Cooperative effects in homogenous water oxidation catalysis by mononuclear ruthenium complexes

Abstract

The homogenous water oxidation catalysis by [Ru(terpy)(bipy)Cl]⁺ (1) and [Ru(terpy)(Me₂bipy)Cl]⁺ (2) (terpy = 2,2′:6′,2′′-terpyridine, bipy = 2,2′-bipyridine, Me₂bipy = 4,4′-dimethyl-2,2′-bipyridine) under the influence of two redox mediators [Ru(bipy)₃]²⁺ (3) and [Ru(phen)₂(Me₂bipy)]²⁺ (4) (phen = 1,10-phenanthroline) was investigated using Ce⁴⁺ as sacrificial oxidant. Oxygen evolution experiments revealed that mixtures of both 2–4 and 2–3 produced more molecular oxygen than catalyst 2 alone. In contrast, the combination of mediator 4 and catalyst 1 resulted in a lower catalytic performance of 1. Measurements of the temporal change in the intensity of a UV transition at 261 nm caused by the addition of four equivalents of Ce⁴⁺ to 2 revealed three distinctive regions-suggested to correspond to the stepwise processes: (i) [Ru^IV=O]²⁺ → [Ru^V=O]³⁺; (ii) [Ru^V=O]³⁺ → [Ru^III–(OOH)]²⁺; and (iii) [Ru^III–(OOH)]²⁺ → [Ru^II–OH₂]²⁺. UV-Visible spectrophotometric experiments on the 1–4 and 2–4 mixtures, also carried out with four equivalents of Ce⁴⁺, demonstrated a faster [Ru(phen)₂(Me₂bipy)]³⁺ → [Ru(phen)₂(Me₂bipy)]²⁺ reduction rate in 2–4 than that observed for the 1–4 combination. Cyclic voltammetry data measured for the catalysts and the mixtures revealed a coincidence in the potentials of the Ru^II/Ru^III redox process of mediators 3 and 4 and the predicted [Ru^IV=O]²⁺/[Ru^V=O]³⁺ potential of catalyst 2. In contrast, the [Ru^IV=O]²⁺/[Ru^V=O]³⁺ process for catalyst 1 was found to occur at a higher potential than the Ru^II/Ru^III redox process for 4. Both the spectroscopic and electrochemical experiments provide evidence that the interplay between the mediator and the catalyst is an important determinant of the catalytic activity.

---

Introduction

The splitting of water into oxygen and hydrogen using sunlight has a great potential for developing an alternative renewable energy source that will help to overcome the looming shortfall in fossil fuels. However, despite the promise, to achieve water splitting (2H₂O → O₂ + 2H₂) a substantial energy input additional to the thermodynamic requirement (1.23 V or 267 kJ mol⁻¹ per mole of H₂ produced) is needed to compensate for losses arising from, for example, cell resistance and various activation barriers. In this regard, the mechanistically complex water oxidation half reaction (2H₂O → O₂ + 4H⁺ + 4e⁻) presents particular challenges as it involves the extraction of four electrons and four protons from two water molecules and the formation of a stable diatomic O–O bond. Thus, efficient catalysts need to be developed such that the energy required to drive the water-splitting reaction is as close as possible to the thermodynamic value (1.23 V). There has been much interest in this field of research and efforts have been made to develop heterogeneous photo-catalysts as well as photo-electrocatalysts for water splitting.¹ Single semiconductors or combinations of semiconductors with appropriate valence and conduction band positioning (Z-scheme) have been used to split water directly and more active catalysts have been introduced to enhance performance (these include metals and a variety of transition metal oxides).^1,2 In parallel with these studies, there has been a long-standing interest in molecular catalysts that oxidise water or reduce water, with a focus on the former being driven in part by the fact that a tetramanganese cluster in water oxidation complex (WOC) of photosystem II is the only catalyst known to catalyse water oxidation in nature.³ Amongst the range of molecular water oxidation catalysts that have been reported,⁴ however ruthenium complexes have been investigated for several decades now and, in contrast to some purported molecular catalysts incorporating other transition metals, there is clear evidence that such complexes are indeed molecular catalysts operating in homogeneous solution and when attached to surfaces.^5–9 Initially, the focus was on dinuclear complexes owing to the discovery that the ruthenium blue dimer (cis,cis-[(bipy)₂(H₂O)Ru^III(μ-O)Ru^III(OH₂)(bipy)₂]⁴⁺; bipy = 2,2′-bipyridine) catalysed this reaction,⁶ and a mechanistic understanding of the pathways leading to oxygen formation was developed from these studies.⁷ More recently, mononuclear ruthenium complexes have also been shown to catalyse water oxidation.^5,8–11 They do so by forming a reactive [Ru^V=O] intermediate, generated through successive oxidation processes [Ru^II–OH₂] → [Ru^III–OH] → [Ru^IV=O] → [Ru^V=O].^5,8 The formation of the [Ru^V=O] moiety is central to the catalytic evolution of oxygen, which has been proposed to occur through either the attack of [Ru^V=O] by water or, alternatively, through intra- or inter-molecular coupling of two [Ru^V=O] species.^5,8 These mononuclear ruthenium complexes have largely been investigated using Ce⁴⁺ as a sacrificial oxidant,^5,8,9,11 and although investigations of the interactions between potential photosensitisers (or dyes) and the ruthenium catalysts is very important for the development of a light-induced water-splitting system, to date only a handful of reports have focussed on this aspect in homogenous systems.^9,10 For example, the catalytic rate of Ce⁴⁺-assisted water oxidation by the blue dimer in homogenous solution was enhanced by a factor of ∼30 on addition of mediators such as [Ru(bipy)₂L]²⁺, where L = 2,2′-bipyridine, 2,2′-bipyrimidine and 2,2′-bipyrazine.⁹ In a more recent report, a covalently-linked mediator/catalyst dinuclear complex [(bipy)₂Ru(4-Mebpy-4′-bimpy)Ru(terpy)(OH₂)]⁴⁺ or [Ru_a^II–Ru_b^II–OH₂]⁴⁺ (4-Mebpy-4′-bimpy = 4-(methylbipyridin-4′-yl)-N-benzimid-N′-pyridine; terpy = 2,2′:6′,2′′-terpyridine) displayed a faster catalytic rate than the individual mononuclear complex [Ru(terpy)(Mebim-py)(OH₂)]²⁺ (Mebim-py = 2-pyridyl-N-methylbenzimidazole).¹⁰

