Li-rich layered Li_1.2Mn_0.54Ni_0.13Co_0.13O₂ derived from transition metal carbonate with a micro–nanostructure as a cathode material for high-performance Li-ion batteries

Abstract

Compared to commercialized cathode materials, Li-rich layered oxide exhibits a superior mass energy density. However, owing to its low tap/press density, the advantage of its volume energy density is not as obvious as that of its mass energy density, which limits its applications in some volume-constrained fields. It has been shown that the morphology of the precursor is critical to the performances of the final product. Here, solvothermal and co-precipitation methods were adopted to synthesize transition metal carbonate balls with micro-size particles to obtain high-density Li-rich layered oxides. The solvothermal synthesized carbonate showed a micro–nano hierarchical structure composed of nanoplates as subunits, and the co-precipitated synthesized carbonate just presents a micrometer quasi-ball morphology. The Li_1.2Mn_0.54Ni_0.13Co_0.13O₂ derived from the above solvothermal synthesized carbonate (ST-LMNCO) demonstrated an improved volume density of ∼14% compared to the one derived from the co-precipitated synthesized carbonate (CP-LMNCO). As for electrochemical performances, the ST-LMNCO exhibited a higher discharge specific capacitance (296.6 mA h g⁻¹ for the first discharge), a better rate performance (201.6 mA h g⁻¹ at 1C rate) and a better capacity retention capability (86.2% after 80 cycles) than the CP-LMNCO. The morphologies of the transition metal carbonates as starting materials significantly impacted the morphologies of the derived Li_1.2Mn_0.54Ni_0.13Co_0.13O₂ particles. Therefore, the carbonate with a hierarchical micro–nanostructure obtained from the solvothermal method is a promising precursor for high performance Li_1.2Mn_0.54Ni_0.13Co_0.13O₂.

---

1. Introduction

Motivated by the demands for high energy and power densities in portable electronic devices and electric vehicles, Li-ion batteries (LIBs) have attracted attention because of their high energy density and long cycle life.^1–3 The cathode material is one of the key constituents and largely determines the whole capacity of LIBs.^4–6 Among the commercialized cathode materials so far, Li-rich layered oxides, represented as xLi₂MnO₃·(1 − x)LiMO₂ (M = Co, Ni, Mn, etc.), is considered to be one of the most promising candidates for LIBs because of its high discharge capacity and high operating voltage.^7,8 However, some disadvantages still need to be solved before their practical application as a cathode material in LIBs, such as low tap/press density,^2,9 low initial coulombic efficiency,^8,10 poor rate performance,^11–13 and cycling capacity fading.^14–16

Compared to other reported cathode materials, Li-rich layered oxide shows a great mass energy density (>250 mA h g⁻¹).¹⁶ However, the advantage of its volume energy density is not obvious because of its low tap/press density.^2,17 Therefore, it would be worthwhile to increase the volume energy density while simultaneously increasing or maintaining the other values for Li-rich layered oxides. It has been shown that the morphology of the starting materials has a great influence on the morphologies of the Li-rich layered oxides and their electrochemical performances.^13,18,19 The unique micro–nanostructure combines both the advantages of microstructure and nanostructure,²⁰ making it a feasible and sensible strategy to improve the Li-rich layered oxide volume density by synthesizing a starting precursor with a designed micro–nano size.^21,22

Among the various synthesis methods,^23–29 hydrothermal and solvothermal methods are effective ways to realize an intentional structure.^5,29–32 Shi et al. synthesized a spherical Li-rich Li_1.2Mn_0.56Ni_0.16Co_0.08O₂ compound through a facile microwave hydrothermal method followed by a sintering reaction and found that the compound exhibited a discharge capacity of 235.6 mA h g⁻¹ even at a 1C rate.⁵ Fan et al. fabricated Li-rich layered microspheres with a high capacity and superior rate-capability through hydrothermal-assisted synthesis.³¹ Sun et al. synthesized a Li-rich layered material with a hierarchical micro/nanostructure using a solvothermal method, and the material showed a high reversible capacity and an excellent rate capability.³⁰ They also found that the samples synthesized by the solvothermal method exhibited a porous hierarchical microstructure and stoichiometric components compared to the partially aggregated nanoplates and Mn-deficiency for the samples created by the co-precipitation method.¹⁰ Li et al. reported that hollow LiMn₂O₄ could be synthesized at a relatively low sintering temperature from hydrothermal obtained MnO₂ nanocones/CNTs.³²

