Self-Supported Sheets-on-Wire CuO@Ni(OH) 2 /Zn(OH) 2 Nanoarrays for High-Performance Flexible Quasi-Solid-State Supercapacitor

: Transition metal hydroxides have attracted a lot of attention as the electrode materials for supercapacitors owing to their relatively high theoretical capacity, low cost, and facile preparation methods. However, their low intrinsic conductivity deteriorates their high-rate performance and cycling stability. Here, self-supported sheets-on-wire CuO@Ni(OH) 2 /Zn(OH) 2 (CuO@NiZn) composite nanowire arrays were successfully grown on copper foam. The CuO nanowire backbone provided enhanced structural stability and a highly efﬁcient electron-conducting pathway from the active hydroxide nanosheets to the current collector. The resulting CuO@NiZn as the battery-type electrode for supercapacitor application delivered a high capacity of 306.2 mAh g − 1 at a current density of 0.8 A g − 1 and a very stable capacity of 195.1 mAh g − 1 at 4 A g − 1 for 10,000 charge–discharge cycles. Furthermore, a quasi-solid-state hybrid supercapacitor (qss HSC) was assembled with active carbon, exhibiting 125.3 mAh g − 1 at 0.8 A g − 1 and a capacity of 41.6 mAh g − 1 at 4 A g − 1 for 5000 charge–discharge cycles. Furthermore, the qss HSC was able to deliver a high energy density of about 116.0 Wh kg − 1 . Even at the highest power density of 7.8 kW kg − 1 , an energy density of 20.5 Wh kg − 1 could still be obtained. Finally, 14 red light-emitting diodes were lit up by a single qss HSC at different bending states, showing good potential for ﬂexible energy storage applications.


Introduction
Environmental problems urge us to develop highly efficient and environment-friendly energy technologies to replace conventional fossil fuels [1][2][3][4][5][6]. Supercapacitors, as some of the most promising electrochemical energy storage devices, have been extensively studied due to their high power density and long cycle life [7][8][9][10][11][12]. Based on the different mechanisms of charge storage, the first category could be electrochemical double-layer capacitors (EDLCs). Carbon-based materials with high porosity, large surface area, and relatively high electronic conductivity have usually been considered suitable choices for EDLCs, as they store charges at the surface of the electrodes via electrostatic interaction [13][14][15]. The second type of supercapacitors stores charges based on faradaic reactions [16][17][18], and battery-type materials such as Ni(OH) 2 are usually used as the electrode materials [11,19]. Because of these electron-transfer processes at the surface or the near-surface region of the electrodes, they usually present a higher charge storage capacity than EDLCs.
For advanced battery-type supercapacitors, the electrode materials should possess high conductivity, abundant active sites, and a relatively low cost. Various materials have been designed and synthesized with enhanced electrochemical performance, such as transition metal-based compounds [20][21][22][23][24][25][26]. Among them, transition metal hydroxides have attracted a lot of attention on account of their battery-type feature, high theoretical capacity, facile preparation methods, and low cost. However, their intrinsically low conductivity deteriorates the cycling stability and high-rate performance [21,[27][28][29].
