Decorating MnO2 nanosheets on MOF-derived Co3O4 as a battery-type electrode for hybrid supercapacitors

Metal–organic framework-derived materials are now considered potential next-generation electrode materials for supercapacitors. In this present investigation, Co3O4@MnO2 nanosheets are synthesized using ZIF-67, which is used as a sacrificial template through a facile hydrothermal method. The unique vertically grown nanosheets provide an effective pathway for rapidly transporting electrons and ions. As a result, the ZIF-67 derived Co3O4@MnO2-3 electrode material shows a high specific capacitance of 768 C g−1 at 1 A g−1 current density with outstanding cycling stability (86% retention after 5000 cycles) and the porous structure of the material has a good BET surface area of 160.8 m2 g−1. As a hybrid supercapacitor, Co3O4@MnO2-3//activated carbon exhibits a high specific capacitance (82.9 C g−1) and long cycle life (85.5% retention after 5000 cycles). Moreover, a high energy density of 60.17 W h kg−1 and power density of 2674.37 W kg−1 has been achieved. This attractive performance reveals that Co3O4@MnO2 nanosheets could find potential applications as an electrode material for high-performance hybrid supercapacitors.


Introduction
Over the last few years, much emphasis has been placed on developing lightweight, versatile, and environmentally safe solidstate energy storage technologies in consumer electronics, slide displays, and miniature medical implants. [1][2][3][4][5][6][7] The battery and the supercapacitor are the most effective sources of energy. However, the batteries' bulkiness, slow charge-discharge rate, and short life period restrict their use in wearable and portable devices. [8][9][10] Supercapacitors have recently gained much attention, particularly in the automotive industry, because of their key characteristics like high power density, lightweight, fast charging-discharging rates, reliable handling, and long lifetime. Based on their charge storage mechanism, supercapacitors are grouped into two categories: electrical double-layer capacitors (EDLCs), which generally use carbon-active materials, and pseudocapacitors, which use redox-active materials. 11,12 Due to their high energy density with rapidly reversible surface redox processes, pseudocapacitors have considerable potential as supercapacitor options in the future. On the other side, several pseudocapacitive materials support one of two inadequate cycling stability and poor conductivity. [13][14][15][16][17] Due to their characteristics of considerable specic capacitance and highrate capacitance, transition metal oxides (RuO 2 , Co 3 O 4 , SnO 2 , and MnO 2 , among others) have prompted widespread attention in the eld of pseudocapacitive electrode materials with substantial specic capacitance. [18][19][20] MnO 2 is one of the most used materials for supercapacitors. It has a wide selection of high attributes such as low cost, environmental friendliness, abundant reserves, and a high theoretical potential capacity. 21,22 However, the weak electrical conductivity of MnO 2 and the practical, specic capacitance of the product is signicantly lower than its theoretical result (1370 F g −1 ). 23 The distinctive structural properties can be combined in such electrodes to improve rate and cycle capability.
Additionally, MnO 2 materials have a limited loading of active materials, resulting in low energy density because of the low number of active sites. As a result, increasing the electrochemical utilization of MnO 2 's pseudocapacitance by rationally constructing MnO 2 -based electrodes with innovative architectures and dependable electric connections remains a signicant problem. Directly growing innovative integrated array mechanisms are fascinating in conducting substances for supercapacitors. It will provide synergistic effects from their respective materials, achieving high power density, energy density, and long cycle life. 24 Co 3 O 4 has become the subject of extensive research and development due to its low cost and high theoretical capacitance. Aside from that, introducing benecial ions into Co 3 O 4 and its composites can improve electrochemical performance due to the increased electric conductivity and the enhancement of faradaic redox reactions that result from this process. However, single-phase Co 3 O 4 cannot meet the demands of actual applications due to its low capacity and inferior cycling stability, which are caused by few electroactive sites, weak ion diffusion, and limited electric conductivity. 25,26 An alternative solution is a metal-organic framework (MOF) based synthesis of Co 3 O 4 to satisfy these difficulties. It is possible to enhance the performance of the material by using the MOF's porous structure. [27][28][29][30][31] This study aimed to investigate the electrochemical performance of MnO 2 nanosheets decorated on MOF-derived Co 3 O 4 synthesized in a facile two-step procedure. To construct Co 3 -O 4 @MnO 2 nanosheets, a sacricial template (ZIF-67) was used to prepare the Co 3 O 4 . Then MnO 2 nanosheet arrays were anchored to its surface using a hydrothermal technique. This unique structural design makes it possible to store a lot of energy and improve electrochemical performance by increasing the surface area.

