Introduction

Electrochemical capacitors (EC’s) (or supercapacitors or ultracapacitors or pseudocapacitors) are the capacitor, having high power density, high energy density, and long cycle life as compared to electrostatic capacitors [1]. In EC’s, charges stored in between electrode and electrolyte interface, hence electrochemical mechanism of energy storage gives new system of charge storage and carries the more charges as compared to electrostatic capacitor. The EC’s are classified in two types depending on charge storage mechanisms such as (1) electric double-layer capacitors in which charge separation occurs at electrode–electrolyte interface [2, 3]—E.g., electrodes of carbon nanotubes (CNT’s), activated carbon, and graphene [4], and (2) pseudocapacitor, where charge separation takes place at the surface of electrode—E.g., electrodes of conducting polymers and transition metal oxides [5–9]. For the improvement of capacitance of supercapacitor, ruthenium oxide is the most studied material [10, 11] formed by sol–gel and electrodeposition reporting large SC 720 and 788 F/g, respectively, with excellent reversibility [11, 12]. Because of highly poisonous nature and high cost of ruthenium, alternative materials to ruthenium oxide were tried. The other electrode materials like Fe2O3 [13], Co3O4 [14], NiO [15, 16], MoO2 [17], MnO2 [18, 19], and CuO [20] shows good response to electrochemical performance. Out of these electrode materials, copper oxide is found to be cost effective, easy to prepare, and non-toxic. Copper oxide shows different nanosized dimensions such as, nanospheres, nanoflowers, nanorods, and nanotubes. In rechargeable Li+ ion batteries, the capacity for Li+ storage depends upon the shape and structure of the nanosized copper oxide [21]. The low cost of CuO electrode futurizes better potential application for supercapacitor and also alternative for high-cost materials. Anodization of metals is a simple and well-established technique to form highly self-ordered metal oxide nanostructures [22]. Anodization was widely used in making different nanostructures of several metal hydroxides/oxides [23, 24]. Anodized Cu(OH)2 thin films of nanoneedle-type morphology with 115 F/gm SC in 1 M NaOH have been reported by Jadhav et al. [25]. Zhang et al. and Wen et al. reported the synthesis of Cu(OH)2 nanoribbons and nanotubes in a single-crystal form by using a process of surface oxidation of copper foil in alkaline solution [26, 27]. Patake et al. reported the formation of electrodeposited porous and amorphous copper oxide thin films from alkaline sulfate bath. The prepared electrode shows 36 F/g SC in 1 M Na2SO4 [28]. Shaikh et al. reported the synthesis of CuO–PAA hybrid thin films by a spin coating technique. Coated films were annealed at different annealing temperatures such as 300, 400, and 500 °C. The obtained values of SC increase from 41 to 136 F/g carried in 1 M H2SO4 [29], which clearly indicates post annealing temperature dependent of the electrochemical properties of the electrode materials.

Herein, the attempt was made for the preparation of copper hydroxide thin films by anodization of copper at various anodic potentials. Also, the copper oxide was obtained from post annealing treatment at various temperatures on the deposited copper hydroxide thin films. The formation mechanism of copper hydroxide thin films from anodization of copper is discussed briefly and also a systematic investigation on the post annealing temperature-dependent structural, morphological, and electrochemical properties is successfully done.

Experimental

The anodization setup includes, the two-electrode systems in which copper substrate (1.1 cm2) acting as a working electrode at anode and platinum wire acts as a counter electrode at cathode. Deposition was carried in 10 ml, 1 M NaOH electrolyte. Thin film samples of copper hydroxide were obtained by varying the deposition potential as 0.1, 0.5, 0.9, 1.1, and 1.3 V for the period of 30 min (the deposited samples were denoted as D 1, D 2, D 3, D 4, and D 5, respectively). On the basis of capacitive nature and adherency, sample gets optimized. The optimized copper hydroxide samples (deposited at 0.9 V, for 30 min) get annealed separately in air for 1 h by varying the temperatures as 523, 548, 573, 598, 623, and 673 K to get copper oxide thin films. The annealed samples were further denoted as T 1, T 2, T 3, T 4, T 5, and T 6, respectively. Weight of all deposited samples (mg/cm2) was measured by gravimetric weight difference method which is about ~0.1 mg/cm2. The internal geometry and crystal parameters of all samples were analyzed by using X-ray diffraction patterns carried out by X-ray diffractometer (XRD) (Rigaku D/max2550Vb +18 kw with CuKα, λ = 1.54056 Å) in the range of diffraction angle 2θ from 20° to 80°. Surface morphology was observed using a scanning electron microscope (SEM) JEOLJSM-6360. The electrochemical properties of all prepared samples were carried out using computer-controlled potentiostat (CHI 600AD—electrochemical analyzer/workstation, CH instruments, USA) with standard three-electrode cell having platinum wire as a counter electrode and saturated Ag/AgCl as a reference electrode.

