the case for related homologues revealed that the newly synthesized Ru-bda linear oligomer containing the Ru-tda unit,30 has a strong affinity for graphitic {[RuII(bda-κ-N2O2)(4,4′-bpy)]10(4,4′-bpy)}, 10, can be suc- surfaces due to its multiple interaction sites. Indeed, upon cessfully anchored on CNTs via noncovalent CH−π addition of a reddish solution of 10 in TFE to a interactions, and the respective molecular hybrid anode tetrahydrofuran (THF) suspension containing CNTs at exhibited excellent catalytic performance in water oxidation, room temperature (RT), the solution immediately decolorizes, possessing outstanding stability. ■ clearly indicating adsorption of 10 onto the CNTs, forming the molecular hybrid material 10@CNT (Figure S4). Similarly, RESULTS AND DISCUSSION when a THF solution of 10 is in contact with a glassy carbon Synthesis, Characterization, and Graphitic Affinity of electrode disk (GCd; S = 0.07 cm2), it also strongly adsorbs at 10. The Ru-bda-based oligomer 10 (for structure see Figures the electrode, forming the material 10@GC. The presence of 1, top, and 3) was synthesized by refluxing an equimolar 10 anchored on the graphitic surfaces can be readily observed solution of [Ru(bda)(DMSO)2] and 4,4′-bpy in TFE−MeOH, using electrochemical techniques, as can be seen in Figures S1 9:1 (TFE is 2,2,2-trifluoroethanol), at 78 °C for 3 days (details and S2. The electrochemical performance of 10 anchored on are provided in the Supporting Information (SI)). The new graphitic supports was analyzed based on cyclic voltammetry oligomer was characterized by 1D and 2D NMR spectroscopy (CV), differential pulse voltammetry (DPV), and controlled (Figures 1 and S5), UV−vis (Figure S12), mass spectrometry potential electrolysis (CPE) techniques. All electrochemical (Figures S10 and S11), elemental analysis, resonance Raman experiments were carried out in 1 M phosphate buffer at pH 7 spectroscopy (Figures S33 and S34), powder X-ray diffraction and reported vs NHE. The CVs of 10@GC or 10@CNT@GC (Figure S13), and electrochemical techniques in organic (where 10@CNT is drop-casted at the GCd electrode; see the solvents (Figures S14 and S15). 1H NMR is particularly useful SI for details) display the typical redox waves associated with because it allows differentiating between resonances of the the III/II, IV/III, and V/IV couples that have been previously bridging and terminal 4,4′-bpy ligands and thus enables the described for the [Ru(bda)(pic)2], 1-pic,22 monomer in calculation of the number of repetitive units that turned out to homogeneous phase, although with smaller peak splitting due be 10 and thus is labeled as 10 (Figure 1). Thus, oligomer 10 to its restricted mobility (see SI, Figures S17, S20A, and S24 has a length of approximately 12.0 nm and a width of 0.9 nm and Tables S1 and S2). This confirms that the nature of the based on the typical bonding distances for Ru-bda complexes oligomeric species anchored on the graphitic surface is the 11652 https://doi.org/10.1021/jacs.1c04738 J. Am. Chem. Soc. 2021, 143, 11651−11661 Journal of the American Chemical Society pubs.acs.org/JACS Article suggesting the formation of less than a monolayer and thus excluding agglomeration of oligomers at the surface. Furthermore, X-ray absorption near edge structure (XANES) and extended X-ray absorption fine structure (EXAFS) techniques were applied in order to spectroscopically characterize the new hybrid materials. Figure 2 shows XANES spectra of 10 as a powder and anchored on CNTs, which in turn have been deposited on the surface of a glassy carbon plate (GCp) by drop-casting to give 10@CNT@GCp. The half-edge energies obtained revealed that the Ru center remains at oxidation state II before and after anchoring onto the CNTs. The simulated and experimental EXAFS spectra for both samples show the expected octahedral coordination environment surrounding the Ru center with two Ru−Nbda at 1.90−1.91 Å and four Ru−N/O at 2.07−2.08 Å, which are slightly elongated to Ru−N/O at 2.07 Å after anchoring to the CNTs (Tables S5−S7). These distances are also in proper agreement with the X-ray structure of the monomeric [Ru(bda)(pic)2]21 and with the calculated structures of monomeric model compound [Ru(bda)(py)2] (Figure S42) and oligomer 10 (Figure 3). This new material was also characterized by microscopy techniques including atomic force microscopy (AFM), scanning electron microscopy (SEM), and energy dispersive X-ray spectroscopy (EDX). The AFM experiments in Figure 4 Figure 4. Height (a) and phase (b) AFM images of 10@CNT. Z scale is 200 nm. The images show a fragment of a carbon nanotube covered Figure 2. (a) Normalized Ru K-edge XANES for 10 (black line), 10@ by thin rod-like particles of about 11−12 nm length. CNT@GCp (red line), 10′(H2O)2@CNT@GCp (cyan line), 10′(H2O)2@CNT@GCp after catalysis (magenta line), [Ru(tda)- (Py)2] (purple line), and RuO2 (green line). (b) Fourier transforms of show images of 10@CNT with the fragments of CNTs k2-weighted Ru EXAFS for 10 (black line), 10@CNT@GCp (red covered by thin rod-like particles with a length of ca. 11−12 line), 10′(H2O)2@CNT@GCp (cyan line), and 10′(H2O)2@CNT@ nm, which is within the 11.96 nm expected distance of 10 (see GCp after catalysis (magenta line) and RuO2 (green line). Inset: Back Figure 3 and Figure S2 in the SI). The oligomers are oriented Fourier transformed experimental (solid lines) and fitted (dashed parallel to one another on the CNT surface, and the distance lines) k2[χ(k)] of Ru complexes together with RuO2 (green). between two neighboring particles measured middle-to-middle Experimental spectra were calculated for k values of 2−11 Å−1. varies between 6.5 and 7.5 nm. Considering the tip-broadening effect