XPS was used to determine the changes in the surface charge states. The full spectra of MnO 2 /graphene/Ni composites are shown in a. Clearly, graphene/Ni foam contained two elements (C and O). After electrochemical deposition, three element signals (C, O, and Mn) could be observed, which indicates that MnO 2 was successfully deposited on the graphene/Ni foam. The peak at 654.1 eV was attributed to Mn 2p. The peaks at 285.7 eV (C-O), 284.8 eV (C-C sp 3 ), and 284.2 eV (C=C sp 2 ) were similar to those in the literature [ 28 , 29 , 30 ]. The coexistence bonds of C-C (sp 3 ) and C=C (sp 2 ) were due to the appearance of larger graphite fragments during the process of graphite stripping. The high-resolution Mn(2p) of the MnO 2 /graphene/Ni foam composite was amplified for fitting analysis at different current densities. The peaks at 654.1 eV and 653.3 eV were attributed to Mn 2p 1/2 , while the peaks at 642.7 eV and 641.7 eV were related to Mn 2p 3/2 , as seen in b,c. The Mn 2p peak was fitted into two peaks (2p 1/2 and 2p 2/3 ) at 654.1 eV and 642.7 eV, respectively, indicating that Mn was represented in the chemical state of Mn 4+ . The peak values were 653.3 eV and 641.7 eV ( c), implying that they were not only tetravalent but also bivalent and trivalent manganous oxide. The energy separation was about 11.4 eV, which was related to α- MnO 2 [ 31 ], suggesting that MnO 2 was successfully formed. To obtain pure MnO 2 , the optimum current density was fixed at 1 mA·cm −2 in the following experiment. d,e show high-resolution O(1s) spectra of MnO 2 /graphene/Ni composites at different current densities. The binding energy value was 529.9 eV and 531.3 eV ( d), which was attributed to Mn-O-Mn and Mn-OH, respectively. The Mn-O-Mn and Mn-OH presented at 530 eV and 531.4 eV, and are shown in e, indicating that the structure of manganese was complex. It was found that manganese dioxide had a different composition according to the constant current density. When the current density increased, the signal peak of Mn-OH increased to decrease the content of manganese oxide.
The crystal structures of the prepared samples were characterized by XRD, as shown in a. The obvious characteristic peaks at 44.5°, 51.7°, and 76.5° were attributed to the (111), (200), and (220) crystal planes of the Ni foams. The peak at 26.4° belonged to the diffraction peak of the carbon material, which corresponded to the (002) crystal plane of the graphene. Two weak peaks at 37.5° and 65.6° were attributed to the (211) and (002) diffraction peaks of α-MnO 2 (JCPDS Card NO.44-0141) [ 25 ]. The results could be further confirmed by the Raman spectra of the MnO 2 /graphene/Ni composite, as shown in b. The D band is a common feature for sp 3 defects or disorder in carbon, and the G band provides useful information on in-plane vibration of sp 2 -bonded carbon atoms in a 2-D hexagonal lattice. The D band (~1345 cm −1 ) and G band (~1578 cm −1 ) were assigned to the characteristic bands of graphene. The G band represents the in-plane vibration of the sp 2 carbon atoms. The D band is related to the breathing mode of κ-point phonons of the A1g symmetry [ 26 ]. The relative intensity ratio of the D-band to G-band, defined as R = I D /I G , represents the structural regularity of graphene. The R-value was proposed to be an indication of disorder in graphene. The smaller the R value, the greater the regularity is. As seen from b, the R-value is approximately 1.15, demonstrating an irregular structure in graphene compare with our previous study [ 27 ]. A broad (observed at ~630 cm −1 ) peak could be attributed to the symmetric stretching vibrations of Mn–O.
