Unravelling the effect of crystal lattice compression on the supercapacitive performance of hydrothermally grown nanostructured hollandite α-MnO₂ induced by incremental growth time

Abstract

Manganese oxide (α-MnO₂) nanoparticles are highly recognised for their use in supercapacitor applications. This study demonstrates the successful synthesis of flower-like and nanorods hollandite α-MnO₂ by a simple one-pot hydrothermal technique at various reaction times. The synthesised nanoparticles were characterised by various physicochemical and electrochemical characterisation techniques. The influence of the various reaction times on the structural and morphological properties was evaluated by X-ray diffraction (XRD) and scanning electron microscope. XRD patterns revealed that the synthesized MnO₂ nanoparticles are tetragonal structures with crystallite sizes ranging from 13.69 to 20.37 nm estimated from the Williamson–Hall method. Moreover, the functional groups and surface area were examined by Fourier transform infrared spectroscopy and Bruner–Emmert–Teller, respectively. Furthermore, the compositional elements were studied by X-ray photoemission spectroscopy and energy-dispersive X-ray spectroscopy. Finally, the electrochemical performances were studied using cyclic voltammetry, galvanostatic charge–discharge and electrochemical impedance spectroscopy (EIS). The GCD characteristics revealed that the optimised α-MnO₂ has a good capacitive behaviour, which predicts the potential application in energy storage. Electrochemical studies revealed that the 3 h-MnO₂ sample exhibited a superior electrochemical behaviour and demonstrated a high specific capacitance of 132 F/g at a current density of 1A/g.

Introduction

Energy issues are particularly prominent in the development of modern civilization. Most energy is generated from fossil fuel resources, but blindly using fossil fuel resources would not only cause overheating of the Earth's greenhouse gases but also produce harmful gases that pollute the environment. To address these issues, it is necessary to harvest sustainable energy and develop new technologies to improve energy storage and conversion^1–3 and supercapacitors are one of the solutions. In recent years, electrochemical supercapacitors (ultracapacitors) have attracted researchers in the field of energy storage. They are low in maintenance, high energy density charging storage devices with extended cycle life and rapid charging. They bridge the gap between high-energy batteries and high-power electrolytic capacitors¹. In general, supercapacitors have found applications where large power is required for a short period of time, such as break-in systems of electric and hybrid vehicles, emergency shutdown, and power stabilisers in low-power devices^2,3. Supercapacitors are classified into double-layer electric capacitors (EDLCs) and Pseudocapacitors (PCs). EDLCs are where charges are stored electrostatically by the separation of charges, whereas PCs are where charges are stored faradaically with electrochemical redox reactions, electrosorption, and intercalation of ions or atoms between the electrode and the electrolyte^4,5. Transition-metal oxides are considered ideal pseudocapacitor electrode materials because they exhibit a high intrinsic pseudocapacitive behaviour in fast reversible surface redox reactions⁶. Of the different metal oxides, MnO₂ has emerged as a promising electrode material for supercapacitors, due to various fascinating properties such as reversible electrochemical reactions, abundance in nature, affordability, and environmental compatibility. Manganese oxide (MnO₂) is one example of a pseudocapacitive material^7–9.

The understanding that a material's capabilities are determined by the size and shape of its nanoparticles (NPs) has led researchers to focus on creating a variety of MnO₂ nanostructured materials under controlled conditions to regulate the morphology of NPs^10,11. MnO₂ exists in various polymorphic forms, such as α, β, λ, and ε phases, which are different in terms of how the basic units of the octahedral MnO₆ are linked^12,13. Among these, α-MnO₂ demonstrated decent electrochemical performance compared to its counterpart polymorphs due to its large 2 × 2 tunnel structure and a large surface area¹⁴. Since the electrode properties are influenced by particle size, morphology, strain and synthesis method, several efforts to prepare nanostructured electrodes are reported¹³. In general, the morphology, surface area, porosity, and crystalline structure of MnO₂ have a significant influence on its specific capacitance¹⁵.

