Layer-by-Layer Heterostructure of MnO2@Reduced Graphene Oxide Composites as High-Performance Electrodes for Supercapacitors

In this paper, δ-MnO2 with layered structure was prepared by a facile liquid phase method, and exfoliated MnO2 nanosheet (e-MnO2) was obtained by ultrasonic exfoliation, whose surface was negatively charged. Then, positive charges were grafted on the surface of MnO2 nanosheets with a polycation electrolyte of polydiallyl dimethylammonium chloride (PDDA) in different concentrations. A series of e-MnO2@reduced graphene oxide (rGO) composites were obtained by electrostatic self-assembly combined with hydrothermal chemical reduction. When PDDA was adjusted to 0.75 g/L, the thickness of e-MnO2 was ~1.2 nm, and the nanosheets were uniformly adsorbed on the surface of graphene, which shows layer-by-layer morphology with a specific surface area of ~154 m2/g. On account of the unique heterostructure, the composite exhibits good electrochemical performance as supercapacitor electrodes. The specific capacitance of e-MnO2-0.75@rGO can reach 456 F/g at a current density of 1 A/g in KOH electrolyte, which still remains 201 F/g at 10 A/g. In addition, the capacitance retention is 98.7% after 10000 charge-discharge cycles at 20 A/g. Furthermore, an asymmetric supercapacitor (ASC) device of e-MnO2-0.75@rGO//graphene hydrogel (GH) was assembled, of which the specific capacitance achieves 94 F/g (1 A/g) and the cycle stability is excellent, with a retention rate of 99.3% over 10000 cycles (20 A/g).


Introduction
Under the background of rapid global economic development, continuous consumption of fossil energy, and increasingly serious environmental pollution, seeking "green" and renewable energy has become the most urgent challenge in society nowadays. Moreover, the breakthrough and popularization of large-scale energy storage technology is strong support for the development of renewable energy. In many forms of energy storage, electrochemical energy storage (EES) has been highly focused on because of its high theoretical conversion efficiency of chemical energy to electrical energy, as well as the high energy density and power density. Accordingly, further technological innovation requires continuous improvement of performance, which also drives the researchers' exploration of new materials and mechanisms. Compared to batteries, supercapacitors can offer higher power densities and have longer cycle life, faster charge-discharge capabilities and better safety. Therefore, the application of supercapacitors in the field of energy storage has attracted wide attention [1,2].
According to the mechanism of charge storage, supercapacitors can be divided into electrical double-layer capacitors (EDLC) and pseudocapacitors. The electrode materials of EDLC are mainly based on carbon materials, such as activated carbon (AC) [3], carbon nanotube (CNTs) [4][5][6], and graphene [7,8]. For pseudocapacitors, a variety of different materials, such as metal oxides [9][10][11] or hydroxides [12,13], conductive polymers [14,15] and metal sulfides [16], are all candidates for electrode materials. As the first pseudocapacitance electrode material, RuO 2 has excellent electrochemical performance [17]. However, the toxicity and high cost limit its large-scale application [18]. Some transition metal oxides with low cost also exhibit pseudocapacitive behaviors and can be used in supercapacitors instead of RuO 2 , such as MnO 2 . However, due to poor electronic conductivity, charge storage is limited to a thin layer on the surface, resulting in much lower actual capacitance than its theoretical value. In addition, low charge transfer kinetics and slow ion diffusion also affect the rate performance [19,20]. The effective strategies to improve the electrochemical performance are to increase the specific surface area of the materials by designing various nanostructures to increase the active sites or to construct hierarchical porous structures to improve mass transfer [18]. When the MnO 2 electrode is processed into ultra-thin film, the specific surface area is greatly increased and the specific capacitance can reach more than 1000 F·g −1 [21]. Lang et al. [22] proposed a composite structure of nanoporous gold (NPG) with nanocrystalline MnO 2 . NPG allows electrons to pass through MnO 2 and promotes rapid ion diffusion between MnO 2 and electrolyte, thus obtaining a high specific capacitance of~1145 F·g −1 , which is very close to the theoretical value.
Recombination and doping are representative approaches to improving the electrochemical performance of MnO 2 . The recombination could produce a synergistic effect; thus, the properties of the composite will be much better than that of the single component. The research priority is mainly to load or grow MnO 2 on porous carbon materials with large surface areas or metal substrates with good conductivity, such as graphene [23][24][25], carbon nanotubes [26], carbon fiber [27], wood-derived carbon (WC) [28], Ag [29] and Ni [30]. Especially for graphene, it is popular in composites, which could improve the electrical conductivity of the composite and reduce the solution resistance [31]. Composite nanostructure also provides an interconnected pathway for electron transportation and electrolyte diffusion [32,33], as well as inhibits the agglomeration of the individual components [34]. In addition, the heterojunction could be reasonably designed and constructed to adjust the electron structure and improve the rate of ion transport and electron transfer [35]. Metal doping (Au, Ag, Co, Al and Na) can also improve the inherent conductivity of MnO 2 and promote the electrochemical reaction, which is mainly based on the adjustment of the electronic structure [36]. Zong et al. [37] produced a positive electrode of Na-doped MnO 2 nanosheets@carbon nanotube fibers (CNTFs) with high performance. The thin nanosheets afford a large surface area for the electrode, as well as inserting Na + into MnO 2 improves the conductivity to deliver a large specific capacitance (743.3 mF·cm −2 ), leading to a broad potential window extended up to 0-1.2 V.
