Introduction

Supercapacitors are designed to meet the need for high power in a short time. In supercapacitors, the electrochemical energy is stored in the electrical field at the interface between the electrode and the electrolyte [1,2,3,4,5]. The key component of the supercapacitor electrode is its electrode material because all storage mechanisms depend on its type, structure, morphology, and amount in the electrode [6]. Electrical double-layer capacitors (EDLCs), pseudocapacitors (PCs), and battery-like supercapacitors are three types of supercapacitors, which are classified according to the charge-storage mechanism of the electrode material [6,7,8]. Transition metal oxides (TMOs) and their spinel structure as an electrode material to fabricate pseudocapacitors (PCs) and battery-like supercapacitors have attracted widespread research interest [9,10,11]. Transition metal oxides participate in energy storage through faradaic redox reactions, and their ability to perform these reactions is promoted by reducing their density and exposing a high surface of these materials to electrolytes [12]. Metal oxide spinel structures have attracted a lot of attention as a supercapacitor electrode material compared to single-component metal oxides due to their large ion diffusion pathways and the high surface of electrolyte exposure [13,14,15,16,17,18]. Along with research on other metal oxides, it has been widely demonstrated that cobalt spinels, especially copper–cobalt spinel (CuCo2O4), can be used as a suitable electrode material to improve the electrochemical performance of supercapacitors [19,20,21]. CuCo2O4 is a promising material for use in supercapacitors because this structure enables to exchange of a high number of electrons per mole of oxides as well as higher conductivity than that of copper oxide and cobalt oxide [22]. In this structure, copper compounds show high conductivity and cobalt compounds enhance energy storage capacity. Among the various methods deployed for spinel synthesis [23,24,25,26,27,28,29,30], electrochemical deposition is considered a fast and direct preparation of the supercapacitor electrode method. Electrochemical deposition methods have been widely used due to easy composition control and high deposition rate as well as low cost compared to other multistep routs [30,31,32]. The formation and percentage of spinel phase in the coating structure depend on the synthesis method and the ratio of cobalt and copper elements during deposition. Despite many efforts have been devoted to the synthesis and use of CuCo2O4 spinel nanostructures for supercapacitor electrode materials, a clear identification of the best ratio of cobalt and copper that brings about the better spinel structure is still missing.

In this study, for the first time, we investigate the effect of different ratios of cobalt and copper on the electrochemical performance of CuCo2O4 electrodes for hybrid supercapacitors. We electrodeposited different compositions of Cobalt–Copper hydroxides on Ni foam (NF). The electrochemical deposition was carried out in a solution containing different rations of cobalt (II) and copper (II) sulfates in nitric acid. The reaction mechanism for the formation of Co–Cu hydroxide films is described by Eqs. (1)–(3) [32]:

$${\mathrm{NO}}_{3}^{-} + 7{\mathrm{H}}_{2}\mathrm{O }+ 8{\mathrm{e}}^{-} \to \mathrm{ N}{\mathrm{H}}_{4}^{+} +10{\mathrm{OH}}^{-}$$
(1)
$${2\mathrm{H}}^{+}+ 2{\mathrm{e}}^{-} \to {\mathrm{H}}_{2(\mathrm{g})}$$
(2)
$$x{\mathrm{Cu}}^{2+} + y{\mathrm{Co}}^{2+} + 2\left(x+y\right){\mathrm{OH}}^{-}\to {\mathrm{Cu}}_{x}{\mathrm{Co}}_{y}{(\mathrm{OH})}_{2(x+y)}$$
(3)

CuCo2O4 was obtained by thermal annealing of electrodeposited Cobalt–Copper hydroxide films on NF as described by Eq. (4):

$${\mathrm{Cu}}_{x}{\mathrm{Co}}_{2x}{(\mathrm{OH})}_{6x} + {1/2x\mathrm{O}}_{2} \to x{\mathrm{CuCo}}_{2}{\mathrm{O}}_{4} + 3{\mathrm{H}}_{2}\mathrm{O }$$
(4)

XRD and SEM analysis were used to investigate the phase structure and the surface morphology of electrodeposited electrodes, respectively. Cyclic voltammetry (CV), galvanostatic charge/discharge (GCD), cyclic stability test, and electrochemical impedance spectroscopy (EIS) were conducted to characterize the electrochemical performance of all prepared electrodes.