MnO₂ can be synthesized by different synthesis methods such as chemical coprecipitation, sol–gel approach, electrochemical deposition, hydrothermal method, combustion method and so^16–19. Of these methods, the hydrothermal technique is one of the simplest and cost-effective methods of obtaining the desired shape and structure of MnO₂¹². Feng et al., have reported the one-pot hydrothermal synthesis of α-MnO₂ nanorods with a diameter of 300 ± 200 nm and a length of up to 1.2 ± 0.2 μm²⁰. Shan et al. have reported on the simple hydrothermal synthesis of α-MnO₂ nanowires²¹. Shen et al., reported on the synthesis of flower-like MnO₂ nanostructures²². Herein, we report a simple hydrothermal method for the synthesis of α-MnO₂. The ultimate intent is to investigate different reaction times and correlate its effect on the physicochemical and electrochemical performance of the α-MnO₂ nanoparticles. To optimise the reaction time, the temperature was kept constant at 140 °C and the reaction time was varied between 3, 9, and 15 h. Furthermore, the pseudocapacitive performance of the samples was studied using Na₂SO₄ electrolyte. In this study, the electrochemical results agree well with the structural results which revealed the compression of the unit cell volume seen by the reduction of lattice parameters a = b and c of the crystal lattice due to longer reaction time. It should be noted that the observed physical compression of the crystal lattice causes the reduction in surface area as confirmed by BET study. Thus, the reduction of the specific capacitance which decreases with reaction time was found to be dependent on the reduction in the surface area on the one hand which is also the result of the compression of the crystal lattice. Finally, the present study revealed a considerable energy saving during the synthesis of MnO₂ while maintaining the electrochemical performance at a reasonable level relative to previous published work. Hence, our study demonstrated the critical aspect of controlling unit cell volume electrode/electrolyte interface coupling effects when lattice defects are involved in the synthesis of electrode materials (Supplementary Fig. S1).

Results and discussion

Structural analysis

To determine the crystal structure and phase of the prepared MnO₂ nanoparticles, XRD measurements were conducted. Figure 1a shows the XRD pattern of the as-synthesized MnO₂ nanoparticles prepared at different reaction times. All samples exhibited high crystallinity, with no additional peaks detected and this indicated a good crystallization of MnO₂ by the hydrothermal method. Compared to the standard PDF, the diffraction peaks can be indexed to the α-MnO₂ phase which presents a hollandite-type structure (tetragonal with space group I4/m symmetry) and is well matched with the PDF card no. 44-0141. Moreover, the reaction time did not significantly alter the hollandite polymorph as confirmed through the same preferred growth orientation. Figure 1b shows an enlarged diffraction peak (211), clearly showing a slight shift to higher angles of the diffraction peak 2theta as the reaction time increases. The shift of diffraction peaks to higher angles shows that the prepared samples are under compressive stress that naturally lead to dense crystal lattice of α-MnO₂ nanoparticles as the growth time increased^19,20. Interestingly, the observed compression in all (h,k,l) directions resulted in the volume of the unit cell decreased with increased reaction time while all samples were grown in an autoclave in which comparable temperature and pressure conditions prevailed.

Figure 1.

(a) XRD patterns of α-MnO₂ with various reaction time and (b) peak shift at 211.

The average crystallite sizes were calculated using the Scherer equation (Eq. 1)²³.

$$D=\frac{K\lambda }{\beta cos\theta }$$ 1

where λ is the wavelength of the CuKα X-ray sources (1.54 nm), K is the shape factor (0.9), β is the full width at the half-maximum intensity (FWHM), D is the average crystallite size and θ is the diffraction angle. In another manner, to ascertain the micro strain and average crystallite size, we used the Williamson–Hall (W–H) technique and compared it with the Debye–Scherrer method. The micro strain (ε) and the average crystallite size of the W–H method are given in Eq. (2)²⁴.

