In this study, a multicore-shell ternary composite electrode material (MnO2@Ppy@NPCNFs) with excellent electrochemical performances was prepared by using surface modification, in which core-shell Ppy@N-doped porous carbon nanofibers (Ppy@NPCNFs) with large specific surface area and high conductivity were used as the substrate (a multicore layer), and MnO2 was loaded on the substrate by hydrothermal synthesis to form a shell layer, further improving the SC of electrode material. The parameters of hydrothermal growth of MnO2 on Ppy@NPCNFs were explored by means of the control variable method and response surface methodology, and the optimal parameters were predicted and verified. Electrochemical test results showed that the SC of MnO2@Ppy@NPCNFs prepared under the optimal reaction parameters was as high as 595.77 F g-1, and its capacitance retention was 96.2% after 1000 cycles. Moreover, a symmetric supercapacitor prepared with the optimal multicore-shell electrode showed an energy density of 9.36 Wh kg−1 at a power density of 1000 W kg−1 and a retention rate of 93.75% after 1000 cycles, indicating the promising application of multicore-shell ternary composite electrode material in high-performance supercapacitors.
Research Article
Multicore-shell MnO2@Ppy@N-doped porous carbon nanofiber ternary composites as electrode materials for high- performance supercapacitors
https://doi.org/10.21203/rs.3.rs-2690092/v1
This work is licensed under a CC BY 4.0 License
Version 1
posted
You are reading this latest preprint version
Nanocomposites
Nano-structures
Electrical properties
Supercapacitor electrode
The development of sustainable and renewable energy storage systems with high power and energy density has become a global concern to address the critical issues of rapid depletion of fossil fuels worldwide and environmental pollution [1–4]. Among the current energy storage systems, supercapacitors offer superior performances over batteries, such as high power density, rapid charge/discharge rates and long cycle stability [5–9], which make them attractive alternatives or complements to batteries for a variety of applications from portable electronic devices and hybrid electric vehicles. Potential candidates for electrode materials, which are essential components of supercapacitors, include metal oxides, carbon materials and conductive polymers [10–12].
In recent years, MnO2 in metal oxides has been widely studied because of its low cost, wide source, safety, no pollution and theoretical specific capacitance (SC) up to 1370F g-1 [13–14]. However, due to its weak conductivity and the diversity of its crystals, MnO2 is often massive and powdered, leading to lower actual SC of the final electrode material. Therefore, MnO2 is usually compounded with other electrode materials to improve its utilization efficiency and give full play to its advantages. For example, it can be combined with conductive polymers with high conductivity or carbon-based materials with stable structure and strong cycle stability to form composite electrode materials [15–17]. In our previous work [18], a core-shell Ppy@N-doped porous carbon nanofibers (Ppy@NPCNFs) with large specific surface area and high conductivity were fabricated by a combination of electrospinning, carbonation, nitrogen doping and chemical polymerization. However, due to the poor stability of conductive polymer Ppy, the symmetric supercapacitor prepared with the Ppy@NPCNFs electrodes displayed a low energy density (3.10 Wh kg− 1 ).
Accordingly, in order to obtain a high-performance symmetric supercapacitor with high energy density, MnO2 was grown on Ppy@NPCNFs by hydrothermal synthesis to prepare a multicore-shell ternary composite electrode material (MnO2@Ppy@NPCNFs) with excellent electrochemical performances in this work. Among the MnO2@Ppy@NPCNFs electrode, the 3D network-structured NPCNFs with larger specific surface area and high surface activity were used as the inner core layer, which provided more space for the loading of Ppy and MnO2; as the outer core layer, the loaded Ppy could not only enhance the conductivity of the inner core layer and form a highly conductive network, but also participate in the charge storage process; as the outermost shell layer, the grown MnO2 could improve the capacitance, energy density and stability of the multicore layer (Ppy@NPCNFs). Therefore, the synergistic effect between these layers led to the excellent electrochemical performance of the multicore-shell ternary composite electrode.
In addition, MnO2 generally exists in five crystal types, namely α-MnO2, β-MnO2, γ-MnO2, δ-MnO2 and λ-MnO2. Among them, α-MnO2, β-MnO2 and γ-MnO2 have a 1D tunnel structure, δ-MnO2 is a 2D lamellar structure, and γ-MnO2 is a 3D spinel structure [19–20]. Therefore, the crystal type of MnO2 has a great influence on its application in supercapacitors. In the process of MnO2 synthesis, the synthesis parameters have a marked impact on the crystal type and final performance of products. Accordingly, the response surface method (RSM) can be used to adjust and optimize the growth of the MnO2, which is a set of mathematical and statistical methods described by Box and Wilson in 1951. When the response to be studied is influenced by multiple variables, RSM can help model and analyze the process and get the best conditions to implement the required response [21–23]. Finally, the multicore-shell ternary composite electrode with the best electrochemical performance was obtained under the optimal reaction parameters, and then a symmetric supercapacitor was prepared with the multicore-shell electrode, as shown in Fig. 1.
2.1 Materials
PAN powder (MW =150,000) and PMMA (MW=97,000), were supplied by Lark Branch Co. Ltd (Beijing, China). N, N-dimethylformamide (DMF) (Analytical Reagent) was purchased from Chemical Reagent Co. Ltd (Shanghai, China). Melamine (Extra pure, ≥ 99%), Pyrrole (Extra pure, ≥ 99.9%), FeC13·6H2O (Extra pure, ≥ 99%), KMnO4 (Extra pure, ≥ 99%) and MnSO4·H2O (Extra pure, ≥ 99%) were purchased from Aladdin Biochemical Technology Co. Ltd (Shanghai, China). Syringes with capacities of 10 ml and needles with diameter of 0.7 mm and length of 32 mm were purchased from Misawa Medical Industry Co.Ltd (Shanghai, China). All materials were used without any further purification.
