Facile preparation of Mn₃O₄ octahedra and their long-term cycle life as an anode material for Li-ion batteries

Abstract

We describe the simple preparation of octahedral Mn₃O₄ nanomaterials with a typical diameter around 300–400 nm using a one step dealloying of MnAl alloy at room temperature. The as-made sample exhibits high performance as an anode material for Li-ion batteries. Electrochemical measurements reveal that the Mn₃O₄ octahedra have an ultralong cycle life with capacity retentions of 81.3% and 77.8% after 500 cycles at 100 and 300 mA g⁻¹, respectively. Moreover, the Mn₃O₄ octahedra deliver a stable capacity at a high rate of 1000 mA g⁻¹ with a good rate capability. The as-made Mn₃O₄ octahedra exhibit great potential for application as anode materials for Li-ion batteries with the advantages of unique performance and easy preparation.

---

1. Introduction

Li-ion batteries (LIBs) are regarded as the most promising candidates to be the major power source for mobile applications.¹ At present, the main anode material for current commercial LIBs is graphite, but its low capacity (372 mA h g⁻¹) restricts the application of LIBs in high-power fields.^2,3 Therefore, great effort has been devoted to searching for alternative anode materials with high capacity, good rate capability, and long cycle life. Recently, transition metal oxides such as manganese, cobalt and iron oxides,^4–9 have attracted increasing attention due to their respectable performance as anode materials in LIBs. Among these oxides, Mn₃O₄ delivers a high theoretical capacity of 937 mA h g⁻¹,^7,10,11 which is almost three times as high as the value of graphite materials. In addition, manganese has the advantages of a relatively low electromotive force, high natural abundance, low cost, environmental benignity, etc. Considering the properties of Mn, Mn₃O₄, one of its oxides, has great application potential as a promising candidate anode material. However, there are still few investigations on this candidate material related to its electrochemical properties so far.

In previous reports, Mn₃O₄ has usually exhibited poor lithiation activity and cyclability, which is a result of the significant volume change that occurs during Li insertion–extraction.^12–15 In order to tackle this problem, researchers have designed Co-doped Mn₃O₄ samples. For example, pure Mn₃O₄ delivered a capacity of only 200 mA h g⁻¹, which was improved to 400 mA h g⁻¹ through Co doping.¹⁶ However, this is still far below the theoretical capacity. In addition, Co-based materials are confronted with high-cost and toxicity problems. Consequently, engineering the morphology of Mn₃O₄ nanomaterials has been an effective strategy for enhancing their electrochemical performance in LIBs. For example, Wang et al. found that an order-aligned Mn₃O₄ anode delivered an initial capacity of up to 637 mA h g⁻¹ and retained 494 mA h g⁻¹ after 100 cycles.¹⁰ Gao et al. fabricated sponge-like Mn₃O₄ nanomaterials using a precipitation method, which exhibited a high initial reversible capacity of 869 mA h g⁻¹ and good cyclability over 40 cycles.¹² Despite achieving better performance, it should be noted that these Mn₃O₄ nanomaterials are usually prepared using hydrothermal/solvothermal, co-precipitation, electrospinning and sol–gel techniques, etc.^10–20 These methods always involve high-temperature processing and excessive use of capping or organic agents resulting in low yields. In addition, although the capacity and cyclability of Mn₃O₄ materials have been optimized to some extent, the performance has usually only been tested within 100 cycles. From the perspective of practical applications, there needs to be further investigations that explore the simple preparation of highly active Mn₃O₄ anode materials for LIBs.

Therefore, it is highly desirable to develop a simple method with mild operating conditions, good control of the morphology, and high through-put. A corrosion route was previously reported by our group to prepare nanostructured metal oxides, such as Fe₃O₄ and Co₃O₄.^21,22 This method only requires the immersion of the M/Al (M = Fe, Co, etc.) alloy in a NaOH solution under ambient atmospheric conditions and at room temperature. It has been demonstrated to be an effective and simple method for fabricating metal oxides in high yields with no organic agents used.

The aim of this paper is to design a Mn₃O₄ anode nanomaterial in order to obtain an excellent electrochemical response. In a typical experiment, Mn₃O₄ octahedra were conveniently prepared through dealloying of an Mn/Al alloy in NaOH solution at room temperature. In addition, the as-made sample was then used as an anode material for a LIB in order to investigate its electrochemical properties, such as capacity, cycle life, rate capability, etc. Detailed characterization demonstrated that the Mn₃O₄ octahedra performed excellently, in particular they exhibited an ultralong cycle life.