In this paper, a Li-rich layered cathode material with an improved density and excellent electrochemical performance was created using a solvothermal synthesized transition metal carbonate. Li_1.2Mn_0.54Ni_0.13Co_0.13O₂ was synthesized from a solvothermal carbonate (ST-LMNCO) and presented a higher tap density by ∼14% compared to the contrast sample obtained from the co-precipitated carbonate (CP-LMNCO) based on the results of the tap density tests. ST-LMNCO showed a higher initial discharge capacitance (296.6 mA h g⁻¹) and a better rate performance (201.6 mA h g⁻¹ at 1C rate) than the CP-LMNCO (275.9 and 172.3 mA h g⁻¹ for the corresponding part) under the same conditions. The main difference in the two samples comes from the morphology of the precursor carbonates. The ST-LMNCO is sintered from the carbonate with a hierarchical micro–nano structure. Thus, the designed transition metal carbonates with a micro–nanostructure obtained from the solvothermal method are a promising carbonate platform for Li_1.2Mn_0.54Ni_0.13Co_0.13O₂. The derived Li-rich layered oxide designed structure exhibits a high tap density and outstanding electrochemical performances.

2. Experimental

2.1 Synthesis of Li-rich layered cathode material

The carbonate for the Li-rich layered oxide with a micro–nanostructure was prepared by a solvothermal method, which is similar to the description in literature.¹¹ In a typical experiment, stoichiometric acetates (Mn(CH₃COO)₂, Ni(CH₃COO)₂, Co(CH₃COO)₂) were dissolved in ethanol under grinding within an agate mortar. Subsequently, the mixed solution was transferred into Teflon-lined autoclaves, and urea powder was then added into the solution under electromagnetic stirring (The amount of urea used refers to the literature.¹¹). Then, the sealed autoclaves were set in an oven, and the temperature was held at 200 °C for 20 h followed by cooling to room temperature. Finally, the purple carbonate was obtained after separating the precipitate from the mother liquid in a high speed centrifuge and drying the precipitate in a vacuum oven at 80 °C overnight. Scheme 1 illustrates the synthesis procedure for the carbonate by the solvothermal method.

Scheme 1 The schematic illustration of the carbonate with a micro–nanostructure synthesized by the solvothermal method.

For comparison, another carbonate was also synthesized by a co-precipitation method. Water solutions of the stoichiometric acetates and Na₂CO₃ were dropped into a three-necked bottle at the same time with strong stirring at a 50 °C under N₂. After co-precipitation, the reaction system was aged for 10 hours, followed by filtration and washing several times. After drying in a vacuum oven at 80 °C, the precipitate was obtained as a co-precipitated carbonate.

Li_1.2Mn_0.54Ni_0.13Co_0.13O₂ was prepared by sintering the as-produced carbonate with 5%-excess LiOH in a muffle at a temperature of 500 °C for 5 h and, subsequently, 850 °C for 20 h under air. Then, the Li-rich cathode material was obtained after naturally cooling to room temperature. The sample sintered from the solvothermal synthesized carbonate was labeled as ST-LMNCO, and the sample sintered from the co-precipitated carbonate was labeled as CP-LMNCO.

2.2 Characterization of structure and morphology

X-ray diffraction was carried on a diffraction instrument of Bruker (D8 Advance, Cu Ka radiation) with a 2θ range from 10° to 80°. The morphologies of the as-synthesized samples were examined by a Hitachi SU8010 with a working voltage of 3 kV and working distance of 5 mm. Transmission electron microscope images were obtained on a JOEL JEM-2010 electron microscope with an accelerating voltage of 200 kV. The tap density of the samples was determined by a tapping tester (BT-300, Dandong Bettersize Instrument Co. Ltd.). The tapping rate was 85 times per minute with an amplitude of 2 cm, and the cylinder was tapped 500 times. The tap density was then calculated by dividing the weight by the volume of the powder.