Results and discussion

Initially, there is a formation of Cu(OH)2; after annealing it gets converted into Cu2O and then CuO. The phase transition occurs due to the change in +2 to +1 oxidation state and vice versa [28]. The possible reactions at anode to form copper hydroxide and then formation of copper oxide (I) are given as

$$ {\text{Cu}} \to 2 {\text{e}}^{ - } {\text{ + Cu}}^{{ 2 { + }}} $$
$$ {\text{Cu}}^{{ 2 { + }}} {\text{ + 2OH}}^{ - } \to {\text{Cu}}\left( {\text{OH}} \right)_{ 2} $$
$$ 4 {\text{Cu}}\left( {\text{OH}} \right)_{ 2} \;\underrightarrow \Delta \; 2 {\text{Cu}}_{ 2} {\text{O + 4H}}_{ 2} {\text{O}} \uparrow {\text{ + O}}_{ 2} \uparrow $$
$$ 2 {\text{Cu}}_{ 2} {\text{O + 2O}}_{ 2} \;\underrightarrow \Delta \; 4 {\text{CuO + O}}_{ 2} \uparrow $$

Structural and morphological characterizations

Copper oxide can exist in two crystalline phases i.e., cuprous oxide or cuprite (Cu2O) and cupric oxide or tenorite (CuO) [30]. Both these materials are semiconductors having band gap energy of 1.21–1.51 eV and 2.10–2.60 eV, respectively [31, 32].

Figure 1a shows the XRD patterns of pure Cu (Do) and as-deposited Cu(OH)2 samples at various anodic potentials (D 1, D 2, D 3, D 4, and D 5). Here, it is observed that all samples show crystalline behavior. Sample Do shows orientations along (111), (200), and (220) planes. Value of interplanner spacing ‘d’ for highly oriented (200) plane is 2.12 Å, while for (111) plane is 2.087 Å. and for (220) plane it is 1.278 Å. The observed ‘d’ values properly matches with standard ‘d’ values taken from JCPDS data card no.85-1326 for pure Cu showing cubic crystal structure and JCPDS data card no. 72-0140 for Cu(OH)2 showing orthorhombic crystal structure.

Figure 1
figure 1

a The XRD patterns of as deposited Cu(OH)2 samples as a function of deposition potential b the effect of annealing temperature on Cu(OH)2 samples deposited at 0.9 V and for 30 min

Full size image

The optimized sample D 3 gets annealed in air for 1 h by varying the temperatures from 523 to 673 K by the interval of 25 K. The annealed samples are denoted as T 1, T 2, T 3, T 4, T 5, and T 6. Figure 1b shows XRD patterns of these samples. Here, it is observed that all samples show crystalline behavior. The observed ‘d’ values properly match with standard ‘d’ values taken from JCPDS data card no.44-0706 exhibiting monoclinic crystal structure for CuO and for JCPDS data card no.74-1230, exhibiting cubic crystal structure for Cu2O. At lower annealing temperatures, from 523 to 598 K (samples T 1 to T 4), the oxidation occurs at the top surface and Cu2O grows down to the lower layer so that only Cu2O (111) is able to form [33]. As temperature increases to 673 K (sample T 6), the phase transformation was observed and the mixture of CuO and Cu2O phases is obtained due to the change in +2 oxidation state to +1 oxidation state and vice versa [28].The oxidation process promotes the phase transformation from pure cubic Cu to cubic Cu2O to cubic-monoclinic Cu2O–CuO mixed phase [34]. The value of crystallite size ‘D’ was calculated for predominant (111) plane using Scherer’s formula [35].

$$ D = \frac{0.9\lambda }{\beta \cos \theta } $$
(1)

where 0.9 is the Scherer’s constant, ‘λ’ is the wavelength of X-ray, ‘β’ is the full-width at half maximum, and ‘θ’ is the diffraction angle. With increase in annealing temperature from T 1 to T 6, the average crystallite size D increases from 31.029 to 44.04 nm. The increase in crystallite size can be attributed due to the change in crystallographic phase from Cu2O to CuO [35]. The relative intensity of (111) peak for T 4 is high as compare to other electrodes that may be due to better crystalline nature of the Cu2O.