2. Experimental section

2.1 Sample preparation

Mn/Al alloy foils with thicknesses of about 50 μm were prepared by refining high purity Mn and Al (99.9%) in an arc-furnace, followed by melt-spinning under a protective nitrogen atmosphere. The atomic content of Al in all alloy foils is 85%. And then Mn/Al alloy foil was placed into NaOH solution with different concentrations at room temperature for 48 h. Finally, the products were washed several times with ultra-pure water (18.2 MΩ) and dried at room temperature in air.

2.2 Sample characterization

Powder X-ray diffraction (XRD) was carried out with a Bruker D8 advanced X-ray diffractometer using Cu Kα radiation at a step rate of 0.04° s⁻¹. The morphology and structure of the samples were observed through field emission scanning electron microscopy (SEM, JEOL JSM-7600F). Transmission electron microscopy (TEM) images were obtained with a JEM-2100 high-resolution transmission electron microscope (200 kV). Cyclic voltammetry (CV) tests were performed using a Lanlike LK2005A Electrochemical Interface with a scan rate of 0.1 mV s⁻¹. Electrochemical impedance spectroscopy (EIS) was carried out using a Princeton Applied Research spectrometer by applying an alternating current voltage of 10 mV in the frequency range from 0.01 to 100 kHz. The open circuit voltage for the EIS measurements was about 2.86 V.

2.3 Electrode preparation and electrochemical tests

The electrochemical behavior versus Li was measured using coin-type cells (size: 2032). To prepare the working electrodes, the active Mn₃O₄ powders, acetylene black and carboxymethyl cellulose (CMC) were mixed in a weight ratio of 6 : 2 : 2 in ultrapure water. After the slurry was milled (QM-3SP2 Planetary Ball Mill) for 2 hours, it was coated onto a piece of Cu foil and dried under vacuum at 80 °C for 12 h. The separator was a Celgard 2300 microporous membrane. The electrolyte was composed of 1 mol L⁻¹ LiPF₆ dissolved in ethylene carbonate (EC)–dimethyl carbonate (DMC)–ethylene methyl carbonate (EMC) with the volume ratio of 1 : 1 : 1. The cells were constructed in an argon-filled humidity-free glove box. The cells were cycled galvanostatically between 0.05 and 3 V using a Land CT2001 battery tester at 25 °C.

3. Results and discussion

Fig. 1a shows the XRD pattern of the Mn/Al source alloy, which suggests that the alloy consists of an Al₈Mn₅ alloy phase (JCPDS card 49-1279) and pure Al (JCPDS card 65-2869). The atom content of the feed is around 85% Al during the refining process of the source alloy, so the excess of Al is actually expected, which could benefit the corrosion process by accelerating the reaction rate in NaOH solution. This is in accordance with our early work concerning an Fe/Al alloy.²¹ In the experimental process, NaOH solution was used to selectively remove aluminum from the Mn/Al alloy. Fig. 1b shows the XRD pattern of the resulting sample after etching of the source alloy foil with a 1 mol L⁻¹ NaOH solution for 48 h at room temperature. Based on the XRD analysis, it is interesting to note that the product has a Mn₃O₄ (JCPDS 24-0734) crystal structure instead of pure elemental Mn. According to our earlier study on preparing Fe₃O₄ octahedra using a similar method,²¹ when the Al atoms in the alloy are selectively etched using OH⁻, the remaining metal atoms are less coordinated and exposed to the OH⁻ ions and oxygen-containing atmosphere. In such an alkaline environment, these fresh metal sites are so active that they instantaneously combine with OH⁻ and undergo natural oxidation by O₂, resulting in the formation of oxides.

Fig. 1 XRD patterns of (a) the MnAl source alloy and (b) the product after dealloying the MnAl alloy in 1 mol L⁻¹ NaOH for 48 h.

