0 Introduction

As an important transition metal oxide, MnO₂ exists in different structural forms. The crystal structure of MnO₂ includes three major categories: one-dimensional tunnel structure, two-dimensional layered structure, and three-dimensional network structure. Birnessite MnO₂ which is made up of edge-sharing MnO₆ octahedron layers belongs to two-dimensional MnO₂. Between the MnO₆ layers distribute a certain amount of water molecules and charge balance ions^[1-2]. Birnessite MnO₂ has been widely studied as electrode materials used for the field of energy storage because of its abundant reserves, unique structure, large theoretical specific capacity (1232 mAh/g), and environmental benignity^[3-8]. It has been reported that layered birnessite MnO₂ can be delaminated into nanosheets^[9]. Compared with MnO₂ bulks, MnO₂ nanosheets have a higher specific surface area, which can provide more location for electrochemical reaction and is in favor of the electrochemical performance^[10-11]. Besides, MnO₂ nanosheets can be used as components to form nanocomposites^[12-15].

It is a traditional and effective method to delaminate birnessite MnO₂by insertion of tetramethylammonium (TMA⁺) ions before water washing^[9]. But the above delamination method has a main shortcoming, i.e., the periodic time is longer than a week. Cui et al.^[16] reported the delamination of birnessite MnO₂ by ultrasonic concussion (300 W, 40 kHz) of layered MnO₂ in solution of tetramethylammonium hydroxide (TMAOH). It only took 20 min to transform the bulk materials to the colloidal nanosheets, so the whole period significantly reduced. However, high power ultrasound tends to fragment the MnO₂ nanaosheets into smaller chips, which is undesirable because of losing intrinsic properties in the final assemble nanosheet structure.

Ogino et al.^[17] successfully delaminated the layered graphite oxide into GO by rapid quenching method without sonication. Chakravarty and Late^[18] also delaminated bulk layered transition metal dichalcogenides into nanosheets by rapid cyclic quenching method. The rapid quenching method is time-saving and mild, which consists of rapidly freezing an aqueous solution containing the bulk layered material and subsequently thawing the resultant solution. These reports have inspired us to adopt the rapid quenching method to delaminate birnessite MnO₂ into nanosheets. Herein the preparation of MnO₂ nanosheets through repetitive freezing and thawing cycles was reported. The electrochemical properties of the prepared MnO₂ nanosheets used as an anode material for lithium-ion batteries was also studied.

1 Experimental

All chemical reagents used in our experiment were analytically pure and used without further treatment. The water used in the experiment was deionized water.

A solution was prepared by mixing 0.6 M NaOH with 1.1 M H₂O₂, and then the mixed solution was quickly added into a 0.3 M Mn(NO₃)₂ solution to form precipitate. The precipitate was collected and washed until it is neutral. The precipitate was transferred to Teflon sealed autoclave and heated at 150 ℃ for 16 h in a 2 M NaOH solution. The Na-birnessite obtained was immersed in 0.1 M HCl solution for three days at room temperature, washed with deionized water, and dried at 70 ℃. The product was designated as H-Birnessite.

0.02 g H-Birnessite and 5 mL aqueous solution of TMAOH (0.35 M) were charged into a propylene tube (inside diameter of 13 mm, length of 128 mm). The tube was immersed in a liquid nitrogen bath for about 30 s to freeze the mixture. Then the tube was immediately moved to a water bath (set at 70 ℃) and stayed there for 30 min to thaw the mixture. After the freezing-thawing cycle was repeated for 50 times, a colloid suspension was obtained. The suspension was washed with water for 4 times through centrifuging at 8000 r·min^-1 for 10 min every time to obtain the delaminated birnessite MnO₂ slurry. Finally the birnessite MnO₂ slurry was freeze-dried.

XRD patterns were collected on a Rigaku D/max-2200PC powder diffractometer equipped with a Cu Kα source (λ=1.5418 Å). The morphology of the products was characterized by employing a scanning electron microscope (S-4800) as well as a transmission electron microscope (Tecnai G2 F20 S-TWIN).

The electrochemical performances of MnO₂ nanosheets and Na-birnessite were measured by assembling 2032 coin cells. The mixture of 40% active materials, 50% carbon black, and 10% polyvinylidene difluoride (PVDF) was ground for 30 min, and then moderate N-methyl-2-pyrrolidinone (NMP) was added for grinding for 30 min to form a slurry. The slurry was coated on copper foils and then dried in a vacuum drying oven at 80 ℃ for 24 h. The copper with active material was cut into wafers with a diameter of 16 mm. The mass loading of the active material was about 0.5-0.7 mg/cm². Lithium plate was selected as the cathode. The electrolyte solution in this test was 1M LiPF₆ dissolved in a mixture of ethylene carbonate (EC), ethyl methyl carbonate (EMC), and dimethyl carbonate (DMC) with a 1∶1∶1 volume ratio. The button cells were assembled in a glove box filled with argon. Nyquist plots and cyclic voltammetry were recorded using an electrochemical work station (CHI660e) at selected voltages range of 0.01-3.0 V and frequencies from 0.001 Hz to 100 kHz at room temperature. Galvanostatic charge-discharge was measured by battery system (LANHE CT-2001A) between 0.01 and 3.0 V.