Chen et al. [38] successfully prepared a novel petal-like MnO 2 nanosheet@carbon sphere (CS) core-shell structure by in situ growth of MnO 2 on the surface of the carbon sphere by adjusting the amount of KMnO 4 precursor. Porous carbon spheres have a high specific surface area and suitable pore size distribution, which are suitable for energy storage and electrolyte conversion. In 1 M Na 2 SO 4 electrolyte, it has a specific capacitance of 231 F·g −1 at a current density of 0.5 A·g −1 and good cycle stability. The excellent electrochemical performance is due to the unique core-shell structure and the synergistic effect between MnO 2 and carbon spheres. Ma et al. [35] prepared layered α-MnO 2 nanowire@ultrathin δ-MnO 2 nanosheet core-shell nanostructure by a simple liquid phase technique. The novel hierarchical nanostructure is composed of ultrathin δ-MnO 2 nanosheets grown on the surface of the α-MnO 2 ultralong nanowire. When the discharge current density is as high as 20 A·g −1 , the initial specific capacitance of the composite reaches 153.8 F·g −1 , and the stability remains at 98.1% after 10,000 charge-discharge cycles. The good rate performance and stability of the composite are attributed to the structural characteristics of the two MnO 2 crystals. A 1D α-MnO 2 nanowire as the core provided a stable skeleton structure, and ultra-thin 2D δ-MnO 2 nanosheets as the shell formed more active sites. Therefore, the synergistic effect of different dimensions is of great benefit to the improvement of the electrochemical performance. In conclusion, constructing special composite microstructure by compounding materials with different properties is considered to be an effective way to obtain excellent performance as supercapacitor electrodes on account of their good synergistic effects, including material synergistic effect, dimensional synergistic effect and heterostructural synergistic effect. Researchers have carried out extensive exploration of the controllable synthesis, morphology control, structural design and electrochemical performance improvement of the composites of MnO 2 with different dimensional carbon materials [39,40].
In general, MnO 2 with relatively poor conductivity is coated on the surface of a conductive substrate, such as graphene, which makes the conductive substrate materials unable to contact each other directly, resulting in the increase of the contact resistance between the particles, which seriously affects the electrochemical performance of the composite. At present, the MnO 2 layer on the surface of the conductive substrate is often very thin. If the thickness increases, the electrochemical performance of the composite will decrease significantly, but the thin oxide layer will reduce the available active material, which is not beneficial to the full play of its excellent electrochemical properties. Together with the low specific capacitance of carbon materials, it is important to optimize the ratio and the structure of carbon and MnO 2 in the composites and design an MnO 2 /carbon interface. In addition, constructing a layer-by-layer structure is a viable choice. Research on layerby-layer structure are mostly focused on multilayer films by layer-by-layer self-assembly in the presence of substrates, such as carbon cloth [41], indium-tin-oxide (ITO) [42], Ni foam [43] and gold-coated poly-(ethylene terephthalate) (PET) [44].
In this work, on the basis of the preparation of δ-MnO 2 with a layered structure, MnO 2 laminas were obtained by ultrasonic exfoliation, which was positively charged after surface charge modification by polydiene dimethyl ammonium chloride (PDDA). As a commonly used cationic polyelectrolyte, PDDA has many advantages, such as safety, non-toxicity, easy solubility in water, strong cohesion, good hydrolysis stability and low cost. There is also another positively charged polyelectrolyte always used for charge regulation, such as polyethyleneimine (PEI). As a strong polyelectrolyte, the electrostatic interaction between PDDA and rGO is stronger than that produced by PEI [45], so PDDA was chosen in our research. Then, the composites with the 2D layer-by-layer structure were acquired with no substrate by the self-assembly of MnO 2 laminas with the surface positively charged and graphene oxide (GO) nanosheets with the surface negatively charged through electrostatic attraction. The final product, MnO 2 /reduced graphene oxide (rGO), was obtained by the reduction of GO to rGO with glucose as the reductant under mild conditions. As expected, rGO in the layer-by-layer structure has acted as a conductive layer and bridge to improve the electrical conductivity of the MnO 2 /rGO composite and relieved the stacking of MnO 2 nanosheets. A series of self-assembled MnO 2 /rGO composites were prepared by adjusting the concentration of cationic polymer for surface charge modification. As supercapacitor electrodes, the composite designed with layer-by-layer heterostructure shows high performance.

Reagents and Materials
All the reagents used in this part were of analytical grade. KMnO [46]. The product was cooled to room temperature and then centrifuged at 5000 rpm. The centrifuged product was freeze-dried after washing, which was MnO 2 . 400 mg of MnO 2 above prepared was dispersed in 250 mL deionized water to be exfoliated under ultrasonication (40 KHz, 240 W) for 3.5 h. Every 0.5 h, a stirring of 5 min was needed, and the ice bath environment was always maintained during the ultrasonication process. After ultrasonication, the solution was centrifuged at 5800 rpm, and the upper liquid was freeze-dried to obtain the exfoliated MnO 2 nanosheets, denoted as e-MnO 2 . The exfoliated MnO 2 dispersion presents Tyndall effect through testing.
The tested Zeta potential of e-MnO 2 was −20.  Table S1 (Supplementary Information). It can be seen that when the concentration of PDDA is 0.5 g/L, the Zeta potential of e-MnO 2 -0.5 reaches 27 mV. When the concentration increases to 0.75 g/L, the Zeta potential of e-MnO 2 -0.75 increases to 35.7 mV, whereas the concentration of PDDA increases to 1 g/L, the Zeta potential of e-MnO 2 -1 decreases to 30.2 mV. It is mainly because when the amount of PDDA is low, it cannot effectively prevent the coagulation effect of electrolytes on the sol system, but when the amount of PDDA is too high, it will affect the amount of charge in the diffusion layer of the micelle and also cause the decrease of the Zeta potential of the colloid system, which is not beneficial to the stable existence of the sol [47].
