Binder-Free MnO2/MWCNT/Al Electrodes for Supercapacitors

Recently, significant progress has been made in the performance of supercapacitors through the development of composite electrodes that combine various charge storage mechanisms. A new method for preparing composite binder-free MnO2/MWCNT/Al electrodes for supercapacitors is proposed. The method is based on the original technique of direct growth of layers of multi-walled carbon nanotubes (MWCNTs) on aluminum foil by the catalytic pyrolysis of ethanol vapor. Binder-free MnO2/MWCNT/Al electrodes for electrochemical supercapacitors were obtained by simply treating MWCNT/Al samples with an aqueous solution of KMnO4 under mild conditions. The optimal conditions for the preparation of MnO2/MWCNT/Al electrodes were found. The treatment of MWCNT/Al samples in a 1% KMnO4 aqueous solution for 40 min increased the specific capacitance of the active material of the samples by a factor of 3, up to 100–120 F/g. At the same time, excellent adhesion and electrical contact of the working material to the aluminum substrate were maintained. The properties of the MnO2/MWCNT/Al samples were studied by electron probe microanalysis (EPMA), Raman spectroscopy, cyclic voltammetry (CV), and impedance spectroscopy. Excellent charge/discharge characteristics of composite electrodes were demonstrated. The obtained MnO2/MWCNT/Al electrodes maintained excellent stability to multiple charge-discharge cycles. After 60,000 CVs, the capacitance loss was less than 20%. Thus, this work opens up new possibilities for using the MWCNT/Al material obtained by direct deposition of carbon nanotubes on aluminum foil for the fabrication of composite binder-free electrodes of supercapacitors.


Introduction
The development and use of environmentally friendly renewable energy sources is becoming an extremely urgent task all over the world. Over the past decade, there has been tremendous progress in the development of solar energy, hydropower, wind power, tidal energy, and other renewable energy sources [1,2], as well as in technologies of energy harvesting [3]. However, in most cases, clean, renewable energy from these sources cannot be applied directly due to the instability of its generation. Therefore, reliable electrochemical energy storage devices (ECS), including fuel cells, ion batteries, and supercapacitors, are necessary for efficient storage, conversion, and further use of the energy of these sources [4]. Among them, electrochemical capacitors, also known as supercapacitors (SC), are considered a new generation of green energy storage [5]. They have a simple design and are made from cheap, widely used materials, have high power density, fast charging, and longer service life than traditional batteries [6]. Due to this, the areas of application of supercapacitors are extremely wide, from hybrid electric vehicles to miniature current sources in smart things technologies [7][8][9].
In terms of charge storage mechanisms, energy storage/conversion processes for supercapacitors occur in two different ways, which can be generally divided into electrochemical double-layer capacitors (EDLC) and pseudocapacitors [10][11][12]. EDLCs store energy electrostatically by surface adsorption/desorption of ions at the electrode/electrolyte carbon fiber cloth [48], stainless steel mesh [36,37], nickel mesh [24], nickel foam [27], and aluminum foil [49] served as the conductive substrates for such electrodes. It should be noted that aluminum is considered the best metal for a current collector in SC due to its low cost, low specific gravity, excellent electrical conductivity, and high plasticity [50][51][52]. However, there are practically no works on binder-free MnO 2 /CNT/Al electrodes for supercapacitors in the literature. Recently, we have demonstrated a new method for the deposition of MWCNT layers on aluminum foil, which provides excellent adhesion of MWCNTs to the substrate [53]. The resulting MWCNT/Al material showed good performance as SC electrodes. In the present work, we studied the possibility of preparing binder-free MnO 2 /MWCNT/Al composite electrodes using the developed method of MWCNTs grown on aluminum foil.

Materials and Methods
The aluminum substrate (99.0%, Mikhailovsk, Russia) was a foil 50 µm thick. Strips (0.75 × 3 cm 2 ) were preliminarily cleaned with isopropyl alcohol (99.8%, EKOS-1, Staraya Kupavna, Russia) and distilled water. Then they were sonicated for 10 min in a suspension of 45 mg of abrasive powder (28 µm, corundum) in 40 mL of a 20 wt.% Ni(NO 3 ) 2 (99.0%, EKOS-1, Staraya Kupavna, Russia) solution. After that, they were washed with distilled water and placed in a 20 wt.% Ni(NO 3 ) 2 aqueous solution for 20 h for further mild oxidation. Finally, nickel nitrate residues were removed by washing the samples three times with deionized water. The substrates thus prepared after drying in air were used to deposit MWCNTs. The general features of the deposition of MWCNTs on aluminum foil by the catalytic pyrolysis of ethanol vapor are described in a previous article [53]. The masses of the deposited MWCNTs and MnO 2 /MWCNT composite were calculated from the difference in the masses of the substrates before and after synthesis.