The relative lack of investigations focussing on cooperativity between ruthenium catalysts and redox-active ruthenium polypyridyl complexes has led us to study the interplay between two tris(diimine)ruthenium complexes {[Ru(bipy)₃]²⁺ (3) and [Ru(phen)₂(Me₂bipy)]²⁺ (4) (phen = 1,10-phenanthroline, Me₂bipy = 4,4′-dimethyl-2,2′-bipyridine)} as redox mediators and two closely-related mononuclear catalysts [Ru(terpy)(bipy)Cl]⁺ (1) and [Ru(terpy)(Me₂bipy)Cl]⁺ (2) in Ce⁴⁺-activated homogenous water oxidation catalysis. Building on previous related investigations of mononuclear complexes of type [Ru(terpy)(xbipy)(H₂O)]²⁺ (xbipy = derivatives of 2,2′-bipyridines with various substitutions on the 4,4′ position),¹¹ we report an investigation of the influence of redox mediators 3 and 4 on water oxidation mononuclear catalysis by 1 and 2 (Scheme 1).

Scheme 1

Results and discussion

Following our reports describing the synthesis of the stereochemically-resolved Δ-[Ru(phen)₂(Me₂bipy)]²⁺ and the non-chiral [Ru(terpy)(Me₂bipy)Cl]⁺ for the study of their biological activity,^12,13 in the present work we adopted the same synthetic pathway to obtain complexes 1 and 2. These compounds were prepared by refluxing mixtures of [Ru(terpy)Cl₃] and 2,2′-bipyridine or 4,4′-dimethyl-2,2′-bipyridine in ethanol–water (4 : 1), while 4 was obtained from the reaction between [Ru(phen)₂Cl₂] and 4,4′-dimethyl-2,2′-bipyridine in ethanol–water (1 : 1). Size-exclusion chromatography was used to separate each complex from the unreacted starting material and any oxidised ruthenium by-products. The purity of the compounds was confirmed by ¹H NMR spectroscopy.

A series of oxygen evolution experiments using catalysts 1 and 2 as a 1 : 1 mixture with either 3 or 4 at the concentration 73 μM was carried out in an air-tight vessel using 0.1 M HClO₄ solution as the reaction medium and (NH₄)₂[Ce(NO₃)₆] (0.073 M or 1000 equivalents) as the oxidant. The molecular oxygen evolved from the reaction was detected in the headspace over the course of several hours by a Clark-type micro-sensor. The oxygen measurement for the individual catalysts revealed that 2 produced approximately 16 μmol oxygen after six hours whereas 1 produced twice as much (30 μmol). In order to test the effect of the redox mediators 3 and 4 on the catalytic performance of 1 and 2, the oxygen evolution of each catalyst in the presence of the mediators was investigated. Upon mixing with 4 in a 1 : 1 ratio, oxygen production catalysed by 1 was found to decrease slightly to 23 μmol over the six hours of testing. In contrast, the addition of mediator 4 to catalyst 2 resulted in a significant increase in oxygen production from 16 μmol to 38 μmol over the same period. No molecular oxygen could be detected from a control experiment using only mediator 4 (Fig. 1). The enhancement of the catalytic properties of 2 affected by the redox mediator prompted us to perform further analyses in order to elucidate the mode of interaction between the tris(diimine)ruthenium(II) complexes and the catalyst. Our initial investigations using UV-Visible, ¹H NMR and electrochemical techniques are discussed in the following sections.

Fig. 1 (Top) Oxygen evolution from the reaction between the complexes and Ce⁴⁺ in 0.1 M HClO₄ detected in the headspace. (Bottom) Turnover numbers (TONs). Blue = 2–4 mixture; green = 1 only; black = 1–4 mixture; orange = 2–3 mixture; purple = 2 only; red = 4 only. The error bars of the full and crossed lines are in grey. In all cases, the concentration of each complex and Ce⁴⁺ were 73 μM and 0.073 M, respectively.