2.2 Preparation method of composite electrode materials
2.2.1 Preparation of core-shell composite materials
According to our previous work [18], PAN/PMMA nanofibers (NFs) with 10 wt% PMMA relative to PAN were fabricated by electrospinning, and then they were preoxidized (280°C for 8 h in air) and carbonized (800°C for 2 h in nitrogen atmosphere) to obtain porous carbon nanofibers (PCNFs). Next, PCNFs were mixed with melamine evenly in the ratio of 1:10 and placed in a tube furnace for chemical vapor deposition (800°C for 4 h in nitrogen atmosphere) to prepare NPCNFs. Afterward, Ppy was loaded on NPCNFs by chemical polymerization at 0°C to obtain Ppy@NPCNFs with a core-shell structure.
2.2.2 Preparation of multicore-shell ternary composites
MnO2 was grown on Ppy@NPCNFs by hydrothermal synthesis to form a multicore-shell ternary composite material (MnO2@Ppy@NPCNFs). The procedures were as followings: a certain amount of KMnO4 and MnSO4·H2O were weighed and configured into aqueous solutions, respectively; MnSO4·H2O aqueous solution was slowly dropped into KMnO4 aqueous solution and fully mixed; 80 ml of mixed solution was transferred to a 100 ml high-pressure reactor liner, and a certain amount of Ppy@NPCNFs was added; the reaction kettle was sealed and put into the oven preheated to a certain temperature; after a period of reaction, the reaction kettle was taken out and naturally cooled to room temperature; the multicore-shell ternary composite was taken out and repeatedly washed with deionized water to neutral.
2.3 Characterization
The morphology and structure of MnO2@Ppy@NPCNFs were acquired using a field emission scanning electron microscopy (FESEM, Regulus8100 and S4800) and a field emission transmission electron microscope (FETEM, TecnaiG2F20). X-ray diffraction (XRD, ESCALAB 250Xi) analyses were performed to elucidate the crystalline structure of MnO2@Ppy@NPCNFs with a diffraction angle 2θ range of 10°- 80°.
2.4 Electrochemical measurements
The electrochemical properties of MnO2@Ppy@NPCNFs as working electrode materials were measured using CHI660E electrochemical workstation (Shanghai Chenhua Instrument, China) in a three-electrode configuration, in which the Pt wire and saturated calomel electrode (SCE) were used as the counter electrode and reference electrode, respectively. The applied potential window of cyclic voltammetry (CV) and galvanostatic charging/discharging (GCD) curves ranged from − 1.0 V to 0 V in the 6 M KOH electrolyte. Electrochemical impedance spectroscopy (EIS) was performed at open-circuit potential with a perturbation amplitude of 5 mV in the frequency range of 100 kHz to 10 MHz. The SC was calculated according to the GCD curve [24]. A MnO2@Ppy@NPCNFs symmetric supercapacitor was assembled and tested, in which 1 mol l-1 NaSO4 was used as the electrolyte. Its energy density and power density are determined as the following equations:
$$E=\frac{1000\times {C}_{m}\times {\varDelta V}^{2}}{2\times 3600}$$
1
$$P=\frac{E\times 3600}{\varDelta t}$$
2
where \({C}_{m}\)(F g− 1), \(\varDelta V\)(V), \(\varDelta t\)(s), \(E\) (Wh kg− 1) and \(P\) (W kg− 1) are SC, discharge potential, discharge time, energy density and power density of symmetric supercapacitor, respectively.
2.5 Control variable method (single factor analysis method)
During the synthesis process of MnO2, the effects of three main reaction parameters, including the molar ratio of KMnO4 and MnSO4·H2O (1:3, 2:3, 3:3, 5:3 and 9:3), reaction time (2 h, 4 h, 6 h, 8 h and 10 h) and reaction temperature (120 ℃, 140 ℃, 160 ℃, 180 ℃ and 200 ℃), on the morphology, crystal type and electrochemical properties of the grown MnO2 were investigated by the control variable method (also single factor analysis method). Each time one of the reaction parameters was studied, the other reaction parameters remained unchanged.
2.6 Response surface methodology
Response surface methodology (RSM) is a comprehensive optimization method, which is developed by the combination of modern mathematical statistics method and parameter optimization theory. Using RSM analysis, the experimental values can be fully simulated and analyzed to establish a model, which can predict the optimal parameters and response values [25–27]. In this work, according to the results of control variable method, the RSM based on a three-factor and three-level Box-Behnken design was used to determine the optimal experimental conditions. The three factors selected in the experiment were the reaction temperature (A), molar ratio of KMnO4 to MnSO4·H2O in the reactants (B) and reaction time (C), and their actual values, and levels were shown in Tables S1. Based on the Box-Behnken design, a model was established to explore the interaction effects of these three factors and their influence on the response value (SC of MnO2@Ppy@NPCNFs), which could be approximated by the following quadratic regression model:
Y = β0 + β1A + β2B + β3C + β4AB + β5AC + β6BC + β7A2 + β8B2 + β9C2 (3)
where β0, β1, β2, β3, β4 and β5 are coefficients to be determined from the experiment data.
To analyze the statistical significance of the items in the established model for the response value, a P-value was defined. The P-value less than 0.05 meant that the item had a significant impact on the response value. The F-value described the significance of the interaction between each variable. The larger the F-value, the more significant the item was. R2 was applied to determine the fitting degree between the model and the response values obtained from experiments. The higher the value of R2, the higher the fitting degree of the model. However, when the number of independent variables in the model increases, R2 will increase, but the actual fitting degree may not increase. Therefore, the adjusted correction coefficient Adj-R2 was defined for the added variables in the model, which will only increase when the new variables change the model beyond expectations; if unnecessary variables were added, Adj-R2 will decrease [28–30].
3.1 Control variable method
3.1.1 Effects of the molar ratio of KMnO4 and MnSO4·H2O
The effects of the molar ratio of KMnO4 and MnSO4·H2O in the growth solution on the morphology, structure and electrochemical properties of MnO2@Ppy@NPCNFs were investigated. In the hydrothermal synthesis of MnO2, the reaction temperature was maintained at 160 ℃ and the reaction time was kept at 6 h. The molar ratios of KMnO4 and MnSO4·H2O were set to 1:3, 2:3, 3:3, 5:3 and 9:3, respectively.
