Fig. 1.Structures of catalysts 1-5. RuII(bda)(pic)2 1; RuII(Br-bda)(pic)2 2; RuII(2Br-bda)(pic)2 3; RuII(bda)(isoq)2 4; RuII(Br-bda)(isoq)2 5.

Scheme 1Synthetic routes for the preparation of complexes 2, 3 and 5.

Cyclic voltammetry (CV) and differential pulse voltammetry (DPV) were recorded on an IJCambria CHI-660 potentiostat or a Bro-Logic SP-150 potentiostat using a three-electrode setup. CV measurements were carried out at a scan rate of 100 mV/s. DPV experiments were carried out with the parameters: pulses height = 50 mV, pulses width = 50 ms, step high = 4 mV. A glass carbon electrode (2 mm in diameter) was used as a working electrode, a platinum wire as an auxiliary electrode, and an Ag/AgCl electrode as a reference electrode. The CVs were obtained in a CF₃SO₃H aqueous solution (pH = 1.0) containing 33% of CF₃CH₂OH. [Ru(bpy)₃]²⁺ was used as a reference with the E_1/2 (Ru^II/Ru^III) being 1.26 V vs. NHE.

The Pourbaix diagrams were obtained in an aqueous Britton-Robinson buffer solution (0.1 mol/L) containing catalysts (1.0 mmol/L) in the pH range of 1.5-10. The pH values of the solutions were adjusted by the addition of an aqueous NaOH solution (0.2 mol/L).

The oxygen evolution was recorded with an oxygen sensor (Ocean Option, Oxygen Sensing System, NEFOX-GT, NFGO 540), and the concentration of oxygen was calibrated by gas chromatography (GC-2014 Shimadzu) equipped with a thermal conductive detector, a 5 Å molecular sieve column and with Ar as a carrier gas.

To determine the kinetic order, we recorded oxygen evolution as a function of time at the initial stages of the reactions by adding 1.5 mmol/L Ce(IV) oxidant to solutions containing 0.2-6.0 μmol/L of the catalyst. The reaction orders of catalysts were determined by plotting the initial logarithmic [rate] vs. logarithmic [Ru]. All the experiments were carried out in a 0.1 mol/L aqueous triflic acid solution at 25 °C with a total volume of 2 mL. Ultraviolet visible-Vis data were obtained using a Cary 300 Bio UV-Vis spectrophotometer.

All chemicals and solvents were purchased from Sigma-Aldrich (St. Louis, MO, USA). All synthetic reactions were carried out under an N₂ atmosphere. The compounds Ru(DMSO)₄Cl₂, 2,2′-bipyridine-6,6′-dicarboxylic acid, 4-Br-6,6′- dimethyl-2,2′-bipyridine, and 4,4′-diBr-6,6′-dimethyl-2,2′- bipyridine were prepared according to the procedures reported in the literature ^{[30, 31, 32, 33]}. All the chemicals used are commercially available.

2.2.2. Synthesis of 4, 4′-diBr-6, 6′-dicarboxylic acid-2, 2′-bipyridine (H₂L2)

Using the same synthesis and purification methods of compound L1, compound L2 was obtained as a white solid with a yield of 67.5%. ¹H NMR (CD₃OD, 400 MHz): δ 8.71 (s, 2H), 7.36 (s, 2H). HRMS (ESI): found m/z = 401.9950 [M+H]⁺, calcd.: 401.9954.

Fig. 2.(a) Cyclic voltammograms (CVs) of complexes 1-3. (b) Differential pulse voltammograms (DPVs) of complexes 1-3. (c) CVs of complexes 4 and 5. (d) DPVs of complexes 4 and 5. (Conditions: [catalyst] = 1.0 mM, a mixed CF3CH2OH/H2O (v/v = 1/2, pH = 1.0) as solvent, scan rate = 100 mVs−1, Ag/AgCl (3.0 M, KCl, aqueous) as the reference electrode and ENHE = EAg/AgCl + 0.21 V.

The CVs and DPVs of complexes 1-5 were measured in a mixed aqueous solution of CF₃CH₂OH/H₂O (1/2, v/v, pH = 1.0) with Ag/AgCl (3.0 mol/L, KCl) electrode as the reference electrode and Pt wire as the counter electrode and are shown in Fig. 2. Detailed electrochemical data of these complexes are shown in Table 1. Complexes 1-5 displayed two redox waves from 0.4 to 1.1 V (vs. NHE) in the CV measurements (Fig. 2(a) and 2(c)). The first reversible peaks from 0.4 to 0.6 V are assigned to Ru^II/[Ru^III-OH₂]⁺ redox couples. The other reversible couples at about 1.0 V (vs. NHE) are assigned to [Ru^III-OH₂]⁺/[Ru^IV-OH]⁺ redox couples. The electronic effect of Br substituent on the equatorial ligand results in the anodic shift of the oxidation potentials of Ru^II/Ru^III and Ru^III/Ru^IV couples.