The changes in the UV-Visible spectrum of the individual catalysts and the mixtures, before and after the addition of four equivalents of Ce⁴⁺, were studied in 0.1 M HClO₄. The initial UV-Visible spectrum of catalyst 2 (before Ce⁴⁺ was added) showed three absorption bands (480 nm, 313 nm and 281 nm). The absorption at 480 nm, assigned as MLCT (metal-to-ligand charge transfer) band, was found to disappear upon addition of four equivalents of Ce⁴⁺ (Fig. 2, panel a). This spectral assignment has been reported earlier for the reaction of the closely-related [Ru(terpy)(bpm)(OH₂)]²⁺ complex with Ce⁴⁺ in 0.1 M HClO₄.⁸ The addition of two equivalents of Ce⁴⁺ to the solution of [Ru(terpy)(bpm)(OH₂)]²⁺ resulted in the two oxidation steps that could be followed spectroscopically:

[Ru^II–OH₂]²⁺ + Ce⁴⁺ → [Ru^III–OH]²⁺ + Ce³⁺ (1)

[Ru^III–OH]⁺ + Ce⁴⁺ → [Ru^IV=O]²⁺ + Ce³⁺ (2)

Fig. 2 Spectral changes before and after addition of four equivalent Ce⁴⁺ in (a) 2 only; (b) 2–4 mixture and (c) 1–4 mixture. In all cases the complex concentration was 24 μM.

Further addition of one equivalent of Ce⁴⁺ to the [Ru^IV=O]²⁺ species (or three equivalents to [Ru^II–OH₂]²⁺) triggered three stepwise redox processes. Three distinctive regions based on the absorbance changes at 283 nm were observed and attributed to the reactions given in eqn (3)–(5).⁸

[Ru^IV=O]²⁺ + Ce⁴⁺ → [Ru^V=O]³⁺ + Ce³⁺ (3)

[Ru^V=O]³⁺ + H₂O → [Ru^III–(OOH)]²⁺ + H⁺ (4)

[Ru^III–(OOH)]²⁺ → [Ru^II–OH₂]²⁺ (5)

The transient [Ru^V=O]³⁺ ion has been proposed to react further with water to give what is tentatively assigned as the [Ru^III–OOH]²⁺ species which slowly decays to [Ru^II–OH₂]²⁺ possibly by releasing O₂.⁸ In the present investigation, three regions were indeed observed at the slightly lower wavelength (261 nm), compared with 283 nm in the previous study, suggesting that catalyst 2 also undergoes the stepwise redox processes described by eqn (3)–(5) after the addition in this case of four equivalents of Ce⁴⁺ (see Fig. 3, top panel). Oxygen can be released if further Ce⁴⁺ is added to the peroxido species [Ru^III–OOH]²⁺ according to eqn (6) and (7).⁸ This was demonstrated in the present study by using an excess (1000 equivalents) of Ce⁴⁺ in the oxygen evolution experiment mentioned earlier.

[Ru^III–(OOH)]²⁺ + Ce⁴⁺ → [Ru^IV–(O₂)]²⁺ + Ce³⁺ + H⁺ (6)

[Ru^IV–(O₂)]²⁺ + H₂O → [Ru^II–(OH₂)] + O₂ (7)

Fig. 3 (Top) Absorbance changes at 261 nm in 2 after the addition of four equivalent of Ce⁴⁺ showing three distinctive changes with time. (Bottom) Absorbance changes at 261 nm measured on addition of four equivalents of Ce⁴⁺ to a mixture of catalyst 2 and 4. No clear evidence was found for the redox processes identified for 2 alone.

The UV-Visible spectrum of the mixture of catalyst 2 and 4 before the Ce⁴⁺ addition showed the MLCT band at 455 nm and three other bands in the UV regions (313, 281 and 264 nm). As observed for the individual catalyst, the MLCT band also disappeared upon adding the oxidant (Fig. 2, panel b). In addition to the three redox processes observed for the individual catalyst, oxidation of the mixture with four equivalents of Ce⁴⁺ is also expected to trigger the formation of [Ru(phen)₂(Me₂bipy)]³⁺ according to eqn (8).

[Ru(phen)₂(Me₂bipy)]²⁺ + Ce⁴⁺ → [Ru(phen)₂(Me₂bipy)]³⁺ + Ce³⁺ (8)

As in the study with 2 alone, the absorbance changes at 261 nm was also monitored for the 2–4 mixture to see whether the features associated with (3), (4) and (5) could be identified on the same time scale. These features were absent (Fig. 3, bottom) partly because of the coincidental rise in a strong ligand-centred absorption at 264 nm as is typically observed in this region for similar tris(diimine)ruthenium(II) complexes.¹⁴ Alternatively, it is possible that the oxidation step (3) is greatly accelerated by [Ru(phen)₂(Me₂bipy)]³⁺ such that it could not be detected on the minute time scale. The acceleration of reaction (3) by [Ru(phen)₂(Me₂bipy)]³⁺ (see eqn (9)) during the oxidation of 2 by Ce⁴⁺ is consistent with the enhanced oxygen production by 2 in the presence of 4, as shown in Fig. 1.

