Effect of Mn²⁺ concentration on the growth of δ-MnO₂ crystals under acidic conditions

Abstract

δ-MnO₂ is an important component of environmental minerals and is among the strongest sorbents and oxidants. The crystalline morphology of δ-MnO₂ is one of the key factors affecting its reactivity. In this work, δ-MnO₂ was initially synthesized and placed in an acidic environment to react with Mn²⁺ and undergo a crystalline transformation. During the transformation of crystalline δ-MnO₂, kinetic sampling was conducted, followed by analyses of the structures and morphologies of the samples. The results showed that at pH 2.5 and 4, δ-MnO₂ nanoflakes spontaneously self-assembled into nanoribbons via edge-to-edge assembly in the initial stage. Subsequently, these nanoribbons attached to each other to form primary nanorods through a face-to-face assembly along the c-axis. These primary nanorods then assembled along the (001) planes and lateral surfaces, achieving further growth and thickening. Since a lower pH is more favorable for the formation of vacancies in δ-MnO₂, δ-MnO₂ can rapidly adsorb Mn²⁺ directly onto the vacancies to form tunnel walls. At the same time, the rapid formation of the tunnel walls leads to a quick establishment of hydrogen bonding between adjacent nanoribbons, enabling the assembly of these nanoribbons into primary nanorods. Therefore, in a solution with the same concentration of Mn²⁺, the structure transformation and morphology evolution of δ-MnO₂ to α-MnO₂ occur faster at pH 2.5 than at pH 4. These findings provide insights into the mechanism for crystal growth from layer-based to tunnel-based nanorods and methods for efficient and controlled syntheses of nanomaterials.

Introduction

α-MnO₂ is the predominant manganese oxide in the supergene oxidation zones of manganese-bearing crusts, manganese deposits, and lateritic weathered profiles [1,2,3,4]. Representing a significant group within the transition metal oxides (TMOs), α-MnO₂ features 2 × 2 and 1 × 1 tunnel structures (with occasional intergrowth of 2 × 3, 2 × 4, and “T” junction structures), which arise from double chains of edge-sharing [MnO₆] octahedra corner-sharing with neighboring chains [5,6,7]. These larger tunnels are typically stabilized by cations such as K⁺, Ba⁺, and Na⁺ [8,9,10,11,12]. The unique physicochemical properties of α-MnO₂, including its distinct structural and morphological characteristics, enable a broad spectrum of potential applications in environmental pollution abatement, supercapacitors, and molecular sieves [13,14,15,16]. The efficacy of these applications is significantly influenced by the specific structure and morphology of α-MnO₂ [7].

Previous research has revealed that α-MnO₂ forms from the transformation of δ-MnO₂ in acidic environments (pH < 4) [7, 17,18,19]. δ-MnO₂ is naturally formed and widely present in nature [20]. It has poorly crystalline, c-axis disorder stacked [MnO₆] octahedral layered structure, strong adsorption and oxidation activity [21,22,23,24]. Our previous study on the one-step synthesis of α-MnO₂ revealed that the initially formed δ-MnO₂, driven by an increase in internal Mn(III), undergoes a multi-stage oriented assembly of δ-MnO₂ nanosheets, ultimately leading to the formation of α-MnO₂ nanorods [7]. Grangeon et al. through aging experiments of δ-MnO₂ with pH 3–10, found that only the δ-MnO₂ at pH 3 transformed into α-MnO₂ after four years of aging [17]. Above all, the coupled structural and morphological evolution from δ-MnO₂ to α-MnO₂ is influenced by both pH and Mn(III), although the specifics of this influence remain unclear. Understanding the morphological evolution throughout the entire synthetic process and its relationship with structural transformation, as well as the influence of pH and Mn(III) on the morphological evolution during the transformation from δ-MnO₂ to α-MnO₂, is critical for α-MnO₂ material synthesis in various applications and for understanding the mineralogy and behavior of different types of MnO₂ minerals in nature [8,9,10,11,12,13,14,15,16, 20, 23,24,25,26,27].

