Comparative Study of Nanocarbon-Based Flexible Multifunctional Composite Electrodes.

Although nanocarbon-based nanofillers have been widely used to improve the energy-storing and sensing functions of porous materials, the comparison of the effects of different nanocarbon-based fillers on the capacitive and flexible sensing properties of nanocarbon-based porous sponge composite supercapacitor electrodes by combining a carbon nanotube, graphene, and graphene oxide with porous sponge is incomplete. The specific capacitance of carbon nanotube-based electrodes is 20.1 F/g. The specific capacitance of graphene-based electrodes is 26.7 F/g. The specific capacitance of graphene oxide-based electrodes is 78.1 F/g, and the capacity retention rate is 92.99% under 20 000 charge-discharge cycles. Under a bending load of 180°, the capacitance retention rate of graphene oxide sponge composite electrodes is 67.46%, which indicates that the prepared electrodes of supercapacitor have the advantages of high capacitance and good flexibility at the same time. To demonstrate their performance, an array of three graphene oxide supercapacitors in series was constructed, which could light up a red light-emitting diode (LED). The tensile strength of carbon nanotube sponge composite electrodes is 0.267 MPa, and the tensile linearity is 0.0169. The experimental results show that graphene oxide-based sponge composite supercapacitor electrodes have the best capacitance performance and carbon nanotube sponge composites have the most potential as a flexible sensor.

Introduction

Recently, porous materials have been widely used in superhydrophobic materials, adsorption and catalytic materials, energy, electronics, and other fields due to their porosity, flexibility, and strong adsorption capacity.[1−3] Conductive porous materials have unique pore structure, large specific surface area, excellent conductivity, and electrochemical performance, as well as broad application prospects in the fields of supercapacitors, electrochemistry, and flexible sensors.[4,5] However, conductive porous materials have their limitations, such as uneven pore size distribution and poor adhesion between filler and pore structure.[6] Porous materials prepared by pyrolysis,[7] polymer curing[8] or freeze drying method, and foaming technique[9,10] will produce low porosity and weak adhesion.[11,12] In addition, the preparation process of some porous materials is complicated and their properties may be unstable. In this regard, commercial sponge with uniform pore size and stable structure shows great potential in the rational design and design of porous materials.[13]

Adding conductive fillers to the sponge is a common strategy to fabricate a conductive porous sponge material with capacitive energy storage performance.[14,15] Metal particle filler has a high conductivity, but the combined ability of metal filler and sponge porous material is weak, leading to poor performance scalability and low charge–discharge cycle stability, which limits its application in the preparation of high-performance porous electrodes material for supercapacitor.[16,17] Compared with metal packing, carbon packing is more effective in ion-transport channels due to the strong expansibility, good compatibility,[18] conductivity coupling,[19] and strong charge–discharge cycle stability.[20] Therefore, nanocarbon-based materials exhibit promising potential for the electrodes of supercapacitor.[21]

Among various kinds of nanocarbon materials, carbon nanotubes (CNTs), graphene, and graphene oxide show good electrical and mechanical properties.[22,23] Carbon nanotube sponge has been widely studied as an electrode material and a flexible sensor of supercapacitor.[24,25] The sponge porous composite with graphene as the filler has strong electronic carrying capacity,[26,27] and its capacitance performance has been widely studied.[28,29] Graphene oxide has many functional groups containing oxygen and hydrogen. It has a strong ability to transport electrons and ions, as well as excellent electrochemical performance.[30,31] It is a good choice to use graphene oxide as the active material to prepare electrodes of supercapacitor, which has caused extensive research.[32,33]

The combination of nanocarbon-based fillers and porous materials will greatly improve the capacitance, conductivity, and mechanical properties of nanocomposites, which is worthy of further exploration.[34−36] However, previous works pay little attention to compare the capacitance of porous nanocomposites with nanocarbon-based fillers such as carbon nanotubes, graphene, and graphene oxide.[37−39] Yang Zhang et al. prepared a carbon nanotube/graphene/polypyrrole ternary composite surge by electropolymerization and obtained high specific capacitance and excellent capacitance retention. However, the capacitance of the prepared CNT sponge has the potential to be further improved.[40] Cui et al. prepared a kind of multifunctional graphene sponge nanocomposite. It can be used as an electrode material of flexible sensor and supercapacitor, and its electrochemical performance has been tested.[41] Moussa et al. fabricated graphene sponge with manganese dioxide, poly (3,4-ethylenedioxythiophene), and graphene in situ to obtain high specific capacitance.[42] El-Kady et al. deposited graphene and polyaniline on sponge. The interconnected holes of sponge provided enough inner surface between GnPs/PANI composite and electrolyte, which promoted ion diffusion during charging and discharging, and obtained higher energy density and power density.[43] The specific capacitance of graphene sponge obtained from the above research is relatively low, and the number of tested graphene filler components is small, which needs further research and improvement (Chee et al).[44] Polypyrrole/graphene oxide/zinc oxide nanocomposites were prepared on soft nickel foam surface by potentiostatic electrochemical polymerization. Using the pseudocapacitance behavior of polypyrrole and zinc oxide and the double-layer capacitance behavior of graphene oxide, higher energy and power density are obtained.[44] The specific capacitance of graphene oxide obtained in the above study has room for improvement, and the capacitance of graphene oxide is not compared with other nanocarbon-based materials such as carbon nanotubes and graphene. In this experiment, these problems have been solved to a certain extent.

In this paper, three kinds of nanocarbon-based supercapacitor composite electrodes are shown. The porous sponge nanocarbon-based composite electrodes were prepared by the soaking method to combine carbon nanotubes (CNTs), graphene platelets (GnPs), and graphene oxide (GO) with porous sponge. The effects of different carbon nanofillers on the electrochemical and flexible sensing properties of porous nanocomposite electrodes for supercapacitors were studied. The electrochemical performance of three kinds of flexible nanocarbon-based supercapacitors is tested and studied. The tensile strength, elongation at break, and linearity under tensile load of three kinds of nanocarbon-based sponge porous composites were studied. The nanocarbon-based sponge porous nanocomposites electrodes also have the potential to serve as flexible sensors. The flexible graphene oxide-based supercapacitor prepared in this paper has good electrochemical performance.