GO was prepared using the modified Hummers method [48]. 40 mg of GO was dispersed in 40 mL of deionized water to obtain GO assembly solution (1 mg/mL). The tested Zeta potential of GO is −50 mV. As known, when the absolute value of the potential exceeds 30 mV, stable dispersion can be formed [49]. Under continuous stirring, GO assembly solution was added into e-MnO 2 -0.75 assembly solution slowly, also into e-MnO 2 -0.5 and e-MnO 2 -1 dispersion for comparison, and then stirred for 40 min. During the process, it could be observed that with the addition of GO, coagulation occurred. Positively charged e-MnO 2 -0.75 and negatively charged GO completed electrostatic self-assembly. After static settlement, the supernatant was removed, and then the product was filtrated and freeze-dried, denoted as e-MnO 2 -0.5@GO, e-MnO 2 -0.75@GO and e-MnO 2 -1@GO, respectively. The prepared e-MnO 2 -0.75@GO was dispersed in 40 mL of deionized water. Under magnetic stirring, 100 µL NH 3 ·H 2 O solution (25% w/w) was added to adjust pH to~9-10, then 640 mg glucose was added. After stirring for 15 min, the mixture was transferred into a Teflon-lined stainless steel autoclave of 50 mL and reacted at 95 • C for 1.5 h in order to reduce GO to rGO. The precipitate was repeatedly washed and freezedried to acquire e-MnO 2 -0.75@rGO composite. In the same way, e-MnO 2 -0.5@rGO and e-MnO 2 -1@rGO were synthesized. The schematic synthesis procedure for the e-MnO 2 @rGO composites is illustrated in Figure 1.

Characterization
The phase structures of the as-prepared materials were performed using the powder Xray diffraction (XRD, Rigaku D-max-2500/PC) with Cu Kα radiation (λ = 0.15406 nm) over a range 2θ = 5-70 • . The elemental analysis was detected by X-ray photoelectron spectroscopy (XPS, Thermo Fischer, ESCALAB Xi+) with Al Kα radiation (hγ = 1486.6 eV). The analysis of chemical bond was completed via Fourier transform infrared spectroscopy (FTIR, Thermo Nicolet iS10), and the preparation process of samples for FTIR is detailed in supplementary information. The morphologies were characterized by scanning electron microscopy (SEM, Zeiss Supra55 and Hitachi Regulus SU8230), transmission electron microscopy (TEM, Hitachi HT-7700) and atomic force microscopy (AFM, Bruker Multimode 8). The X-ray energy disperse spectra (EDS) of the samples were recorded on Oxford Instruments (Ultim Max170). The specific surface area was measured by Brunauer-Emmett-Teller (BET) method (Micromeritics Tristar II 3020).

Characterization
The phase structures of the as-prepared materials were performed using the powder X-ray diffraction (XRD, Rigaku D-max-2500/PC) with Cu Kα radiation (λ = 0.15406 nm) over a range 2θ = 5-70°. The elemental analysis was detected by X-ray photoelectron spectroscopy (XPS, Thermo Fischer, ESCALAB Xi+) with Al Kα radiation (hγ = 1486.6 eV). The analysis of chemical bond was completed via Fourier transform infrared spectroscopy (FTIR, Thermo Nicolet iS10), and the preparation process of samples for FTIR is detailed in supplementary information. The morphologies were characterized by scanning electron microscopy (SEM, Zeiss Supra55 and Hitachi Regulus SU8230), transmission electron microscopy (TEM, Hitachi HT-7700) and atomic force microscopy (AFM, Bruker Multimode 8). The X-ray energy disperse spectra (EDS) of the samples were recorded on Oxford Instruments (Ultim Max170). The specific surface area was measured by Brunauer-Emmett-Teller (BET) method (Micromeritics Tristar Ⅱ 3020).

Electrochemical Measurements
The as-prepared material was evenly mixed with acetylene black and polytetrafluoroethylene (PTFE, 60 wt.%) at a mass ratio of 80:15:5, and then a modicum of anhydrous ethanol was added to make a paste, which was smeared on the surface of nickel foam, and dried at 60 °C. The nickel foam loaded with active material was pressed under a pressure of 10 MPa, and then soaked in 6 M KOH solution for 24 h for activation. The electrochemical performance was tested in a three-electrode system with 6M KOH aqueous as electrolyte. The as-prepared active material was used as the working electrode, platinum plate electrode as the counter electrode, and Hg/HgO electrode as the reference electrode. Galvanostatic charge-discharge (GCD) tests were recorded on a charge-discharge instrument (Neware CT-4008T, Shenzhen, China) with a potential range of approximately −0.2-0.5 V. Cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) were carried out on a CHI660E electrochemical workstation (Chenhua, Shanghai, China). The potential window of CV tests was from −0.2 V to 0.5 V, and the frequency range of EIS was from 10 −2 Hz to 10 5 Hz with the amplitude of 5 mV.