To prepare MnO 2 /MWCNT/Al composites, the MWCNT/Al samples were placed in an aqueous solution of KMnO 4 (high purity grade, Russia) for a certain time at room temperature. Then, the samples were washed several times with deionized water and dried in air for 24 h. In a series of experiments, KMnO 4 solutions with concentrations of 0.2, 1, and 2 wt.% were used. The duration of treatment varied from 10 to 120 min.
Electrochemical oxidation of MWCNT/Al samples was carried out in a 0.05 M aqueous sodium sulfate solution for 10 min in a two-electrode cell at a potential of 4 V. A counter electrode was a platinum wire.
A P-40X potentiostat-impedance meter (Electrochemical Instruments, Chernogolovka, Russia) was used for electrochemical measurements. Cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) of the initial and modified samples were carried out in a three-electrode cell. The electrolyte was 0.5 M Na 2 SO 4 aqueous solution. A saturated calomel electrode (SCE) and platinum wire were used as reference and counter electrodes. Electrochemical impedance experiments were conducted in the presence of 0.5 M aqueous Na 2 SO 4 solution at a DC potential of 0 V, superimposed by an AC potential of 20 mV peak-to-peak amplitude over a frequency range of 50 kHz to 10 mHz. Scanning electron microscopy (SEM) and the chemical composition of the samples were studied using a JSM 6490 scanning electron microscope (Jeol, Tokyo, Japan) equipped with an INCA Oxford Instruments Electron probe microanalysis (EPMA) system. Raman spectra were recorded using a Sentera Raman microscope (Bruker, Berlin, Germany) under excitation with a solid-state laser with a wavelength of 532 nm.

Binder-Free MnO 2 /MWCNT/Al Electrodes from As-Prepared MWCNT/Al Samples
As is known, aluminum does not possess the catalytic properties necessary for growing CNTs on its surface by the CVD method. Previously, we developed an original technique for making catalytic activity on an aluminum surface [53]. To deposit MWCNT layers on aluminum foil, we used the catalytic pyrolysis of ethanol vapor. This process does not require complex equipment and proceeds at a relatively low temperature, which is important for an aluminum substrate (Tmp = 660 • C). The features of growing MWCNT layers on aluminum foil are described in detail in a previous article [53]. MWCNT layers grown by this procedure have excellent adhesion to an aluminum substrate and can be used directly as SC electrodes [53]. As noted in the introduction, MWCNTs can serve as the basis for the preparation of MnO 2 /MWCNT composite materials, which have a significantly higher specific electrochemical capacitance compared to the capacitance of MWCNTs due to the pseudocapacitance of MnO 2 . An analysis of the literature shows that the most common way to obtain MnO 2 /MWCNT composites is the direct interaction of carbon nanotubes with a solution of potassium permanganate. The heterogeneous character of the reaction ensures the deposition of the formed MnO 2 directly on the surface of the nanotubes. According to these reasons, in the present work, we have chosen this simple method to obtain the MnO 2 /MWCNT/Al composite material. An additional argument was the fact that highly defective CNTs quite easily react with KMnO 4 at room temperature [36]. In previous works, we have shown that the MWCNTs obtained by our method are highly defective [53,54]. The SEM image of the MWCNT/Al composite ( Figure 1a) showed a dense fibrous layer on the aluminum substrate. Numerous bends testified to the high defectiveness of MWCNTs. The thickness of the MWCNT layer is about 10 µm. Transmission electron microscopy of MWCNTs ( Figure 1b) showed many basal planes (0001) of sp 2 carbon ("turbostratic structure"). They are visible on the TEM image of the tube ( Figure 1b) as a series of parallel lines with an average interval of 0.36 nm. It can be seen that the sp 2 carbon planes are directed at an angle to the nanotube axis. Thus, the edges of the planes come to the surface of the nanotube, creating a large number of defects [53].