Results and Discussion

Morphology of Nanocarbon-Based Sponge Composite Electrodes

Figure shows the comparison of scanning electron microscopy (SEM) images of carbon nanotube sponge composite electrodes with different filling fractions. Figure a1–a3 shows the morphology of carbon nanotube sponge composite electrodes at a filling fraction of 0.08 wt %. In Figure a1–a3, the connected pores and network structure of sponge and the carbon nanotubes attached to the sponge skeleton are clearly visible. Figure a2 enlarges the representative region of Figure a1. Figure a3 enlarges the details in Figure a2. In Figure a2, CNT is relatively sparse on the surface of sponge pore, forming an electronic channel with a certain degree of conductivity. However, a small number of carbon nanotubes may not optimize the ability of the entire conductive network to carry electrons and ions.

Figure 1

Scanning electron microscopy (SEM) images of (a1–a3) 0.08 wt % carbon nanotube sponge composite electrode, (b1–b3) 0.11 wt % carbon nanotube sponge composite electrode, and (c1–c3) 0.14 wt % carbon nanotube sponge composite electrode.

Figure b1–b3 shows the morphology of carbon nanotube sponge composite electrodes with a filling fraction of 0.11 wt %. Figure b2 enlarges a representative region of Figure b1. Figure b3 enlarges the details in Figure b2. CNT is well distributed on the surface of sponge pores, which will form electronic channels with strong conductivity. Compared with that shown in Figure a2, the carbon nanotubes in Figure b2 are more evenly distributed and more dense on the sponge skeleton, which greatly improves the ability of the whole conductive network to carry electrons and ions.

Figure c1–c3 shows the morphology of carbon nanotube sponge composite electrodes at a filling fraction of 0.14 wt %. Figure c2 enlarges a representative region of Figure c1. Figure c3 enlarges the details in Figure c2. It can be found that CNTs are densely distributed on the surface of sponge skeleton, many carbon nanotubes are stacked together, and serious agglomeration occurs. The aggregation of carbon nanotubes increases the internal resistance of the conducting network and weakens the conductivity of the electron channel. Compared with Figure b2, Figure c2 shows that the number of carbon nanotubes is too high and the distribution of carbon nanotubes is too dense on the sponge skeleton. The agglomeration of carbon nanotubes also reduces the ability of the whole conductive network to carry electrons and ions. Therefore, from Figure , we find that when the filling fraction is 0.11 wt %, the attachment on the sponge pore structure is the most uniform, and the formed conductive network may have the strongest ability to carry electrons and ions.

Figure shows the comparison of SEM images of different filling fractions of graphene sponge composite electrodes. Figure a1–a3 shows the morphology of the graphene sponge composite electrodes at a filling fraction of 0.16 wt %. In Figure a1–a3, the connection holes and network structure of the sponge and the GnPs attached to the sponge skeleton are clearly visible. Figure a2 enlarges the representative area of Figure a1. Figure a3 enlarges the details in Figure a2. In Figure a2, GnPs are relatively sparsely distributed on the surface of sponge pores, forming electronic channels with certain conductivity. However, it is equivalent to Figure a2 that a small number of GnPs may not optimize the ability of the entire conductive network to carry electrons and ions.

Figure 2

Scanning electron microscopy (SEM) images of (a1–a3) 0.16 wt % graphene sponge composite electrode, (b1–b3) 0.24 wt % graphene sponge composite electrode, and (c1–c3) 0.32 wt % graphene sponge composite electrode.

Figure b1–b3 shows the morphology of graphene sponge composite electrodes filled with 0.24 wt %. Figure b2 enlarges a representative region of Figure b1. Figure b3 enlarges the details in Figure b2. GnPs are evenly distributed on the surface of sponge pore structure, forming a highly conductive electron channel.[45] Compared with Figure a2, GnPs in Figure b2 are more evenly distributed and more numerous on the sponge skeleton. The GnPs with large number and uniform distribution on the surface of sponge skeleton will form a large specific surface area conductive network with good ability to carry electrons and ions.

Figure c1–c3 shows the morphology of graphene sponge composite electrodes filled with 0.32 wt %. Figure c2 enlarges the representative region of Figure c1. Figure c3 enlarges the details in Figure c2. The results show that it is equivalent to Figure c2 that GnPs are densely distributed on the surface of sponge skeleton and a large number of GnPs are also stacked together, causing serious agglomeration. The agglomeration of GnPs increases the internal resistance of the conductive network, affects the quantum tunneling effect between GnPs, and weakens the conductivity of the electronic channel. Compared with Figure b2, Figure c2 shows that the number of GnPs is too high and the distribution of GnPs on the sponge skeleton is too dense. The agglomeration of GnPs also reduces the ability of the whole conducting network to carry electrons and ions. Therefore, it can be seen from Figure that when the filling rate is 0.24 wt %, the attachment on the sponge pore structure is the most uniform, and the formed conductive network may have the strongest ability to carry electrons and ions.

Figure shows the comparison of SEM images of different filling fractions of graphene oxide sponge composite electrodes. Figure a1–a3 shows the morphology of the graphene oxide sponge composite electrodes at a filling fraction of 0.04 wt %. In Figure a1–a3, the connection holes and network structure of the sponge and the GO attached to the skeleton are clearly visible. Figure a2 enlarges the representative area of Figure a1. Figure a3 enlarges the details in Figure a2. In Figure a2, GO is sparsely distributed on the surface of sponge pores, forming channels with the certain ability to carry electrons and ions. However, comparing Figures a2 and 2a2, a small amount of GO may not be able to make the whole conductive network have enough ability to carry electrons and ions.