The asymmetric supercapacitor (ASC) device was assembled by using the asprepared e-MnO2-0.75@rGO as the positive electrode, graphene hydrogel (GH) as the negative electrode and 6 M KOH as electrolyte. The preparation of GH is described in supplementary information. Wherein the mass ratio of positive and negative electrodes was obtained by the equation below [50]: where m is the mass of active materials (g), Cs is the specific capacitance of electrodes (F·g −1 ) and ∆V is the potential window (V), and the sign of "+" and "−" represents the

Electrochemical Measurements
The as-prepared material was evenly mixed with acetylene black and polytetrafluoroethylene (PTFE, 60 wt.%) at a mass ratio of 80:15:5, and then a modicum of anhydrous ethanol was added to make a paste, which was smeared on the surface of nickel foam, and dried at 60 • C. The nickel foam loaded with active material was pressed under a pressure of 10 MPa, and then soaked in 6 M KOH solution for 24 h for activation. The electrochemical performance was tested in a three-electrode system with 6M KOH aqueous as electrolyte. The as-prepared active material was used as the working electrode, platinum plate electrode as the counter electrode, and Hg/HgO electrode as the reference electrode. Galvanostatic charge-discharge (GCD) tests were recorded on a charge-discharge instrument (Neware CT-4008T, Shenzhen, China) with a potential range of approximately −0.2-0.5 V. Cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) were carried out on a CHI660E electrochemical workstation (Chenhua, Shanghai, China). The potential window of CV tests was from −0.2 V to 0.5 V, and the frequency range of EIS was from 10 −2 Hz to 10 5 Hz with the amplitude of 5 mV.
The asymmetric supercapacitor (ASC) device was assembled by using the as-prepared e-MnO 2 -0.75@rGO as the positive electrode, graphene hydrogel (GH) as the negative electrode and 6 M KOH as electrolyte. The preparation of GH is described in supplementary information. Wherein the mass ratio of positive and negative electrodes was obtained by the equation below [50]: where m is the mass of active materials (g), C s is the specific capacitance of electrodes (F·g −1 ) and ∆V is the potential window (V), and the sign of "+" and "−" represents the positive and negative electrodes, respectively.  200) and (020) crystal planes, respectively [51]. Simultaneously, rGO was prepared using the same reduction process of the composite. Moreover, XRD patterns of rGO and GO are displayed in Figure S1 (Supplementary Information). The XRD curve of GO has a strong diffraction peak at 2θ = 11.8 • , corresponding to (001) crystal plane. After reduced by hydrothermal reduction with glucose, a broad diffraction peak (002) appears at 2θ = 24.5 • , indicating the reduction of GO to rGO. The position of which is close to the (002) plane of δ-MnO 2 , so there may be an overlap of (002) peaks for MnO 2 and rGO [52]. Moreover, the existence of rGO in the composite will be further proved by subsequent SEM, TEM and XPS characterization. According to the Bragg equation, the basal plane spacing calculated from the (001) plane is about 0.72 nm. Compared with MnO 2 , there is no obvious change for the position of the diffraction peaks of e-MnO 2 , indicating that the phase structure of MnO 2 has no change after ultrasonic exfoliation, but the intensity of the diffraction peaks weakened, especially the peaks corresponding to (001) and (002) planes. For the e-MnO 2 @rGO composites with different concentrations of PDDA (0.5, 0.75 and 1 g/L), they all present the diffraction peaks of δ-MnO 2 only, which is probably because MnO 2 laminas covered on the surface of rGO [53]. Moreover, with the change of the concentration, the XRD patterns of e-MnO 2 -0.5@rGO, e-MnO 2 -0.75@rGO and e-MnO 2 -1@rGO have little change, suggesting no effect on the phase structure of the composites for the charge regulation on the surface of e-MnO 2 . FTIR spectra were used to further represent the structure of MnO 2 , e-MnO 2 and e-MnO 2 -0.75@rGO, as shown in Figure 2b [54]. The intensity of the band of e-MnO 2 is a little stronger than MnO 2 , mainly because more functional groups are exposed on the surface of the nanosheets.

Results and Discussion
1@rGO are shown in Figure 2a. For MnO2, the diffraction peaks of 2θ at 12.3°, 24.9°, and 65.5° can be indexed to birnessite-type MnO2 (δ-MnO2) (PDF# 43-1456), cor sponding to (001), (002), (200) and (020) crystal planes, respectively [51]. Simultaneous rGO was prepared using the same reduction process of the composite. Moreover, XR patterns of rGO and GO are displayed in Figure S1 (Supplementary Information). T XRD curve of GO has a strong diffraction peak at 2θ = 11.8°, corresponding to (0 crystal plane. After reduced by hydrothermal reduction with glucose, a broad diffract peak (002) appears at 2θ = 24.5°, indicating the reduction of GO to rGO. The position which is close to the (002) plane of δ-MnO2, so there may be an overlap of (002) pea for MnO2 and rGO [52]. Moreover, the existence of rGO in the composite will be furth proved by subsequent SEM, TEM and XPS characterization. According to the Bra equation, the basal plane spacing calculated from the (001) plane is about 0.72 n Compared with MnO2, there is no obvious change for the position of the diffracti peaks of e-MnO2, indicating that the phase structure of MnO2 has no change after ult sonic exfoliation, but the intensity of the diffraction peaks weakened, especially peaks corresponding to (001) and (002) [54]. The intensity of the band of e-MnO2 is a little stronger than MnO2, mainly becau more functional groups are exposed on the surface of the nanosheets.   (Figure 3d), which are assigned to Mn 4+ (2p 1/2 ), Mn 3+ (2p 1/2 ), Mn 4+ (2p 3/2 ) and Mn 3+ (2p 1/2 ), respectively [59,60]. The existence of Mn 3+ is probably to maintain charge neutrality and oxygen vacancies in the MnO 2 lattice [60]. The XPS results further confirm that it has been synthesized successfully of the e-MnO 2 -0.75@rGO composite. are attributed to the adsorbed water molecules and surface functional groups of rGO in the composite, respectively, and the Mn-O bond belongs to MnO2. For Mn 2p core-level, it could be fitted into four peaks at 654.8 eV, 652.9 eV, 644.4 eV and 641.8 eV (Figure 3d), which are assigned to Mn 4+ (2p1/2), Mn 3+ (2p1/2), Mn 4+ (2p3/2) and Mn 3+ (2p1/2), respectively [59,60]. The existence of Mn 3+ is probably to maintain charge neutrality and oxygen vacancies in the MnO2 lattice [60]. The XPS results further confirm that it has been synthesized successfully of the e-MnO2-0.75@rGO composite. The SEM image of the as-prepared MnO2 is shown in Figure 4a. It exhibits a large area and continuous lamellar morphology with the size of several hundred nanometers, and the nanosheets cross with each other. AFM was used to represent the morphology of e-MnO2. As shown in Figure 4b, smooth nanosheets are observed, which have a large surface with a dimension of ~800 nm from the AFM image (Figure 4b, on the left), and a thickness of ~1.2 nm from the height profile (Figure 4b, on the right). The theoretical thickness of the single-layer MnO2 nanosheet is 0.52 nm [42], but it should be considered the existence of hydration on both sides of the single-layer nanosheet [61]. So the thickness of the obtained e-MnO2 nanosheet is approximately two layers. The morphologies of e-MnO2@rGO composites under different concentrations of PDDA were characterized by SEM. When the concentration of PDDA is 0.5 g/L, the nanosheets of e-MnO2-0.5@rGO are stacked, and the structure is relatively compact (Figure 4c), indicating that the dispersed lamellas failed to attract effectively and aggregated again. Although rGO could The SEM image of the as-prepared MnO 2 is shown in Figure 4a. It exhibits a large area and continuous lamellar morphology with the size of several hundred nanometers, and the nanosheets cross with each other. AFM was used to represent the morphology of e-MnO 2 . As shown in Figure 4b, smooth nanosheets are observed, which have a large surface with a dimension of~800 nm from the AFM image (Figure 4b, on the left), and a thickness of~1.2 nm from the height profile (Figure 4b, on the right). The theoretical thickness of the single-layer MnO 2 nanosheet is 0.52 nm [42], but it should be considered the existence of hydration on both sides of the single-layer nanosheet [61]. So the thickness of the obtained e-MnO 2 nanosheet is approximately two layers. The morphologies of e-MnO 2 @rGO composites under different concentrations of PDDA were characterized by SEM. When the concentration of PDDA is 0.5 g/L, the nanosheets of e-MnO 2 -0.5@rGO are stacked, and the structure is relatively compact (Figure 4c), indicating that the dispersed lamellas failed to attract effectively and aggregated again. Although rGO could increase the conductivity of the composite, the compact structure makes it difficult for the electrolyte to enter the material. When the concentration of PDDA is 0.75 g/L, it can be observed from Figure 4d that e-MnO 2 -0.75@rGO composite displays a layer-by-layer structure with relatively uniform compounding, demonstrating that rGO and e-MnO 2 were assembled well. As can be seen from the TEM image of e-MnO 2 -0.75@rGO (Figure 4f), e-MnO 2 lamellas are distributed on the surface of rGO nanosheets. These e-MnO 2 lamellas interlace with each other, and the material shows a relatively transparent state, indicating a less-layer structure. For the novel layer-by-layer heterostructure, it has many advantages: Firstly, the compact combination of e-MnO 2 and rGO not only improves the conductivity of the composite but also mitigates the powder dropping caused by the expansion of MnO 2 during the charge-discharge process. Furthermore, the cross-linking between e-MnO 2 nanosheets results in a large number of pores, which make ions in electrolytes easily accessible to the layers of the nanosheets, thus greatly increasing active sites. When the concentration of PDDA increased to 1 g/L, e-MnO 2 -1@rGO presented a lamellar accumulation structure (Figure 4e), indicating that e-MnO 2 and rGO have not formed a good assembly. As shown in Figure 4g of SEM mapping images, Mn, O and C elements distribute homogeneously over the e-MnO 2 -0.75@rGO architecture, which further proves the existence and interlacing distribution of MnO 2 and rGO in the composite. 4f), e-MnO2 lamellas are distributed on the surface of rGO nanosheets. These e-MnO lamellas interlace with each other, and the material shows a relatively transparent stat indicating a less-layer structure. For the novel layer-by-layer heterostructure, it ha many advantages: Firstly, the compact combination of e-MnO2 and rGO not only im proves the conductivity of the composite but also mitigates the powder dropping cause by the expansion of MnO2 during the charge-discharge process. Furthermore, the cross linking between e-MnO2 nanosheets results in a large number of pores, which make ion in electrolytes easily accessible to the layers of the nanosheets, thus greatly increasin active sites. When the concentration of PDDA increased to 1 g/L, e-MnO2-1@rGO pre sented a lamellar accumulation structure (Figure 4e), indicating that e-MnO2 and rG have not formed a good assembly. As shown in Figure 4g of SEM mapping images, Mn O and C elements distribute homogeneously over the e-MnO2-0.75@rGO architectur which further proves the existence and interlacing distribution of MnO2 and rGO in th composite. The specific surface area and pore structure of the e-MnO2-0.75@rGO composit were investigated by analyzing N2 adsorption-desorption isotherms and the pore siz distribution curve, as shown in Figure 5. It shows a type IV isotherm with a hysteres loop at the relative pressure of ~0.6-1.0, indicating the existence of mesoporous structur in e-MnO2-0.75@rGO (Figure 5a) [62]. The specific surface area of e-MnO2-0.75@rGO 154.3 m 2 /g, which is higher or comparable compared with that reported in the literatur such as MnO2 nanowires/rGO (139.9 m 2 /g) [62], reduced graphene/MnO2 (120.2 m 2 /g [63] and high-reduced graphene (HRGO)/MnO2 (159.1 m 2 /g) [63]. For a comparison, ad sorption-desorption isotherms of MnO2 and rGO are displayed in Figure S2a,c (Sup The specific surface area and pore