Advance Aqueous Asymmetric Supercapacitor Based on Large 2D NiCo2O4 Nanostructures and the rGO@Fe3O4 Composite.

NiCo2O4 nanostructure is a widely studied pseudocapacitor material because of its high specific capacitance value. Most of the time, the thickness of the nanostructure inhibits the electrode material from whole-body participation and causes sluggish charge transportation. These phenomena directly interfere with the electrochemical performance of the electrode, such as specific capacitance value, stability, energy density, and so forth. Here, two different thin two-dimensional morphologies (nanosheet and nanoplate) of the NiCo2O4 nanocomposite with a large lateral size are reported using ammonia as a hydrolyzing agent. The large size and flat surface of the as-synthesized materials offer enormous active sites during the electrochemical reaction, and the thin wall makes the ion penetration and transportation very effective and facile. Therefore, the NiCo2O4 nanosheet and nanoplate structures exhibited high specific capacitance values of 1540 and 1333 F/g, respectively, with excellent rate and good cycling stability. Here also, two different advance aqueous asymmetric supercapacitors have been reported utilizing two NiCo2O4 nanostructure materials as positive electrodes and the rGO@Fe3O4 composite as a negative electrode, which exhibited excellent rate and high specific energy without sacrificing the specific power. We also studied the electrochemical activity of the rGO@Fe3O4 composite at different compositions.

Introduction

Our existing energy sources are fixed and are gradually depleting day by day. This demands new research for energy harvesting or storage. In both cases, the quality of electrode materials becomes the most important concern, which actually administers the excellence of the device. The electrochemical capacitor, which is also known as the supercapacitor, is a promising energy storage device for future applications because of its several advantages, such as high power density, high cyclic durability, fast charge–discharge (ch–dch) process, and so forth.[1] Till now, it has not been commercialized properly because of some of its limitations, mainly low energy density.[1] In the early stages of research, carbon was the only material that was used as the electrochemical capacitor. Different carbonaceous materials such as graphene, graphene oxide (GO), carbon nanotube (both single-walled and multiwalled), activated carbon, and so forth have been used so far. In these cases, electrostatic charge separation on the electrode−electrolyte interface is the reason behind the storage of energy. Those materials have limitations too, specifically low specific capacitance value and low energy density. Energy density (E = 0.5CV2; C = specific capacitance value and V = voltage window) of an energy storage device can be increased by two ways, either by increasing the specific capacitance value or by increasing the voltage window or both.[2−4] Hence, scientists started working on transition metal oxides (TMOs) or transition metal hydroxides (TMHs) or mixed transition metal oxides (MTMOs) as an electrode material for energy storage, which is termed as the pseudocapacitor.[5−11] In most cases, energy has been stored, exploiting the surface redox reactions of the electrode materials. Because of the fast faradic redox reaction on the surface, theoretically though these compounds exhibit a high specific capacitance value, in practice, they do not exhibit such a high value and fall flat. The thickness of the electrode material limits the electrolyte ions to access all active sites of the electrode material during the electrochemical reaction.[2−11] Thus, morphology is an important factor for TMOs/TMHs/MTMOs to behave as an advance pseudocapacitor.[2−15] This problem can be solved by using a material of two-dimensional (2D) morphology.[2−11] Aqueous electrolyte ions can penetrate up to 20 nm of an electrode material.[2−4] Thus, if an ultrathin 2D material can be used as an electrode material for an energy storage device, it can participate through its whole body during the electrochemical reaction. Simultaneously, fast ion transportation through the electrode material can also increase the rate capability as well as cyclic durability of the energy storage device.[16,17] Another problem with most of the pseudocapacitors is that they exhibit a small potential window during the electrochemical reaction in an aqueous electrolyte.

