Angewandte Research Articles Chemie How to cite: Angew. Chem. Int. Ed. 2021, 60, 18639 – 18644 Water Oxidation Catalysis Hot Paper International Edition: doi.org/10.1002/anie.202104020 German Edition: doi.org/10.1002/ange.202104020 Consecutive Ligand-Based Electron Transfer in New Molecular Copper-Based Water Oxidation Catalysts Marcos Gil-Sepulcre+, Pablo Garrido-Barros+, Jan Oldengott, Ignacio Funes-Ardoiz, Roger Bofill, Xavier Sala, Jordi Benet-Buchholz, and Antoni Llobet* Dedicated to Prof. Dr. Wolfgang Kaim on the occasion of his 70th birthday Abstract: Water oxidation to dioxygen is one of the key mechanistic analysis, including the isolation and spectroscopic reactions that need to be mastered for the design of practical characterization of relevant reaction intermediates.[9] Further, devices based on water splitting with sunlight. In this context, the molecular toolkit provides an ample variety of ligands water oxidation catalysts based on first-row transition metal that—if properly used—can help in the rational design of complexes are highly desirable due to their low cost and their extremely efficient and oxidatively rugged molecular water synthetic versatility and tunability through rational ligand oxidation catalysts (WOCs).[10, 11] At present, there is a large design. A new family of dianionic bpy-amidate ligands of body of WOCs based on Ru complexes with very good general formula H2LNn@ (LN is [2,2’-bipyridine]-6,6’-dicar- performances,[12–14] and a few based on first raw transition boxamide) substituted with phenyl or naphthyl redox non- metals.[15–20] innocent moieties is described. A detailed electrochemical Most of the WOCs described today achieve reactive analysis of [(L4)Cu]2@ (L4 = 4,4’-(([2,2’-bipyridine]-6,6’-di- species using metal-based Proton Couple Electron Transfer carbonyl)bis(azanediyl))dibenzenesulfonate) at pH 11.6 (PCET) processes from M-OHx (x = 0–2) groups.[21] A new shows the presence of a large electrocatalytic wave for water body of transition metal complexes capable of oxidizing water oxidation catalysis at an h = 830 mV. Combined experimental to dioxygen is starting to emerge, where redox non-innocent and computational evidence, support an all ligand-based ligands play a key role.[22–27] This is particularly important for process with redox events taking place at the aryl-amide first-row transition metal complexes where the energy differ- groups and at the hydroxido ligands. ence between different oxidation states is generally much wider than for instance second-row transition metals such as Introduction Ru.[9, 10, 28, 29] The achievement of highly reactive species by a combination of metal- and ligand-based electron transfer Molecular water oxidation catalysis driven by transition processes is also a strategy followed by nature to achieve metal complexes[1–4] is an active field of research at present highly reactive species at relatively low energies for oxidative due to its implication in the generation of solar fuels.[5–8] The transformations, as is the case of Cyt-P450.[30–32] molecular approach offers the possibility of a thorough We have recently shown that for the Cu-amidate complex containing a redox active ligand (N1,N1’-(1,2-phenylene)- bis(N2-methyloxalamide)), its overpotential (h) for the [*] Dr. M. Gil-Sepulcre,[+] Dr. P. Garrido-Barros,[+] Dr. J. Oldengott, catalytic water oxidation reaction could be reduced by more Dr. I. Funes-Ardoiz, Dr. J. Benet-Buchholz, Prof. A. Llobet than 500 mV, simply changing the electron donating nature of Institute of Chemical Research of Catalonia (ICIQ) the organic substituents on the parent ligand.