The crystal types of MnO2 prepared using different molar ratios of KMnO4 and MnSO4·H2O were characterized by X-ray diffraction, and the XRD spectra of MnO2@Ppy@NPCNFs were shown in Fig. 2a. It could be seen that when the molar ratios were 1:3 and 2:3, the products had obvious diffraction peaks, indicating that they had good crystallinity and almost no amorphous structure. Compared with the standard card JCPDS NO.24-0735, the products belonged to the tetragonal system of β-MnO2 single-chain structure [31-32]. When the molor ratio was 3:3, there were characteristic diffraction peaks at 2θ of 12.2°, 17.9°, 28.7°, 37.3°, 41.8°, 49.6°, 59.7 ° and 69.5° in the products, which were consistent with the diffraction peaks of the standard card JCPDS NO.44-0141. Therefore, the products generated at 3:3 belonged to the tetragonal α -MnO2 system, and their structure were I4/m (87) spatial point group . When the molar ratio increased to 5:3, the products had characteristic peaks at 12.2°, 24.9°, 36.9° and 66.4°. Compared with the standard card JCPDS NO.80-1098, these four diffraction peaks corresponded to (001), (002), (201) and (021) crystal planes of δ-MnO2, respectively, illustrating the formation of δ-MnO2. The δ-MnO2 was a rhomboid crystal system with a layered structure, which belonged to potassium-containing hydrated MnO2 [33]. When the molar ratio reached 9:3, the products were still δ-MnO2, but the crystallinity of the products was lower than that of the products at 5:3.
SEM images of MnO2@Ppy@NPCNFs obtained from the growth solution with different molar ratios of KMnO4 and MnSO4·H2O were shown in Fig. 2b-f. According to XRD patterns of MnO2@Ppy@NPCNFs (Fig. 2a), Fig. 2b and c indicated that the products at 1:3 and 2:3 molar ratios were β-MnO2 rods, and they are similar in morphology and size, with lengths of 16-30 μm and widths of 2-3 μm. However, the surface of the products at 2:3 was smoother than that at 1:3. When the molar ratio increased to 3:3, the products were α-MnO2, which presented a needle-like structure with lengths of 200-900 nm and diameters of 25-40 nm, as displayed in Fig. 2d. Furthermore, as shown in Fig. 2e, the products at 5:3 belonging to δ-MnO2 appeared many flower-like nanospheres composed of a large number of nano-flakes interleaved and stacked with each other. The thickness of nano-flakes was about 10 nm, the lamellar spacing was about 4 nm, and the diameter of flower-like nanospheres was 170-240 nm. When the molar ratio reached 9:3, the products were still flower-like nanospheres, but compared with the products at 5:3, the folds and lamellar structure on the surface of nanospheres were reduced, the lamellar spacing increased to about 20 nm, and the diameter of flower-like nanospheres increased to 200-400 nm, as illustrated in Fig. 2f.
Accordingly, the above XRD and SEM characterization indicated that the molar ratio of KMnO4 and MnSO4·H2O in the growth solution had a great influence on the crystal type of MnO2 generated on Ppy@NPCNFs. When the content of oxidizer KMnO4 was higher than MnSO4·H2O, δ-MnO2 with the lamellar structure was produced. With the increased of MnSO4·H2O addition, the crystal type of MnO2 generated gradually changed from δ-MnO2 to α-MnO2, illustrating that the coexistence of δ-MnO2 required sufficient K+ to maintain the lamellae structure. With the decrease of K+, the lamellar structure of δ -MnO2 began to collapse and form α-MnO2 with [2X2] tunnel structure. Then, after the further decrease of K+, α-MnO2 transformed into β-MnO2 with [1X1] tunnel structure with higher crystallinity. Therefore, K+ could be regarded as a template for the lamellar structure and tunnel structure of MnO2 formed in the reaction process [34], and the final products were controlled by changing the K+ concentration in the reaction process [35, 36].
The electrochemical performance of MnO2@Ppy@NPCNFs prepared using different molar ratios of KMnO4 and MnSO4·H2O were displayed in Fig. 3. Fig. 3a showed the CV curves of the above products at the scanning speed of 10 mV s-1, which were approximately symmetrical rectangles, suggesting their good electrochemical reversibility and electrical double-layer capacitance (EDLC) characteristics [37-39]. Furthermore, these CV curves exhibited a broad weak redox peak, indicating that they had the characteristics of Faraday pseudocapacitance [40]. Moreover, the enclosed area of CV at 5:3 was the largest, suggesting that it had the highest SC. Fig. 3b displayed the GCD curves of each product at a constant current of 1 A g-1, which had good symmetry and presented a good linear relationship, further indicating their excellent cycling performance and electrochemical reversibility [41, 42]. According to the GCD curves of MnO2@Ppy@NPCNFs obtained from different molar ratios of KMnO4 and MnSO4·H2O at 1 A g-1 , their SC calculated was 562.12 F g-1 (1:3), 569.33 F g-1 (2:3), 574.35 F g-1 (3:3), 588.76 F g-1 (5:3) and 580.98 F g-1 (9:3), respectively.
In conclusion, the electrochemical performance of the obtained MnO2@Ppy@NPCNFs at 5:3 was the best. This was because at this time, the MnO2 loaded on Ppy@NPCNFs was flower-like nanosphere composed of δ-MnO2 with the best lamellar structure, and its surface had more folds, which was beneficial to increase the specific surface area, the contact area between electrode material and electrolyte, and the energy storage site of electrochemical reaction, thus improving its electrochemical performance. In addition, the interlayer spacing of the product δ-MnO2 at 5:3 was about 4 nm (Fig.2e), while the hydration radius of K+ was 0.33 nm, which was much smaller than the interlayer spacing of the product δ-MnO2. Therefore, the free interpenetration of K+ between the layers of δ-MnO2 could be carried out [43]. The special interlayer structure of δ-MnO2 can also form EDLC inside the δ-MnO2, which greatly increased the EDLC of MnO2@Ppy@NPCNFs, thus enhancing the overall electrochemical performance of MnO2@Ppy@NPCNFs. In contrast, α-MnO2 and β-MnO2 with tunnel structure were not conducive to the penetration of K+ in the electrolyte, resulting in relatively smaller capacitance [44].