In this study, poorly crystalline δ-MnO₂ was synthesized. We investigate the influence of pH and Mn²⁺ on the structural transformation and morphological evolution of δ-MnO₂ into α-MnO₂ by introducing varying concentrations of Mn²⁺ at pH 2.5 and pH 4 solutions. The transformation process was observed with morphology and kinetic analyses to study the mechanism. The changes in the nanoparticle morphology and structure during the transformation of δ-MnO₂ to α-MnO₂ were observed by X-ray diffraction (XRD) and high-resolution transmission electron microscopy (HRTEM). This provides not only insights into the formation process of α-MnO₂ nanorods but also new possibilities for controlled synthesis of α-MnO₂ materials.

Experimental section

Experimental materials and methods

Reagents used in the experiment

The reagents used in this experiment included NaMnO₄·H₂O (> 97%), MnSO₄·H₂O (> 99%), and Hac (> 97%), all of which were purchased from Sigma-Aldrich (St Louis, MO). Distilled deionized water (DDW) was used in the preparations of mineral reagents and for washing the products. The DDW was obtained by treating distilled water with a Labconco Water Pro Ps system, and the resulting resistivity was greater than 18.4 MΩ•cm.

Synthesis of poorly crystalline δ-MnO2

In a typical experiment, 10.00 g NaMnO₄ was dissolved in 680 mL of 0.25 mol/L NaOH solution. Then, 320 mL of 0.3 mol/L MnCl₂ solution was added to the NaMnO₄ solution dropwise at a rate of 10.00 mL/min. The suspension was vigorously stirred for an additional 2 h and then standing for 4 h. The precipitate which obtained after centrifugation was mixed with 1 M NaCl₂ solution for 1 h. The suspension was filtered and washed until the conductivity of the supernatant was less than 20.0 µS cm^− 1 [7].

Isothermal adsorption

The isothermal adsorption experiments were controlled variable experiments, and the controllable factors in this experiment were the concentration of Mn²⁺ and the pH of the solution. To meet the quantity requirements for subsequent characteristics, we ultimately adopted the 3 g/L δ-MnO₂ suspension concentration reported by Zhao et al., as determined through our preliminary experiments [28]. The pH setting is based on the findings reported by Grangeon et al. that transformation from δ-MnO₂ to α-MnO₂ is more favorable at pH values below 4, and does not occur at pH 5 [19]. Therefore, we set the pH levels to 2.5, 4, and 5. Our previous research found that δ-MnO₂ can transform into α-MnO₂ when the Mn(III) content above and below the vacancies is at 30%. We used this as the median concentration to set higher and lower Mn²⁺ concentrations to study their effects on the transformation process [7]. The concentrations of Mn²⁺ were 0 mol/L, 0.05 mmol/L, 0.1 mmol/L, 0.3 mmol/L, 0.7 mmol/L, 1 mmol/L, 5 mmol/L, 10 mmol/L, 25 mmol/L, and 50 mmol/L. There were 3 groups of experiments (Table 1). Each set of experiments was repeated three times, and the average value was taken.

Table 1 Experimental settings of the parameters in the syntheses

For each sample, 2.5 mL of 36 g/L suspension of poorly crystalline δ-MnO₂ was added to a centrifuge tube with a certain amount of MnSO₄ solution and a certain volume of deionized water. The total volume of the solution was 30 mL. The pH of the mineral suspensions was adjusted to 0.01 mol/L, 0.05 mol/L, 0.1 mol/L, 0.5 mol/L, 1.00 mol/L NaOH and H₂SO₄. The changes in the suspension pH were regularly monitored, the set pH was maintained, and the mixture was allowed to react for several days until the pH change was less than ± 0.05 within 7 days. After 7 days of stirring, the sample was centrifuged once in a centrifuge tube, the supernatant was removed, and the sample was washed and centrifuged several times until the conductivity of the supernatant was 20 µS/cm. After centrifugation, a high-speed freezing centrifuge (BECKMAN J2-MC) was used at relative centrifugal force (RCF) of 17,800 g, and the centrifugation time was 6 min. The washed paste was freeze-dried, ground in an agate mortar and then bottled for later use. The concentration of Mn²⁺ in the supernatant after centrifugation was determined with flame atomic absorption spectroscopy, and the amount of Mn²⁺ adsorbed and its maximum adsorption capacity were determined at different pHs. Each set of experiments was repeated three times, and the average value was taken.