MWCNT layers on aluminum foil are described in detail in a previous article [53]. MWCNT layers grown by this procedure have excellent adhesion to an aluminum substrate and can be used directly as SC electrodes [53]. As noted in the introduction, MWCNTs can serve as the basis for the preparation of MnO2/MWCNT composite materials, which have a significantly higher specific electrochemical capacitance compared to the capacitance of MWCNTs due to the pseudocapacitance of MnO2. An analysis of the literature shows that the most common way to obtain MnO2/MWCNT composites is the direct interaction of carbon nanotubes with a solution of potassium permanganate. The heterogeneous character of the reaction ensures the deposition of the formed MnO2 directly on the surface of the nanotubes. According to these reasons, in the present work, we have chosen this simple method to obtain the MnO2/MWCNT/Al composite material. An additional argument was the fact that highly defective CNTs quite easily react with KMnO4 at room temperature [36]. In previous works, we have shown that the MWCNTs obtained by our method are highly defective [53,54]. The SEM image of the MWCNT/Al composite ( Figure 1a) showed a dense fibrous layer on the aluminum substrate. Numerous bends testified to the high defectiveness of MWCNTs. The thickness of the MWCNT layer is about 10 µm. Transmission electron microscopy of MWCNTs ( Figure  1b) showed many basal planes (0001) of sp 2 carbon ("turbostratic structure"). They are visible on the TEM image of the tube ( Figure 1b) as a series of parallel lines with an average interval of 0.36 nm. It can be seen that the sp 2 carbon planes are directed at an angle to the nanotube axis. Thus, the edges of the planes come to the surface of the nanotube, creating a large number of defects [53]. To select the optimal conditions for the preparation of the composite, a preliminary series of experiments was carried out with KMnO4 solutions of various concentrations and various processing times. No pretreatment of the MWCNT/Al samples was performed. Next, changes in the elemental composition of the active layer of the MnO2/MWCNT/Al samples were compared with changes in specific capacitance and weight gain. Elemental analysis of the surface layer of all treated samples showed the To select the optimal conditions for the preparation of the composite, a preliminary series of experiments was carried out with KMnO 4 solutions of various concentrations and various processing times. No pretreatment of the MWCNT/Al samples was performed. Next, changes in the elemental composition of the active layer of the MnO 2 /MWCNT/Al samples were compared with changes in specific capacitance and weight gain. Elemental analysis of the surface layer of all treated samples showed the presence of carbon, oxygen, and manganese as the main components. Data on their concentrations are given in Table 1. As can be seen, the selected mild processing conditions are quite sufficient for the effective interaction of MWCNTs and permanganate. The data obtained by the EPMA method, which is local, should be considered as semi-quantitative. Despite some dispersion of experimental results, the following conclusions can be drawn. With an increase in permanganate concentration, the rate and degree of MWCNT oxidation increase. For 0.2% and 1% solutions, the main increase in the manganese concentration in the active layer occurred during the first 30-40 min. Further treatment in a 0.2% solution even led to a slight decrease in the concentration of manganese, and in a 1% solution, an insignificant gradual increase was observed. In both cases, the oxygen concentration gradually increased. When processed in a 2% solution, an increase in the concentration of manganese and oxygen occurred throughout the entire time. During prolonged treatment (more than 90 min), the layer of MnO 2 /MWCNT was destroyed, and the surface of the aluminum substrate was partly exposed.
Raman spectroscopy confirmed the EPMA data. Figure 2 shows the Raman spectra of the MnO 2 /MWCNT/Al samples obtained after treatment of the initial MWCNT/Al in a KMnO 4 solution of various concentrations for 90 min. Bands at 1347 and 1604 cm −1 belong to MWCNTs (D and G peaks, respectively) [14], and the bands at 505, 580, and 643 cm −1 correspond to vibrations of the Mn-O bond [36]. As can be seen from Figure 2, with an increase in the KMnO 4 concentration, the relative height of the peaks corresponding to manganese increased. After prolonged treatment in a 2% KMnO 4 solution, their intensity became higher than the intensity of the MWCNTs peaks.
Cyclic voltammetry showed a significant increase in the capacitance of MWCNT/Al samples after treatment in a KMnO 4 solution. Figure 3 shows an example CV of samples before and after treatment of various durations. The quasi-rectangular shape of the CV curves demonstrates the good capacitance characteristics of the composite electrodes.
The specific capacitance of MWCNTs and the MnO 2 /MWCNT hybrid material on the electrode was measured according to the generally accepted procedure [51]. The cell capacitance was calculated by Equation (1): where C cell is the capacitance of the cell, IdV is the area under the CV characteristic at I > 0 (V · A) (this is half of the area inside the complete loop in Figure 3), ∆V is the voltage range (V), and ν is the voltage scanning rate (V/s). The capacitance of the working electrode in a three-electrode cell is equal to the capacitance of the cell. Therefore, the specific capacitance (C spm ) of active material (MWCNTs or MWCNT/MnO 2 composite) is C cell /m, where m is the mass of active material on the electrode. As quantitative characteristics of composite electrodes, we used the values of capacitance per total mass of the working material (C spm ) and capacitance per electrode surface area (C sps ). Cyclic voltammetry showed a significant increase in the capacitance of samples after treatment in a KMnO4 solution. Figure 3 shows an example C before and after treatment of various durations. The quasi-rectangular sha curves demonstrates the good capacitance characteristics of the composite e  Cyclic voltammetry showed a significant increase in the capacitance samples after treatment in a KMnO4 solution. Figure 3 shows an example before and after treatment of various durations. The quasi-rectangular s curves demonstrates the good capacitance characteristics of the composite The specific capacitance of MWCNTs and the MnO2/MWCNT hyb the electrode was measured according to the generally accepted procedu As expected, the composition of the active material and its specific capacitance after treatment of the MWCNT/Al samples in a KMnO 4 solution depended on both the concentration and duration of treatment. Data on the evolution in the composition of the MnO 2 /MWCNT material, the increase in the mass of the active material, and the increase in specific capacitance depending on the treatment time are collected in Figure 4. range (V), and ν is the voltage scanning rate (V/s).