3.1.2 Effects of the reaction temperature
According to the optimal molar ratio of KMnO4 and MnSO4·H2O determined by the above experiments, the effects of the reaction temperature on the MnO2 crystal form, morphology and electrochemical properties of MnO2@Ppy@NPCNFs were discussed. In this section, the molar ratio of KMnO4 and MnSO4·H2O in the growth solution was 5:3, and the reaction time was kept at 6 h. The reaction temperatures were set at 120 ℃, 140 ℃, 160 ℃, 180 ℃ and 200 ℃, respectively.
Fig.4a showed the XRD patterns of MnO2@Ppy@NPCNFs prepared at different reaction temperatures. It could be found when the reaction temperature was between 120-160 ℃, there were characteristic peaks at 12.2°, 24.9°, 36.9° and 66.4°, which corresponded to (001), (002), (201) and (021) crystal planes of δ-MnO2, respectively. But when the temperature was 120 ℃, the intensity of the diffraction peaks was weak, which was because the low reaction temperature led to the low nucleation activity of MnO2, resulting in incomplete crystal nucleus growth of MnO2. As the temperature increased, the characteristic diffraction peak intensity of δ-MnO2 increased gradually, indicating that its crystallinity increased. When the temperature rose to 180-200 ℃, the characteristic peaks of products illustrated that they were β-MnO2 with higher crystallinity, in which the diffraction peak intensity of the product fabricated at 200 ℃was stronger, indicating that β-MnO2 with a good crystal structure was formed.
Fig. 4b-f showed the SEM images of MnO2@Ppy@NPCNFs obtained at different reaction temperatures. It was found that the products prepared at 120-160 ℃ (Fig.4b-d) all consisted of a large number of flower-like nanospheres, which were δ-MnO2 according to the XRD results. And with the increase of temperature the diameter and lamellar spacing of nanospheres decreased. As shown in Fig.4b-d, the diameter of nanospheres was reduced from 280-450 nm (120 ℃) to 190-270 nm (140 ℃) and then to 170-240 nm (160 ℃) , and their corresponding lamellar spacings decreased from about 100 nm to about 20 nm and then to about 4 nm. When the temperature reached to 180 ℃ (Fig.4e), the products were transformed from δ-MnO2 to β-MnO2 by XRD analysis, the morphology of the products changed from nanospheres to nanorods with lengths of 0.7-1.3 μm and widths of about 300 nm, which had unstable structure and uneven surface. This might be due to the collapse of nanosphere lamellar structure caused by the higher reaction temperature. With the temperature further increased to 200 ℃ (Fig.4f), the products changed from unstable nanorods with rough surface to quite complete and stable microrods with smooth surface, and their lengths and widths were 15-30 μm and 2-3 μm, respectively. It might be because the shorter nanorods were dissolved due to the high reaction temperature, which made the micron structure uniform and stable.
Electrochemical performance
Fig. 5a showed the CV curves of products prepared at different reaction temperatures in a three-electrode system at a scanning rate of 10 mV s-1, which were similar to the CV curves in the above section, indicating that they had obvious EDLC characteristics and good electrochemical reversibility. And the MnO2@Ppy@NPCNFs obtained at 160 ℃ had the largest enclosed area of CV curve, suggesting its largest SC. Fig. 5b exhibited the GCD curves of these products at 1 A g-1, which had good linear symmetry in the potential range of -1.0-0 V (vs.SCE), further indicating their good coulombic efficiency and electrochemical reversibility. Based on these GCD curves, the SC of the products was 556.43 F g-1 (120 ℃), 578.90 F g-1 (140 ℃), 588.76 F g-1 (160 ℃), 539.12 F g-1 (180 ℃) and 511.32 F g-1 (200 ℃), respectively.
In conclusion, when the reaction temperature was 160 ℃, the electrochemical performance of MnO2@Ppy@NPCNFs was the best. This was because based on the XRD and SEM results, the flower-like δ-MnO2 nanospheres loaded on Ppy@NPCNFs at 160 ℃ had the smallest size, resulting in the smallest lamellar spacing and the largest number of nano-flakes, which provided the largest specific surface area and the largest number of active sites to promote their electrochemical performance.
3.1.3 Effects of the reaction time
Base on the previous experiment results, the effects of the reaction time on the MnO2 crystal form, morphology and electrochemical performances of MnO2@Ppy@NPCNFs were explored. Therefore, the molar ratio of KMnO4 and MnSO4·H2O was fixed at 5:3, the reaction temperature remained 160 ℃, and the reaction time was 2 h, 4 h, 6 h, 8 h and 10 h, respectively.
Combined with XRD and SEM analysis, as shown in Fig. 6, when the reaction time was 2-6 h (Fig. 6b-d), the products were all flower-like δ-MnO2 nanospheres , and the diameter and lamellar spacing of nanospheres decreased as the reaction time increased. When the reaction time was 8-10 h (Fig. 6e and f), the MnO2 crystal form of the products changed from flower-like δ-MnO2 nanospheres to needle-like α-MnO2 nanowires, and the diameter and length of nanowires decreased with the increase of reaction time.
The above experimental results were due to the spontaneous reaction of KMnO4 and MnSO4·H2O at a certain temperature and pressure during the whole reaction process, and the MnOx structural unit was first formed [45]. When the reaction time was long enough, nanospheres with large lamellar structure were gradually formed due to the concentration reaction, and their lamellar spacing was large. As the reaction time increased, the nanosheets gradually gathered and the lamellar spacing decreased, resulting in the formation of flower-like nanospheres. With the further increase of reaction time, the flower-like nanosphere size continued to decrease. Subsequently, in order to maintain the stable structure of the products and reduce the surface free energy, the lamellar structure of the flower-like nanospheres gradually collapsed and shrank, and then the nanowire structure was formed.
