1 Introduction

Energy exhaustion has prompted more and more researchers to study how to better manage and store energy [1]. People are advancing the concept of renewable energy, looking for better technology, and researching the storage of renewable energy at faster speeds. Supercapacitors, batteries, fuel cells, etc., are the most important electrochemical energy storage devices [2,3,4]. Among them, as a portable energy storage system, supercapacitors (SCs) stand out, with many unique advantages, such as long cycle capability, high-power output, excellent reversibility and a wide operating temperature range [5, 6]. In the past few years, due to its potential application in portable electronic devices, research on the new flexible supercapacitors has become one of the most focuses in the field of electrical energy storage [7]. Carbon materials, conductive polymers and transition metal oxides are conventional electrode materials for supercapacitors [8,9,10,11,12,13,14]. However, the rapid development of high-performance supercapacitors limits the application of traditional materials. Therefore, the development of new high-performance supercapacitor electrode materials is imminent [15].

Metallic organic framework (MOF) materials has become the darling of supercapacitor materials because of its many special properties, such as high accessible surface areas, tunable pore sizes, open metal sites and ordered crystalline structures [16, 17]. There are two main types of applications of MOFs in terms of supercapacitors. One is that MOFs are applied as templates for metal oxides [18, 19], mixed metal oxides [20, 21], metal nanoparticles [22,23,24], and porous carbon compounds [25, 26]. Another is that the original MOFs are directly applied as active electrode materials. In fact, in recent years, the direct use of the original MOFs as the active material of supercapacitors has been rapidly developed, which not only expands the application of MOFs materials, but also provides more development prospects for new electrode materials about supercapacitors. For instance, Yang et al. [27] synthesized a layered structure Ni-MOF, which has good specific capacitance (1127 F g−1 at 0.5 A g−1) and cycling performance. Lee et al. [28] successfully synthesized Co-MOF film with a specific capacitance beyond 206.76 F g−1 and only lost 1.5% after 5000 cycles. Yan et al. [29] synthesized an electrode material Ni-MOF-like accordion showing specific capacitances of 988 F g−1 at 1.4 A g−1. Liu et al. [30] synthesized a layered Cu-MOF structure with an excellent specific capacitance of 1274 F g−1, and it lost 12% of the capacitance during the cycling. However, the current number of related reports is limited, and the supercapacitor performance of these materials needs to be further studied due to poor conductivity and large steric hindrance [31, 32].

Therefore, in order to develop more kinds of MOF electrode materials and further improve their supercapacitance, it is necessary to design a MOF material with a good structure. Not only does it have a stable structure, but it also has a conductive network frame, which ensures a capacitive process with easy electrolyte diffusion and rapid electron transfer and increases the chances of achieving enhanced energy storage performance. Recently, the Behera project from India has synthesized a new MOF structure, which brings some ideas for our research [33]. Co-HKUST (Hong Kong University of Science and Technology) is a new MOF made up of cobalt nodes with 1,3,5-benzenetricarboxylic acid struts between them. The Co-HKUST has the same structure as Cu-HKUST-1 and has a high degree of stability. The application of energy storage is one of the important applications of MOF materials, and it is also one of the most studied properties in the performance of MOF materials. Co-HKUST, which has the same structure as Cu-HKUST-1, should have the same excellent electrochemical performance as Cu-HKUST-1, but its electrochemical performance needs to be studied urgently.

Herein, we first successful loaded the Co-HKUST nanobundles on the foamed nickel via a simple hydrothermal method and then applied to supercapacitors. The Co-HKUST nanobundles are similar in structure to Cu-HKUST, providing good electron and ion transport pathways for the storage of electrolyte ions. In addition, the Co-HKUST nanobundles loaded on the foamed nickel can effectively increase the electrical conductivity of Co-HKUST and enhance the specific surface area. Meanwhile, the asymmetric supercapacitors (Co-HKUST//Activated carbon ASCs) were assembled. The Co-HKUST//activated carbon (AC) devices had an excellent specific energy of 71.3 Wh kg−1 at a specific power of 809.9 W kg−1. Even more interesting is that Co-HKUST//AC ASCs exhibit a good cycling performance at 5000 cycles. All the results of tests show that the Co-HKUST nanobundles are promising and innovative energy storage materials.

