In Situ-Grown Heterostructured Co3S4/CNTs/C Nanocomposites with a Bridged Structure for High-Performance Supercapacitors.

As one of the most competitive candidates for energy storage devices, supercapacitors have attracted extensive research interest due to their incomparable power density and ultralong cycling stability. However, the large surface area required for charge storage is an irreconcilable contradiction with the requirement of energy density. Therefore, a high energy density is a major challenge for supercapacitors. To solve the contradiction, Co3S4/CNTs/C with a bridged structure is designed, where CNTs generated in situ serve as a bridge to connect a porous carbon matrix and a Co3S4 nanoparticle, and Co3S4 nanoparticles are anchored on the topmost of CNTs. The porous carbon and Co3S4 are used for electrochemical double-layer capacitors and pseudocapacitors, respectively. This bridged structure can efficiently utilize the surface of Co3S4 nanoparticles to increase the overall energy storage capacity and provide more electrochemically active sites for charge storage and delivery. The materials show an energy density of 41.3 Wh kg-1 at 691.9 W kg-1 power density and a retaining energy density of 33.1 Wh kg-1 at a high power density of 3199.9 W kg-1 in an asymmetrical supercapacitor. The synthetic technique provides a simple method to obtain heterostructured nanocomposites with a high energy density by maximizing the effect of pseudocapacitor electrode active materials.

Introduction

Supercapacitors, as energy storage devices, exhibit incomparable power density and ultralong cycling stability than secondary batteries.[1−7] However, its energy density is unsatisfactory. Based on the charge storage mechanism, supercapacitors can be divided into electrochemical double-layer capacitors and pseudocapacitors. Electrochemical double-layer capacitors (EDLCs) are generally characterized by a rapid charge–discharge process, high-power density, and long-life cycle. EDLCs employ carbon-based materials, such as porous carbon, as active electrode materials. Charge storage mainly depends on the electrochemical adsorption/desorption of ions at the electrode/electrolyte interface. The energy density and energy storage capacity of those supercapacitors are limited by the specific surface area.[8−12] Pseudocapacitors are based on the rapid surface faradic redox reactions to store electric energy. They can provide much great energy storage capacity than EDLCs, as the materials used in the type of supercapacitor, such as transition metal oxides, could perform at a specific capacitance close to their theoretical value.[13−15] However, their power density and life cycle are lower due to the chemical reactions involved in the charge–discharge process and low electrical conductivity. Therefore, many efforts have been made to increase the energy density by combining the advantages of EDLCs and pseudocapacitors.[16−18]

The energy density of a supercapacitor depends on its capacitance and the working voltage, following E = 0.5CU2, where E is the energy density (Wh kg–1), C is the specific capacitance (F g–1), and U is the overall voltage (V). One approach to increase energy density is to develop electrode active materials with a large specific capacitance.[1−3] Based on the charge surface-storage mechanism, a porous structure in nanoscale is indispensable for electrode materials that could combine EDLC and pseudocapacitor features, which have exhibited its potential in this regard.[15−17] The emergence of new structures such as embedded and wrapped structures provides new concepts for the combination of EDLC and pseudocapacitor electrodes to improve energy density.[18−21] Among numerous supercapacitor electrode materials, Co3S4 is regarded as a promising electrode material for commercial supercapacitors due to its environmental-friendly feature and low cost.[20] The performance of Co3S4 depends on its structure, morphology, and components, such as nanostructure, porous structure, or composite with other materials, and it is critical for Co3S4 with encouraging electrochemical performances such as high rate capability and long cycle life.[21]

As a typical family of novel materials, metal–organic frameworks (MOFs) are porous materials with a high specific surface area, large pore volumes, and tunable pore size. MOFs and MOF-derived materials (porous carbon materials and nanostructured metal or metal oxide materials) have demonstrated promising performances in the field of energy storage and conversion, such as Li-based batteries, Na-ion batteries, fuel cells, solar cells, and supercapacitors.[22] In the present work, Co3S4/CNTs/C nanocomposites with a bridged structure are constructed through the in situ growth of MOF-derived Co/C nanoparticles. ZIF-67 was carbonized to form Co nanoparticles in a porous carbon matrix first, and then it acts as a catalyst to prepare in situ CNTs synchronously. With the growth of CNTs, Co nanoparticles located on the endmost of CNTs are shifted from the inside to the outside; as a result, the location of Co nanoparticles is re-adjusted, which is convenient for the following sulfuration with l-cysteine used as a vulcanizing agent. In the nanocomposites, CNTs act as the bridges, Co3S4 nanoparticles located at the endmost of the bridges act as pseudocapacitors, and porous carbons are used for EDLCs.

