Structural, optical, morphological, and elemental analysis of mixed ternary CNC
The CNC samples, synthesised by one pot hydrothermal method, are obtained as green coloured powder. The TG-DTG curves in the temperature range of 30–800°C is given in supplementary (Fig S1). The first weight loss (~ 2.5%) within 200°C is primarily attributed to evaporation of moisture. The major weight loss (~ 18%) occurred in between 200–500°C is due to removal of hydroxides/structural water and practically no further weight loss was observed up to 800°C. The structural, morphological and optical properties of the synthesized material were studied by several characterization techniques. The powder XRD, recorded to identify the phases and crystallinity of the sample, is displayed in Fig. 2(a). The diffraction peaks, observed at 2θ values 11.54, 2,2.89, 47.38, and 59.87°, are ascribed to (003), (006), (018), and (110) Ni(OH)2 crystal planes, respectively as per matching with JCPDS card no. 00-038-0715. The sharp peaks at 13.19, 26.04, and 36.71° with crystal planes (001), (002), and (121) can be attributed to Cu(OH)2 (JCPDS No.00-042-0746). The presence of CoO phase is confirmed by distinct peaks at 34.16 (111), 39.86 (200), and 56.76 (220) as referred to JCPDS No. 00-042-1300. Thus, the sharp well defined diffraction pattern of the sample confirmed that there is a mixed phase of CoO, Ni(OH)2, and Cu(OH)2. Using the high intensity peaks for three phases, the average crystallite size using Scherrer’s equation is 17.99 nm. The FTIR absorption spectra as given in Fig. 2(b), showed sharp peaks for different metal oxide/hydroxide bonds in 4000 − 500 cm-1 range. The peaks at 1294.2 cm-1 and 827.9 cm-1 are indexed to Cu-OH, 662.3 cm-1 to Co-O, and 461.6 cm-1 to Ni-OH stretching vibrations (Devamani et al. 2013; Saleh et al. 2021; Salavati-Niasari et al. 2013). The peaks at 1502.2 cm-1 and 1602 cm-1 indicated the carbonate ions and bending mode of H2O respectively. Similarly, 2348.4 cm-1 peak for C\(\equiv \text{N}\) stretching vibrations mode in OCN- anion. The broad peak in the range of 3100–3330 cm-1 was aroused due to O-H stretching vibration mode (Salavati-Niasari et al. 2013). The Raman spectrum, presented in Fig. 2(c), revealed the structural properties of CNC. The peak at 293.4 cm-1 is assigned to Cu-OH stretching vibration, 515 cm-1 refers to F2g CoO stretching mode, and peaks at 794 cm-1 and 1055.8 cm-1 denote the Raman active vibrational modes of Ni(OH)2. The O-H bending modes are observed at 1509.4 cm-1 and 1716.7 cm-1 (Bulakha et al. 2018; Vidhya et al. 2020; Hall et al. 2015). Thus, FTIR and Raman spectra validated the findings of XRD which confirmed the mixed-phase formation.
Further, the optical property of the synthesized sample was investigated by UV-Visible DRS studies. Figure 2(d) displays the absorption spectrum of CNC sample with an inset of Tauc’s plot. The optical band gap energy (Eg) can be calculated by using Tauc’s relation (Khan et al. 2020).
α = K (hν – Eg)n (5)
where α is the absorption coefficient, K is constant, and hν is the photon energy. From the inset graph, it is found that the CNC has a direct transition with a band gap energy of 2.36 eV that assisted in improving the photocatalytic efficiency by harnessing the visible light range.
The morphological and elemental analyses are crucial for catalyst samples to understand their activity for different applications. The SEM images of ternary CNC composites in Fig. 3(a, b) are noticeable as small sheets that are beautifully stacked against each other giving the shape of a ball. The edges of the balls resemble the sea urchin like structure throughout the surface. The 3D-image shows that low-temperature hydrothermal synthesis with urea fuel acting as a reducing agent has helped in the smooth formation of CNC samples. The nano spikes at the surface of CNC are arranged gradually to form the structure of the ball. The nanospikes are free to surface which would furnish high distribution of active sites for the reaction. The sea urchin structured small balls of size around 1 µm are connected together to form bigger diameter balls and spread over all, which would also provide high surface area. Wei et al. reported the formation of an In2O3/ZnO composite using a hydrothermal approach showing an urchin-like hollow microsphere morphology. The photocatalytic performance and stability of the composite were enhanced in contrast to pure In2O3 and ZnO (Wei et al. 2017). Zhang et al. synthesized sea urchin ZnCo2O4 microspheres deposited on nickel foam and showed superior performance as electrode materials for supercapacitor applications (Wu et al. 2015). Such morphology is expected to increase the activity of the samples when exposed to either photocatalytic and/or electrocatalytic reactions (Ma et al. 2022). The EDX carried out for the CNC sample as given in Fig. 3 (c), shows 36.64, 50.92, 5.13, 5.56, and 1.75 wt.% for C, O, Co, Ni, and Cu, respectively. The atomic wt.% matches well with the ratio of the precursor of metals taken for the synthesis of the CNC (Co:Ni:Cu::3:3:1) sample.