Figure 3

Scanning electron microscopy (SEM) images of (a1–a3) 0.04 wt % graphene oxide sponge composite electrode, (b1–b3) 0.06 wt % graphene oxide sponge composite electrode, and (c1–c3) 0.08 wt % graphene oxide sponge composite electrode.

Figure b1–b3 shows the morphology of graphene oxide sponge composite electrodes filled with 0.06 wt %. Figure b2 enlarges a representative region of Figure b1. Figure b3 enlarges the details in Figure b2. It can be found that GO is evenly distributed on the surface of sponge pore structure and can form high-performance channels for carrying electrons and ions. Compared with Figures b2 and 2b2, GO is more evenly distributed and has good dispersion performance than GnPs and CNT. Compared with Figure a2, the GO in Figure b2 is more evenly distributed and has more quantities on the sponge skeleton. Moreover, GO has many functional groups containing oxygen and hydrogen, which has a strong ability to carry electrons. Extensive GO with uniform distribution on the surface of sponge skeleton will form a conductive network with a large specific surface area and strong ability to carry electrons and ions.

Figure c1–c3 shows the morphology of the graphene oxide sponge composite electrodes filled with 0.08 wt % graphene oxide. Figure c2 enlarges the representative region of Figure c1. Figure c3 enlarges the details in Figure c2. The results show that comparing Figures c2 and 2c2, extensive GO are piled up together, which will also cause serious agglomeration. The agglomeration of GO increases the internal resistance of the conductive network, weakens the conductivity of the electronic channel, and affects the carrying capacity of GO. Compared with Figure b2, Figure c2 shows that the number of GO is too high and the distribution of GO on sponge skeleton is too dense. The agglomeration of GO also reduces the ability of the whole conductive network to carry electrons and ions. Therefore, it can be seen from Figure that when the filling fraction is 0.06 wt %, the attachment on the sponge pore structure is the most uniform and the formed conductive network may have the strongest ability to carry electrons and ions.

Carbon Nanotube-Based Sponge Composite Supercapacitor Electrodes

The carbon nanotube-based sponge composite supercapacitor electrodes have the characteristics of uniform conductive network, which means that it has the potential to act as electrodes material of supercapacitor. The electrochemical properties of carbon nanotube sponge composite electrodes were studied by cyclic voltammetry (CV), electrochemical impedance spectroscopy (EIS), and constant current charge–discharge (GCD). A three-electrode system was used in all electrochemical measurements, which could more accurately measure the electrochemical properties of carbon nanotube sponge composite electrodes.

Figure a shows the CV curves of the carbon nanotube sponge composite electrodes at a scan rate of 100 mv/s within 0–0.8 V. For different filler fractions of carbon nanotubes sponge electrodes, CV curves are almost rectangular in shape, implying typical capacitive behavior. This is due to the good conductivity of CNT. The CV curve of 0.11 wt % carbon nanotube sponge composite electrode has higher current density than 0.05 and 0.08 wt % carbon nanotube sponges, which shows that it has better electrochemical performance. This is because sponges with high CNT fractions have more conducting channels and interparticle pores. Noteworthy is that when the CNT’s filler fraction is more than 0.11 wt %, agglomeration begins to occur, which will reduce the electrochemical performance of carbon nanotube sponge, meaning that more conducting channels and interparticle pores are formed within a certain CNT filling score. The specific capacitance values calculated from the CV curves are 7.2, 9.1, 10.3, 6.7, and 4.5 F/g, respectively, for carbon nanotube sponge containing 0.05, 0.08, 0.11, 0.14, and 0.17 wt % CNT.

Figure 4

Electrochemical characterization of carbon nanotube-based sponge composite supercapacitor electrodes containing 0.05, 0.08, 0.11, 0.14, and 0.17 wt % CNT. (a) CV curves obtained at 100 mV/s. (b) CV curves at various scan rates. (c) GCD curves of the 0.11 wt % carbon nanotube sponge composite electrode. (d) Nyquist diagram of carbon nanotube sponge composite electrode.

The performance of 0.11 wt % carbon nanotube sponge composite electrode was specifically investigated at 20–100 mV/s. In Figure b, the CV curves of carbon nanotube sponge electrodes maintain similar shapes at different scan rates, indicating remarkable electrochemical performance. The specific capacitances calculated from the CV curves are 20.1, 18.1, 14.5, 12.8, 11.2, and 10.3 F/g when the scan rates increase from 20 to 100 mV/s, respectively. With the decrease of scanning speed, the specific capacitance of carbon nanotube sponge composite electrode increases slightly. This intertesting rate capability is caused by the interconnected holes in the electrodes that affect the diffusion of electrolyte ions.

Figure c shows the GCD curves of the carbon nanotube sponge composite electrode tested at 0.5, 1, 3, and 5 A/g. These carbon nanotube sponges show specific capacitances at 19.3, 16.5, 12.1, and 10.3 F/g of different current densities from 0.5 to 5 A/g. We can find that the specific capacitances calculated from the CV and GCD curves are similar. The GCD curves of 0.11 wt % are almost linear and symmetrically mirrored to its discharge counterparts, which indicates a perfect electrochemical capacitive behavior. The resemblance of the specific capacitances obtained from the CV curve is increasing with a decrease in scanning rate; the specific capacitance obtained from GCD curve is also increasing with the decrease in current density.

Figure d shows the Nyquist curve of the carbon nanotube sponge composite electrode. The obtained Nyquist plots consist of a semicircle in the high-frequency region and a straight line in the low-frequency region. The slope of the low-frequency straight line represents the diffusion resistance and the contact interface capacitance of the electrolyte ions in the electrode, and the semicircle in the high-frequency region represents the double-layer capacitance and the charge transfer resistance. The diffusion resistance of graphene sponge electrode is 72.3 Ω.