$$\beta cos\theta =\epsilon (4sin\theta )+\frac{K\lambda }{D}$$ 2

where, λ, ε, D, β, and K are the wavelength, strain, crystallite size, full-width at half maximum, and shape factor, respectively. It must be noted that Eq. (2) represents a straight-line graph. The straight line is found by plotting βcosθ (y-axis) versus 4sinθ (x-axis). The crystallite size value was determined from the y-intercept which is $\frac{K\lambda }{D}$ and the value of strain was determined from the slope (gradient) of the graph. Plots showing the relationship between βcosθ and 4sinθ are drawn for all samples as shown in Fig. 2. Crystallite size values and strain for all samples extracted from the linear fit data are shown in Table 2. The results show that both methods have similar crystallite sizes while showing a slight size difference.

Figure 2.

Williamson Hall plots of MnO₂ (a) 3 h, (b) 9 h and (c) 15 h.

Table 2.

Surface area, pore volume and pore sizes of α-MnO₂ at different reaction time.

Sample name	Surface area (m2/g)	Pore volume (cm3/g)	Pore sizes (nm)
3 h	49.2124	0.0823	13.88
9 h	34.8546	0.0636	12.50
15 h	22.8398	0.0324	9.98

The lattice parameters a = b and c were determined using Eqs. (3) and (4)^25,26

$$2d_{\mathit{hkl}}sin\theta =n\lambda$$ 3

$$\frac{1}{d_{\mathit{hkl}}^2}=\frac{h^2+k^2}{a^2}+\frac{l^2}{c^2}$$ 4

where λ is the wavelength of the X-ray source, n is the order of diffraction (for first order n = 1), $d_{\mathit{hkl}}$ is the interplanar spacing that is calculated from Eq. (4) and hkl is the Miller indices. The a = b and c are the lattice parameters of the unit cell. The estimated lattice parameters for all samples were found to be close to the reported values of bulk MnO₂ (PDF card No. 44-0141). The values of lattice parameters are depicted in Table 1 and were found to decrease with reaction time, which may be attributed to samples being under compressive strain along the a-axis. The volume, and dislocation density of the prepared MnO₂ nanopowder samples are also computed and shown Table 1 in following equations^24,27,28

$$V=a^{2}c$$ 5

$$\delta =\frac{1}{D^2}$$ 6

Table 1.

Lattice parameters, average crystallite sizes, volume, dislocation density and strain of the samples prepared at different reaction time.

Sample name	a = b (Å)	c ( Å)	D (nm)		V ( Å)3	δ (× 10−3)	Strain (× 10−3)
			Scherer	W–H plots
3 h	9.87	2.86	17.07	13.69	278.61	3.43	− 0.995
9 h	9.86	2.85	23.66	22.95	277.08	1.79	0.267
15 h	9.61	2.85	20.37	17.03	263.20	2.41	0.268

SEM and EDS analysis

The morphological characteristic aspect of the MnO₂ nanoparticles is depicted by the SEM images in Fig. 3a–f. The SEM images are shown in a low and high magnifications respectively. Figure 3a, b shows the formation of nano-flower-like MnO₂ nanoparticles with few nanorods on the surface at a reaction time of 3 h. Figure 2c–f shows that the flower-shaped morphology tends to collapse and leads to more individual nanorod particles, showing a wider loosened surface at a 6 h reaction time. At a subsequent increase in the reaction time of 15 h as shown in Fig. 3e,f, more agglomerated surfaces of the nanorods were observed. The transition from nano-flower to nanorod appearance is due to the slow growth kinetics of nanorods under stable conditions, which are linked to the progressive dissolution and recrystallization processes⁸. The SEM images show a strong correlation between the yielded samples and the reaction time. More interestingly, the calculated aspect ratio (SI) were found to be 13.94, 14.02 and 16.57 for 3 h, 9 h and 15 h samples, respectively, which relatively agreed with XRD crystallite size results.

Figure 3.

SEM images of (a,b) 3 h, (c,d) 9 h and (e,f) 15 h at low and high magnifications, separately.