Electrochemical performance
Fig. 7 showed the CV (a) and GCD (b) curves of the products obtained after different reaction times. It could be found these products had good EDLC characteristics, electrochemical reversibility and coulombic efficiency, and MnO2@Ppy@NPCNFs prepared after 6 h had the largest enclosed area of CV curve, indicating its largest SC. In addition, according to their GCD curves, the SC of MnO2@Ppy@NPCNFs was 574.23 F g-1 (2 h), 581.98 F g-1 (4 h), 588.76 F g-1 (6 h), 558.32 F g-1 (8 h) and 543.35 F g-1 (10 h), respectively. Therefore, the electrochemical performance of MnO2@Ppy@NPCNFs obtained after 6 h was the best because of their flower-like δ-MnO2 nanospheres with the smallest size. In conclusion, when the molar ratio of KMnO4 and MnSO4·H2O was 5:3, the reaction temperature was 160 ℃ and the reaction time was 6 h, the electrochemical performance of MnO2@Ppy@NPCNFs obtained was the best and its SC was the largest .
3.2 Response surface methodology and experimental verification
Based on the above results of control variable method, the actual values of factors and their corresponding response values were shown in Table S2. The synthesis process of MnO2 was designed by Design-Expert software to study the effects of reaction temperature, molar ratio and reaction time on the electrochemical performance of MnO2@Ppy@NPCNFs, and a statistical model was established according to Eq. (3). Subsequently, the established model was applied to predict the optimal reaction parameters, and the experimental verification was carried out, thus obtaining the MnO2@Ppy@NPCNFs with the best electrochemical performance.
3.2.1 Statistical analysis
Analysis of variance was performed on the experimental data in Table S2, and the analysis results were shown in Table S3. According to the analysis results and Eq. (3), a quadratic regression model of the SC (Y) against the reaction temperature (A), the molar ratio of reactants (B) and the reaction time (C) could be written as follows:
Y = 563.83-17.40A+ 15.33B-6.58C+4.92AB-4.87AC-2.19BC-14.23A2-24.44B2-17.24C2 (4)
It could be seen from Table S3 that the F-value of this model was 82.82 and its P-value<0.0001, indicating an extremely significant level. P-value of the lack of fit greater than 0.05 showed the difference of the term was not significant, indicating that the model had a high fitting degree with the actual experimental values and was highly reliable. Moreover, the R2 value of 0.9907 and the Adj-R2 value of 0.9787 further exhibited this conclusion. In addition, it was found from Fig. S1 that there was a linear correlation between the predicted values obtained by the model and the experiment data, and the fitting degree between them was high.
3.2.2 Response surface analysis
The regression function can usually be represented by contour and 3D response surface plots, which is mainly used to characterize the interaction type between the two variables and the relationship between the response of each variable and experimental level. The elliptical contour shows that the interaction between variables is obvious, while the circular contour is the opposite [46,47]. In this experiment, the influences of three reaction parameters on the SC of MnO2@Ppy@NPCNFs were shown in Fig. S2. It could be found that all 3D response surface plots indicated that with the increase of each single reaction parameter, the SC of MnO2@Ppy@NPCNFs increased first and then decreased. In addition, the elliptical contour plots in Fig. S2 (a and b) illustrated that the interaction between the reaction temperature and molar ratio was significant when the reaction time remained constant, and there was a significant interaction between the reaction time and reaction temperature when the molar ratio was kept constant. Fig. S2c showed when the reactant temperature was kept constant, the interaction between the molar ratio and reaction time was not significant due to the circular contour plot. The results of response surface analysis were consistent with the regression equation results in Table S2, and there was an extreme value of SC in Fig. S2, indicated that the parameter range of the experiment was reasonable and the designed model had strong reliability.
3.2.3 Prediction of optimal reaction parameters
According to the established regression model (Eq. (4)), the optimal reaction parameters were obtained: the molar ratio of KMnO4 to MnSO·H2O was 1.93, the reaction temperature was 140.64 ℃, and the reaction time was 6.26 h. It was predicted that the SC of MnO2@Ppy@NPCNFs obtained under these optimal reaction parameters was 593.809 F g-1.
3.2.4 Experimental verification
Three groups of repeated experiments were carried out to verify the optimal reaction parameters predicted by the model. The GCD curves at 1 A g-1 of MnO2@Ppy@NPCNFs (1.93/140.64℃/6.26h) obtained by three verification experiments were shown in Fig. 8a. Based on these GCD curves, the SCs of the three experimental products calculated were 597.21 F g-1, 590.45 F g-1, and 596.65 F g-1, respectively, and the average SC was 595.77 F g-1. The error between the average experimental value and the theoretical value was very small, which indicated that the optimization results were reliable, and the optimal reaction parameters obtained had practical application value.
The MnO2@Ppy@NPCNFs (1.93/140.6℃/6.26h) was characterized by XRD. Compared with the sample with the largest SC obtained in the previous experiments (the molar ratio of 5:3 (1.67), the reaction temperature of 160 ℃, the reaction time of 6 h), it had a similar diffraction pattern. As shown in Fig. 8b, their characteristic diffraction peaks corresponded to (001), (002), (201) and (021) crystal planes, indicating that the outmost layer of MnO2@Ppy@NPCNFs (1.93/140.64℃/6.26h) was also δ-MnO2. According to its SEM images, the MnO2@Ppy@NPCNFs (1.93/140.64/6.26h) showed an obvious characteristic morphology of δ-MnO2, which were flower-like nanospheres with diameters of 70-180 nm and lamellar spacing of about 2.5 nm (Fig. 9a. And the product had an obvious multicore-shell ternary structure, as indicated in Fig. 9b, in which NPCNFs with the fiber diameter of about 280 nm served as the core layer, Ppy was used as the intermediate layer with a thickness of about 50 nm, and the loaded δ-MnO2 was the outermost shell layer with a thickness of about 100 nm. The innermost NPCNFs provided porous structure and conductive skeleton, and the intermediate layer Ppy enhanced the conductivity of the material. The outermost MnO2 not only improved the disadvantages of low life and unstable performance of the middle layer Ppy, but also provided more active sites. Therefore, the final product MnO2@Ppy@NPCNFs would have excellent electrochemical performances under the synergistic action of the three internal materials. As shown in (Fig. 9c), the lattice fringes with the spacing of 0.35 nm and 0.71 nm could be found in the HR-TEM image of MnO2@Ppy@NPCNFs, which were indexed as the (002) and (001) planes of δ-MnO2, respectively. The elemental distribution of MnO2@Ppy@NPCNFs was further verified by the elemental mapping image (Fig. 9d, where O was distributed in the whole fiber, C and N were evenly distributed in the inner layer of the fiber, while Mn was only distributed in the outermost layer of the fiber, indicating the multilayer structure of the fiber and the outermost layer was MnO2.