structure of the e-MnO 2 -0.75@rGO composite were investigated by analyzing N 2 adsorption-desorption isotherms and the pore size distribution curve, as shown in Figure 5. It shows a type IV isotherm with a hysteresis loop at the relative pressure of~0.6-1.0, indicating the existence of mesoporous structure in e-MnO 2 -0.75@rGO (Figure 5a) [62]. The specific surface area of e-MnO 2 -0.75@rGO is 154.3 m 2 /g, which is higher or comparable compared with that reported in the literature, such as MnO 2 nanowires/rGO (139.9 m 2 /g) [62], reduced graphene/MnO 2 (120.2 m 2 /g) [63] and high-reduced graphene (HRGO)/MnO 2 (159.1 m 2 /g) [63]. For a comparison, adsorptiondesorption isotherms of MnO 2 and rGO are displayed in Figure S2a,c (Supplementary Information). The specific surface area of MnO 2 and rGO is 78.7 and 207.8 m 2 /g, respectively. It can be seen that the higher specific surface area of rGO plays a positive role in the composite, and the specific surface area of e-MnO 2 -0.75@rGO composite is significantly improved compared with that of pristine MnO 2 . Additionally, the pore size distribution curve reveals that the average pore size calculated by Barrett-Joyner-Halenda (BJH) model is concentrated from 20 to 40 nm (Figure 5b), demonstrating the mesoporous structure of the as-prepared composite. While the pore size distribution of MnO 2 and rGO is both concentrated at~3-4 nm ( Supplementary Information, Figure S2b,d), so the porosity is owing to the composite layer-by-layer architecture formed by MnO 2 and rGO, which is ascribed to the cross-linked structure constructed by self-assembly of the folded nanosheets. High specific surface area and suitable pore size are favorable for ions transport during the charge-discharge process. rett-Joyner-Halenda (BJH) model is concentrated from 20 to 40 nm (Figure 5b), demonstrating the mesoporous structure of the as-prepared composite. While the pore size distribution of MnO2 and rGO is both concentrated at ~3-4 nm (Supplementary Information, Figure S2b,d), so the porosity is owing to the composite layer-by-layer architecture formed by MnO2 and rGO, which is ascribed to the cross-linked structure constructed by self-assembly of the folded nanosheets. High specific surface area and suitable pore size are favorable for ions transport during the charge-discharge process.

Electrochemical Performance
For the evaluation of electrochemical performance, CV curves at various scan rates and GCD curves at various current densities were tested, as shown in Figure 6. Figure  6a-c show CV curves of e-MnO2-0.5@rGO, e-MnO2-0.75@rGO and e-MnO2-1@rGO, respectively. CV curves of e-MnO2@rGO obtained at different PDDA concentrations are similar in shape and approximate in rectangle, indicating that the materials present good pseudocapacitance characteristics. At the same scan rate, the CV curve of e-MnO2-0.75@rGO has the maximum current response, demonstrating the maximum specific capacitance. Moreover, GCD curves of e-MnO2-0.5@rGO, e-MnO2-0.75@rGO and e-MnO2-1@rGO are shown in Figure 6d-f), respectively. These GCD curves at different current densities are close to symmetric triangles, indicating good reversibility. At the same current density, the GCD curve of e-MnO2-0.75@rGO has the longest discharge time, also manifesting the highest specific capacitance, which is consistent with the CV results. At a current density of 1 A/g, the specific capacitance of e-MnO2-0.5@rGO, e-MnO2-0.75@rGO and e-MnO2-1@rGO calculated by the GCD curves is 236 F/g, 456 F/g and 298 F/g, respectively. Combined with the above Zeta potential results after surface charge regulation, the higher the absolute value of Zeta potential is, the more stable the system is, so the self-assembly effect is better. When the concentration of PDDA was 0.5 g/L, due to the low concentration of the charge regulating solution, the self-assembly via electrostatic gravity was not ideal, and the nanosheets were aggregated and stacked, which reduced the number of the active sites of MnO2 and affected its specific capacitance. However, Zeta potential at 1 g/L of PDDA decreased compared with that at 0.75 g/L of PDDA, which did not achieve complete assembly. Furthermore, the higher concentration of PDDA made the long molecular chain at the outer end cause micelle expansion and sedimentation. While excess PDDA polymers mixed in the composite

Electrochemical Performance
For the evaluation of electrochemical performance, CV curves at various scan rates and GCD curves at various current densities were tested, as shown in Figure 6.  Figure 6d-f), respectively. These GCD curves at different current densities are close to symmetric triangles, indicating good reversibility. At the same current density, the GCD curve of e-MnO 2 -0.75@rGO has the longest discharge time, also manifesting the highest specific capacitance, which is consistent with the CV results. At a current density of 1 A/g, the specific capacitance of e-MnO 2 -0.5@rGO, e-MnO 2 -0.75@rGO and e-MnO 2 -1@rGO calculated by the GCD curves is 236 F/g, 456 F/g and 298 F/g, respectively. Combined with the above Zeta potential results after surface charge regulation, the higher the absolute value of Zeta potential is, the more stable the system is, so the self-assembly effect is better. When the concentration of PDDA was 0.5 g/L, due to the low concentration of the charge regulating solution, the self-assembly via electrostatic gravity was not ideal, and the nanosheets were aggregated and stacked, which reduced the number of the active sites of MnO 2 and affected its specific capacitance. However, Zeta potential at 1 g/L of PDDA decreased compared with that at 0.75 g/L of PDDA, which did not achieve complete assembly. Furthermore, the higher concentration of PDDA made the long molecular chain at the outer end cause micelle expansion and sedimentation. While excess PDDA polymers mixed in the composite slowed down the agglomeration, it was bound to reduce the electrical conductivity of the material, thus affecting the electrochemical performance, resulting in the decline of the specific capacitance. The results correspond with SEM analysis.