Preparation of Co 3 O 4
The ZIF-67 precursor was used as a sacricial template in a twostep process to synthesize Co 3 O 4 . The rst step involves the synthesis of the precursor ZIF-67 using a standard preparation method. Where 1.74 g of cobalt nitrate and 1.968 g of 2-methylimidazole have dissolved in 60 mL and 20 mL of methanol, respectively. The two solutions were then combined and gently shaken for 7 minutes. The mixed solution was then stored at room temperature for 48 hours. The resulting precipitate was separated using a centrifuge, rinsed with methanol and dried for 12 hours at 80 C to obtain the precursor ZIF-67. The second step entails the utilization of the synthesized ZIF-67 as a sacri-cial template. Then the sample was calcinated in an argon atmosphere for 4 hours with a heating rate of 1 C min −1 up to 550 C and in an air atmosphere for 4 hours up to 350 C. The nal collected dark powder was named ZIF-67 derived Co 3 O 4 .

Preparation of Co 3 O 4 @MnO 2
To prepare Co 3 O 4 @MnO 2 , 45 mg of prepared Co 3 O 4 powder were ultrasonically dispersed in 30 mL of DI water. Then the prepared solutions were mixed with 30 mL DI water containing 30 mg, 45 mg, and 60 mg of potassium permanganate, respectively and placed into the stainless-steel Teon reaction kettle to react for 14 hours at 145 C. The nal grey powders collected by centrifugation and drying at overnight are named as Co 3 O 4 @-MnO 2 -1, Co 3 O 4 @MnO 2 -2 and Co 3 O 4 @MnO 2 -3 corresponding to 30 mg, 45 mg, and 60 mg of KMnO 4 respectively.

Material characterizations and electrochemical measurements
The crystallographic structure and surface element compositions of prepared samples were characterized by XRD (BRUKER USA D8 Advance, Davinci) and X-ray photoelectron spectroscopy (PHI Versaprobe III). A eld-emission scanning electron microscope (Thermosceintic Apreo S) and highresolution transmission electron microscope (JEOL Japan, JEM-2100 Plus) were used to examine the microstructure and morphology of the synthesized materials. A thermogravimetric analyzer determines the thermal stability of the sample. Thermogravimetric measurements are taken in a nitrogen atmosphere from 50 to 800 degrees Celsius at a linear heating rate of 10 degrees Celsius per minute. Furthermore, FTIR was used to characterize the various bonds present on the surfaces of the prepared material (SHIMADZU, IRTRACER 100). The electrochemical performances of all prepared electrodes were performed in 1 M KOH using an electrochemical workstation (Biologic-SP200 Potentiostat). The three-electrode assessment used the active material as the working electrode, platinum as the counter electrode and Hg/HgO as the reference electrode. The two-electrode evaluation was carried out with Co 3 O 4 @MnO 2 -3 as the positive electrode and activated carbon as the negative electrode. The electrode material for the assessment was prepared by evenly blending the active material, conductive substance (carbon black), and binder (NMP) in an 8 : 1 : 1 ratio and then coating it on half of the 0.5 Â 1 cm nickel foam.
The following formula determines the specic capacitance (C p ) from the chronopotentiometry charge-discharge curves. 32 where Im is the current density (A g −1 ), Dt is the discharge time (s), and DV is the voltage window (V).
The following equation is to determine energy density (E) and power density (P). 33 where E, C, DV, P and Dt are the energy density, specic capacitance, potential window, power density and discharge time.