Fig. 10 showed the comparison of electrochemical properties of the three electrode materials, including MnO2@Ppy@NPCNFs (1.93/140.64℃/6.26h), MnO2@Ppy@NPCNFs (1.67/160.00℃/6.00h) and Ppy@NPCNFs). According to the CV curves (Fig. 10a) and GCD curves (Fig. 10b), it could be concluded that the three products all had good EDLC characteristics and electrochemical reversibility, and their CV curves all displayed a broad and weak redox peak at 10 mV s-1, especially MnO2@Ppy@NPCNFs, indicating that they had the characteristics of Faraday pseudocapacitance. In addition, according to the GCD curves, the SC of MnO2@Ppy@NPCNFs (1.93/140.64℃/6.26h) (596.65 F g-1) was 74.2% higher than that of Ppy@NPCNFs (342.13 F g-1). It was mainly due to the excellent capacitive characteristics of a large number of flower-like nanospherical δ-MnO2 loaded on the Ppy@NPCNFs surface, which mdae the electrode material have a large number of active sites and larger specific surface area, thus promoting the adsorption and desorption of ions in the electrolyte on the material surface.
Meanwhile, the electrochemical impedance spectra of the three products (Fig. 10c) showed that the semicircle diameter of MnO2@Ppy@NPCNFs (1.93/140.64℃/6.26h) in the high-frequency region was the smallest, and its slope of the inclined line in the low-frequency region was the largest, illustrating that it has the smallest ion transfer resistance and charge diffusion resistance, which made the electrode material had better capacitance performance [48,49]. The SC attenuation and coulomb efficiency variation of MnO2@Ppy@NPCNFs (1.93/140.64℃/6.26h) after 1000 cycles of GCD at 1 A g-1 (Fig. 10d) displayed that its SC retention rate reached 96.2% and its coulomb efficiency was stable at 91±3%, which were higher than those of Ppy@NPCNFs (95.4% and 86±2%) [18]. The main reason was that the metal oxide MnO2 made up for the structural instability of conductive polymer Ppy in a long time cycle. All of the above indicated that MnO2@Ppy@NPCNFs had excellent electrochemical performances, and its electrochemical performances were superior to most of the MnO2-loaded carbon materials previously reported (Table S4). According, it was an ideal choice for supercapacitor electrode materials.
The capacitance contribution of materials mainly includes two parts: EDLC (Cd) and pseudocapacitance (Cp), in which Cd is attributed to the contribution of specific surface area and inherent structural defects with micropores, while Cp is derived from the contribution of active sites. The total capacitance contribution of materials will be the sum of Cd and Cp. Trassatti method can be performed to investigate the capacitive storage behavior of electrode materials , and the total voltammetric charge (q) measured can be expressed as follows [50]:
where k1 and k2 are constants, v is a scan rate, q∞ is the storage charge at v→∞, which is equivalent to Cd, and q0 is the storage charge at v = 0.
The variation curve of specific capacitance for MnO2@Ppy@NPCNFs was determined through GCD curves at different current densities (Fig. S3), and its specific capacitance of MnO2@Ppy@NPCNFs (595.53 F g− 1 at 1 A g-1) was much higher than other samples [18], as shown in Fig. 11d. Then, according to the CV curves of the four samples at different scan rates (Fig. S4), their specific capacitances were determined by formulas (5) and (6), respectively, as shown in Fig. 11a-c. It could be seen that the MnO2@Ppy@NPCNFs electrode displayed higher Cd and Cp values than other samples, especially Cp value, illustrating that the large number of active sites provided by MnO2 on its outmost layer could significantly enhance the charge storage efficiency through the pseudocapacitance mechanism.
Consequently, a symmetric supercapacitor was assembled using MnO2@Ppy@NPCNFs as electrodes for evaluating its practicability. The CV and GCD curves of the symmetric supercapacitor measured in a two-electrode configuration were exhibited in Fig. 12a and b, respectively. Its energy density and power density (Fig. 12c) were calculated from the GCD curve according to Eqs. (1 and 2), and illustrated an energy density of 9.36 Wh kg−1 at a power density of 1000 W kg−1, which were significantly higher than those of the Ppy@NPCNFs-based symmetric supercapacitor (3.10 Wh kg−1 at 50 W kg−1) [18]. In addition, the retention rate of specific capacitance could reach 93.75% after 1000 cycles at 10 A g−1 (Fig.12d). The electrochemical performance comparison of MnO2@Ppy@NPCNFs-based symmetric supercapacitor with the MnO2-loaded supercapacitors previously reported was shown in Table S5.
In conclusion, the MnO2@Ppy@NPCNFs ternary composite electrode material with a multicore-shell prepared by simple hydrothermal synthesis reaction had excellent electrochemical properties, in which the multifunctional Ppy nanocoating was evenly formed on the 3D NPCNF skeleton, and then a large number of flower-like nanospherical δ-MnO2 were grown on the outermost layer, thus jointly constructing the multicore-shell hierarchical structure. This unique structure enables the ternary nanocomposites to exhibit excellent electrochemical properties far superior to individual components, which is due to the synergistic effect of flower-like nanospherical δ-MnO2 layer, highly conductive Ppy layer and interwoven NPCNF conductive networks.