2 Experimental

2.1 Synthesis of Co-HKUST nanobundles

All drugs were obtained from Aladdin Shanghai without further purification. The porosity of the foamed nickel is 95–98%, the thickness is 1.7 mm, and the purity is 99.9%. The foamed nickel used was cleaned before the experiment, and ultrasonically washed with acetone, 1 M HCl and ethanol in that order. 0.0184 g of 1,3,5-benzenetricarboxylic acid (BTC) was dissolved in 35 ml of water and 35 ml of ethanol, and stirred for 1 h, after which 0.002 mol of cobalt chloride was added and dispersed for 1 h. The previously treated foamed nickel was cut into some 1 × 5 cm strips into a polytetrafluoroethylene reactor, and the dispersed reaction solution reacted at 120 °C during 16 h. The reaction vessel was naturally cooled and the foamed nickel ultrasonicated with distilled water and ethanol for 5 min, respectively. Last, the sample placed in an oven and dried at 60 °C for spare.

2.2 Characterization

The phase analysis of the obtained sample was carried out using a Bruker X-ray powder diffractometer (D8 Fous in Germany, Cu Kα, λ = 0.1541 nm). The sample was pressed on the sample stage. The morphology and structure of the prepared materials were characterized by a Hitachi S-4800 field emission scanning electron microscope (SEM). X-ray photoelectron spectroscopy (XPS) was obtained on the ESCALAB 250Xi from Thermo Fisher Scientific with Al Kα radiation. TEM and HRTEM are tested on a copper target by FEI F30 at a voltage of 200 kV. Fourier transform infrared (FTIR) transmission spectra were recorded on a Nicolet iS10 IR spectrophotometer. I–V curves were recorded by two-point probe measurement (Keithley 2400 semiconductor parameter) at room temperature. The surface area of the as-synthesized sample is measured by the Brunauer–Emmett–Teller (BET) method and the pore size distribution by the Barrett–Joyner–Halenda (BJH) method on Tristar II 3020 from Micromeritics Instrument Corporation.

2.3 Electrochemical measurements

In the electrolyte of KOH solution (1 M), a platinum electrode and a saturated calomel electrode (SCE) electrode were used as counter electrode and reference electrode, respectively, and the Co-HKUST was used as working electrode for the three-electrode electrochemical test. The electrochemical performance of the two electrodes was tested using asymmetric supercapacitor assembly, assembled with 6 M KOH as the electrolyte. In the two-electrode test system, the positive electrode is Co-HKUST and the activated carbon is the negative electrode to prepare an aqueous asymmetric supercapacitor (Co-HKUST//AC). All cyclic voltammetry (CV), constant current charge and discharge (GCD), alternating current impedance (EIS) and cycle were performed on the Shanghai Chenhua CHI660E electrochemical workstation. The asymmetric supercapacitor (Co-HKUST//AC) was tested on Wuhan Lan He CT2001A.

3 Results and discussion

A schematic of the Co-HKUST synthesis is given in Fig. 1. Two sheets of the foamed nickel were added to the pink solution in which 1,3,5-benzenetricarboxylic acid (BTC) and cobalt chloride were dissolved. The crystal structure of Co-HKUST is composed of Co2(COO)4 units during the high temperature hydrothermal reaction. Each Co2(COO)4 unit is locked by four carboxyl groups from four equivalent BTC molecules, and each BTC molecule is linked to three Co2(COO)4 units to form a Co-HKUST three-dimensional framework during the hydrothermal reaction [33].

Fig. 1
figure 1

Synthetic schematic of the Co-HKUST

Full size image
Fig. 2
figure 2

a XRD pattern and b FTIR spectra of Co-HKUST

Full size image

The XRD characterization of the samples is shown in Fig. 2a. It can be seen from the XRD pattern that the peaks of the samples are at 12.5, 9.2, 7.6, 25.3, 44.5, 51.9, and 76.4, while the peaks of 44.5, 51.9, and 76.4 are the characteristic peaks of the foamed nickel. Other peaks are each comparable to the Co-HKUST characteristic peak. Additionally, the Co-HKUST was characterized by infrared in Fig. 2b. The peaks at 3428 cm−1 and 1637 cm−1 are respectively due to the O–H and H–O–H stretching peaks in the water, with the presence of peaks at 1104 cm−1 and 760 cm−1 corresponding to the C=C of aromatic benzene. Further, the other two peaks at 1430 cm−1 and 1366 cm−1 can be indexed as the presence of Co–O and –COO− bonds, and the peak at 721 cm−1 is a meta-substitutional absorption peak of the benzene ring which further confirms the formation of the Co-HKUST. These results are consistent with previous research results [32, 33].

