Carbon-coated SnO₂ thin films developed by magnetron sputtering as anode material for lithium-ion batteries

Abstract

The magnetron sputtering technique was employed to produce carbon-coated SnO₂ (CCSO) thin films, which were applied as an anode material for lithium-ion batteries (LIBs). The CCSO shows better electrochemical properties than the pure SnO₂ material. The CCSO thin film has a specific discharge capacity of 650 mA h g⁻¹ with a capacity retention rate of 72% after 100 cycles, while the pure SnO₂ has only 207 mA h g⁻¹ with a capacity retention rate of 19%. High-resolution transmission electron microscopic analysis provides crucial information that the magnetron-sputtered carbon coating has an amorphous feature and shows excellent durability because the carbon film is coated tightly on the surface of the SnO₂ active materials through the discharging and charging cycles. The amorphous carbon coating on the surface of SnO₂ thin film reduced the electrode deterioration, and effectively improved the conductivity and electrochemical stability of the electrode in LIBs.

---

1. Introduction

Lithium-ion batteries (LIBs) are the most popular rechargeable batteries in consumer electronics. Portable electronic devices rely largely on the high energy density, high electromotive force and long cycle life of these batteries, whose performance, on the other hand, is unlikely to meet the requirements for future technological improvements on grid storage, electric vehicles and hybrid electric vehicles.^1–3 Despite the fair stability and rate performance of the currently commercialized graphite anode, its low specific capacity has hindered its further development, thus stimulating extensive research into anode substitutes.

Much work has been devoted to alloy-type tin-based anodes, SnO₂ in particular. Two principle electrochemical processes occur in SnO₂ based anodes during charge–discharge cycling:⁴

SnO₂ + 4Li⁺ + 4e⁻ ↔ Sn + 2Li₂O (1)

Sn + xLi⁺ + xe⁻ ↔ Li_xSn (0 ≤ x ≤ 4.4) (2)

The complete discharging processes comprise the reduction of SnO₂ and the alloying of metallic Sn to Li_4.4Sn, which lead to a whole discharge capacity as high as 1491 mA h g⁻¹ (782 mA h g⁻¹ theoretically reversible as shown in eqn (2)).⁵ The high theoretical capacity is quite intriguing yet alloying–dealloying processes are always accompanied by an intractable volume change of up to 250%.^6,7 Inevitably, the mechanical structure of the anode material fails and active materials lose contact with the current collector, causing severe capacity fading.

Hybridizing SnO₂ with buffering materials such as carbon, Cu and conducting polymers is an effective method to tackle the aforementioned problems whilst improving the conductivity of SnO₂.^8–12 Carbon has won a reputation in both anode or cathode surface coating owing to its high stability and compatibility, which has stimulated extensive study and utilization in composite electrodes by various techniques. Typical carbon-coating techniques include spray pyrolysis, the wet chemical method, hydrothermal/solvothermal methods, calcination after polymer coating, sol–gel reaction, etc.^13–17 These approaches generally involve a mixture of carbon precursor and base material in the preparation process, during which the precursor might induce unwanted phases or morphologies.¹⁸ Some methods, for instance, the hydrothermal method, usually require an additional treatment that adds liability to the product.¹⁹ It has been reported that nano-scaled Sb-doped SnO₂ deposited by magnetron sputtering on copper foil has excellent electrochemical performance as anode material.²⁰ With the magnetron sputtering technique, thin films of different composition and thickness can be easily obtained under well-controlled conditions.

In an attempt to incorporate a SnO₂ thin film with carbon coating, SnO₂ is deposited on copper foil by radio frequency (RF) magnetron sputtering and carbon is coated on the surface of the thin film by the same technique in the present work. This method provides a new angle on thin film carbon coating to obtain composite materials. The electrochemical properties of the carbon-coated SnO₂ thin film are investigated and its microstructure is inspected by transmission electron microscopy (TEM).

2. Experimental

The preparation of an SnO₂ target for sputtering was conducted by a traditional ceramic fabrication process.²⁰ After sol–gel synthesis, calcination, cold pressing, sintering and polishing, a target was obtained, 50 mm in diameter and 4 mm in thickness. Copper foil substrates were initially ultrasonically cleaned in ethanol and then deposited with SnO₂ thin films by RF magnetron sputtering (JGP 450, China) at room temperature. The sputtering chamber was evacuated to a base pressure of 5.0 × 10⁻⁴ Pa and then maintained at 0.8 Pa by adjusting the Ar/O gas mixture at a flux ratio of 4 : 1 during sputtering, which includes 30 minutes of pre-sputtering to ensure a clean target surface and stable glow and 60 minutes of film production. Next, carbon was coated on top of the SnO₂ film with a graphite target by the same technique; except that the sputtering atmosphere was pure argon and the sputtering time was 3 minutes. The carbon-coated SnO₂ (CCSO) thin films were cut into round pieces of 12 mm in diameter and assembled in half coin cells (CR 2032), in an argon-filled glove box, with Li metal foil as counter electrode and reference electrode. The electrolyte is 1 M LiPF₆ solution in a mixture of ethylene carbonate (EC) and diethyl carbonate (DEC) with a volume ratio of 1 : 1.