Additionally, the phase and morphological insights were explored by HRTEM images. Figure 4(a and b) displays the TEM images of the CNC sample which is composed of several spikes stacked at each other to form ball type structure. It also shows thin sheets stacked with each other. The edges of the ball are clearly visible in Fig. 4(b) where the image goes well with SEM image. Further, the SAED pattern captured in Fig. 4(c) consists of bright dots arranged in rings which evidenced the crystallinity nature of the CNC sample. The appearance of well-defined lattice fringes also supports the mixed phase of the CNC sample as depicted in Fig. 4(d-f). In Fig. 4(d), 0.68 nm fringes ascertained to (001) crystal plane of Cu(OH)2 phase. The Fig. 4(e) and 4(f) with 0.81 nm and 0.27 nm fringes match to (003) and (111) crystal planes of Ni(OH)2 and CoO, respectively. The crystal planes of CoO and Ni/Cu(OH)2 phases, as in the XRD pattern, are also validated from the lattice fringes of HRTEM. Further, the mapping images, presented in Fig. 4(g-k), illustrate a clear vision of the elemental distributions of CNC. Thus, HRTEM corroborates the XRD, Raman, FTIR, and SEM analysis of the CNC sample. Such morphology and crystallinity with high active sites and surface area can provide a platform for improved catalytic activity, more wettability, and easy interaction of samples in a reaction process.
As the ternary mixed phases of the CNC sample are confirmed by XRD and HRTEM, so we have carried out the XPS survey to understand the elemental and chemical states in hydrothermally prepared products. The XPS survey spectrum (Fig. 5) confirmed the existence of Co, Ni, Cu, C, and O elements in the product. The deconvolution of Co 2p, Ni 2p, and Cu 2p are given in Fig. 5(b), (c), and (d). The Co 2p shows four sharp peaks that appeared for Co2+ and Co3+. These peaks can be further deconvoluted into two characteristic peaks (782.5 and 797.8 eV) related to Co 2p3/2 and Co 2p1/2 with two satellite peaks at 786.7 and 803.6 eV (Khan et al. 2020). Similarly, the Ni 2p consists of characteristic peaks at 856.8 eV (Ni 2p3/2) and 874.2 eV (Ni 2p1/2) along with other two satellite peaks at 862.5 and 880.2 eV. The Cu 2p3/2 and Cu 2p1/2 peaks, centered at 935.3 and 955.7 eV, are ascribed to the Cu2+ state forming Cu(OH)2 (Cho et al. 2022). The C 1s deconvoluted into three peaks at 284.3, 285.4, and 289.4 eV, seen in Fig. 5(e) reflect C-C, O-C = O, and C-O-C bonds, respectively. The sources of carbon might be from the base or urea in the CNC product. Figure 5(f) displayed the deconvoluted O 1s spectra with peaks at 531.08 (C-O bond), 532.1 (C-O-H bond), and 533.1 (C-O-H bond) eV allied to metal oxide and hydroxide (viz. Co-O, Ni/Cu-OH) bonds and OH bond of absorbed water (Wu et al. 2019). Thus, the observed results from characterisations analysis confirmed the presence of mixed phases that can be utilized for dual functionality applications.
Photocatalytic activity of CNC for CR degradation
As adsorption of dyes on catalyst surface is believed to be first step in photocatalytic process, the maximum adsorption capacity of the CNC was evaluated under dark condition before the visible light treatment for removal of CR as a dye pollutant. The extent of adsorption with varying contact time was performed to reach the adsorption-desorption equilibrium as given in Fig. 6(a) at different concentrations of CR. As evident the maximum adsorption was attained within 60 min and further saturation was reached in 120 min in all concentrations. To elucidate the adsorption mechanism of CR adsorption, the experimental data were fitted to three linearised forms of isotherm models and found best fit (correlation coefficient, R2 = 0.998) to the Langmuir isotherm among all (Fig. 6(b)) yielding an adsorption capacity of 86.9 mg g− 1 indicating monolayer coverage of CR molecules onto the catalyst surface having a finite number of adsorption sites. The linear fittings of time dependent adsorption data at different CR concentrations to different kinetic models are shown in Fig. 6 (c, d) and the calculated kinetic parameters are listed in Table 1. The values of R2 revealed that the experimental data fitted well with the pseudo first-order kinetic model which is further supported from the closeness of calculated and experimental qe values. The multilinearity curves not passing through the origin was obtained in the plots of qt versus t0.5 (supplementary Fig. S2) indicate adsorption process is controlled by other mechanism along with intra-particle diffusion.
