β-MnO₂ sacrificial template synthesis of Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ for lithium ion battery cathodes

Abstract

Lithium-rich manganese-based oxide Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ nanorods and polyhedrons have been successfully prepared using precursor β-MnO₂ nanorods as sacrificial templates at a calcination temperature of 750, 800, 850 or 900 °C. X-ray diffraction (XRD) and scanning electron microscope (SEM) measurements show that the elevated calcination temperatures help to improve the layered structure and average particle size of the target products. Rod-like Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ can be obtained at low calcination temperatures (i.e., 750 and 800 °C), which becomes polyhedral at 850 or 900 °C. As lithium ion battery cathodes, the Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ obtained at 850 °C shows the highest discharge capacity of 239.2 mA h g⁻¹ at 20 mA g⁻¹, and a stable discharge capacity of 92.8 mA h g⁻¹ at 1000 mA g⁻¹. The good electrochemical performances of the 850 °C sample should be attributed to the better crystal structure and/or the more appropriate particle size compared with those of other samples.

---

1. Introduction

The development of cathode materials with high capacity, long cycle life and low cost has been one of the most important subjects for high-performance lithium ion batteries (LIBs).^1–3 Recently, the lithium-rich manganese-based oxides xLi₂MnO₃·(1 − x)LiMO₂ (M = Ni, Co, Mn, Fe, Cr, Ni_1/2Mn_1/2, Ni_1/3Co_1/3Mn_1/3…) have attracted more and more attention owing to their high reversible capacity (230–300 mA h g⁻¹).^4–9 As for a solid solution electrode, a Li₂MnO₃ component is presented to stabilize the crystal structure and enhance discharge capacity through extracting the lithium ions, which is concomitant with the release of oxygen at a charge voltage within 4.5–4.6 V and facilitates the formation of active MnO₂. Therefore, these solid solutions could reach a high special capacity and retain a good cycling stability upon cycling.^4–9

Li_1.2Ni_0.13Co_0.13Mn_0.54O₂, also denoted as 0.5Li₂MnO₃·0.5LiNi_1/3Co_1/3Mn_1/3O₂, stands out from other solid solution materials due to its high discharge capacity, moderate cycling performance and good rate capability.^10,11 Its improved electrochemical performances can be attributed to a reasonable ratio of Li₂MnO₃ to LiNi_1/3Co_1/3Mn_1/3O₂ components and the appropriate existence of the element Co.^12–16 Therefore, conventional methods such as sol–gel,^17,18 sucrose combustion,¹⁹ molten salt,^20,21 co-precipitation^19,22,23 and polymer gel²⁴ have been successfully introduced for the preparation of this most promising solid solution. These show that the structural and electrochemical properties of the as-prepared materials are determined by the synthesis methods and/or preparation conditions.

Sacrificial template technology has proven to be one of the most successful routes to prepare electrode materials with a controlled morphology and/or a tunable particle size.^25–28 As far as the synthesis of Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ is concerned, the co-precipitated precursors of Ni_0.13Co_0.13Mn_0.54(OH)_0.8 and Ni_0.13Co_0.13Mn_0.54(CO₃)_0.8 have been successfully treated as sacrificial templates therein.²³ Considering the high manganese content of lithium-rich manganese-based oxides, a manganese compound (e.g., MnCO₃ or MnO₂) can also be regarded as a sacrificial template through the chemical insertion of Li and other metal ions during preparation.²⁹ Inspired by the fact that rod-like β-MnO₂ can function as a sacrificial template to generate the crystallographically single crystal of LiMn₂O₄, LiNi_1/3Co_1/3Mn_1/3O₂ or LiNi_1/2Mn_3/2O₄,^25–28 this approach to the synthesis of Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ solid solution has been performed for the first time. The effects of calcination temperature on the structures, morphologies, particle sizes and electrochemical properties of the as-prepared products are comparatively investigated and discussed in the context.