Graphene-Based Sponge Composite Supercapacitor Electrodes

Figure a shows the CV curves of the graphene sponge composite supercapacitor electrodes at a scan rate of 100 mV/s within 0–0.8 V. For different filler fractions of graphene sponge composite electrodes, the CV curves are almost rectangular in shape, implying typical behavior of double-layer supercapacitor. In contrast, the CV curve of graphene sponge composite electrodes is closer to rectangle than that of carbon nanotube sponge, which means more typical electrode characteristics of supercapacitors, and it has a higher current density. This may be because the GnPs have a stronger ability to carry ions and electrons than CNT, and CNT is more likely to agglomerate and thus affect the electrochemical performance. Therefore, the mechanisms of the capacitance performance can be classified as electrochemical double-layer capacitances (EDLCs).[46]

Figure 5

Electrochemical characterization of graphene-based sponge composite supercapacitor electrodes containing 0.08, 0.16, 0.24, 0.32, and 0.40 wt % GnPs. (a) CV curves obtained at 100 mV/s. (b) CV curves at various scan rates. (c) GCD curves of the 0.24 wt % graphene sponge composite electrode. (d) Nyquist diagram of graphene sponge composite electrode.

The CV curve of the 0.24 wt % graphene sponge composite electrode has a maximum current density representing the best electrochemical performance. This is because sponges within a certain GnPs filling score have more conducting channels and interparticle pores. However, when the filler fraction of GnPs reaches a certain degree, agglomeration will occur. The stacking of GnPs will increase the internal resistance of graphene sponge and reduce the ability to conduct network to carry ions and electrons. EDLCs play an important role in capacitance enhancement for the graphene sponge.

The specific capacitance values calculated from CV curves are 11.2, 16.7, 20.1, 9.9, and 4.6 F/g, respectively, for graphene sponge composite electrode containing 0.08, 0.16, 0.24, 0.32, and 0.40 wt % GnPs. Through the specific calculation of capacitance of different filler fractions of graphene, it can be found that the filler fraction corresponding to the best capacitance performance of graphene sponge composite electrode is also 0.24 wt %. This is consistent with the conclusion of CV curves analysis. When the filler fraction is 0.24 wt %, the conductive channel of graphene sponge is perfect, and there is not a lot of agglomeration in GnPs, so the capacitance performance of graphene sponge is the best.

The performance of 0.24 wt % graphene sponge composite electrode was specifically investigated at 20–100 mV/s. In Figure b, the CV curves of graphene sponge composite electrodes maintain similar shapes at different scan rates, indicating remarkable electrochemical performance. The specific capacitances calculated from the CV curves are 26.7, 25.9, 24.3, 23.8, 21.4, and 20.1 F/g when the scan rates increase from 20 to 100 mV/s. With the decrease of scanning speed, the specific capacitance of graphene sponge composite electrode increases slightly. This intertesting rate capability is caused by the interconnected holes in the electrodes that affect the diffusion of electrolyte ions.

Figure c shows the GCD curves of the graphene sponge composite electrode tested at 0.5, 1, 3, and 5 A/g. These graphene sponge composite electrodes show specific capacitances at 25.1, 21.5, 11.4, and 10.3 F/g of different current densities from 0.5 to 5 A/g. We can find that specific capacitances calculated from CV and GCD curves are similar. The GCD curves of 0.24 wt % are almost linear and symmetrically mirrored to its discharge counterparts, which indicates a perfect electrochemical capacitive behavior. The resemblance of the specific capacitance obtained from the CV curve is increasing with a decrease in scanning rate, and the specific capacitance obtained from the GCD curve is also increasing with a decrease in current density. The increase of capacitance with a decrease in either scan rate or current density is caused by the time-dependent ion diffusion of the electrode materials. Figure d shows the Nyquist curve of the graphene sponge composite electrode. The diffusion resistance of the graphene sponge composite electrode is 51.1 Ω, lower than that of carbon nanotube sponge composite electrode, which is due to the lower resistance and reactance of the conductive network formed by GnPs and porous materials than CNTs.

Graphene Oxide-Based Composite Supercapacitor Electrodes

Graphene oxide has many chemical functional groups, which have a strong ability to carry ions. Therefore, the capacitive properties of graphene oxide-based composite supercapacitor electrodes were studied. Figure a shows the CV curves of the graphene oxide sponge composite electrodes at a scan rate of 100 mv/s within 0–0.8 V.

Figure 6

Electrochemical characterization of graphene oxide-based composite supercapacitor electrodes containing 0.02, 0.04, 0.06, 0.08, and 0.10 wt % GO. (a) CV curves obtained at 100 mv/s. (b) CV curves at various scan rates. (c) GCD curves of the 0.06 wt % graphene oxide sponge composite electrode. (d) Nyquist diagram of graphene oxide sponge composite electrode.

For different filler fractions of graphene oxide sponge composite electrodes, CV curves are almost rectangular in shape, implying typical supercapacitor behavior. Comparing the CV curves of three kinds of nanocarbon-based porous sponge composite supercapacitor electrodes, carbon nanotube sponge composite electrodes, graphene sponge composite electrodes, and graphene oxide sponge composite electrodes, it can be found that the CV curve of graphene oxide sponge is the closest to rectangle, and has the maximum current density, which represents the best electrochemical performance. This is because GO has a large number of functional groups containing oxygen and hydrogen, which are rarely found in GnPs and CNTs. The functional groups of graphene oxide mainly include epoxides, alcohols, ketone carbonyls, and carboxylic groups. These functional groups have strong ability to carry electrons and ions and have excellent electrochemical performance. The conductive network of graphene oxide sponge composite electrodes has more excellent electrochemical and capacitance properties.