Kinetics experiments

Briefly, 3 g of wet δ-MnO₂ (the weight of the solid sample is 0.624 g) was transferred to 200 mL of deionized water containing 25 mM MnSO₄. Then the pH of suspension was maintained at ~ 2.5 or 4 for 15 d under vigorous stirring by addition of 0.1 M NaOH or 0.1 M H₂SO₄. Aliquots were taken at various time intervals and the solids were collected by centrifugation at 16 000 g (Neofuge 23R). The collected solids were then rinsed with DDW until the supernatant conductivity was below 20 µS/cm.

Characterization

X-ray diffraction (XRD) analyses of minerals

The synthesized manganese oxide minerals or intermediates were completely washed with DDW, and then X-ray diffraction (XRD) analyses were carried out with a D8 ADVANCE instrument from the BRUKER Company in Germany. Powder tableting and directional tableting processes were used. The test conditions were as follows: Cu Ka radiation (λ = 0.15406 nm), a 40 kV test voltage, a 40 mA test current, steps, 10°/min scanning speed, and 0.01° step sizes.

Transmission electron microscopy (TEM)

The morphologies and crystalline characteristics of the minerals were analyzed with an FEI-Talos-F200 analytical transmission electron microscope (TEM) and identified by electron diffraction (ED). The acceleration voltage was 200 kV, and the emission current was 10 µA. Ten mL of the sample was added to anhydrous ethanol in a 20 mL capacity serum bottle, small amounts of minerals were added to anhydrous ethanol, and after ultrasonic dispersion, the solution was slightly discolored. Samples were taken from the suspension with an ultrathin carbon membrane for computer analyses.

Flame atomic absorption spectrometry

The concentrations of Mn²⁺ in the solutions were determined with a Varian AAS240 FS atomic absorption spectrometer. The corresponding elemental lamps and standard curves were used for analyses.

Results and analysis

Mn2+ adsorption at different pH

As shown in Fig. 1a, the amount of Mn²⁺ adsorbed under acidic conditions increased with the addition of the same amount of Mn²⁺ and with an increase in pH. When the initial concentration of Mn²⁺ in the solution was less than 5 mmol/L, δ-MnO₂ was completely adsorbed at both pH 2.5 and 4. However, as the concentration of Mn²⁺ in the solution continued to increase, the adsorption of Mn²⁺ by δ-MnO₂ reached saturation at pH 2.5. At pH 4, the adsorption of Mn²⁺ by δ-MnO₂ reached saturation when the concentration of Mn²⁺ reached 22.5 mmol/L. This is attributed to the point of zero charge (PZC) of δ-MnO₂, which ranges between 1.5 and 2.5. With the increased pH, the negative charge on δ-MnO₂ also increased, subsequently enhancing its Mn²⁺ adsorption capacity. Notably, the maximum adsorption capacity for Mn²⁺ is lower at pH 2.5 than that at pH 4.

Fig. 1

Isothermal adsorption experiments of δ-MnO₂ on Mn²⁺ in solutions with pH 2.5 and 4 (a), adsorption kinetics of δ-MnO₂ in a 25 mM Mn²⁺ solution (b)

We conducted kinetic experiments using 25 mM Mn²⁺ with 3 g/L δ-MnO₂, and the results are shown in Fig. 1b. When the solution had a pH of 2.5 and 4, the concentration of Mn²⁺ in the solution rapidly decreased from 25 mM to 2.5 mM and 3.6 mM after the addition, respectively. The kinetic process of δ-MnO₂ adsorbing Mn²⁺ can be divided into two parts. Initially, in the first 48 h, the adsorption of Mn²⁺ by δ-MnO₂ showed minor fluctuations, indicating instability in the adsorption of Mn²⁺ during this phase. This is likely due to Mn²⁺ continuously undergoing electron transfer with δ-MnO₂ and exchanging with Mn (IV) in bulk, which promotes the crystallization and structural transformation of δ-MnO₂ [29, 30]. Subsequently, in the second part, from 48 h to 15 days, the amount of Mn²⁺ adsorbed by δ-MnO₂ gradually increased over time. The distinctly different two-stage adsorption process is likely due to the structural and morphological evolution of δ-MnO₂ caused by Mn²⁺.