2. Experimental

Synthesis of Li_1.2Ni_0.13Co_0.13Mn_0.54O₂

All the chemical reagents are of A.R. grade and were used without further purification. Firstly, precursor β-MnO₂ nanorods are prepared by a hydrothermal route according to Wang's report.³⁰ Next, 0.2934 g freshly prepared β-MnO₂, 0.2022 g Ni(CH₃COO)₂·4H₂O, 0.2024 g Co(CH₃COO)₂·4H₂O and 0.2936 g Li₂CO₃ (6% excess) are well mixed in an agate mortar using a small amount of ethanol as a dispersant. Finally, after the evaporation of ethanol, the resulting mixtures are decomposed at 450 °C for 4 h, and subsequently crystallization-treated at a temperature of 750, 800, 850 or 900 °C for 10 h in a muffle furnace, and then cooled naturally to room temperature.

Crystal characterization

X-ray diffraction (XRD) patterns are collected on a Rigaku D/max-2400 powder X-ray diffractometer (XRD, 0.04° step per s). Scanning electron microscope (SEM) (JEOL JSM-7600F, 5 kV) measurements are conducted to characterize the morphologies and particle sizes of precursor β-MnO₂ and the series of Li_1.2Ni_0.13Co_0.13Mn_0.54O₂.

Electrochemical characterization

CR 2032 coin cells of Li_1.2Ni_0.13Co_0.13Mn_0.54O₂/Li are used for electrochemical experiments performed at room temperature. The working electrodes with a loading density of 2.6 ± 0.4 mg cm⁻² are prepared as described in the following: after the mixing of the electrode materials, acetylene black and poly(vinyl difluoride) (PVDF) at a weight ratio of 80 : 10 : 10, the resulting mixtures are slurried with N-methyl-2-pyrrolidone, pasted onto aluminum foils, cut into discs with a diameter of 12 mm, and then dried at 80 °C for 5 h. Glass fiber and commercial LBC 305-01 LiPF₆ solution are used as the separator and electrolyte, respectively, and the model cells are assembled in an argon-filled glove box. Galvanostatic cycling tests are conducted on a Land CT2001A battery system within a voltage range of 2.0–4.7 V vs. Li⁺/Li.

3. Results and discussion

The XRD pattern and SEM image of the precursor β-MnO₂ nanorods are shown in Fig. 1. All the diffraction peaks in Fig. 1a can be clearly assigned to the standard data of tetragonal MnO₂ (JCPDS 24-0735). Fig. 1b exhibits a rod-like structure with the average length of 0.6–1.5 μm and the average width of about 0.1 μm. Under SEM observation, the nanometer dimension of the rod-like precursors can be further proved using the wide XRD diffraction peaks shown in Fig. 1a.

Fig. 1 (a) XRD pattern and (b) SEM image of the precursor β-MnO₂ nanorods.

In order to investigate the sacrificial templating effect of the β-MnO₂ nanorods on the formation of lithium-rich manganese-based oxide Li_1.2Ni_0.13Co_0.13Mn_0.54O₂, the calcination temperature is of crucial importance. XRD patterns of the series of Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ samples obtained at different temperatures are shown in Fig. 2, with each diffraction peak indexed according to those of crystalline α-NaFeO₂. That is, the XRD pattern of each target sample coincides well with the standard data of the hexagonal structure of layered LiMO₂ (i.e., M = Li_1/3Mn_2/3, Ni_1/3Co_1/3Mn_1/3).^4–11 Also, the asterisk-marked weak diffraction peaks in the 2θ range of 20° and 25° belong to the XRD characteristics for the periodic occupation of Li⁺ ions in the transition metal layers of crystalline LiMO₂, and the resulting LiMn₆-type cation arrangements indicate the coexistence of both crystalline Li₂MnO₃ (also referred to as layered Li(Li_1/3Mn_2/3)O₂) and LiNi_1/3Co_1/3Mn_1/3O₂ phases in the solid solutions.^4–11

Fig. 2 XRD patterns of the series of Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ materials obtained at the calcination temperatures of (a) 750, (b) 800, (c) 850 and (d) 900 °C.

