1 Introduction

The oxygen evolution reaction (OER) is an important reaction in modern industry. This reaction plays an important role in hydrogen production, electro-organic syntheses, and metal electrowinning processes. OER requires the distribution of four redox processes: the coupling of multiple proton and electron transfers, and the formation of two O–O bonds [1]. In an electrochemical system, OER occurs at high positive potentials when aqueous solution is used as the electrolyte. The slow reaction kinetics due to the involvement of four electrons in the reaction mechanism is the main obstacle of OER. As OER is the primary reaction at the anode, a large amount of the energy in the electrochemical system will be dedicated to it. Therefore, the continued development of catalysts with low overpotential and fast reaction kinetics with high stability to improve OER efficiency remains a scientific and technological challenge [2, 3].

Coated electrodes are now extensively used in the electrochemical field because they have both the mechanical properties of the substrate and the electrochemical properties of the coating. As one of the more traditional coatings, lead dioxide (PbO2) has attracted considerable attention due to its demonstrated advantages, including low electrical resistivity, low cost, ease of preparation, good chemical stability, and relatively large surface area [4,5,6,7]. Thus, lead dioxide has already been used in waste water treatment [8,9,10], ozone generation [11, 12], lead–acid batteries [13,14,15], analytical sensors [16], and the metal electrowinning process [17]. It is well-known that PbO2 shows two phases: the α phase and β phase. Electrodeposition is a traditional way to prepare these two phases of PbO2 [18]. The conditions (primarily composition and temperature) of the synthesizing bath determine the phase of the deposited material [19]. To the best of our knowledge, the electrochemical activity and electrical conductivity of β-PbO2 is better than that of α-PbO2 [20]. However, as an interlayer of the electrode, the binding force for β-PbO2 and its underlayer can be improved by using α-PbO2 [21].

The OER is sensitive to the intrinsic structure of the catalysts [22]. It has been demonstrated that conductive transition metal oxides can promote the OER with high efficiency and current density in either highly acidic or highly alkaline solutions [23]. Among these metal oxides, Co3O4 shows excellent catalytic activity and with a lower cost. Co3O4 is a p-type semiconductor material with a spinel structure, which is coordinated by Co2+ in a tetrahedron and Co3+ in an octahedron [24]. Due to its high capacity (~ 890 mAh/g), good redox reversibility, strong corrosion resistance, low price, and environmental friendliness, Co3O4 has been widely used for catalysis, gas sensors, lithium batteries, and capacitors [25,26,27,28,29,30,31,32]. As one of the high catalytic activity metal oxides, Co3O4 was incorporated into the PbO2 matrix to improve the reaction rate of the oxygen evolution process [33]. However, PbO2 composite coatings containing Co3O4 particles were found to be considerably rougher compared to the pure PbO2 coatings, which may be caused by the preferred orientation change in the crystal growth process [34].

In this study, β-PbO2–Co3O4 composite coatings were synthesized on a substrate using the electrodeposition method in an electrolyte containing Pb2+ and Co3O4 particles. To improve the binding force of the substrate and the β-PbO2–Co3O4 outer layer, the substrates were covered with an α-PbO2 layer via electrodeposition pretreatment. The influence of the preparation parameters for the composite coating, such as ultrasonic dispersion time and particle concentration of the electrolyte, was investigated using a Zeta potential test and anodic polarization test. The morphology and structure of the β-PbO2–Co3O4 composite layer for varying nucleation times was measured using XRD, SEM, and EDS. To the best of our knowledge, the nucleation process for the β-PbO2–Co3O4 composite on the surface of the α-PbO2 layer has not been thoroughly elucidated to date. In addition, several of the conclusions obtained in this study can explain the unusual roughness of the β-PbO2–Co3O4 composite.