The capacitance of the working electrode in a three-electrode cell is equal to the capacitance of the cell. Therefore, the specific capacitance (Cspm) of active material (MWCNTs or MWCNT/MnO2 composite) is Ccell/m, where m is the mass of active material on the electrode. As quantitative characteristics of composite electrodes, we used the values of capacitance per total mass of the working material (Cspm) and capacitance per electrode surface area (Csps).
As expected, the composition of the active material and its specific capacitance after treatment of the MWCNT/Al samples in a KMnO4 solution depended on both the concentration and duration of treatment. Data on the evolution in the composition of the MnO2/MWCNT material, the increase in the mass of the active material, and the increase in specific capacitance depending on the treatment time are collected in Figure 4.  It is believed that the reaction between CNTs and permanganate proceeds according to Equation (2) [24]: According to this equation, carbon is oxidized to the maximum possible state and removed from the CNT surface. However, it is well known that the strong oxidizing agents (hydrogen peroxide, concentrated nitric and sulfuric acids) react with CNTs to form oxygen-containing functional groups (-C=O, -COOH, etc.) on their surface [14]. Why can a similar process not occur when CNTs are treated with permanganate? Thus, the MnO 2 deposition may be accompanied by functionalization of the CNT surface. In particular, Chen et al. used a permanganate solution to functionalize CNTs [55]. On the other hand, the oxidation of CNTs also leads to a significant increase in their specific capacitance. This effect is usually associated with the pseudocapacitance of functional groups and an increase in the specific surface area of CNTs due to etching [14]. Thus, an increase in the specific capacitance of MWCNTs after treatment in a KMnO 4 solution can be associated with both the deposition of MnO 2 and the oxidation of the MWCNTs surface. Since it is difficult to isolate the contribution of each of the factors, we used the capacitance related to the total mass of MnO 2 /MWCNT material to characterize the samples.
The data in Figure 4 show that, at all permanganate concentrations, the treatment of MWCNT/Al samples in a KMnO 4 solution led to an increase in the specific capacitance of the samples by several times. The use of 0.2% KMnO 4 solution gave the least increase in capacitance. For a 2% solution, the increase in capacitance was the largest. However, the value of the specific capacitance of the active material is an important characteristic of the SC electrode, but not the only one. For reliable operation of the MnO 2 /MWCNT/Al binder-free electrode, it is necessary that after its preparation, the aluminum substrate and the MWCNT layer retain the integrity and good adhesion, which ensures good electrical contact between MnO 2 and the current collector. From this point of view, a 2% solution is less suitable. Considering the above, we have chosen treatment in a 1% KMnO 4 solution for 40 min as the optimal conditions for MnO 2 deposition. These conditions were used to prepare MnO 2 /MWCNT/Al samples for further studies. According to the literature, the synthesis of MnO 2 on CNTs by immersing them in KMnO 4 solutions has already been published. However, in this work, we had to solve problems related to the activity of MWCNTs synthesized from alcohol vapors. In addition, it was not known how the aluminum substrate would behave in the presence of a strong oxidizing agent. As a result of the experiments performed, the optimal processing conditions were chosen, which make it possible to keep the Al substrate and the MWCNT layer intact and, at the same time, to provide a significant improvement in the characteristics of the electrodes.
In classical supercapacitors, the electrical charge accumulates in an electrical double layer near the electrode's surface. The process of formation of such a layer is very fast; therefore, SCs are characterized by high charge-discharge rates. The mechanism of charge accumulation in pseudocapacitors is associated with the occurrence of a reversible electrochemical redox reaction [56]. For MnO 2 , Equation (3) describes this mechanism as follows [19]: C + (H + , Li + , Na + , K + ) is an electrolyte cation that is embedded in MnO 2 on the electrode surface. The process involves a reversible redox reaction between Mn 3+ and Mn 4+ [33,[41][42][43]. Due to this process, a larger charge can accumulate on the electrode than in a conventional SC. However, this process is slower than the formation of an electrical double layer. In addition, MnO 2 has a low electrical conductivity, and its deposition can lead to an increase in the ohmic resistance of the electrode, which can also affect the rate of charge/discharge [19,22]. To evaluate the rate of the charge/discharge performance of the obtained binder-free MnO 2 /MWCNT/Al electrodes, a CV study of the dependence of capacitance on the scanning rate was carried out. Figure 5 shows the dependence of the specific capacitance of the active material of the initial MWCNT/Al electrode and the same sample treated in 1% KMnO 4 solution for 40 min (MnO 2 /MWCNT/Al).