The morphology of Mn₃O₄ was characterized by SEM as shown in Fig. 2. It is found that the sample consists of well dispersed octahedra (Fig. 2a), and the detailed structure is shown in the high magnification SEM image (Fig. 2b). We can easily observe that the surfaces of the octahedra are smooth and their edge lengths are mainly distributed between 300–400 nm. In order to understand the formation process of this nanostructure, the morphological evolution of the sample was also monitored at an earlier corrosion stage (Fig. 2c). It can be seen that some thick sheets with rough surfaces emerged in the product after dealloying the Mn/Al alloy for 12 h (marked with ellipses in Fig. 2c). It should be noted that many particles existed on the surfaces of the sheets. Meanwhile, octahedral-like particles have begun to appear (see the arrow in Fig. 2c) with sizes of about 100–300 nm, which is because the freshly dealloyed Mn atoms gradually grow into particles upon aggregation. These rough sheet-like aggregates should be the incompletely corroded Mn/Al alloy. When the reaction time is prolonged to 48 h, these thick sheets disappear and the Mn particles are gradually transformed into lots of octahedra as the final product (Fig. 2b). It is considered that, during the whole dealloying process, Al is gradually dissolved, while Mn goes on to be oxidized into Mn₃O₄ by OH⁻ in the alkaline environment and dissolved O₂, which is accompanied by a process of crystal growth. Therefore, based on the above analysis, the whole experimental process and corresponding mechanism can be illustrated by the following scheme in Fig. 3.

Fig. 2 SEM images of the Mn₃O₄ samples after dealloying for (a and b) 48 h and (c) 12 h in 1 mol L⁻¹ NaOH solution.

Fig. 3 The schematic fabrication of the Mn₃O₄ octahedra.

The structure, formation and evolution of the Mn₃O₄ octahedra were further explored by carrying out a series of experiments in NaOH solutions with different concentrations. In a typical experiment, Mn/Al alloy was dispersed in 2, 5 and 10 mol L⁻¹ NaOH solutions for 48 h, and the corresponding SEM images of the products are shown in Fig. 4. From Fig. 4a, a caking phenomenon can be obviously observed for the octahedra under the corrosive conditions of the 2 mol L⁻¹ NaOH solution. This is because the higher concentration of OH⁻ accelerates the etching of Al, resulting in more fresh and active Mn atoms being exposed to the alkaline solution. The faster etching rate and oxidation are not beneficial for the formation of regular Mn₃O₄ octahedra, therefore, the product tends to aggregate rather than to form well-dispersed octahedra. This phenomenon becomes more evident with further increases in the concentration to 5 and 10 mol L⁻¹. As shown in Fig. 4b and c, the octahedral morphology no longer exists, and large aggregates become the main products. Therefore, the lower concentration of NaOH solution is favorable for the fabrication of Mn₃O₄ octahedra. It is clear that the concentration of the NaOH solution plays an important role in the formation of this octahedral structure.

Fig. 4 SEM images of the Mn₃O₄ samples after dealloying for 48 h in (a) 2, (b) 5 and (c) 10 mol L⁻¹ NaOH solutions.

The as-prepared Mn₃O₄ octahedral sample was used as an anode material in order to investigate its electrochemical properties for Li-ion batteries (LIBs). Fig. 5a shows its cycling performance at the rates of 100 and 300 mA g⁻¹ from 0.05 to 3.0 V over 500 cycles. When the current density was 100 mA g⁻¹, the electrode delivered an initial capacity of 918.3 mA h g⁻¹. In addition, the Coulombic efficiency (the ratio between the charge and discharge capacities) for the first cycle was 58.5%, and the related first reversible capacity was 536.8 mA h g⁻¹. The reason for such a big capacity loss is mainly due to two factors: one is the large volume change that occurs during the reaction between Mn₃O₄ and Li, which is common for anode materials based on conversion reactions for lithium storage;^23,24 the other is the irreversible formation of a solid electrolyte interphase (SEI) film by degradation of the electrolyte on the surface of Mn₃O₄. The capacity then dropped gradually to ∼335 mA h g⁻¹ at the ∼50^th cycle, but the Coulombic efficiency increased sharply to 93.6% at the second cycle, and was maintained between 97–100% in the following several hundred cycles (see Fig. 5b). It is interesting to note that the capacity exhibited an increasing trend from ∼50^th cycle upwards, and remained at ∼746 mA h g⁻¹ between the 300^th–500^th cycles, indicating an excellent cycling stability. The retentive capacity at 500^th cycle was calculated to be 81.3% relative to the initial value. Such a rising trend of capacity between the 50^th–300^th cycles is normally observed for transition metal oxides and can be attributed to the reversible growth of a polymeric gel-like film that results from kinetically activated electrolyte degradation. This has been widely reported and is well documented throughout the literature.^25–28

Fig. 5 (a) Cycling behavior and (b) Coulombic efficiency of the Mn₃O₄ anode at different current densities.