[Ru(phen)₂(Me₂bipy)]³⁺ + [Ru^IV=O]²⁺ → [Ru^V=O]³⁺ + [Ru(phen)₂(Me₂bipy)]²⁺ (9)

A ¹H NMR experiment was carried out on the mixture of 2–4 in D₂O–CD₃CN to further probe the processes occurring on addition of Ce⁴⁺ (Fig. 4). The characteristic C6 proton resonance of the bipy ligand, located between 9 to 10 ppm, is affected by the adjacent chlorido or aqua ligand in [Ru(terpy)(xbipy)Cl]⁺ or [Ru(terpy)(xbipy)(OH₂)]²⁺ is a useful tool to identify the presence of [Ru^II–Cl]⁺ or [Ru^II–OH₂]²⁺ species.^11,13 As shown in Fig. 4, the C6 proton at 9.45 ppm of 2 disappears on addition of Ce⁴⁺ to the mixture because of the formation of paramagnetic higher-valency ruthenium species. Over a period of 180 minutes, [Ru^II–OD₂]²⁺ was clearly generated as indicated by the reappearance of the C6 proton signal.

Fig. 4 Time dependence of the evolution of the ¹H NMR spectrum of a 1 : 1 mixture of 2 and 4 in CD₃CN–0.1 M HClO₄ in D₂O (1 : 4) (4.22 mM) before and after addition of four equivalent Ce⁴⁺.

The spectrum of the mixture before Ce⁴⁺ addition shows a proton resonance at 9.45 ppm, attributed to the C6 Me₂bipy proton adjacent to the chloro ligand in [Ru(terpy)(Me₂bipy)Cl]⁺ (2). The proton resonance located nearby (9.4 ppm) is attributed to the corresponding C6 proton of Me₂bipy of the solvation product, [Ru(terpy)(Me₂bipy)(OD)₂)]²⁺.^11,13 Integration of the two proton resonances reveals that the mixture contains approximately 90% [Ru(terpy)(Me₂bipy)Cl]⁺ and 10% [Ru(terpy)(Me₂bipy)(OD₂)]²⁺. Both proton resonances disappear following the addition of Ce⁴⁺. The spectrum measured at 180 minutes after the addition of Ce⁴⁺ shows the appearance of the C6 proton at 9.4 ppm, suggesting that the [Ru(terpy)(Me₂bipy)(OD₂)]²⁺ was regenerated but not [Ru(terpy)(Me₂bipy)Cl]⁺.

Although the stepwise processes (3) to (5) in the 2–4 mixture could not be identified from the UV-Visible spectral data, process (5) or the regeneration of [Ru^II–OH₂]²⁺ (or the deuterated forms) was evident from the time-course ¹H NMR data.

In order to investigate the kinetics of the reduction of [Ru(phen)₂(Me₂bipy)]³⁺ to the initial [Ru(phen)₂(Me₂bipy)]²⁺ state with respect to catalyst 1 and 2, the absorbance changes at 264 nm in 1–4 and 2–4 mixtures after Ce⁴⁺ addition were monitored. The UV-Visible spectrum of catalyst 2 before Ce⁴⁺ addition showed no absorption below 270 nm (Fig. 2), and thus the strong 264 nm band is more specific for [Ru(phen)₂(Me₂bipy)]²⁺ and was chosen for the kinetic evaluation. As can be seen from Fig. 5 (top panel), the absorbance changes at 264 nm reflecting the regeneration of [Ru(phen)₂(Me₂bipy)]²⁺ follow first-order exponential kinetics and the rate constants for the [Ru(phen)₂(Me₂bipy)]³⁺ to [Ru(phen)₂(Me₂bipy)]²⁺ reduction in the 2–4 and 1–4 combinations were calculated to be 1.11 (±0.03) × 10⁻³ s⁻¹ and 5.2 (±0.2) × 10⁻⁴ s⁻¹, respectively (Fig. 5, bottom panel). The faster rate of reduction of [Ru(phen)₂(Me₂bipy)]³⁺ to [Ru(phen)₂(Me₂bipy)]²⁺ in the 2–4 combination in comparison to 1–4 suggests that the redox processes of 2 and 4 under Ce⁴⁺ activation may be inter-dependent, thus supporting the cooperative interaction given in eqn (9). In contrast to the ‘cooperative’ effect in the 2–4 mixture, 1 and 4 appear to consume Ce⁴⁺ independently, such that a ‘competitive’ effect leads to a slight decrease in the oxygen evolution performance of 1 (Fig. 1). Accordingly, in this case 4 merely reacts with some of the available Ce⁴⁺ but the [Ru(phen)₂(Me₂bipy)]³⁺ complex produced is not capable of oxidising the catalyst 1 from [Ru^IV=O]²⁺ to [Ru^V=O]³⁺ state.

Fig. 5 (Top) Absorbance changes reflecting the regeneration of the [Ru(phen)₂(Me₂bipy)]²⁺, measured over time at 264 nm after four equivalents of Ce⁴⁺ were added to mixtures of 2–4 (blue dotted line) and 1–4 (red dotted line). The concentration of each complex was 24 μM. The final absorbance values are denoted as A_∞ and the absorbance at a time t as A_t. (Bottom) Analysis of the linear plots of log[A_∞ − A_t] vs. time in 2–4 (blue dotted line) and in 1–4 (red dotted line), assumed to correspond to the decay of [Ru(phen)₂(Me₂bipy)]³⁺, gave first-order rate constants of 1.11 (±0.03) × 10⁻³ s⁻¹ and 5.2 (±0.2) × 10⁻⁴ s⁻¹ for the [Ru(phen)₂(Me₂bipy)]²⁺ recovery in the 2–4 and in 1–4 combinations, respectively.