XRD patterns of the isothermal adsorption sample

The XRD pattern of the sample obtained after 15 d of reaction is shown in Fig. 2. The XRD pattern of the samples when the Mn²⁺ concentrations were 0, 1 mmol/L and 5 mmol/L shows two broad diffraction peaks at 37° (d₍₁₀₀₎ = 0.24 nm) and 65° (d₍₁₁₀₎ = 0.14 nm) (Fig. 2), which can be attributed to δ-MnO₂ with poor crystallinity, small-sized and randomly stacked [MnO₆] octahedral layers [21, 22]. The d-spacing ratio of d₍₁₀₀₎ to d₍₁₁₀₎ is 1.73, which indicates a hexagonal layer symmetry [21]. When Mn²⁺ was added up to 10 mmol/L, the peak at d₍₁₀₀₎ = 0.24 nm began to sharpen and a new shoulder peak appeared at 42.2° 2θ, indicating that Mn²⁺ was gradually adsorbed on the top and bottom of the octahedral vacancies (Fig. 2). When Mn²⁺ was added to more than 25 mmol/L, (110), (200), and (310) α-MnO₂ peaks appeared (ICDD No. 00-29-1020, d₍₁₁₀₎ = 0.69 nm, d₍₂₀₀₎ = 0.48 nm, and d₍₃₁₀₎ = 0.31 nm) (Fig. 2). In the isothermal adsorption curve, when the concentration of Mn²⁺ in the solution reached 10 mmol/L, the adsorption capacity of δ-MnO₂ for Mn²⁺ was close to saturation (Fig. 1a). In the XRD patterns, even in this state of adsorption saturation, δ-MnO₂ did not undergo structural transformation. As more Mn²⁺ was added, the amount of Mn²⁺ adsorbed by δ-MnO₂ did not change significantly, but δ-MnO₂ transformed into α-MnO₂ structure. It might be because the appropriate amount of Mn²⁺ did not directly integrate into δ-MnO₂, but rather facilitated the transformation of δ-MnO₂ to α-MnO₂ through its effect on the electron transfer of surface Mn (III/IV).

Fig. 2

The XRD patterns of solid samples in isothermal adsorption experiments of δ-MnO₂ on Mn²⁺ in solutions with pH 2.5 (a) and 4 (b)

XRD patterns of the adsorption kinetics samples

The XRD patterns of the obtained samples under pH 2.5 are shown in Fig. 3. The initial mineral was poorly crystalline δ-MnO₂. After 2 days of reaction, the intensity of the diffraction peak at d = 0.24 nm increased. And at the same time, the characteristic diffraction peak of α-MnO₂ (d₍₁₁₀₎ = 0.70 nm) appeared at 12° 2θ, which indicated that δ-MnO₂ was transformed into α-MnO₂ after 2 days (Fig. 2a) [7, 11]. When the reaction time was extended to 7 days, the intensity of the diffraction peak gradually increased indicated that the δ-MnO₂ transformed into α-MnO₂ (Fig. 2a). Before the transformation, δ-MnO₂ continuously exchanged with Mn²⁺ in the solution, preparing for the conversion to α-MnO₂, such as the generation and arrangement of Mn (III) [7]. Therefore, the adsorption amount of Mn²⁺ by δ-MnO₂ was constantly fluctuating. When the reaction proceeded for two days, δ-MnO₂ transformed into the more stable α-MnO₂, and α-MnO₂ further adsorbed Mn²⁺ from the solution.