[17] Barcelona Institute of Science and Technology (BIST) In the present work, we take a step forward and we Av. Pa"sos Catalans 16, 43007 Tarragona (Spain) E-mail: allobet@iciq.cat present a family of molecular WOCs based on Cu complexes Dr. R. Bofill, Dr. X. Sala, Prof. A. Llobet where all the electron transfer processes involved in the Departament de Qu&mica, Universitat Autlnoma de Barcelona catalytic cycle are ligand-based only. This strategy is attractive Cerdanyola del Valles, 08193 Barcelona (Spain) because there is a huge variety of ligands that can be used and Dr. I. Funes-Ardoiz therefore it can provide the means for an exquisite control of Departamento de Qu&mica the kinetics and thermodynamics involved in water oxidation Centro de Investigacikn en S&ntesis Qu&mica (CISQ) catalysis. Here on, we describe the synthesis, spectroscopic, Universitad de La Rioja redox properties and the water oxidation mechanism of 26006 LogroÇo(Spain) a family of Cu complexes, [(LN)CuII] (N = 1–3) and [+] These authors contributed equally to this work. [(LN)CuII]2@ (N = 4–8), containing a dianonic bpy-amidate Supporting information and the ORCID identification number(s) for scaffold [2,2’-bipyridine]-6,6’-dicarboxamide (LN) with a vari- the author(s) of this article can be found under: ety of aromatic substituents (neutral and anionic) as depicted https://doi.org/10.1002/anie.202104020. in Figure 1, that can fine tune the water oxidation properties T 2021 The Authors. Angewandte Chemie International Edition of the catalyst. published by Wiley-VCH GmbH. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. Angew. Chem. Int. Ed. 2021, 60, 18639 – 18644 T 2021 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH 18639 Angewandte Research Articles Chemie copper complexes.[17, 22] Whereas the Cu–Nbpy bond lengths fall within 1.94 to 1.97 c, the Cu–Namide bonds are slightly larger, 1.96 to 2.01 c. (Table S2). The cis N-Cu-N angles are within the expected range of 118.488 to 120.888 for Namide-Cu-Namide and 77.488 to 78.588 for Nbpy-Cu-Nbpy (Table S3). Redox Properties and Water Oxidation Catalysis The redox properties of the ligands and complexes Figure 1. Drawings of H2LNn@ (N = 1–3, n = 0; N = 4–8; n = 2) bpy- described have been investigated in aqueous solution by amidate ligands used in this work. using Cyclic Voltammetry (CV), Differential Pulse Voltam- metry (DPV) and Controlled Potential Electrolysis (CPE) experiments. Results and Discussion The [(L4)Cu]2@ complex is selected as the model for initial evaluation of the electrochemical properties and catalytic Synthesis and Structure performances with regard to water oxidation catalysis be- cause of its higher solubility in aqueous solutions thanks to The synthesis of a family of ligands H2LNn@ (N = 1–3, n = the sulfonate group when compared to their analogues with 0; N = 4–8; n = 2; see Figure 1) and their CuII complexes is no sulfonate, [(L1–L3)Cu]. described in detail, together with their characterization based In aqueous solution at pH 11.6 the CV of complex on NMR (Figures S1–S9), UV/Vis (Figures S29,S30), Elec- [(L4)Cu]2@ shows only a large electrocatalytic wave with its tron Paramagnetic Resonance (EPR, Figure S38), MS (Fig- onset potential at approx. 1.3 V (Figure 2 B and Figures S42– ures S11–S26) and EA that is reported in the Supporting S44), which is related to the electrocatalytic oxidation of Information (SI). The ZnII analogue with the H2L42@ depro- water to dioxygen. The DPV shows a wave at 1.35 V (h = tonated ligand, [(L4)Zn]2@, has also been prepared in a similar 830 mV) that can be associated with the redox couple manner and its synthesis and characterization is reported in responsible for the catalytic activity. the Supporting Information (see Experimental Section and The generation of dioxygen was confirmed by a CPE Figure S10). In this family of ligands, the aromatic delocali- experiment where a gas phase Clark-type electrode had been zation along the bipyridyl-di-amidate groups places their four placed at the headspace of the working compartment (Fig- coordinating N-atoms in a nearly planar manner, so that all ure S50). A Faradaic efficiency (FE) of 76 % was obtained for their CuII complexes obtained display a distorted square [(L4)Cu]2@. No Cu-based heterogeneous oxide species planar geometry, as can be seen in the X-ray crystal structure (CuOx) were detected at the electrode as ascertained for [(L1-L7)Cu]n@ (n = 0 or 2), whose ORTEP are drawn in electrochemically by CV when placing the working electrode Figure 2 A and Figures S31–S37.