c–f shows SEM images of MnO 2 /graphene/Ni composites with different constant currents for 10 min. The diameter of the MnO 2 nanoparticles ranged from 10 nm to 15 nm, with a constant current of 1 mA cm −2 . The diameter of MnO 2 nanoparticles ranged from 10 nm to 20 nm, with a constant current of 5 mA cm −2 . The diameter of MnO 2 nanoparticles was about 30 nm, with a constant current of 10 mA cm −2 . The diameter of MnO 2 nanoparticles was about 50 nm, with a constant current of 15 mA cm −2 . These results indicated that the MnO 2 was homogeneously dispersed and uniformly deposited on the skeleton of the graphene/Ni composite. Actually, the thickness of MnO 2 varied in different places of the sample. It could be that only scale 200 μm was flat. The diameters of MnO 2 nanoparticles increased with the increase in the constant current. g presents an SEM image of MnO 2 /graphene/Ni composite with a current of 1 mA·cm −2 for a 30 min deposition. When the deposition time was longer, the particle size increased continuously, which greatly reduced the surface area of the particles, leading to a decrease in the contact area of the electrolyte. C, Mn, O, Au, and Ni were observed in the EDX spectrum (based on the SEM image of g) of the MnO 2 /graphene/Ni composite (prepared at 1 mA cm −2 , 30 min deposition), which were attributed to graphene, MnO 2 , Au, and Ni. However, Au derived from the Au-coating of the sample for SEM image measurement, as shown in h. Furthermore, the Mn and C in the electrode dispersed well in the material ( i,j,k), indicating that MnO 2 has good dispersibility in the MnO 2 /graphene/Ni electrode. TEM was also utilized to identify the images of MnO 2 and MnO 2 /graphene. It could be clearly observed that the prepared MnO 2 structure was loosely arranged. The wrinkle-like morphology for graphene could be observed. Considering the MnO 2 /graphene was prepared by the deposition process, the graphene could restack, resulting in excessive surface agglomeration ( Supplementary Materials, Figure S3 ).
SEM was used to study the morphologies of the prepared samples. a shows an SEM image of the nickel foams. The insert image shows that such a smooth substrate not only supplies the platform for the deposition of graphene and MnO 2 , but also facilitates the fast channel of electrolytes to the electrode. It could be observed that the pore distribution of nickel foams was approximately between 100 μm and 300 μm, in which the porous structure of the nickel foams had a higher surface area. Therefore, the fabricated electrode and electrolyte had a greater contact area to allow for a higher super capacitance. The SEM image of the graphene/Ni foam composite with 20,000× magnification is shown in b. It could be observed that graphene prepared by electrolytic-stripping had a sheet nanostructure. However, the coated graphene/Ni was obtained in an oven at 100 °C for 1 h to adhere the graphene onto the nickel foam, thus the surface exhibited densely packed but porous morphology. It was indicated that the number of layers of graphene can be revealed by the Raman shift of the 2D band at approximate 2700 cm −1 of the Raman spectra. The 2D band for an extremely few layer of grsaphene exhibits sharper and a more symmetrical shape [ 24 ]. For the graphene/Ni in the present study, the 2D band at 2700 cm −1 was not sharp enough, indicating that the number of graphene layers obtained was relatively high ( Supplementary Materials, Figure S2 ).
Due to ultrasonic cavitation, ultrasonic-assisted impregnation generated special physical states, such as a higher temperature and higher pressure locally than without the apparatus, which can increase graphene loading and improve the dispersity, thus enhancing the activity. The deposition technology followed the steps to grow nucleation sites and then ions reacted to produce the deposition material on the substrate. In this case, MnO 2 obtained by oxidation of Mn 2+ from manganese acetate was adsorbed onto graphene coated on the Ni foam. The MnO 2 /graphene/Ni composite was fabricated using a facile method, as shown in .
In the study, Ni foam with high conductivity was used as the current collector. The CV characteristic curve was narrow and small, indicating a very small specific capacitance (Figure S4). a–d show the CV curves of MnO2/graphene/Ni electrodes at different current densities. As the scanning rate increased from 10 mV s−1 to 100 mV s−1, the shapes of the CV curves changed slighty for all the samples, implying they possessed excellent electrochemical reversibility and stability. The specific capacitances of MnO2/graphene/Ni foam prepared at different scanning rates under different current densities are presented in . Each experimental data point was measured several times and calculated using a statistical analysis method; the mean and standard error of the mean were obtained; and the mean was taken from three significant figures.
It was clearly observed that the specific capacitance of the MnO2/graphene/Ni electrode decreased as the current density increased. The highest specific capacitance came from the lowest current density. At lower scan rates, the diffusion of Na+ from the electrolyte could transport ions to the interior and interface between MnO2 and graphene, leading to a completed reaction and process. At a high scan rate, Na+ is pushed onto the outer surface layer of the electrode, where effective interaction between ions and electrodes is reduced. Therefore, the specific capacitance would be decreased. The specific capacitance of the electrodes can be calculated according to Formula (1).
From Formula (1), ν was lower and Cs was higher. By increasing ν, the current became larger and the ions in the electrolyte did not have enough time to diffuse into the electrode materials. The capacitance performance of MnO2/Ni and MnO2/graphene/Ni electrodes were investigated using circle voltammetry at a scan rate of 100 mV s−1 for different current densities using the electrodeposition method, as seen in e–h. Since the specific capacitance was directly proportional to the area of the CV curves, the results showed that the CV area of the MnO2/graphene/Ni foam electrodes was much larger than that of the MnO2/Ni foam electrode. Moreover, the electrodeposition current density was high, causing the deposited mass of MnO2 to be large. However, the CV characteristic showed that MnO2/graphene/Ni did not change much, demonstrating that they were almost quasi-rectangular shapes. It was obvious that graphene could improve the conductivity of MnO2. In addition, due to the increase of MnO2 mass deposition on MnO2/Ni by increasing with the current density of the electrodeposition, the CV characteristic deviated from the rectangular shape ( g,h). As seen in e–h, due to the fact that the area of the CV curve of graphene was much less than that of MnO2, in this work, only the mass of MnO2 was used as the active material mass.
