Increasing energy consumption and environmental pollution caused by burning of traditional fossil fuels have stimulated global efforts to develop clean energy alternatives to fossil fuels [1, 2]. Water splitting, involving the cathodic hydrogen evolution reaction and anodic oxygen evolution reaction (OER), is considered to be a highly efficient method for producing renewable and clean hydrogen energy [3-7]. However, because of its intrinsically sluggish reaction kinetics and the requirement for a high overpotential, the OER is the limiting step in water-splitting reactions [8, 9]. The development of electrocatalysts with high OER activities is therefore needed to improve the reaction rate and reduce the overpotential. At present, complexes and oxides based on iridium and ruthenium are the best performing OER catalysts [10-16]. However, the scarcity and high cost of these noble metals greatly limits their large-scale practical applications. It is therefore important to identify new efficient and cheap electrocatalysts for the OER. Recently, water oxidation catalysts based on non-precious metals, such as cobalt [17-23], nickel [24-28], copper [29-40], iron [41-43], and manganese [44, 45] have been reported.

Because of their good redox properties and low cost, copper-based electrocatalysts show great potential in various catalytic applications. Several Cu(Ⅱ)-based homogeneous catalysts for the OER have been reported in various buffered solutions with different pHs, including Cu(Ⅱ)-bipyridyl [46, 47] and Cu(Ⅱ)-polypeptide [48] complexes. Highly efficient, stable copper-based heterogeneous electrocatalysts for the OER have also been reported and are listed in Table 1. In 2014, Du's group [29] reported that heterogeneous copper oxide electrodeposited from copper-tris(2-pyridylmethyl)amine was an active catalyst for the OER, with an overpotential of ∼600 mV (1.0 mA/cm²). Du's group [31] also reported the use of an electrodeposited copper-based film as a bifunctional catalyst; the catalytic overpotential was 749 mV (1.0 mA/cm²) for the OER. In 2015, Sun's group reported a robust copper oxide film[32], prepared by a simple electrodeposition of Cu²⁺ from a borate buffer solution, which featured a current density of 1.0 mA/cm² at an overpotential of 530 mV in the OER. More recently, Chen's group [33, 36] reported that CuO films electrodeposited from Cu-en and Cu-Gly complexes gave overpotentials of 550 and 450 mV, respectively, at a catalytic current density of 1.0 mA/cm² in 0.2 mol/L phosphate buffer solution (PBS, pH = 12.0).

All chemical reagents used in this study, including CuSO₄·5H₂O, Na₂HPO₄·2H₂O, NaOH, N-[tris(hydroxymethyl) methyl] glycine (tricine), and Amberlite® IRC-748 were purchased from commercial suppliers. The chemicals were of analytical reagent grade and used without further purification. Indium tin oxide-(ITO-) coated glass slides were purchased from Zhuhai Kaivo Optoelectronic Technology Co., Ltd. (thickness ~1.6 mm, transmittance > 70%, resistance < 15 mΩ/cm²). All aqueous solutions were prepared with Milli-Q ultrapure water (> 18 MΩ·cm) unless stated otherwise. The ITO-coated glass was washed by ultrasonication in aqueous detergent solution, acetone, ethanol, and deionized water for 30 min each, and then dried in an oven for future use.

The PBS (0.2 mol/L, pH = 12.0) was prepared by dissolving Na₂HPO₄ (0.2 mol/L) in deionized water and the pH value of the solution was adjusted to 12.0 by addition of NaOH (solid). To rule out the influence of other metal impurity ions on the experiment, the PBS solution was washed with Amberlite® IRC-748. The washing procedure for the electrolyte was as follows. First, the Amberlite resin was stirred for 45 min in ultrapure water and then filtered to remove any unbound iminodiacetic acid. After drying in air, 5.0 g of the cleaned Amberlite resin was added to 400 mL of electrolyte and the mixture was stirred for 5 min. In this time, the resin was removed by filtration and the electrolyte was reserved for further use.