[33] In addition, they also in a clean electrolyte solution after the O2 detection experi- display a contact in the apical position with an oxygen atom ment (Figure S50B) and neither by Scanning Electron coming from a H2O or MeOH molecule, or a sulfonate from Microscopy (SEM) or Energy Dispersive X-ray (EDX) a neighbor complex unit (the Cu–O distances span between spectroscopy measurements of the electrode (Figures S54A 2.33–2.61 c). The Cu@N bond lengths range from 1.94 to and S55A). The relatively low FE can be associated with the 2.01 c, similar to other previously reported tetraamidate potential catalyst degradation via hydroxylation of the aromatic ring, specially at the cationic radical stage, and further oxidative pathways when the phenolate group is generated.[14, 34, 35] A kinetic analysis of the electrocatalytic wave based on the Foot-of-the-wave analysis (FOWA)[36–38] methodology was carried out (see Figure S48 and SI for details). A TOFmax of 10.5 s@1 was obtained for [(L4)Cu]2@, which is within the range observed for previous Cu-based catalysts at basic pH.[17, 22, 39, 40] Further, first order behavior with regard to the concentration of [(L4)Cu]2@ was also obtained that points to a hydroxide nucleophilic attack, mimicking traditional water nucleophilic attack (WNA) mechanism and discarding the interaction of two MO units (I2M mechanism), as can be observed in the Figure 2. A) ORTEP drawing of [(L4)Cu]2@ at 50 % probability level. The inset of Figure 2 B. This is also consistent with the mechanistic solvent molecules, sodium and hydrogen atoms have been omitted for scenario proposed for previous Cu WOCs containing redox clarity. Color code: C, black; N, blue; O, red; S, yellow; Cu, light blue. non-innocent ligands.[29] B) Cyclic Voltammogram of 1 mM solution of [(L4)Cu]2@ in 0.1 M phosphate buffer at pH 11.6 at scan rate of 100 mVs@1. Inset: Plot of A computational analysis by means of DFT calculations kobs vs. [(L4)Cu]2@ calculated by FOWA analysis assuming a WNA using B3LYP-D3 in combination with the SMD solvation mechanism (green trace) or an I2M mechanism (purple trace). model was carried out to further characterize the catalytic 18640 www.angewandte.org T 2021 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH Angew. Chem. Int. Ed. 2021, 60, 18639 – 18644 Angewandte Research Articles Chemie The next step involves the coordination of an -OH ligand at the equatorial position, replacing the just described elongated Cu-amidate bond, thus generating [(L4C-k-N3)CuII- (OH)]2@, which is exergonic by 9.1 kcal mol@1. A second oxidation occurs now at 1.26 V giving rise to the formation of a diradical, [(L4CC-k-N3)CuII(OH)]@ . The similarity of the two redox potentials reflects the relative independence of the amidate fragments within the complex with no or very small electronic interaction, and is consistent with the absence of precatalytic waves. The doubly oxidized quartet species [(L4CC- k-N3)CuII(OH)]@ features again an increased Cu–Namide dis- tance, from 2.07 c to 2.21 c. In this case, its substitution by a second incoming hydroxo group to form [(L4CC-k-N2)CuII- (OH)2]2@ is exergonic by 2.7 kcal mol@1. Electronically, both doubly oxidized structures are quartets, presenting an un- paired electron on the CuII center and two other unpaired electrons on each aryl-amidate fragment. We have considered both mono- and di-hydroxido doubly oxidized species as possible intermediates for the O@O bond formation, since both of them give similar energy penalties. The monohydrox- ido mechanism is described below whereas the dihydroxido species is reported in the SI (Figures S61–S63). Approaching an OH@ group to [(L4CC-k-N3)CuII(OH)]- at distances as long as 4 c already promotes a single electron transfer (SET) process, reducing one of the aryl-amidate groups (SET-WNA; Figure S60), similar to the previously described [Cu(L)(OH2)2] (L = 6,6’-dihydroxy-2,2’-bipyridine) catalyst.