Table 1
The determined constants of pseudo-first-order and pseudo-second-order kinetic model with correlation coefficients (R2) at four different initial concentrations.
| CR (mg L− 1) | qe, exp(mg g− 1) | Pseudo first order | Pseudo second order |
| | k1 (min− 1) | qe (mg g− 1) | R2 | k2 (g mg− 1 min− 1) | qe(mg g− 1) | R2 |
| 20 | 19.6032 | 0.0344 | 21.55 | 0.99 | 0.0021 | 22.52 | 0.912 |
| 40 | 33.4137 | 0.0317 | 34.98 | 0.996 | 0.0012 | 38.31 | 0.910 |
| 60 | 44.5893 | 0.0287 | 45.86 | 0.999 | 0.0008 | 51.28 | 0.900 |
| 100 | 57.9327 | 0.0276 | 56.19 | 0.996 | 0.0009 | 64.1 | 0.937 |
Understanding the adsorption process, the photocatalytic performance of CNC towards CR degradation was studied in visible light range and the results obtained are as shown in Fig. 7(a). A complete degradation of CR at 20 mg L− 1 was achieved within 120 min while it degraded 90, 75, and 57% with CR concentrations of 40, 60, and 100 mg L− 1, respectively after attaining adsorption-desorption equilibrium. This shows a decrease in overall CR removal with increase in initial CR concentrations. The declined photocatalytic performance at higher concentrations of dye molecules could be due to reduction in active sites due to the adsorption of more CR molecules thus hindering the generation of charge carriers and limiting the amount of photons reaching the photocatalyst surface. Figure 7(b) shows the progressive decrease in intensity of absorption over entire wavelength range with the increase of exposure time without any noticeable shift in peak positions. The progressive decrease of CR colour intensity is also visible in Fig. 7(b) (inset).
To get further insight into the mechanism of CR photodegradation, the scavenger trapping experiments were performed and results obtained are shown in Fig. 8(a). As seen in the figure, photodegradation was increased by about 4% in the presence of K2Cr2O7 acting as an e− scavenger while there was a very small decrease (< 5%) of CR degradation with addition of EDTA and IPA as h+ and OH. scavengers, respectively. These observations indicate the minimal role of e−, h+ and OH. radicals in the degradation process. On the other hand, a significant decline (> 40%) in the dye degradation was seen in the presence of p-BQ which implies that the superoxide radicals (O2−.), generated through reduction of adsorbed oxygen molecules on the catalyst surface by photoexcited e−, is the principal active species involved in the photodegradation of CR into non-harmful substances. To further substantiate the role of O2−. as the major species, the positions of the conduction and valence bands of CNC was estimated using Mott–Schottky analysis. The analysis was done in 3 M KOH (pH ~ 14) electrolyte within the potential range of − 1 to 0.6 V vs. Ag/AgCl. All the potentials are shown against the reversible hydrogen electrode (RHE), and the conversion between Ag/AgCl and RHE was calculated by the following equation (Jiang et al. 2018) :
ERHE = EAg/AgCl + 0.059 pH + 0.197 V (6)
The extrapolation of linear portion of the Mott- Schottky plot (Fig. 8(b)) to the x-axis shows the flat-band potential value of − 0.047 eV (vs. Ag/AgCl), which is equivalent to 0.97 eV (vs. RHE) as per Eq. (6). This flat-band potential corresponds to the conduction band potential of the CNC. Further, the potential of the valence band was estimated using equation (Jiang et al. 2018) :
EVB = Eg - ECB (7)
Where EVB is valence band potential, Eg is the band gap energy of the semiconductor material, and ECB is conduction band potential. Using the estimated band gap energy (Eg) value of CNC (2.36 eV) from Tauc’s plot, the position of the valence band is calculated to be 1.39 eV. As such, the lower potential of the valence band is not effective in oxidation of H2O to OH. (EH2O/OH. = 2.32 eV), thereby confirmed relatively less involvement of OH. radical in the overall photodegradation process (Li et al. 2019). However, the high positive potential of the conduction band, relative to the standard reduction potential of O2/O2−. (-0.33 eV), supports the active participation of superoxide radical in the degradation process (Li et al. 2019). On the basis of experimental results and theoretical analysis, a plausible schematic mechanism for the photodegradation CR is illustrated in Fig. 8(c). On absorption of photons by CNC from the irradiated visible light, the photoholes (h+) and photoelectron (e−) are generated respectively at valence and conduction band of the catalyst. The photoholes further react with H2O/OH− to give OH. and participate in oxidative degradation of dye while the photogenerated electron in the conduction band reduces the O2 molecules adsorbed on the catalyst surface to generate O2−. radicals and involved as active species in the CR decolorization process. It is believed that the heterojunction derived photocatalyst significantly reduce the recombination of the photogenerated charge carriers which leads to enhancement in the photocatalytic performance of the ternary composite. Table 2 presents a comparative degradation ability of ternary CNC with other photocatalysts toward CR degradation. It showed the effective degradation of CR at higher concentration with minimal loading of catalyst CNC for the nearly same exposure time as compared to the reported studies.