The copper complex was formed in situ by adding an aqueous solution of ligand (10.0 mL of 0.3-mol/L tricine) to an aqueous solution of CuSO₄ (10.0 mL, 0.3 mol/L) in molar ratio corresponding to the Cu/ligand stoichiometry of the final complex. The formation of the copper(Ⅱ) complex was confirmed by UV/Vis spectrophotometry. Immediately, prior to deposition, the solutions were diluted in PBS (0.2 mol/L, pH = 12.0) to form a 1.0 mmol/L Cu-tricine complex in 0.2-mol/L PBS.

Solution UV/Vis spectra of the Cu-tricine complex were measured on a Varian 300 UV/Vis spectrophotometer in 1.0 cm cuvettes. The copper(Ⅱ) complex was diluted in PBS (0.2 mol/L, pH = 12.0) for analysis.

The electrodeposition of the Cu-tricine films was performed in a two-compartment electrochemical cell with a glass frit junction to separate the two compartments. In the auxiliary chamber, the counter electrode was immersed in 30.0 mL PBS (0.2 mol/L, pH = 12.0). In the working electrode compartment, an ITO-coated glass electrode, with a 1.0-cm² electroactive area, was immersed in 30 mL PBS (0.2 mol/L, pH = 12.0) containing 1.0 mmol/L Cu-tricine complex. An Ag/AgCl reference electrode was also immersed in the working electrode compartment. After controlled-potential electrolysis for 6.0 h at 1.15 V (vs. normal hydrogen electrode (NHE)), without iR compensation or stirring, a layer of the catalyst film was formed on the ITO substrate.

The morphology of the prepared Cu-tricine films was observed with FEI Nova Nano SEM 450 instrument equipped with an energy-dispersive X-ray (EDX) analysis detector. Scanning electron microscope (SEM) images were obtained at an acceleration voltage of 3–5 kV and EDX spectra were obtained at accelerating voltages between 15 and 20 kV. Transmission electron microscopy (TEM) and high resolution TEM (HRTEM) images were obtained with FEI TF30 equipment. The sample used for this measurement was carefully scraped off with the use of a doctor blade. The combined material was dispersed in absolute ethanol by ultrasonication, and a drop of the mixture was dried on a carbon-coated copper grid for analysis. X-ray diffraction (XRD) results were collected with a D/max-2400 diffractometer. The measurements were performed from 10° to 90° at a scan-rate of 8°/min. X-ray photoelectron spectroscopy (XPS) was performed on a Thermo VG ESCALAB 250X. The results were obtained with the use of an aluminum anode (Al K_α = 1486.6 eV) operating at 150 W and a spot size of 500 μm. Typically, the hydrocarbon C 1s line at 284.6 eV from adventitious carbon was used for energy referencing. Survey scans were measured at a constant pass energy of 50.0 eV.

All electrochemical experiments were performed at room temperature (25 ± 2 ℃) with a CH Instruments (model CHI 660E Electrochemical Analyzer). The electrochemical measurements were performed in aqueous solution (35.0 mL) with a conventional three-electrode configuration: with the ITO (1.0 cm²) and catalyst film as the working electrode, saturated Ag/AgCl (0.20 V vs. NHE) as the reference electrode, and a Pt mesh as the auxiliary electrode. The reference electrode was calibrated against Ru(bpy)₃ (1.26 V vs. NHE) before use. Cyclic voltammetry (CV) was performed in static PBS (0.2 mol/L, pH = 12.0) between 0.2 and 1.5 V vs. NHE (without iR compensation). The scan rate was 50 mV/s. Long-term controlled-potential electrolysis (CPE) was recorded in 35.0 mL PBS (0.2 mol/L, pH = 12.0) at 1.15 V vs. NHE in a fritted cell with gentle stirring. The Tafel plot of the Cu-Tris film was constructed by measuring the steady-state current density (after 120 s of electrolysis) with iR compensation in 25.0 mL PBS (0.2 mol/L, pH = 12.0) as a function of the applied overpotential. The solution was gently stirred during this experiment. The working electrode was prepared by depositing the Cu-tricine film on 1.0 cm² ITO via CPE for 6 h at 1.15 V vs. NHE. The experimental setup was a single gas-tight cell equipped with copper-coated ITO (1.0-cm² active area) as the working electrode, Ag/AgCl as the reference electrode, and Pt mesh as the counter electrode in PBS (0.2 mol/L, pH = 12.0). First, the PBS was degassed by bubbling Ar for 1.0 h. Then electrolysis was performed at 6.0 mA/cm² without iR compensation in PBS (0.2 mol/L, pH = 12.0) for 3.0 h. The amount of oxygen evolved in the headspace was measured by a gas chromatograph (GC 7890T instrument equipped with a thermal conductive detector) to be 166.5 μmol and the charge passed was 64.8 C (namely, 168.2 μmol of oxygen gas) during electrolysis. The Faradic efficiency (%) was calculated as O₂(actual)/O₂(theoretical)×100.