Therefore, in the last few years, scientists have been working on an asymmetric supercapacitor (ASC) where two materials of different working potentials are judiciously coupled together resulting in a broadened voltage window.[2−11] A large voltage window provides a high energy density.[2−15] Thus, both the positive electrode (mainly pseudocapacitors) and negative electrode (generally carbonaceous materials, but sometimes V2O5, MoO3, and Fe-oxides) in combination play a crucial role in the energy storage activity of the ASC.[18−20] Because of the high ionic mobility, low cost, and safety issues, scientists are more interested in the aqueous electrolyte.[2−20] Binary metal cobaltites (MCo2O4, M = Ni, Mn, Zn etc.) are widely used as popular pseudocapacitors.[21−25] Among them, NiCo2O4 is a well-documented one as both Ni and Co exhibit more than one oxidation state. In most cases, NiCo2O4 has been chosen as a positive electrode in the ASC because of its interconnected structure, which contemplates high electrical conductivity and high electrochemical activity.[21−30] Several morphologies (nanosheet, nanowire, nanoflake, mesoporous etc.) of NiCo2O4 have been studied so far to fabricate an advance pseudocapacitor electrode.[23−30] But in most of the cases, bare NiCo2O4 does not perform well to exhibit the expected result. As for example, Zhang and Lou reported NiCo2O4 of nanorod and nanosheet morphologies, which exhibited specific capacitance values of 905 and 889 F/g at 2 A/g specific current, respectively.[29] Some researchers have also studied the composite of NiCo2O4 with carbon dots, graphene, and so forth, and some have also prepared some heterostructures of NiCo2O4 with the addition of other materials or NiCo2O4 itself to fabricate a pseudocapacitor of high energy density with excellent rate capability and cyclic durability to meet the demand.[24] There is a report where Liu et al. have synthesized NiCo2O4@NiCo2O4 core–shell nanoflake arrays for high-performance supercapacitors. They reported maximum areal specific capacitance of 2.20 F/cm2 at a discharge current of 5 mA/cm2.[24] Till now, several groups have reported the pseudocapacitance activity of the NiCo2O4 nanosheet.[26−30] Few researchers have got excellent result also.[27] For example, Du and co-workers prepared the NiCo2O4 nanosheet over a flexible carbon fiber, which displayed very high specific capacitance value of 2658 F/g at 2 A/g specific current.[27] But, in most of the cases, nanosheets were of short sizes with an aggregated surface, which limits their electrochemical activity.[26−30] Though several studies have been done on the 2D nanosheets of NiCo2O4, yet synthesis of the large, flat, and smooth surface of the NiCo2O4 nanosheet is not an easy task. On the other hand, use of Fe3O4 as a negative electrode has drawn much attention of the researchers because of its high over potential for the hydrogen evolution reaction.[31−35] Environmental friendliness, low cost, and high abundance in the earth’s crust make Fe3O4 an excellent replacement for activated carbon for a negative electrode.[31−35] However, low electrical conductivity is a major concern for Fe3O4 to be used as a pseudocapacitive electrode.

Here, we present the synthesis of two different 2D NiCo2O4 nanostructures of high lateral size (micron level). Both the 2D morphologies have been synthesized using two different sets of precursor salts, resulting in two important materials. However, in both the cases, ammonia is used as the hydrolyzing agent. Both the materials exhibited excellent electrochemical activity. For the sake of complete assembly, we have also synthesized rGO@Fe3O4 composites of different compositions. During the fabrication of the ASC, based on the range of electrochemical activities, NiCo2O4 nanostructures and the rGO@Fe3O4 composite have been co-jointly considered as the positive electrode and negative electrode, respectively.

Results and Discussion

Positive Electrode

Here we used two different 2D NiCo2O4 nanostructure materials as positive electrodes for the ASC; one is NiCo2O4 nanosheet (NCS) and another is NiCo2O4 nanoplate (NCP). For the synthesis of NCS, we used cobalt sulfate (CoSO4·7H2O) and nickel acetate [Ni(OOCCH3)2·4H2O] as precursor salts, and for NCP, nickel nitrate [Ni(NO3)2·6H2O] and cobalt nitrate [Co(NO3)2·6H2O] were used (Figure ). In both the cases, ammonia was used as the hydrolyzing agent. In our previous study, we have discussed how ammonia supports the evolution of the 2D morphology of TMOs and TMHs.[2,3] Here also, the different hydrogen-bonding ability of ammonia with different metal ions and the anisotropic growth-oriented attachment under vigorous reaction conditions guided the formation of 2D mixed TMOs.[2,3]

Figure 1

Pictorial presentation of the synthesis of NCS and NCP samples.