The DPV studies of complexes 1-5 show similar results to those of the CVs. The DPV curves are shown in Figure 2b and 2d. Another redox wave at 1.2 V is assigned to the Ru^IV/Ru^V redox couples of complexes 3 and 5. For the Ru-bda catalysts, a catalytic current is expected after [Ru^V=O]⁺ was generated, but the active intermediate [Ru^V =O]⁺ can be overlapped by the catalytic current of a more active catalyst that cannot be captured. In addition, complexes 2 and 3 exhibited lower catalytic currents than complex 1, indicating that the catalytic efficiency of catalyst 1 was slightly higher than those of 2 and 3. A similar trend was also displayed by 4, showing a slightly higher efficiency than that of 5. The electrochemical data suggest that the Br substituent causes an anodic shift of the oxidation potentials of catalysts, and the onset potentials of catalytic water oxidations.

Table 1 Redox potentials of complexes 1-5. Cat. E1/2(RuII/III) (V vs. NHE a) E1/2(RuIII/IV) (V vs. NHE a) E1/2(RuIV/V) (V vs. NHE a) Eonset (V vs. NHE a) 1 0.458 0.938 — 1.070 2 0.502 0.954 — 1.080 3 0.534 1.066 1.180 1.190 4 0.556 1.040 — 1.240 5 0.600 1.052 1.250 1.270 Reaction conditions: CF3SO3H (pH = 1.0)/CF3CH2OH = 2/1 mL/mL, reaction temperature 20 °C. a Ru(bpy)32+ was used as a reference with E1/2 = 1.26 V vs. NHE.

Table 1 Redox potentials of complexes 1-5.

Fig. 4. Oxygen evolution curves for catalysts 1-5. (a) 5.88×10−5 mol/L of catalysts 1-3; (b) 1.5×10−5 mol/L of catalysts 4 and 5. Conditions: 3.4 mL of aqueous solution (0.1 mol/L CF3SO3H, pH = 1.0) containing 0.526 mol/L of CeIV.

The Pourbaix diagrams (E_1/2 vs pH) of complexes 2, 3 and 5 were studied by DPV in aqueous Britton-Robinson buffer solutions at different pH values (2-10) and are shown in Fig. 3. Similar to the reported results of catalysts 1 and 4 ^{[34, 35, 36]}, the segments of Ru species of different valency are divided by the redox potential verses pH lines. The redox processes of Ru^II/Ru^III, Ru^III/Ru^IV and Ru^IV/Ru^V of these complexes occurred in a large pH range ^[37]. The Ru^II/Ru^III oxidation process of catalyst 2 involves electron transfer at pH < 6.2 (slope = 0 mV) and accompanies proton-coupled electron transfer at 6.2 < pH < 10 (slope = 52 mV/pH). The subsequent oxidation process is accompanied with a single-proton-coupled electron transfer (Ru^III/Ru^IV) at 2 < pH < 10 (slope = 42 mV/pH). Further oxidation from [Ru^IV =O]⁺ to [Ru^V =O]⁺ only involves electron transfer as expected, whereas the oxidation of [Ru^IV-OH]⁺ is accompanied by proton transfer.

The potential/pH diagrams for complexes 3 and 5 display almost identical variation trends relative to that of complex 2. The introduction of Br substituent groups results in the enhanced electron withdrawing ability of the equatorial ligands and leads to a more positive potential deflection of the four segments of Ru species. However, the redox potential of the Ru^IV/Ru^V wave is not clear in catalyst 3 at pH < 5, possibly because the catalyst was deposited on the electrode surface. We conclude that the electron withdrawing groups on the equatorial ligand substantially increase the redox potential of catalysts 2, 3, and 5 in an acidic solution. Therefore, the redox processes of complexes 2, 3, and 5 at pH 1.0 are proposed to be [Ru^II] → [Ru^III-OH₂]⁺ → [Ru^IV-OH]⁺ → [Ru^V=O]⁺ before water oxidation.

Fig. 3.Pourbaix diagrams of complexes (a) 2, (b) 3 and (c) 5.