[39] This highly favored electron transfer is likely a result of a low energy reorganization process and thus Figure 3. Catalytic cycle for [(L4)Cu]2@ ([(L4-k-N4)CuII]2@) via the SET- precludes the acquisition of a transition state. This new WNA mechanism. Free energy changes are indicated in black in electronic structure evolves to an intermediate [(L4C-k- kcal mol@1 and the calculated redox potentials are shown in orange in N3)CuII(HO···OHC)]2@ where both oxygen atoms form a two V vs. NHE. Spin density distribution is shown for selected intermedi- center-three electron bond (2c-3e@) with a formal order of 0.5 ates. Bond lengths are indicated in Angstroms. The radical character of the ligands is indicated in the formula. and a bond length of 2.23 c (Figure 3 and Figure S62). Further decreasing the distance between the two oxygen cycle to support and complement available experimental centers fosters a second single electron transfer that reduces data. The main results obtained for the [(L4-k-N4)CuII]2@ the second phenyl-amidate moiety concomitant with the catalyst are summarized in Figure 3 and described in detail formation of a neutral hydroperoxo species bonded to the Cu in the SI. Note that in the DFT analysis the calculated species center, [(L4-k-N3)CuII(HOOH)]2@, where the Cu center has involved in the catalytic cycle contain the L4 ligand bonded in doublet electronic state. The initial quartet species are different coordination modes. Therefore, the number and connected by a minimum energy crossing point (MECP).[41] nature of the atoms bonded to the metal center will be A 9.4 kcal mol@1 energy barrier is obtained for this second indicated from now on in the corresponding formula. Also, electron transfer as estimated from the potential energy the ligands based radical species will also be indicated with relaxed scan (Figure 3 and Figure S60). DFT calculation of a dot. the OH@ using the experimental value for the energy of The calculated structure for complex [(L4-k-N4)CuII]2@ at solvation (@102.8 kcal mol@1)[42] suggests a likely underesti- pH 11.6 is similar to the one obtained by X-ray diffraction mation of this value using SMD, which in turn would be analysis (see Table S4 and Figure 2 A; S56). Upon one reflected in a lower activation energy for the O@O bond electron removal, the most stable oxidized species consists formation (see Figure S59). of a triplet CuII complex with a radical cation mainly centered The all ligand-based ET process suggested by DFT is in on one of the aryl-amidate moieties, as shown in Figures S57 agreement with the need of highly anionic/s-and p-donating and S58, [(L4C-k-N4)CuII]@ . The calculated oxidation potential ligand environment coordinated to the Cu metal center, in of 1.27 V is in good agreement with the experimental value of order to be able to observe the III/II redox couple at relatively 1.37 V obtained from DPV (Table S5). As expected, this low potentials. For instance the [CuII-OPBAN]2@ (OPBAN is ligand oxidation causes a decrease in the electron-donating N1,N1’-(1,2-phenylene)bis(N2-methyl-oxalamide), where the character of the amidate group that is reflected in a significant OPBAN ligand provided four negatives charges directly increase in the Cu–Namide distance from 2.05 c to 2.14 c, and bonded to the metal center, has a CuIII/II