Different physical methods have been used for the characterization of the as-synthesized samples. Figure a exhibits the powdered X-ray diffraction (XRD) spectra of the two samples. In both the cases, the XRD patterns are in good agreement with the JCPDS file no. 73-1702, which is of cubic lattice with the Fd3m space group of the nickel cobaltite (NiCo2O4) spinel structure.[26] Compositional analysis of the energy-dispersive X-ray (EDX) spectra indicates the presence of Ni, Co, and O in both the as-synthesized samples (Figure S1a,b). Quantitative analysis of the EDX for both the samples reveals that the stoichiometric ratio of Ni/Co/O is 1:2:4, which is also in agreement with the XRD analysis of the samples. X-ray photoelectron spectroscopy analysis points out the surface elemental composition and oxidation state of the as-prepared two samples. Figure b exhibits the X-ray photoelectron survey spectra of NCS and NCP, which also confirms the presence of Ni, Co, and O in both the samples (C 1s has come from the reference).

Figure 2

Comparative (a) XRD and (b) wide-range X-ray photoelectron spectroscopy results of the NCS and NCP samples.

Deconvoluted XPS spectra (Figure a,b) reveal that in both the cases, Ni 2p consists of four peaks; 2p3/2, 2p1/2, and other two are satellite peaks. For both the samples, Ni 2p3/2 and Ni 2p1/2 are situated at ∼855.6 and ∼872.8 eV, respectively. Deconvoluted Co 2p (Figure c,d) also exhibits four peaks. For Co, 2p3/2 and 2p1/2 are situated at ∼780.2 and ∼795.8 eV, respectively. These experimental results are well in agreement with the literature report of the binding energies for these peaks, which indicate the formation of the cubic NiCo2O4 spinel structure.[36] Deconvolution of the O 1s (Figure e,f) spectra for both the samples consist of three peaks at ∼529.8, ∼531.4, and 533.7 eV, which signify the typical metal–oxygen bond, oxygen ions at low coordination on the surface, and hydroxyl group of the surface-adsorbed water molecules, respectively, for both the as-synthesized samples.[38] Transmission electron microscopy (TEM) images of the as-synthesized samples indicate that both the materials are of thin 2D morphologies with lateral size in the micron level. Figure a depicts the 2D thin nanosheet of the NiCo2O4 sample whose lateral size is greater than 1 μm (∼1.9 μm), and Figure b depicts the uniform 2D thin hexagonal nanoplate morphology of the as-synthesized NiCo2O4, which is of ∼970 nm lateral size. Figure S2 displays the field emission scanning electron microscopy (FESEM) images of the two as-prepared samples, which also depict the 2D nanosheet and hexagonal nanoplate morphology, respectively. Selected area electron diffraction (SAED) patterns of the NCS and NCP samples specify the crystallinity of the samples. The calculated lattice spacing from the SAED images is in good agreement with the (311), (422), (511), and (220) planes for the NCS sample (Figure c) and the (311), (422), and (220) planes for the NCP sample (Figure d). These results are well in agreement with the XRD of the samples. Brunauer–Emmett–Teller (BET) study reveals that both the materials are mesoporous in nature, which is an important parameter for an electrode to behave as an advance pseudocapacitor. Figure S3a,c demonstrates the N2 adsorption–desorption isotherm curves for the NCS and NCP samples, respectively, at a relative pressure from 0.1 to 1 (P/P0). The calculated BET surface areas for the NCS and NCP samples are 99 and 115 m2/g, respectively. Figure S3b,d displays the Barrett–Joyner–Halenda pore size distribution curves, where we can see that the pore diameter for NCS is 3.8 nm and that for NCP is 3.4 nm. These results clearly indicate the mesoporous nature of the samples. The pore volumes of NCS and NCP are 0.282 and 0.339 cm3/g, respectively.

Figure 3

High-magnification X-ray photoelectron spectra (XPS) of (a,b) Ni 2p, (c,d) Co 2p, and (e,f) O 1s of NCS and NCP samples, respectively.

Figure 4

(a,b) TEM images and (c,d) SAED patterns of NCS and NCP samples, respectively.