Fig. 3

The XRD patterns of solid samples in adsorption kinetics experiments of 25 mM Mn²⁺ on δ-MnO₂ in solutions with pH 2.5 (a) and 4 (b)

The XRD patterns of the samples, capturing the dynamic interaction between δ-MnO₂ and 25 mM Mn²⁺ in a pH 4 solution over a period of 15 days, are shown in the Fig. 3b. Compared to the rate at which δ-MnO₂ begins to convert to α-MnO₂ within 2 days in a pH 2.5 solution, δ-MnO₂ in a pH 4 solution only starts to transform into α- MnO₂ after 7 d. When the reaction time was 4 days, the peak intensity of d₍₁₀₀₎ = 0.24 nm increased, indicating that the crystallinity of the δ-MnO₂ increased with increasing reaction time [18]. When the reaction time was extended to 7 days, the peak intensity at d = 0.24 nm increased, and the characteristic diffraction peak for α-MnO₂ (d₍₁₁₀₎ = 0.70 nm) appeared at 2θ = 12°. This indicated that after 7 days, δ-MnO₂ transformed into α-MnO₂ [7, 11]. Therefore, compared to the reaction system at pH 2.5, δ-MnO₂ in pH 4 exhibits a fluctuation cycle (preparation period) of 7 days in Mn²⁺ adsorption. An increase in adsorption is observed when δ-MnO₂ begins its transformation into α-MnO₂ on the 7 days.

Morphological changes of δ-MnO2 under different pH

Morphological of initial δ-MnO2

Figure 4a illustrates the initial formation of nanoflakes aggregates upon mixing the KMnO₄ and MnSO₄ solutions. Further captured at a higher resolution (Fig. 4b), reveals poorly crystalline δ-MnO₂ nanosheets ranging in size from 3 to 5 nm, featuring a lattice spacing of d₍₁₀₀₎ = 0.24 nm. Furthermore, the selected area diffraction (SAED) pattern (inset of Fig. 4a) shows two diffuse diffraction rings at ∼0.24 nm and ∼0.14 nm, which is consistent with the d₍₁₀₀₎ = 0.24 nm and d₍₁₁₀₎ = 0.14 nm spacings of δ-MnO₂, respectively. This observation consists with the results obtained from XRD pattern in Fig. 3.

Fig. 4

TEM images of the intermediate product of initial δ-MnO₂ (a and b) and the SAED pattern (an inset) were recorded by focusing the electron beam in the area of image b

Morphological changes of δ-MnO2 under pH 2.5

When δ-MnO₂ reacted with Mn²⁺ for 1 h at pH 2.5, as demonstrated in Fig. 5a and b, nanoribbons with lengths of 10–20 nm were formed. Upon magnifying nanoribbons, it becomes revealed that these nanoribbons are assembled from several δ-MnO₂ nanoflakes. They display lattice with a spacing of d₍₁₀₀₎ = 0.24 nm, while lacking the typical lattice spacing of d = 0.7 nm characteristic of α-MnO₂. This indicates that these nanoribbons are assembled δ-MnO₂.

Fig. 5

TEM images of δ-MnO₂ react with 25 mM Mn²⁺ at pH 2.5 about 1 h (a and b), and 1 d (c and d). The red line and circles represent hexagonal δ-MnO₂ nanoflakes in image b. The yellow dashed lines within the red rectangular area show the assembly of several δ-MnO₂ nanoflakes into nanoribbons in image d

When the reaction time was extended to 1 day, as shown in Fig. 5c and d, the δ-MnO₂ nanoflakes gradually disappeared, while the nanoribbons increased in length to 100 nm. The δ-MnO₂ nanoflakes were gradually transforming into δ-MnO₂ nanoribbons as there were no lattice stripes of α-MnO₂ observed. The side perspective of a nanoribbon highlight by yellow dashed line in Fig. 5d demonstrates that an δ-MnO₂ nanoflake assemble at the end surface of a nanoribbon.

When the reaction time was extended to 4 days, the δ-MnO₂ nanoflakes almost disappeared, resulting in the formation of nanorods with good crystallinity and lengths ranging from 100 nm to 300 nm (Fig. 6a). Upon further magnification of the nanorods, the lattice with a spacing of d₍₁₁₀₎ = 0.70 nm characteristic of α-MnO₂ structures were observed internally (Fig. 6b), consistent with the XRD pattern (Fig. 3a). Small nanorods were assembled side-to-side with each other, increasing the thickness of the original nanorods observed in Fig. 6b. Previous studies also described this side-to-side assembly of α-MnO₂ nanorod along the (110) plane driven by surface energy [11, 21].