2 Experimentation

2.1 Materials and reagents

A Pb–0.3%Ag alloy plate with dimensions of 40 mm × 20 mm × 2 mm was selected as the substrate. All chemicals were of analytical grade and purchased from Aladdin Industrial Corporation (Shanghai, China). All solutions were prepared by using deionized water supplied from an EPED water purification system. Bath solutions were prepared by dissolving Pb(NO3)2 in HNO3 solution. The concentrations of Pb2+ and HNO3 were controlled at 0.8 M and 0.2 M, respectively.

2.2 Electrode preparation

The Co3O4 doped β-PbO2 electrode consisted of a Pb–0.3%Ag alloy substrate, α-PbO2 inter layer, and β-PbO2–Co3O4 outer layer. First, the Pb–0.3%Ag alloy substrates were pretreated by polishing, degreasing, and acid etching. Second, the α-PbO2 inter layer was electrodeposited in alkaline solution, which consisted of 3 M NaOH and 0.15 M Pb(II). The current density and temperature was controlled at 13.5 mA/cm2 and 40 °C, respectively. The β-PbO2 plating bath consisted of 0.2 M HNO3 + 0.8 M Pb2+ with varying concentrations of Co3O4 particles. Figure 1 shows a schematic diagram for the electrode preparation device. The substrate was contacted to the anode and the stainless steel plates were contacted to the cathode.

Fig. 1
figure 1

Schematic diagram for the electrode preparation device

2.3 Zeta potential and particle size test

The Zeta potential and particle size was measured by analyzing Co3O4 (0.1 g) in solution(10 ml) using the ZetaPALS (Brookhaven). All samples were sonicated for 0–50 min before Zeta potential measurements. The frequency and power of the ultrasonicator was controlled at 40 kHz and 300 W, respectively. The Zeta potential was measured from the electrophoretic mobility obtained using the Smoluchowski equation.

2.4 Anodic polarization curve measurement

The anodic polarization curves were obtained in a synthetic zinc electrowinning simulated electrolyte composed of 50 g/L Zn2+ (added as ZnSO4) and 150 g/L H2SO4, at 35 °C. The scan rates were controlled at 5 mV/s. All of the electrochemical measurements were conducted using a PARSTAT2273 electrochemical workstation with a three-electrode system. The working electrode with an effective area of 1 cm2 was first treated by wax-sealing. The reference electrode was a saturated calomel electrode (SCE), and the counter electrode was a graphite electrode. The reference electrode and working electrode were linked by a Luggin capillary filled with agar and potassium chloride. In addition, the distance between the capillary and working electrode surface was approximately 2d (d is the diameter of the capillary) with d = 0.5 mm.

2.5 Characterization of the surface and phase composition

The phase composition and surface microstructure characteristics for the PbO2 composite layer were measured by a D/Max-2200 X-ray diffractometer (XRD) and Nova NanoSEM450 scanning electron microscope (SEM), respectively.

3 Results and discussion

3.1 Zeta potential measurement for the electrolyte

The Zeta potential and particle size measured in a solution of 0.2 M HNO3 + 0.8 M Pb2+ with different ultrasonic dispersion time are shown in Fig. 2. The measured particle size was reduced from 7.86 to 2.63 μm for an ultrasonic dispersion time of 20 min, which indicates a significant decrease. Following this decrease, the particle size was stabilized at approximately 2.7 μm. This phenomenon is observed because the particles with larger size gradually sank to the bottom of the containing vessel in the first 20 min. The blue line shows the Zeta potential value along with the ultrasonic dispersion time. The Zeta potential value for the Co3O4 particles in this solution gradually increased from 0 to 30 min. When the ultrasonic dispersion time exceeded 30 min, the value for the Zeta potential started to decrease. The absolute value of the Zeta potential is mainly used to characterize the stability of the solution. A greater absolute value for the Zeta potential indicates a more stable solution [35, 36]. Therefore, the β-PbO2 bath containing Co3O4 particles becomes more stable for an ultrasonic dispersion time within 30 min, with the stability beginning to decrease after the ultrasonic time exceeds 30 min. In addition, the Zeta potential values are negative in all the ultrasonic time ranges, indicating that Co3O4 particles are positively charged and that anions are adsorbed onto the particles. Considering the variation law for the particle size and Zeta potential with ultrasonic time, an ultrasonic dispersion time of 30 min is more suitable.