When the current density was increased to 300 mA g⁻¹, the corresponding initial capacity was 806.6 mA h g⁻¹, with a first cycle Coulombic efficiency of 57.3%. In the following cycles, the capacity gradually decreased to 300.5 mA h g⁻¹ at the 100^th cycle, but the Coulombic efficiency reached 93.7% for the second cycle, and was no lower than 97% in the following several hundred cycles (see Fig. 5b). It is found that the capacity increased quickly during 101^th–300^th cycles. This rising trend was similar to the performance at the rate of 100 mA g⁻¹. In the latter 200 cycles, the capacity was stable with the value holding at ∼620 mA h g⁻¹. After the 500^th cycle, the sample exhibited a capacity retention of 77.8% relative to the value of the 1^st cycle. Based on the above results, the Mn₃O₄ octahedra have an ultralong cycle lifetime, indicating their great application potential as anode materials for LIBs with the advantages of unique performance and easy preparation.

In addition, we made a comparison of the electrochemical performance of our Mn₃O₄ octahedra with other previously reported Mn₃O₄ materials, which is summarized in Table 1. As shown, most of the previously reported Mn₃O₄ materials were tested at current densities below 200 mA g⁻¹. Although the initial capacity of the Mn₃O₄ octahedra is not the highest of all of the Mn₃O₄ materials, it performs better than most samples at similar rates. The cycling stabilities of the Mn₃O₄ anodes were investigated in these reports, but all of the tests were limited to within 100 cycles. With the exception of the order-aligned nanomaterial, which maintains 96% of its initial capacity (No. 4, ref. 10) after 85 cycles, the other materials capacity retentions barely exceed 60% after 100 cycles with some even less than 10%. However, both in terms of cycling stability and discharge capacity, our Mn₃O₄ octahedra perform excellently. For example, its capacity retention reaches up to 81.3% and 77.8% at rates of 100 and 300 mA g⁻¹, respectively, even after 500 cycles. We would emphasise that the ultralong cycle life of our Mn₃O₄ octahedra is outstanding among all of the previously published results on Mn₃O₄ anode materials.

Table 1 Comparison of the electrochemical performance of Mn₃O₄ materials between this work and the previous reports

No.	Type of pure Mn3O4	Current density	Cycle number	Discharge capacity (mA h g^−1)	Capacity retention	Ref.
1	Octahedra	100 mA g^−1	500	918/1^st cycle	81.3%/500^th cycle	This work
				746/500^th cycle
2	Octahedra	300 mA g^−1	500	807/1^st cycle	77.8%/500^th cycle	This work
				628/500^th cycle
3	Nanorods	0.1 C (<100 mA g^−1)	100	1050/1^st cycle	10.3%/100^th cycle	15
				108/100^th cycle
4	Order-aligned nanostructure	940 mA g^−1	85	919/1^st cycle	96.0%/85^th cycle	10
				882/85^th cycle
5	Mesoporous stacked nanosheets	0.1 C (<100 mA g^−1)	60	824/1^st cycle	48.5%/60^th cycle	13
				400/60^th cycle
6	Fibers	0.5 mA cm^−2 (≤208 mA g^−1)	50	∼860/1^st cycle	52.3%/50^th cycle	14
				450/50^th cycle
7	Spongelike nanostructure	30 mA g^−1	40	1327/1^st cycle	60.3%/40^th cycle	12
				800/40^th cycle
8	Nanoparticles	40 mA g^−1	100	∼300/1^st cycle	38.3%/100^th cycle	7
				115/100^th cycle
9	Nanoparticles	40 mA g^−1	50	1300/1^st cycle	38.2%/50^th cycle	29
				496/50^th cycle
10	Nanoparticles	100 mA g^−1	50	∼900/1^st cycle	16.4%/50^th cycle	8
				148/50^th cycle
11	Nanoparticles	40 mA g^−1	50	974/1^st cycle	15.9%/50^th cycle	30
				155/50^th cycle
12	Nanoparticles	200 mA g^−1	50	∼710/1^st cycle	6.1%/50^th cycle	31
				43/50^th cycle
13	Unknown morphology	0.2C (<200 mA g^−1)	10	∼900/1^st cycle	27.8%/50^th cycle	16
				∼250/10^th cycle
14	Co-doped Mn3O4	0.2C (<200 mA g^−1)	10	∼950/1^st cycle	42.1%/50^th cycle	16
				∼400/10^th cycle