for AFM with a nominal tip radius of 7 nm used for these same as in the homogeneous phase. The electrochemical experiments, we can assume that each rod-like nanoparticle experiments are also instrumental since they allow calculating corresponds to an oligomer of 10. Further combined SEM/ quantitatively the amount of sample deposited at the surface of EDX investigations of the thin 10@CNT film spin-coated onto the electrode as well as its surface coverage. For 10@GC, a a silicon wafer are also consistent with the presence of the surface coverage of 73% is obtained (see Figure S2 in the SI), oligomer anchored on the CNT (Figure S41). Figure 3. Space-filling structure of oligomer 10 in the gas phase calculated based on the PM6-D3H4 method. Color code: Ru, light blue; O, red; N, blue; C, gray; H, white. In this model the distance between the N atoms in the extremes of the oligomer (nonbonded to Ru) turns out to be 11.96 nm. 11653 https://doi.org/10.1021/jacs.1c04738 J. Am. Chem. Soc. 2021, 143, 11651−11661 Journal of the American Chemical Society pubs.acs.org/JACS Article As proposed earlier for a related homologue,30 a nearly perpendicular CH−π interactions between the 60 CH groups (H1, H2, and H3 in Figure 1, top, and their symmetric counterparts) of the 10 bda ligands in 10 with the delocalized π-system of the graphitic material are responsible for the strong affinity of 10 toward graphitic surfaces such as graphite and CNTs (see Figure S46 for a drawing of this interaction). The existence of this CH−π interaction was monitored in solution by 1H NMR spectroscopy using 1-pic and 1-pyrenebutanol (pyn-OH) as low molecular weight models of 10 and the graphitic surface, respectively. Figures S7−S9 show that when a d4-MeOD solution of 1-pic and pyn-OH are mixed, their aromatic resonances shift within the range of 0.01 ppm due to the mentioned interactions, whereas the aliphatic ones remain virtually unaltered. Grazing incident wide-angle X-ray scattering (GIWAXS) experiments were employed to characterize the interaction of 10 with graphitic surfaces. As can be observed in Figure 5 and Figure 6. Schematic representation of noncovalent interactions in a dimer model of 10@CNT@GC before (top) and after activation (bottom) with semiempirically calculated interaction energies per Ru center (ΔEi) in kJ/mol. (assigned to the V/IV) decreases, while the onset of the foot of the catalytic wave progressively increases, which is a clear indication that the initial active species is transformed into a new catalytic species (Figure S17). Furthermore, the non- catalytic waves that appear at potentials E1/2 = 0.73 V (III/II, ΔE = 28 mV) and E1/2 = 0.89 V (IV/III ΔE = 23 mV) similar Figure 5. (Left) 2D GIWAXS blank-subtracted scattering pattern of 10@GCp and (right) the corresponding qz 1D cut profile near the to the monomer disappear and a new set of redox couples specular direction. appears in the voltammogram at E1/2 = 0.62 V (III/II, ΔE = 34 mV), E1/2 = 0.70 V (IV/III, ΔE = 30 mV), and E1/2 = 0.88 V (V/IV ΔE = 39 mV), as shown in Figures S18, S19, and S21. Figure S38, 2D GIWAXS patterns obtained for 10 anchored on The redox potential of these couples was analyzed as a a flat glassy carbon plate electrode display an anisotropic broad function of pH, and a Pourbaix diagram was built, which is band associated with a preferential face on orientation of 10 on presented in Figure 7b. The diagram shows three consecutive GCp, also consistent with the results obtained from AFM and PCET events in the range of pH 2−12, with slopes close to the NMR and further supported by DFT calculations. The 1D cut expected ones for one electron and multiple proton transfer displayed on the right-hand side reveals an oligomer−surface processes associated with multiple Ru-aquo groups. Surpris- distance in the range of 2.8−6.2 Å (10−22 nm−1 in reciprocal ingly, the Pourbaix obtained is very similar to that of cis- space).30 These observations, together with the parallel [Ru(bpy)2(H2O)2]2+,36 with the same number of species but orientation of the monomers revealed by AFM, and a coverage with some slight differences in the redox potentials (see Figure below a monolayer strongly support that CH−π interactions S20B in the SI). Thus, we associate this transformation to the are responsible for the high affinity of 10 for graphitic surfaces. progressive formation of {cis-[RuII(H2O)2(bda-κ-NO)(4,4′- To further corroborate these experimental evidence and bpy)]10(4,4′-bpy)}, 10′(H2O)2, at the surface of the electrode complement them, we carried out density functional theory where the initial tetradentate bda ligands change their (DFT) as well as semiempirical calculations. In this work all coordination mode from κ-N2O2 to κ-NO with concomitant DFT calculations were performed using the B3LYP func- coordination of aquo ligands to the positions partially liberated tional,33−35 together with a proper implicit solvation model by the bda ligands (Figure 6, bottom). It is also interesting to and, where needed, explicit water molecules, as described in mention that this transformation only occurs during the detail in the SI. Figure 6 (top) shows an image of the electrocatalytic process at the graphitic surface and it does not calculated interactions of a dimeric Ru model {[RuII(bda-κ- take place at lower oxidation states (III/II or IV/III; see Figure N2O2)(py)]2(μ-4,4′-bpy)}, 2, anchored on a hexagonal S23 in the SI) nor in the homogeneous phase for monomeric graphene sheet surface, 2@GS, where the CH−π interactions homologue [Ru(bda)(pic)2],22 thus manifesting the key role of can be visualized. The resulting interaction between the surface the graphitic surface to drive this transformation. and the dimer is exergonic by 