Using the control variable method and the Box-Behnken experimental design model in RSM, the interaction between the reaction parameters for the growth of MnO2 was explored in detail. On this basis, the optimal reaction parameters were predicted by the established model and then verified by experiments. Under the optimum reaction parameters, MnO2@Ppy@NPCNFs (1.93/140.64℃/6.26h) had the best electrochemical performance, that was, its SC was the largest (595.77 F g-1), and its SC retention rate and coulomb efficiency were 96.2% and 91 ± 3% after 1000 cycles, respectively, indicating that it had good rate performance and long-term cycling stability. Subsquently, a symmetric supercapacitor assembled using MnO2@Ppy@NPCNFs as electrodes displayed an energy density of 9.36 Wh kg− 1 at a power density of 1000 W kg− 1and a retention rate of 93.75% after 1000 cycles. The preparation method of the multicore-shell electrode materials took 3D NPCNFs as the core substrate, and then the different multi-functional components were uniformly coated on the substrate layer by layer, thus constructing multicore-shell electrode composites for high-performance supercapacitors.
Author contribution
All authors contributed to the study conception and design. Lan Xu contributed to the study conception and design. Material preparation, data collection and analysis were performed by Yi Wang, Jie Wang and Dong Wei. The first draft of the manuscript was written by Yi Wang. All authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Funding
This study was funded by the National Natural Science Foundation of China (Grant No.11672198), Jiangsu Higher Education Institutions of China (Grant No.20KJA130001), and PAPD (A Project Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions) .
- Mohapatra D, Parida S, Badrayyana S, et al (2017) High performance flexible asymmetric CNO-ZnO//ZnO supercapacitor with an operating voltage of 1.8 V in aqueous medium. Appl Mater Today, 7: 212-21.
- Sevilla M, Mokaya R (2014) Energy storage applications of activated carbons: supercapacitors and hydrogen storage . Energy Environ Sci 7(4): 1250-80.
- Kumar N, Yu YC, Lu YH, et al (2016) Fabrication of carbon nanotube/cobalt oxide nanocomposites via electrophoretic deposition for supercapacitor electrodes. J Mater Sci 51(5): 2320-9.
- Lv W, Li Z, Deng Y, et al (2016) Graphene-based materials for electrochemical energy storage devices: Opportunities and challenges. Energy Storage Materi 2: 107-38.
- Adams K, A F González, Mallows J, et al (2019) Facile synthesis and characterization of Bi13S18I2 films as a stable supercapacitor electrode material. J Mater Chem A 7(4): 1638-46.
- Shayeh J S, Ehsani A, Nikkar A, et al (2015) Physioelectrochemical investigation of supercapacitive performance of ternary nanocomposite by common electrochemical methods and fast fourier transform voltammetry. New J Chem 39(12): 9454-60.
- Hao J, Lee P S, Li CD (2012) Carbon based nanostructures for advanced supercapacitors. Energ Environ Sci 6(1): 41-53.
- Lang X, Hirata A, Fujita T, et al (2011) Nanoporous metal/oxide hybrid electrodes for electrochemical supercapacitors. Nat Nanotechnol 6(4): 232-6.
- Ranganatha S, Munichandraiah N (2018) γ-MnS nanoparticles anchored reduced graphene oxide: Electrode materials for high performance Supercapacitors. J Sci-Adv Mater Dev 359-65.
- Wei, W, Cui, et al (2012) Manganese oxide-based materials as electrochemical supercapacitor electrodes. Chem soc rev 40(3): 1697-721.
- Hui P, Li J, Yuan PF (2010) Carbon nanotubes for supercapacitor. Nanoscale Res Lett 5(3): 654-68.
- Ran L, Duay J, Sang BL (2011) Heterogeneous nanostructured electrode materials for electrochemical energy storage. Chem Commun 47(5): 1384-404.
- Zhu J, Zhang D, Zhu Z, et al (2021) Review and prospect of MnO2-based composite materials for supercapacitor electrodes. Ionics 27(9): 3699-714.
- Zhao Q, Song A, Ding S, Qin R, Cui Y, Li S, Pan F (2020) Preintercalation strategy in manganese oxides for electrochemical energy storage: review and prospects. Adv Mater 32(50): 2002450.
- Liu P, Zhu Y, Gao X, et al (2018) Rational construction of bowl-like MnO2 nanosheets with excellent electrochemical performance for supercapacitor electrodes. Chem Eng J 350: 79-88.
- Jafri R, Mishra A, Ramaprabhu S (2011) Polyaniline–MnO2 nanotube hybrid nanocomposite as supercapacitor electrode material in acidic electrolyte. J Mater Chem 21(4): 17601-5.
- Kim JH, Lee KH, Overzet LJ, Lee GS (2011) Synthesis and electrochemical properties of spin-capable carbon nanotube sheet/MnOx composites for high-performance energy storage devices. Nano Lett 11(7): 2611–7.
- Gao L, Wang Y, Liu Y , Xu L (2023) Core-shell Ppy@N-doped porous carbon nanofiber-based electrodes for high-property supercapacitors. Colloids Surf Physicochem Eng Aspects 366: 131056.
- Brock SL, Duan N, Tian ZR, Giraldo O, Zhou H, Suib S (1998) A review of porous manganese oxide materials. Chem Mater 10: 2619–28.
- Shi K, Zhitomirsky I (2013) Polypyrrole nanofiber–carbon nanotube electrodes for supercapacitors with high mass loading obtained using an organic dye as a co-dispersant. J Mater Chem A 1(38): 11614-22.
- Maba B, Res A, Epo A, et al (2008) Response surface methodology (RSM) as a tool for optimization in analytical chemistry. Talanta 76(5): 965-77.
- Hill WJ, Hunter WG (1966) A review of response surface methodology: A literature survey. Technometrics 8(4): 571-90.
- Pezeshki-Modaress M, Zandi M, Mirzadeh H (2015) Fabrication of Gelatin/Chitosan Nanofibrous Scaffold: Process Optimization and Empirical Modeling. Polym Int 64(4): 571-80.