The stability and recyclability of the photocatalyst are considered to be vital in the photocatalytic process in view of on-filed application and economic perspectives. The recyclability of CNC was examined for successive five cycles at the similar experimental settings and the results are displayed in Fig. S3. It exhibits very minimal decrease in the percentage of CR degradation (< 3%) after five cycles indicating the stability as well as retaining the activity of the photocatalyst. The higher dye degradation activity in comparison to other binary and ternary systems (Table 2) with adequate stability of the present ternary composite signifies its potential for practical applications in remediation of dyes and other pollutants.
Table 2
Summary of several works using mixed oxide/hydroxide toward CR degradation.
| Photocatalyst | Amount of photocatalyst (g L− 1) | [CR], mg L− 1 | Light source | Time (min) | % Degradation | Ref. |
| TiO2 doped CoFe2O4 | 0.8 | 20 | Visible | 120 | 50 | (Magdalane et al. 2021)] |
| NiO-NiFe2O4 | 1 | 50 | Visible | 156 | 98.45 | (Hariani et al. 2022) |
| Graphene-CdO/SnO2 | 5 | 70 | Visible | 120 | 82 | (Sirohi et al. 2019) |
| Co1 − xCuxFe2O4 | 0.2 | 10 | Visible | 90 | 71.23 | (Kirankumar et al. 2017) |
| CoO/Ni(OH)2/Cu(OH)2 | 0.33 | 40 | Visible | 120 | 90 | This work |
Applicability of CNC for electrochemical energy storage applications
The electrochemical behaviour of CNC was evaluated by CV, GCD and EIS techniques in 3 M KOH electrolyte. The redox peaks in CV curves at different scan rates in a potential range of 0 to 0.6 V are shown in Fig. 9(a), attributed to the oxidation and reduction of metals oxide/hydroxide in the KOH electrolyte. The current follows the trend of scan rate, i.e. peak current increases with the increase of scan rate from 10 to 100 mV s− 1. The redox peaks in CV curves visible around 0.3 V and 0.5 V may be due to the synergistic reactions of ternary metals (Co-Ni-Cu) during the charging/discharging process (Gurav et al. 2013; Hu et al. 2019) :
CoO + OH− ↔ CoOOH + e− (8)
CoOOH + OH− ↔ CoO2 + H2O + e− (9)
Ni(OH)2 + OH− ↔ NiOOH + H2O + e− (10)
Cu(OH)2 + OH− ↔ CuOOH + H2O + e− (11)
Further, the charge/discharge process at different current densities for the CNC electrode is shown in Fig. 9(b) as GCD profile. The plateau region found in the GCD profile during the charge-discharge process confirms the battery-type behaviour of the CNC electrode in KOH electrolyte. The low capacity degradation even at high scan rate may be ascribed to the structural and morphological stability of the CNC electrode. The specific capacity of the CNC electrode was estimated using Eq. 2, represented in Fig. 9(c). The specific capacity at 0.5 A g− 1 is as high as 405 C g− 1 whereas 280 C g− 1 at 5 A g− 1. As usual, the discharge time decreased with the rise in the current rate owing to the rapid surface scanning of electrode material during charge-discharge. A cobalt-nickel carbonate hydroxide with sea-urchin morphology grown on a 3D carbon sponge using the hydrothermal method was reported with a maximum specific capacity of 289.3 C g− 1 at 1 A g− 1 (Das et al. 2017). The CNC electrode was subjected to an electrochemical impedance study and the spectra are displayed in Fig. 9(d) as a Nyquist plot. The EIS reflects the conductive nature of the CNC electrode with very low solution resistance (Rs). The reduced resistance may be due to better wettability of the electrode in the electrolyte and friendly electrode-electrolyte interaction. This substantiated the CV and GCD curves stating the high charge-transfer kinetics of CNC electrode during the electrochemical reactions. The assembled nano-spikes and sea-urchin morphology provide high surface area and ample active sites for the charge storage in the electrode materials (Shang et al. 2020).