Structural characteristics
An X-ray diffraction spectrometer (XRD) was used to examine the as-prepared samples' crystal structure and phase purity. Fig. 1 Fig. 2(b) shows the TGA analysis of ZIF-67. For ZIF-67, three decomposition steps occur under the N 2 atmosphere TGA analysis. The absorption of methanol molecules observed on the surface of ZIF-67 at temperatures below 150 C is the rst stage of weight loss. The carbonization of 2-methylimidazole molecules in ZIF-67 pores from 250 to 490 C causes the second stage (9.2%). The third stage's signicant loss (33.3%) occurs when the temperature reaches a specic point. At this point, the organic groups and ZIF-67 dodecahedrons break down, revealing the nal phase above 500 C. The resulting calcined materials' oxidation states and chemical compositions were conrmed using XPS. The XPS survey spectra of Co 3 O 4 @MnO 2 -3 are shown in Fig. 2(c). As shown in Fig. 2(d), the high-resolution XPS spectra of Co 2p can be tted into two primary peaks at 780.06 and 795.25 eV and can be associated with the binding energies of Co 2p 3/2 and Co 2p 1/2 , respectively. The lower two peaks, 789 and 805.35 eV, can be assigned to the binding energies of 2p 3/2 and 2p 1/2 of Co(II) and Co(III). These results reveal the presence of the Co 3 O 4 phase in the prepared sample. 43,44 The high-resolution XPS spectra obtained from Mn 2p are shown in Fig. 2(e). The primary two peaks are centred at 642.24 and 653.84 eV, respectively, with a spin energy separation of 11.6 eV corresponding to Mn(IV). These ndings are based on the electronic orbits of Mn 2p 3/2 and 2p 1/2 , indicating that the compounds are in the Mn(IV) state. 45 In Fig. 2(f), the binding energy peak at 532.2 eV is attributed to the oxygen atoms in the hydroxyl groups. In contrast, the intense peak at 529.8 belongs to the oxygen atoms in the Co 3 O 4 @MnO 2 -3 chemical compositions. 46,47

Morphological analysis
As shown in the schematic illustration in Fig. 3, The dodecahedral ZIF-67 was used as a template for the synthesis of Co 3 O 4 aer pyrolysis at 550 C in an argon atmosphere and calcinated at 350 C in air. Then, Co 3 O 4 was used as the substrate to grow MnO 2 nanosheets via a hydrothermal process at 140 C to form a Co 3 O 4 @MnO 2 . This formation is well proved by SEM and HR-TEM results. As illustrated in Fig. 4(a and b), the ZIF-67 has a standard form and a smooth surface. In Fig S1 †(b), we show the TEM image of the Co/C sample and it is found that the sample is stable aer calcination at 550 C under an inert gas atmosphere. In contrast, the Co 3 O 4 driven by ZIF-67 has a rough surface, illustrated in Fig. 4(c) and the TEM image is illustrated in Fig. 4(d) due to the collapse of a portion of the MOF frame during the calcination process. As shown in Fig. 4(e-g), the staggered MnO 2 nanosheets clusters vertically grown on the surface of Co 3 O 4 as the concentration of KMnO 4 increases, forming a structure similar to dodecahedral. The TEM images reect the unique hierarchical Co 3 O 4 @MnO 2 nanostructure containing a core of Co 3 O 4 dodecahedral and a shell of MnO 2 nanosheets. The interface contacts between the black core Co 3 O 4 and grey shell MnO 2 nanosheet arrays were seen vertically. Surprisingly, the concentration of KMnO 4 affects the distribution of MnO 2 nanosheets on the Co 3 O 4 surface during the hydrothermal process. As illustrated in Fig. 4(e), the surface of the Co 3 O 4 @MnO 2 -1 sample was only covered by partial and uneven MnO 2 nanosheets due to insufficient KMnO 4 . However, excessive KMnO 4 results in the overlapping of MnO 2 nanosheets on a portion of the Co 3 O 4 surface, as well as partially formed MnO 2 nanosheets, as illustrated in Fig. 4(f) Co 3 O 4 @MnO 2 -2 sample.
In comparison, the composite Co 3 O 4 @MnO 2 -3, MnO 2 nanosheet arrays with staggered and orderly vertical growth exhibit an appealing and satisfying morphology, as illustrated in Fig. 4(g), which corresponds to the superior electrochemical performance. Additionally, as illustrated in Fig. 4(h), the interplanar crystal spacing of the well-dened lattice fringes is 0.23 nm, which corresponds to the (3 1 1) plane of cubic Co 3 O 4 and amorphous MnO 2 overlayer on the surface of Co 3 O 4 core could be clearly identied. Furthermore, the energy dispersive spectroscopy (EDS) indicates Co, Mn, O, and C presented in the Co 3 O 4 @MnO 2 -3 shown in Fig. 4(i), which is also consistent with the XPS results. In Fig. S3, † the element mapping images revealed a homogeneous distribution of all elements, conrming that the Co 3 -O 4 @MnO 2 -3 was successfully synthesized.