2.3 Electrochemical measurements

CR2025 type button-cells were assembled in an argon-filled glove box. To prepare the positive electrodes, the cathode materials, acetylene black and polyvinylidene fluoride (PVDF) with a ratio of 75 : 15 : 10 were blended into a slurry with N-methylpyrrolidone (NMP), and the slurry was coated onto Al foils followed by vacuum drying at 80 °C for 24 h. The pasted foil was cut into a disc (12 mm in diameter) with an active material loading of ∼3 mg cm⁻². LiPF₆ was dissolved in DEC/EC/DMC (1 : 1 : 1 volume ratio) as the electrolyte, and Celgard 2300 was the cell separator. Li metal foil worked as the negative electrode and reference electrode. Galvanostatic charge/discharge curves were tested in the range of 2.0–4.8 V (vs. Li⁺/Li) on Land CT2001A battery testers (Wuhan Jinnuo Electronics Co. Ltd, China) at different C rates, and the nominal capacity was selected as 200 mA h g⁻¹ as in literature³³ (1C = 200 mA g⁻¹). Cyclic voltammetry (CV) was carried out on Battery Test Equipment (BT2000, Arbin Instruments, USA) with a scan rate of 0.1 mV s⁻¹.

3. Results and discussion

3.1 Structure and morphology

The XRD patterns of carbonates and samples of Li_1.2Mn_0.54Ni_0.13Co_0.13O₂ are shown in Fig. 1. For the carbonate synthesized by the solvothermal method, the XRD pattern matches those of the standard XRD patterns for MnCO₃ (JCPDS, no. 85-1109),¹¹ CoCO₃ (JCPDS, no. 78-0209),¹¹ and NiCO₃ (JCPDS, no. 78-0210). However, the XRD pattern of the carbonate obtained by the co-precipitated method is rather weak. As two carbonates have the same composition, the solvothermal carbonate should be well-crystallized compared to the co-precipitated carbonate based on the intensity of the corresponding peaks. It was deduced that urea undergoes a slow decomposition process during the solvothermal synthesis and accordingly, the primary particles of carbonate can grow into nanoplates. However, the primary particles of carbonate obtained from the co-precipitated method are apt to form small grains because the nucleation rate is higher than the growth of the particles in the supersaturation solution. This deduction was further confirmed by the SEM images.

Fig. 1 XRD patterns of Li_1.2Mn_0.54Ni_0.13Co_0.13O₂ and carbonates.

The XRD patterns of the Li_1.2Mn_0.54Ni_0.13Co_0.13O₂ cathode materials (Fig. 1B) indicated that both patterns present the characteristic peaks of a typical α-NaFeO₂ with a layered hexagonal structure.^33,34 Except for the peaks between 20–25°, all the peaks can be indexed to the characteristic peaks of Li-rich layered oxide with a space group of R3̄m (no. 166). Peaks in the range of 20–25° reflect the order structure from an alternate arrangement of Li⁺ and Mn⁴⁺ ions in transition metal layers, which also could be seen as the integrated monoclinic phase of Li₂MnO₃ (C2/m).^33,35

Fig. 2 shows SEM images of two carbonates and samples of ST-LMNCO and CP-LMNCO. Both carbonates are composed of micro-size balls measured in Fig. 2A and D. For the solvothermal carbonate, the secondary particles are relatively neat, although the particles are partly integrated with each other. Nevertheless, for the co-precipitated carbonate, the secondary particles are relatively irregular (Fig. 2C and D). Interestingly, the magnified image of the solvothermal carbonate shows the micro-size secondary particles are composed of parallel nanosize plates (Fig. 2B and inset). The particles of the co-precipitated carbonate do not have such special features.

Fig. 2 SEM images of carbonate synthesized by the solvothermal method (A, B and the scale bar for the inset is 100 nm) and by the co-precipitated method (C and D), and the derived ST-LMNCO (E and F) and CP-LMNCO (G and H).

As mentioned in the discussion on XRD patterns, the carbonate synthesized by the solvothermal method is crystallized relative to the one obtained from the co-precipitation method. Results from the SEM images confirmed that the particles of the solvothermal synthesized carbonate might have opportunities to grow along certain directions and form the nanosize layered primary particles. Secondary particles of carbonate from the co-precipitated method are assembled by irregular primary particles to reduce the surface free energy of the small particles.