From Fig. 2, the estimated lattice parameters of the unit cell, the peak intensity ratio of I₍₀₀₃₎/I₍₁₀₄₎ and the full width at half maximum (FWHM) of the (003) crystal plane are summarized in Table 1. For the series of Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ materials, the lattice parameters a and c are about 2.845 and 14.210 Å, respectively, which are close to those of layered LiNi_1/3Co_1/3Mn_1/3O₂.³¹ Generally speaking, the peak intensity ratio of I₍₀₀₃₎/I₍₁₀₄₎ can be used to measure the structure order and cation mixing; the higher ratio means the better layered structure and the lower cation mixing.^32,33 As shown in Table 1, when the calcination temperature is raised from 750 to 900 °C, the ratio increases from 0.86 to 1.67, suggesting an improvement of layered structure therein.³⁴ According to the FWHM₍₀₀₃₎ values in Table 1, the following estimated particle sizes can be calculated for each sample: 23.28 (750 °C), 39.81 (800 °C), 50.72 (850 °C) and 60.73 nm (900 °C). This is consistent with a previous report, that the increase of solid-state reaction temperature helps to improve the average particle size of the layered compound.³⁴

Table 1 Lattice parameters, I₍₀₀₃₎/I₍₁₀₄₎ ratios and FWHM₍₀₀₃₎ values of the series of Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ materials estimated from XRD data and space group R3̄m

Temperature (°C)	a (Å)	c (Å)	v (Å^3)	I(003)/I(104)	FWHM(003)
750	2.843	14.183	99.28	0.86	0.342
800	2.847	14.203	99.70	1.19	0.200
850	2.848	14.217	99.87	1.52	0.157
900	2.847	14.215	99.78	1.67	0.129

---

SEM images of the series of Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ materials obtained from sacrificial β-MnO₂ nanorods at different calcination temperatures are shown in Fig. 3. At a low temperature (i.e., 750 °C), the resulting Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ retains the rodlike structure of precursors (Fig. 3a and b). These nanorods are endowed with a larger width and rougher surface compared with those of precursor β-MnO₂, which can be attributed to the effective insertion of metal ions (i.e. Li⁺, Ni²⁺ and Co³⁺ ions) during the calcination process. At a moderate calcination temperature of 800 °C, the resulting Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ nanorods become the relatively big in width and the relatively small in length, and simultaneously the surface of each particle becomes smooth along with the increasing calcination temperature (Fig. 3c–f). Thus, at or 850 or 900 °C polyhedral nanoparticles are sampled, and the nanocrystalline Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ exhibits clear edges and corners (Fig. 3g and h).

Fig. 3 SEM images of Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ obtained at the calcination temperature of (a and b) 750, (c and d) 800, (e and f) 850 and (g and h) 900 °C.

According to the results listed above, a plausible functionalization of precursor β-MnO₂ nanorods for the formation of different Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ samples can be schematically shown in Fig. 4. During the first stage of a calcination treatment, the β-MnO₂ nanorods play a role as sacrificial templates, and the metal elements of Li, Ni and Co appropriately insert into these rod-like structures. That is, even at a temperature of 800 °C or higher, thermodynamically unstable Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ nanorods may be initially formed through the gradual warming. Because the rod-like structure of layered compounds is relatively unstable,^31–34 during the second stage the intermediate Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ nanorods may experience further crystal growth and turn into polyhedral nanoparticles (Fig. 4).

Fig. 4 A schematic drawing for the sacrificial template effect of the β-MnO₂ nanorods on the formation of Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ polyhedra.

It is well-known that: (i) the orderly layered structure of Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ may provide a solid framework upon cycling and improve the discharge capacity by decreasing the initial irreversible capacity loss; (ii) to a certain extent the enlargement of the average particle size will block the effective diffusion of lithium ions and electrons within the layered lattices and decrease the specific capacity of the electrodes. Therefore, the obtained Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ series could exhibit both structure- and size-dependent electrochemical performances and give a maximum initial discharge capacity. Fig. 5 shows a comparison of the initial charge–discharge profiles of Li_1.2Ni_0.13Co_0.13Mn_0.54O₂/Li cells recorded at a low current density of 20 mA g⁻¹ within the electrochemical window of 2.0 and 4.7 V (vs. Li⁺/Li), and parts of the obtained electrochemical parameters are summarized in Table 2. At a calcination temperature of 750, 800, 850 or 900 °C, the resulting Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ separately gives a specific discharge capacity of 207.0, 227.9, 239.2 or 205.7 mA h g⁻¹ with a coulombic efficiency of 61.7%, 69.0%, 71.6% or 71.7% respectively (Table 2). These indicate an optimal calcination temperature for the sacrificial template synthesis of Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ of 850 °C.