The C spm obtained for the MnO 2 /MWCNT/Al electrode was 99.4 F/g at a scan rate of 2 mV/s, which is three times higher than the C spm obtained for the original MWCNT/Al electrode (33.3 F/g). As the scan rate increased to 100 mV/s, the corresponding capacitance retentions were 88.6% for the original MWCNT/Al electrode and 74.7% for the MnO 2 /MWCNT/Al electrode. Although the retention of the MnO 2 /MWCNT/Al electrode was slightly lower, this result shows an excellent rate capability. As noted above, the mass of the active material of the MnO 2 /MWCNT/Al electrode was about 60% higher than that of the original MWCNT/Al. Accordingly, the observed increase in capacitance per electrode surface area (C sps ) was also greater. The C sps of the MnO 2 /MWCNT/Al electrode was 22.3 mF/cm 2 at a scan rate of 2 mV/s, which was five times higher than the C sps of the original MWCNT/Al electrode (4.4 mF/cm 2 ). capacitance on the scanning rate was carried out. Figure 5 shows the dependen specific capacitance of the active material of the initial MWCNT/Al electrode same sample treated in 1% KMnO4 solution for 40 min (MnO2/MWCNT/Al). The Cspm obtained for the MnO2/MWCNT/Al electrode was 99.4 F/g at a scan mV/s, which is three times higher than the Cspm obtained for the original MW electrode (33.3 F/g). As the scan rate increased to 100 mV/s, the corresponding tance retentions were 88.6% for the original MWCNT/Al electrode and 74.7% MnO2/MWCNT/Al electrode. Although the retention of the MnO2/MWCNT/Al e was slightly lower, this result shows an excellent rate capability. As noted ab mass of the active material of the MnO2/MWCNT/Al electrode was about 60% than that of the original MWCNT/Al. Accordingly, the observed increase in cap per electrode surface area (Csps) was also greater. The Csps of the MnO2/MWCNT trode was 22.3 mF/cm 2 at a scan rate of 2 mV/s, which was five times higher than of the original MWCNT/Al electrode (4.4 mF/cm 2 ).
Further study of the electrochemical characteristics of MnO2/MWCNT/Al el was carried out using electrochemical impedance spectroscopy. Figure 6a  Further study of the electrochemical characteristics of MnO 2 /MWCNT/Al electrodes was carried out using electrochemical impedance spectroscopy. Figure 6a,b show the Nyquist plots of the original MWCNT/Al sample and the MnO 2 /MWCNT/Al sample obtained by its treatment in 1% KMnO 4 solution for 40 min. Both Nyquist plots are almost linear at low frequencies, indicating good capacitance characteristics (Figure 6a). On the Nyquist plot, the ohmic resistance of a cell is estimated by the offset Z' along the x-axis in the high-frequency region. Accordingly, the ohmic resistance was 4.5 Ω for the original MWCNT/Al sample and 4.6 Ω for the MnO 2 /MWCNT/Al sample (Figure 6b). These results indicate that the deposition of MnO 2 did not lead to an increase in the ohmic resistance of the cell. Therefore, the MnO 2 in the MnO 2 /MWCNT/Al composite electrode had good electrical contact with the conductive substrate. At the same time, the slope of the Nyquist plot for the MnO 2 /MWCNT/Al sample in the low-frequency region decreased compared to the initial MWCNT/Al sample (Figure 6a), which can be explained by the pseudocapacitance of MnO 2 .
From the data of impedance spectroscopy, it was also possible to calculate the frequency dependences of the real and imaginary capacitances (C'(ω) and C"(ω), respectively), which provide valuable information about the characteristics of the SC electrodes. The following Equation (4) was used for these purposes [26]: where ω is the angular frequency defined as ω = 2πf and |Z(ω)| is the impedance modulus. From the data of impedance spectroscopy, it was also possible to calculate the frequency dependences of the real and imaginary capacitances (C'(ω) and C"(ω), respectively), which provide valuable information about the characteristics of the SC electrodes. The following Equation (4) was used for these purposes [26]: where ω is the angular frequency defined as ω = 2πf and |Z(ω)| is the impedance modulus.