Fig. 2
figure 2

Particle size and Zeta potential for Co3O4 particles in a β-PbO2 plating bath at different ultrasonic dispersion time

3.2 Nucleation of the composite layer

The substrates discussed in this section are Pb–0.3%Ag/α-PbO2, with all of the studies on the nucleation process being conducted for the surface of the α-PbO2 interlayer. In the nucleation study, the oxygen evolution reaction energy-saving(OERES) electrodes were prepared in a β-PbO2 bath with a Co3O4 particle concentration of 15 g/L. The OERES electrodes were obtained using six different electrodeposition time of 0 s, 10 s, 30 s, 2 min, 5 min and 60 min. The XRD patterns for the OERES electrodes prepared at six different electrodeposition time are shown in Fig. 3. Figures 4, 5, 6 and 7 shows the SEM images and EDS analysis for the OERES electrodes prepared at different electrodeposition time.

Fig. 3
figure 3

XRD patterns for OERES electrodes obtained at different deposition time

Fig. 4
figure 4

SEM images for OERES electrodes obtained at different deposition time (1000 × , 10,000 ×)

Fig. 5
figure 5

Energy spectrum analysis for the OERES electrodes for 10 s of electrodeposition

Fig. 6
figure 6

Energy spectrum analysis for OERES electrodes for 5 min of electrodeposition

Fig. 7
figure 7

Energy spectrum analysis for OERES electrodes for 60 min of electrodeposition

It can be seen from Fig. 3 that the XRD pattern at a deposition time of 0 s is actually the diffraction pattern of the substrate; thus, the primary component is α-PbO2, with the α-PbO2 crystal being preferentially crystallized on the (200) crystal plane. When the deposition time was 10 s, the main phase is still α-PbO2, but its diffraction intensity at the (200) plane becomes weak. In addition, there are two peaks for β-PbO2 with low diffraction intensity. According to the SEM images in Fig. 4, the β-PbO2 crystal grains basically cover the surface of the substrate at a deposition time of 10 s. However, since the XRD has a certain depth of penetration, the reflected phase information is still based on the substrate material. When the deposition time was 30 s, the main phase of the XRD pattern is still α-PbO2, and its diffraction intensity at the (200) crystal plane is further reduced. In addition, although the intensity of the β-PbO2 diffraction peak is not high, the number of β-PbO2 peaks increases significantly. For a deposition time of 2 min, the main phases are α-PbO2 and β-PbO2. The diffraction intensity for α-PbO2 in the (200) crystal plane continues to decrease, and the intensity and quantity of the β-PbO2 characteristic peaks are similar to those found at 30 s. Moreover, a small amount of Co3O4 diffraction peaks were captured during this deposition time. When the deposition time was 5 min, the XRD pattern does not notably change compared with the sample obtained using a deposition time of 2 min. The diffraction peak for α-PbO2 does not change significantly. The quantity of the β-PbO2 diffraction peak continues to increase, but the intensity does not show a clear change. When the deposition time was extended to 60 min, the XRD pattern changed very clearly. The main phase was changed to β-PbO2, with a clear preferred crystal orientation for β-PbO2 on the (211), (101) and (110) crystal planes. In addition, the diffraction peaks for Co3O4 can be clearly detected in the composite deposition layer. However, since Co3O4 enters the deposition layer using co-deposition methods, the content is not very high and the intensity of the diffraction peak is still at a low level. In summary, as the deposition time is extended, the intensity and number of α-PbO2 diffraction peaks show a decreasing trend, while the intensity and number of β-PbO2 diffraction peaks show a gradually increasing trend. For a deposition time of 2 min, a diffraction peak for Co3O4 was detected for the first time.