Results and Discussion

Structure Characterization

The Co3S4 phase is observed from the X-ray diffraction (XRD) patterns of Co3S4/CNTs/C nanocomposites (Figure a), according to the standard card PDF42-1448 (cubic structure; space group: Fd-3m (22); a = 9.437 Å; vol = 840.4 Å3). The peaks at 31.4°, 37.9°, and 55.2° correspond to the (311), (400), and (440) planes, respectively. The particle sizes calculated from the half-width of those diffraction peaks are 11.1, 12.2, and 11.6 nm, respectively. The morphology and phase constitution of the Co3S4/CNTs/C nanocomposites were revealed by transmission electron microscopy (TEM) and the electron diffraction pattern (EDP) (Figure b,c). The Co3S4/CNTs/C nanocomposites indicated a bridged structure. Those CNTs indicated monodispersed characterization without entangling. In addition, the Co3S4/CNTs/C nanocomposites have a rhombic dodecahedron shape, which is inherent from the ZIF-67 precursor (Figure b,d). The scanning electron microscopy (SEM) image and corresponding EDS mapping of the Co3S4/CNTs/C nanocomposites are shown in Figure e–g, indicating the even distribution of Co and S elements.

Figure 1

Structure characteristics of the Co3S4/CNTs/C nanocomposites derived from ZIF-67. (a) XRD of the nanocomposites; (b, c) TEM and EDP of the nanocomposites; (d, e) SEM images and (f, g) EDS mappings of the nanocomposites.

The TEM image and high-resolution images of the CNT and Co3S4 in the Co3S4/CNTs/C nanocomposites are shown in Figure . The monodispersed CNTs have a length of about 100 nm with a diameter of about 15 nm (Figure b). Figure c shows the high-resolution image of a Co3S4 nanoparticle anchored on the endmost of the CNTs. Figure d shows a sealed multiwalled CNT with an inner diameter of about 5 nm and a thickness of about 5 nm. Figure d also shows that the CNT wall is composed of graphite carbon with a high defect density and a multilayer structure with an interlamellar spacing of 0.35 nm (Figure e). The nanoscale size and monodispersed characteristic of the Co3S4 particles explicitly revealed that they have a large specific surface area, which would be convenient for the permeation of the electrolyte during the process of charge storage/delivery and result in a high specific capacitance. The monodispersed nanosized Co3S4 particles may present a typical capacitance behavior with a linear dependence of the charge stored on the width of the potential window.

Figure 2

(a) TEM image of Co3S4/CNTs/C nanocomposites, (b) morphology of a CNT with a Co3S4 nanoparticle anchored on its end, and (c–e) high-resolution images of a Co3S4 nanoparticle with a crystal structure, the starting point, and the wall (about 10 layers, d ≈ 0.35 nm) of the CNT.