Figure 2 shows scanning electron micrographs (SEM) of different samples. Figure 2a shows dense surfaces for hydroxide thin film (sample D 3), while copper (I) oxide (Fig. 2b–f) shows uniformly distributed rough granular type of morphology. With increase in the annealing temperature from 523 to 598 K, the increase in granular size was clearly observed which may facilitate the increase in active surface area of the copper oxide and hence may useful in improvement of the electrochemical performance of the copper oxide thin film samples. With further increase in temperature to 623 K, grains merge together forming mud-like compact surfaces. The grain size increases from 130 to 833 nm from sample (b) to (f), respectively. AFM scanning of the optimized T 4 sample was made to study surface topography and roughness of the films. The 2D and 3D images of AFM scanning are shown in Fig. 3a, b, respectively. For measuring the surface roughness of the films, 4.39 × 4.24 μm2 area was utilized. The roughness of the sample is observed to be 125 nm. Weight loss of the deposited material with annealing temperature is demonstrated in Fig. 4 deposited mass (for T 1-0.0028 gm) continuously deceases with increase in temperatures (for T 2-0.0024 gm, T 3-0.0016 gm, T 4-0.0008 gm), the observed loss in mass may be due to desorption of the physisorbed water molecules [29]. For T 5 and T 6, no much loss is observed.

Figure 2
figure 2

a SEM images for as deposited Cu(OH)2 sample and the effect of annealing temperatures b 523 K, c 548 K, d 573 K, e 598 K, f 623 K on Cu(OH)2

Full size image
Figure 3
figure 3

a 2D and b 3D AFM images of Cu(OH)2 sample deposited at 0.9 V, 30 min, and annealed at 598 K

Full size image
Figure 4
figure 4

Deposited mass versus temperature curve for Cu (OH)2 samples carried at 0.9 V, for 30 min at different annealing temperatures

Full size image

Electrochemical characterizations

The electrochemical performance of all copper hydroxide and copper oxide samples was studied by using three-electrode cell containing 1 M aqueous NaOH solution. Cyclic voltammetry (CV) characterization was carried within potential windows −0.6 to +0.2 V for Cu(OH)2 and −0.7 to +0.2 V for Cu2O/CuO electrodes at 2 mV/s. CV curves of Cu(OH)2 and Cu2O/CuO shows large current, small potential window, large area under curve, and also asymmetric nature having larger tail of reduction. It may be due to the electrochemical reduction reaction of Cu(OH)2 and Cu2O/CuO with Na+ ions and the formation of solid electrolyte interface (SEI) films [36].

For the samples D 1 to D 5 (Fig. 5a), the main cathodic peak was observed for all deposition potentials. For the sample D 1, reduction peak was observed at −0.18 V with small oxidation peak, while reduction peak gets shifted to right at −0.13 V for D 2 sample. This shifting of peak occurs due to the presence of some irreversible processes in the electrode material [16]. For sample D 3, the reduction peak was observed at −0.15 V with little large anodic peak than D 1 and D 2 samples, which gives good reversibility of the electrochemical reaction [21]. Hence the value of SC was maximum for D 3 sample i.e., 604.5 F/g at 2 mV/s. For sample D 4 and D 5, reduction peak shifted towards −0.18 V with a small oxidation peak and with small area under the curve. The pseudocapacitance of copper oxide electrode is the result of transition between +2 and +1 oxidation state and vice versa [28, 35]. The capacitance (C) and specific capacitance (SC) values of the samples carried at different deposition potentials are tabulated in Table 1a.

Figure 5
figure 5

a CV curves of as deposited Cu(OH)2 samples and b CV curves of annealed Cu(OH)2 samples at 2 mV/s in 1 M NaOH

Full size image
Table 1 Variation of C and SC for the samples carried at different deposition potentials (a) and temperatures (b)
Full size table

Figure 5b shows the CV curves of sample T 1 to T 6 carried at different temperatures exhibiting asymmetric nature of the curves. The reduction peaks of all the samples T 1 to T 6 were observed at −0.1 V with small oxidation peak. For sample T 4, the area under curve increases with large oxidation peak. Hence the T 4 sample gives maximum value of SC i.e., 6000 F/g at 2 mV/s. Samples from T 1 to T 5 show stable phase of Cu2O, while Cu2O–CuO mixed phase was observed for sample T 6. Which may be the reason for further decrease in the electrochemical performance and hence the SC values of T 6 sample. The values of C and SC for all the samples annealed at different temperatures are tabulated in Table 1b. The values of C and SC associated with electrode were calculated from the following relations.