Fig. 3
figure 3

a–c Different multiplication of SEM images, d TEM images and HRTEM images, e element mapping of Co-HKUST

Full size image

The SEM images of the Co-HKUST are shown in Fig. 3a, with the information on the nanobundles structure on the substrate of the foamed nickel. It can be shown on the low-magnification images that Co-HKUST nanobundles grows more uniformly on the surface of the foamed nickel, and the loose structure remains good with uniform size in Fig. 3b. The smooth surface of the nanobundles can be clearly seen under the high-magnification electron microscope. The morphology is close to the straw bundle. Meanwhile, it can be found that the diameter of Co-HKUST is approximately between 50 and 100 nm from Fig. 3c. Additionally, the Brunauer–Emmett–Teller (BET) test can also prove. The N2 adsorption–desorption isotherms at 77 K show that Co-HKUST possesses a large BET surface area of around 20.0777 m2 g−1 and a large number of mesopores with average pore diameter at about 2–5 nm in Fig. S1a, 1b. The N2 adsorption–desorption isotherms and pore size distribution of the foamed nickel are also exhibited in Fig. S1c, 1d. The results show that the presence of the Co-HKUST enhances the specific surface area. To evaluate the electrical behavior of Co-HKUST, the I–V responses were measured at room temperature in Fig. S2. The conductance of the pure foamed nickel and the Co-HKUST are, respectively, 261.9 ± 0.1 mS and 696.5 ± 0.3 mS. Due to the growth of the Co-HKUST structure on the foamed nickel, the conductance becomes larger and the resistance becomes smaller, thereby having better electrochemical performance.

Fig. 4
figure 4

a Full XPS spectra, b Co, c C, d O of Co-HKUST

Full size image

TEM analysis also confirmed its nanobundles characteristics in Fig. 3d. The HRTEM image exhibits well-resolved lattice fringes, revealing good crystallization characteristics of the Co-HKUST in Fig. 3d3. This is consistent with the XRD pattern. Meanwhile, the element mapping of the Co-HKUST nanobundles is characterized in Fig. 3e. In Fig. 3e, it can be seen that all elements of the Co-HKUST are distributed over the sample and are more evenly distributed. In a word, it can be found that Co is mainly distributed on the sample, and C, N, O, which are distributed in the air, are distributed on the sample and the substrate. The results of TEM were consistent with results of the surface elemental samples, confirming the successful synthesis of the Co-HKUST.

The surface chemical state of the prepared Co-HKUST and the elemental composition were confirmed by XPS measurement. It can be found that cobalt, carbon and oxygen are distributed on the surface of the Co-HKUST in Fig. 4a. In the element spectrum of cobalt in Fig. 4b, the peaks at six positions correspond well to Co 2p1/2, Co 2p3/2, and the position of the satellite peak, which proves the presence of Co (II) and Co(III) in the Co-HKUST [34,35,36]. In the C 1s spectrum in Fig. 4c, the peak at 284.84 eV corresponds to C–C, and 288.14 eV corresponds to O–C = O bond, and the results are consistent with the structural skeleton of Co-HKUST [35, 37, 38]. In the spectrum of O1s in Fig. 4d, the peaks at 531.4 eV and 533.1 eV are consistent with the previously reported C–O bond, O=C–O bond, demonstrating the presence of Co2(COO)4 structural units in the Co-HKUST [35, 39]. In summary, all results of test are consistent with the Co-HKUST structure, demonstrating the successful synthesis of the Co-HKUST structure on the foamed nickel.

Fig. 5
figure 5

a Cyclic voltammetry curves at 2, 5, 10, 25, 50 mV s−1, b Galvanostatic charge–discharge curves at 1, 2, 5, 10, 25 A g−1, c the rate performance, d electrochemical impedance spectroscopy curve of Co-HKUST

Full size image

Figure 5a exhibits the electrochemical performance of Co-HKUST through three-electrode tests. The area of the Co-HKUST exposed to the electrolyte is 1 cm2, and the mass is about 2–3 mg. From the cyclic voltammogram of Co-HKUST, as the sweep speed increases, the shape of the CV curve has no clear changes. The shape of the cyclic voltammetry curve does not show the characteristics of pure electric double-layer capacitors (EDLC), and a pair of significant redox absorption peaks appearing between 0.2 and 0.3 V show a typical pseudo-capacitor type [40]. The possible charge storage mechanisms are in the redox reactions of equations [40,41,42]:

$$ \begin{aligned} & {\text{Co}}_{s}^{{{\text{II}}}} + {\text{OH}}^{ - } \leftrightarrow {\text{Co}}_{s}^{{{\text{II}}}} ({\text{OH}})_{ad} + e^{ - 1} \\ & {\text{Co}}_{s}^{{{\text{II}}}} ({\text{OH}})_{ad} \leftrightarrow {\text{Co}}_{s}^{{{\text{III}}}} ({\text{OH}})_{ad} + e^{ - 1} \\ \end{aligned} $$
Fig. 6
figure 6

a Cycle performance of Co-HKUST at 7.5 A g−1 (insets: the SEM of the Co-HKUST after cycling), b the EDS of the Co-HKUST after cycling