We have tested the half-cell electrochemical activities of the samples (NCS and NCP) in a three-electrode system using a 3 M aqueous solution of KOH as the electrolyte. Figure exhibits a series of cyclic voltammetric (CV) curves and ch–dch curves of the NCS and NCP samples at different scan rates and different specific currents, respectively, between −0.1 and 0.4 V relative to the saturated calomel electrode (SCE). In both the cases, the CV curves are peak-shaped, which indicate the pseudocapacitive nature of the electrodes (Figure a,b). The almost symmetric ch–dch curves point to the excellent kinetic reversibility of the samples during the electrochemical reactions [Co(II) ↔ Co(III) ↔ Co(IV)]. We calculated the specific capacitance values of the samples from the ch–dch curves (Figure c,d). At 1 A/g specific current, the specific capacitance values of NCS and NCP are 1540 and 1333 F/g, respectively. Here, it is observed that the specific capacitance value of NCS is greater than that of NCP. This is because the thickness of NCP is ∼75 nm and that of NCS is <20 nm (Figure S2). Thus, NCS offers more redox active sites than NCP during the electrochemical reaction. We calculated the specific capacitance values at different specific currents and different scan rates to explore the rate capability of the as-prepared material. Figure S4a,b displays the plot of specific capacitance values of NCS and NCP samples as a function of specific currents and scan rates, which indicates the superior rate capability of the samples. Compared to the reported specific capacitance values of different NiCo2O4 nanosheets, it can be concluded that the as-synthesized NCS and NCP exhibit excellent electrochemical activities (Table S1). We performed the ch–dch experiments for both samples up to 10 000 cycles at 30 A/g specific current. Figure e,f exhibits the graphical presentation of the experimental results for the cyclic performances. Here, we can see that after 10 000 cycles, the NCS and NCP samples kept their initial specific capacitance up to 89 and 81%, respectively, at such a high specific current. Figure e,f also reveals that after 10 000 cycles, the Coulombic efficiency of the NCS and NCP samples are 98 and 93%, respectively, which indicates the superb kinetic reversibility of the electrode material. The excellent electrochemical activity of the electrode materials could be explained, considering their morphology and structural features. First, both the materials bear a high lateral size with thin and flat surfaces, which offer maximum active sites during the redox reactions that increase the specific capacitance value largely. Additionally, the thin wall of both the materials facilitates the first charge transportation and excels ion diffusion, which increase the rate capability of the material. Furthermore, the mesoporous nature and high pore volume increase the wettability of the electrode materials, which also facilitates the fast ion transportation through the material during electrochemical reactions. On the other hand, NiCo2O4 with its mixed oxide structure increases the electrical conductivity and physical robustness. This physical robustness supports material prevention and inhibits physical degradation during the ch–dch cycles at high specific currents, which have clearly been observed from our experimental results. To support these discussions, we performed electrochemical impedance spectroscopy. Figure S4c,d exhibits the electrochemical impedance spectra () of the NCS and NCP samples before and after 10 000 ch–dch cycles. From Table S2, it is observed that the values of RS (internal resistance), RCT (charge transfer resistance), and W (diffusive resistance) for NCS are lower than those for NCP. This also justifies the high specific capacitance value of NCS than that of NCP. Again, before and after 10 000 cycles, both the samples maintain excellent stability. There is no such significant change in RS and RCT values after stability performance for the NCS sample. The only significant change is for W. For NCP, the differences are not very high, but larger compared to NCS. This also supports the high stability of NCS over NCP.

Figure 5

(a,b) CV curves, (c,d) ch–dch curves, and (e,f) stability curves of the NCP and NCS samples.

Negative Electrode

To fabricate an advance ASC, a suitable negative electrode is also an important factor. Here we prepared the rGO@Fe3O4 composite (GFe), which at a suitable mass-loading of Fe3O4 behaves as an excellent pseudocapacitor electrode. We synthesized Fe3O4 nanocubes using an alternative redox pathway, which has been developed by our group.[38] We prepared several composites (GFe1, GFe2, GFe3, and GFe4) by varying the Mohr’s salt concentration, keeping the amount of GO fixed. We performed the electrochemical analysis using all these four composites, and it was found the GFe2 exhibits the best pseudocapacitance activity (discussed later). Hence, we performed all experiments using GFe2 as the standard composite and mentioned as GFe in the whole manuscript. Figure a illustrates the comparative XRD spectra of GO, Fe3O4 cube, and the rGO@Fe3O4 (GFe) composite. A broad peak at 10.3° in the spectrum in blue is the characteristic peak for GO. In black, most of the distinct peaks arise from Fe3O4 [JCPDS no. 86-1362, magnetite, face-centered cubic lattice, cell parameter a = 8.396 Å, space group Fd3m (227)], which are in good agreement with the spectrum in red (which stands for the pure Fe3O4 nanocube) and the literature report.[38] A broad hump is observed at around 23°, which is due to the formation of reduced GO (rGO). An interesting fact is that no peak at 10.3° is observed, which confirms the transformation of GO to rGO. Thus, from the above discussion, it is clear that the as-prepared composite is an exclusive rGO@Fe3O4 composite.