32.6 kJ/mol. We used an EPR spectroelectrochemical cell37 to follow the Surface-Promoted Transformation of 10@CNT to paramagnetic signature of 10 and 10′(H2O)2 as well as their 10′(H2O)2@CNT. The performance of 10@CNT was initially derived species immobilized on the electrode surface of a analyzed based on 50 repetitive CV scans in the potential range carbon paper (10@C-paper), at different oxidation states. The of 0.25−1.45 V at pH 7 (Figure 7a and Figure S16). After each main spectra obtained are displayed in Figure 8. Measurement scan the intensity of the catalytic wave at Eonset = 1.02 V of a freshly formed film (RuII-10@C-paper) reveals only an 11654 https://doi.org/10.1021/jacs.1c04738 J. Am. Chem. Soc. 2021, 143, 11651−11661 Journal of the American Chemical Society pubs.acs.org/JACS Article Figure 8. EPR spectra (upper panel, thick shaded traces) measured in situ before and after electrochemical activation (by running 50 CVs between 0.2 and 1.2 V), on a carbon paper electrode functionalized with 10 (10@C-paper); see the SI and Figure S37 for details. The spectra can be simulated with a combination of two main components, a signal centered around g = 2.0, assigned to the carbon paper (Figure S36), and the anisotropic Ru signals (present only after catalyst adsorption). Spectra were measured at 10 K in 0.1 phosphate buffer (pH 7.0). increase of anisotropy is consistent with the formation of {[RuIII(OH)2(bda-κ-NO)(4,4′-bpy)]10(4,4′-bpy)} species, RuIII-10′(OH)2@C-paper. It is interesting to note that for [Ru(bda)(py)2] complexes the species responsible for catalysis needs to reach oxidation state V. However, in cis-[Ru- (bpy)2(H2O)2]2+, the presence of two aqua ligands bonded to the Ru aqua center stabilizes the Ru(V) oxidation state and water oxidation catalysis occurs only at the formal oxidation state VI.38 This allows the spectroscopic characterization of the Ru(V) species before catalysis occurs.39 Similarly, the oxidation of RuIII-10′(OH)2@C-paper at 1.1 V generates the RuV-10′(O)2@C-paper species (g = [2.07, 2.00, 1.88]), the EPR spectrum of which is reported in Figure 8. These g- factors are in good agreement with previously reported RuIII Figure 7. (a) CV evolution of 10@CNT@GC (Γ10 = 2.3 nmol·cm−2) and RuV complexes.39−41 to form 10′(H2O)2@CNT@GC (Γ10′(H2O)2 = 1.6 nmol·cm−2) during The proposed transformation was further analyzed by DFT 50 repetitive CV cycles in 1 M phosphate buffer (pH 7) at a scan rate of 100 mV/s. The black line corresponds to the first cycle, the gray calculations32b,35 where several options were considered. DFT line to the second to 49th cycles, and the red line to the 50th cycle. discarded both a seven-coordinated Ru species and κ-N2- Inset: enlargement of the noncatalytic redox waves. (b) Pourbaix coordinated bda in 10′(H2O)2@CNT, since the respective diagram of 10′(H2O)2@CNT@GC. minima on the potential energy surface are much higher than for bda-κ-NO coordination (Table S9) at the Ru(V) oxidation state, where the crucial transformation step occurs according to EPR spectrum identical to when the bare carbon paper is CV. Indeed, the addition of two water molecules to Ru(II) is measured, with a sharp signal assigned to radical species on C- strongly endergonic, with a calculated free energy difference of paper (Figure S36). 101.7 kJ/mol, while a value of only 9.1 kJ/mol is obtained for After applying 0.7 V for 10 min an additional anisotropic Ru(V) (Figure S43). Since the employed monomeric model signal with g = [2.27, 2.12, 1.85] appears, in agreement with compounds do not include the effects of changing noncovalent the formation of the S = 1/2 {[RuIII(OH)(bda)(4,4′- interaction energy between catalyst and graphitic surface bpy)]10(4,4′-bpy)} species, RuIII-10@C-paper. We did not during the activation procedure, the latter has to be evaluated identify any spin−spin interaction between neighboring Ru separately. For this purpose, we performed quantum chemical centers, probably due to the long distance (11.3 Å). Increasing calculations on a semiempirical level employing a dimeric the potential to 0.9 V just lowers the intensity of this signal, as model of 10′(H2O)2, {[RuII(H2O)2(bda-κ-NO)(py)]2(μ-4,4′- EPR-silent RuIV species are formed. Interestingly, the g values bpy)}, 2(H2O)2, anchored on a sufficiently large hexagonal of the signal assigned to the RuIII species shift to g = [2.45, graphene sheet, 2(H2O)2@GS. As illustrated in Figure 6, now 2.24, 1.85] after the electrode was cycled 50 times. This besides the CH−π interactions observed for 2@GS, the 11655 https://doi.org/10.1021/jacs.1c04738 J. Am. Chem. Soc. 2021, 143, 11651−11661 Journal of the American Chemical Society pubs.acs.org/JACS Article formation of 2(H2O)2@GS introduces additional π−π interactions between the graphitic surface and the bda axial pyridyl moiety, which is further favored by the hydrogen bonding between the Ru-OH group and the dangling carboxylate (see Figure 6). Additionally, free carboxylates are stabilized by anion−π interactions, although this effect is assumed to play a minor role in higher oxidation states due to proton sharing with aquo ligands via hydrogen bonds. In total, the interaction is enhanced during this transformation by 12.9 kJ/mol per Ru center. Although this value has been determined for Ru(II), it can be assumed to be approximately constant in all oxidation states and, as a result, compensating the energy needed to incorporate a second water molecule in the Ru(V) state (9.1 kJ/mol) as described above. This explains why this transformation process only occurs under the influence of the graphitic surface. Since a large barrier of 150.2 kJ/mol (Figure S43b) is involved in this transition, this process is slow, which explains