As can be seen from Figure 6c, at a frequency of 0.1 Hz, the real capacitance of the MnO2/MWCNT/Al electrode was almost 4.5 times higher than the capacitance of the original MWCNT/Al electrode. This result is in good agreement with the data of cyclic voltammetry (in this case, one should compare the capacitance per electrode area, Csps). The rate of reversible charge/discharge performance can be estimated from the evolutions of the imaginary capacitance of the MWCNT/Al and MnO2/MWCNT/Al electrodes versus the frequency (Figure 6d). The dependences had peaks corresponding to the determined relaxation frequency (fr) and the response time constant τr, defined as 1/fr. The As can be seen from Figure 6c, at a frequency of 0.1 Hz, the real capacitance of the MnO 2 /MWCNT/Al electrode was almost 4.5 times higher than the capacitance of the original MWCNT/Al electrode. This result is in good agreement with the data of cyclic voltammetry (in this case, one should compare the capacitance per electrode area, C sps ). The rate of reversible charge/discharge performance can be estimated from the evolutions of the imaginary capacitance of the MWCNT/Al and MnO 2 /MWCNT/Al electrodes versus the frequency (Figure 6d). The dependences had peaks corresponding to the determined relaxation frequency (fr) and the response time constant τr, defined as 1/fr. The τr is a quantitative parameter to evaluate the fast reversible charge/discharge performance [26]. The MWCNT/Al electrode had a τr of 0.42 s, which showed fast ion transfer capability and excellent rapid charge/discharge performance. The pseudocapacitance restricts the charge/ion rapid transport and leads to a large response time. Therefore, the MnO 2 /MWCNT/Al electrode had a greater value of τr of 3.0 s. However, as applied to pseudocapacitors, this value indicates a very good charge/discharge performance of the obtained composite binder-free electrodes [26].

Binder-Free MnO 2 /MWCNT/Al Electrodes from Oxidized MWCNT/Al Samples
In many works on the preparation of composite MnO 2 /CNT materials for SC electrodes, the surface of initial CNTs was preliminarily oxidized (functionalized). In this work, we also compared the results of the interaction of non-oxidized and oxidized MWCNTs with permanganate solution. As was found earlier, the electrochemical oxidation of MWCNT/Al samples leads to an increase in the concentration of functional oxygen-containing groups on the surface of MWCNTs and an increase in specific capacitance by 4-5 times in the range from -800 to 10 mV and by 1.5-2 times in the range from −10 to 800 mV [54]. At the same time, the electrochemical oxidation of MWCNT/Al somewhat reduces their rate of reversible charge/discharge performance. In the present work, the effect the pre-oxidation has on the properties of MnO 2 /MWCNT/Al binder-free electrodes obtained by treating MWCNT/Al with a KMnO 4 solution was studied. The initial MWCNT/Al samples were oxidized in a two-electrode cell at 4 V for 10 min in a 0.05 M Na 2 SO 4 solution. After that, the specific capacitance of MWCNT in the range of −10 to 800 mV increased, on average, by a factor of 1. capability and excellent rapid charge/discharge performance. The pseudocapacitance restricts the charge/ion rapid transport and leads to a large response time. Therefore, the MnO2/MWCNT/Al electrode had a greater value of τr of 3.0 s. However, as applied to pseudocapacitors, this value indicates a very good charge/discharge performance of the obtained composite binder-free electrodes [26].

Binder-Free MnO2/MWCNT/Al Electrodes from Oxidized MWCNT/Al Samples
In many works on the preparation of composite MnO2/CNT materials for SC electrodes, the surface of initial CNTs was preliminarily oxidized (functionalized). In this work, we also compared the results of the interaction of non-oxidized and oxidized MWCNTs with permanganate solution. As was found earlier, the electrochemical oxidation of MWCNT/Al samples leads to an increase in the concentration of functional oxygen-containing groups on the surface of MWCNTs and an increase in specific capacitance by 4-5 times in the range from -800 to 10 mV and by 1.5-2 times in the range from −10 to 800 mV [54]. At the same time, the electrochemical oxidation of MWCNT/Al somewhat reduces their rate of reversible charge/discharge performance. In the present work, the effect the pre-oxidation has on the properties of MnO2/MWCNT/Al binder-free electrodes obtained by treating MWCNT/Al with a KMnO4 solution was studied. The initial MWCNT/Al samples were oxidized in a two-electrode cell at 4 V for 10 min in a 0.05 M Na2SO4 solution. After that, the specific capacitance of MWCNT in the range of −10 to 800 mV increased, on average, by a factor of 1.5. Thus, the starting characteristics of the oxidized MWCNT/Al samples differed significantly from the characteristics of the as-prepared samples. Next, the samples were treated in 1% KMnO4 solution for various times. Data on the atomic concentrations of Mn and O, weight increase, and increase in the specific capacitance for the MnO2/MWCNT/Al samples thus obtained are shown in Figure 7. The comparison of these data with data for the MnO2/MWCNT/Al samples obtained from as-prepared MWCNT/Al samples (Figure 4e,f,t) shows a greater increase in the concentration of Mn and O in the active layer (Figure 7a), as well as a greater weight gain in the working material (Figure 7b). At the same time, there was no significant improvement in specific capacitance (Cspm) and (Csps) (Figure 7c, curves 1 and 2, respectively; curves 1' and 2' show the growth relative to the oxidized sample).