The elemental state of the as-synthesized Co3S4/CNTs/C nanocomposites was confirmed by X-ray photoelectron spectroscopy (XPS) (Figure a–e). For cobalt (Co) 2p shown in the high-resolution spectrum (Figure b), signals of Co 2p3/2 and Co 2p1/2 were observed at the binding energy range of 775–810 eV. The Co 2p3/2 spectrum was deconvoluted into two peaks at 778.1 and 780.1 eV, which can be assigned to the binding energies of Co–S bonding. The Co2+/Co3+ ratio is around 1:2, which coincides well with that in Co3S4 in the Co3S4/CNTs/C nanocomposites. Different from XRD (Figure a), no signals of the Co crystal phase were detected by XPS, indicating that the surface layer of Co3S4/CNTs/C is sulfurated by l-cysteine completely. Note that the small nanoparticles (∼11 nm) located on the top of CNTs can be penetrated by XPS. This evidence suggests that the nanoparticles located on the top of CNTs are Co3S4 nanoparticles. The S 2p peaks in Figure c consist of the sulfur (S) 2p3/2 peak (161.5 eV) and the S 2p1/2 peak (162.1 eV), which confirmed the Co–S bonding. Besides Co and S elements, nitrogen (N) and C were also detected by XPS, and the N 1s spectrum was deconvoluted into pyrrolic nitrogen, graphitic nitrogen, and pyridinic nitrogen (Figure d). Carbon includes graphite-like carbon, graphite carbon, and C–N bonding carbon (Figure e), which was also confirmed by the Raman spectrum (Figure f) for the Co3S4/CNTs/C nanocomposites. As seen from Figure f, the D-band associated with disorders or defects (1350 cm–1) and the G-band associated with highly ordered graphite (1580 cm–1) were observed at a ratio of the intensity ID/IG = 1.09, where the D-band corresponds to sp3-bonded carbon atoms and the G-band corresponds to sp2-bonded carbon atoms. A D + D′ peak at 2930 cm–1 was also observed. Generally, the D + D′ band is a specific signal associated with defect density so that Figure f shows that the graphite carbon in the MOF-derived Co3S4/CNTs/C nanocomposites has a high defect density. In addition, the functional group of graphite carbon was also confirmed by FT-IR, as shown by the signal at 1620 cm–1 in Figure g. The surface area and porosity of these Co3S4/CNTs/C nanocomposites were studied by using nitrogen adsorption–desorption isotherms (Figure h,i). The Co3S4/CNTs/C nanocomposites exhibit a specific surface area of 62.74 cm2 g–1, a total pore volume of 0.17 cm3 g–1, and an average pore diameter of 10 nm, suggesting that the Co3S4/CNTs/C nanocomposites have a mesoporous structure.

Figure 3

(a–e) XPS spectra of the Co 2p level, S 2p level, and N 1s level on Co3S4/CNTs/C nanocomposites; (f) Raman, (g) FT-IR, (h) nitrogen adsorption–desorption isotherms, and (i) porosity of the MOF-derived Co3S4/CNTs/C nanocomposites.

Formation Mechanism

To clarify the formation mechanism, the ZIF-67 precursor was carbonized at 600 °C, and no dicyandiamide was used in the synthesis (Figure a). Co/C nanocomposites were obtained via this conventional pyrolysis process. Figure b reports the XRD pattern of the Co/C nanocomposites with the Co crystal phase (JCPDS no. 15-0806; cubic structure; space group: Fm-3m (225); a = 3.545 Å; vol = 44.5 Å3). The peaks at 44.3°, 51.4°, and 75.8° correspond to the (111), (200), and (220) planes, respectively. The size of the Co nanoparticles was calculated based on the half-width of (111) and (200) diffraction peaks, and the values are 11.1 and 11.6 nm, respectively. The structure of the Co/C nanocomposites was also characterized by TEM, EDP, and high-resolution images (Figure c,d). Co nanoparticles dispersed in the carbon matrix have a diameter of about 10 nm. Figure e shows a high-resolution image of a Co/C nanocomposite, showing the monocrystal structure of the Co nanoparticle and the C-matrix around it. Based on the aforementioned results, Co nanoparticles acting as a catalyst for the growth of CNTs were generated in situ in the carbonization of the ZIF-67 precursor. The monodispersed characterization of Co nanoparticles led to monodispersed CNTs, and the diameter of the CNTs also depended on the size of the Co nanoparticles, indicating a size-confinement effect.

Figure 4

Structure characteristics of Co/C nanocomposites derived from ZIF-67. (a) Morphology of the Co/C nanocomposites with a rhombic dodecahedron shape; (b) XRD; (c) high-resolution image of the Co/C nanocomposites with monodispersed Co nanoparticles dispersed in a porous carbon matrix; (d) EDP of the Co/C nanocomposites; (e) amplified high-resolution image of a Co/C nanoparticle, showing the Co nanoparticle with a monocrystal structure and the carbon matrix with a lamellar graphene structure.