The elemental composition of the prepared samples was determined by EDS analysis. Figure 4 shows the EDS evaluation of the samples that confirms the existence of Mn, O, C, and K within the material. The presence of K in the spectrum is due to the potassium permanganate (KMnO₄) chemical utilized during the synthesis, indicating that the samples were not completely washed to get rid of K. Furthermore, during sample preparation for EDS analysis, a carbon coating to minimize sample charging was applied, which shows that the C present in the material is due to the coating.

Figure 4.

EDS spectrum of (a) 3 h, (b) 9 h and (c) 15 h.

FTIR analysis

Fourier Transform Infrared Spectroscopy (FTIR) was used for the analysis of functional group present in the samples. FTIR spectroscopy allows the identification of both organic and inorganic components in the sample. An infrared spectrometer creates an infrared absorption spectrum by measuring the light absorbed at different frequencies by the bonds between distinct components. The FTIR spectroscopy was carried out over a spectral range of 400 cm⁻¹ to 4000 cm⁻¹. Figure 5 shows the FTIR spectra of the various samples, which reveal that the intensities of surface functional groups are decreasing with increasing reaction time. Among the observed functional groups, the broad peak located at 3444 cm⁻¹ was assigned to O–H stretching vibration is due to the hydroxyl group. Moreover, the bands located approximately at 1653, 1384, and 1105 cm⁻¹ may be assigned to the bending vibrations of O–H combined with Mn atoms²⁹. Furthermore, the peaks located around 518 and 592 cm⁻¹ can be ascribed to the metal–oxygen (Mn–O) bending vibrations of [MnO₆] octahedral in α-MnO₂³⁰. Finally, the thorough observation of FTIR intensities revealed a decreasing trend of all major peaks relative to the reaction time increase, which confirmed unit cell compression as observed in XRD.

Figure 5.

FTIR spectra of MnO₂ at different reaction time.

BET analysis

The Brunauer–Emmett–Teller (BET) which is physical gas adsorption is the best technique for examining the porosity characteristics of the electrode materials. Surface area, pore volume, and pore size distribution were calculated using the isotherm data acquired from these adsorptions³¹. Figure 6a–c shows the nitrogen adsorption and desorption of prepared at different reaction temperatures. All prepared samples are structurally characterized with the collection of isotherms in the relative pressure range of ~ 0 to 1.0 P/Po using nitrogen at 77 K. The gas sorption measurements showed a characteristic type IV Brunauer isotherm according to the IUPAC classification for all of the prepared samples. It showed a characteristic hysteresis loop with capillary condensation that occurred at a relative pressure of 0.8–1.0. This reveals that the samples are mesoporous^32,33. The specific surface area of the samples was observed to decrease with increasing reaction time, interestingly it agreed well with the unit cell volume compression as observed in XRD. The longer the hydrothermal reaction time, the smaller the surface area. The specific surface areas were found to be 49.21, 34.85, and 22.84 m²/g for 3 h, 9 h, and 15 h respectively. The larger surface area resulting from the more expanded crystal lattice of the 3 h sample may allow an efficient synergistic effect of Na ions at the electrode/electrolyte interface which may result in better electrochemical performance than other samples (9 h and 15 h). According to BJH plots (Fig. 6d), the pore size curve distribution for all samples has a broad pore size distribution, which may contribute to both micropore and mesoporous. Both the micropores and the mesopores play a crucial role in charge accumulation. For charge storage and ion adsorption, the micropores are responsible, whereas the mesopores act as a channel for ions through transport from the bulk of the electrolyte to the interface between the electrode and the electrolyte³⁴. On the basis of this, the electrolyte ions should ideally compliment the pore sizes of the electrode material to attain convincing results of charge propagation and avoid some steric restrains. Table 2 shows the values for surface area, pore volume, and size for all samples.

Figure 6.

BET adsorption–desorption isotherms at (a) 3 h, (b) 9 h and (c) 15 h. (d) BJH pore size distribution at different reaction times.