CVs of the MnO2/MWCNT/Al samples prepared from pre-oxidized MWCNT/Al samples showed a similar result. Figure 8a shows the evolution of the CV curves for the same sample: initial (curve 1), electrochemically oxidized (curve 2), and final The comparison of these data with data for the MnO 2 /MWCNT/Al samples obtained from as-prepared MWCNT/Al samples (Figure 4e,f,t) shows a greater increase in the concentration of Mn and O in the active layer (Figure 7a), as well as a greater weight gain in the working material (Figure 7b). At the same time, there was no significant improvement in specific capacitance (C spm ) and (C sps ) (Figure 7c, curves 1 and 2, respectively; curves 1' and 2' show the growth relative to the oxidized sample).
CVs of the MnO 2 /MWCNT/Al samples prepared from pre-oxidized MWCNT/Al samples showed a similar result. Figure 8a shows the evolution of the CV curves for the same sample: initial (curve 1), electrochemically oxidized (curve 2), and final MnO 2 /MWCNT/Al obtained by treating the oxidized MWCNT/Al sample in 1% KMnO 4 solution for 40 min (curve 3). It can be seen that the shape and area of the CV loop changed after oxidation. After treatment in a KMnO 4 solution, a significant increase in the area inside the CV curve was observed; this indicates a significant increase in the electrode capacitance.
MnO2/MWCNT/Al obtained by treating the oxidized MWCNT/Al sample in 1% KMnO4 solution for 40 min (curve 3). It can be seen that the shape and area of the CV loop changed after oxidation. After treatment in a KMnO4 solution, a significant increase in the area inside the CV curve was observed; this indicates a significant increase in the electrode capacitance. Changes in specific capacitance as a function of scanning rate for the sample after different stages of processing are shown in Figure 8b. At a rate of 2 mV/s, the specific capacitance of the initial sample was 39 F/g; after oxidation, it increased to 64 F/g; after subsequent treatment in a KMnO4 solution, it reached 118 F/g. Thus, the specific capacitance of the active material in the final MnO2/MWCNT/Al sample increased by a factor of 3 compared to the initial MWCNT/Al sample. With an increase in the scanning rate to 100 mV/s, the specific capacitances of the sample were 36 F/g for the original MWCNT/Al sample, 47 F/g for the oxidized sample, and 90 F/g for the final MnO2/MWCNT/Al sample. The capacitance retentions were 92.3%, 73.4%, and 76.3% respectively. These results are close to those obtained by treating as-prepared MWCNT/Al samples with a permanganate solution (see Section 3.1).
The results of impedance spectroscopy of these samples are shown in Figure 9. At low frequencies, Nyquist plots of initial, oxidized, and final MnO2/MWCNT/Al samples were almost linear, indicating good capacitance characteristics (Figure 9a). In the high-frequency region, the Nyquist plots were shifted towards lower Z' values for the oxidized and final sample compared to the original (Figure 9b). This indicates a decrease in the ohmic resistance of the electrode after electrochemical oxidation from 4.7 to 3.7 Ω.  Changes in specific capacitance as a function of scanning rate for the sample after different stages of processing are shown in Figure 8b. At a rate of 2 mV/s, the specific capacitance of the initial sample was 39 F/g; after oxidation, it increased to 64 F/g; after subsequent treatment in a KMnO 4 solution, it reached 118 F/g. Thus, the specific capacitance of the active material in the final MnO 2 /MWCNT/Al sample increased by a factor of 3 compared to the initial MWCNT/Al sample. With an increase in the scanning rate to 100 mV/s, the specific capacitances of the sample were 36 F/g for the original MWCNT/Al sample, 47 F/g for the oxidized sample, and 90 F/g for the final MnO 2 /MWCNT/Al sample. The capacitance retentions were 92.3%, 73.4%, and 76.3% respectively. These results are close to those obtained by treating as-prepared MWCNT/Al samples with a permanganate solution (see Section 3.1).