In this study, a Cu-tricine complex, formed in situ, was used as the precursor for electrodepositing a nanostructured catalyst film (i.e., a Cu-tricine film), which featured a high OER activity in 0.2 mol/L PBS. Fig. 1(a), (b), and (c) show the ultraviolet-visible spectra and electrochemical CV curves measured for various Cu²⁺/tricine ligand ratios in PBS. The results show that the optimum concentrations were 1.0 mmol/L Cu²⁺ and 1.0 mmol/L tricine. These values were used to electrodeposit the catalyst film in subsequent experiments. Fig. 1(d) shows that a low background current density was measured in the absence of Cu²⁺ and/or the tricine ligand. When Cu²⁺ ions alone were added to PBS, Cu(OH)₂ and Cu₃(PO₄)₂ precipitates formed, which were not active towards water oxidation. Similarly, the addition of tricine alone did not enhance the OER activity. Water oxidation reactivity was clearly observed in the CV curve when both tricine and Cu²⁺ were added to PBS. A copper-derived electrocatalyst film was formed in PBS (0.2 mol/L, pH = 12.0) containing 1.0 mmol/L Cu²⁺ and 1.0 mmol/L tricine within 6.0 h, while a working electrode potential of 1.15 V vs. NHE was maintained in the controlled-potential experiments. Photographs (the inset in Fig. 2(a)) show the black catalyst film that was electrodeposited on the ITO substrate.

The morphology of the prepared Cu-tricine catalyst film was examined by SEM and the images are shown in Fig. 2. A dense nanostructure can be clearly observed on the ITO substrate. A cross-sectional image of the film, shown in Fig. 3, indicated that the average thickness of the catalyst film was approximately 1.4 μm. TEM and high-resolution TEM images (Fig. 4) of the Cu-tricine film scraped from the ITO substrate showed that the catalyst film was amorphous. The amorphous nature of the electrodeposited film was confirmed by XRD. Fig. 5 shows that the XRD pattern of the prepared film contained no peaks indicative of crystalline phases other than the peaks associated with the ITO substrate.

The composition of the prepared Cu-tricine film was determined by EDX and XPS measurements. The EDX data for the catalyst film (Fig. 6) showed that the film deposited in PBS contained mainly the elements Cu, O, and In. The Cu and O peaks likely derived from decomposition of the Cu complex, and In derived from the ITO substrate. The peaks in the XPS survey in Fig. 7(a) were all accounted for by the elements detected by EDX. Fig. 7(b) shows a high-resolution spectrum of the Cu 2p. Peaks from 2p_3/2 and 2p_1/2 occurred at 934.2 and 953.9 eV, respectively, with additional intense satellite peaks from Cu(Ⅱ). The binding energies of these Cu 2p peaks were intermediate between those of CuO and Cu(OH)₂ and the signals could be attributed to a mixture of CuO/Cu(OH)₂. The C signal in the XPS survey data (Fig. 7(c)) can presumably be attributed to surface contamination of the sample by air. The two sub-peaks at 529.5 and 531.2 eV in the high-resolution O 1s spectrum (Fig. 7(d)) were assigned to oxygen atoms of oxide and hydroxide, respectively. This result confirms that the electrodeposited Cu-based catalyst film was a mixture of CuO and Cu(OH)₂.