The specific capacitance values calculated from CV curves are 21.8, 26.3, 31.2, 23.7, and 18.2 F/g, respectively, for graphene oxide sponge composite electrodes containing 0.02, 0.04, 0.06, 0.08, and 0.10 wt % GO. Through the specific calculation of the capacitance value of different filler fractions of graphene oxide, it can be found that the filler fraction corresponding to the best capacitance performance of graphene oxide sponge composite electrode is also 0.06 wt %. This is consistent with the conclusion of CV curves analysis. When the filler fraction is 0.06 wt %, the conductive channel of graphene sponge is perfect, and there is not much agglomeration in GO, so the capacitance performance of graphene oxide sponge composite electrode is the best.

In Figure b, the capacitance of 0.06 wt % graphene oxide sponge composite electrode was measured at 20–100 mV/s. Compared with those of graphene sponge composite electrode and carbon nanotube sponge composite electrode, the CV curves of graphene oxide sponge composite electrode are more similar at different scanning speeds, showing the most significant electrochemical performance. The specific capacitances calculated from the CV curves are 78.1, 57.4, 48.5, 40.3, 34.8, and 31.2 F/g when the scan rates increase from 20 to 100 mV/s, respectively.

Figure c contains the GCD curves of the graphene oxide sponge composite electrode tested at 0.5, 1, 3, and 5 A/g. These graphene sponge composite electrodes show specific capacitances of 68.5, 50.1, 31.3, and 27.5 F/g at different current densities of 0.5–5 A/g. We can find that specific capacitance calculated from CV and GCD curves are similar. The GCD curves of 0.06 wt % are almost linear and symmetrically mirrored to its discharge counterparts, which indicates a perfect electrochemical capacitive behavior. Figure d shows the Nyquist curve of the graphene oxide sponge electrode. The diffusion resistance of graphene oxide sponge electrode is 35.2 Ω. Among the three kinds of nanocarbon-based fillers, graphene oxide has the lowest diffusion resistance and the smallest charge transfer impedance, which is conducive to the performance of its capacitance. In general, among the three kinds of nanocarbon-based sponges, graphene oxide sponge composite electrodes have the highest specific capacitance, the best electrode performance of supercapacitor, and the best electrochemical impedance performance.

Cycle Stability

The cycle stabilities of the carbon nanotube sponge composite electrode, the graphene sponge composite electrode, and the graphene oxide sponge composite electrode were tested. In the capacitance cycling stability test, the packing composition of carbon nanotube sponge composite electrode is 0.11 wt %, that of graphene sponge composite electrode is 0.24 wt %, and that of graphene oxide sponge composite electrode is 0.06 wt %. The capacitance decay was examined between 0 to 0.8 V at 0.5 A/g for 20 000 cycles. Figure shows that the capacitance retention rate of carbon nanotube sponge composite electrode is 94.12%, that of graphene sponge composite electrode is 95.56%, and that of graphene oxide sponge composite electrode is 92.99%. Specific capacitance gradually decreases and remains relatively stable after the initial 20 000 cycles. This can be attributed to porous structure evolved itself through the first few cycles to obtain a stable capacitance for the following cycles. The results of cyclic tests show that the sponge samples of graphene have the highest stability, which may be due to the high structural integrity of graphene. However, the test results show that the cycle stability of carbon nanotube sponge composite electrode and graphene oxide sponge composite electrode resembles that of graphene sponge composite electrode, and the capacity retention rate of the three sponge is more than 90%. Therefore, the three kinds of nanocarbon-based sponge composite electrodes all have the potential to be used as electrodes of various energy storage devices.

Figure 7

Cycle stability of carbon nanotube sponge composite electrode, graphene sponge composite electrode, and graphene oxide sponge composite electrode.

In Figure , the specific capacitance values of three kinds of nanocarbon-based porous sponge composite supercapacitor electrodes under different components are compared. The maximum specific capacitance values of carbon nanotube sponge composite electrodes, graphene sponge composite electrodes, and graphene oxide sponge composite electrodes under different packing components are 20.1, 26.7, and 78.1 F/g, respectively. The graphene oxide composite electrode has abundant functional groups with a strong ability to carry electrons and ions, so the peak specific capacitance of graphene oxide sponge composite electrode is the largest of the three nanocarbon-based sponge composite electrodes. The packing components corresponding to the peak specific capacitance of the three nanocarbon-based sponge composite electrodes are 0.11, 0.24, and 0.06 wt %, respectively. The packing components corresponding to the peak specific capacitance of carbon nanotube sponge composite electrodes and graphene oxide sponge composite electrodes are relatively small, which means that the agglomeration of the two nanocarbon-based fillers occurs under the relatively small packing components.

Figure 8

Specific capacitance curve. (a) Curve of specific capacitance of carbon nanotube sponge composite electrodes under different filler fractions. (b) Curve of specific capacitance of graphene sponge composite electrodes under different filler fractions. (c) Curve of specific capacitance of graphene oxide sponge composite electrodes under different filler fractions.

Flexible Nanocarbon-Based Porous Sponge Composite Supercapacitor Electrodes

Due to the unique structure of the nanocarbon-based porous sponge composite supercapacitor electrodes, the supercapacitor can make 180° bending pairs, and it can restore its original shape after removing the load. Regardless of the kind of nanocarbon-based sponge composite electrodes, the specific capacitance will decrease under bending load. This is mainly due to the influence of the bending structure and the deformation of the material on the uniformity of the conductive channel and the ion channel of the nanocarbon-based sponge composite electrodes. In Figure a–d, the capacitance retention of graphene sponge composite electrode under 180° bending is 77.63%, that of carbon nanotube sponge composite electrode is 72.52%, and that of graphene oxide composite electrode is 67.46%. The largest capacity retention of graphene sponge composite electrode under bending condition is due to the higher structural regularity of graphene single-layer carbon atomic structure.

Figure 9

Nanocarbon-based porous sponge composite supercapacitor electrodes under bending load: (a) bending angles, (b) carbon nanotube sponge flexible supercapacitor electrodes, (c) graphene sponge flexible supercapacitor electrodes, (d) graphene oxide sponge flexible supercapacitor electrodes, and (e) series supercapacitor based on graphene oxide sponge electrode.