The cyclic voltammograms of 1, 2, 3 and 4 as individual complexes and in mixtures were measured in CH₃CN–0.1 M HClO₄ (1 : 6) at the concentration 0.48 mM and the data are presented in Fig. 6, 7 and Table 1. The Ru^II/Ru^III oxidation potentials (E_pavs. NHE) of 1 and 2 of 0.97 V and 0.93 V, respectively, lie within the range of those reported for closely-related complexes.¹¹ The Ru^II/Ru^III oxidation potentials (E_pavs. NHE) for 3 and 4 were observed at 1.37 V and 1.39 V, respectively. There were no significant differences in these potentials when measured as mixtures of complexes.

Fig. 6 (Top) Comparison of the cyclic voltammetry (CV) of 1 (black) and 2 (red) showing the Ru^II/Ru^III redox wave and the predicted Ru^III to Ru^IV and Ru^IV to Ru^V oxidations leading to the onset of water oxidation. (Bottom) Comparison of the CV traces showing the [Ru(bipy)₃]^2+/3+ redox wave for 3 (green) and [Ru(phen)₂(Me₂bipy)]^2+/3+ in 4 (red). The traces were recorded at 0.48 mM complex concentrations in CH₃CN–0.1 M HClO₄ (1 : 6) using a three electrode system comprising a glassy carbon working electrode, platinum auxiliary electrode and Ag/AgCl reference electrode at the scan rate 50 mV s⁻¹ at 23 °C.

Fig. 7 (Top) Comparison of the cyclic voltammetry of mixtures of 1–3 (green) and 1–4 (red) in top panel and of 2–3 (green) and 2–4 (red) in bottom panel. Complex concentrations, electrode configuration and CV measurement conditions as indicated in caption to Fig. 6.

Table 1 Assignment of the redox potentials obtained from the CV experiments of the individual complexes and the mixtures, at 0.48 mM concentrations, carried out in CH₃CN–0.1 M HClO₄ (1 : 6) on a glassy carbon electrode at the scan rate 50 mV s⁻¹. E_1/2 = (E_pa − E_pc)/2 in volts relative to NHE; ΔE = |E_pa − E_pc| in mV

Complexes and mixtures	Assignment	E pa (V)	E pc (V)	E 1/2 (V)	ΔE (mV)
a An increase in current starting from 1.37 V for catalyst 2 (Fig. 6, top) reflects the Ru^III/Ru^IV and Ru^IV/Ru^V oxidation processes. b Ru^III/Ru^IV and Ru^IV/Ru^V oxidation processes in 1 predicted to start at 1.55 V.
4 only	Ru^II/Ru^III (4)	1.39	1.23	1.31	160
2–4	Ru^II/Ru^III (4)	1.39	1.23	1.31	160
	Ru^II/Ru^III (2)	0.92	0.86	0.89	60
1–4	Ru^II/Ru^III (4)	1.39	1.24	1.32	150
	Ru^II/Ru^III (1)	0.97	0.91	0.94	60
3 only	Ru^II/Ru^III (3)	1.37	1.30	1.34	70
2–3	Ru^II/Ru^III (3)	1.39	1.32	1.36	70
	Ru^II/Ru^III (2)	0.92	0.86	0.89	60
1–3	Ru^II/Ru^III (3)	1.38	1.30	1.34	80
	Ru^II/Ru^III (1)	0.96	0.90	0.93	60
2 only^a	Ru^II/Ru^III (2)	0.93	0.86	0.90	70
1 only^b	Ru^II/Ru^III (1)	0.97	0.90	0.94	70

---

A close analysis of the cyclic voltammetry of the individual catalysts reveals a significant increase in current starting from 1.37 V (Fig. 6, bottom) in 2 which, as described in eqn (2) and (3), may be attributed to the formation of [Ru^IV=O]²⁺ and [Ru^V=O]³⁺ leading to the onset of water oxidation. The redox waves of the [Ru^IV=O]²⁺ and [Ru^V=O]³⁺ are generally not well-defined as they coincide with the onset of water oxidation. On the other hand, the cyclic voltammetry of 1 reveals that the oxidation processes occurs at much higher potential (1.55 V). The predicted [Ru^IV=O]²⁺ and [Ru^V=O]³⁺ species that are accessible at a reasonably lower potential in 2 may be ‘switched on’ by both redox mediators 3 and 4 as the Ru^II/Ru^III potentials of the two complexes occur in the same range. In contrast, this may be more difficult for catalyst 1 because the higher potential results in a slight increase in the reaction driving force. This postulate is consistent with both the oxygen evolution and the UV-Visible spectrophotometric data presented earlier.

The significant difference in the oxygen evolution performance of 2 in the presence of 3 and 4 is interesting (Fig. 1), and worthy of further comment. From the electrochemistry point of view, although the anodic potentials (E_pa) of both 3 and 4 are identical, the cathodic potential (E_pc) of 4 was significantly lower (see Table 1). When compared with [Ru(bipy)₃]²⁺, it is feasible that the [Ru(phen)₂(Me₂bipy)]^2+/3+ redox process in the acidified, and predominantly aqueous, solution may be accompanied by structural changes that lead to electrochemical irreversibility which manifests itself in a larger ΔE. In addition to the potential matching described earlier, the acceleration of the oxidation of the [Ru^IV=O]²⁺ conversion to [Ru^V=O]³⁺ by [Ru(phen)₂(Me₂bipy)]³⁺ (eqn (9)), we tentatively propose that this may be due to a greater thermodynamic stability of [Ru(phen)₂(Me₂bipy)]²⁺, resulting in faster reduction of [Ru(phen)₂(Me₂bipy)]³⁺ (compared with [Ru(bipy)₃]³⁺ to [Ru(bipy)₃]²⁺), and a concomitantly faster oxidation of [Ru^IV=O]²⁺ to [Ru^V=O]³⁺.