Figure 6

(a) Comparative XRD curves of GO, GFe, and Fe3O4, (b) FESEM, and (c) TEM images of GFe samples.

To complete the compositional investigation of the composite, we performed the X-ray photoelectron spectroscopy analysis. Figure S5a depicts the wide-range XPS of the composite, where all characteristic peaks for Fe, O, and C were observed. Figure S5b exhibits the high-resolution XPS of Fe. The peak positions at 711.9 eV (Fe 2p3/2) and 725.6 eV (Fe 2p1/2) confirm the formation of Fe3O4 in the system, which is in good agreement with the literature value.[38] The deconvoluted C 1s spectrum is composed of four peaks, which are situated at 284.6, 286.6, 287.7, and 288.57 eV (FigureS5c). The peak at 284.6 eV stands for the sp2 C.[4] The other peaks define that even after thermal reduction, the rGO state contains some oxygen-containing groups (286.6 eV for the −C–O bond, 287.7 eV for the −C=O bond, and 288.57 eV for the −C(O)O bond). Figure S5d demonstrates the high-resolution XPS of O 1s positioned at 531 eV, which is due to the Fe–O bond. The above discussion confirms the formation of the rGO@Fe3O4 composite (GFe).

We performed Raman analysis to confirm the transformation of GO to rGO. Figure S6a,b exhibits the Raman spectra of pure GO and the GFe composite. In both the spectra, two major peaks were observed at ∼1592 and ∼1351 cm–1, which are referred as the G band and D band, respectively. In the case of GO (Figure S6a), ID/IG is 0.94, and for the rGO@Fe3O4 composite, it is 1.14 (Figure S6b), which confirms the transformation of GO to rGO during the synthesis of the Fe3O4-based composite. Here, the D band stands for the disorder and defect in the atomic arrangement, and the G band stands for the plane vibrations of the sp2 hybridized carbon atom of the 2D layer.[4] This clearly confirms the transformation of GO to rGO during the hydrothermal reaction.

To identify the morphology of the as-synthesized composite, we performed the FESEM and high-resolution TEM analyses. Figure b,c exhibits the FESEM and TEM images of the GFe sample, where we can see small Fe3O4 nanocube decoration over the rGO surface. Figure displays the FESEM and TEM images of all different rGO@Fe3O4 composites of various compositions (GFe1, GFe2, GFe3, and GFe4). From Figure , it is clear that the Fe3O4 nanocubes decorate the flat 2D rGO sheet, and as the concentration of the Mohr’s salt is increased, the distribution of nanocubes over the 2D rGO increases. For GFe1, most of the rGO surface lies vacant (Figure a,e). Again, for GFe3 and GFe4, we observe that Fe3O4 is randomly distributed over the rGO surface, which causes aggregation of the nanocubes (Figure c,d,g,h). But, for GFe2, the Fe3O4 nanocubes are uniformly distributed all over the flat rGO surface (Figure b,f). To study the morphology of the pure Fe3O4 nanocubes, we use the same procedure without using GO. Figure S7a exhibits the FESEM images of pure Fe3O4 nanocubes. Figure S7b exhibits the FESEM image of pure rGO. Inductively coupled plasma-mass spectrometry was used to quantify the amount of Fe3O4 in the composites, and it was found that Fe3O4 loading in GFe1, GFe2, GFe3, and GFe4 are 42, 61, 79, and 81%, respectively.

Figure 7

(a–d) FESEM and (e–h) TEM images of rGO@Fe3O4 nanoparticles with increasing concentration of Mohr’s salt.