Materials and experimental methods

Materials

Cobalt (II) sulfate heptahydrate (CoSO4·7H2O) and copper (II) sulfate pentahydrate (CuSO4·5H2O), and Nitric acid (≥ 65%) were purchased from Sigma-Aldrich. Nickel foam was provided by Goodfellow.

Preparation of cobalt oxide (CoO), copper oxide (CuO), and Cobalt–Copper mixed oxides coatings (CC)

Electrodeposition of Cobalt–Copper mixed oxides coating (CC; Table 1) was carried out from electrolyte solutions prepared from analytical grade chemicals (Sigma-Aldrich) and distilled water. Cobalt-containing coatings were obtained from aqueous solutions containing variable concentrations of CoSO4·7H2O and CuSO4·5H2O (Table 1). Pure cobalt oxide coating (CoO) was deposited from a 0.1 M CoSO4·7H2O aqueous solution. In order to adjust the pH to 1.4, HNO3 was added to all solutions. Before electrodeposition, the Ni foam substrates measuring 0.25 × 0.25 cm2 were cleaned with ethanol in an ultrasonic bath and dried in the atmosphere. The electrochemical process was performed in three-electrode cell consisting of a Pt mesh as a counter electrode, a Ni foam (NF) working electrode, and a Ti grid as a quasi-reference electrode. Under galvanostatic conditions with a current density of 1 A cm−2 for 45 s, the electrodeposition was run at room temperature. The average deposited mass of the oxide electrodeposited on Ni foam substrates was measured using Mettler Toledo balance and was 0.8 ± 0.1 mg. After drying, thermal annealing of the as-obtained samples was performed at 400 °C for 2 h under an unprotected atmosphere.

Table 1 Sample identification and chemical composition
Full size table

Material characterization

The phase structure of the electrodeposited coatings was investigated using high-resolution X-ray diffraction (XRD; PANalytical X'Pert PRO powder diffractometer equipped with an X’Celerator detector CuKα radiation, λ = 1.5406 Å, 40 mA, 40 kV). Morphological features and chemical composition of the samples were studied using scanning electron microscopy (SEM; Zeiss EVO 50 electron scanning microscope (SEM) equipped with an energy-dispersive X-ray analyzer (EDS) from Oxford INCA Energy 350 system.).

Electrochemical characterization

The electrochemical performances of all samples have been characterized in a three-electrode cell by cyclic voltammetry (CV) at different scan rates ranging from 5 to 200 mV s─1 at a potential window of − 0.1 and 0.45 V (vs. Ag/AgCl), and GCD tests were performed in a 3 M KOH electrolyte between − 0.1 and 0.45 V at a current density of 1 A g─1 to evaluate cycling stability. The electrochemical impedance spectroscopy (EIS) was measured in a 3 M KOH electrolyte over the frequency range 0.01–100 kHz, with 5 mV AC. All electrochemical measurements have been run with a (VSP multichannel potentiostat/galvanostat/FRA (BioLogic, Seyssinet-Pariset, France)) in a three-electrode cell configuration in which, the electrodeposited electrode was the working electrode, a Pt spring the counter, and Ag/AgCl the reference one.