Fig. 5 Initial charge–discharge voltage profiles of Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ electrodes recorded at 20 mA g⁻¹ between 2.0 and 4.7 V.

Table 2 Electrochemical parameters of the series of Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ electrodes recorded at 20 mA g⁻¹ between 2.0 and 4.7 V

Temperature (°C)	Charge capacity (mA h g^−1)	Discharge capacity (mA h g^−1)	Coulombic efficiency (%)	Midpoint discharge voltage (V vs. Li^+/Li)
750	335.8	207.0	61.7	3.47
800	330.3	227.9	69.0	3.53
850	334.0	239.2	71.6	3.58
900	287.1	205.7	71.7	3.64

---

As shown in Fig. 5, all charge–discharge profiles show typical electrochemical behaviors of various xLi₂MnO₃·(1 − x)LiMO₂ (e.g., M = Ni_1/3Co_1/3Mn_1/3 or Ni_yCo_zMn_1−y−z) solid solutions. During the charging at a potential below 4.4 V, the capacity can be ascribed to the deintercalation of Li⁺ ions accompanied by the oxidation of Ni²⁺ ions within the active LiMO₂ phases. The plateau near 4.5–4.6 V corresponds to both the loss of Li₂O from Li₂MnO₃ phases and the oxidation of Co³⁺ ions within the active components,^4–12 however, the corresponding 4.6 V plateau cannot be found from the discharge profiles (Fig. 5). These indicate that the component Li₂MnO₃ won't be recovered and the substitutional MnO₂ → LiMnO₂ transformation can account for the high-capacity feature of lithium rich manganese-based oxides.^4–14

Owing to the initial activation of Li₂MnO₃ at 20 mA g⁻¹ within 2.0 and 4.7 V, it is better to study the cycling performances of these Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ samples. As shown in Fig. 6, the 2nd specific discharge capacities of the Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ series obtained at 750, 800, 850 and 900 °C are 167.1, 170.3, 181.3 and 154.6 mA h g⁻¹ respectively at a current density of 100 mA g⁻¹. Over 70 cycles, the residual values are 118.2, 147.6, 163.1, and 148.3 mA h g⁻¹, giving a retention ratio of 70.7%, 86.7%, 90.0% and 95.9%. Apparently, the elevated temperature can effectively improve the electrochemical cycling stability of Li_1.2Ni_0.13Co_0.13Mn_0.54O₂, and the improvement should be attributed to the layered structure which is enhanced with the increasing calcination temperature.

Fig. 6 Cycling performances of different Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ electrodes.

The 2nd, 10th, 25th and 50th charge–discharge profiles and corresponding differential capacity (dQ/dV) profiles of Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ obtained at 750 and 850 °C are shown in Fig. 7. Both the midpoint discharge (or charge) voltages (Fig. 7a and c) and the differential capacities (Fig. 7b and d) change upon cycling, and the decays of the 750 °C sample are more serious than those of 850 °C sample, as marked by arrows. Three pairs of redox peaks can be clearly observed in Fig. 7d, corresponding to redox reactions of Co⁴⁺/Co³⁺, layered Ni⁴⁺/Ni²⁺ and spinel Mn⁴⁺/Mn³⁺. As the cycles increase, the reduction peak of layered Ni⁴⁺/Ni²⁺ near 3.8 V weakens gradually, and the peak of spinel Mn⁴⁺/Mn³⁺ around 3.0 V becomes stronger, indicating a continuous phase transformation from layered LiMnO₂ to spinel LiMn₂O₄.¹⁸ This phase transformation is inevitable, resulting from the same oxygen stacking of layered and spinel structures and the existence of vacancies.^35–37 What's more, the 850 °C sample has clearly distinct redox peaks compared with 750 °C sample, suggesting that the former has a relatively solid framework upon cycling.