Electrochemical activities were carried out for all composites in the three-electrode system using the KOH electrolyte. Figure a shows the series of CV curves for the GFe composite in the potential range of −1.2 to 0 V, where a pair of redox peaks in the region of −0.7 to −1.2 V is observed because of the reversible faradaic redox transformation between Fe2+ & Fe3+.[4,27−30] We calculated the specific capacitance value of the composite from the ch–dch curves at different specific currents (Figure b). The calculated specific capacitance values of the GFe composite are 890 and 252 F/g at 2 and 50 A/g specific currents, respectively. We also calculated the specific capacitance value of the GFe composite at different scan rates (Figure S8a). Figures c and S8a demonstrate the specific capacitance values of the GFe composite at different specific currents and scan rates. These results clearly reveal that the GFe composite possesses excellent rate capability. We also scrutinized the pseudocapacitance activity of the other three samples, and Figure c displays the plot of specific capacitance values of the composites as a function of the specific current. From this figure, it can be concluded that 61% Fe3O4 loading is optimum for the composite to behave as a stable, high rate capable pseudocapacitor. Comparison to the ch–dch curves of pure Fe3O4 nanocube and rGO (Figure S8b) also indicates that because of the synergistic effect between Fe3O4 and rGO in the composite, GFe exhibits better electrochemical activity. We performed 10 000 ch–dch cycles at 40 A/g specific current, and it was observed that the composite maintained 90% of its original specific capacitance values after such a large number of cycles and also maintained ∼98% Coulombic efficiency (Figure S8c). Electrochemical impedance spectroscopy analysis also supports the high electrochemical performance of the GFe composite. Figure S8d exhibits the EIS of the GFe composite before and after 10 000 ch–dch cycles, which support the stability of the composite after the cycles. From the above discussion, it is clear that rGO plays an important role in the fabrication of a stable electrode, and the proportion of composition in the composite has a major contribution during the electrochemical activity.[4]

Figure 8

(a) CV curves and (b) ch–dch curves of the GFe sample, respectively. (c) Plot of specific capacitance values of the samples as a function of specific current.

To evaluate the practical utility of the above-mentioned materials, we fabricated two aqueous ASCs using NCS and NCP as positive electrodes separately and GFe as a negative electrode in 3 M KOH electrolyte. Figure S9a demonstrates the comparative CV curves of NCS, NCP, and GFe at 50 mV/s, which suggests that the maximum voltage window for the ASCs can be 1.6 V. Figure a,c represents the CV curves of NCS//GFe and NCP//GFe cells at different scan rates, respectively. Figure b,d corresponds to the ch–dch curves of NCS//GFe and NCP//GFe cells at different specific currents, respectively. The shapes of the CV curves are well-maintained even at very high scan rates, which suggest very fast and reversible reaction kinetics. We calculated the specific capacitance values of the asymmetric cells using ch–dch curves and CV curves. Figure e,f demonstrates the plot of specific capacitance values as a function of specific currents and scan rates for both the two-electrode systems. From Figure e,f, it is clear that both the two-electrode cell is highly rate capable. Here, it is observed that NCS//GFe and NCP//GFe exhibit specific capacitance values of 505 and 406 F/g at a specific current of 1 A/g, respectively. Again at 12 A/g specific current, the values are 316 and 252 F/g, respectively. These results suggest that after a 12-fold increase in the current, both the asymmetric cells exhibit a rate capability of ∼62%, which is an extraordinary result for a two-electrode system. To confirm the reversibility, we calculated the Coulombic efficiency (η) at different currents for both the cells, and it was found that even at a low specific current, Coulombic efficiency is ∼100% for both the cells. Figure S9b shows that NCS//GFe maintained its stability up to 86% after 10 000 ch–dch cycles, whereas the stability of NCP//GFe (Figure S9c) is 84%. These results stand for the excellent cyclic durability of the asymmetric cells. We also calculated the Coulombic efficiency in each cycle, and Figure S9b,c shows that both cells maintained it up to 10 000 cycles. Figure a displays the Ragone plot for the NCS//GFe and NCP//GFe cells, where both the cells exhibit a high specific energy with a high specific power. The NCS//GFe cell shows a maximum specific energy of 44.89 W h/kg at a specific power of 800 W/kg (discharge time 202 s, specific current 1 A/g) and a maximum specific power of 15 362 W/kg at a specific energy of 24.41 W h/kg (discharge time 5.72 s, specific current 12 A/g). For the NCP//GFe cell, the maximum specific energy is 42.5 W h/kg at a specific power of 1279 W/kg (discharge time 120 s, specific current 1 A/g) and the maximum specific power is 12 789 W/kg at a specific energy of 22.43 W h/kg (discharge time 6.31 s, specific current 12 A/g). These results clearly reveal that both the asymmetric cells can store very high energy without sacrificing the power, which is utmost important for an advance energy storage device. Table S3 also reveals that compared to the reported Ni–Co and Fe-based asymmetric supercapacitor, these two asymmetric cells exhibit a better energy storage and delivery capability. To support the superiority of the asymmetric cells, we performed electrochemical impedance spectroscopy analysis for both the two-electrode cells at the initial and after 10 000 ch–dch cycles (Figure b,c). From the plot, it is clear that in both cases, there is not enough change in the RS and RCT values (Table S4) after the cyclic experiments for both the cells. The only changes are in the diffusive resistance, which was increased after 10 000 cycles. This is the reason for the capacity fade after the cyclic treatment. Figure d exhibits the Bode plot for the as-fabricated asymmetric cell, where we found that for NCS//GFe, the phase angle at 0.01 Hz is −73 and for NCP//GFe, it is −70, suggesting the capacitive property of the systems. Another important parameter for an energy storage device is the relaxation time (τ0, τ0 = 1/f0, f0 = frequency at −45°), which stands for the low charging time for a supercapacitor to attain its maximum specific capacitance value. For NCS//GFe and NCP//GFe, the values of τ0 are ∼1 and ∼1.26 s, respectively, which also support the high specific energy and specific power of the asymmetric sample.