The proposed mechanism for the homogenous catalysis by the recently-reported dinuclear complex [(bipy)₂Ru(4-Mebpy-4′-bimpy)Ru(terpy)(OH₂)]⁴⁺, referred to as [Ru_a^II–Ru_b^II–OH₂]⁴⁺ (a = mediator and b = catalyst) highlighted the role of the redox mediator in accelerating the catalytic process.¹⁰ Following the oxidation of [Ru_a^II–Ru_b^II–OH₂]⁴⁺ to [Ru_a^III–Ru_b^IV=O]⁵⁺ by Ce⁴⁺ (eqn (10)), an intra-molecular electron transfer between the two ruthenium centres leads to a redox equilibrium described in eqn (11).¹⁰ Further reactions (eqn (12)–(14)) follow what was described earlier in eqn (4), (6) and (7).¹⁰

[Ru_a^II–Ru_b^II–OH₂]⁴⁺ + 3Ce⁴⁺ → [Ru_a^III–Ru_b^IV=O]⁵⁺ + 3Ce³⁺ + 2H⁺ (10)

[Ru_a^III–Ru_b^IV=O]⁵⁺ ⇄ [Ru_a^II–Ru_b^V=O]⁵⁺ (11)

[Ru_a^II–Ru_b^V=O]⁵⁺ + H₂O → [Ru_a^II–Ru_b^III(OOH)]⁴⁺ + H⁺ (12)

[Ru_a^II–Ru_b^III(OOH)]⁴⁺ + Ce⁴⁺ → [Ru_a^II–Ru_b^IV(O₂)]⁴⁺ + Ce³⁺ + H⁺ (13)

[Ru_a^II–Ru_b^IV(O₂)]⁴⁺ + H₂O → [Ru_a^II–Ru_b^II–OH₂]⁴⁺ + O₂ (14)

In keeping with this study, an (inter-molecular) electron transfer between [Ru(phen)₂(Me₂bipy)]³⁺ in 4 and [Ru^IV=O]²⁺ in 2 was demonstrated in the present study suggesting a similar mechanism of the cooperative effect proposed in Fig. 8.

Fig. 8 (Top) Proposed mechanism of the mediator/catalyst cooperative effect observed in the 2–4 mixture.^9,10 The catalytic mechanism involving only Ce⁴⁺ is given in the bottom diagram.⁸ The green colour indicates the catalyst and the red indicates the redox mediator.

Conclusions

This study has provided evidence of the enhancement of the water oxidation catalytic properties of 2 under the influence of the redox mediator 4, [Ru(phen)₂(Me₂bipy)]³⁺. The ability of 4 to accelerate the critical oxidation of the catalyst, [Ru^IV=O]²⁺ → [Ru^V=O]³⁺, may be due in part to the slightly better potential match between the two redox events when compared to [Ru(bpy)₃]²⁺ (3). Work is currently underway to further probe such electron transfer processes using ruthenium catalysts and mediators appropriate for immobilisation on electrode surfaces. This accessibility of the electron transfer between the redox mediator (dye) and the catalyst is significant for the design of catalytic assemblies for the photo-induced oxidation of water. For example, mediator-catalyst assemblies can be attached to, for example, dyed-titania semiconductor films to construct photoanodes capable of splitting water with visible light.¹⁵

Experimental section

Materials and synthesis

[Ru(bipy)₃]Cl₂ and (NH₄)₂[Ce(NO₃)₆] were purchased from Sigma Aldrich and used as received. All other chemicals were sourced from commercial suppliers and used without further purification except where otherwise indicated.

The synthesis and characterisation of the two precursors [Ru(terpy)Cl₃] and [Ru(phen)₂Cl₂]^16,17 and the [Ru(phen)₂(Me₂bipy)]Cl₂,¹² [Ru(terpy)(Me₂bipy)Cl]Cl and [Ru(terpy)(bipy)Cl]Cl complexes¹³ have been reported previously and the same methods were used in the present work. In a typical synthesis, solid [Ru(terpy)Cl₃] (0.20 g, 0.45 mmol) and appropriate amounts of the ligands (0.084 g, 0.45 mmol for Me₂bpy and 0.071 g, 045 mmol for bipy) were refluxed in EtOH–H₂O (4 : 1; 40 ml) for 4 h. After cooling, the solvent mixture was evaporated to dryness. The crude product was dissolved in methanol and loaded onto a Sephadex LH-20 exclusion column and the pure compounds were separated from the impurities using methanol as the eluent. Yield of [Ru(terpy)(Me₂bipy)Cl]Cl 0.16 g (60%). To prepare [Ru(phen)₂(Me₂bipy)]Cl₂, [Ru(phen)₂Cl₂] (0.20 g, 0.38 mmol) and Me₂bipy (0.069 g, 0.38 mmol) were refluxed in EtOH–H₂O (1 : 1, 30 ml) for 6 h, after which the bright orange solution was evaporated to dryness and loaded onto a Sephadex LH-20 column. The pure [Ru(phen)₂(Me₂bipy)]Cl₂ compound was separated from the impurities using methanol as the eluent. The yield was 0.24 g (90%). The ¹H NMR data of the compounds were consistent with those reported in the literature.^12,13