Results and discussion

Structural and morphological analyses

XRD analysis was administrated on the coatings to see the crystalline phases present in each of them and therefore the results are depicted in Fig. 1. In Fig. 1, the peaks associated with the Ni foam substrate are indicated by “NF”. Regarding the CoO sample (Fig. 1a), diffraction peaks can be observed at 31.2°, 36.8°, 44.8°, 59.3°, and 65.2° indexed to the (220), (311), (400), (511), and (440) planes of the cubic phase of Co3O4 which is in agreement with standard JCPDS data [31, 32]. CuO sample shows well-defined peaks at different angles which are marked with a black symbol in Fig. 1a. consistent with the XRD pattern of CuO, there are two sharper diffraction peaks located at 35.5° and 38.6° corresponding to the (002) and (111) planes of monoclinic CuO structure. Figure 1b presents the patterns of the Cobalt–Copper mixed oxides (CC) coatings, which are drawn in several colors and from CC8 to CC2, respectively. As shown within the Fig. 1b, peaks associated with copper oxide appeared more sharply in samples obtained from Cu-rich solutions (CC8 and CC6), whereas with the rise of cobalt in solution (CC4 and CC2), the intensity of those peaks gradually decreases, and peaks associated with cobalt oxide and Cobalt–Copper spinel structure appear. The peaks indicated with a black symbol are indexed to the (111), (220), (311), (400), (511), and (440) planes of the spinel cubic CuCo2O4 phase, respectively (JCPDS card no. 01–1155). Additionally, the main peaks of the CuxCo3-xO4 spinel structure appeared within the diffraction pattern of all mixed oxides samples. Table 2 shows the intensity, location and corresponding plane of the three main peaks of CuxCo3-xO4 spinel structure for each sample. According to the table and comparison of intensity values, it is well understood that the spinel phase peaks are more evident within the CC4 sample, indicating that more spinel phase has been formed in this coating. These results are consistent with previous reports [32,33,34].

Fig. 1
figure 1

X-ray diffraction patterns of a cobalt oxide (CoO) and copper oxide (CuO) and b Cobalt–Copper mixed oxides (CC2, CC4, CC6 and CC8) deposited on NF, compared with the reference XRD pattern of the spinel CuxCo3-xO4 (in red)

Full size image
Table 2 The intensity, location and corresponding plane of the three main peaks of CuxCo3-xO4 spinel structure for all samples
Full size table

Figure 2 shows low and high magnification SEM images of all electrodeposited coatings, which are arranged from the CoO electrode to the CuO electrode, respectively, based on the increase in the amount of copper in the coating. Morphological features such as particle size, growth process, structure, and pores created are different depending on the composition of cobalt and copper. Figure 2b reveals that in the CoO layer, particles are placed alongside one another to create interconnected fractal structures with a length of 100 µm and grow into cauliflower form. As shown in Fig. 2a, b, and c, samples with higher cobalt content display cauliflower-like structures with sharp surfaces and smaller pores, while samples with more copper content show larger pores with open dendritic walls. (Fig. 2d, e, and f). Dendrites do not present any preferential growth (in relevancy the substrate); they are randomly oriented. In addition, the porous structure created within the coatings is due to hydrogen bubbles on the surface during electrodeposition. The constant evolution of H2 bubbles during the deposition limits the free volume available for metal(s) growth, forcing it to occur around the evolving bubbles. The concentration of elements in all samples which is taken by EDS can be observed in Fig. 2g. Sample identification and average Co/Cu composition ratio of CC2 to CC8 electrodes as determined by EDS are presented in Table 3. The EDS results confirm the targeted compositions in cobalt and copper.

Fig. 2
figure 2

SEM images for a CoO, b CC2, c CC4, d CC6, e CC8, f CuO, and g elements composition in all samples using EDS

Full size image
Table 3 Targeted and determined Co/Cu composition ratio of CC2 to CC8 electrodes
Full size table

Electrochemical characterization

To evaluate the electrochemical performance of all samples, cyclic voltammetry (CV), galvanostatic charge/discharge (GCD), and electrochemical impedance spectroscopy (EIS) measurements were conducted in 3 M KOH aqueous solution employing a three-electrode system. CV experiments have been run to investigate both the occurrence of reversible faradaic reaction at the electrode and an acceptable potential range with high coulombic efficiency. Figure 3a shows CV curves of all electrodes at 5 mV/s at which redox peaks contributed from the faradaic reactions are observed. The non-rectangular behavior of all CV curves confirms that Faradaic reactions occur at the surface of the electrodes and also the charge-storage mechanism is principally because of these reactions. Within the cathodic scan, redox peaks are clearly observed in copper–cobalt mixed oxide (CC) and CoO electrodes at the potentials of approximately 0.36 and 0.28 V vs. Ag/AgCl which indicates the characteristic of a battery-like electrode for all of them. These redox peaks are attributed to the couples Co3+/Co2+ and Cu2+/Cu+ reactions taking place in KOH electrolyte. [35]. The expected reaction equations are as follows (5)–(7) [32, 36]:

Fig. 3
figure 3

a comparison of the CV of all electrodes at 5 mV/s, b CV of CC4 electrode at different scan rates, c CV capacity of electrodes at different scan rates, d coulombic efficiency of electrodes at different scan rates, and e plots of the cathodic and anodic peaks of current densities obtained from (a, b) versus (scan rate)1/2

Full size image
$${\mathrm{CuCo}}_{2}{\mathrm{O}}_{4} +{\mathrm{H}}_{2}\mathrm{O }+{\mathrm{e}}^{-} \leftrightarrow 2{\mathrm{CoOOH}}^{-} +\mathrm{Cu}(\mathrm{OH}{)}^{+}$$
(5)
$${\mathrm{CoOOH}}^{-} +\mathrm{ O}{\mathrm{H}}^{-}\leftrightarrow {\mathrm{CoO}}_{2}+ {\mathrm{H}}_{2}\mathrm{O}+2{\mathrm{e}}^{-}$$
(6)
$${\mathrm{CuOH}}^{+} + {\mathrm{e}}^{-}\leftrightarrow \mathrm{ Cu}\left(\mathrm{OH}\right)$$
(7)

The cyclic voltammogram curves (CV) for the CC4 electrode at different scan rates starting from 5 to 200 mV/s are shown in Fig. 3b. A slight shift of the anodic and cathodic peaks towards higher and lower potential is observed with increasing scan rate. This was due to the slow kinetics of electrolyte ion transport within the active layer at high scan rates [37]. In addition, as the scan rate is increased from 5 to 200 mV s−1, the shapes of these CV curves do not show significant change implying the wonderful reversibility of the Faradaic redox reaction. The enclosed area of the CV curves is indicative of the electrode capacity, which is broadener for the CC4 electrode, confirming the maximum capacity provided by CC4 compared to other electrodes. Based on the integral of the CV cathodic current over time, the specific capacity of the electrodes was calculated, using Eq. (8) [32]:

$${Q}_{s}=\frac{1}{(3.6 \, m)}{\int }_{t_{i}}^{{t}_{f}}I \, dt$$
(8)

where m is the mass of the materials.

Since these electrodes exhibited redox peaks in the CV curve and the capacitance of the electrodes varies over the whole potential window, the interpretation of these electrodes must be different from that of pseudocapacitive, and the term of specific capacity Qs (mAh/g) must be used to evaluate their electrochemical performance [38]. The calculated specific capacity of the electrodes as a function of scan rate has been shown in Fig. 3c. The specific capacity decreases with increasing scan rate due to the redox reaction on the electrode surface [39]. According to Fig. 3c, the electrodes containing more copper than cobalt showed less capacity, and the capacity of the CC4 electrode had the highest value at each scan rate. The maximum specific capacity of the CC4 electrode is 28.3 mAh/g at 5 mV s−1. The calculated specific capacity values for CuO, CoO, CC8, CC6, and CC2 were found to be 8.3, 12.7. 15.4, 23.3, and 18.7 mAh/g, respectively at 5 mV/s.

Figure 3d reports the values of coulombic efficiency of the electrodes at different scan rates. The highest coulombic efficiencies, of 96%, for CC4, were achieved in the potential range of − 0.1 to 0.45 V vs Ag/AgCl. By increasing the scan rate, the coulombic efficiency of all electrodes increases, but the CC4 electrode features higher efficiency at all scan rates and underwent a uniform trend at higher scan rates.

The change in anodic and cathodic peak currents with respect to square roots of scan rates of the CC4 electrode are plotted and shown in Fig. 3e and it showed linear in nature which is expounded to the mass diffusion control process [40, 41].