Figure 9

(a,c) CV and (b,d) ch–dch curves of NCS//GFe and NCP//GFe asymmetric cells, respectively. (e,f) Plot of specific capacitance values as a function of specific currents and scan rates.

Figure 10

(a) Ragone plot of NCS//GFe and NCP//GFe, (b,c) EIS curves of NCS//GFe and NCP//GFe at the initial and after 10 000 ch–dch cycles, respectively, and (d) comparative phase angle plot of NCS//GFe and NCP//GFe samples.

Conclusions

In summary, two as-fabricated two-electrode electrochemical systems are reported, and they are proved to be excellent ASCs because of their high specific energy, specific power, and extraordinary rate. High electrochemical performances of the as-fabricated large 2D NiCo2O4 nanostructures ensure to be the ideal candidate for the positive electrode in an asymmetric cell. The large surface of the 2D NiCo2O4 provides innumerable active sites, and the thin wall shortens the diffusion path, which augments the electrochemical activities of both the as-synthesized samples. The large potential window and stability of the rGO@Fe3O4 composite (at an appropriate composition) offers an alternative negative electrode for the two-electrode system. rGO present in the composite enhances the electrical conductivity of the composite, which also overcomes the stability issue of the Fe3O4 electrode as a pseudocapacitor. The above discussion reveals that the above-mentioned asymmetric cells could be an ideal combination for hybrid energy storage devices for future application.

Experimental Section

Materials and Instruments

The related information is briefly discussed in the Supporting Information.

Synthesis of 2D NiCo2O4 Nanostructures

NCS

First, a mixture of 5 mL of 0.05 M Ni(OAc)2·4H2O and 10 mL of 0.05 M CoSO4·7H2O aqueous solution was prepared. The mixture was then taken in a 100 mL beaker, and 5 mL of 2% (w/v) aqueous solution of polyvinylpyrrolidone (PVP) was added slowly. The resulting mixture was stirred for 4 h. Then, the mixture was taken in a 15 mL screw cap test tube (10 mL in each test tube). An aliquot of 400 μL liquor ammonia solution was injected into the solution and shaken to make the reaction mixture homogeneous. After that, the test tube containing the reaction mixture was capped and subjected to modified hydrothermal treatment (MHT) at 180 °C. After 24 h of heating, the obtained product was centrifuged and washed with distilled water and then with ethanol to remove all impurities. Finally, the product was dried under vacuum and stored.