the large number of necessary CV cycles in order to achieve complete conversion into 10′(H2O)2@CNT, but also, once formed, it displays high stability at the electrode surface. Further, the X-ray absorption spectroscopy (XAS) experiments (red and cyan traces in Figure 2; Tables S5−S7) are also consistent with this transformation. Using the information from the Pourbaix diagram in Figure 7b, DFT has been used to study the complete water oxidation mechanism starting from the monomer cis-[RuII(H2O)2(bda-κ- NO)(py)2], 1′(H2O)2, as depicted in Figure 9a. All calculated redox potentials are remarkably low, supporting the assumption that 1′(H 2 O) 2 is a suitable model for 10′(H2O)2@CNT@GC and the respective coordination environment of Ru (Table S8). Especially, the obtained values for RuIV/RuIII, RuV/RuIV, and RuVI/RuV redox pairs are properly fitting to the experimental data at neutral pH. Notably, a strong hydrogen bond between one of the aquo ligands and the dangling carboxy group helps intramolecular charge delocalization. The deviation in the calculated RuIII/ RuII potential can be explained by the neglect of the graphitic surface in the employed model system. Missing anion−π interaction leads to protonation of the carboxy group already Figure 9. (a) Water oxidation mechanism of 1(H2O)2 proposed in Ru(II) and overestimation of the free energy of the latter based on calculations on the B3LYP-D3/def2-SVP level of theory species, therefore suggesting a 1e−/1H+ proton-coupled including implicit water solvation. Redox potentials are provided vs electron transfer (PCET) for the Ru III /Ru II pair in standard hydrogen electrode. Free energies are given in kJ/mol. Blue 1′(H 2 O) 2 , contrary to the experimental findings for letters correspond to the lowest spin state. Pyridine axial ligands have 10′(H2O)2@CNT. The intramolecular proton transfer is been omitted for clarity in all drawn structures. (b) Side view and top especially important in higher oxidation states, since it allows view of the optimized transition state of the WNA with omitted axial ligands for clarity. Relevant bond lengths in Å are provided as black access to an exceptionally stable RuV(O)2 species. Formation numbers. Blue labels correspond to the numbering of oxo groups used of this complex is accompanied by further decoordination of in the text. For a detailed description of all calculations and chosen the equatorial ligand to form a trigonal bipyramidal bda-κ-O model systems, see the SI. complex, reducing the formal number of electrons in the coordination environment from 19 to 17. WNA would be energetically strongly uphill in this state (ΔG = 129.7 kJ/mol, therefore unlikely to occur and could not be identified by Figure S45a); thus a fourth oxidation up to formal RuVI is our optimization attempts. Instead, it can be seen from the necessary (as observed experimentally here and for [Ru- optimized transition state depicted in Figure 9b that the (bpy)2(H2O)2]2+ as well)38 to allow formation of the O−O oxygen atom of the water attacks O1 while O2 serves as an bond. In this step, two different oxo groups, O1 or O2, could be immediate proton acceptor, which is in line with former attacked by water according to the numbering in Figure 9b. studies of WNA on RuVI(O)2.38 Although it cannot be excluded that in this step WNA happens In acidic media, however, the ability of intramolecular on oxo group O2 to some extent in the equilibrium, this is less proton transfer to the carboxylate is reduced because of probable because of its more negative partial charge compared external protonation of the latter. Thus, O2 would be already to O1 (Figure S45b). A transition state with the dangling protonated and as a consequence unable to support the WNA carboxylate directly acting as a base for the attacking water, as step, which explains the loss in activity of the presented system it has been discussed for [Ru(tda)(py)2] catalysts,42 is below pH 5.5 (Figure S22). Further, it is worth noting that the 11656 https://doi.org/10.1021/jacs.1c04738 J. Am. Chem. Soc. 2021, 143, 11651−11661 Journal of the American Chemical Society pubs.acs.org/JACS Article Scheme 1. Summary of Proposed Redox Events for 10@CNT@GC and 10′(H2O)2@CNT@GC at pH 7a a Pyridine axial ligands have been omitted for clarity. electrochemical oxidation up to RuVI(O)2 directly leads to a S25−S27 and Table S3). To kinetically characterize the high-spin triplet species. Since this spin state is preserved in the electrocatalytic performance of these materials, we carried out following intermediates, no additional intersystem crossing is a foot of the wave analysis (FOWA), which is shown in Figure necessary to close the catalytic cycle and release triplet 10a and Figures S29 and S30.43 A TOFmax average value of 3.7 dioxygen. This way, a potential additional barrier associated × 103 ± 0.36 × 103 s−1 is obtained for several samples of with change of the spin state is avoided. different degrees of surface loading with a kinetic rate law that The proposed scenario for 10 is summarized in Scheme 1, is first order with regard to the [Ru] at the surface (ΓRu). A where after three consecutive one-electron transfers 10@CNT CPE experiment was carried out using 10′(H2O)2@CNT@ is able to oxidize water to dioxygen during initial CV cycles. GC (Γ10′(H2O)2 = 16.7 nmol·cm−2) at an applied potential of However, upon repetitive cycling, the intensity of the catalytic Eapp = 1.35 V (Figure 10b). wave located at Eonset ≈ 1.02 V decreases drastically, A very stable current density of 14 mA/cm2 was generated accompanied by a shift of the foot of the electrocatalytic for 12 h, which implies a turnover number TON = 9.4 × 104 wave to Eonset ≈ 1.10 V. This fact is provoked by a per