couple at 0.56 V at thus weakening the Cu@N bond. pH 11.5. In the present case the LN ligands are bonded to the metal center by two neutral N of the bpy group and two Angew. Chem. Int. Ed. 2021, 60, 18639 – 18644 T 2021 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH www.angewandte.org 18641 Angewandte Research Articles Chemie anionic N of the amide groups, thus effectively act as a dianonic ligand to the metal center. For Ru complexes, where a large number of complexes have been electrochemi- cally characterized, the effect of exchanging an anionic ligand by a neutral one increases the redox potential by roughly 400– 500 mV per anionic group.[43, 44] Thus, in the present case the LN ligands will not have sufficient electron-donation capacity to stabilize the CuIII/II couple below or at the potential where catalysis is occur- ring.[17, 45, 46] An additional evidence for the ligand based process is provided by the electrochemical properties of the homologue Zn complex, [(L4)Zn]2@, where no metal based ET can occur. A CV of [(L4)Zn]2@ (see Figure S41) under the Figure 5. A) CVs of a 1 mM solution of [(L3)Cu] (blue line), [(L4)Cu]2@ same conditions as [(L4)Cu]2@ shows a ligand-based wave (green line), [(L7)Cu]2@ (red line) and [(L8)Cu]2@ (black line) in 0.1 M (non-catalytic) at the same potential where the Cu homo- phosphate buffer at pH 11.6. Note: In case of complex [(L3)Cu] the logue shows catalytic activity. experiments were performed in a mixture of 0.1 M phosphate buffer/ Finally, a compelling evidence for the all ligand-based ET TFE (6:4) for solubility reasons. B) plot of kobs (s@1) for the water in the water oxidation catalysis is provided by the correlation oxidation catalysis vs. h (mV) for selected cases. between the substituent effect on the amidate ligand in the potential observed for water oxidation catalysis, that is described in the next section. whereas electron-withdrawing groups increase it, as expected from phenyl-centered oxidations. Similarly, the same trend occurs for the naphthyl substituted cases. In addition, the p- Substituent Effects delocalization (phenyl vs. naphthyl) results in a significant decrease of h, as expected.[47] As described above, both electrochemical experiments It is interesting to see that all complexes [(L1-L8)Cu]n@ and DFT calculations clearly point out that the O@O bond (n = 0 or 2@) display a catalytic wave in the range of 1.0–1.6 V, formation and thus water oxidation catalysis that occurs using except for [(L5)Cu]2@, that basically shows no activity [(L4)Cu]2@ is an all ligand-based process where the Cu center (Figures S45 and S51). This can be attributed to the coordi- remains at oxidation state II. Hence, the tuning of the redox nation ability of the sulfonate group situated in the ortho active aryl group should have a strong impact into the position, which has the right spatial disposition to generate catalytic properties. Due to the high modularity of the ligand a weak coordination bond with the Cu center, as can be seen synthesis, the family of complexes indicated in Figure 4 in its crystal structure (Figure S35). This contact can partially constitutes an excellent platform to test how electronic disfavor the OH@ coordination needed for catalysis by perturbations exerted by electron-donating and electron- stabilizing the Cu–Namide bond (Figure S64). withdrawing groups as well as by p-extended systems such as A kinetic analysis of the electrocatalytic performance of naphthyls influence the electrocatalytic water oxidation catalysts [(LN)Cu]n@ was undertaken based on FOWA. performance. Complexes [(L4)Cu]2@, [(L6)Cu]2@, [(L7)Cu]2@ and [(L8)Cu]2@ were analyzed in aqueous solution at pH 11.6 whereas for complexes [(L1-L3)Cu] we used a mixture of 40 % TFE and 60 % aqueous pH 11.6, due to the their low solubility in water, and all the results are shown in