This conclusion is further confirmed by the subsequent galvanostatic charge–discharge tests. Figure 4a reports the galvanostatic charge/discharge curves of all electrodes at a density of 1 A/g with a potential range of − 0.1 to 0.45 V. GCD is a reliable method for evaluating the electrochemical capacity of materials under controlled conditions. Nonlinear discharge curves of the electrode potential profile confirmed the Faradaic behavior of the battery-like electrode. In order to calculate the specific capacity of the electrodes from the galvanostatic discharge curves, Eq. (9) was used [32]:

$${Q}_{s}=\frac{I \times \Delta t}{m}$$
(9)

where I is the current (A), m is the mass loading of the oxide materials on the electrode (g) and Δt is the discharge time (s).

Fig. 4
figure 4

a GCPL profiles of electrodes at 1 A/g, b capacity retention of CC4 electrode, and c Nyquist plot of the CC4 electrode

Full size image

According to the GCD curves, there is a linear change of potential with the charge in the potential range of 0.4–0.2 V vs Ag/AgCl. Therefore, only within this range the electrodes can be interpreted as a pseudocapacitor. Using the slope of the discharge curve, the specific capacitance of all electrodes is estimated to be 315, 219, 170, 157, 113, and 56 F/g for CC4, CC2, CoO, CC6, CC8, and CuO, respectively. These results are fairly consistent with the results observed in the CV curves.

Since these results clearly indicated that the electrochemical performance of the CC4 electrode was much better than other electrodes, the performance of the long-term cycling stability and ion transport properties of the CC4 electrode was evaluated and is shown in Fig. 4b, c. The cycling measurements were carried out at a current density of 1 A g−1 for 5000 cycles in 3 M KOH, at which the CC4 electrode shows good cycling stability with a capacity retention of ∼79.2% in 3 M KOH.

The electronic and ionic conductivities and diffusive behavior of the electrodes were investigated using EIS analysis. Figure 4c compares the Nyquist plots of the CC4 and CC2 electrodes. A semicircle in the high-frequency region and a straight line in the low-frequency region show that the Nyquist plots of these electrodes are closer to an ideal supercapacitor. The figure inset is the magnification of the plots at the highest frequencies. The high-frequency intercept of the semicircle on the real axis indicates the uncompensated resistance including the electrode electronic resistance and the electrolyte ionic resistance, and that depends on the electrochemical cell geometry. The straight line in the low-frequency region is related to the Warburg impedance (ZW) and the semicircle is associated with the diffusion of ions into the electrode and the charge transfer resistance (Rct), respectively [42]. The CC4 electrode has lower high-frequency resistance (1.1 ohms) than the CC2 electrode (1.4 ohms). According to these data and considering that these two measurements were performed using the same electrolyte and cell configuration, it can be argued that the lower resistance of CC4 is related to its better conductivity achieved by more spinel phase and the resulting defects in the coating structure. In addition, the slope of the Warburg line of the CC4 sample was slightly higher than that of the CC2, which can be attributed to the difference between the two samples in porosity, reactivity, and the number of defects due to the spinel phase, which decreased the resistance to electron/ion diffusion and increased the redox reactions at the electrode.

Conclusion

In summary, supercapacitor electrodes of the Cobalt–Copper mixed oxides (CC) with different ratios of Co/Cu were synthesized on nickel foam by electrochemical deposition. X-ray diffraction showed the presence of Cobalt–Copper oxide spinel structure and the electrode synthesized from 0.06 CoSO4·7H2O + 0.04 M CuSO4.5H2O solution (CC4) featured a higher content of the spinel phase of CuCo2O4. SEM images of samples approved the simultaneous presence of Co and Cu and morphological and structural changes due to differences in the composition of cobalt and copper in the coatings. Electrochemical examinations indicated that the CC4 electrode had a higher specific capacity and better electrochemical performance compared to other electrodes. The initial capacity of the CC4 electrode was 28.3 mAh/g at a scan rate of 5 mV/s with a coulombic efficiency of 94%. In addition, it showed capacity retention of 79.2% at a constant current density of 1 A/g after 5000 cycles.