For a comparison, CV curves at 5 mV/s and GCD curves at 1 A/g of MnO 2 , e-MnO 2 and e-MnO 2 -0.75@rGO are displayed in Figure 7a,b). The CV curve area of e-MnO 2 is larger than that of pristine MnO 2 . The composite of e-MnO 2 -0.75@rGO shows the largest closed area, indicating the highest specific capacitance among the three. The specific capacitance of MnO 2 , e-MnO 2 and e-MnO 2 -0.75@rGO is 268, 360 and 456 F/g, respectively, at a current density of 1 A/g, calculated from the charge-discharge curve in Figure 7b. (The average values and standard deviations were given in Figure S3 of supplementary information.) Compared with pristine MnO 2 , e-MnO 2 has a larger specific surface, and more active sites are exposed to the electrolyte, so it has higher specific capacitance than pristine MnO 2 . Furthermore, e-MnO 2 -0.75@rGO composite has the highest specific capacitance, 70% higher than that of pristine MnO 2 , mainly because rGO and e-MnO 2 were laminated in a layer-bylayer structure, which improves the conductivity of the material and inhibits the stacking of the nanosheets. According to Figure 7c, for the Nyquist plots of EIS, the intercept at the real axis represents the equivalent series resistance (R s ), including electrolyte resistance, contact resistance between the electrode material and the current collector, or the internal resistance of the material [64]. The diameter of the semi-arc intersecting at the real axis represents the charge transfer resistance (R ct ), the value of which is proportional to the R ct . As can be seen in Figure 7c, the diameter of the semi-arc for e-MnO 2 -0.75@rGO is very small, indicating lower R ct . The plots were fitted according to the equivalent circuit given in Figure 7d, and the fitting results are also shown in Figure 7d. The R s of MnO 2 and e-MnO 2 -0.75@rGO is 0.69 Ω and 0.46 Ω, respectively. Moreover, the corresponding R ct is 0.42 Ω and 0.26 Ω, manifesting that the R s and R ct of e-MnO 2 -0.75@rGO are both lower than pristine MnO 2 . It is further demonstrated that e-MnO 2 -0.75@rGO composite with lamellar structure has good conductivity, and the lamellar structure could contribute to the transfer of electrolyte ions, while rGO is also conducive to reducing the internal resistance of the material. Figure 7e shows the rate performance of MnO 2 , e-MnO 2 -0.5@rGO, e-MnO 2 -0.75@rGO and e-MnO 2 -1@rGO at the current densities increasing from 1 to 10 A/g. Obviously, the capacitance retention of e-MnO 2 -0.5@rGO, e-MnO 2 -0.75@rGO and e-MnO 2 -1@rGO are 16.9%, 44% and 34.6%, respectively, which is more superior than that of pristine MnO 2 (10.6%), owing to the role of rGO in stabilizing the architecture. In particular, the rate performance of e-MnO 2 -0.75@rGO is significantly the best in the e-MnO 2 @rGO composites with different concentrations of PDDA, and it has the highest specific capacitance at different current densities of 1, 2, 5, 8 and 10 A/g. Figure 7f shows the cycle performance and coulombic efficiency of the e-MnO 2 -0.75@rGO composite. It can be seen that the specific capacitance retention remains at 98.7% of the initial value after 10,000 charge-discharge cycles at a current density of 20 A/g, and coulombic efficiency is about 98.7%.  For a comparison, CV curves at 5 mV/s and GCD curves at 1 A/g of MnO2, e-Mn and e-MnO2-0.75@rGO are displayed in Figure 7a,b). The CV curve area of e-MnO larger than that of pristine MnO2. The composite of e-MnO2-0.75@rGO shows the larg closed area, indicating the highest specific capacitance among the three. The specific pacitance of MnO2, e-MnO2 and e-MnO2-0.75@rGO is 268, 360 and 456 F/g, respective at a current density of 1 A/g, calculated from the charge-discharge curve in Figure  (The average values and standard deviations were given in Figure S3 of supplement The specific capacitance of e-MnO 2 -0.75@rGO composite synthesized by electrostatic self-assembly combined with hydrothermal reduction after exfoliating MnO 2 nanosheets in this work was compared with the composites of MnO 2 with graphene prepared via a variety of methods reported in the literature [47,[63][64][65][66][67][68][69]. As shown in Table 1, in terms of the specific capacitance, the result of our work is comparable. respectively, which is more superior than that of pristine MnO2 (10.6%), owing to the role of rGO in stabilizing the architecture. In particular, the rate performance of e-MnO2-0.75@rGO is significantly the best in the e-MnO2@rGO composites with different concentrations of PDDA, and it has the highest specific capacitance at different current densities of 1, 2, 5, 8 and 10 A/g. Figure 7f shows the cycle performance and coulombic efficiency of the e-MnO2-0.75@rGO composite. It can be seen that the specific capacitance retention remains at 98.7% of the initial value after 10,000 charge-discharge cycles at a current density of 20 A/g, and coulombic efficiency is about 98.7%. The specific capacitance of e-MnO2-0.75@rGO composite synthesized by electrostatic self-assembly combined with hydrothermal reduction after exfoliating MnO2 Compared with pristine MnO 2 , the improved electrochemical performance of e-MnO 2 -0.75@rGO composite is mainly due to the following reasons. Briefly, a layer-by-layer heterostructure of e-MnO 2 -0.75@rGO facilitates the diffusion of electrolyte ions. Meanwhile, the architecture was developed by the self-assembly of graphene nanosheets and e-MnO 2 nanoflakes, which could effectively inhibit the restacking of e-MnO 2 and graphene. In addition, the intersecting e-MnO 2 nanosheets formed a large number of pores, promoting rapid faradaic reactions. Furthermore, the synergistic effect of components must be mentioned. As a conductive layer in the heterostructure, rGO improves the electrical conductivity of the e-MnO 2 -0.75@rGO composite, which is beneficial to the improvement of the overall electrochemical performance [70,71].