Full size image

The capacitance of the material is mainly derived from a pseudo-capacitor based on a redox mechanism. In addition, as the scanning rate increases, the surrounding area of the CV curves increases, indicating that the Co-HKUST material is favorable for the fast Faraday reaction. It can be found that the redox peak can be observed even if the sweep speed is increased to 50 mV s−1, indicating that the material has good rate performance, which is consistent with the test results [32]. In addition, the potentials of the oxidation and reduction peaks are shifted to larger positive and negative values, respectively, which may be mainly due to an increase in the internal resistance of the electrode.

Meanwhile, the galvanostatic charge–discharge (GCD) curves of Co-HKUST nanobundles electrode was further investigated for the chronopotentiometry (CP) curve in Fig. 5b. There is a significant discharge platform at about 0.2–0.3 V reflecting the strong pseudo-absorption behavior of the reversible Faraday redox reaction, which is in accordance with the results of the CV curves at all current densities in Fig. 5a. In addition, Co-HKUST powder was synthesized without adding the foamed nickel under the same conditions, and its electrochemical performance was tested in Fig. S3. The similar redox peaks also appear in the CV curve of Co-HKUST powder in Fig. S3a. Simultaneously, after the calculation, the specific capacitance is 578.6 F g−1, 568.2 F g−1, 500.3 F g−1, 480.7 F g−1, 452.3 F g−1 at 1, 2, 5, 7.5, 10 F g−1, respectively. The rate performance curve is shown in Fig. 5c. It can be found that the electrode material has a large surface area and retains a capacitance of 452.3 F g−1 at 10 A g−1, indicating that the material has a good rate capability. The rate performance is 78.2%, indicating that the Co-HKUST has a large current charge and discharge capability and is an excellent electrode material. From the GCD curve in Fig. S3b, the specific capacitance of the Co-HKUST powder is 367.4 F g−1, which is lower than the Co-HKUST. Because the Co-HKUST powder has added a binder when synthesizing an electrode, its electrochemical performance is reduced. It also proved the superiority of the Co-HKUST growing on the foamed nickel. In order to evaluate the conductivity and charge transport properties at the electrode/electrolyte interface, electrical impedance spectroscopy (EIS) is performed in the frequency range of 0.01 Hz to 10 kHz. The EIS curve of the Co-HKUST composes a semicircle in the high-frequency region and a straight line in the low-frequency region, which indicates that the reaction process of the electrode reaction under steady state conditions is a mixed reaction control process composed of charge transfer and diffusion migration. The EIS curve of the Co-HKUST powder is also shown in Fig. S3c. Generally, Rs refers to the solution resistance, including electrolytic resistance, inherent resistance of the material, and contact resistance between active material and current collector [43]. Rct is a charge transfer resistance existing due to the Faraday reaction. The smaller the semicircle in the high-frequency region, the smaller the charge transfer resistance [44, 45]. The impedance tests in Fig. 5d show that the Co-HKUST has a small solution resistance and a small ion transfer resistance in 1 M KOH, which is a good electrochemical energy storage material.

One of the more important factors in evaluating the electrochemical performance of electrode materials is the cycle performance. Thus, the cycling performance of Co-HKUST was also tested in 1 M KOH solution at 7.5 A g−1 shown in Fig. 6. At the first lap of the cycling, the specific capacitance of Co-HKUST is 804.7 F g−1. In the 1000th lap, the specific capacitance of Co-HKUST is 713.5 F g−1, and the retention is 88.7% of the initial capacitance, which shows the excellent cycling performance. The cycling performance of the Co-HKUST powder was also given in Fig. S3d with less specific capacitance. Additionally, the Co-HKUST after cycling was characterized. The result of SEM was showing that the nanobundles structure did not collapse in Fig. 6a insets. The XRD pattern of the Co-HKUST after cycling is shown in Fig. S4. It is seen that the electrodes cycled retained its structure. Meanwhile, the surface elements have not changed in the EDS test in Fig. 6b. The above results were also the reason for maintaining high retention.