XPS analysis

The surface composition and oxidation state of the elements were analyzed by XPS, and the results are shown in Fig. 7a–d. The wide survey spectra of the samples in (Fig. 7a), show the presence of manganese, potassium, oxygen, and carbon, that corresponded to the EDS results. The potassium might be due to the KMnO₂ precursor, while the carbon peak is atmospheric contamination due handling of the sample. The high-resolution XPS spectra of the Mn 2p core level are shown in Fig. 7b–d. The two distinct peaks of binding energies at about 641.9 eV and 653.62 eV with spin orbital splitting of 11.6 eV were observed for all samples which correspond to Mn 2p_3/2 and 2p_1/2, which implies that it is indeed MnO₂^35,36. The Mn 2p peaks were fitted with five peaks. The fitted peaks of Mn 2p_3/2 and 2p_1/2 revealed the presence of a mixed valent manganese system, indicating that oxygen vacancies exist on the surface of α-MnO₂³⁷. The Mn 2p_3/2 for all samples were resolved into three components at about 641 eV, ~ 643 eV and 646 eV. The peak at 641 eV might be attributed to the Mn³⁺ and the one at about 643 eV was attributed³⁸ to Mn⁴⁺. A peak at about 646 eV for might be attributed to the satellite peak^39,40. The observed satellite line (the third component) is spaced by the ~ 5 eV from the first component is due to dominant contribution^40,41 from Mn²⁺. The Mn 2p_1/2 were resolved into two components. The first peak at about 653 eV could be attributed to Mn³⁺ and the other peak at about 654 eV could be associated^42,43 with Mn⁴⁺. Furthermore, the peak that corresponded to the Mn³⁺ state has a higher peak intensity and a larger area under the curve than the peak corresponding to the Mn⁴⁺, this may suggest that α-MnO₂ has the possibility to improve electrochemical performances⁴⁴. Interestingly, the decrease in both Mn³⁺ and Mn⁴⁺ area with increasing reaction time resulting from the compressed crystal lattice indicated that there were reduced redox-active sites at the electrode/electrolyte interface which would destructively affect the specific capacitance. The contents of the chemical bonds of all samples are presented in Table 3. Furthermore, the high-resolution spectra of the O 1s core level are shown in Fig. 8. The spectra constitute a main characteristic peak at approximately 529 eV that can be associated with oxygen in the MnO₂ lattice (Mn–O–Mn) and another shoulder at approximately 531 eV which belongs to an oxygen atom in the hydroxyl group (MnOOH)³⁶. Finally, amount of the oxygen atom in the hydroxyl group related peak is decreasing with increase in reaction time in accordance with the volume of the unit cell on the α-MnO₂ as presented with XRD.

Figure 7.

(a) Survey spectrum of MnO₂. (b) High resolution of Mn 2p core level at 3 h, (c) Mn 2p core level at 9 h and (d) Mn 2p core level at 15 h.

Table 3.

XPS high resolution for Mn chemical composition of MnO₂.

Samples	Ion species	Mn 2p3/2		Mn 2p1/2
		Peak position (eV)	Area (%)	Peak position (eV)	Area (%)
3 h	Mn3+	641.02	36.37	652.49	23.00
	Mn4+	642.83	31.13	654.15	12.01
	Satellite peak	646.21	7.49
9 h	Mn3+	641.11	39.09	652.64	24.94
	Mn4+	642.94	18.33	654.53	9.35
	Satellite peak	646.14	8.29
15 h	Mn3+	641.02	31.96	652.72	25.31
	Mn4+	642.49	27.49	654.44	9.09
	Satellite peak	646.04	6.14

Figure 8.

High resolution spectra of the O 1s core level at (a) 3 h, (b) 9 h and (c) 15 h.