Surface area analysis
Furthermore, the specic surface area and porous characteristics of ZIF-67-derived Co 3 O 4 , Co 3 O 4 @MnO 2 -1, Co 3 O 4 @MnO 2 -2 and Co 3 O 4 @MnO 2 -3 were determined using N 2 isothermal adsorption-desorption measurements. The prepared samples were typical Type-IV isotherms with an H3 hysteresis loop ( Fig. 5(a)). According to the Brunauer-Emmett-Teller (BET) method, the determined specic surface area of Co 3 O 4 @MnO 2 -3 is 160.8 m 2 g −1 , while Co 3 O 4 @MnO 2 -1 (121.3 m 2 g −1 ) and Co 3 O 4 @MnO 2 -2 (138.5 m 2 g −1 ) and much better than ZIFderived Co 3 O 4 (109 m 2 g −1 ), emphasizing the superiority of the design of the core-shell structure. It can be observed that  the specic surface area and porosity of the prepared materials are greatly improved aer decorating MnO 2 nanosheets on the surface of cobalt oxide. Also, the mesopores structure can provide a plentiful ion transport/charge storage, which enhances the pseudocapacitance. The Barrett-Joyner-Halenda (BJH) technique determined the pore size distribution, as shown in Fig. 5(b) and reveals the mesoporous nature of all prepared samples. And the respective average pore size was obtained for Co 3 O 4 @MnO 2 -3 at around 8 nm, whereas ZIFderived Co 3 O 4 , Co 3 O 4 @MnO 2 -1 and Co 3 O 4 @MnO 2 -2 show pore size of 10, 9.2 and 8.6 nm respectively. The large specic surface area provides abundant opportunities for the electrode and the electrolyte to make complete contact, which builds a strong foundation for the excellent electrochemical performance of Co 3 O 4 @MnO 2 , which can be attributed to the material.