The morphology of the carbonates influences the morphology and structure of the final sintered cathode material. Fig. 2E and G shows the SEM images of ST-LMNCO and CP-LMNCO, respectively. Obviously, the shape of the microsize spheres is partly preserved for ST-LMNCO (quasi balls marked with broken yellow circles in Fig. 2E), but not for CP-LMNCO. To differentiate the difference in the morphology between the two samples, the original SEM images are shown in the ESI S1.† Owing to the close packing, the secondary particles with a ball-like structure show a high tap density. Two possible reasons are proposed to explain why the secondary particles of ST-LMNCO keep the ball-like structure of the precursor carbonate, while CP-LMNCO loses its ball-like structure, resulting in a relative low tap density. One is the irregular morphology of the co-precipitated precursor carbonate, and the other (more important) is the low degree of crystallization. The sub-units of the solvothermal synthesized carbonate are neat nanosheets, and they closely compact with each other. However, the sub-units of the carbonate synthesized through co-precipitation are irregular and loosely coupled. Note that the crystallization degree of the solvothermal synthesized carbonate is higher than the co-precipitated synthesized carbonate, and the structure of the solvothermal synthesized carbonate is more stable. The reaction between the solvothermal synthesized carbonate and Li₂CO₃ is also more uniform, and the ball-like structure is more likely to be retained (ESI S1†). As for the co-precipitated synthesized carbonate, its loosely compact structure and less uniform reaction with Li₂CO₃ make it less possible to inherit the structure of its precursor. A moderate sintering temperature (600 °C) could allow for the retention of shape, but, for our systems, a moderate temperature means poor crystallization and poor electrochemical performances. A high sintering temperature (900 °C or higher) leads to molten particles and poor shape maintenance. Our following work will focus on searching sintering technology to obtain a real, ball-like microsize sample and with a high electrochemical performance.

TEM images clearly show that the particles of ST-LMNCO are only partly adhered (Fig. 3A), and each particle should be a complete crystal as it consists of uniform lattice fringes throughout the whole particle (Fig. 3B). The lattice space is 0.37 nm in Fig. 3B, which matches the interplanar distance of the (111) plane for a Li-rich layered oxide. Fig. 3C and D show that each particle of CP-LMNCO is composed of even smaller crystals rather than one crystal for one particle. Lattice fringes for CP-LMNCO are in different directions and belong to divided small crystals, and few lattice fringes run through a whole particle, even after a slight adjusting in the direction of the sample. There are many paralleled thick lines in Fig. 3D, but they are not real lattice fringes as they are too large and not as straight. The smaller crystals in the particles must lead to grain boundaries between crystals, which reduces the electrochemical performances of the cathode materials.

Fig. 3 TEM images of ST-LMNCO (A and B) and CP-LMNCO (C and D).

3.2 The tap density

One of the advantages of the quasi ball-like particles of ST-LMNCO is the increasing tap density. The results of the tap density tests showed that the values for ST-LMNCO and CP-LMNCO were 1.40 and 1.23 g cm⁻³, respectively. The quasi-spherical secondary particles of ST-LMNCO had an increasing ratio of ∼14%, which means the volume energy density improved by ∼14% for ST-LMNCO even with the same mass energy density for both sintered samples. The tap densities of both samples are still low relative to other commercial cathode materials. The values increase after optimizing the sintering technology, and this will be reported in future work. After all, this result confirmed the speculation from the SEM image results.

3.3 Electrochemical performances

The electrochemical performances of Li_1.2Mn_0.54Ni_0.13Co_0.13O₂ are presented in Fig. 4. The first charge/discharge curves clearly show that ST-LMNCO possesses a higher specific capacity compared to CP-LMNCO. Within a cutoff voltage of 2.0–4.8 V at room temperature, the initial discharge capacities are 296.6 and 275.9 mA h g⁻¹ at a 0.1C rate for ST-LMNCO and CP-LMNCO, respectively (Fig. 4A). If the theoretical volume energy density was considered (Volume energy density was calculated using the product of capacity and tap density.), the values for ST-LMNCO and CP-LMNCO are 415 and 339 mA h cm⁻³, respectively. The volume energy density of ST-LMNCO is ∼22.4% higher than that of CP-LMNCO.

Fig. 4 The first charge/discharge curves at 0.1C (A), rate performances (B), CV curves of ST-LMNCO (C) and CP-LMNCO (D), the Nyquist plots (E) and cycle capacities (F).

The initial coulombic efficiencies have a great influence on the possible applications of Li-rich layered cathode materials. The values for ST-LMNCO and CP-LMNCO are 85% and 83%, respectively, and their corresponding charge capacities are 349.6 and 331.4 mA h g⁻¹. The higher efficiencies for ST-LMNCO should be attributed to the low surface area as the larger particles may reduce the side reactions with the electrolyte compared to smaller particles.