Ru unit assuming 94% faradaic efficiency. The latter was rearrangement of coordination environment for Ru-bda on calculated using a Clark-type electrode in the gas phase, as graphitic surfaces, which favor the high-stabilized open-ring shown in Figure S31. The integrity of the catalyst after 9.4 × species due to the strong CH−π and π−π interactions with 104 TONs was evaluated based on CV experiments that CNTs. Successively, the new “bda-κ-NO” coordination showed no additional species generated during turnover and environment promotes the introduction of a second aquo thus manifests its high stability (Figure S28). In addition, XAS ligand in the first coordination sphere, forming 10′(H2O)2@ experiments displayed in Figure 2 further support that the CNT. Thus, a new catalytic cycle comes into play, where molecular species before and after catalysis are the same (cyan Ru(VI) is the species responsible for O−O bond formation, and magenta traces in Figure 2) and, importantly, also ruling thanks to the high stabilization effect of three consecutive out the formation of RuO2 (see also Figures S32 and S33). PCET events, the intramolecular hydrogen bonding, and the This is remarkable considering that in the homogeneous phase additional stabilization energy provided by the graphitic the cis-[Ru(bpy)2(H2O)2]2+ complex has a high tendency to surface. isomerize to its trans form, which does not have the sufficiently Performance of 10′(H2O)2@CNT as a Molecular Water high redox potentials to undergo water oxidation catalysis. This Oxidation Anode. In the previous section, we have shown isomerization phenomenon is therefore responsible for the low that upon repetitive CVs the molecular water oxidation catalyst activity of cis-[Ru(bpy)2(H2O)2]2+ as a water oxidation 10 anchored on the CNT is progressively transformed into a catalyst.38,39 In sharp contrast, here the cis-bis-aquo complex more powerful catalyst, 10′(H2O)2, which reaches current 10′(H2O)2@CNT@GC is perfectly stable under turnover and densities of 42 mA/cm2 at 1.45 V, as shown in Figure 7a. does not isomerize, thanks again to the stabilizing effect of the Increasing the amount of mass deposited on the electrode surface interactions. surface, we obtained 10′(H2O)2@CNT@GC molecular The performance of 10′(H2O)2@CNT@GC as a molecular anodes with a surface density of Γ10′(H2O)2 = 12.2 nmol· anode for water oxidation is comparable to that of the recently cm−2, which reach remarkable current densities in the range of reported activated (Ru-tda)15@CNT@GC in terms of TONs 180 mA/cm2 at 1.45 V, as shown in Figure 10a (see Figures and a bit lower regarding TOFs.30 The latter also relies on 11657 https://doi.org/10.1021/jacs.1c04738 J. Am. Chem. Soc. 2021, 143, 11651−11661 Journal of the American Chemical Society pubs.acs.org/JACS Article only obtained at pH 13 or above with a very large amount of metal oxide deposited at the surface of the electrode.48 Further, coordination polymers based on bridging ligands and using Fe and Ni as transition metals have been reported as efficient water oxidation anodes using approximately Γ = 17 μmols/cm2 of Fe2Ni on top of a Ni foam electrode.49 At pH 14, these electrodes achieve current densities of 100 μA/cm2 with approximately 300 mV overpotential. In sharp contrast, here in our work we deposit 0.1 nmol/cm2 of Ru at the GC electrode and 16.7 nmols/cm2 at the CNTs, which is 5 and 3 orders of magnitude less mass as compared to the previously mentioned Ni foam electrode. ■ CONCLUSIONS A Ru-bda oligomer with 10 repetitive units of formula {[RuII(bda-κ-N2O2)(4,4′-bpy)]10(4,4′-bpy)}, 10, has been prepared, thoroughly characterized, and successfully anchored onto CNTs taking advantage of CH−π interactions between the bda ligand and graphitic surfaces. Anchored on CNTs, 10 behaves as a molecular electrocatalytic anode but progressively evolves to a new complex where the bda ligands change their coordination mode from bda-κ-N2O2 to bda-κ-NO, generating the new molecular hybrid material, 10′(H2O)2@CNT. This transformation is facilitated by the interaction of the bda ligand with graphitic surfaces favoring a partial decoordination via additional π−π and anion−π interactions. This ligand flexibility is in turn responsible for the high stability of the oligomer on the surface of CNTs. The new molecular hybrid material 10′(H2O)2@CNT achieves current densities in the range of 200 mA/cm2 at pH 7 under an applied potential of 1.45 V, using very small amounts of mass (0.1 nmol/cm2 of Ru at the GC electrode and 16.7 nmol/cm2 at the CNTs) with TOFs and TONs in the range of 3.7 × 104 s−1 and 9.4 × 104, respectively. The high activity of the complex is associated with the ligand flexibility and the second coordination sphere effects exerted by the decoordinated arm of the bda ligand that provides the Figure 10. (a) CV of 10′(H2O)2@CNT@GC in 1.0 M phosphate necessary stabilization by H-bonding for a low-energy catalytic buffer (pH 7) at a scan rate of 100 mV/s (gray trace). The black line cycle. On the other hand, the high stability of the Ru water corresponds to the experimental data used for FOWA analysis, and oxidation catalyst is associated with the molecular catalyst− the red line shows the region used for the determination of the surface interactions especially during turnover. The oligomeric TOFmax value. Inset: Plot of TOFmax vs Γ10′(H2O)2. (b) CPE of nature of the molecule provides multiple sites of anchorage at 10′(H2O)2@CNT@GC (Γ10′(H2O)2 = 16.7 nmol·cm−2) at Eapp = 1.35 the surface, which strongly enhances the overall stability of the V vs NHE in a 1 M phbf (pH 7) solution for 12 h (black trace) and hybrid catalyst−CNT anode via a dynamic bonding. ■ blank (blue trace). ASSOCIATED CONTENT