A single supercapacitor is often insufficient to power practical devices such as a light-emitting diode (LED), and a common solution is to adopt a tandem cell that connects several supercapacitor units either in series or parallel depending on the applications. Herein, our graphene oxide supercapacitor units were connected in series using carbon cloth as current collectors. It extended the operation potential window from 0.8 V for a single unit to 2.4 V for a three-unit tandem cell. The three-unit supercapacitor tandem cell can deliver sufficient energy to light up a red LED (Figure e) when it was charged at 2.4 V for 10 s.

Mechanical Property

The nanocarbon-based sponge porous composite electrodes are also expected to be used in flexible sensors, so we have studied the mechanical properties and tensile linearity of nanocarbon-based sponge composite. Tensile strength and elongation at break of sponges with different filler fractions are tested. Figure a shows that the tensile strength of the sponge nanocomposites varies with CNT filling fraction from 0 to 0.17 wt %. When the filling ratio is 0.11 wt %, the maximum tensile strength of graphene sponge composite is 0.267 MPa, which is 52.1% higher than that of pure sponge. Compared with GnPs, CNT has a stronger toughening effect on nanocomposites and a stronger stress transfer ability under load. However, the maximum tensile strength at 0.17 wt % of the filler component is 16.5% lower than that at 0.11 wt % of the filler component. This may be because CNT agglomerates with the increase of filler composition, which leads to the decrease of mechanical properties of the sponge nanocomposites, thus reducing the maximum tensile strength of CNT sponge composite. The elongation at break of the sponge nanocomposite decreases with the increase of the filler composition, and the filling rate of graphene increases. When the filling ratio of graphene is 0.17 wt %, the elongation at break of sponge nanocomposites decreases by 97.3%. This is because CNT nanopacking has a stronger toughening effect than GnPs. To summarize, CNT filler can improve the mechanical properties of sponge nanocomposites better than GnPs.

Figure 10

(a) Tensile strength and breaking elongation of carbon nanotube sponge with different filler fractions. (b) Tensile strength and breaking elongation of graphene sponge with different filler fractions. (c) Tensile strength and breaking elongation of graphene oxide sponge with different filler fractions.

Figure b demonstrates the changes in the tensile strength of sponge nanocomposites with an increase of GnP filler fractions ranging from 0 to 0.40 wt %. The maximum tensile strength of graphene sponge has reached 0.236 MPa at 0.24 wt % filler fraction, depicting an increase by 45% compared with that of pure sponge. This increase is due to the ability of the GnPs to provide more specific surface area and interfacial structure that can effectively prevent stress concentrations and facilitate stress transfer across the interface under loading.[47−49] However, the maximum tensile strength at 0.40 wt % filler fraction is 22.03% lower than that at 0.24 wt % of filler fraction. This may be due to the aggregation of graphene platelets with the increase of filler fraction, which leads to the decrease of the structural and mechanical properties of the attached sponge skeleton, thus reducing the maximum tensile strength of graphene sponge composite. The elongation at break of sponge nanocomposites decreases with the increase of graphene filler fraction. When the graphene filler fraction is 0.40 wt %, the elongation at break of the sponge nanocomposites decreases by about 79.9%. This is attributed to the enhancement effect of graphene on the sponge matrix, which experienced hardening and toughening effects during the accumulation of nanofillers. In conclusion, GnP fillers increase the tensile strength and decrease the elongation of sponge nanocomposites.

Figure c shows that the tensile strength of the sponge nanocomposites varies with the filling ratio of graphene oxide, from 0 to 0.10 wt %. When the filling ratio is 0.06 wt %, the maximum tensile strength of graphene sponge is 0.187 MPa, which is 31.6% higher than that of pure sponge. Compared with GnPs and CNT, the toughening effect of GO on nanocomposite porous materials is weaker, as well as the stress transfer ability under load. However, the maximum tensile strength at 0.10 wt % of the filler component is 23.5% lower than that at 0.06 wt % of the filler component. This may also be due to the aggregation of GO with the increase of filler composition, resulting in the decrease of mechanical properties of the sponge nanocomposites, thus reducing the maximum tensile strength of the carbon nanotube sponge. With the increase of filler composition, the elongation at break of sponge nanocomposites decreases and the filling rate of GO increases. When the filling ratio of GO was 0.10 wt %, the elongation at break of sponge nanocomposites decreased by 59.8%. This is because the oxygen- and hydrogen-containing functional groups in GO may weaken the toughening effect of GO on sponge porous composites. The toughening effect of the GO nanofiller is weaker than that of GnPs and CNT. In conclusion, the improvement of the mechanical properties of sponge nanocomposites by GO fillers is weaker than that by GnPs and CNT.

Tensile Linearity

Signal linearity is an important parameter to measure when determining the sensitivity of materials. The cross-sectional area of nanocarbon-based sponge composite decreases during tension, but due to the special high elastic skeleton structure of sponge, the distance between conductive paths decreases and the paths become denser in the process of cross-sectional shrinkage. These factors lead to a decrease in the resistance of nanocarbon-based sponge composite during stretching.[50]

The linearity of carbon nanotube sponge composite with a filler fraction of 0.11 wt % is shown in Figure a; the strain increases from 0 to 13.52% and the rate of change in relative resistance decreases uniformly with the increase of strain, which indicates that the material has a good linear response. The sensitivity coefficient (slope) is 0.0169. Based on the analysis, it can be highlighted that conductivity of the nanocomposite sponge is inversely related to the cross-sectional area of the specimen. Figure b shows the linearity of the graphene sponge composite filled with 0.24 wt % filling fraction; the strain increases from 0 to 21.19%, and the change rate of relative resistance decreases uniformly with the increase of strain. The sensitivity coefficient (slope) is 0.0127. The reason why the linearity of graphene sponge composite is lower than that of carbon nanotube sponge composite is that the order of graphene is weaker than that of carbon nanotube sponge, and the conductive network formed by porous materials is less sensitive to tensile load. Figure c shows the linearity of graphene oxide sponge composite filled with a 0.0625 wt % component, the strain increases from 0 to 4.38%, and the change rate of relative resistance decreases uniformly with the increase of strain. The sensitivity coefficient (slope) is 0.0409. The linear degree of graphene oxide sponge composite is lower than the other two kinds of nanocarbon-based sponge composites, which may be due to the fact that the mechanical and electrical properties of graphene oxide sponge composite is weaker than those of carbon nanotube sponge composite and graphene sponge composite, so the sensitivity of the conductive network formed under tensile conditions is relatively low.