The rate capacity is an important property for Li-rich layered oxide. Fig. 4B shows that the capacities of ST-LMNCO are 289.7, 277.1, 256.8, 201.6, 141.1 and 272.3 mA h g⁻¹ at 0.1C, 0.2C, 0.5C, 1C, 2C, and 0.1C rates, respectively. The capacities of CP-LMNCO are 267.5, 242.4, 225.1, 172.3, 107.1, 238.1 mA h g⁻¹ at the same C rates. At each C rate, ST-LMNCO shows higher capacities than CP-LMNCO. By the end of first 50 charge/discharge cycles in the C rate tests, the capacity difference between the two samples is 34.2 mA h g⁻¹, which is larger than the initial capacity difference (22.2 mA h g⁻¹). The results predict a different cycle performance for the two cathode materials.

The rate performance of the cathode material is related to the current density of the charge/discharge and can be further confirmed by the results of the CV curves. Fig. 4C and D show the CV curves of Li-rich layered oxides for the first three charge/discharge cycles. The two samples have similar peak shapes and peak positions, which means that they have the same electrochemical reaction processes. However, after the second cycle, ST-LMNCO shows a nearly overlapped curve while the peak intensity of CP-LMNCO is changed. Such results showed that CP-LMNCO was not activated completely after the first cycle even at a scan rate of 0.1 mV s⁻¹. The most interesting feature of the Li-rich layered cathode is the high voltage plateau of ∼4.5 V during the initial charge process. ST-LMNCO exhibited a broad strong peak from ∼4.5 V (Fig. 4C) compared to the thin, weak peak of CP-LMNCO (Fig. 4D). The results of the dQ/dV plots obtained from the initial charge/discharge curves also supported this conclusion (ESI S2†).

The activation results were decided by the fine structure of the electrode material and had a great influence on the electrochemical performances of the material. The AC impedance spectra can give us some information on the surface reaction and intrinsic resistances. Fig. 4E shows the Nyquist plots for two samples, and the inset is the corresponding equivalent circuit diagram. The simulated results from the equivalent circuit show the value of the interphase resistance for ST-LMNCO is 369.9 Ω, which is only ∼57% of that for CP-LMNCO (643.6 Ω). The result means an easier interphase reaction for ST-LMNCO. Interestingly, the intrinsic resistance of ST-LMNCO (10.58 Ω) is higher than that of ST-LMNCO (6.85 Ω). This may be related to the small particles, which have enough contact with the conductive agent (carbon black).

Another challenge for the Li_1.2Mn_0.54Ni_0.13Co_0.13O₂ cathode materials is the capacity retention after long charge/discharge cycles. The cycle performances of the two layered oxides are shown in Fig. 4F. Clearly, under the same experimental conditions (0.1C rate), the capacity retention of ST-LMNCO is superior to CP-LMNCO. After 80 charge/discharge cycles, the final capacities are 247.0 mA h g⁻¹ and 215.8 mA h g⁻¹, and the calculated capacity retentions are 86.2% and 80.9% for ST-LMNCO and CP-LMNCO, respectively.

The electrochemical performances of the cathode materials are generally determined by the structure and morphology. These two samples possess the same structure but different morphologies. For further clarifying the relationship between the performances and the particle size (and special surface area) of the materials, the size distribution of the two samples is shown in Fig. 5 using a statistical method (SEM images in ESI S3†). Obviously, ST-LMNCO shows a broad size distribution, as its primary particles present a flat shape (A different side shows a different size.). Even so, the average particle size of ST-LMNCO (195 nm) is still larger than that of CP-LMNCO (172 nm). Moreover, the special surface area of ST-LMNCO is 4.08 m² g⁻¹, while the area of CP-LMNCO is 6.04 m² g⁻¹. Combining the electrochemical performances of the two samples with the particle size and special surface area, the results are consistent with those reported in literature.¹⁸ The large size secondary particles might be able to inhibit oxygen release because of their comparatively low surface area. Besides, larger particles may also reduce side reactions with the electrolyte compared to smaller particles.

Fig. 5 The particle size distribution of ST-LMNCO (A) and CP-LMNCO (B).