aromatic CH−π interactions with a CNT surface for anchoring purposes. Additionally, the present system benefits from the * sı Supporting Information fact that it does not need an activation process in basic The Supporting Information is available free of charge at solution, as is the case for (Ru-tda)15@CNT@GC. It is worth https://pubs.acs.org/doi/10.1021/jacs.1c04738. mentioning that the CH−π interaction used here achieves Experimental procedures, additional spectroscopic, surface densities that are at least 2 orders of magnitude higher microscopy, and electrochemical data, and computa- than traditional anchoring strategies based on π−π stacking or tional details (PDF) ■ on the use of diazonium salts, as observed for other molecular water oxidation catalysts (see Table S4).44−46 This in turn allows reaching current densities that are more than 2 orders of AUTHOR INFORMATION magnitude higher than the ones obtained using these Corresponding Authors traditional methods, and it provides a remarkable stability.26 Antoni Llobet − Institute of Chemical Research of Catalonia This is in sharp contrast with the paradigm established when (ICIQ). Barcelona Institute of Science and Technology using oxides as water oxidation catalysts, where an increase in (BIST), 43007 Tarragona, Spain; Departament de Quimica, the amount of mass generates an interfacial instability that Universitat Autonoma de Barcelona, 08193 Barcelona, limits the anode performance.47 Additionally, current densities Spain; orcid.org/0000-0002-6176-5272; in the range of 0.2 A/cm2 using metal oxides as catalysts are Email: allobet@iciq.cat 11658 https://doi.org/10.1021/jacs.1c04738 J. Am. Chem. Soc. 2021, 143, 11651−11661 Journal of the American Chemical Society pubs.acs.org/JACS Article Frank Würthner − Center for Nanosystems Chemistry, 97074 funding from the Severo Ochoa Excellence Program from the Würzburg, Germany; Institut für Organische Chemie, Instituto IMDEA Nanociencia, the Acciones de Dinamización Universität Würzburg, 97074 Würzburg, Germany; “Europa Investigacion” grant (EIN2019-103399), and the orcid.org/0000-0001-7245-0471; Email: wuerthner@ Spanish Ministerio de Ciencia, Innovacion y Universidades uni-wuerzburg.de grant (PID2019-111086RA-I00). O.R. and S.D. thank the Max Planck Society for funding and Eckhard Bill for support with Authors the EPR simulations. This research used resources of the Marcos Gil-Sepulcre − Institute of Chemical Research of Advanced Photon Source (Beamline 9 μB-B), an Office of Catalonia (ICIQ). Barcelona Institute of Science and Science User Facility operated for the U.S. Department of Technology (BIST), 43007 Tarragona, Spain; orcid.org/ Energy (DOE) Office of Science by Argonne National 0000-0002-2839-8108 Laboratory, and was supported by the U.S. DOE under Joachim O. Lindner − Center for Nanosystems Chemistry, Contract No. DE-AC02-06CH11357. Synchrotron X-ray 97074 Würzburg, Germany scattering experiments were performed at NCD-SWEET Dorothee Schindler − Institut für Organische Chemie, beamline at ALBA synchrotron with the collaboration of Universität Würzburg, 97074 Würzburg, Germany ALBA staff (Proposal 2020094629). ■ Lucía Velasco − Instituto Madrileño de Estudios Avanzados en Nanociencia (IMDEA-Nanociencia), 28049 Madrid, REFERENCES Spain Dooshaye Moonshiram − Instituto Madrileño de Estudios (1) Steffen, W.; Rockström, J.; Richardson, K.; Lenton, T. M.; Folke, Avanzados en Nanociencia (IMDEA-Nanociencia), 28049 C.; Liverman, C.; Summerhayes, C. P.; Barnosky, A. D.; Cornell, S. E.; Madrid, Spain; orcid.org/0000-0002-9075-3035 Crucifix, M.; Donges, J. F.; Fetzer, I.; Lade, S. J.; Scheffer, M.; Winkelmann, R.; Schellnhuber, H. J. Trajectories of the Earth System Olaf Rüdiger − Max Planck Institute for Chemical Energy in the Anthropocene. Proc. Natl. Acad. Sci. U. S. A. 2018, 115, 8252− Conversion, 45470 Mülheim an der Ruhr, Germany; 8259. orcid.org/0000-0002-5148-9083 (2) Balmaseda, M. A.; Trenberth, K. E.; Källén, E. Distinctive Serena DeBeer − Max Planck Institute for Chemical Energy climate signals in reanalysis of global ocean heat content. Geophys. Res. Conversion, 45470 Mülheim an der Ruhr, Germany; Lett. 2013, 40, 1754−1759. orcid.org/0000-0002-5196-3400 (3) Edenhofer, O.; Pichs-Madruga, R.; Sokona, Y.; Minx, J. C.; Vladimir Stepanenko − Center for Nanosystems Chemistry, Farahani, E.; Kadner, S.; Seyboth, K.; Adler, A.; Baum, I.; Brunner, S.; 97074 Würzburg, Germany; Institut für Organische Chemie, Eickemeier, P.; Kriemann, B.; Savolainen, J.; Schlömer, S.; Stechow, Universität Würzburg, 97074 Würzburg, Germany C. V.; Zwickel, T. The Intergovernmental Panel on Climate Change Eduardo Solano − NCD-SWEET beamline, ALBA (IPCC). Climate Change 2014: Mitigation of Climate; Cambridge Univ. Press, 2014. synchrotron light source, 08290 Barcelona, Spain; (4) Frischmann, P. D.; Mahata, K.; Würthner, F. Powering the future orcid.org/0000-0002-2348-2271 of molecular artificial photosynthesis with light-harvesting metal- Complete contact information is available at: losupramolecular dye assemblies. Chem. Soc. Rev. 2013, 42, 1847− https://pubs.acs.org/10.1021/jacs.1c04738 1870. (5) Sala, X.; Maji, S.; Bofill, R.; García-Antón, J.; Escriche, L.; Llobet, Author Contributions A. Molecular Water Oxidation Mechanisms Followed by Transition Metals: State of the Art. Acc. Chem. Res. 2014, 47, 504−516. M.G.