Electrochemical performance
Cyclic voltammetry (CV) curves of all electrodes with a potential window of 0 V to 0.6 V at scan rates 5, 10, 20, 30, 40, 50, 70 and 100 mV s −1 were taken. Fig. 6(a-d) represents the CV curve of Co 3 O 4 , Co 3 O 4 @MnO 2 -1, Co 3 O 4 @MnO 2 -2 and Co 3 O 4 @MnO 2 -3. The two mild redox peaks are observed in the CV curve of Co 3 O 4 ( Fig. 6(a)), while redox peaks become more evident aer adding the Mn element. Moreover, the redox peak location varies with different Mn concentrations. The redox peaks in CV curves are mainly associated with the faradaic redox behaviour. The Co 3 -O 4 @MnO 2 -3 electrode has a strong CV curve, exhibiting its maximum capacitance. It has been revealed that when the scan rate increases with current increases, the shape of the CV curves follows a similar pattern. The appearance of redox peaks and the deviation of the curves indicate that the storage mechanism is owing to the faradaic redox behaviours. The redox peaks are caused by electrolyte cations intercalating or de-intercalating in MnO 2 nanosheets, which relates to eqn (4). 48 The electrode material absorbs K + ions from the electrolyte during charging. Then, K + ions are released from the electrode material and released to the electrolyte during discharge. The cathodic peaks shied towards lower negative potential due to polarization with increasing scan rates.
CP curves demonstrated the typical faradaic behaviour of all prepared electrodes in the charge storage process at various current densities (1, 2, 3, 4, 5 and 6 A g −1 ). Discharge curves are nearly symmetric in pattern, with a slight IR drop at the beginnings of discharge, implying high redox reversibility. As  clearly observed in the CP results shown in Fig. 7(a-d), eqn (1) follows. The Co 3 O 4 @MnO 2 -3 electrode reveals a longer discharge duration than the other prepared electrodes and archives specic capacitance around 768 C g −1 at 1 A g −1 current density shown in Fig. 7(d). This maximum specic capacitance is ascribed to the nearly complete redox reaction achieved by the Co 3 O 4 @MnO 2 -3 electrode material. Its initial IR drop is relatively low, conrming intense contact of the active material with the current collector. The other electrode materials are Co 3 O 4 , Co 3 O 4 @MnO 2 -1 and Co 3 O 4 @MnO 2 -2, which achieve lower capacitance around 309, 415 and 585 C g −1 at 1 A g −1 current density, respectively shown in Fig. 7(a-c). The specic capacitances of all electrode materials are measured and plotted in Fig. 8. (a) Using the CP curves, indicating that Co 3 O 4 @MnO 2 -3 (768 C g −1 ) is statistically superior to that of bare Co 3 O 4 and other composite materials. The specic  capacitance of Co 3 O 4 @MnO 2 -3 is signicantly higher than the most oen reported Co 3 O 4 -based electrode materials in Table 1. Fig. 8(b) shows that the cycling stability of all electrodes was examined for 5000 cycles at a constant current density of 6 A g −1 . The Co 3 O 4 , Co 3 O 4 @MnO 2 -1, Co 3 O 4 @MnO 2 -2 and Co 3 -O 4 @MnO 2 -3 exhibit cycling stability around 77, 72, 83 and 86%, respectively. The Co 3 O 4 @MnO 2 −3 electrode suggests high cycling stability and electrical conductivity compared to other electrodes. Furthermore, specic capacitance increases during the rst few cycles due to the electrode material's activation inuence and increased mobility of the surface charge and electrolyte ions. 60,61 The relationship between peak current and sweep rate is determined to understand the charge storage kinetics process of all electrodes in 1 M KOH electrolyte. The peak current (I) measured from CV curves at various scan rates is calculated using the power-law equation. 61-63 where n is the scan rate, i is the peak current, a, b, k 1 and k 2 are adjustable parameters and i(V) is the current response at a xed potential V. The square root of the scan rate and the corresponding current response correlates to the diffusioncontrolled and capacitive control processes. The value b ¼ 1 implies that the capacitive-controlled charge storage mechanism provides a rapid capability primarily responsible for the power density usually seen in carbon-based materials. In contrast, b ¼ 0.5 suggests a diffusion-controlled charge storage mechanism. The anodic and cathodic peaks for the Co 3 O 4 @-MnO 2 −3 are shown in Fig. 9(a). The linear relationship obtained illustrates the diffusion characteristics of the materials. The anodic and cathodic peak current values contain an Rsquare value close to one, indicating that the material has redox behaviour, which is one of the criteria for battery-type electrode material. The slope in Fig. 9(b) is 0.55, indicating the pure diffusion-controlled and battery-type electrode. 63 Dunn's method can quantify the signicant contribution of the diffusion and the capacitance mechanism. This approach allows for the quantitative determination of the CV curves contributed by the capacitive and diffusion control processes at varied scanning speeds. At a scan speed of 5 mV s −1 , the red portion of the CV curve in Fig. 9(c) reects the contribution ratio (47%) occupied by the capacitance control mechanism and the contribution ratio (53%) occupied by the diffusion control mechanism. In Fig. 9(d), the percentage of capacitance and diffusion contribution at each scan rate is given as a histogram.