Additionally, an asymmetric supercapacitor Co-HKUST//AC was assembled and electrochemically tested. As for a supercapacitor, the charge balance follows the relationship Q+ = Q−. The charge stored by each electrode usually depends on specific capacitance (C), the potential range for charge/discharge process and the mass of the electrode (m) following equation: m+/m− = C− × V−/C+ × V+. Therefore, to design the Co-HKUST//AC ASC, loading mass ratio of active materials (m(Co-HKUST)/m(AC)) is 0.38 according to formula [46, 47]. Figure 7a shows the CV curve for the Co-HKUST and AC in the two-electrode electrochemical test at a sweep speed of 10 mV s−1. It can be seen that the AC electrode applies in a potential range of − 1 to 0 V, while the Co-HKUST electrode material operates between 0 and 0.44 V. Therefore, the fabricated asymmetrical device can work in the range of 0–1.44 V. However, in practice, we observed that the electrode active material can operate stably within a wide potential window of 1.6 V, so the operating potential window of the device is limited to 1.6 V. CV curves were performed on the ASC apparatus fabricated in Fig. 7b. The samples were tested at different scanning speeds (2, 5, 10, 25, 50 mV s−1), and the CV curves were typical and not deformed at all scanning rates. The pseudo-capacitor characteristics show good rate performance [29]. The constant current discharge curve of the Co-HKUST//AC device at different current densities is shown in Fig. 7c. As shown in Fig. 7c, the Co-HKUST//AC devices reach a high specific capacitance of 200.5 F g−1 at 1 A g−1. The specific capacitance of Co-HKUST//AC device remains 123.7 F g−1 at 10 A g−1 with a high rate performance (61.7%). The EIS curve of Co-HKUST//AC was also composed of a semicircle in the high-frequency region and a straight line in the low frequency region in Fig. 7d. The analog equivalent circuit that meets the experimental data exhibits a very low Rs of 0.83 Ω and the Rct of only 1.60 Ω in the Fig. 7d inset. These values, as well as the low-frequency region near the line, exhibited a well supercapacitor performance of the material [48].

Fig. 7
figure 7

a Cyclic voltammetry curves of AC and Co-HKUST, b cyclic voltammetry curves at 2, 5, 10, 25, 50 mV s−1, c Galvanostatic charge–discharge curves at 1, 2, 5, 7.5, 10 A g−1 (inset: LED lights), d electrochemical impedance spectroscopy curves (inset: The equivalent circuit fitted to the experimental EIS data), e cycling stability of the Co-HKUST//AC (insets: LED lights at 5 min and 30 min) at 10 A g−1, f Ragone plot of energy density and power density compared with other reported data

Full size image

The four Co-HKUST//AC devices were used to illuminate a string of LED lights with 6 V rated voltage, the string can last more than 30 min. Meanwhile, after 5000 cycles of Co-HKUST//AC devices, the capacitance retention rate reached a staggering 91.8% in Fig. 7e. After calculation, the Co-HKUST//AC devices reach a high-power density and a high-energy density at 1 A g−1 (809.9 W kg−1, 71.35 Wh kg−1, respectively). In addition, at a high current density of 10 A g−1, the device maintains an energy density of 44.0 Wh kg−1 at 8.5 kW kg−1. For comparison, we list the power and energy density of other recently reported electrochemical energy storage material of similar structures. The Co-HKUST is one of the best reported recently in flexible supercapacitors, such as Co/Ni-MOF//CNTs-COOH (49.5 Wh kg−1, 2 A g−1) [44], AC//Ni/Co-MOF-rGO (72.8 Wh kg−1, 1 A g−1) [48], Ni-MOF/CNTs//rGO/g-C3N4 (36.6 Wh kg−1, 0.5 A g−1) [49] and so on [40, 50,51,52,53,54,55,56,57].

4 Conclusions

In this work, the Co-HKUST nanobundles were first loaded on the foamed nickel and used it as an electrical material for supercapacitors. The prepared Co-HKUST nanobundles have excellent electrochemical properties. After testing, the material has a specific capacitance of 578.6 in 1 A g−1. When the current density is 10 A g−1, the specific capacitance is 452.3 F g−1, and the rate performance is amazing 78.2%. After 1000 cycles at 7.5 A g−1, the cycle retention rate of Co-HKUST nanobundles is 88.7%. When assembled with an asymmetric supercapacitor (Co-HKUST//AC), the specific capacitance is 200.5 F g−1 at 1 A g−1. The capacitance retention can keep a staggering 91.8% after 5000 cycles at 5 A g−1. In addition, at current density of 1 A g−1, the energy density is 71.35 Wh kg−1 at power of 809.9 W kg−1. When the current increased to 10 times, the energy density can still be kept up to 44.0 Wh kg−1 at 8.5 kW kg−1. In summary, the Co-HKUST is an excellent electrochemical material with unlimited potential for use in materials science.