The catalytic activities for water oxidation were evaluated in aqueous solutions (pH = 1.0) containing a catalyst and Ce^IV. The evolved oxygen was measured with an oxygen sensor and gas chromatography, and is shown in Fig. 4. Catalyst and Ce^IV (0.526 mol/L) were dissolved in 3.4 mL CF₃SO₃H solution (pH = 1.0) containing 1% CF₃CH₂OH. The turnover numbers of catalysts 1, 2 and 3 were 1807, 1059 and 773, respectively. Under the same conditions, catalyst 1 showed a higher catalytic activity that stops increasing after 30 min, but catalysts 2 and 3 showed lower catalytic activities. The maximum turnover frequencies of catalysts 1, 2 and 3 were 4, 2.5, and 2 s⁻¹, respectively. These results show that the catalytic abilities become less efficient after replacing the -H atoms with Br groups.

Compared with catalyst 4, a notable decrease of reaction rate and catalytic activity were observed for catalyst 5 as shown in Fig. 4(b). The turnover number of catalyst 5 was 6680, which was approximately 2000 less than the corresponding value for catalyst 4 under the same experimental conditions. The maximum turnover frequencies of catalysts 4 and 5 were 40 and 20 s⁻¹, respectively, which also indicates the undesirable effect of Br groups in the catalysts. The results show that the -Br substituent on the equatorial ligand of catalyst 5 causes a negative effect on the catalytic activity of water oxidation.

Fig. 5. CeIV decay as monitored at 360 nm after the addition of different concentrations of catalysts 1-5 into a CeIV (1.5 mmol/L) aqueous solution (pH 1.0, containing 0.1 mol/L CF3SO3H).

Considering the catalytic water oxidation performances of complexes 1-5, the electronic effect of Br groups is a disadvantage. To explore the reasons of this low efficiency, kinetic studies were performed. Using Ce^IV as the oxidant, we have previously demonstrated that the rate law of Ce^IV consumption is zero order under low [Ce^IV] conditions ([Ce^IV] < 2.0 mmol/L), whereas the mechanism of water oxidation is a bimolecular coupling reaction (second-order) in [Ru(bda)(L)₂] (L = N-donor ligands) ^{[34, 35]}. As shown in Fig. 5, the kinetic measurements were carried out by monitoring the decay of the absorbance of Ce^IV at a wavelength of 360 nm upon addition of catalysts. Catalysts 1-3 at concentrations of 0.8, 2, 4 and 6 μmol/L, and catalysts 4 and 5 at concentrations of 0.2, 0.5, 0.8 and 1.0 μmol/L were injected in a Ce^IV (1.5 mmol/L) aqueous solution (pH 1.0, containing 0.1 mol/L CF₃SO₃H), and the initial rates were calculated by linear fitting the data from 2 to 10 s. No data were collected in the first 2 s because of the operations for injecting the catalyst and stirring. The rate vs. [Cat.] order was estimated by tracking the slope of ln(Rate) vs. ln([Cat.]), which are shown in Fig. 6. The slope value of catalyst 1 was 2.08. A second-order reaction process was confirmed, which agrees with our previous report. The slope values of catalysts 2 and 3 were 1.79 and 1.61, which were lower than that of catalyst 1. This result implies that the catalytic processes of 2 and 3 do not follow a normal second-order reaction. The lower kinetic order indicates that catalysts 2 and 3 involve first-order catalytic processes. Thus, the Br groups in the catalysts exhibit more activity as a first-order reaction process. The possible explanation is that the electronic effect of the substituted groups on the equatorial ligand influences the mechanisms of O-O bond formation of water oxidation. The electron withdrawing effect of Br groups can partially change the mechanism of water oxidation from a bimolecular coupling reaction to a water nucleophilic attack reaction.

Similar to the kinetic measurements of complexes 2 and 3, the slope of catalyst 5 was 1.90, which is only slightly lower than that of catalyst 4. This result shows that catalyst 5 inclines to follow a bimolecular coupling reaction, but some follow a water nucleophilic attack reaction. The second-order kinetics of complex 5 can be explained by the fact that the water oxidation reaction is limited by the electronic effect of Br substituent on the equatorial ligand and the π bond conjugated effect of isoquinoline. Based on a previous report, the π-bond conjugation effect of isoquinoline is conducive to bimolecular coupling reactions ^[28]. Therefore, the kinetic order reveals that the electronic effect of Br substituents on the equatorial ligand in catalyst 5 is weaker than the influence of the conjugation effect isoquinoline on the mechanism of water oxidation.