The activity of the Cu-tricine catalyst film for the OER was evaluated by CV. Fig. 8(a) shows that the catalyst film exhibits excellent performance for electrocatalytic water oxidation. The catalytic current density grew rapidly and reached > 15.0 mA/cm² at 1.4 V vs NHE. A bare ITO electrode, which had no appreciable catalytic activity from 0.2 to 1.50 V vs NHE, was used for a control experiment. The long-term performance and stability of the Cu-tricine electrocatalyst film were investigated by performing controlled-potential water electrolysis in PBS (0.2 mol/L). Under an applied potential of 1.15 V vs NHE, a stable catalytic current density for water oxidation of approximately 7.5 mA/cm² was achieved (Fig. 8(b)). No obvious OER catalytic activity was observed with the use of a bare ITO plate as the working electrode during electrolysis under the same conditions (Fig. 8(b)). These results clearly show that the electrodeposited Cu-tricine film performed as a highly efficient and stable water oxidation catalyst in PBS.

Gas bubbles were clearly observed on the surface of the Cu-tricine film electrode during electrolysis. The electrolysis products were further explored by calculating the Faradaic efficiency, based on the catalytic performance of the Cu-tricine film catalyst at 6.0 mA/cm² in a gas-tight electrochemical cell in PBS (0.2 mol/L). Gas chromatography showed that the amount of oxygen evolved during the OER during electrolysis for 3 h was 166.5 μmol. The theoretical amount of oxygen evolution at a current density of 6.0 mA/cm² was calculated to be 168.2 μmol, assuming that the total charge arose from 4e^– oxidation of water, based on Faraday's law. Thus, the amount of oxygen produced matched the theoretical amount well, and corresponded to a Faradaic efficiency of ∼99% (Fig. 9). This result shows that the catalytic current density was mainly consumed by water oxidation leading to oxygen production.

The kinetics of the reaction catalyzed by the Cu-tricine film were explored based on a Tafel plot for the electrocatalyst in PBS (0.2 mol/L), with 100% iR compensation, under stirring. The steady-state current densities (i) obtained by electrolysis for 120 s at various potentials were evaluated as a function of the overpotential (η) as η = V_appl – iR – E_pH, where V_appl is the applied potential, i is the steady-state current, R is the uncompensated resistance, and E_pH is the thermodynamic potential for water oxidation at this pH (E_pH = 1.23 V – 0.059 pH V vs. NHE). Fig. 10 shows that a current density of 1.0 mA/cm² required η = 395 mV. A nearly linear relationship, with a slope of 46.7 mV/decade, was observed from 0.315 to 0.395 V, indicating that the water oxidation catalyst featured efficient kinetics. Thus, the catalytic activities and Tafel slopes of the reported Cu-based catalysts (Table 1) indicate that the Cu-tricine catalyst films exhibited efficient activity for water oxidation.

A Cu-tricine complex formed in situ served as a precursor for electrodepositing a nanostructured amorphous catalyst film for water oxidation. The results show that Cu-tricine is a good and robust catalyst in PBS (0.2 mol/L, pH 12.0), giving a catalytic current density of 1.0 mA/cm² at an overpotential of 395 mV. A Faradaic efficiency of ∼99% for water oxidation was achieved with the use of the Cu-tricine film as the working electrode at a current density of 6.0 mA/cm². A Tafel plot at pH 12.0 was constructed with a slope of 46.7 mV/decade. The use of metal complexes as precursors for electrodeposition can avoid metal salt precipitation in buffer solution and enable deposition of highly efficient and stable catalyst films.