Figure S49. A plot of kobs (kobs = TOFmax) vs. h is offered in Figure 5 B. As can be observed, the kobs obtained are in the range of 5–35 s@1. For the two phenyl substituted complexes [(L4)Cu]2@ and [(L6)Cu]2@ the increase in rates correlates with an increase of Figure 4. Influence of the substituents in [(LN)Cu]n@ (n = 0 or 2) overpotential, as expected from linear free energy relation- complexes on the h (mV) of the water oxidation reaction. ships. However, from the graph it is further evident that the naphthyl substituted complexes have higher rates than the phenyl substituted ones in spite of the higher overpotential The CV of complexes [(L1-L8)Cu]n@ (except [(L5)Cu]2@) and thus higher thermodynamic driving force for water recorded at pH 11.6 are similar to that of [(L4)Cu]2@, showing oxidation found for the latter. This is consistent with the an electrocatalytic wave at anodic potentials associated with lower reorganization energy of the naphthyl fragment versus the water oxidations catalysis (Figures S45 and S47). How- the phenyl group upon electron transfer due to the more ever, the substituent clearly influences the overpotential at extended p-delocalized system and also with the potentially which catalysis occurs, as can be seen in Figure 4 and closer proximity of the Cu-HOOHC group to the aromatic ring Figure 5 A (also in Figures S45–S47). Within the family of in the 2c-3 e@ species.[48–50] Both complexes [(L7)Cu]2@ and phenyl substituents (H2LNn@, where N = 1, 2, 4 and 6) [(L8)Cu]2@ feature subsequent oxidations in the ligands electron-donor moieties decrease the overpotential (h) where the resulting spin distribution is delocalized within 18642 www.angewandte.org T 2021 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH Angew. Chem. Int. Ed. 2021, 60, 18639 – 18644 Angewandte Research Articles Chemie the naphthalenyl amides as shown by DFT (Figure S68), Acknowledgements following similar trends to [(L3)Cu] complex. Faradaic efficiencies of 52 % and 40 % for [(L7)Cu]2@ and The authors gratefully acknowledge the support from Min- [(L8)Cu]2@ respectively were obtained that are a bit lower isterio de Ciencia, Innovaciln y Universidades and FEDER than that of [(L4)Cu]2@, indicating again that the naphthyl (PID2019-111617RB-I00 and PID2019-104171RB-I00), groups are easier to oxidize/degrade than the corresponding AGAUR 2017-SGR-1631, Ministerio de Ciencia e Inovaciln phenyl group (Figures S52–S55 and S27, S28). for a Severo Ochoa Excellence Accreditation grant 2020– 2023 (CEX2019-000925-S, MIC/AEI), DFG research fellow- ship (OL-562) and EU-funded LICROX FET project (Grant Conclusion agreement 951843). P.G.-B. acknowledges the “La Caixa” Foundation for the Ph.D. grant. The CuII complexes of general formula [(LN)CuII] (N = 1–3) and [(LN)CuII]2@ (N = 4–8) containing aryl substituted bpy-diamidate ligands have proven to be an excellent plat- Conflict of interest form to evaluate ligands effects on water oxidation catalytic properties. The electron transfer processes to enable O@O The authors declare no conflict of interest. bond formation in this family of CuII complexes occurs exclusively on the ligand while the Cu centers remain at Keywords: first-row transition metal complexes · oxidation state II. reaction mechanisms · redox non-innocent ligand · At the initial stages of the catalytic cycle, the CuII water oxidation catalysis · water splitting complexes undergo two one-electron oxidation process followed by the critical O@O bond formation step. The role of the Cu center here is mainly as a scaffold that holds the two [1] R. Matheu, P. Garrido-Barros, M. Gil-Sepulcre, M. Z. Ertem, X. bpy-amide ligand bonded to the metal center as it undergoes Sala, C. Gimbert-SuriÇach, A. Llobet, Nat. Rev. Chem. 2019, 3, different denticity along the intermediates involved in the 331 – 341. catalytic cycle, and at the