Electrochemical properties

Cyclic voltammetry (CV) analysis

The electrochemical study of the prepared electrodes was conducted in a three-electrode system. Electrochemical tests were used to determine the capacitance of MnO₂. To examine the behaviour of the charge storage mechanism, CV was performed with a diverse scan rate between 10 and 100 mV/s (10, 20, 50, 75, and 100 mV/s) and a potential voltage window of − 1 to 0 V in 1 M Na₂SO₄. Figure 9a–c shows the CV curves corresponding to samples prepared at 3, 9, and 15 h. The curves demonstrate capacitive performance with a distorted quasi-rectangular shape with redox peaks. These redox peaks can be attributed to the Mn³⁺/Mn⁴⁺ for MnO₂ ions and contribute to the electron transfer mechanism⁴⁴. It also represents the ideal pseudocapacitive behaviour of the prepared MnO₂ electrodes. Basically, the pseudocapacitive performance is due to the faradaic reaction that occurs at the electro-electrolyte interface⁴⁵. Figure 9a of the sample prepared at 3 h reaction time display a better performance in terms of output current compared to other samples. The displayed higher current response for the 3 h sample indicates a higher specific capacitance against other samples. This finding agrees well with BET and XRD results.

Figure 9.

CV curve of MnO₂ at (a) 3 h, (b) 9 h and (c) 15 h.

Galvanostatic charge–discharge (GCD) analysis

The galvanostatic charge–discharge technique is a reliable electrochemical method for the evaluation of the electrocapacitive behaviour of electrode materials in an energy storage device. Therefore, this approach is adopted for further analysis of the electrode material. The room temperature charge–discharge curves of α-MnO₂ electrode materials are shown in Fig. 10a–c. The GCD studies were carried out at various current densities of 1, 2, and 5 A/g. The GCD curves are in agreement with the CV curves. The plots of the as-prepared electrodes also show pseudocapacitive behaviour with redox peaks which is expected of material like the MnO₂. Figure 10a (3 h sample) produced a longer discharging time, which means it has a much higher specific capacitance other than other samples as we noticed in the case of the CV measurements. In addition, Table 4 compares the supercapacitor performance of the prepared MnO₂-3 h electrode to that of other comparable results published. Specific capacitance as evaluated from GCD curves at 1 A/g using Eq. (7) was 132, 41, and 39 F/g for 3, 9, and 15 h MnO_2, respectively. Interestingly, the specific capacitance trend is consistent with the unit cell volume of MnO_2. The high specific capacitance of the 3 h sample can be attributed to its large surface area, providing more active sites and internal space to which electrolyte ions could easily access relatively to the relaxed crystal structure⁴⁶. Hence, this pivotal observation may be considered as a fundamental indication that a more expanded α-MnO₂ crystal lattice is a favourable electrochemical platform for Na ions intercalation. Finally, this result is in accordance with the BET higher surface area leading to a favourable synergistic effect at the electrode/electrolyte interface for fast ion transport.

Figure 10.

GCD curve of MnO₂ at (a) 3 h, (b) 9 h and (c) 15 h.

Table 4.

Summary of various MnO₂ electrode material used in supercapacitor.

Material	Method of synthesis	Crystal structure	Morphology	Surface area (m2/g)	Specific capacitance at current density/Scan rate	References
α-MnO2	Chemical route	Tetragonal	Hollow sphere	417	220 F/g at 0.5 A/g	49
δ-MnO2	Hydrothermal	Hexagonal birnessite	Flower-like nanospheres	78.3	535 F/g at 1A/g	50
α-MnO2	Facile chemical method	Orthorhombic	Rough nanowires	83.17	171 F/g at 1A/g	51
α-MnO2	One-pot ultrasonication	Tetragonal	Nanoflake	–	145 F/g at 0.5 A/g	52
α-MnO2	Hydrothermal	Tetragonal	Nanowires	–	138 F/g at 1 A/g	48
α-MnO2	Liquid phase method	Tetragonal	Flower-like	–	148 F/g at 1A/g	22
α-MnO2	Hydrothermal	Tetragonal	Nanorods	180	262 F/g at 5 mV/s	32
β MnO2	Facile hydrothermal	–	Nanowires	–	213 F/g at 0.2 A/g	53
α-MnO2	Hydrothermal	Tetragonal	Nanoflowers and nanorods	49.21	132 F/g at 1 A/g	This work