---

Fig. 6a shows the first and second charge–discharge curves for the tested cell at 300 mA g⁻¹ between 0.05 and 3 V. During the first discharge process, a sloping voltage is found in the voltage range higher than 0.39 V, which may be due to the formation of SEI film and the initial reduction of Mn₃O₄.¹² An obvious, well-defined voltage plateau is then observed around 0.39 V, corresponding to the reaction of lithium and Mn₃O₄.¹⁰ As shown in the first charge curve, a voltage plateau appears around 1.2 V. However, in the second charge–discharge cycle, the discharge profile is different from that in the first cycle. As demonstrated in Fig. 6a, only one voltage plateau is observed at ∼0.7 V, which is evidently higher and has more severe slope than the voltage plateau in the first discharge curve. In addition, an irreversible capacity loss is observed, which is mainly caused by the large volume expansion of Mn₃O₄ and the formation of SEI film. This is commonly observed in anode materials that are based on conversion or alloying reactions during lithium storage.^23,32–35 However, the difference between the first and second charge profiles is slight, both in terms of the voltage plateau and charge capacity.

Fig. 6 (a) Charge–discharge and (b) CV curves of the Mn₃O₄ octahedra.

In order to further understand the electrochemical details of the lithium insertion–extraction process, cyclic voltammetric tests (CV) were carried out between 0.05 and 3 V at a scan rate of 0.1 mV s⁻¹. Fig. 6b shows the first three CV curves of the cell with the octahedral Mn₃O₄ sample used as an anode material. In the first cathodic scan, there are weak peaks locate at 0.4–1.6 V, which disappear in the following cycles and correspond to the formation of an SEI layer due to the decomposition of the electrolyte on the Mn₃O₄ surface. In addition, the strong peak that appears at 0.17 V is related to the decomposition of Mn₃O₄ into Mn and Li₂O. Upon charge, one peak appears at 1.22 V in the first anodic scan, which can be assigned to the oxidation of Mn and decomposition of Li₂O. In subsequent cycles the position of the reduction peak shifts to ∼0.38 V, which is mainly related to drastic lithium driven structural modifications.¹⁰ However, the oxidation peak is very similar to that of the first anodic scan. When the cell is tested during a third cycle, the intensity and position of all of the peaks are approximately equal to those in the second cycle, suggesting the high reversibility of the Mn₃O₄ anode. These results are in accordance with the relevant data in previous reports.^10,30 Based on the aforementioned analysis and with reference to the relevant literature,^{6,7,10–13,29,30} the reaction mechanism between Li and Mn₃O₄ can be summarized by the following equation, which has been generally accepted:

Mn₃O₄ + 8Li⁺ + 8e⁻ ⇌ 3Mn + 4Li₂O

Electrochemical impedance spectroscopy (EIS) measurements were carried out and the typical Nyquist plots of the cells before and after charge–discharge cycling are shown in Fig. 7. In the case of the cell before cycling tests, the Nyquist plot comprises a depressed semicircle in the high frequency region and a slopping line in the low frequency region. The semicircle reflects the charge-transfer resistance (R_ct), and the inclined line represents the Warburg impedance (Z_w), which is related to the solid-state diffusion of Li⁺ in the electrode material.³⁶ After the cell is charged and discharged for 50 cycles, the composition of the Nyquist plot is different from that before cycling. This Nyquist plot comprises three parts: a semicircle that appears in the high frequency region, corresponding to the resistance due to Li⁺ migration through the SEI film (R_sf); another semicircle in the medium frequency region related to R_ct; and a sloping line in the low frequency range, which is related to Z_w. These results indicate that SEI films have formed during the charge–discharge process, which is in keeping with the above charge–discharge curves and CV analysis (Fig. 6).

Fig. 7 Impedance spectra of the Mn₃O₄ octahedra obtained before and after cycling.