The choice of aqueous electrolyte was based on the size of the hydrated cations and anions, and on the mobility of ions [32]. In addition, the corrosiveness of the electrolyte with regard to the electrode must be considered. The Na2SO4 solution was approximately a neutral electrolyte, which is friendly to the environment and was utilized as the electrolyte in this study. The MnO2/graphene/Ni electrodes showed a quasi-rectangular shape in the CV curves, indicating the capacitance characteristics of the MnO2 deposited onto the Ni foam electrode [33]. The CV curve for the MnO2/graphene/Ni electrode in the Na2SO4 electrolyte was unlike that expected from an EDLC; the CV characteristic curve for an EDLC shows a nearly ideal rectangle [34].
In this article, the capacitances are compared with other calculation methods, which depends on the mass ratio of graphene and the MnO2 of the composite electrode. It was calculated and the specific capacitance (Cs) is listed based on Cs=∫|i|dV2mνΔV ; the total loaded mass (MnO2+graphene, M) as active materials was calculated by Cs′=∫|i|dVMνΔV to obtain the specific capacitance (Cs′) and areal capacitance CsA′ based on Cs′, which were also calculated for the comparison (Supplementary Materials, Table S2). In this study, Ni foam was used as the substrate material and this material has a porous structure. The pores have a significant influence on the capacitance, thus it is not suitable to discuss the volume capacitance.
The GCD test for MnO2/graphene/Ni electrodes obtained from different electrodeposition current densities were examined at 1 A g−1. The longer discharge time of the MnO2/graphene/Ni electrode prepared at an electrodeposition current density of 5 mA cm−2 exhibited a capacitance higher than that of the other electrodes, which is consistent with the results obtained from the CV characterizations. It could be seen that the MnO2/graphene/Ni electrode exhibited highly linear and almost symmetrical triangular curves, revealing that the IR potential drop for MnO2/graphene/Ni is less noticeable. However, the GCD curve was slightly bent at the end of the discharge curve for the pseudo-capacitor material. This was mainly caused by the redox reaction inside the material. Thus, the graphene composite MnO2 electrode has good electrochemical performance (Supplementary Materials, Figure S5).
To study the electrochemical mechanism for the MnO2 electrode materials obtained from various electrodeposition conditions, they were subjected to EIS analysis, as shown in a. EIS measurements were taken in a frequency range from 100 kHz to 0.01 Hz.
The results are displayed using Nyquist plots, which were divided into three different regions, as follows.
In the high frequency region, the intercept at the real axis (Z0) represents the equivalent series resistance (ESR), including the ionic resistance of the electrolyte, the inherent resistance of the substrate, and the contact resistance of the active material/current collector interface [35]. The span of the semicircular arc in the mid–high frequency region represents the charge transfer resistance (Rct) at the electrode/electrolyte interface [36]. The slope in the low frequency region was attributed to semi-infinite Warburg impedance, which played an important role in the frequency dependence of diffusion/transport for electrolyte ions in the holes of the electrode [37]. If the impedance graph increases sharply and tends to become a vertical line, the characteristic of a pure capacitance behavior is indicated. The impedance values of the electrodes are summarized in . In the table, each experimental data point was measured several times and calculated using a statistical method; the mean and standard error of the mean were obtained; and the mean was taken with two significant figures.
At high frequencies, the intercepts (RE) for the curve and real axis for the MnO2/graphene/Ni electrode (prepared at 5 mA cm−2 current density of the electrodeposition), the MnO2/graphene/Ni electrode (prepared at 15 mA cm−2 current density of the electrodeposition), the MnO2/Ni electrode (prepared at 5 mA cm−2 current density of the electrodeposition), and the MnO2/Ni electrode (prepared at 15 mA cm−2 current density of the electrodeposition) were determined to be 2.8 Ω, 3.1 Ω, 3.0 Ω, and 3.1 Ω, respectively. In particular, the MnO2/graphene/Ni electrode obtained from the electrodeposition at a current density of 5 mA cm−2 showed the smallest equivalent resistance, demonstrating that the electrode had better conductivity. Moreover, the as-obtained MnO2/graphene/Ni electrode (5 mA cm−2) showed the smallest semicircular arc-shaped impedance (0.15 Ω) in the high–medium frequency region, indicating that the charge transfer resistance (Rct) for the electrode was extremely low and that the ion diffusion path was very short.