In this work, we reported a self-supported sheets-on-wire composite nanoarray for high-performance battery-type supercapacitor. CuO nanowire array was first fabricated via an electrochemical anodization followed by thermal annealing, which served as support for the hydrothermal growth of Ni(OH) 2 /Zn(OH) 2 nanosheets, forming a core-shell CuO@Ni(OH) 2 /Zn(OH) 2 (denoted as CuO@NiZn). The CuO nanowire backbone provided enhanced structural stability and a conducting pathway for highly efficient electron from the hydroxide nanosheets to the current collector. As a result, the as-obtained CuO@NiZn showed a high capacity of 306.2 mAh g −1 at a current density of 0.8 A g −1 , which is significantly higher than that of CuO (235.8 mAh g −1 ) and Cu@NiZn (132.4 mAh g −1 ). After a long charge-discharge test at 4 A g −1 for 10,000 cycles, the CuO@NiZn exhibited a retention of 97.3%. Moreover, when a quasi-solid-state hybrid supercapacitor (denoted as qss HSC) was assembled with CuO@NiZn as the positive electrode and activated carbon (AC) as the negative electrode, the device demonstrated a capacity of 125.3 mAh g −1 at 0.8 A g −1 and a 93.6% retention after 5000 cycles at 4 A g −1 . The highest energy density of this qss HSC could reach up to 100.20 Wh kg −1 . Even at the highest power density of 16.26 kW kg −1 , an energy density of 43 Wh kg −1 could still be obtained. Finally, 14 red light-emitting diodes (LEDs) were lit up by a single qss HSC at different bending states, showing good potential for flexible energy storage applications. Figure 1 shows the morphology of different samples prepared after respective treatments. The inset photographs clearly show the distinct transformation of the Cu foam, from light blue to brown and finally to black, indicating the corresponding phase transformations and the formation of new material. It is apparent that the Cu(OH) 2 nanowire arrays (denoted as NAs) completely covered the entire surface of the Cu foam (Figure 1a), which showed a uniform 1D nanostructure morphology (Figure 1b). Under a higher magnification (Figure 1c), it is clear that Cu(OH) 2 possessed a rodlike structure with a width of about 200 nm. After the thermal conversion, the corresponding CuO maintained the original NA structure (Figure 1d,e) with no apparent collapse. However, a distinct difference could also be observed ( Figure 1f) that CuO showed much thinner wirelike subunits with sharp tips. Such a change in the structure could probably be attributed to the decomposition of Cu(OH) 2 and the loss of H 2 O during the heat treatment. After the hydrothermal growth of Ni/Zn hydroxides, the entire surface of the nanowires was covered with additional materials, giving rise to a much thicker 1D structure (Figure 1g,h). The CuO backbones appeared to be quite robust, which could withstand the hydrothermal condition without loss in structural integrity. The CuO nanowires were covered by small nanosheets with a size of about only 100 nm (Figure 1i), leading to the formation of a sheet-on-wire CuO@NiZn core-shell structure. It should be mentioned that directly using the original Cu(OH) 2 NAs for the growth of the hydroxides led to nanosheets growing on the Cu foam ( Figure S1) with no presence of the parental 1D NAs. This could be due to the strong destructive effect of the hydrothermal environment, which damaged the Cu(OH) 2 NAs. Conversion was thus necessary from the hydroxide to the oxide in order to achieve a high structural stability. Directly growing the metal hydroxides onto the Cu foam substrate would lead to the formation of large nanosheets ( Figure S1), which is in accordance with previous reports [30]. Furthermore, the surface structures of the samples were characterized by the Brunner−Emmet−Teller method ( Figure S2). It is clear that all the samples showed a type III isotherm, while only CuO@NiZn possessed a type H3 hysteresis loop, suggesting the presence of a porous structure. On the other hand, the other two samples did not exhibit a noticeable hysteresis. This gave rise to a higher surface area of CuO@NiZn of 12.08 m 2 g −1 and a larger total pore volume of 0.05 cm 3 g −1 than those of CuO (3.34 m 2 g −1 , 0.02 cm 3 g −1 ) and Cu@NiZn (1.33 m 2 g −1 , 0.04 cm 3 g −1 ). The crystallographic characterization was collected by X-ray diffraction (XRD; Figure 3). It can be observed that in addition to the three very strong peaks due to the Cu foam, there were two minor peaks at about 35 • and 39 • in CuO. Such weak intensities in these two peaks could be due to the very low heat treatment temperature during the hydroxide-tooxide conversion, thus giving rise to a relatively poor crystallinity in the resulting CuO. After the hydrothermal treatment, the peaks due to Ni(OH) 2 (JCPDS No. 03-0177) and Zn(OH) 2 (JCPDS No. 48-1066) could be clearly identified with relatively low intensities [31], and this is quite understandable considering the low hydrothermal temperature employed during synthesis, which may not be able to promote the formation of a highly crystallized structure. At the same time, the peaks due to CuO increased in intensity with one more peak at 46 • [32], suggesting that the hydrothermal process helped in the formation of better crystalline CuO. Additionally, directly growing these hydroxides onto the Cu foam gave rise to diffraction peaks with poor intensities (Figure S3), and according to a previous report [33], the current synthesis has been shown to generate phase-pure Ni-based hydroxide with low crystallinity.  The valence states of different elements of CuO@NiZn were further analyzed by X-ray photoelectron spectroscopy (XPS). The full survey spectrum ( Figure S4) confirmed the existence of Cu, O, Ni, and Zn elements. The high-resolution XPS spectrum of Cu 2p (Figure 4a) can be deconvoluted into Cu 2p 1/2 and Cu 2p 3/2 with three satellite peaks located at 962.3, 943.5, and 940.6 eV. These satellite peaks are usually caused by the "shake up" process when excess electrons are excited to higher energy states [34][35][36]. The Cu 2p 1/2 signal can be fitted into two peaks at 954.4 and 952.2 eV, which can be ascribed to the Cu 2+ and Cu 0 , respectively. The Cu 2p 3/2 signal can also be fitted into two peaks at 934.5 and 932.3 eV, attributing to the Cu 2+ and Cu 0 , respectively. These results are consistent with previous reports [28,29] and can be assigned to CuO (for Cu 2+ ) and the Cu foam substrate (for Cu 0 ). In the O 1s spectrum (Figure 4b), the peak shows three contributions: the peak at 531.9 eV is typical for adsorbed water (H-O-H), the peak at 531.1 eV for a hydroxyl group (M-OH), and that at 530.3 eV for a metallic oxide (M-O) [37]. Two distinct peaks appear in the Zn 2p spectrum (Figure 4c) at 1044.7 and 1021.6 eV, indicating the existence of Zn 2+ [38,39]. The high-resolution spectrum of Ni 2p (Figure 4d) shows three satellite peaks located at 880.1, 865.6, and 862.1 eV [40], following the two main peaks of Ni 2p 1/2 and Ni 2p 3/2 at 856.0 and 873.6 eV, respectively, which can be attributed to the state of Ni 2+ [28,29,41]. The two peaks at 875.3 and 857.8 eV belong to Ni 3+ , which might be caused by the partial surface oxidation and has been widely observed in transition metal-based compounds [42,43]. The supercapacitor performance of the as-obtained CuO@NiZn, CuO, and Cu@NiZn were all tested in an alkaline electrolyte, where the potential of the working electrode was relative to that of the reference electrode (saturated calomel electrode, SCE), with a Pt counter electrode. Figure 5a shows the cyclic voltammograms (CVs) of CuO@NiZn at different scan rates from 2 to 50 mV s −1 , with oxidation peak at 0.36 V vs. SCE (anodic sweep) and reduction peak at 0.18 V vs. SCE (cathodic sweep) at the scan rate of 2 mV s −1 . Such a pair of redox peaks correspond to the following reactions: [41,44,45]:

Results
The contribution of Zn(OH) 2 was suggested to accelerate the transport of ions, endowing more efficient faradaic processes [44,46,47]. These redox peaks clearly exhibit the battery-type properties of CuO@NiZn. The capacities (Figure 5b) of 483.1, 400.4, 344.2, 200.5, and 113.8 mAh g −1 at the various scan rates can be calculated, which are much higher than the capacity of CuO and Cu@NiZn at the same scan rates ( Figure S5), exhibiting the advantages of the current sheet-on-wire composite NAs.