- Fang C, Zhang D (2020) A large areal capacitance structural supercapacitor with a 3D rGO@MnO2 foam electrode and polyacrylic acid–Portland cement–KOH electrolyte. J Mater Chem A 8(25): 12586-93.
- Shirvan KM, Mamourian M, Mirzakhanlari S & Ellahi R (2017) Numerical investigation of heat exchanger effectiveness in a double pipe heat exchanger filled with nanofluid: a sensitivity analysis by response surface methodology. Powder Technol (313): 99-111.
- Abdulgader M, Yu J, Zinatizadeh AA, Williams P & Zahra R (2019) Process analysis and optimization of single stage flexible fibre biofilm reactor treating milk processing industrial wastewater using response surface methodology (RSM). Chem Eng Res Des (7): 149169-81.
- Su D, Ahn H J, Wang G (2013) Hydrothermal synthesis of α-MnO2 and β-MnO2 nanorods as high capacity cathode materials for sodium ion batteries. J Mater Chem A 1(15): 4845-50.
- Fasuan TO, Akanbi CT, Betiku E (2018) Acetylation of amaranthus viridis starch: modeling and process parameters optimization. Food Sci Nutr 6: 1287-97.
- Tang H, Fan S, Li Y, Dong S (2019) Amylose: acetylation, optimization, and characterization. J Food Sci 84: 738-45.
- Priya DD, Nandhakumar M, Shanavas S, et al (2021) Aerva lanata-mediated bio-treated production of copper oxide nanoparticles, optimization by BBD–RSM method and it behaviour against water related mosquito. Appl Nanosci 11: 207-16.
- Hill LI, H Arrivé, Guyomard D (2002) Effect of synthesis conditions on the morphology of MnO2 (α-, β-, γ-) synthesized by the hydrothermal-electrochemical method. Ionics 8(3): 161-71.
- Yang Y, and C Huang (2021) Effect of synthetical conditions, morphology, and crystallographic structure of MnO2 on its electrochemical behavior. J Solid State Electr 14(7): 1293-301.
- Zhu G, Deng L, Wang J, et al (2012) The simple preparation of birnessite-type manganese oxide with flower-like microsphere morphology and its remarkable capacity retention. Mater Res Bull 47 (11): 3533-7.
- Hashemzadeh F, Mehdi Kashani Motlagh M, Maghsoudipour A (2009) A comparative study of hydrothermal and sol–gel methods in the synthesis of MnO2 nanostructures. J Sol-Gel Sci Techn 51(2): 169-74.
- Yin B, Zhang S, He J, et al (2015) Phase-controlled synthesis of polymorphic MnO2 structures for electrochemical energy storage. J Mater Chem A 3(10): 5722-9.
- Wu Y, Lin G, Ma Z, et al (2014) Structural and morphological evolution of mesoporous α-MnO2 and β-MnO2 materials synthesized via different routes through KMnO4/H2C2O4 reaction. Mater Lett 125(jun 15): 158-61.
- Li P, Yang Y, Shi E, et al. Core-double-shell carbon nanotube@polypyrrole@MnO(2) sponge as freestanding compressible supercapacitor electrode. ACS Appl Mater. Interfaces 6: 5228-5234.
- Lei ZB, Shi FH, Lu L (2012) Incorporation of MnO2-Coated carbon nanotubes between graphene sheets as supercapacitor electrode. ACS Appl Mater Inter 4: 1058-64．
- Liu C, Wang J, Li JS, Zeng ML, Luo R, Shen JY, Sun XY, Han WQ, Wang LJ (2016) Synthesis of N-doped hollow-structured mesoporous carbon nanospheres for high-performance supercapacitors, ACS Appl Mater. Interfaces 8: 7194-204.
- Chengwei Y, Lan X (2021) Heteroatom-doped porous carbon derived from zeolite imidazole framework/polymer core-shell fibers as an electrode material for supercapacitor. Compos B Eng 225: 109256.
- Qu Q, Zhang P, Wang B, et al (2009) Electrochemical performance of MnO2 nanorods in neutral aqueous electrolytes as a cathode for asymmetric supercapacitors. J Phys Chem C 113(31): 14020-27.
- Jang J, Bae J, Choi M, Yoon SH (2005) Fabrication and characterization of polyaniline coated carbon nanofiber for supercapacitor. Carbon 43: 2730-6.
- E R Nightingale Jr (1958) Phenomenological theory of ion solvation. Effective radii of hydrated ions. J Phys Chem 63(9): 1381-7.
- Devaraj S, Munichan (2008) draiah N. Effect of crystallographic structure of MnO2 on its electrochemical capacitance properties . J Phys Chem C 112(11): 4406-17.
- Yong QG, Zheng HW, Jun XW (2005) A facile route to synthesize uniform single-crystalline α-MnO2 nanowires. J Cryst Growth 279: 415-9.
- Asadi N, Taran M, Rad M, Alavi M (2019) Effects of glucose, metformin, and protein on formation of flower-like nanocomposites of struvite in infected artificial urine medium by methicillin-resistant Staphylococcus aureus (MRSA): new report. Nano Biomed Eng 11(1): 91-7.
- Zhang Z, Li Q, Guo B, et al (2020) Optimization of process parameters for preparation of polystyrene PM2.5 particles by supercritical antisolvent method using BBD-RSM. Sci Rep-UK 10(1): 11187.
- Hou L-M, Wen Z, Li Y-X, Hu H-Y, Kan Y-H, Su Z-M (2015) Molecular design of indolizine derivative as sensitizers for organic dye-sensitized solar cells . Acta Phys-Chim Sin 31(8): 1504-12.
- Yoo HD, Jang JH, Ryu JH (2014) Impedance analysis of porous carbon electrodes to predict rate capability of electric double-layer capacitors. J Power Sources 267: 411-20.
- Ardizzone S, Fregonara G, and Trasatti S (1990) “Inner” and “outer” active surface of RuO2 electrodes. Electrochim Acta 35(1): 263-7.
No competing interests reported.