Fig. 6

TEM images of δ-MnO₂ react with 25 mM Mn²⁺ at pH 2.5 about 4 d (a and b), and 7 d (c and d). The yellow dashed lines show the side-to-side assembly in image b. The yellow dashed lines show the end-to-end assembly in image d

When the reaction time reached 7 days, the nanorods grew to approximately 400 nm and tended to grow in the same direction shown in Fig. 6c. As shown by the amplified image in Fig. 6d, short nanorods with length of about 150 nm connect with each other by end-to-end assembly along the α-MnO₂ (001) planes to form secondary nanorods.

Morphological changes of δ-MnO2 under pH 4

When all other conditions remained unchanged and only the pH value was adjusted from 2.5 to 4, nanoribbons with weak crystallinity appeared after a reaction time of 12 h shown in Fig. 7a. The internal lattice of the nanoribbon is inconsistently oriented, with spacing of d₍₁₀₀₎ = 0.24 nm, indicating that it is a δ-MnO₂ structure. These nanoribbons are approximately 40 nm in length, which is longer than the nanoribbons (length about 20 nm) formed after 1 h at a pH of 2.5. It is noteworthy that the d₍₁₀₀₎ = 0.24 nm lattice directions within the nanoribbons are inconsistent, which may be due to the δ-MnO₂ nanoribbons being newly assembled by δ-MnO₂ nanoflakes and not having the opportunity to adjust orientations (Fig. 7b).

Fig. 7

TEM images of δ-MnO₂ react with 25 mM Mn²⁺ at pH 4 about 12 h (a and b), and 4 d (c and d). The yellow lines show the assembly of δ-MnO₂ nanoflakes on the end of nanoribbons in image d

When the reaction time reached 4 days, δ-MnO₂ nanoflakes gradually transform into nanoribbons with length from 100 to 200 nm (Fig. 7c and d). δ-MnO₂ nanoflakes were observed to assemble on the end faces of the nanoribbons (Fig. 7d). This may be due to the outside nanoflakes of the initial δ-MnO₂ nanoflakes aggregation were first assembled into long nanoribbons. Then the inner δ-MnO₂ nanoflakes were gradually involved as the reaction progressed. Therefore, δ-MnO₂ was still present after 4 days. Figure 7d shows several nanosheets stacking and thickening each other, resulting in the formation of nanorods.

After 7 days of reaction, TEM observations revealed that the δ-MnO₂ nanoflakes had disappeared, and nanorods approximately 500 nm in length had emerged (Fig. 8a). The SAED with d₍₁₁₀₎ = 0.70 nm characteristic of α-MnO₂ was observed, indicating the beginning of the transformation from δ-MnO₂ to α-MnO₂ (Fig. 8b). In Fig. 8b, numerous nanoribbons (30 nm in length and 5 nm in width) were observed assembled at the end of nanorods. Notably, these nanoribbons did not exhibit lattice with a spacing of d₍₁₁₀₎ = 0.70 nm, suggesting they are still of the δ-MnO₂ structure. It is probable that these nanoribbons stacked upon each other during the assembly process, ultimately extending along the assembled faces of the nanorods and forming α-MnO₂ nanorods.

Fig. 8

TEM images of δ-MnO₂ react with 25 mM Mn²⁺ at pH 4 about 7 d (a and b), and 15 d (c and d). The SAED pattern (an inset) was recorded by focusing the electron beam in the area of image b. The yellow lines show the side-to-side assembly in image d

After 15 days of reaction, there are nanorods with 300–500 nm in length and 30 nm in width were formed (Fig. 8c). Upon further magnification of the nanorods, the lattice with a spacing of d₍₁₁₀₎ = 0.70 nm characteristic of α-MnO₂ structures were observed (Fig. 8d), which was consistent with the XRD findings after 15 d (Fig. 3b). Additionally, it was observed that the nanorods underwent side-to-side assembly along the (110) crystal planes, leading to an increase in width.