Fig. 7 The 2nd, 10th, 25th and 50th charge–discharge voltage profiles and corresponding dQ/dV curves of Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ samples obtained at (a and b) 750 and (c and d) 850 °C.

The rate capabilities of different Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ electrodes are revealed in Fig. 8. It is shown that, after the initial activation of Li₂MnO₃ at the 1st cycle, the discharge capacity of each sample gradually decreases with the increasing current rate from 20 mA g⁻¹ to 1000 mA g⁻¹. The 22nd specific discharge capacity is 118.2 (750 °C), 127.5 (800 °C), 128.4 (850 °C) or 110.2 (900 °C) mA h g⁻¹ at 400 mA g⁻¹, and the 32nd reversible capacity at 1000 mA g⁻¹ retains a value of 86.2 (750 °C), 87.9 (800 °C), 92.8 (850 °C) or 86.9 (900 °C) mA h g⁻¹. Also from Fig. 8, when the current density goes back to 20 mA g⁻¹, the 52nd cycle discharge capacity returns to a high value of 195.8 (750 °C), 209.0 (800 °C), 225.0 (850 °C) or 220.3 (900 °C) mA h g⁻¹. The calcination temperature exerts an effect on the rate capability of a solid solution, and the recovery ability of 850 °C or 900 °C sample is better than those of the other two samples.

Fig. 8 Rate capability of different Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ electrodes.

4. Conclusions

In summary, the β-MnO₂ sacrificial template synthesis, a plausible formation mechanism and the structure–function relationships of Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ nanostructures are presented. With the increase of calcination temperature, the resulting decreased particle size and improved layered structure can help to enhance the electrochemical performance of Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ as a LIB cathode. The optimal temperature is approximately set around 850 °C and the resulting solid solution possesses major polyhedral shapes and a high discharge capacity of 239.2 mAh g⁻¹ at 20 mA g⁻¹. However, the relatively low coulombic efficiency of optimal Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ (i.e., 71.6%) still deserves to be improved in future.

Acknowledgements

The authors thank the financial support from the National Basic Research Program of China (2011CB935900) and from Shandong Province (ZR2012BM001).