Figure a illustrates the schematic process of synthesizing the MOF-derived Co3S4/CNTs/C nanocomposites, where Co nanoparticles generated in the dicyandiamide condition act as a catalyst first in preparing CNTs and serve as the source for the formation of Co3S4 nanoparticles at the end of the process. Co/CNTs/C composites were obtained first by a one-step carbonization of the ZIF-67 precursor, and Co3S4/CNTs/C nanocomposites were obtained by sulfuration with l-cysteine as the vulcanizing agent. The growth and formation mechanisms of the Co3S4/CNTs/C nanocomposites can be described as follows (Figure b). (1) ZIF-67 was obtained by using the method reported by our previous paper.[23] (2) The ZIF-67 precursor was then carbonized at 600 °C, and Co nanoparticles were obtained synchronously. Meanwhile, as the source of CNTs, a g-C3N4 atmosphere was produced by the pyrolytic elimination of dicyandiamide. (3) The Co nanoparticles captured g-C3N4 to produce CNTs, and the nanoparticles themselves were staying on the endmost of CNTs in the process. The CNTs grew from the inside to the surface of the porous carbon matrix and extended to the outside space. Following the top-growth mechanism, Co nanoparticles migrated along with the growth of CNTs. Monodispersed CNTs without entangling were then formed due to the monodispersion of Co nanoparticles. (4) Co3S4 nanoparticles were produced by sulfuration with l-cysteine as the vulcanizing agent at 400 °C, and they anchored on the topmost of CNTs. It is noteworthy that the Co nanoparticles in the porous carbon migrated from the inside to the outside and redistributed in the in situ growth of CNTs, which facilitated the sulfuration of Co to Co3S4.

Figure 5

Schematic of the mechanism for the fabrication of Co3S4/CNTs/C nanocomposites as electrode materials with a combination function of EDLCs and pseudocapacitors. (a) Fabrication process of ZIF-67-derived Co3S4/CNTs/C nanocomposites; (b) growth and formation mechanisms of Co3S4/CNTs/C nanocomposites following the top-growth mechanism.

We note that the residual Co phase is observed from the XRD patterns and the EDP of Co3S4/CNTs/C. To evaluate the efficiency of sulfuration, the characteristics of Co/C and Co3S4/CNTs/C were studied by hysteresis loop analysis (Figure ). Both samples are ferromagnetic. The saturation magnetization, remanent magnetization, and coercivity of Co/C are 21.6 emu/g, 2.70 emu/g, and 111 Oe, respectively. The saturation magnetization, remanent magnetization, and coercivity of Co3S4/CNTs/C are 3.92 emu/g, 0.33 emu/g, and 80 Oe, respectively. Note that the Co/C and Co3S4/CNTs/C samples contain approximately 23.5 wt % Co. The normalized saturation magnetization values of Co/C and Co3S4/CNTs/C by the content of Co are 91.8 and 16.7 emu/g, respectively. The saturation magnetization of Co3S4/CNTs/C is significantly smaller than that of Co/C. This is understandable since most Co atoms (∼82%) combined with S atoms to form a Co3S4 compound in Co3S4/CNTs/C. Note that the saturation magnetization of Co is greatly affected by the particle size. The saturation magnetization of the bulk Co crystal is 161 emu/g,[24] whereas the previously measured saturation magnetization values of Co nanoparticles with average particle sizes of ∼15 and 27 nm are 79 and 158 ± 7 emu/g,[25] respectively. The decreased magnetization can be explained with the model of the spin-canted surface layer.[26] Briefly, the fine Co particles consist of two parts, a surface layer and an inner part. The magnetic moment of the surface layer cannot be turned entirely along the direction of the applied field, but it makes an average canting angle with the field, whereas the magnetic moment of the inner part can be aligned along the direction of the applied field. As a result, the saturation magnetization of the Co nanoparticles decreases with the decrease in particle size since the finer particles have a higher area of surface layer. The saturation magnetization of Co nanoparticles with an average particle size of ∼20 nm in Co/C is only 91.8 emu/g, which is approximately 57% of the saturation magnetization of the bulk Co crystal. These data suggest that the saturation magnetization measured for Co/C should be reliable. In addition, the Co3S4 phase is paramagnetic and has a low susceptibility (in an order of 4 × 10–6).[27] Based on the aforementioned results, saturation magnetization can be used to evaluate the residual Co content in Co3S4/CNTs/C nanocomposites.

Figure 6

Hysteresis loops for the Co/C and Co3S4/CNTs/C nanocomposites measured at room temperature.