Figure 11

(a) Tensile linearity of carbon nanotube sponge composite with a filler fraction of 0.11 wt %. (b) Tensile linearity of graphene sponge composite with a filler fraction of 0.24 wt %. (c) Tensile linearity of graphene oxide sponge composite with a filler fraction of 0.06 wt %. (d) Strain fatigue test of carbon nanotube sponge composite with a filler fraction of 0.11 wt %. (e) Strain fatigue test of graphene sponge composite with a filler fraction of 0.24 wt %. (f) Strain fatigue test of graphene oxide sponge composite with a filler fraction of 0.06 wt %.

Strain Fatigue Property

It is critical to demonstrate the durability and stability of the composites by monitoring their performance under repeated stretch loading–unloading tests.[51,52] Three kinds of nanocarbon-based sponge composites were continuously stretched and released for 10 000 cycles at the frequency of 3.33 Hz under the deformation of 0.5 mm, and the corresponding resistance change rate (ΔR/R0) was recorded, as shown in Figure d–f. As shown in Figure d, the carbon nanotube-based sponge composite shows excellent electrical response stability after cyclic tensile test. Although the value of ΔR/R0 fluctuates slightly, the maximum ΔR/R0 is basically stable at −0.1. Figure e shows that after cyclic tensile test, the maximum ΔR/R0 of the graphene-based sponge composite fluctuates around −0.05, and the overall ΔR/R0 value fluctuates more than that of the carbon nanotube-based sponge composite, indicating that the electrical response stability of the graphene-based sponge composite under cyclic tensile load is weaker than that of the carbon nanotube-based sponge composite. This may be due to the better mechanical properties of carbon nanotubes and the better support and shape retention of conductive network of nanocomposites under cyclic loading. Figure f shows that after cyclic tensile test, the maximum ΔR/R0 value of graphene oxide-based sponge composite fluctuates around −0.025 and the overall ΔR/R0 value fluctuation is the largest among the three kinds of nanocarbon-based porous sponge composites, indicating that the electrical response stability of the graphene oxide-based sponge composite is weak under cyclic tensile load. This may be due to the relatively weak electrical and mechanical properties of graphene oxide.

Electrical Response to Bending and Torsion

The electrical response of flexible sensors under bending and torsion load is an important parameter to test the performance of a flexible sensor. The electrical responses of three kinds of nanocarbon-based sponge composites under bending and torsion were compared. Figure shows the resistance changes of three kinds of nanocarbon-based sponge composites at different bending and torsion angles. Under 180° bending and torsion loading, the maximum resistance change rates of carbon nanotube sponge composite are 40.6 and 87.2%, respectively. Under 180° bending and torsion load, the maximum resistance change rates of graphene sponge composite are 25.2 and 60.9%, respectively. Under 180° bending and torsion load, the maximum resistance change rates of graphene oxide sponge composite are 8.1 and 22.8%, respectively. The results of electrical response of bending and torsion load are consistent whether clockwise or counterclockwise. It can be found that the carbon nanotube sponge composite has the strongest sensing ability for bending and torsion load changes, and flexible sensing performance is the best. This is due to the strong conductivity of CNT and the good support for the conductive structure of the porous framework of sponge composites. This characteristic provides the possibility for the application of carbon nanotube sponge composites in sensors and material health detection, especially when the sensor is faced with complex loadings such as bending and torsion.

Figure 12

(a) Resistance variations with different bending angles and (b) resistance variations with different twist angles.

Sound Absorption

The nanocarbon-based sponge porous composite electrode in this paper is a typical porous material with a large specific surface area and high porosity. When the sound wave passes through the nanocarbon-based sponge porous composite, it will diffuse reflection and scattering in its dense porous structure, and then part of the sound wave will be absorbed by it. Therefore, the nanocarbon-based sponge porous composite materials can not only be used as flexible supercapacitor electrode materials and sensor materials but also as porous sound-absorbing materials. This characteristic also expands the multifunction of nanocarbon-based sponge composite electrodes.[53−55] In this paper, the sound absorption properties of three kinds of nanocarbon-based sponge composites were studied. Figure a–e shows the schematic diagrams of three types of experimental devices: common darkroom, pure sponge darkroom, and nanocarbon-based sponge composite material-lying darkroom. The sound absorption effect is compared by recording the audio from the music player in three types of containers. The research methods in Figure a–e are as follows: the anechoic chamber, the pure sponge darkroom, the graphene oxide sponge composite darkroom, the carbon nanotube sponge composite darkroom, and the graphene sponge composite darkroom, respectively, to compare and study the sound absorption effect of three kinds of nanocarbon-based sponge composites (Figure ).

Figure 13

(a–e) Audio in different darkrooms.

Figure 14

Fabrication of graphene oxide sponge nanocomposites.

The experimental results show that the absorption rate of a pure sponge is 18%, that of graphene oxide sponge composite is 41%, that of carbon nanotube sponge composite is 59%, and that of graphene sponge composite is 68%. Among the three kinds of nanocarbon matrix composites, graphene sponge composite has the best sound absorption effect. This may be because the regular two-dimensional nanostructure of graphene greatly improves the diffuse reflection effect of the sponge skeleton attached to the graphene on the acoustic wave (reflected wave in multidirectional scattering), thus optimizing the sound absorption effect of the sponge nanocomposites. Therefore, graphene sponge nanocomposites are expected to be used in vehicles, aircraft, submarine, and other fields to reduce noise.