Thus, the one pot synthesized CNC as a promising candidate for a charge storage device was utilised as the positive electrode against the AC as the negative electrode in an electrochemical bath with aqueous KOH as electrolyte. The fabricated HSC was tested with different potential windows at scan rate of 30 mV s− 1 as given in Fig. 10(a). The Fig. 10(b), CV curves at different scan rates at an optimised potential of 1.6 V. The CV curves are quasi-rectangular during the reaction and gradually increase when the scan rate increases which depicts a stable and compatible charge-discharge process in the device. The potential window was also tested by GCD at 0.5 A g− 1 current density. From the GCD profile, the device was found to work best in the range of 0- 1.6 V (Fig. 10(c)) in an aqueous basic electrotype. Figure 10(d) shows irregular triangular-type GCD for the HSC constituting CNC (+ ve) and AC (-ve) electrodes at different current densities.
The HSC device delivered a maximum capacity of 161.89 C g− 1 at 0.5 A g− 1 as shown in Fig. 11(a). The ED and PD were calculated by using Eqs. 3 and 4 from the GCD of the HSC device reflected by the Ragone plot in Fig. 11(b). The device prototype delivered 58.02 W h kg− 1 ED for 645 W kg− 1 PD at 1 A g− 1 current density whereas 42.7 W h kg− 1 ED for 5160.9 PD at 5 A g− 1 which was comparatively higher than the other reported ternary metal oxide/hydroxide as positive electrode material for HSC device (Xiong et al. 2015; Abebe et al. 2022; Sivakumar et al. 2019; Wu et al. 2017; Sha et al. 2018). The battery-type CNC and capacitive-type AC electrodes resulted in high ED and PD for the designed device (Vinoth et al. 2021). For the practical applications, HSC demands high ED and PD with long-term cycling stability. Even though the metal oxides/hydroxides have high specific capacity but its low stability and rapid capacity degradation confines the device performance (Chandrasekaran et al. 2017).
Fascinatingly, the fabricated device has exhibited improved capacity retention (110%) and a high Coulombic efficiency of around 98% after continuous 5000 cycles of charge/discharge process in Fig. 11(c). Such an outcome is only due to the combined effect of metals in the CNC nanocomposites. The presence of copper avoids the capacitance decay due to phase transformation of cobalt and nickel oxides/hydroxides which results in establishing the strong stability of the electrode during the long-term cycling process (Biswal et al. 2020). The stability process was also investigated by an impedance study of the device (Chinnapaiyan et al. 2023). The before and after 5000 cycles EIS was recorded for the ASC device prototype as given in Fig. 11(d). It was found that the resistance was depleted after the 5000th cycle. Some of the reported research work has been listed in Table 3 that verified an enhanced performance of the device due to CNC electrode.
Table 3
Shows the comparison of electrochemical performance for various electrode materials in different supercapacitors.
| Electrode material | Specific charge capacity (C g− 1) | Electrolyte (KOH in M) | Capacity retention (number of cycles) | Energy density (W h kg− 1) | Power density (W kg− 1) | Reference |
| Ni0.5Cu0.5Co2O4 | 147 at 1 A g− 1 | 3 | 89% (5000) | 53.08 | 3500 | (Samanta et al. 2018) |
| CoO/Co3O4 | 54.04 at 0.2 A g− 1 | 3 | 105% (2000) | 10.52 | 140 | (Pang et al. 2015) |
| Co3O4@Ni(OH)2 | 150 at 0.5 A g− 1 | 3 | 87% (6000) | 112.5 | 1350 | (Hu et al. 2019) |
| Zn-Ni-Co oxide | 170.8 at 1 A g− 1 | 6 | 71.2% (6000) | 35.6 | 187.6 | (Wu et al. 2015) |
| CoO/Ni(OH)2/ Cu(OH)2 | 162 at 0.5 A g− 1 | 3 | 110% (5000) | 58 | 645 | This work |
Thus, an extraordinary hybrid supercapacitor performance is delivered by CNC (+)|KOH|AC(-). The research findings certainly garnered the researchers to explore other ternary combinations of metal oxides/hydroxides developed by a facile eco-friendly method in cost-effective approach. The mass production and dual functionality of the synthesised materials can also encourage industries to scale up production.