-S. performed the synthesis, characterization, and electro- (6) Smith, R. D. L.; Prévot, M. S.; Fagan, R. D.; Zhang, Z.; Sedach, chemical experiments and coordinated the tasks with all P. A.; Siu, M. K. J.; Trudel, S.; Berlinguette, C. P. Photochemical authors. J.O.L. performed the computational studies. D.S. Route for Accessing Amorphous Metal Oxide Materials for Water assisted in the analysis of the results. L.V. and D.M. performed Oxidation Catalysis. Science 2013, 340, 60−63. the XANES and EXAFS measurements and data analysis. O.R. (7) McCrory, C. C. L.; Jung, S.; Peters, J. C.; Jaramillo, T. F. performed EPR measurements and data analysis. V.S. Benchmarking Heterogeneous Electrocatalysts for the Oxygen performed the AFM and SEM experiments. E.S. assisted in Evolution Reaction. J. Am. Chem. Soc. 2013, 135, 16977−16987. the analysis of X-ray scattering results. F.W. and A.L. conceived (8) McCrory, C. C. L.; Jung, S.; Ferrer, I. M.; Chatman, S. M.; the idea of the project and wrote the paper. All authors Peters, J. C.; Jaramillo, T. F. Benchmarking Hydrogen Evolving Reaction and Oxygen Evolving Reaction Electrocatalysts for Solar contributed to the design of experiments, analysis of the Water Splitting Devices. J. Am. Chem. Soc. 2015, 137, 4347−4357. results, and preparation of the manuscript. (9) Mohammed-Ibrahim, J. A. review on NiFe-based electrocatalysts Notes for efficient alkaline oxygen evolution reaction. J. Power Sources 2020, The authors declare no competing financial interest. 448, 227375. ■ (10) Walter, M. G.; Warren, E. L.; McKone, J. R.; Boettcher, S. W.; Mi, Q.; Santori, E. A.; Lewis, N. Solar Water Splitting Cells. Chem. ACKNOWLEDGMENTS Rev. 2010, 110, 6446−6473. This project has received funding from the European Research (11) Kärkäs, M. D.; Verho, O.; Johnston, E. V.; Åkermark, B. Council (ERC) under the European Union’s Horizon 2020 Artificial Photosynthesis: Molecular Systems for Catalytic Water research and innovation program (grant agreement No. Oxidation. Chem. Rev. 2014, 114, 11863−12001. 787937, recipient F.W.). We thank the Alexander von (12) Berardi, S.; Drouet, S.; Francàs, L.; Gimbert-Suriñach, C.; Guttentag, M.; Richmond, C.; Stoll, T.; Llobet, A. Molecular artificial Humboldt Foundation for supporting this collaborative photosynthesis. Chem. Soc. Rev. 2014, 43, 7501−7519. research with a Humboldt research award for A.L. A.L. also (13) Blakemore, J. D.; Crabtree, R. H.; Brudvig, G. W. Molecular acknowledges support from Ministerio de Ciencia e Catalysts for Water Oxidation. Chem. Rev. 2015, 115, 12974−13005. Innovación, Severo Ochoa Excellence Acreditation, FEDER, (14) Garrido-Barros, P.; Gimbert-Suriñach, C.; Matheu, R.; Sala, X.; and AGAUR through grants PID2019-111617RB-I00, Llobet, A. How to make an efficient and robust molecular catalyst for CEX2019-000925-S, and 2017-SGR-163. D.M. acknowledges water oxidation. Chem. Soc. Rev. 2017, 46, 6088−6098. 11659 https://doi.org/10.1021/jacs.1c04738 J. Am. Chem. Soc. 2021, 143, 11651−11661 Journal of the American Chemical Society pubs.acs.org/JACS Article (15) Matheu, R.; Garrido-Barros, P.; Gil-Sepulcre, M.; Ertem, M. Z.; (31) Schindler, D.; Gil-Sepulcre, M.; Lindner, J. O.; Stepanenko, V.; Sala, X.; Gimbert-Suriñach, C.; Llobet, A. The development of Moonshiram, D.; Llobet, A.; Würthner, F. Efficient Electrochemical molecular water oxidation catalysts. Nat. Rev. Chem. 2019, 3, 331− Water Oxidation by a Trinuclear Ru(bda) Macrocycle Immobilized 341. on Multi-Walled Carbon Nanotube Electrodes. Adv. Energy Mater. (16) Li, X.; Lei, H.; Liu, J.; Zhao, X.; Ding, S.; Zhang, Z.; Tao, X.; 2020, 10, 2002329. Zhang, W.; Wang, W.; Zheng, X.; Cao, R. Carbon Nanotubes with (32) (a) Schulze, M.; Kunz, V.; Frischmann, P. D.; Würthner, F. A. Cobalt Corroles for Hydrogen and Oxygen Evolution in pH 0−14 supramolecular ruthenium macrocycle with high catalytic activity for Solutions. Angew. Chem., Int. Ed. 2018, 57, 15070−15075. water oxidation that mechanistically mimics photosystem II. Nat. (17) Xie, L.; Li, X.; Wang, B.; Meng, J.; Lei, H.; Zhang, W.; Cao, R. Chem. 2016, 8, 576−583. (b) Kunz, V.; Lindner, J. O.; Schulze, M.; Molecular Engineering of a 3D Self-Supported Electrode for Oxygen Röhr, M. I. S.; Schmidt, D.; Mitrić, R.; Würthner, F. Cooperative Electrocatalysis in Neutral Media. Angew. Chem., Int. Ed. 2019, 58, water oxidation catalysis in a series of trinuclear metallosupramo- 18883−18887. lecular ruthenium macrocycles. Energy Environ. Sci. 2017, 10, 2137− (18) Xu, J.; Murphy, S.; Xiong, D.; Cai, R.; Wei, X.-K.; Heggen, M.; 2153. Barborini, E.; Vinati, S.; Dunin-Borkowski, R. E.; Palmer, R. E.; Liu, L. (33) Becke, A. D. A new mixing of Hartree−Fock and local density- Cluster Beam Deposition of Ultrafine Cobalt and Ruthenium Clusters functional theories. J. Chem. Phys. 1993, 98, 1372−1377. for Efficient and Stable Oxygen Evolution Reaction. ACS Appl. Energy (34) Lee, C.; Yang, W.; Parr, R. G. Development of the Colle- Mater. 2018, 1, 3013−3018. Salvetti correlation-energy formula into a functional of the electron (19) Li, W.; Gao, X.; Wang, X.; Xiong, D.; Huang, P.-P.; Song, W.- density. Phys. Rev. B: Condens. Matter Mater. Phys. 1988, 37, 785−789. G.; Bao, X.; Liu, L. From water reduction to oxidation: Janus Co-Ni-P (35) Schilling, M.; Cunha, R. A.; Luber, S. Enhanced Ab Initio nanowires as high efficiency and ultrastable electrocatalysts for over Molecular Dynamics Exploration Unveils the Complex Role of 3000 h water splitting. J. Power Sources 2016, 330, 156−166. Different Intramolecular Bases on the Water Nucleophilic Attack (20) Garrido-Barros, P.; Gimbert-Suriñach, C.; Moonshiram, D.; Mechanism. ACS Catal. 2020, 10, 7657−7667. Picón, A.; Monge, P.; Batista, V. S.; Llobet, A. Electronic π- (36) Dobson, J. C.; Meyer, T. J. Redox Properties and Ligand Loss Delocalization Boosts Catalytic Water Oxidation by Cu(II) Molecular Chemistry in Aqua/Hydroxo/Oxo Complexes Derived from cis- and Catalysts Heterogenized on Graphene Sheets. J. Am. Chem. Soc. 2017, trans-[(bpy)2RuII(OH2)2]2+. Inorg. Chem. 1988, 27, 3283−3291. 