Electrochemical impedance spectroscopy analysis

The resistive behaviour of α-MnO₂ was characterized by electrochemical impedance spectroscopy in the frequency range of 10 mHz–100 kHz at room temperature to understand the ion transport at the MnO₂/Na₂SO₄ interface. The impedance data are examined by using the Nyquist plot fitted via analyzer software. Basically, the Nyquist plot is represented on the X-axis with the real part (Z′) and on the Y-axis with the imaginary part (−Z″) and the area is magnified in the high-frequency range in the inset as shown in Fig. 11. The intercept in the real part (Z′) responds to the combined contact resistance at the interface of the active material, the current collector with the ionic resistance of the electrolyte and the intrinsic resistance of the substrate, namely the equivalent series resistance (ESR). The ESR values of 3, 9, and 15 h of MnO₂ were calculated to be 8.94, 4.34, and 8.77 Ω respectively, suggesting an excellent ionic response in the high frequency range of the sample. Meanwhile, the diameter of the quasi-semicircle in the high-frequency range of the curve represents the low electron transfer resistance (R_ct) arising from the double layer capacitance and faradaic redox reaction on the surface of the materials^9,47. Furthermore, the Nyquist plot (Fig. 11) shows the absence of the complete semicircle, which could be due to the Faradaic reaction. In addition, the nearly straight line in the low-frequency region indicated a favourable diffusion of electroactive species rather than the charge transfer resistance⁴⁸. The impedance spectra of the prepared MnO₂ samples were fitted using the same equivalent circuit model that contains R1 (Equivalent series resistance), R2 (electron transfer resistance), CPE (Constant phase element of double layer capacitance), and W1 (Warburg which is the transition from high to low frequency).

Figure 11.

Nyquist plot of MnO₂ at various reaction time.

Conclusions

In conclusion, an α-MnO₂ nanoparticle were successfully synthesized by hydrothermal method at 3, 9 and 15 h reaction times. The structure, morphology, chemical composition, surface area, and electrochemical properties were investigated. X-ray diffraction (XRD) showed that the reaction time does not affect the phase, but it only shifts the peaks slightly to the higher diffraction angles. It was noticed that the surface morphology changed from nanoflower to nanorods with reaction time. BET analyses confirmed the presence of the mesoporous structure with a typical type IV isothermal hysteresis loop in all samples. The surface area of the samples decreases with an increase in reaction time. Electrochemical performance was tested by CV, GCD, and EIS in a 3-electrode system. The shapes of the CV curves hardly changed with an increase in the scan rate, suggesting the excellent reversibility of the α-MnO₂. Specific capacitance was calculated from the GCD curve with a current density of 1A/g for all samples and the 3 h sample possessed high specific capacitance of 132 F/g. The results obtained in the EIS are in good agreement with the CV and GCD analysis. The higher specific capacitance obtained in the 3 h sample might be due to its higher surface area and the nanoflower-like morphology. These results demonstrate that the MnO₂ prepared by the hydrothermal method at low reaction temperatures is a promising method to improve the electrochemical performance of MnO₂-based electrode materials.

Methods

Materials

All chemical reagents were of pure analytical and were used as received without further purification. The following chemicals and materials were used in this study: Manganese sulphate monohydrate [MnSO₄·H₂O] Potassium permanganate [KMnO₄], polyvinylidene fluoride, sodium sulphate (Na₂SO₄), carbon black, ethanol [C₂H₆O], hydrochloric acid [HCl], deionised water which was extracted using DRAWEL water purification device, Nickel foam and N-Methyl-2-pyrrolidone [NMP].