Electrochemical Properties

The electrochemical performance of the MOF-derived Co3S4/CNTs/C nanocomposites as electrode materials was measured in 3 mol L–1 KOH. Figure a shows the cyclic voltammetry (CV) curves of the Co3S4/CNTs/C electrode materials at different scan rates (from 1 to 100 mV s–1) with a potential range of 0.0 to 0.5 V. The near-rectangular shape of the CV curves revealed that the Co3S4/CNTs/C nanocomposites fabricated displayed a linear dependence of the charge on the charging potential. The charges were stored in the Co3S4 pseudocapacitive electrode through surface faradic redox reactions, rather than through the simple accumulation of ions on the surface. The calculated capacitances of the Co3S4/CNTs/C electrode materials are 524, 389, 363, 336, 303, 237, and 117 F g–1 at 1, 2, 5, 10, 20, 50, and 100 mV s–1, respectively. Figure b gives the galvanostatic charge–discharge (GCD) curves of the Co3S4/CNTs/C electrode materials at current densities from 0.5 to 5 A g–1. It can be clearly seen that the MOF-derived Co3S4/CNTs/C nanocomposites have a typical capacitance behavior with a nearly linear dependence of the stored charge on the width of the potential window, which coincides well with that of the CV curve. The Co3S4/CNTs/C nanocomposites presented a high specific capacitance of 380 F g–1 at a current density of 0.5 A g–1, which remained at 330 F g–1 when the discharge current density was increased by five times (5 A g–1), indicating a good high-rate discharge ability (Figure c). The specific capacitance of MOF-derived Co/C nanoparticles at a current density of 0.5 A g–1 was 146 F g–1,[23] which was about 38% of the specific capacitance (380 F g–1) of Co3S4/CNTs/C nanocomposites. Based on these studies, the capacity contributions of the EDLCs (carbon materials) and pseudocapacitors (Co3S4) might be 38 and 62%, respectively.

Figure 7

Electrochemical performance of the Co3S4/CNTs/C nanocomposite electrode: (a) cyclic voltammetry curves, (b) galvanostatic charge–discharge curves, (c) HRD, and (d) EIS. Electrochemical performance of the assembled asymmetric supercapacitor: (e) CV curves of the Co3S4/CNTs/C nanocomposites and activated carbon, (f) CV curves in different potential ranges, (g) CV curves in different scan rates, and (h) galvanostatic charge–discharge curves at different current densities. (i) Ragone plots of the assembled asymmetric supercapacitor (inset: cycling stability of the assembled asymmetric supercapacitor).

Figure d shows the electrochemical impedance spectroscopy (EIS) of the Co3S4/CNTs/C electrode materials. An equivalent circuit was used (see the inset in Figure d), where Rs, Rct, and C1 represent the internal resistance, the charge-transfer resistance of the electrochemical reaction, and the constant phase element in the EIS, respectively. The semicircle at the high-frequency region reflects the impedance of the electrochemical reaction. The impedance spectra were fitted with an equivalent circuit model (inset in Figure d) by using the least-square method with ZVIEW electrochemical impedance software. The fitted results show that the Rct is about ∼0.21 Ω for the Co3S4/CNTs/C nanocomposites.

An asymmetric Co3S4/CNTs/C∥AC supercapacitor was assembled to further evaluate the Co3S4/CNTs/C electrode materials, where the Co3S4/CNTs/C nanocomposites and activated carbon were used as the positive and negative electrode materials (Figure e), respectively. The CV curves shown in Figure f were obtained in different potential windows at a scan rate of 10 mV s–1, indicating that the optimal operating voltage is 1.6 V for the asymmetric supercapacitor. Figure g shows the CV curves of the asymmetric supercapacitor at various current densities in the potential range of 0–1.6 V. The specific capacitance calculated from CV of the asymmetric supercapacitor is 130 at 1 mV s–1. The asymmetric supercapacitor presented a high specific capacitance of 105 F g–1 at a scan rate of 1 mV s–1, which still remained at 79 F g–1 when the scan rate was increased to 100 mV s–1. Figure h shows the GCD curves of the asymmetric supercapacitor at various current densities in the potential range of 0–1.6 V. Both the symmetrical shape of CV curves and the linear dependence of the GCD indicate a typical capacitance behavior as well as good reversibility. The energy density (Wh kg–1) and power density (W kg–1) of the Co3S4/CNTs/C nanocomposite electrode were evaluated by this constructed symmetrical supercapacitor. The Ragone plots of the asymmetric supercapacitor (Figure i) indicate that the specific energy density is 41.3 Wh kg–1 at a power density of 691.9 W kg–1, with a higher maintenance of the energy density of 33.1 Wh kg–1 at a high power density of 3199.9 W kg–1. In addition, the asymmetric supercapacitor exhibited high cycle stability with 100% capacitance retention after 10,000 cycles at 2 A g–1 (inset in Figure i), showing potentially wide applications.[28−32]