139, 12907−12910. (37) Kutin, Y.; Cox, N.; Lubitz, W.; Schnegg, A.; Rüdinger, O. In (21) Duan, L.; Fischer, A.; Xu, Y.; Sun, L. Isolated Seven-Coordinate Situ EPR Characterization of a Cobalt Oxide Water Oxidation Ru(IV) Dimer Complex with [HOHOH]− Bridging Ligand as an Catalyst at Neutral pH. Catalysts 2019, 9, 926. Intermediate for Catalytic Water Oxidation. J. Am. Chem. Soc. 2009, (38) Sala, X.; Ertem, M. Z.; Vigara, L.; Todorova, T.; Chen, W.; 131, 10397−10399. Rocha, R. C.; Aquilante, F.; Cramer, C. J.; Gagliardi, L.; Llobet, A. (22) Duan, L.; Bozoglian, F.; Mandal, S.; Steward, B.; Privalov, T.; The cis-[RuII(bpy)2(H2O)2]2+ Water-Oxidation Catalyst Revisited. Llobet, A.; Sun, L. A molecular ruthenium catalyst with water- Angew. Chem., Int. Ed. 2010, 49, 7745−7747. oxidation activity comparable to that of photosystem II. Nat. Chem. (39) Planas, N.; Vigara, L.; Cady, C.; Miró, P.; Huang, P.; 2012, 4, 418−423. Hammarström, L.; Styring, S.; Leidel, N.; Dau, H.; Haumann, M.; (23) Zhang, B.; Sun, L. Ru-bda: Unique Molecular Water-Oxidation Gagliardi, L.; Cramer, C. J.; Llobet, A. Electronic Structure of Catalysts with Distortion Induced Open Site and Negatively Charged Oxidized Complexes Derived from cis-[RuII(bpy)2(H2O)2]2+ and Its Ligands. J. Am. Chem. Soc. 2019, 141, 5565−5580. Photoisomerization Mechanism. Inorg. Chem. 2011, 50, 11134− (24) Matheu, R.; Ertem, M. Z.; Gimbert-Suriñach, C.; Sala, X.; 11142. Llobet, A. Seven Coordinated Molecular Ruthenium−Water (40) Ezhov, R.; Ravari, A. K.; Page, A.; Pushkar, Y. Water Oxidation Oxidation Catalysts: A Coordination Chemistry Journey. Chem. Rev. Catalyst cis-[Ru(bpy)(5,5′-dcbpy)(H2O)2]2+ and Its Stabilization in 2019, 119, 3453−3471. Metal−Organic Framework. ACS Catal. 2020, 10, 5299−5308. (25) Matheu, R.; Ertem, M. Z.; Gimbert-Suriñach, C.; Benet- (41) Lebedev, D.; Pineda-Galvan, Y.; Tokimaru, Y.; Fedorov, A.; Buchholz, J.; Sala, X.; Llobet, A. Hydrogen Bonding Rescues Kaeffer, N.; Copéret, C.; Pushkar, Y. The Key RuV=O Intermediate of Overpotential in Seven-Coordinated Ru Water Oxidation Catalysts. Site-Isolated Mononuclear Water Oxidation Catalyst Detected by in ACS Catal. 2017, 7, 6525−6532. Situ X-ray Absorption Spectroscopy. J. Am. Chem. Soc. 2018, 140, (26) Li, F.; Zhang, B.; Li, X.; Jiang, Y.; Chen, L.; Li, Y.; Sun, L. 451−458. Highly Efficient Oxidation of Water by a Molecular Catalyst (42) Matheu, R.; Ertem, M. Z.; Benet-Buchholz, J.; Coronado, E.; Immobilized on Carbon Nanotubes. Angew. Chem., Int. Ed. 2011, Batista, V. S.; Sala, X.; Llobet, A. Intramolecular Proton Transfer 50, 12276−12279. Boosts Water Oxidation Catalyzed by a Ru Complex. J. Am. Chem. (27) Wang, L.; Fan, K.; Chen, H.; Daniel, Q.; Philippe, B.; Rensmo, Soc. 2015, 137, 10786−10795. H.; Sun, L. Towards efficient and robust anodes for water splitting: (43) Matheu, R.; Neudeck, S.; Meyer, F.; Sala, X.; Llobet, A. Foot of Immobilization of Ru catalysts on carbon electrode and hematite by the Wave Analysis for Mechanistic Elucidation and Benchmarking in situ polymerization. Catal. Today 2017, 290, 73−77. Applications in Molecular Water Oxidation Catalysis. ChemSusChem (28) Creus, J.; Matheu, R.; Peñafiel, I.; Moonshiram, D.; Blondeau, 2016, 9, 3361−3369. P.; Benet-Buchholz, J.; García-Antón, J.; Sala, X.; Godard, C.; Llobet, (44) Wu, L.; Nayak, A.; Shao, J.; Meyer, T. J. Crossing the bridge A. A Million TurnoverMolecular Anode for CatalyticWater Oxidation. from molecular catalysis to a heterogenous electrode in electro- Angew. Chem., Int. Ed. 2016, 55, 15382−15386. catalytic water oxidation. Proc. Natl. Acad. Sci. U. S. A. 2019, 116, (29) Matheu, R.; Francàs, L.; Chernev, P.; Ertem, M. Z.; Batista, V.; 11153−11158. Haumann, M.; Sala, X.; Llobet, A. Behavior of the Ru-bda Water (45) Wang, L.; Fan, K.; Daniel, Q.; Duan, L.; Li, F.; Philippe, B.; Oxidation Catalyst Covalently Anchored on Glassy Carbon Electro- Rensmo, H.; Chen, H.; Sun, J.; Sun, L. Electrochemical driven water des. ACS Catal. 2015, 5, 3422−3429. oxidation by molecular catalysts in situ polymerized on the surface of (30) Hoque, Md. A.; Gil-Sepulcre, M.; de Aguirre, A.; Elemans, J. A. graphite carbon electrode. Chem. Commun. 2015, 51, 7883−7886. A. W.; Moonshiram, D.; Matheu, R.; Shi, Y.; Benet-Buchholz, J.; Sala, (46) Li, F.; Fan, K.; Wang, L.; Daniel, Q.; Duan, L.; Sun, L. X.; Malfois, M.; Solano, E.; Lim, J.; Garzón-Manjón, A.; Scheu, C.; Immobilizing Ru(bda) Catalyst on a Photoanode via Electrochemical Lanza, M.; Maseras, F.; Gimbert-Suriñach, C.; Llobet, A. Water Polymerization for Light-Driven Water Splitting. ACS Catal. 2015, 5, oxidation electrocatalysis using ruthenium coordination oligomers 3786−3790. adsorbed on multiwalled carbon nanotubes. Nat. Chem. 2020, 12, (47) Weber, D.; Tripkovic, D.; Kretschmer, K.; Bianchini, M.; 1060−1066. Brezesinski, T. Surface modification strategies for improving the 11660 https://doi.org/10.1021/jacs.1c04738 J. Am. Chem. Soc. 2021, 143, 11651−11661 Journal of the American Chemical Society pubs.acs.org/JACS Article cycling performance of Ni-rich cathode materials. Eur. J. Inorg. Chem. 2020, 2020, 3117−3130. (48) Xu, J.; Murphy, S.; Xiong, D.; Cai, R.; Wei, X.-K.; Heggen, M.; Barborini, E.; Vinati, S.; Dunin-Borkowski, R. E.; Palmer, R. E.; Liu, L. Cluster Beam Deposition of Ultrafine Cobalt and Ruthenium Clusters for Efficient and Stable Oxygen Evolution Reaction. ACS Appl. Energy Mater. 2018, 1, 3013−3018. (49) Ma, J.; Bai, X.; He, W.; Wang, S.; Li, L.; Chen, H.; Wang, T.; Zhang, X.; Li, Y.; Zhang, L.; Chen, J.; Meng, F.; Fu, Y. Amorphous FeNi-bimetallic infinite coordination polymers as advanced electro- catalysts for the oxygen evolution reaction. Chem. Commun. 2019, 55, 12567−12570. 11661 https://doi.org/10.1021/jacs.1c04738 J. Am. Chem. Soc. 2021, 143, 11651−11661