4. Conclusion

In this work, Li-rich layered Li_1.2Mn_0.54Ni_0.13Co_0.13O₂ particles derived from a transition metal carbonate with a micro–nanostructure were obtained. The Li_1.2Mn_0.54Ni_0.13Co_0.13O₂ particles derived from the transition metal carbonate balls synthesized by a solvothermal method (ST-LMNCO) were well-crystallized and showed a higher tap density than the one from the transition metal carbonate balls synthesized by the co-precipitation method (CP-LMNCO). When evaluated as cathode materials for lithium ion batteries, ST-LMNCO delivers a higher discharge specific capacitance (296.6 mA h g⁻¹), a better rate performance (201.6 mA h g⁻¹ at 1C rate), a better capacity retention capability (86.2% after 80 cycles), and an improved volume energy density (∼22.4% higher) compared to CP-LMNCO. The as-obtained Li-rich layered Li_1.2Mn_0.54Ni_0.13Co_0.13O₂ with a micro–nanostructure is well-crystallized with a high density and is a promising cathode candidate for high-performance lithium ion batteries. It is clear that the morphology of the transition metal carbonate significantly impacts the morphology of the derived Li_1.2Mn_0.54Ni_0.13Co_0.13O₂. Considering the morphology impact is critical to the performance of Li-rich layered Li_1.2Mn_0.54Ni_0.13Co_0.13O₂ particles, the designed transition metal carbonates with a micro–nanostructure obtained from a solvothermal process is a promising carbonate platform for Li-rich layered oxides. The strategy of using transition metal carbonates from solvothermal processes as a starting material for Li_1.2Mn_0.54Ni_0.13Co_0.13O₂ might be used to synthesize other types of Li-rich layered oxides for high-performance lithium ion batteries.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (51672071, 21203056, 21303042), National High Technology Research and Development Program (2013AA032002, 2015AA034601), Key Scientific Research Project of Henan Province (15A150056), the Program for Innovative Research Team in University of Henan Province (17IRTSTHN001), Beijing National Laboratory for Molecular Sciences (20140144), and Xinxiang key scientific and technological projects (ZG15004).