2 Results and Discussion

Fig. 1 displays the mechanism diagram of delamination of birnessite MnO₂ into nanosheets through rapid quenching method. In this method, water rapidly changed to ice. The volume of water molecules between the MnO₆ octahedron layers expanded and contracted repeatedly during rapid freezing and subsequent thawing cycles, which caused fast change in the interlayer distance between layers, and thus weakened the binding force between layers. This way, the intercalation of TMA⁺ cations into layers became much easier, which enabled a rapid and efficient delamination of birnessite MnO₂.

Fig. 2 exhibits the XRD patterns of Na-birnessite and delaminated MnO₂ slurry. Na-birnessite showed three obvious diffraction peaks at 12.3 ° (001 plane), 24.9 ° (002 plane), and 36.1° (110 plane), suggesting that the Na-birnessite had the typical monoclinic lamellar structure with a basal spacing of 0.7 nm. The delaminated MnO₂ slurry displayed only an amorphous halo compared with Na-birnessite, which was similar to the case of other works^{[9, 11, 19]}. The halo indicated the original crystal structure of birnessite MnO₂ disappeared, suggesting that the layered H-birnessite was delaminated to scattered MnO₂ nanosheets. It follows that H-birnessite can be delaminated to nanosheets effectively by the rapid quenching method.

The SEM image of Na-birnessite is given in Fig. 3(a). The sample showed platelike pattern with a lateral dimension of 1-2 μm. The TEM micrograph (Fig. 3(b)) of the delaminated MnO₂ slurry manifested that the targets of observation had homogeneous brightness contrast, which indicated that they were ultrathin and uniform. Besides, it was easily observed that the width of the ultrathin platelets was about 1.5 μm, which was in line with the crystal size of Na-birnessite. The TEM result indicated that the delamination took place without destroying the MnO₂ nanoplates. Fig. 3(c) displays the morphology of the delaminated MnO₂ slurry followed by freeze-drying. It could be found that the nanosheets were uniform with a lateral dimension of 1-2 μm.

The samples of MnO₂ nanosheets and Na-birnessite were respectively employed as anodes to assemble lithium ion batteries to test their electrochemical performances. Fig. 4(a) exhibits the cyclic voltammetry (CV) curves of MnO₂ nanosheets for the initial three cycles measured between 0.01 and 3 V at a scan rate of 0.1 mV/s. Three main redox peaks were observed at around 0.28, 1.05, and 1.2 V. The CV curves exhibited a high overlap after the first cycle, which meant that the MnO₂ nanosheets prepared by rapid quenching method had a good electrochemical reversibility. During the cathodic process, Li⁺ ions entered into MnO₂ crystal lattice and Mn⁴⁺ was reduced. The cathodic peak at 1.05 V corresponded to the reduction of MnO₂ to Mn²⁺, and the sharp peak at 0.28 V accounted for the reduction of Mn²⁺ (MnO) to metallic Mn^0[20]. During the anodic process, lithium ions left the crystal lattice of MnO₂ and Mn⁰ was oxidized. The oxidation peaks were at about 1.2 V in all the cycles, which was contributed to the oxidation of Mn⁰ to Mn⁴⁺. Fig. 4(b) shows the charge-discharge curves of MnO₂ nanosheets at the 1st, 10th, 50th, and 100th cycles at 100 mA/g from 0.01 to 3.0 V. The initial discharge and charge capacities were 1938.9 and 760 mAh/g respectively, and the coulombic efficiency was about 39%. The large capacity loss in the 1st cycle for MnO₂ nanosheets was ascribed to the formation of solid electrolyte interface (SEI) layer, which was common for most electrode materials. The coulombic efficiency reached up to 94% at the 50th cycle and increased to 99% at the 100th cycles.

The cycle performance of MnO₂ nanosheets and Na-birnessite at 100 mA/g are exhibited in Fig. 5(a). Obviously, MnO₂ nanosheets showed a clearly increased cyclic capacity. The reversible capacity of the MnO₂ nanosheets retained 750 mAh/g after cycling for 20 times, and then increased with cycling and reached 1040.6 mAh/g after 100 cycles, which was much higher than those reported in Refs.[21-23]. However, the Na-birnessite can only reach a lower value of 411.8 mAh/g after 100 cycles. The cycle property at 1000 mA/g was also tested. As can be seen from Fig. 5(b), the specific charge capacity of MnO₂ nanosheets tested at 1000 mA/g still stabilized at 438.3 mAh/g after cycling for 100 times. Similarly, after cycling for 100 cycles at 1000 mA/g, the delivered capacity of MnO₂ nanosheets was much larger than that of the Na-birnessite. Fig. 5(c) shows the rate performance of MnO₂ nanosheets and Na-birnessite. The average specific charge capacities of MnO₂ nanosheets were 740, 600, 500, 420, and 320 mAh/gwhen cycling under 100, 200, 500, 1000, and 2000 mA/g, respectively. After the current density returned to 100 mA/g, the average specific charge capacities of MnO₂ nanosheets and Na-birnessite recovered to 800 and 500 mAh/g, respectively.