The results of impedance spectroscopy of these samples are shown in Figure 9. At low frequencies, Nyquist plots of initial, oxidized, and final MnO 2 /MWCNT/Al samples were almost linear, indicating good capacitance characteristics (Figure 9a). In the high-frequency region, the Nyquist plots were shifted towards lower Z' values for the oxidized and final sample compared to the original (Figure 9b). This indicates a decrease in the ohmic resistance of the electrode after electrochemical oxidation from 4.7 to 3.7 Ω. Further deposition of MnO 2 on the MWCNTs surface did not lead to an increase in the ohmic resistance. The frequency dependence of the real capacitance confirms an increase in the electrode capacitance by about 1.5 times after electrochemical oxidation and by about 4 times after MnO 2 deposition (Figure 9c). The dependence curves of the imaginary capacitance have maxima, the position of which characterizes the charge-discharge rate. The value of τr for the initial MWCNT/Al electrode was 0.5 s, for the oxidized sample, 0.9 s, and for the final MnO 2 /MWCNT/Al sample, 2 s. Thus, the MnO 2 /MWCNT/Al electrode prepared using intermediate electrochemical oxidation has slightly better charge/discharge performance. However, in general, the characteristics of MnO 2 /MWCNT/Al electrodes obtained from the as-prepared MWCNT/Al samples do not differ much from the characteristics of samples obtained with intermediate electrochemical oxidation. Thus, the intermediate stage of electrochemical oxidation of MWCNT/Al samples does not significantly improve the characteristics of binder-free MnO2/MWCNT/Al electrodes. In addition, the study of the resistance of MnO2/MWCNT/Al electrodes to multiple charge-discharge cycles showed significant advantages of electrodes obtained from the as-prepared MWCNT/Al samples. As Figure 10 shows, the MnO2/MWCNT/Al electrode prepared from the as-prepared MWCNT/Al sample retained approximately 80% of the specific capacitance after 60,000 charge-discharge cycles. At the same time, the specific capacitance of the electrode obtained with the intermediate stage of electrochemical oxidation of MWCNT/Al dropped to 60% of the initial one after 16,000 cycles. Perhaps this is due to the fact that electrochemical oxidation results in partial etching of the surface of MWCNTs, which makes them less stable. In addition, the study of the resistance of MnO 2 /MWCNT/Al electrodes to multiple chargedischarge cycles showed significant advantages of electrodes obtained from the as-prepared MWCNT/Al samples. As Figure 10 shows, the MnO 2 /MWCNT/Al electrode prepared from the as-prepared MWCNT/Al sample retained approximately 80% of the specific capacitance after 60,000 charge-discharge cycles. At the same time, the specific capacitance of the electrode obtained with the intermediate stage of electrochemical oxidation of MWCNT/Al dropped to 60% of the initial one after 16,000 cycles. Perhaps this is due to the fact that electrochemical oxidation results in partial etching of the surface of MWCNTs, which makes them less stable. For comparison, the as-grown MWCNT/Al electrode after 20,000 cha cycles in a three-electrode cell almost did not lose capacity. In a two-electr 20,000 cycles, the capacity loss was approximately 5% [53].

Conclusions
It has been shown that SC binder-free MnO2/MWCNT/Al electrodes ca by a simple treatment in an aqueous solution of KMnO4 under mild MWCNT/Al samples obtained by the direct MWCNTs deposition on alumi optimal processing conditions were chosen to ensure a significant increas trode capacitance and at the same time, preserve the MWCNT layer and much as possible. The treatment of MWCNT/Al samples in a 1% KMnO4 a tion for 40 min increased the specific capacitance of the active material of th a factor of three, up to 100-120 F/g. At the same time, excellent adhesion contact of the working material to the aluminum substrate were maintai tained MnO2/MWCNTs/Al electrodes had excellent stability to multiple cha cycles. After 60,000 CVs, the capacitance loss was less than 20%. The f vantages of the obtained binder-free MnO2/MWCNT/Al electrodes can be make them promising for wide applications. The aluminum substrate has e trical conductivity, low specific gravity, and low cost. The excellent ad MWCNT layer to the substrate ensures good electrical contact between components. The obtained electrodes show good capacitance performance sistance to multiple discharge-charge cycles. Thus, this work opens up new for using the MWCNT/Al material obtained by direct deposition of carbon aluminum foil for the fabrication of composite binder-free electrodes of sup For comparison, the as-grown MWCNT/Al electrode after 20,000 charge-discharge cycles in a three-electrode cell almost did not lose capacity. In a two-electrode cell, after 20,000 cycles, the capacity loss was approximately 5% [53].

Conclusions
It has been shown that SC binder-free MnO 2 /MWCNT/Al electrodes can be obtained by a simple treatment in an aqueous solution of KMnO 4 under mild conditions of MWCNT/Al samples obtained by the direct MWCNTs deposition on aluminum foil. The optimal processing conditions were chosen to ensure a significant increase in the electrode capacitance and at the same time, preserve the MWCNT layer and substrate as much as possible. The treatment of MWCNT/Al samples in a 1% KMnO 4 aqueous solution for 40 min increased the specific capacitance of the active material of the samples by a factor of three, up to 100-120 F/g. At the same time, excellent adhesion and electrical contact of the working material to the aluminum substrate were maintained. The obtained MnO 2 /MWCNTs/Al electrodes had excellent stability to multiple charge-discharge cycles. After 60,000 CVs, the capacitance loss was less than 20%. The following advantages of the obtained binder-free MnO 2 /MWCNT/Al electrodes can be noted, which make them promising for wide applications. The aluminum substrate has excellent electrical conductivity, low specific gravity, and low cost. The excellent adhesion of the MWCNT layer to the substrate ensures good electrical contact between all electrode components. The obtained electrodes show good capacitance performance and high resistance to multiple discharge-charge cycles. Thus, this work opens up new possibilities for using the MWCNT/Al material obtained by direct deposition of carbon nanotubes on aluminum foil for the fabrication of composite binder-free electrodes of supercapacitors.