Discussion

$$\:{2\text{K}\text{M}\text{n}\text{O}}_{4}\:+\:{3\text{M}\text{n}\text{S}\text{O}}_{4}\:+\:{2\text{H}}_{2}O\:=\:{5\text{M}\text{n}\text{O}}_{2}\:+\:{2\text{H}}_{2}{\text{S}\text{O}}_{4}\:+\:{K}_{2}{\text{S}\text{O}}_{4}$$

(1)

The morphological evolution process of δ-MnO₂ to α-MnO₂ in the presence of Mn²⁺ in solutions with pH 2.5 and 4 is shown in Schematic 1. As shown by the chemical reaction in Eq. (1), the initial δ-MnO₂ was produced by reducing KMnO₄ with MnSO₄. The δ-MnO₂ formed small nanoflakes about 3 to 5 nm in size with hexagonal symmetry. In pH 2.5 or pH 4 solution, δ-MnO₂ reacts with Mn²⁺ concentrations greater than 10 mM to transform into α-MnO₂ nanorods. Specifically, δ-MnO₂ nanoparticles with sizes ranging from 3 to 5 nm first assembled edge-to-edge along the (100) plane to form longer δ-MnO₂ nanoribbons [28, 31]. This is consistent with the findings of Liang et al. who reported that the morphology evolution preceded the formation of the mineral phase [7]. In pH 2.5 solution, secondary δ-MnO₂ nanoribbons were assembled along the (110) surface and thickened to form primary α-MnO₂ nanorods. Thirdly, primary α-MnO₂ nanorods assemble through end-to-end along the (001) plane to form longer nanorods. Adjacent nanorods align side-to-side along the (110) surface, driven by the high surface energy, to form wider nanorods.

Schematic 1

Diagram of crystal growth during the transformation of δ-MnO₂ nanosheets to α-MnO₂ nanorods at different pH

Previous studies have shown that the cation adsorption capacity of δ-MnO₂ increases with the rise in pH [32, 33]. This study also reveals that the amount of Mn²⁺ adsorbed by δ-MnO₂ at pH 4 is 0.5 mmol/g greater than that in a pH 2.5 solution. The slight increase in δ-MnO₂ adsorption capacity for Mn²⁺ is the result of the combined changes in the number of vacancies and the amount of negative charge as the pH increases. A previous study revealed that after aging δ-MnO₂ for several years at various pHs, the δ-MnO₂ vacancy content was closely related to the pH, with a lower pH leading to the formation of more δ-MnO₂ vacancies [17]. The vacancies in δ-MnO₂, which have a negative charge due to charge deficiency, can strongly adsorb positively charged ions. Therefore, the number of vacancies in δ-MnO₂ at pH 2.5 is greater than in a pH 4 solution. Conversely, the PZC of δ-MnO₂ is 1.5–2.5, and as the pH of the solution increases from 2.5 to 4, the increase in negative charge leads to stronger adsorption of Mn²⁺ [33]. However, the effect of the vacancies was not as significant as that of the PZC, and the amount of Mn²⁺ adsorbed by δ-MnO₂ at pH 4 was greater than in a pH 2.5.

In this study, when there is a small difference in the amount of Mn²⁺ adsorbed by δ-MnO₂, pH is a key factor in controlling the rate of its transformation to the 2 × 2 tunnel structure of α-MnO₂ [32, 33]. This may be because at pH 2.5, more vacancies are formed directly, and Mn²⁺ is directly adsorbed onto these vacancies after its addition [17]. According to the reaction in Eq. 2, a large amount of [Mn (III)O₆] tunnel walls are formed. These [Mn (III)O₆] octahedra include three unsaturated oxygen molecules that combine with H⁺ to form -OH. When the amount of -OH is large enough, a network of hydrogen bonds (H-bonding) forms between the [Mn (III)O₆] octahedra of adjacent nanoribbons and nanoflakes [7]. However, in a pH 4 solution, fewer vacancies are formed in δ-MnO₂ [34]. The initial adsorption of Mn²⁺ and its reaction with δ-MnO₂ generate Mn(III), which migrates onto the vacancies due to the Jahn-Teller effect [7]. This process is slower than direct adsorption onto vacancies [35]. Therefore, at a solution pH of 2.5, δ-MnO₂ begins to transform into α-MnO₂ within 2 d, while in a pH 4 solution, this process in δ-MnO₂ only starts after 7 days.