Preparation of MnO₂ nanoparticles

In a typical synthesis, 2.5 × 10⁻³ mol of potassium permanganate (KMnO4) and 1.0 × 10⁻³ mol of manganese sulphate monohydrate (MnSO₄·H₂O) were dissolved in 40 ml of distilled water with constant stirring at room temperature for 30 min. The homogeneous solution was transferred to 100 ml of Teflon-lined autoclaves. The autoclave was placed in an oven kept at 140 °C. The reaction time was varied from 3, 9, and 15 h to optimize the synthesis conditions. Then the autoclave was removed, and the mixture was cooled down to room temperature naturally. The brownish precipitates were collected by centrifugation and washed with distilled water several times to remove impurities. Finally, the precipitates were dried in air for 12 h at 80 °C.

Preparation of MnO₂ working electrodes

Working electrodes were prepared by mixing the active material (MnO₂), polyvinylidene fluoride, and carbon black in a ratio of (80:10:10) with N-Methyl-2-pyrrolidone (NMP). The ink was then obtained with the help of ultrasonication for 1 h. Subsequently, the inks were drop-cast on the nickel foam and dried at 100 °C for an hour to remove the solvent.

Materials characterization techniques

XRD equipped with monochromatic CuKα radiation with a wavelength of 0.15406 nm was used to determine the phase and crystallite size of the materials. The surface morphology and chemical composition of the material were analyzed by using a JEOL JSM-7800F Scanning Electron Microscope (SEM) equipped with Oxford Aztec Energy Dispersive X-ray spectroscopy (EDS). Vibrational studies were conducted using PerkinElmer Fourier transform infrared spectroscopy (FTIR). Specific surface area and pore volume analyses were recorded by Brunauer–Emmett–Teller (BET) and Barrett–Joyner–Halenda (BJH) methods employing N₂ adsorption–desorption experiments on a Micromeritics TRISTAR 3000 surface area analyzer. Analysis of the chemical state of the samples was carried out by X-ray photoelectron (XPS) comprising PHI 5000 Versaporbe Scanning ESCA using monochromatic Al Kα radiation (hv = 1486.6 eV) and the data was analyzed using Multipack software. The binding energies were calibrated by using the Cu 2p and Au 4f peaks.

Electrochemical characterization

Electrochemical measurements of the prepared electrode material were performed using a Bio-Logic VMP300 potentiostat/galvanostat controlled by EC-Lab v11.42 software. The measurements were done on the three-electrode system. The measurement setup consists of a working electrode, a platinum wire as a counter electrode, and Ag/AgCl electrode as reference electrodes. A 1.0 M Na₂SO₄ was used as an electrolyte. The cyclic voltammetry (CV) curves were collected in a potential window of − 1 to 0 V vs Ag/AgCl at various voltage scans of 10, 20, 75, and 100 mV/s. In addition, galvanostatic charge–discharge (GCD) analyses were recorded at various current densities between 1 and 10 A/g. The specific capacitance ($C_s$) was evaluated from the GCD curve by applying Eq. (1) as shown below⁵⁴:

$$C_s=\frac{I_d\times \mathrm{\Delta }t}{m\times \mathrm{\Delta }V}[\text{F}·{\text{g}}^{-1}]$$ 7

where $\mathrm{\Delta }V$ is the operating potential window (V), $I_d$ is the discharge current in (mA), m is the mass loading of active material in (mg) and $\mathrm{\Delta }t$ is the electrode discharge time in (s). The electrical resistance of the as-synthesized sample was estimated by using an EIS Nyquist plot at an open circuit potential of 0.0 V in the frequency range of 10 MHz–100 kHz.

Author contributions

Modungwe TM: Investigation, Methodology, writing-original draft, formal analysis, Kabongo GL: Conceptualization, Investigation, Conceptualization, writing—review and editing, Mbule PS: writing, review and editing. Coetsee E: writing—review and editing. Makgopa K: writing—review and editing. Dhlamini MS: supervision, writing—review and editing.

Data availability

The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.

Competing interests

The authors declare no competing interests.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-024-70111-4.