It can be seen from Fig. 4 that the morphology of the substrate is evident for a deposition time of 0 s. The α-PbO2 substrate is composed of circular cells: the circular cell boundaries are clear, with the cell consisting of interwoven rod-shaped crystal grains. For a deposition time of 10 s, the circular shape of the α-PbO2 substrate is still visible, but a layer of β-PbO2 nucleus has formed on the surface, with some positions for the circular cell boundaries not yet fully covered with a β-PbO2 nucleus. This shows that the nucleation of β-PbO2 on the substrate begins with the protrusion of the α-PbO2 circular cell and gradually spreads to the cell boundaries. In addition, a large number of spherical Co3O4 particles are adsorbed onto the surface of the sediment layer. These particles are preferentially gathered at the boundaries of the circular cell. The cell boundaries of α-PbO2 is the position that closer to the power source and thus the electric field force is stronger. In addition, the Zeta potential for the Co3O4 particles in the bath is negative, which has already been demonstrated previously. Thus, the Co3O4 particles with anions adsorbed are affected by the electric field force, leading to preferential adsorption onto the cell boundaries of the α-PbO2 substrate. The analysis indicates that the electric field force around the substrate plays a dominant role in the adsorption of Co3O4 particles. Due to this preferential adsorption, a large number of Co3O4 particles gathered together in the cell boundaries. Moreover, the morphologies of the adsorbed particles and the original Co3O4 particles are identical, indicating that no β-PbO2 is electro-crystallized onto the surface of the particles although some Co3O4 particles have already been adsorbed onto the substrate. It can be concluded that the main process occurring during the deposition time of 0–10 s was nucleation of the β-PbO2 grains and adsorption of Co3O4 particles.

Figure 5 shows the spot scanning energy spectrum for the OERES electrode prepared using a deposition time of 10 s. Point 1 is positioned on the Co3O4 particles close to the substrate, point 2 is positioned on the Co3O4 particles far from the substrate, and point 3 is positioned on the β-PbO2 deposition layer without Co3O4 particles core. Comparing the composition of point 1 and point 2, the Pb content at point 1 is higher than that at point 2, and the Co content is lower than that at point 2, indicating a thicker β-PbO2 deposition layer on the Co3O4 particles close to the substrate. This finding is attributable to the longer growth time of the β-PbO2 grains on the Co3O4 particles closer to the substrate. Conversely, the Co3O4 particles farther from the substrate have a shorter adsorption time; thus, the β-PbO2 crystal grains have a shorter time for nucleation and growth on the surface. Therefore, the β-PbO2 layer on the Co3O4 particles farther from the substrate is thinner.

For a deposition time of 30 s, the circular shape of the α-PbO2 is still visible. Moreover, the β-PbO2 crystal grains completely covered the surface of the substrate and begin to grow. In addition, fine β-PbO2 grains appeared on the Co3O4 particles closest to the surface of the substrate; β-PbO2 grains do not appear on the Co3O4 particles far from the surface of the substrate. It is indicated that the main process occurring during a deposition time of 10–30 s was the growth of β-PbO2 grains on the substrate and the nucleation of β-PbO2 grains on the Co3O4 particles. For a deposition time of 2 min, the circular shape of the α-PbO2 substrate has basically disappeared, with the β-PbO2 crystal grains continuing to grow. Moreover, the grain size for β-PbO2 on Co3O4 particles was not significantly different from that observed at 30 s. This shows that for a deposition time of 30 s–2 min, the main process occurring at the electrode is the growth of β-PbO2 crystal grains on the substrate.

When the deposition time was 5 min, the grain size for β-PbO2 on Co3O4 particles clearly grows, while the β-PbO2 grain size on the substrate does not show obvious changes. This shows that for a deposition time of 2–5 min, the main process occurring at the electrode is the growth of β-PbO2 grains on the Co3O4 particles. Figure 6 shows the spot scanning energy spectrum for the OERES electrode prepared using a deposition time of 5 min. Point 1 is positioned on the β-PbO2 deposition layer without core Co3O4 particles, point 2 is positioned on Co3O4 particles completely covered by β-PbO2, and point 3 is positioned on Co3O4 particles not fully covered by β-PbO2. Comparing points 2 and 3, the Co element content at point 2 is much lower than that at point 3, while the content of Pb at point 2 is higher than that at point 3. It can be inferred that the position of point 3 is a gap in the β-PbO2-coated Co3O4 particles, with the gap gradually closing, similar to point 2, during the deposition process. From the point 2 of Fig. 6 and point 1 of Fig. 5, it can be found that following extension of the deposition time to 5 min, the Pb content of the coated particles became higher and the Co content decreased, indicating that the β-PbO2 layer on the coated Co3O4 particles was significantly thickened at this time, which confirms the analysis based on Fig. 4.