Methods

Oxygen evolution

Oxygen measurements were performed using a Unisense OXY-500 microsensor connected to a Unisense OXY Meter. The sensor was calibrated using air and argon for 100% O₂ and 0% O₂, respectively. The signal was processed using Sensor Trace software. To measure oxygen evolution by the individual complexes and the mixtures (1 : 1 complex to redox mediator), each complex and redox mediator (0.51 μmol) was suspended in 0.1 M HClO₄ (6 ml) in an air-tight vessel. Argon was then purged to obtain zero oxygen reading. A solution of (NH₄)₂[Ce(NO₃)₆] (0.28 g, 0.51 mmol) in 0.1 M HClO₄ (1 ml) was charged through a septum after the reading had stabilised for 20 min. The final concentrations of the complexes and the Ce⁴⁺ in the vessel were 73 μM and 0.073 M, respectively. The oxygen sensor was measured in the headspace and the signal was collected for 6 to 10 h.

UV-Visible spectrophotometry

UV-Visible spectra were recorded on a Varian Cary 300 spectrophotometer. To measure the spectra of either individual complexes or the mixture of complexes, each compound (0.51 μmol) was dissolved in distilled water (0.5 ml) and 0.1 M HClO₄ (6.5 ml) was then added. A 1 ml aliquot was taken from this stock into a 1 cm path length cuvette and 1.95 ml of 0.1 M HClO₄ was added. The initial spectrum was recorded prior to adding 0.05 ml (NH₄)₂[Ce(NO₃)₆] in 0.1 M HClO₄. The concentration of the complex and (NH₄)₂[Ce(NO₃)₆] in the cuvette were 24 μM and 97 μM, respectively. The spectra were then recorded in 2, 5, 10 and 15 min intervals for 15 h.

¹H NMR experiment

¹H NMR experiments were performed on a Bruker Avance 400 MHz. Solid [Ru(terpy)(Me₂bipy)Cl]Cl (2.0 mg, 3.4 μmol) and [Ru(phen)(Me₂bipy)]Cl₂ (2.4 mg, 3.4 μmol) were dissolved in CD₃CN (0.2 ml) in an NMR tube and was topped up by 0.1 M HClO₄ in D₂O (0.6 ml). The concentration of each complex was 4.22 mM. The initial spectrum was measured prior to adding the solid (NH₄)₂[Ce(NO₃)₆] (7.4 mg, 14 μmol). The spectrum was then recorded every 10 min for the first 30 min then every 30 min for additional 2.5 h.

Electrochemistry

Electrochemical measurements were performed on a VSP Bio-Logic potentiostat connected to a three-electrode system. Cyclic voltammetry of the individual complexes and the mixtures were measured in CH₃CN–0.1 M HClO₄ (1 : 6) at the scan rate 50 mV s⁻¹ on a glassy carbon working electrode. A platinum wire was used as an auxiliary electrode and Ag/AgCl electrode was used the reference. The concentration of each complex in all cases was 0.48 mM. The observed potentials were corrected relative to NHE.

Acknowledgements

The authors thank the Australian Research Council for financial support received through the Australian Centre of Excellence for Electromaterials Science (CE0561616).