In comparison to the solution at pH 2.5, the nanoflakes in the pH 4 solution assembled to form longer nanoribbons due to the prolonged transformation process. This is because the formation of [Mn(III)O₆] tunnel walls takes longer at pH 4 compared to pH 2.5, allowing δ-MnO₂ nanoflakes enough time to assemble edge-to-edge and form longer nanoribbons [7, 36, 37]. Therefore, when these long nanoribbons have assembled enough tunnel walls, they directly form primary α-MnO₂ nanorods longer than those formed at pH 2.5. When nanoribbons are assembled to form nanorods in a pH 4 solution, many δ-MnO₂ nanoflakes remain unassembled into nanoribbons and directly assemble on the (100) crystal face of the primary nanorods. As the pH decreases, the H-bonding between -OH groups on the edges of δ-MnO₂ nanoflakes becomes stronger, resulting in the rapid assembly of smaller nanoribbons at pH 2.5. In contrast, when the primary nanorods begin to form at pH 4, some δ-MnO₂ nanosheets are still present because the H-bonding is weakened. This phenomenon continues, and subsequently, the uncompleted nanoribbons continue to assemble directly along the (100) crystal face of the nanorods.

$$\:{\text{M}\text{n}}^{2+}\:+\:{\text{M}\text{n}\text{O}}_{2}\:+\:{2\text{H}}_{2}O\:\to\:\:{2\text{M}\text{n}}^{3+}\:+\:{2\text{H}}^{+}$$

(2)

$$\:{2\text{M}\text{n}}^{3+}\:+\:{2\text{H}}_{2}O\:\to\:\:{\text{M}\text{n}}^{2+}\:+\:{\text{M}\text{n}\text{O}}_{2}\:+\:{4\text{H}}^{+}$$

(3)

Finally, these primary α-MnO₂ nanorods assemble end-to-end and side-to-side, increasing in width and length. Yuan et al. calculated the energy of the α-MnO₂ (110) facets to be 0.74 J/m² and the energy of the (001) facet to be 1.17 J/m². Driven by the increased surface energy, the nanorods connected head-to-tail along the (001) faces and side-to-side via H-bonding at the (110) face, resulting in further thickening and growth of the α-MnO₂ nanorods [11].

In summary, these results showed that different pH values and different Mn²⁺ concentrations had significant impacts on the structural transformation and morphological evolution of δ-MnO₂ to α-MnO₂. Although this study is not exhaustive, it highlights an important area for further researches on the morphological and structural interactions under different conditions. Future studies could involve quantifying the number of vacancies at different pH levels and exploring the interactions of Mn²⁺ within the system using isotope studies. Additionally, in-situ morphological characterization techniques can be employed to determine the kinetics and mechanisms of the transformation from δ-MnO₂ nanoflakes to α-MnO₂ nanorods under different pH conditions.

Conclusion

δ-MnO₂ is a kind of 3–5 nm nanoparticles consisting of [Mn(III)O₆] octahedra and single-layer flaky manganese oxide with hexagonal symmetry and similar biological functions in a supergene environment. The amount of Mn²⁺ adsorbed by δ-MnO₂ increased with increasing pH, indicating that the negative charge on the mineral surface was greater at pH 4 than at pH 2.5, thus the rate of Mn²⁺ adsorption was greater at pH 4 than at pH 2.5. Different pH values and Mn²⁺ concentrations strongly influenced the transformation of δ-MnO₂ into α-MnO₂. As Mn²⁺ increases and pH decreases from pH 4 to 2.5, the transformation of δ-MnO₂ to α-MnO₂ nanorods accelerates. With increasing reaction time, δ-MnO₂ nanoparticles grew and thickened at the tops and edges of the nanorods through oriented attachment.