Fig. 6(a) displays the Nyquist plots of MnO₂ nanosheets and Na-birnessite from 0.001 Hz to 100 kHz. The impedance data were analyzed using the equivalent circuit inset in Fig. 6(a). The intercept at real axis at high frequency corresponds to ohmic resistance (R_Ω). The capacitive loops at high frequencies is put down to the migration resistance (R_SEI) resulted from Li⁺ ions through SEI films. The intercept at mediate-low frequencies corresponds to R_ct, representing the charge transfer resistance. The low-frequency oblique line represents the Warburg impedance (Z_w), resulting from the lithium ions diffusion in the active materials. The curves of MnO₂ nanosheets and Na-birnessite were both composed of a low-frequency oblique line and two high-frequency semicircles. For the MnO₂ nanosheets and Na-birnessite samples, R_Ω values were 24.31 and 84.36 Ω, R_SEI values were 13.69 and 60.4 Ω, R_ct values were 19.58 and 122.1 Ω, and the Z_w values were 0.0327 and 0.003003 Ω, respectively. The Li⁺ ion diffusion coefficient was calculated on the basis of the following equation:

$ {D_{{\rm{L}}{{\rm{i}}^ + }}} = \frac{{{R^2}{T^2}}}{{2{A^2}{n^4}{F^4}{C^2}{\sigma ^2}}} $

(1)

$ {Z'} = {R_\Omega } + {R_{{\rm{ct}}}} + \sigma {\omega ^{ - 1/2}} $

(2)

where R, T, A, n, F, C, and σ denote the gas constant, the absolute temperature, the electrode surface area, the number of reactive electrons during reducing per molecule MnO₂, the Faraday constant, the concentration of Li⁺ ions, and the slope of Z′-ω^-1/2 at low-frequency (Fig. 6(b)), respectively. For MnO₂ nanosheets and Na-birnessite, Z′ and ω^-1/2 presented a good linear relationship at low frequency.

The Li⁺ ion diffusion coefficients of MnO₂ nanosheets and Na-birnessite were calculated according to Eqs. (1) and (2).The records of data are exhibited in Table 1. The obtained Li⁺ ion diffusion coefficient of the MnO₂ nanosheets was 1.3×10^-19cm²/S, larger than that of Na-birnessite (1.5×10^-21 cm²/S). The smaller R_ct and higher Li⁺ ion diffusion coefficient of as-prepared MnO₂ nanosheets indicated the charges and lithium ion transported more easily in them, which was beneficial to lithium ions battery application.

The charging or discharging process of lithium ion battery corresponds to embedding or disembedding lithium ions into or out of anode materials respectively, which is a complex process of multi-step series. It mainly includes 1) the diffusion of Li⁺ ions in electrolyte, 2) the movement of Li⁺ ions in SEI membrane, 3) the charge transfer reaction process taking place at the interface between SEI membrane and anode material, and 4) the Li⁺ ions diffusion in solid phase anode materials^[24-25]. It is generally believed that it is difficult for Li⁺ ions to diffuse in the solid phase than in the liquid phase. Therefore, the diffusion of Li⁺ ion in solid phase is the dynamic controlling factor in the charging and discharging process^[26]. The size of anode particles has direct influence on the length of the solid phase diffusion path of lithium ions, which has an important influence on the electrochemical performance. In our work, the fine electrochemical property of MnO₂ nanosheets could be attributed to three aspects. Firstly, when particles were reduced to nanoscale, the migration path of Li⁺ ions or electrons in solid decreased, thus the R_ct and ion diffusion resistance reduced. Besides, the nanosheet structure can increase the contact area between the electrode surface and the electrolyte, thus increasing the active sites. Furthermore, the nanostructure brought larger specific surface area, which then effectively reduced the current density of electrode, thus decreasing the polarization of electrode.

3 Conclusions

MnO₂ nanosheets has been successfully synthesized by the freezing and thawing method. This method consisted of many cycles of fast freezing of a mixture solution containing H-birnessite and TMAOH, followed by melting of the resultant solid. This simple method for the delamination of layered birnessite MnO₂ was timesaving and environmentally friendly without destroying the initial nanosheet size. Meanwhile, MnO₂ nanosheets synthesized by this method exhibited an excellent cycling performance. Cycled for 100 times at 1000 mAh/g, MnO₂ nanosheets showed a high charge capacity of 438.3 mAh/g, which is 105% of the initial charge capacity. The fine electrochemical performances was ascribed to the nanosheet structure and increased electrical conductivity.