References

1. B. L. Ellis, K. T. Lee and L. F. Nazar, Chem. Mater., 2010, 22, 691.
2. J. B. Goodenough and Y. S. Kim, Chem. Mater., 2010, 22, 587.
3. H. J. Yu, Y. R. Wang, D. Asakura, E. Hosono, T. Zhang and H. S. Zhou, RSC Adv., 2012, 2, 8797.
4. M. M. Thackeray, C. S. Johnson, J. T. Vaughey, N. Li and S. A. Hackney, J. Mater. Chem., 2005, 15, 2257.
5. M. M. Thackeray, S. H. Kang, C. S. Johnson, J. T. Vaughey, R. Benedek and S. A. Hackney, J. Mater. Chem., 2007, 17, 3053.
6. P. He, H. J. Yu and H. S. Zhou, J. Mater. Chem., 2012, 22, 3680.
7. H. J. Yu and H. S. Zhou, J. Phys. Chem. Lett., 2013, 4, 1268.
8. Z. H. Tang, Z. X. Wang, X. H. Li and W. J. Peng, J. Power Sources, 2012, 204, 187.
9. H. J. Yu, R. Ishikawa, Y. G. So, N. Shibata, T. Kudo, H. S. Zhou and Y. Ikuhara, Angew. Chem., Int. Ed., 2013, 52, 5969.
10. J. Wang, B. Qiu, H. L. Cao, Y. G. Xia and Z. P. Liu, J. Power Sources, 2012, 218, 128.
11. Z. Y. Wang, E. Z. Liu, C. N. He, C. S. Shi, J. J. Li and N. Q. Zhao, J. Power Sources, 2012, 236, 25.
12. F. Amairaj, D. Kovacheva, M. Talianker and L. Zeili, J. Electrochem. Soc., 2010, 157, A1121.
13. C. S. Johnson, N. C. Li and M. M. Thackeray, Chem. Mater., 2008, 20, 2095.
14. F. L. Liu, S. Zhang, C. Deng, Q. Wu, M. Zhang, F. L. Meng and Y. H. Sun, J. Electrochem. Soc., 2012, 159, A1591.
15. S. H. Kang, P. Kempgens, S. Greenbaum, K. Amine and M. M. Thackeray, J. Mater. Chem., 2007, 17, 2069.
16. C. H. Zhao, W. P. Kang, R. Liu and Q. Shen, RSC Adv., 2013, 2, 2362.
17. K. C. Jiang, X. L. Wu, Y. X. Yin, J. S. Lee and J. S. Kim, ACS Appl. Mater. Interfaces, 2012, 4, 4858.
18. T. L. Zhao, S. Chen, L. Li, X. F. Zhang, R. J. Chen, I. Belharouak, F. Wu and K. Amine, J. Power Sources, 2013, 208, 206.
19. J. M. Zheng, X. B. Wu and Y. Yang, Electrochim. Acta, 2011, 56, 3071.
20. J. L. Liu, L. Chen, M. Y. Hou, F. Wang, R. C. Che and Y. Y. Xia, J. Mater. Chem., 2012, 22, 25380.
21. S. J. Shi, J. P. Tu, Y. Y. Tang, X. Y. Liu, X. Y. Zhao, X. L. Wang and C. D. Gu, J. Power Sources, 2013, 241, 186.
22. H. J. Kim, H. G. Jung, B. Scrosati and Y. K. Sun, J. Power Sources, 2012, 203, 115.
23. J. H. Lim, H. Bang, K. S. Lee, K. Amine and Y. K. Sun, J. Power Sources, 2009, 189, 571.
24. C. H. Zhao, W. P. Kang, Q. B. Xue and Q. Shen, J. Nanopart. Res., 2012, 14, 1240.
25. H. W. Lee, P. Muralidharan, R. Ruffo, C. M. Mari, Y. Cui and D. K. Kim, Nano Lett., 2010, 10, 3852.
26. E. Hosono, T. Kudo, I. Honma and H. Matsuda, Nano Lett., 2009, 9, 1045.
27. Y. M. Liu, B. L. Chen, F. Cao, X. Z. Zhao and J. K. Yuan, J. Mater. Chem., 2011, 21, 10437.
28. S. W. Kim, H. W. Lee, P. Muralidharan, D. H. Seo, W. S. Yoon, D. Y. Kim and K. Kang, Nano Res., 2011, 4, 505.
29. Y. Jiang, Z. Yang, W. Luo, X. L. Hu and Y. H. Huang, Phys. Chem. Chem. Phys., 2013, 15, 2954.
30. X. Wang and Y. D. Li, J. Am. Chem. Soc., 2002, 124, 2880.
31. X. Y. Han, Q. F. Meng, T. L. Sun and J. T. Sun, J. Power Sources, 2010, 195, 3047.
32. R. Santhanam, P. Jones, A. Sumana and R. Rambabu, J. Power Sources, 2010, 195, 7391.
33. P. Manikandan, M. V. Ananth, T. P. Kumar, M. Raju, P. Periasamy and K. Manimaran, J. Power Sources, 2011, 196, 10148.
34. J. W. Wen, D. W. Zhang, Y. C. Teng, C. H. Chen and Y. Xiong, Electrochim. Acta, 2010, 55, 2306.
35. B. H. Song, Z. W. Liu, M. O. Lan and L. Lu, Phys. Chem. Chem. Phys., 2012, 14, 12875–12883.
36. N. Yabuuchi, K. Yoshii, S. T. Myung, I. Nakai and S. Komaba, J. Am. Chem. Soc., 2011, 133, 4404–4419.
37. A. R. Armstrong, N. Dupre, A. J. Paterson, C. P. Grey and P. G. Bruce, Chem. Mater., 2004, 16, 3106–3118.

---