Conclusions

In this work, MOF-derived Co3S4/CNTs/C nanocomposites with a bridged structure were fabricated for the application in supercapacitors with a high energy density. As a catalyst for the formation of CNTs, Co nanoparticles were generated synchronously in the carbonization of the ZIF-67 precursor, and the monodispersed characterization of Co nanoparticles led to monodispersed CNTs without entangling. With the growth of CNTs, Co nanoparticles located on the endmost of CNTs migrated from the bulk of carbon to the surface, and Co/CNT/C nanocomposites were then produced. Sulfuration was used to convert Co nanoparticles to Co3S4 nanoparticles. As a result, the bridged structure can greatly maximize the efficiency of the Co3S4 nanoparticles acting as active electrode materials for pseudocapacitors and endow the Co3S4/CNTs/C nanocomposites with a high energy density, indicating that the bridged structure may be appropriate to combine the two electrode materials with different charge-storage mechanisms.

Experimental Section

Synthesis of Co3S4/CNTs/C Nanocomposites

Co3S4/CNTs/C nanocomposites were prepared with the following procedures: (1) Co(NO3)2·6H2O (1.1641 g) was dissolved into 25 mL of methanol in a beaker, and 1.3136 g of 2-methylimidazole was dissolved into 25 mL of methanol in another beaker. (2) The two solutions were combined, the mixture was stirred for 10 min, and then the beaker was set for 24 h at 25 °C. (3) The mixture was centrifuged and dried in a vacuum oven at 60 °C for 12 h to obtain the ZIF-67 precursor. (4) Dicyandiamide (1.0 g) and the ZIF-67 precursor (0.1 g) were added in two quartz boats, respectively, and they were placed in a tube furnace. (5) The quartz boat was heated in a N2 atmosphere to 600 °C at a heating rate of 2 °C min–1, and the temperature was maintained for 1 h to obtain ZIF-67-derived Co/CNTs/C. Nitrogen as a carrier gas was injected into the tube furnace from the end close to the quartz boat containing dicyandiamide. (6) Similar to the last procedure, 1.2712 g of l-cysteine and 0.04 g of Co/CNTs/C nanocomposites were added in two quartz boats separately, they were put in the tube furnace, and l-cysteine was positioned at the upstream of the carrier gas. (7) The quartz boat was heated in the N2 atmosphere at a heating rate of 2 °C min–1, and the temperature was kept at 400 °C for 4 h to obtain MOF-derived Co3S4/CNTs/C nanocomposites.

Characterization

The crystal structure and morphology of the Co3S4/CNTs/C nanocomposite samples were determined with a Rigaku D/max 2500pc X-ray diffractometer operated at 40 kV (100 mA), an S-4800 scanning electron microscope at an acceleration voltage of 10 kV, and a JEM-2100 transmission electron microscope at 200 kV. Magnetic properties were measured at room temperature by using a LakeShore 7407 vibrating sample magnetometer with a maximum field of 20 kOe. XPS was performed on an ESCALAB 250Xi system. The Raman spectra were determined on a Renishaw Gloucestershire at a laser wavelength of 514 nm. Nitrogen adsorption–desorption isotherms were obtained using an ASAP-2020e system at 77 K, and Fourier transform infrared spectroscopy was carried out on a Bruker Vector 22 with a wavenumber ranging from 4000 to 400 cm–1.

Electrochemical Characterization

Electrochemical measurements were performed in a standard three-electrode system composed of a Co3S4/CNTs/C electrode, a Pt electrode, and a HgO/Hg electrode. The working electrode was constructed by mixing the Co3S4/CNTs/C nanocomposites with a conductive agent (acetylene black) and binder (polyvinylidene fluoride) at a mass ratio of 80:10:10. The slurry was coated on a nickel cystosepiment and dried at 120 °C for 12 h under a vacuum. The working electrodes have a size of 1 cm × 1 cm, and their mass loading of active materials on the current collector is about 2 mg/cm2. Electrochemical performances were tested on a BTS-5V 10 mA system with the voltage range of 0–0.5 V in 3 mol L–1 KOH electrolyte solution. Both of the CV tests and EIS were conducted on a CHI 660E electrochemical workstation. The asymmetrical supercapacitors were constructed by assembling an asymmetrical supercapacitor with Co3S4/CNTs/C nanocomposites as the positive electrode and homemade activated carbon as the negative electrode.