We also investigated the rate performance of the Mn₃O₄ octahedra in order to evaluate its power capability. The Mn₃O₄ electrode was cycled at various current densities between 0.01 and 3 V, and Fig. 8 shows its related performance. The current density was increased stepwise from 100 to 300, 500, 1000 and 1500 mA g⁻¹, and the cell was tested for 10 cycles at each current density. It can be clearly observed that the capacity performance was stable at each rate. Even at 1500 mA h g⁻¹, the Mn₃O₄ anode could deliver a reversible capacity of about 240 mA h g⁻¹ and this remained steady. Such a high current rate performance is another very attractive feature of the Mn₃O₄ octahedra. Upon altering the current density back to 100 mA g⁻¹, an average capacity of 387 mA h g⁻¹ could be recovered, which was close to the capacity at the initial rate of 100 mA g⁻¹. The data demonstrate that the Mn₃O₄ octahedra have a good rate capability and great potential as a high-rate anode material for LIBs.

Fig. 8 The rate performance of the Mn₃O₄ octahedra (rate = 100–1500 mA g⁻¹).

An excellent cycling stability at a high current density is an important property for a suitable candidate anode material. So, we set the test rate to 1000 mA g⁻¹. Fig. 9 shows detailed capacity and Coulombic efficiency data over 500 cycles. As shown, the Mn₃O₄ octahedra delivered an initial capacity of 557.5 mA h g⁻¹, and exhibited a first cycle Coulombic efficiency of 57.2%. Subsequently, the capacity decreased to 314.3 mA h g⁻¹ with an increased Coulombic efficiency of 84.7%. The capacity then slowly reduced to 238.9 mA h g⁻¹ at the 56^th cycle. However, from then on the capacity began to rise, and reached 255 mA h g⁻¹ by the 70^th cycle. This dropping-rising part is highlighted by a circle and magnified around the arrow in Fig. 9. Over the next 430 cycles, the sample delivered an essentially constant capacity of ∼255 mA h g⁻¹, indicating its outstanding capacity stability. According to the actual capacity, the charge–discharge time is measured to be ∼4 minutes at this current density. However, the capacity of the Mn₃O₄ octahedra is highly stable at this fast charge–discharge rate and over the ultralong cycling process. This indicates the great value of our Mn₃O₄ anode for LIBs in high-power application fields.

Fig. 9 Cycling behaviour and Coulombic efficiency of the Mn₃O₄ anode at 1000 mA h g⁻¹.

4. Conclusions

In summary, a Mn₃O₄ material with an octahedral morphology was straightforwardly prepared at room temperature using a one-step dealloying method. It was found that a lower concentration of NaOH solution was most applicable for the preparation of the octahedral Mn₃O₄ nanostructure. In addition, the Mn₃O₄ octahedra exhibited good electrochemical properties when used as an anode material for Li-ion batteries. For example, they delivered a high initial capacity and underwent only a small loss of capacity during long cycling test up to 500 cycles, even at a high current density of 1000 mA g⁻¹. The performance is superior to other previously reported Mn₃O₄ materials, indicating the great application potential of the Mn₃O₄ octahedra as anode materials with high activity and easy preparation.

Acknowledgements

This work was supported by the National Science Foundation of China (51001053, 21271085) and the Doctoral Foundation of University of Jinan (XBS1314).