When the deposition time was extended to 60 min, the grain size for β-PbO2 on the substrate increased, but the grain size for β-PbO2 on the Co3O4 particles did not show a significant increase. From SEM photographs taken at low magnification, it can be seen that Co3O4 particles are aggregated on the surface, with very serious agglomeration observed. This shows that for a deposition period of 5–60 min, the main process occurring at the electrode is the growth of β-PbO2 grains on the substrate, continuous adsorption of Co3O4 particles, and deposition of β-PbO2 onto the surface of the Co3O4 particles.

Figure 7 shows the spot scanning energy spectrum for the OERES electrode prepared using a deposition time of 60 min. Point 1 is positioned on the Co3O4 particle farther from the substrate, point 2 is positioned on the Co3O4 particle closer to the substrate, and point 3 is positioned on the β-PbO2 deposition layer without core Co3O4 particles. Comparing point 1 and point 2, the content of Co at point 1 is clealry higher than that at point 2, and the content of Pb at point 1 is lower, indicating a thinner β-PbO2 layer on the Co3O4 particles farther away from the substrate. In addition, the fact that the particle size for the β-PbO2-coated Co3O4 farther away from the substrate is significantly smaller may confirm the possibility discussed above.

From the analysis above, several nucleation laws for the β-PbO2–Co3O4 layer can be obtained. In the β-PbO2 bath containing Co3O4 particles, the electric field force plays a dominant role in the adsorption process for the Co3O4 particles. Therefore, the circular cell boundaries for α-PbO2 are the preferential adsorption regions for Co3O4 particles. In addition, the nucleation of β-PbO2 on the substrate starts at the prominent position of the α-PbO2 circular cell and gradually spreads to the cell boundaries. Finally, the nucleation and growth of β-PbO2 on the Co3O4 particles started from the particles closer to the substrate. The thickness of the β-PbO2 layer on the Co3O4 particles was inversely proportional to the distance between the Co3O4 particles and the substrate.

Based on the composite electrodeposition mechanism (electrochemical mechanism [37,38,39]) and the previous analysis, in which the electric field force plays a dominant role, a nucleation model of the β-PbO2–Co3O4 deposition layer was developed, as shown in Fig. 8. The nucleation process for the β-PbO2–Co3O4 deposition layer can be divided into four steps: step 1: the Co3O4 particles suspended in the β-PbO2 plating bath move from the depth of the plating solution to the vicinity of the substrate surface depending on the flow of the plating solution. Step 2: nucleation of β-PbO2 on the substrate starts at the prominent position of the α-PbO2 circular cell and gradually spreads to the cell boundaries. The second step and the first step do not have a sequence and can be performed at the same time. Step 3: the Co3O4 particles with anionic absorbents are electrophoretically pushed to the surface of the anode and adsorbed on the surface of the anode after reaching the double layer of the anode surface, which is mainly affected by the electric force. Due to the strong electric field at the depression of the circular cell boundaries of α-PbO2, the Co3O4 particles preferentially move to this location and undergo adsorption at this location. Step 4: the β-PbO2 on the substrate continues to grow. At the same time, β-PbO2 nucleates on the adsorbed Co3O4 particles, resulting in gradual but complete coating of the Co3O4 particles with β-PbO2.