Conclusions

To summarize, three kinds of porous sponge nanocarbon (CNT, GnPs, GO) composite supercapacitor electrodes were prepared by a simple dip method. The effects of three kinds of nanocarbon-based fillers on the capacitance of sponge porous composite flexible electrode materials were compared. The results show that the peak specific capacitance of carbon nanotube sponge composite electrode is 20.1 F/g, the diffusion resistance is 72.3 Ω, and the capacity retention rate is 94.12% under 20 000 charge–discharge cycles. The peak specific capacitance of the graphene sponge composite electrode is 26.7 F/g, the diffusion resistance is 51.1 Ω, and the capacity retention rate is 95.56% under 20000 charge–discharge cycles. The peak specific capacitance of graphene oxide sponge composite electrode is 78.1 F/g, the diffusion resistance is 35.2 Ω, and the capacity retention rate is 92.99% under 20 000 charge–discharge cycles. In general, graphene oxide-based sponge composite supercapacitor electrode has the highest specific capacitance peak value and the best overall capacitance performance. Under the bending load of 180°, the capacitance retention rate of graphene oxide sponge composite supercapacitor electrodes is 67.46%, which indicates that the prepared supercapacitor has good flexibility. The three kinds of nanocarbon-based sponge composite electrode also have the potential to be used as flexible sensors. The tensile strength of carbon nanotube sponge composite electrodes is 0.267 MPa, and the tensile linearity is 0.0169. Under 180° bending and torsion loading, the maximum resistance change rates of carbon nanotube sponge composite are 40.6 and 87.2%, respectively. The carbon nanotube sponge composite has the best flexible sensing performance and electrical response stability at cyclic load. Among the three kinds of nanocarbon matrix composites, graphene sponge composite has the best sound absorption effect and the absorption rate of graphene sponge composite is 68%. In the whole manufacturing process of nanocarbon-based sponge composites, expensive materials and complex equipment are not used. It is believed that the prepared graphene oxide-based supercapacitor sponge composite flexible electrodes have broad application prospects in aerospace, automobile, energy sector, and other fields.

Experimental Materials and Methods

Experimental Materials and Instruments

Asbury Carbons (Asbury, NJ) kindly supplied graphite intercalation compound (GIC, 1721). Carbon nanotube was purchased from Deko Gold (Beijing, China). The sponge made of melamine with a density of 15–17 kg/m3 was provided by Xijie Co (Hubei, China). The porosity was 88.725% ± 0.53%. KOH was purchased from Tianjin Hengxing Chemical Reagent Manufacturing Co., Ltd. (Tianjin, China). Carbon cloth is provided by Hessen Electric Co., Ltd. (Shanghai, China). All chemicals were used as received without any further purification.

Scanning electron microscopy (SEM) images of the graphene sponge were obtained by SU8010 (Japan). Tensile test measurement was performed using a uniaxial tensile loading machine LTD GX-SF001 (Guangdong, China). Fluke 2638A Hydra Series III Data Acquisition System (Everett) constantly monitored the resistance of the graphene sponge under different conditions. The capacitive properties of nanocarbon-based supercapacitor were analyzed using an electrochemical workstation ChenHua CHI660E B19038 (Shanghai, China). The charge–discharge test of nanocarbon-based supercapacitor was analyzed by a Blue Test System CT3001A 1U (Hubei, China).

Experimental Methods

A piece of sponge (rectangular cuboids of dimensions 2.0 cm × 1.0 cm × 0.5 cm) was used as the matrix of the composite material. Figure shows the preparation method of 0.06 wt % graphene oxide sponge nanocomposites electrodes. GO (0.05 g) was added to deionized water to form a colloidal solution, which was then ultrasonicated for 5 h, resulting in the formation of a homogeneous GO solution. Afterward, the sponge was soaked into the solution and simultaneously ultrasonicated for 15 h for the GO to permeate completely into the sponge skeleton. The graphene oxide sponge was dried in an oven at 100 °C for 10 h. When testing the mechanical properties, flexible sensing properties, and sound absorption properties of carbon-based nanocomposites, the size of the sponge used in the nanocomposites was 6.0 cm × 2.0 cm × 0.25 cm. As the volume of the sponge increased 3 times, the mass of carbon nanomaterials (carbon nanotubes, graphene, graphene oxide) and dispersant (acetone, deionized water) increased 3 times accordingly. The results show that the mass fraction is consistent with that of 2.0 cm × 1.0 cm × 0.5 cm carbon-based sponge composite electrode.

All electrochemical measurements of individual electrodes were conducted in 1 M KOH aqueous solution using a standard three-electrode system with a platinum plate counter electrode and a Ag/AgCl reference electrode. While two-electrode configuration matches more closely the performance of a commercially packaged cell, three-electrode configuration would be more appropriate to accurately evaluate the capacitive performance for a novel material.[56] Cyclic voltammetry (CV), galvanostatic charge–discharge (GCD) and electrochemical impedance spectroscopy (EIS) measurements were performed with an electrochemical workstation (CHI660E). For the calculation method of specific capacitance, see the Support Information Page S2.

The flexible supercapacitor experiment and LED lighting experiment need to assemble a complete supercapacitor. The fabrication of solid electrolyte was started at 85 °C by dissolving 2 g of polyvinyl alcohol in 100 mL of 1 M KOH with stirring. Then, the two sponge electrodes (each 2 cm × 1 cm × 0.5 cm) were immersed in the solid electrolyte, followed by pressing. After drying at room temperature, two electrodes were assembled to a supercapacitor. A piece of filter paper is used as a separator, and two pieces of carbon cloth are used as collectors to make two such electrodes to form a sandwich structure. In this way, a double-electrode system was formed and a complete supercapacitor was assembled.