References

1. H.-K. Song, K. T. Lee, M. G. Kim, L. F. Nazar and J. Cho, Adv. Funct. Mater., 2010, 20, 3818–3834.
2. J. Yan, X. Liu and B. Li, RSC Adv., 2014, 4, 63268–63284.
3. M. M. Thackeray, C. S. Johnson, J. T. Vaughey, N. Li and S. A. Hackney, J. Mater. Chem., 2005, 15, 2257–2267.
4. D. Wang, X. Wang, R. Yu, Y. Bai, G. Wang, M. Liu and X. Yang, Electrochim. Acta, 2016, 190, 1142–1149.
5. S. Shi, T. Wang, M. Cao, J. Wang, M. Zhao and G. Yang, ACS Appl. Mater. Interfaces, 2016, 8, 11476–11487.
6. P. K. Nayak, J. Grinblat, M. Levi, O. Haik, E. Levi, Y.-K. Sun, N. Munichandraiah and D. Aurbach, J. Mater. Chem. A, 2015, 3, 14598–14608.
7. N. Yabuuchi, K. Yoshii, S. T. Myung, I. Nakai and S. Komaba, J. Am. Chem. Soc., 2011, 133, 4404–4419.
8. D. Dai, B. Li, H. Tang, K. Chang, K. Jiang, Z. Chang and X. Yuan, J. Power Sources, 2016, 307, 665–672.
9. Y. Zhang, P. Hou, E. Zhou, X. Shi, X. Wang, D. Song, J. Guo and L. Zhang, J. Power Sources, 2015, 292, 58–65.
10. F. Li, Y. Y. Sun, Z. H. Yao, J. S. Cao, Y. L. Wang and S. H. Ye, Electrochim. Acta, 2015, 182, 723–732.
11. F. Fu, Q. Wang, Y.-P. Deng, C.-H. Shen, X.-X. Peng, L. Huang and S.-G. Sun, J. Mater. Chem. A, 2015, 3, 5197–5203.
12. D. Luo, G. Li, C. Fu, J. Zheng, J. Fan, Q. Li and L. Li, Adv. Energy Mater., 2014, 4, 1400062.
13. C. Fu, G. Li, D. Luo, J. Zheng and L. Li, J. Mater. Chem. A, 2014, 2, 1471–1483.
14. T. Zhao, S. Chen, L. Li, X. Zhang, H. Wu, T. Wu, C.-J. Sun, R. Chen, F. Wu, J. Lu and K. Amine, ACS Appl. Mater. Interfaces, 2014, 6, 22305–22315.
15. C.-H. Shen, Q. Wang, F. Fu, L. Huang, Z. Lin, S.-Y. Shen, H. Su, X.-M. Zheng, B.-B. Xu, J.-T. Li and S.-G. Sun, ACS Appl. Mater. Interfaces, 2014, 6, 5516–5524.
16. F. Zheng, C. Yang, X. Xiong, J. Xiong, R. Hu, Y. Chen and M. Liu, Angew. Chem., Int. Ed., 2015, 54, 13058–13062.
17. P. Oh, S. Myeong, W. Cho, M.-J. Lee, M. Ko, H. Y. Jeong and J. Cho, Nano Lett., 2014, 14, 5965–5972.
18. M. G. Verde, H. Liu, K. J. Carroll, L. Baggetto, G. M. Veith and Y. S. Meng, ACS Appl. Mater. Interfaces, 2014, 6, 18868–18877.
19. W. Zhao, S. Yamamoto, A. Tanaka and H. Noguchi, Electrochim. Acta, 2014, 143, 347–356.
20. Y. Fu, H. Jiang, Y. Hu, L. Zhang and C. Li, J. Power Sources, 2014, 261, 306–310.
21. Z. Xie, S. Ellis, W. Xu, D. Dye, J. Zhao and Y. Wang, Chem. Commun., 2015, 51, 15000–15003.
22. X. He, J. Wang, R. Kloepsch, S. Krueger, H. Jia, H. Liu, B. Vortmann and J. Li, Nano Res., 2014, 7, 110–118.
23. L. Xi, C. Cao, R. Ma, Y. Wang, S. Yang, J. Deng, M. Gao, F. Lian, Z. Lu and C. Y. Chung, Phys. Chem. Chem. Phys., 2013, 15, 16579–16585.
24. Y. Jiang, Z. Yang, W. Luo, X.-L. Hu, W.-X. Zhang and Y.-H. Huang, J. Mater. Chem., 2012, 22, 14964–14969.
25. H. Yu, H. Kim, Y. Wang, P. He, D. Asakura, Y. Nakamura and H. Zhou, Phys. Chem. Chem. Phys., 2012, 14, 6584–6595.
26. H. Yu, Y. Wang, D. Asakura, E. Hosono, T. Zhang and H. Zhou, RSC Adv., 2012, 2, 8797–8807.
27. J. Liu, L. Chen, M. Hou, F. Wang, R. Che and Y. Xia, J. Mater. Chem., 2012, 22, 25380–25387.
28. L. Zhang, B. Wu, N. Li, D. Mu, C. Zhang and F. Wu, J. Power Sources, 2013, 240, 644–652.
29. X. Miao, Y. Yan, C. Wang, L. Cui, J. Fang and G. Yang, J. Power Sources, 2014, 247, 219–227.
30. F. Fu, Y.-P. Deng, C.-H. Shen, G.-L. Xu, X.-X. Peng, Q. Wang, Y.-F. Xu, J.-C. Fang, L. Huang and S.-G. Sun, Electrochem. Commun., 2014, 44, 54–58.
31. J. Fan, G. Li, D. Luo, C. Fu, Q. Li, J. Zheng and L. Li, Electrochim. Acta, 2015, 173, 7–16.
32. H. Jiang, Y. Fu, Y. Hu, C. Yan, L. Zhang, P. S. Lee and C. Li, Small, 2014, 10, 1096–1100.
33. X. Yuan, Q.-j. Xu, C. Wang, X. Liu, H. Liu and Y. Xia, J. Power Sources, 2015, 279, 157–164.
34. J. Zhao, R. Huang, W. Gao, J.-M. Zuo, X. F. Zhang, S. T. Misture, Y. Chen, J. V. Lockard, B. Zhang, S. Guo, M. R. Khoshi, K. Dooley, H. He and Y. Wang, Adv. Energy Mater., 2015, 5, 1401937.
35. C.-H. Shen, L. Huang, Z. Lin, S.-Y. Shen, Q. Wang, H. Su, F. Fu and X.-M. Zheng, ACS Appl. Mater. Interfaces, 2014, 6, 13271–13279.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra21006f

---