same time the Cu center allows for [2] J. D. Blakemore, R. H. Crabtree, G. W. Brudvig, Chem. Rev. OH@ coordination and activation. Further, once the OH@ is 2015, 115, 12974 – 13005. coordinated, the Cu-OH group will be attacked by exogenous [3] M. D. K-rk-s, O. Verho, E. V. Johnston, B. ckermark, Chem. Rev. 2014, 114, 11863 – 12001. OH@ via a radical-nucleophilic pathway, generating an oxy- [4] I. Roger, M. Shipman, M. Symes, Nat. Rev. Chem. 2017, 1, 0003. gen-oxygen bond leading to a Cu-(HOOH) species. This [5] N. S. Lewis, Science 2016, 351, 353 – 361. process involves an initial intramolecular single ET event [6] D. G. Nocera, Acc. Chem. Res. 2017, 50, 616 – 619. between the phenyl (or naphthyl) radical cation and the initial [7] A. Melis, Energy Environ. Sci. 2012, 5, 5531 – 5539. Cu-OH group, and therefore the metal center is responsible [8] I. Godwin, A. Rovetta, M. Lyons, J. Coleman, Curr. Opin. for placing the two groups sufficiently close so that fast ET Electrochem. 2018, 7, 31 – 35. can occur. The aryl substituent plays a critical role both from [9] X. Sala, S. Maji, R. Bofill, J. Garc&a-Antln, L. Escriche, A. Llobet, Acc. Chem. Res. 2014, 47, 504 – 516. a thermodynamic perspective, because it will tune the over- [10] P. Garrido-Barros, C. Gimbert-SuriÇach, R. Matheu, X. Sala, A. potential at which catalysis occurs, and from a kinetic Llobet, Chem. Soc. Rev. 2017, 46, 6088 – 6098. perspective, because of the low reorganization energy of [11] S. Berardi, S. Drouet, L. Franc/s, C. Gimbert-SuriÇach, M. extended polyacenes upon ET, which can ensure fast kinetic Guttentag, C. Richmond, T. Stoll, A. Llobet, Chem. Soc. Rev. processes. 2014, 43, 7501 – 7519. The “all ligand-based” strategy reported here is extremely [12] L. Duan, F. Bozoglian, S. Mandal, B. Stewart, T. Privalov, A. Llobet, L. Sun, Nat. Chem. 2012, 4, 418 – 423. attractive because of the large variety of organic substituents [13] R. Matheu, M. Z. Ertem, J. Benet-Buchholz, E. Coronado, V. S. that can be potentially used together with the straightforward Batista, X. Sala, A. Llobet, J. Am. Chem. Soc. 2015, 137, 10786 – synthesis of both the ligands and their CuII complexes. This 10795. therefore opens up a completely new avenue for the design of [14] N. Vereshchuk, R. Matheu, J. Benet-Buchholz, M. Pipelier, J. new ligand-based water oxidation catalysts that eventually Lebreton, D. Dubreuil, A. Tessier, C. Gimbert-SuriÇach, M. Z. could lead to metal free WOCs. Ertem, A. Llobet, J. Am. Chem. Soc. 2020, 142, 5068 – 5077. A key challenging feature that remains unsolved here is [15] B.-T. Chen, N. Morlan8s, E. Adogla, K. Takanabe, V. O. Rodionov, ACS Catal. 2016, 6, 4647 – 4652. the low FE obtained for these complexes that is due to [16] H. Lee, X. Wu, L. Sun, ChemSusChem 2020, 13, 3277 – 3282. competitive deactivation pathways. We are confident that the [17] P. Garrido-Barros, I. Funes-Ardoiz, S. Drouet, J. Benet-Buch- synthetic versatility of molecular transition metal complexes holz, F. Maseras, A. Llobet, J. Am. Chem. Soc. 2015, 137, 6758 – ensured by the large variety of potential ligands that can be 6761. used will lead to rugged and efficient all ligand-based WOCs [18] S. M. Barnett, K. I. Goldberg, J. M. Mayer, Nat. Chem. 2012, 4, in the near future. 498 – 502. [19] D. L. Gerlach, S. Bhagan, A. A. Cruce, D. B. Burks, I. Nieto, H. T. Truong, S. P. Kelley, C. J. Herbst-Gervasoni, K. L. Jernigan, M. K. Bowman, S. Pan, M. Zeller, E. T. Papish, Inorg. Chem. 2014, 53, 12689 – 12698. [20] K. J. Fisher, K. L. Materna, B. Q. Mercado, R. H. Crabtree, G. W. Brudvig, ACS Catal. 2017, 7, 3384 – 3387. Angew. Chem. Int. Ed. 2021, 60, 18639 – 18644 T 2021 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH www.angewandte.org 18643 Angewandte Research Articles Chemie [21] C. J. Gagliardi, A. K. Vannucci, J. J. Concepcion, Z. Chen, T. J. [35] A. De Aguirre, P. Garrido-Barros, I. Funes-Ardoiz, F. Maseras, Meyer, Energy Environ. Sci. 2012, 5, 7704 – 7717. Eur. J. Inorg. Chem. 2019, 2109 – 2114. [22] P. Garrido-Barros, D. Moonshiram, M. Gil-Sepulcre, P. Pelosin, [36] C. Costentin, S. Drouet, M. Robert, J.