Additionally, an asymmetric supercapacitor (ASC) device was assembled, and its electrochemical performance was evaluated. The as-prepared e-MnO 2 -0.75@rGO was used as the positive electrode, graphene hydrogel (GH) was used as the negative electrode and 6 M KOH was used as the electrolyte. Figure 8a shows CV curves of the GH negative electrode and e-MnO 2 -0.75@rGO positive electrode at the scan rate of 5 mV/s. The potential window of the negative electrode is approximately −1-0 V, and that of the positive electrode is approximately −0.2-0.5 V. Figure 8b displays CV curves under various potential windows. It can be observed that CV curves remain a good rectangle when the voltage windows are~0-1.5 V and~0-1.6 V; while the voltage rises to 1.7 V, slight polarization occurs. Therefore,~0-1.6 V was selected as the potential window of the ASC device. GCD curves of e-MnO 2 -0.75@rGO//GH ASC at current densities of 1, 2, 5, 8 and 10 A/g with the potential window of~0-1.6 V are shown in Figure 8c. All the GCD curves display approximately symmetrical triangles, indicating good reversibility of the device. The calculated specific capacitance could reach 94 F/g at 1 A/g and still keep 35 F/g at 10 A/g. The cycle performance of the device is revealed in Figure 8d. The specific capacitance retention of e-MnO 2 -0.75@rGO//GH ASC can reach 99.3% after 10,000 cycles at 20 A/g, demonstrating good stability.

Conclusions
In summary, MnO2 nanoflakes with a thickness of ~1.2 nm and size of ~800 nm were obtained by ultrasonic exfoliation and charge regulated by the appropriate concentration of PDDA (0.75 g/L). Then MnO2 and GO nanosheets were self-assembled by

Conclusions
In summary, MnO 2 nanoflakes with a thickness of~1.2 nm and size of~800 nm were obtained by ultrasonic exfoliation and charge regulated by the appropriate concentration of PDDA (0.75 g/L). Then MnO 2 and GO nanosheets were self-assembled by electrostatic force. The composite of e-MnO 2 -0.75@rGO with layer-by-layer heterostructure was acquired after hydrothermal reduction by glucose. The composite exhibits excellent electrochemical performance. In 6 M KOH electrolyte, the specific capacitance of e-MnO 2 -0.75@rGO is 456 F/g at a current density of 1 A/g, which is much higher than that of pristine MnO 2 (268 F/g). Even at 10 A/g, the specific capacitance still retains 201 F/g, and the specific capacitance retention is 98.7% after 10,000 charge-discharge cycles at 20 A/g. It shows that rGO improves the conductivity of the material, and the layer structure formed by rGO is conducive to the migration of ions in the electrolyte, so the specific capacitance of the composite is greatly enhanced. The improved electrochemical performance is attributed to the synergistic effect of architecture coupled with components, including the rate of ion transport and faradaic reaction, plenty of active sites, less restacking as well as improved electrical conductivity. Moreover, the assembled e-MnO 2 -0.75@rGO//GH ASC device shows a specific capacitance of 94 F/g at 1 A/g with a potential window of~0-1.6 V and better cycle stability with capacitance retention of 99.3% over 10,000 cycles at 20 A/g. We believe that this work may provide a reference for the synthesis of the composites with layer-by-layer structure by self-assembly method and the basis for the design and comparison of electrode materials for high-performance supercapacitors. Reasonable design and construction of heterostructure have positive effects on the electrochemical performance. In future studies, we need to further optimize the structure and improve the electrochemical performance, as well as carry out researches on the interface mechanism of the heterostructure so as to better exploit the potential of MnO 2 /rGO composites in the application of supercapacitors.
Supplementary Materials: The following supporting information can be downloaded at: https://www. mdpi.com/article/10.3390/membranes12111044/s1, preparation of GH anode materials, preparation of samples for FTIR, Table S1: Zeta potential of e-MnO 2 regulated by different concentrations of PDDA, Figure S1: XRD patterns of GO and rGO, Figure S2. Adsorption-desorption isotherms of (a) MnO 2 and (c) rGO and the pore-size distribution of (b) MnO 2 and (d) rGO and Figure S3