In the low-frequency region, the MnO2/graphene/Ni electrode showed a straight line with a steeper slope, indicating that the capacitance performance was very close to that of an ideal supercapacitor [38]. Additionally, in this region, the slope of the impedance curve for the electrode composited without graphene was not as steep (less steep) as that for the electrode with a graphene buffer layer. In addition, the RN at 2.8 Ω and 3.1 Ω for MnO2/graphene/Ni was smaller than those of 3.4 Ω and 3.9 Ω for MnO2/Ni electrodes ( ). The sum of the impedance values of the three regions for each electrode prepared using different electrodeposition methods as described above is also listed in . The results also showed that Rt at 5.8 Ω and 6.6 Ω for MnO2/graphene/Ni electrodes was smaller than those of 7.0 Ω and 8.5 Ω for the MnO2/Ni electrodes. These results may be due to the relatively better dispersion of using graphene in the MnO2/graphene/Ni electrode.
In this study, the MnO2/graphene/Ni electrode obtained using a 5 mA cm−2 current density electrodeposition exhibited an extremely low impedance, which was attributed to the better homogeneity and nanostructure of the composites grown on the nickel foam. In addition, the MnO2/graphene/Ni (5 mA cm−2) electrode exhibited an equivalent series resistance (RE) that was lower than that of the other electrodes. This result further showed that the MnO2/graphene/Ni (5 mA cm−2) electrode had faster kinetics compared to the other three electrodes, which is beneficial in improving the capacitance performance of the composite material, especially at high charge/discharge rates for the supercapacitor [39].
The MnO2/graphene sample had a clear hysteresis loop, showing highly interconnected holes, and the material had an open-wide structure. The BET specific surface areas for the MnO2 and MnO2/graphene materials were 158.5 m2·g−1 and 179.2 m2·g−1, respectively, and the pore sizes were 3.7 nm and 7.8 nm, respectively. MnO2/graphene showed the largest BET specific surface area, which is attributed to the loosely arranged structure (Supplementary Materials, Figure S6 and Table S3). The electrode possesses a better pore structure and allows the Na2SO4 electrolyte to be easily adsorbed on the electrode, which facilitates migration and diffusion, resulting in a larger CV curve area and a higher specific capacitance value.
b shows the cycling characteristics of the two electrodes under a current density of 5 A g−1. It is clear that the specific capacitance of the MnO2/Ni electrode was maintained at 84%, while the capacitance retention rate of the MnO2/graphene/Ni electrode increased to 90% after 5000 cycles. It is indicated that the introduction of a graphene layer can improve the poor conductivity of MnO2 and increase the reversibility of both absorbing and desorbing electrons. The graphene layer not only improves the whole capacitance but also increases the retention rate of the capacitance. Another reason for this may be that a good adhesion between graphene and MnO2, as well as between graphene and the Ni substrate, is presented. Here, graphene is considered as a buffer layer. c shows the coulombic efficiency of the MnO2/graphene/Ni electrode. The coulombic efficiency is the ratio of the charge passing through an electrode to the reaction charge on the electrode. From the figure, it can be observed that the coulombic efficiency remained at 96% after 5000 cycles during the charge and discharge process. The charge transfer efficiency of the electrode and the decrease in the stored charge of the battery due to the aging effect during the actual charge and discharge process can thus be known.
The total current (i(v)) of the CV measurement at scan rates for the composite material consisted of two parts. One part concerned the current associated with the EDLC at the electrolyte interface or the initial fast Faraday reaction on the exposed electrode surface (icap). The other part concerned the current associated with the slow diffusion control process (idif). The capacitance contribution and diffusion control contribution can be calculated according to Formula (6) [40].
i(v)=icap+idif=a×νb
(6)
where ν is the scan rate, i(v) is the total current of the CV measurement, and a and b are variable parameters. The b value can be estimated from the slope of the log(i(v)) and log(ν). shows the b parameter of the MnO2 /Ni and MnO2/graphene/Ni electrodes obtained from different electrodeposition conditions.
It can be observed that the MnO2/graphene/Ni electrodes indicated that 0.8 < b < 1 electrode was considered to be a pseudocapacitive material with a main capacitance storage, while the MnO2/Ni electrodes showed “0.5 < b < 0.8” as exhibiting the main Faraday (battery type) behavior [40,41]. b = 1 indicates EDLC, as well as the slope in the Nyquist plot 90°, while b = 0.5 indicates battery-type, as well as the slope in the Nyquist plot 45°. The “b parameter” of the four different electrodes were all greater than 0.5, which obviously reveal that the final element will not be a hybrid cell.