Fig. 8
figure 8

Nucleation model of β-PbO2–Co3O4 coating

3.3 OER property for the electrode

The Co3O4 doped β-PbO2 electrodes were prepared in β-PbO2 baths with different Co3O4 particle concentrations(0–20 g/L). The ultrasonic dispersion time for the β-PbO2 baths were controlled at 30 min. In addition, the temperature and time of the deposition process was controlled at 45 °C and 1 h, respectively. The anodic polarization curves for the OERES electrodes measured in a Zn electrowinning simulated system are shown in Fig. 9. The characteristics for the five anodic polarization curves are similar, with the current being almost zero before oxygen evolution and exponentially increasing after oxygen evolution, which are typical OER characteristic curves. With the doping of Co3O4 particles, the initial oxygen evolution potentials for the electrodes were significantly reduced compared with the β-PbO2 electrodes without particles. Furthermore, the initial oxygen evolution potential shows a decreasing trend as the concentration of Co3O4 increases in the β-PbO2 plating bath.

Fig. 9
figure 9

Anodic polarization curves for OERES electrodes in a simulation zinc electrowinning solution obtained from the original plating bath with different Co3O4 concentrations; the Tafel lines in the OER potential range (inset)

Figure 9b shows the Tafel lines (η = a + blgi) in the OER potential range. The kinetic parameters for the Tafel linear fitting and OER overpotential at a current density of 300, 400, 500 and 600 mA/cm2 are shown in Table 1. It can be found from Table 1 that the Tafel intercept (a) and the Tafel slope (b) are significantly reduced as the Co3O4 particles are doped into the electrode. A smaller Tafel slope is more beneficial for energy-saving as it leads to a remarkably increased OER rate with an increase in overpotential [40,41,42]. When the Co3O4 particle concentration in the plating bath is 15 g/L, the oxygen evolution overpotentials calculated at the different current densities were the lowest for the obtained electrode. Compared to the electrode prepared without Co3O4 particles, the oxygen evolution overpotentials for the electrode at 300 A/m2, 400 A/m2, 500 A/m2 and 600 A/m2 are decreased by 346 mV, 363 mV, 377 mV, and 387 mV, respectively, which indicates an energy-saving effect. In addition, a further increase in the particle content in the plating solution will not contribute to an improved energy-saving effect.

Table 1 Kinetic parameters and overpotential for the OER at OERES electrodes in a simulation zinc electrowinning solution obtained from the original plating bath with varying Co3O4 concentration

4 Conclusions

In this study, the stability of a Co3O4 suspension electrolyte was investigated by a Zeta potential test. The β-PbO2 bath containing Co3O4 particles becomes more stable for an ultrasonic dispersion time within 30 min, with the stability beginning to decrease after the ultrasonic time exceeds 30 min. Co3O4 particles in the β-PbO2 bath are positively charged, with anions being adsorbed onto the particles.

XRD, SEM, and EDS were utilized to investigate the nucleation of the composite layer. The electric field force plays a dominant role in the adsorption process for the Co3O4 particles. Therefore, the circular cell boundaries of α-PbO2 are a preferential adsorption region for Co3O4 particles. In addition, the nucleation of β-PbO2 on the substrate starts at the prominent position of the α-PbO2 circular cell and gradually spreads to the cell boundaries. The nucleation and growth of β-PbO2 on the Co3O4 particles started for the particles closer to the substrate. In addition, the thickness of the β-PbO2 layer on the Co3O4 particles was inversely proportional to the distance between the Co3O4 particles and the substrate. Finally, the nucleation process for the β-PbO2–Co3O4 layer can be divided into four steps.

The influence of the Co3O4 concentration on the electrode OER catalytic properties was characterized using anodic polarization curves. The oxygen evolution overpotential for the obtained electrode was the lowest for a Co3O4 particle concentration of 15 g/L in the plating bath. Compared to the pure PbO2 electrode, the oxygen evolution overpotentials for the electrode at 300 A/m2, 400 A/m2, 500 A/m2 and 600 A/m2 are decreased by 346 mV, 363 mV, 377 mV, and 387 mV, respectively, which demonstrates an energy-saving effect.