$$ C = \frac{I}{dv/dt} $$
(2)
$$ {\text{SC}} = C/W $$
(3)

where I is the average current, dv/dt is the voltage scan rate, C is the capacitance, SC is the specific capacitance, W is the weight of the material dipped in electrolyte. Charge–discharge variation of the annealed Cu2O electrodes was observed at different current densities in 1 M NaOH. Figure 6a shows charge–discharge variation of optimized T 4 electrode scanned within potential window −0.6 to 0.2 V for variable current densities 1–6 mA/cm2. The non-symmetrical discharging behavior was observed at all current densities, which clearly indicates the pseudocapacitive behavior of the Cu2O thin film samples. In discharging curve, a sudden voltage drop (linear part parallel to Y-axis) represents a voltage change due to change in internal resistance and a capacitive component related to the voltage change [25]. The electrical parameters such as specific energy (SE), specific power (SP), and columbic efficiency (η) were calculated from discharging curves using the relations [37, 38].

Figure 6
figure 6

a Charge–discharge study and b stability curve of Cu(OH)2 sample deposited at 0.9 V, 30 min, and annealed at 598 K

Full size image
$$ {\text{SE}} = \frac{V \times Id \times td}{W} $$
(4)
$$ {\text{SP}} = \frac{V \times Id}{W} $$
(5)
$$ \eta = \frac{td}{tc } \times 100 $$
(6)

where V is the voltage, Id and td are the discharging current and time, respectively, W is the weight of the material dipped in electrolyte, and tc is the charging time. The values of SE, SP, and η are summarized in Table 2.

Table 2 Variations of SE, SP, and η with applied current densities of T 4 electrode
Full size table

Electrochemical stability (Fig. 6b) of optimized T 4 electrode was measured in 1 M NaOH at 100 mV/s for 1000 CV cycles. The observed capacity retention of 26 % just within 200 cycles indicates poor stability of the Cu2O thin films. This may be due to slight phase change from Cu(OH)2 to CuO through intermediate stage of Cu2O which might cause collapse of its intrinsic lattice structure and show diminution of its capacitance [39].

Electrochemical impedance measurements were made to search internal resistance of the electrode. Figure 7a shows the Nyquist plot between real and imaginary impedance values in the frequency range of 1 mHz–1 MHz, obtained at −0.249 V open circuit potential. At high frequency region, the crossover point of the highest frequency with the real part of the impedance is in general a net resistance of the electrolyte, intrinsic resistance of substrate, and contact resistance between the active material and the current collector. The observed internal resistance is ~1.75 Ω/cm2. In the intermediate region of the frequency, the straight line nature with the inclination of ~45° to the real axis was noticed, which in fact the characteristic of ion diffusion into the electrode materials [2]. In the low frequency range, the straight line part was leaned slowly toward the imaginary axis, indicating that the electrode material experiences capacitive behavior [40]. Figure 7b (Nyquist plot of Z” vs Z’) shows the experimental and standard curves obtained by simulation using ZsimpWin software, and the inset in this figure shows matched equivalent circuit. Evaluated circuitry parameters are as follows: R 1 = 1.851 Ω, R 2 = 1.883 Ω, R 3 = 22.03 Ω, and R 4 = 0.05909 Ω are the charge transfer resistances, and C 1 = 4.994 × 10−5 F is the capacitance with Q = 0.009036 and Warburg impedance (W) = 0.03471.

Figure 7
figure 7

a Nyquist plot and b matched equivalent circuit of Cu(OH)2 sample deposited at 0.9 V, 30 min, and annealed at 598 K

Full size image

Conclusion

Copper hydroxide thin films were successfully obtained by using the anodization of the copper substrate at different deposition potentials. The as-deposited copper hydroxide samples Cu(OH)2 reveal orthorhombic crystal structure, whereas the cubic Cu2O phase changes to cubic-monoclinic Cu2O–CuO mixed phase due to the post annealing treatment at higher temperature. SEM images depict rough and uniformly distributed granular type of morphology; increase in annealing temperature affected granular size. The Cu2O electrode (T4) exhibits maximum SC of 6000 F/g at 2 mV/s, with SE 5.46 Wh/kg, SP 7.714 KW/kg and η 97.15 %. The stability study shows 26 % capacitive retention. Observed internal resistance from the Nyquist plot is about 1.75 Ω/cm2. The overall study indicates that the excellent capacitive performance of the Cu2O electrode will be suitable for energy and power applications in supercapacitor technology by overcoming the stability problem observed during the study.