Notes and references

1. M. S. Whittingham, Chem. Rev., 2004, 104, 4271.
2. W. J. Zhang, J. Power Sources, 2011, 196, 13.
3. H. H. Zheng, Q. T. Qu, L. Zhang, G. Liu and V. S. Battalion, RSC Adv., 2012, 2, 4904.
4. X. F. Li and C. L. Wang, J. Mater. Chem. A, 2013, 1, 165.
5. L. H. Zhuo, Y. Q. Wu, J. Ming, L. Y. Wang, Y. C. Yu, X. B. Zhang and F. Y. Zhao, J. Mater. Chem. A, 2013, 1, 1141.
6. M. A. Lowe, J. Gao and H. D. Abruña, J. Mater. Chem. A, 2013, 1, 2094.
7. H. L. Wang, L. F. Cui, Y. Yang, H. S. Casalongue, J. T. Robinson, Y. Y. Liang, Y. Cui and H. J. Dai, J. Am. Chem. Soc., 2010, 132, 13978.
8. Z. Q. Li, N. N. Liu, X. K. Wang, C. B. Wang, Y. X. Qi and L. W. Yin, J. Mater. Chem., 2012, 22, 16640.
9. M. Gu, I. Belharouak, A. Genc, Z. G. Wang, D. P. Wang, K. Amine, F. Gao, G. W. Zhou, S. Thevuthasan, D. R. Baer, J. G. Zhang, N. D. Browning, J. Liu and C. M. Wang, Nano Lett., 2012, 12, 5186.
10. J. Z. Wang, N. Du, H. Wu, H. Zhang, J. X. Yu and D. R. Yang, J. Power Sources, 2013, 222, 32.
11. C. J. Chae, J. H. Kim, J. M. Kim, Y. K. Sun and J. K. Lee, J. Mater. Chem., 2012, 22, 17870.
12. J. Gao, M. A. Lowe and H. D. Abruña, Chem. Mater., 2011, 23, 3223.
13. D. P. Dubal and R. Holze, RSC Adv., 2012, 2, 12096.
14. Q. Fan and M. Whittingham, Electrochem. Solid-State Lett., 2007, 10, A48.
15. X. P. Shen, Z. Y. Ji, H. J. Miao, J. Yang and K. M. Chen, J. Alloys Compd., 2011, 509, 5672.
16. D. Pasero, N. Reeves and A. R. West, J. Power Sources, 2005, 141, 156.
17. C. Chen, G. J. Ding, D. Zhang, Z. Jiao, M. H. Wu, C. H. Shek, C. M. L. Wu, J. K. L. Lai and Z. W. Chen, Nanoscale, 2012, 4, 2590.
18. K. N. Jung, J. I. Lee, S. Yoon, S. H. Yeon, W. Y. Chang, K. H. Shina and J. W. Lee, J. Mater. Chem., 2012, 22, 21845.
19. K. V. Sankar, D. Kalpana and R. K. Selvan, J. Appl. Electrochem., 2012, 42, 463.
20. Y. Tan, L. Meng, Q. Peng and Y. Li, Chem. Commun., 2011, 47, 1172.
21. C. X. Xu, R. Y. Wang, Y. Zhang and Y. Ding, Nanoscale, 2010, 2, 906.
22. Q. Hao, M. H. Li, S. Z. Jia, X. Y. Zhao and C. X. Xu, RSC Adv., 2013, 3, 7850.
23. X. W. Lou, D. Deng, J. Y. Lee, J. Feng and L. A. Archer, Adv. Mater., 2008, 20, 258.
24. C. M. Ban, Z. C. Wu, D. T. Gillaspie, L. Chen, Y. F. Yan, J. L. Blackburn and A. C. Dillon, Adv. Mater., 2010, 22, E145.
25. S. Laruelle, S. Grugeon, P. Poizot, M. Dolle, L. Dupont and J. M. Tarascon, J. Electrochem. Soc., 2002, 149, A627.
26. S. Grugeon, S. Laruelle, L. Dupont and J. M. Tarascon, Solid State Sci., 2003, 5, 895.
27. G. M. Zhou, D. W. Wang, F. Li, L. L. Zhang, N. Li, Z. S. Wu, L. Wen, G. Q. Lu and H. M. Cheng, Chem. Mater., 2010, 22, 5306.
28. J. S. Do and C. H. Weng, J. Power Sources, 2005, 146, 482.
29. L. Li, Z. P. Guo, A. J. Du and H. K. Liu, J. Mater. Chem., 2012, 22, 3600.
30. C. B. Wang, L. W. Yin, D. Xiang and Y. X. Qi, ACS Appl. Mater. Interfaces, 2012, 4, 1636.
31. S. Y. Liu, J. Xie, Y. X. Zheng, G. S. Cao, T. J. Zhu and X. B. Zhao, Electrochim. Acta, 2012, 66, 271.
32. M. Gu, Y. Li, X. L. Li, S. Y. Hu, X. W. Zhang, W. Xu, S. Thevuthasan, D. R. Baer, J. G. Zhang, J. Liu and C. M. Wang, ACS Nano, 2012, 6, 8439.
33. M. Gu, Z. G. Wang, J. G. Connell, D. E. Perea, L. J. Lauhon, F. Gao and C. M. Wang, ACS Nano, 2013, 7, 6303.
34. J. Y. Huang, L. Zhong, C. M. Wang, J. P. Sullivan, W. Xu, L. Q. Zhang, S. X. Mao, N. S. Hudak, X. H. Liu, A. Subramanian, H. Y. Fan, L. Qi, A. Kushima and J. Li, Science, 2010, 330, 1515.
35. X. H. Liu and J. Y. Huang, Energy Environ. Sci., 2011, 4, 3844.
36. K. M. Shaju, F. Jiao, A. Debart and P. G. Bruce, Phys. Chem. Chem. Phys., 2007, 9, 1837.

---