-M. Sav8ant, J. Am. Chem. C. Gimbert-SuriÇach, J. Benet-Buchholz, A. Llobet, J. Am. Soc. 2012, 134, 11235 – 11242. Chem. Soc. 2020, 142, 17434 – 17446. [37] C. Costentin, J.-M. Sav8ant, ChemElectroChem 2014, 1, 1226 – [23] C. Gimbert-SuriÇach, D. Moonshiram, L. Franc/s, N. Planas, V. 1236. Bernales, F. Bozoglian, A. Guda, L. Mognon, I. Llpez, M. A. [38] R. Matheu, S. Neudeck, F. Meyer, X. Sala, A. Llobet, Chem- Hoque, L. Gagliardi, C. J. Cramer, A. Llobet, J. Am. Chem. Soc. SusChem 2016, 9, 3361 – 3369. 2016, 138, 15291 – 15294. [39] T. Zhang, C. Wang, S. Liu, J. L. Wang, W. Lin, J. Am. Chem. Soc. [24] J. Shi, Y. H. Guo, F. Xie, Q. F. Chen, M. T. Zhang, Angew. Chem. 2014, 136, 273 – 281. Int. Ed. 2020, 59, 4000 – 4008; Angew. Chem. 2020, 132, 4029 – [40] M. T. Zhang, Z. Chen, P. Kang, T. J. Meyer, J. Am. Chem. Soc. 4037. 2013, 135, 2048 – 2051. [25] J. T. Muckerman, D. E. Polyansky, T. Wada, K. Tanaka, E. Fujita, [41] The MECP was not explicitly calculated due to technical Inorg. Chem. 2008, 47, 1787 – 1802. reasons, as the doublet structure always goes automatically to [26] P. Garrido-Barros, C. Gimbert-SuriÇach, D. Moonshiram, A. the broken-symmetry solution. Picln, P. Monge, V. S. Batista, A. Llobet, J. Am. Chem. Soc. 2017, [42] A. A. Voityuk, S. F. Vyboishchikov, Phys. Chem. Chem. Phys. 139, 12907 – 12910. 2020, 22, 14591 – 14598. [27] H.-Y. Du, S.-C. Chen, X.-J. Jiao, M.-T. Zhang, J. Am. Chem. Soc. [43] A. B. P. Lever, Inorg. Chem. 1990, 29, 1271 – 1285. 2018, 140, 1557 – 1565. [44] R. Matheu, M. Z. Ertem, C. Gimbert-SuriÇach, J. Benet- [28] M. D. K-rk-s, B. ckermark, Dalton Trans. 2016, 45, 14421 – Buchholz, X. Sala, A. Llobet, ACS Catal. 2017, 7, 6525 – 6532. 14461. [45] F. C. Anson, T. J. Collins, T. G. Richmond, B. D. Santarsiero, J. E. [29] I. Funes-Ardoiz, P. Garrido-Barros, A. Llobet, F. Maseras, ACS Toth, B. G. T. Treco, J. Am. Chem. Soc. 1987, 109, 2974 – 2979. Catal. 2017, 7, 1712 – 1719. [46] T. Storr, P. Verma, R. C. Pratt, E. C. Wasinger, Y. Shimazaki, [30] A. R. Modi, J. H. Dawson, Adv. Exp. Med. Biol. 2015, 851, 63 – T. D. P. Stack, J. Am. Chem. Soc. 2008, 130, 15448 – 15459. 81. [47] S.-J. Choi, S.-M. Park, J. Electrochem. Sci. Technol. 2010, 1, 1 – 9. [31] J. H. Dawson, Science 1988, 240, 433 – 439. [48] J. L. Br8das, D. Beljonne, V. Coropceanu, J. Cornil, Chem. Rev. [32] I. G. Denisov, T. M. Makris, S. G. Sligar, I. Schlichting, Chem. 2004, 104, 4971 – 5003. Rev. 2005, 105, 2253 – 2277. [49] N. E. Gruhn, D. A. Da Silva Filho, T. G. Bill, M. Malagoli, V. [33] Deposition Numbers 2023402 (for [(L1)CuII]), 2023403 (for Coropceanu, A. Kahn, J. L. Br8das, J. Am. Chem. Soc. 2002, 124, [(L2)CuII]), 2023404 (for [(L3)CuII]), 2023405 (for [(L4)CuII]2@), 7918 – 7919. 2023406 (for [(L5)CuII]2@), 2023407 (for [(L6)CuII]2@), 2023408 [50] V. Coropceanu, M. Malagoli, D. A. da Silva Filho, N. E. Gruhn, (for [(L7)CuII]2@), and 2023409 (for H2L5) contain the supple- T. G. Bill, J. L. Br8das, Phys. Rev. Lett. 2002, 89, 275503. mentary crystallographic data for this paper. These data are provided free of charge by the joint Cambridge Crystallographic Data Centre and Fachinformationszentrum Karlsruhe Access Structures service www.ccdc.cam.ac.uk/structures. [34] H. N. Kagalwala, L. Tong, R. Zong, L. Kohler, M. S. G. Ahlquist, Manuscript received: March 22, 2021 T. Fan, K. J. Gagnon, R. P. Thummel, ACS Catal. 2017, 7, 2607 – Revised manuscript received: May 17, 2021 2615. Accepted manuscript online: May 20, 2021 18644 www.angewandte.org T 2021 The Authors. Angewandte Chemie International Edition published by Wiley-VCH GmbH Angew. Chem. Int. Ed. 2021, 60, 18639 – 18644