The galvanostatic discharge curves (Figure 5c) of CuO@NiZn were also presented in Figure 5c, showing a typical battery-type behavior with much longer discharge time and plateau than CuO and Cu@NiZn ( Figure S4). Thus, the capacity can be calculated to be 306.2, 295.4, 283.8, 281.9, and 281.7 mAh g −1 at current densities of 0.8, 2, 4, 10, and 20 A g −1 , demonstrating higher capacity than CuO and Cu@NiZn as well (Figure 5d). The long-term cycling performance was tested at 4 A g −1 for 10,000 charge-discharge cycles (Figure 5e). Notably, the CuO@NiZn electrode showed a gradual increase in the capacity during the initial few thousand cycles, which could be attributed to a possible activation process [35]. After 10,000 cycles, a remarkable retention of 97.3% could be maintained, demonstrating its excellent reversibility and structural stability. Such a performance is also considered superior when compared with other reported data (Table S1). In contrast, the capacities of the other two samples declined very rapidly to below 30 mAh g −1 for the first 5000 cycles. Electrochemical impedance spectroscopy (EIS) was also carried out with a frequency range of 100 kHz to 0.01 Hz to investigate the change in the interfacial resistance before and after cycling ( Figure S6). Interestingly, the semicircle part of the curve in the high-frequency region even decreased in diameter after 10,000 cycles, suggesting a reduced charge transfer resistance. Furthermore, after the long-term performance, the CuO@NiZn electrode still maintained the sheets-on-wire NAs ( Figure S7), while the CuO and Cu@NiZn demonstrated severe collapse or rearrangement in the structure, further confirming the superior stability of the as-prepared core-shell NAs. Subsequently, a quasi-solid-state hybrid supercapacitor (qss HSC) was assembled with CuO@NiZn as the positive electrode and active carbon (AC) as the negative electrode to further verify the practical application of the as-prepared electrode. The suitable operating voltage window for the qss HSC was chosen to be 0-1.6 V due to the windows of activated carbon and CuO@NiZn ( Figure S8). The CV curves at different scan rates were tested (Figure 6a) with the corresponding specific capacities displayed in Figure 6b. The capacities could be calculated to be 60.5, 59.5, 58.9, 56.0, and 49.5 mAh g −1 from 2 to 50 mV s −1 . The discharge curves (Figure 6c) implied that the corresponding capacities of 125.3, 116.4, 97.0, 70.6, and 53.7 mAh g −1 could be obtained from 0.8 to 20 A g −1 (Figure 6d). Figure 6e exhibits the stability at a current rate of 4 A g −1 , and the capacity remained to be 43.4 mAh g −1 after 5000 cycles. The insets show two photographs with one qss HSC under both flat and bent states powering up 14 red LEDs arranged in the pattern of "SME" (the abbreviation of School of Materials and Energy), indicating good potential for flexible energy storage applications. Furthermore, the maximal energy density of this qss HSC could reach up to 116 Wh kg −1 . Even at the highest power density of 7.8 kW kg −1 , an energy density of 20.5 Wh kg −1 could still be obtained.

Conclusions
In this work, self-supported sheets-on-wire CuO@NiZn nanoarrays were successfully grown on copper foam and applied as battery-type electrode material for a supercapacitor, which delivered a high capacity of 306.2 mAh g −1 at a current density of 0.8 A g −1 , and a superior rate performance of 281.7 mAh g −1 at a high current density of 20 A g −1 and a remarkable cycling stability of 97.3% retention after 10,000 cycles. This superior performance could be attributed to its unique structure: the CuO nanowire backbone provided enhanced structural stability and a conducting pathway for a highly efficient electron transport from the hydroxide nanosheets to the current collector. Furthermore, when a quasi-solid-state hybrid supercapacitor was assembled with CuO@NiZn as the positive electrode and activated carbon as the negative electrode, the device demonstrated a capacity of 125.3 mAh g −1 at 0.8 A g −1 , and a 93.6% retention after 5000 cycles. The maximal energy density of this qss HSC could reach up to 116.0 Wh kg −1 . Even at the highest power density of 7.8 kW kg −1 , an energy density of 20.5 Wh kg −1 could still be obtained. Finally, 14 red LEDs were lit up by a single qss HSC at different bending states, showing good potential for flexible energy storage applications.

Supplementary Materials:
The following are available online at https://www.mdpi.com/article/10 .3390/pr9040680/s1, Materials and Methods: Figure S1: SEM images of Cu@NiZn and Cu(OH) 2 @NiZn, Figure S2: N 2 adsorption-desorption isotherm and pore size distribution, Figure S3: XRD pattern of Cu@NiZn, Figure S4: XPS survey spectra of CuO@NiZn, Figure S5: Electrochemical performance in a three-electrode system of CuO and Cu@NiZn, Figure S6: Nyquist plots of CuO@NiZn before and after cycles, Figure S7: SEM images of three electrodes after test, Figure S8: CV curves of AC and CuO@NiZn in a three-electrode system, Table S1: Comparison of the supercapacitor performances.