NCP

First, a mixture of 5 mL of 0.05 M Ni(NO3)2·6H2O and 10 mL of 0.05 M Co(NO3)2·6H2O aqueous solution was prepared. The mixture was then taken in a 100 mL beaker, and 5 mL of 2% (w/v) aqueous solution of PVP was added slowly. The resulting mixture was stirred for 4 h. Then, the mixture was taken in a 15 mL screw cap test tube (10 mL in each test tube). An aliquot of 400 μL liquor ammonia solution was injected into the solution and shaken to make the reaction mixture homogeneous. After that, the test tube containing the reaction mixture was subjected to MHT at 180 °C. After 24 h of heating, the obtained product was centrifuged and washed with distilled water and then with ethanol to remove all impurities. Finally, the product was dried under vacuum and stored.

Synthesis of the rGO@Fe3O4 Composite

We synthesized GO from graphite using the Hummers method. Briefly, pristine graphite was taken as the precursor and was oxidized by strong oxidizing agents, KMnO4, NaNO3, and concentrated H2SO4. After that, H2O2 was added to the mixture to remove the excess KMnO4 and to convert the generated MnO2 to MnSO4. Then, it was washed with hot water and air-dried. After that, 50 mg of the solid material was dispersed with 50 mL of distilled water through sonication for 3 h, and then it was centrifuged for 30 min at 3000 rpm speed for washing. Finally, the solid solution was taken and used for the preparation of different rGO@Fe3O4 composites.

Afterward, 20 mL of Mohr’s solution of different concentrations (0.025, 0.05, 0.075, and 0.125 M) were dissolved into the above suspensions and stirred for 24 h. Then, hot distilled water was added to the reaction mixture, and excess NaBH4 was added to the solution. The beaker with the reaction mixture was placed on a piece of door magnet so that the reaction evolved the iron/GO composite, which was adhered to the bottom of the beaker and could not come in contact with the air–water interface. The freshly prepared composites were washed with distilled water. Then 50 mL of 2% PVP solution was added to the beaker and subjected to MHT at 180 °C for 24 h. Here, the product appeared as a black precipitate, which was washed very carefully first with distilled water and then with ethanol, so that all PVP was removed from the surface of the composite.

Fabrication of Electrode for the Electrochemical Supercapacitor

The electrochemical measurements for both the samples, NCS and NCP, were carried out using the three-electrode system at room temperature. Aqueous 3 M KOH solution was used as the electrolyte. Electrochemical studies were carried out by cyclic voltammetry, chronopotentiometry or galvanostatic charge–discharge, and electrochemical impedance spectroscopic technique. We used Ni foam as the current collector. Ni foam was first treated with 6 M HCl, followed by washing with ethanol and water. Then it was dried. For electrode fabrication, first we dispersed the as-synthesized samples, NCS and NCP (separately), acetylene black, and polyvinylidene fluoride (PVDF) in the N-methyl-2-pyrrolidone solvent in the weight ratio of 85:10:5. Then, the as-obtained slurry was pasted on 1 × 1 cm2 activated Ni foam using a spatula. After that, the electrodes were dried at 120 °C for 12 h. These electrodes were used as the working electrode. Pt wire and SCE were used as the counter and reference electrode, respectively. Before starting the experiments, the electrodes were dipped in 3 M KOH electrolyte for 15 min. All electrochemical measurements were carried out using the CHI 660E electrochemical workstation. The weight of the working electrode was 1.5 mg (excluding the weight of acetylene black and the PVDF binder).

For the GFe composite, we used the same procedure as mentioned above. We used GFe instead of NCS or NCP. The weight of the working electrode was also 1.5 mg (excluding the weight of acetylene black and the PVDF binder).

Asymmetric Cell

For the fabrication of the ASC, NCS and NCP were used as the positive electrode separately, and GFe was used as the negative electrode. Pt wire was used as the connector between the electrode and the instrument. A dielectric Whatman filter paper was used as the separator. KOH (3 M) was used as the electrolyte. Here, because of the different specific capacitance values of the positive electrodes, mass balance is necessary to get the maximum activity of the asymmetric cell. The mass balance is calculated using the equationwhere Q+ and Q– signify the charge on the positive and negative electrodes, respectively. Againwhere CSC, m, and ΔE signify the specific capacitance value, mass of the active material in the electrode, and the working potential of the ch–dch curve, respectively. Now, comparing eqs and 2, we getHere, for NCS//GFe, the mass ratio of the positive and negative electrode is (NCS/GFe) 1.4:1. For NCP//GFe, the ratio is (NCP/GFe) 2:1. The weight of the active material in the complete cell is 3 mg (excluding the weight of acetylene black and the PVDF binder).