Notes and references

1. M. G. Walter, E. L. Warren, J. R. McKone, S. W. Boettcher, Q. Mi, E. A. Santori and N. S. Lewis, Chem. Rev., 2010, 110, 6446.
2. T. Bak, J. Nowotny, M. Rekas and C. C. Sorrell, Int. J. Hydrogen Energy, 2002, 27, 991; A. Currao, Chimia, 2007, 61, 815; A. Heller, Science, 1984, 223, 1141; R. Brimblecombe, A. Koo, G. C. Dismukes, G. F. Swiegers and L. Spiccia, J. Am. Chem. Soc., 2010, 132, 2892; W. J. Younpblood, S. H. A. Lee, Y. Kobayashi, E. A. Hernandez-Pagan, P. G. Hoertz, T. A. Moore, A. L. Moore, D. Gust and T. E. Mallouk, J. Am. Chem. Soc., 2009, 131, 926; S. Y. Reece, J. A. Hamel, K. Sung, T. D. Jarvi, A. J. Esswein, J. J. H. Pijpers and D. G. Nocera, Science, 2011, 334, 645; D. G. Nocera, Acc. Chem. Res., 2012, 45, 767.
3. K. N. Ferreira, T. M. Iverson, K. Maghlaoui, J. Barber and S. Iwata, Science, 2004, 303, 1831; J. Barber and J. W. Murray, Coord. Chem. Rev., 2008, 252, 233; C. W. Cady, R. H. Crabtree and G. W. Brudvig, Coord. Chem. Rev., 2008, 252, 444; R. Brimblecombe, D. R. J. Kolling, A. M. Bond, G. C. Dismukes, G. F. Swiegers and L. Spiccia, Inorg. Chem., 2009, 48, 7269; R. Brimblecombe, G. F. Swiegers, G. C. Dismukes and L. Spiccia, Angew. Chem., Int. Ed., 2008, 47, 7335.
4. M. Yagi and M. Kaneko, Chem. Rev., 2001, 101, 21; R. Manchanda, G. W. Brudvig and R. H. Crabtree, Coord. Chem. Rev., 1995, 144, 1; W. Ruettinger and G. C. Dismukes, Chem. Rev., 1997, 97, 1; S. Mukhopadhyay, S. K. Mandal, S. Bhaduri and W. H. Armstrong, Chem. Rev., 2004, 104, 3981; M. Yagi, A. Syouji, S. Yamada, M. Komi, H. Yamazaki and S. Tajima, Photochem. Photobiol. Sci., 2009, 8, 139; H. Yamazaki, A. Shouji, M. Kajita and M. Yagi, Coord. Chem. Rev., 2010, 254, 2483.
5. W. Song, A. Ito, R. A. Binstead, K. Hanson, H. Luo, M. K. Brennaman, J. J. Concepcion and T. J. Meyer, J. Am. Chem. Soc., 2013, 135, 1158; K. Kobayashi, H. Ohtsu, T. Wada, T. Kato and K. Tanaka, J. Am. Chem. Soc., 2003, 125, 6729; J. T. Muckerman, D. E. Polyansky, T. Wada, K. Tanaka and E. Fujita, Inorg. Chem., 2008, 47, 1787; Y. M. Badiei, D. E. Polyansky, J. T. Muckerman, D. J. Szalda, R. Haberdar, R. Zong, R. P. Thummel and E. Fujita, Inorg. Chem., 2013, 52, 8845; J. Nyhlen, L. Duan, B. Akermark, L. Sun and T. Privalov, Angew. Chem., Int. Ed., 2010, 49, 1773; L. Duan, A. Fischer, Y. Xu and L. Sun, J. Am. Chem. Soc., 2009, 131, 10397; D. J. Wasylenko, C. Ganesamoorthy, B. D. Koivisto and C. P. Berlinguette, Eur. J. Inorg. Chem., 2010, 3135; L. Duan, F. Bozoglian, S. Mandal, B. Stewart, T. Privalov, A. Llobet and L. Sun, Nat. Chem., 2012, 4, 418.
6. S. W. Gersten, G. J. Samuels and T. J. Meyer, J. Am. Chem. Soc., 1982, 104, 4029; J. A. Gilbert, D. S. Eggleston, W. R. Murphy, D. A. Geselowitz, S. W. Gersten, D. K. Hodgson and T. J. Meyer, J. Am. Chem. Soc., 1985, 107, 3855.
7. X. Yang and M.-H. Baik, J. Am. Chem. Soc., 2006, 128, 7476; D. Moonshiram, J. W. Jurss, J. J. Concepcion, T. Zakharova, I. Alperovich, T. J. Meyer and Y. Pushkar, J. Am. Chem. Soc., 2012, 134, 4625; D. Moonshiram, V. Purohit, J. J. Concepcion, T. J. Meyer and Y. Pushkar, Materials, 2013, 6, 392.
8. J. J. Concepcion, J. W. Jurss, J. L. Templeton and T. J. Meyer, J. Am. Chem. Soc., 2008, 130, 16462; J. J. Concepcion, J. W. Jurss, M. R. Norris, Z. Chen, J. L. Templeton and T. J. Meyer, Inorg. Chem., 2010, 49, 1277.
9. J. J. Concepcion, J. W. Jurss, J. L. Templeton and T. J. Meyer, Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 17632.
10. M. R. Norris, J. J. Concepcion, D. P. Harrison, R. A. Binstead, D. L. Ashford, Z. Fang, J. L. Templeton and T. J. Meyer, J. Am. Chem. Soc., 2013, 135, 2080.
11. D. J. Wasylenko, C. Ganesamoorthy, B. D. Koivisto, M. A. Henderson and C. P. Berlinguette, Inorg. Chem., 2010, 49, 2202.
12. Y. Mulyana, D. K. Weber, D. P. Buck, C. A. Motti, J. G. Collins and F. R. Keene, Dalton Trans., 2011, 40, 1510; F. Li, Y. Mulyana, M. Feterl, J. M. Warner, J. G. Collins and F. R. Keene, Dalton Trans., 2012, 40, 5032; M. J. Pisani, P. D. Fromm, Y. Mulyana, R. J. Clarke, K. Heimann, J. G. Collins and F. R. Keene, ChemMedChem, 2011, 6, 848.
13. Y. Mulyana, J. G. Collins and F. R. Keene, J. Inclusion Phenom. Macrocyclic Chem., 2011, 71, 371.
14. D. Hesek, Y. Inoue, S. R. L. Everitt, H. Ishida, M. Kunieda and M. G. B. Drew, Inorg. Chem., 2000, 39, 308.
15. W. Song, C. R. K. Glasson, H. Luo, K. Hanson, M. K. Brennaman, J. J. Concepcion and T. J. Meyer, J. Phys. Chem. Lett., 2011, 2, 1808; L. Wang, D. L. Ashford, D. W. Thompson, T. J. Meyer and J. M. Papanikolas, J. Phys. Chem. C, 2013, 117, 24250.
16. P. A. Adcock, F. R. Keene, R. S. Smythe and M. R. Snow, Inorg. Chem., 1984, 23, 2336.
17. T. Togano, N. Nagao, M. Tsuchida, H. Kumakura, K. Hisamatsu, F. S. Howell and M. Mukaida, Inorg. Chim. Acta, 1992, 195, 221.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4dt00629a

---