The capacitance contribution increases with increasing the scanning speed because capacitance control's surface effect is a quick process. The hybrid supercapacitor is represented in the diagram Fig. 10(a); the electrodes are Co 3 O 4 @MnO 2 as cathode, activated carbon (AC) as an anode, and 1 M KOH as an electrolyte. According to prior research, the hybrid supercapacitor made of carbon-supported materials has a high energy and power density. Because of its high porosity and conductivity, activated carbon (AC) is used as a negative electrode. Its broad potential window and good specic capacitance allow absorbing more ions from the electrolyte. [64][65][66][67] Furthermore, based on the above CV and CP results, Co 3 O 4 @MnO 2 -3 is assigned as a positive electrode for the two electrode systems. The operating potential window for the two electrode systems of Co 3 O 4 @MnO 2 -3//AC is performed by combining both electrodes. As shown in Fig. 10(a), the CV curves of Co 3 O 4 @MnO 2 -3, activated carbon, and hybrid supercapacitor electrodes were rst recorded independently. The CV was performed at various potential windows for a hybrid supercapacitor to determine the ideal operating range of the potential window. As shown in Fig. 10(b), a broad potential window of 1.45 V was obtained. CV data was recorded at multiple scan rates ranging from 5 to 100 mV s −1 to examine the performance of the two-electrode system shown in Fig. 10(c). The CV curves are quasirectangular to a particular optimum value and then depart signicantly at high potential. This variation from the typical rectangular form is caused by limiting ion transport on the electrode surface during redox processes at high scan rates. The CV curves remain unchanged even at higher scan rates, demonstrating that the hybrid supercapacitor has high-rate   capabilities and stability. Additionally, no signicant peaks have appeared, which indicates the hybrid supercapacitor exhibits dominating capacitive behaviour. Still, a more prominent knob in the curves indicates the existence of faradaic chemical processes. Charge discharge curves for the twoelectrode system are also presented in Fig. 10(d), which shows the charge storage of the hybrid supercapacitor. The CP curves are neither triangular nor humped in the voltage window range of 0 to 1.45 V, but rather a combination of both types. At various current densities, the charge-discharge curves are essentially symmetric. The low IR drop appears to conrm the low internal resistance and good rate capability and validate the high cycle stability of the material. Stability studies are essential to provide insight into the material's lifetime. In this study, the hybrid supercapacitor is subjected to 5000 charge-discharge cycles at 6 A g −1 current density (Fig. 10(e)). The hybrid supercapacitor using Co 3 O 4 @MnO 2 and activated carbon exhibited 85.5 percent capacity retention aer 5000 charge-discharge cycles.
The energy density of a hybrid supercapacitor is a signicant indicator for determining its energy storage ability. The energy density and power density can be calculated from eqn (2) and (3). The Co 3 O 4 @MnO 2 //AC hybrid supercapacitors in the voltage window from 0 to 1.45 V provide a maximum energy density of about 60.17 W h kg −1 at a power density of around 2674.37 W kg −1 . The prominent energy storage properties of Co 3 O 4 @MnO 2 //AC hybrid supercapacitor are mainly the vertically aligned nanosheets like Co 3 O 4 @MnO 2 electrode provide good specic capacitance in a wide voltage window. The Nyquist plot of hybrid supercapacitors given in Fig. 10(f) identied the stability of hybrid supercapacitor before and aer 5000 cycles. At low frequencies, the impedance rises substantially. It becomes nearly vertically parallel to the imaginary y-axis, indicating that the hybrid supercapacitor is pure capacitive. The charge transfer resistance (R ct ) at the electrode-electrolyte interface is represented by the small semicircular part in the high-frequency region for aer stability, which is combined with intrinsic resistance (R s ) due to ionic resistance of the electrolyte and intrinsic resistance of the current collector. The equivalent circuit corresponding to the EIS data (inset) shows a slight increase of R ct from 6.8 to 7.6 U obtained aer 5000 cycles.

Conclusion
In conclusion, we effectively synthesized vertically aligned nanosheets like Co 3 O 4 @MnO 2 with a distinct core-shell structure employing Co 3 O 4 synthesized by sacricing the ZIF-67 template as the precursor. As a result of the dense MnO 2 nanosheets on the surface covering, the specic capacitance of Co 3 O 4 @MnO 2 -3 reveals around 768 C g −1 , approximately two times that of the bare Co 3 O 4 , and exhibited good cycle stability and the porous structure of the material has a excellent BET surface area of 160.8 m 2 g −1 . Furthermore, a unique hybrid supercapacitor with positive and negative electrodes has been constructed with Co 3 O 4 @MnO 2 -3 and activated carbon, respectively. The hybrid supercapacitor provides high specic capacitance and long cycle life. Meanwhile, the energy and power densities were 60.17 W h kg −1 and 2674.37 W kg −1 , respectively. This method provides a compelling alternative for preparing MOF-derived Co 3 O 4 -based composites as highperformance supercapacitor electrodes.

Conflicts of interest
There are no conicts to declare.