Galvanostatic charge–discharge processes were performed on a battery measurement system (Land CT2001A, China) at room temperature. The cells were cycled between 0.01 V and 2.00 V vs. Li/Li⁺ at a current density of 0.1 C. For rate performance, the cells were tested on the same system at current densities of 0.1 C, 0.2 C, 0.4 C, 0.8 C, 1 C in sequence and 0.1 C again, for 10 cycles at each current density. Cyclic voltammetry (CV) was carried out on an electrochemistry workstation (Gamry Reference 600, USA) at a scanning rate of 0.1 mV s⁻¹ between 0.01 V and 2.00 V. The instrument was also used to measure the electrochemical impedance in order to determine the conduction characteristics of the assembled cells.

A scanning electron microscope (SEM, FEI Quanta 650 FEG) was employed to study the configuration and thickness of the CCSO composite thin film from both plain and cross-sectional views. The sample for the cross-sectional view was obtained by ripping the thin film along a cutting crack. The surface morphology of a clean copper foil was also observed under SEM for comparison. The microstructures and phase components of the SnO₂ thin films and CCSO thin films were inspected by a transmission electron microscope (TEM, FEI Tecnai G² F20) before and after being electrochemically induced. Cells discharged to 0.01 V or charged to 2.00 V were dismantled respectively, and the active materials were elaborately prepared on holey carbon grids for TEM investigation, as was the pristine CCSO material.^21,22 Fast Fourier Transform (FFT) analysis was also performed on the high-resolution images to determine the reaction products after the cycling.

3. Results and discussion

3.1 Microstructure of carbon-coated SnO₂ thin film

Fig. 1(a) shows the SEM image of a pristine CCSO thin film, where the right part is the surface morphology of the CCSO film, the upper-left inset is the cross-sectional view of the CCSO thin film, and the lower-left inset is the plain-view surface morphology of a clean copper foil. The CCSO thin film exhibits the form of cauliflower-shaped protrusions which are uniformly distributed on the surface of the copper foil. Compared with the clean copper foil (lower-left inset), it can be deduced that these protrusions stemmed from the crude surface of the copper foil when SnO₂ was sputtered onto it. The thickness of the active material layer is uniformly less than 1 μm as can be seen in the upper-left inset, while the size of protrusions varies under 5 μm with substantial ravines between them, which can increase the contact area of active material and electrolyte. The carbon layer cannot be identified in either plain or cross-sectional images because of its limited thickness. The SEM results provide evidence that the as-prepared CCSO thin film is uniform on the substrate and that its structural characteristic with cauliflower-shaped protrusions and ravines should be advantageous for application in lithium-ion batteries.²³

Fig. 1 Microstructure investigations of the pristine carbon-coated SnO₂ (CCSO) thin film sputtered on copper foil, (a) SEM observations, where the right part is the surface morphology of the CCSO film, the upper-left inset is a cross-sectional view of CCSO thin film, and the lower-left inset is the plain-view surface morphology of a clean copper foil. (b) TEM observations, the upper-right inset is a related SEAD pattern and the lower-left inset is an HRTEM image.

Fig. 1(b) presents the TEM investigations of a pristine CCSO thin film. The upper-right inset is the selected area electron diffraction (SEAD) pattern and the lower-left inset is the high-resolution TEM (HRTEM) image featuring the carbon layer. The diffraction rings in the SEAD are indexed as polycrystalline SnO₂ (P4₂/mnm, a = b = 4.74 Å, c = 3.18 Å) with a nanostructured characteristic. The average size of SnO₂ crystals is estimated to be about 5 nm from the HRTEM image. An evenly distributed thin layer of amorphous carbon can be seen in Fig. 1(b). The thickness of the coated carbon layer is about 3–5 nm, an insignificant amount to affect the structure of the mass SnO₂ material. Through FFT analysis of the HRTEM image, the lattice fringe of 3.350 Å is indexed to the {110} plane family of tetragonal SnO₂, confirming the pristine structure of tin dioxide.

3.2 Electrochemical performance

Fig. 2 illustrates the first three cycles of CV of the CCSO (Fig. 2(a)) and pure SnO₂ (Fig. 2(b)) thin films. The CV characteristic of the CCSO thin film is generally in accordance with that of the SnO₂ one without carbon coating, indicating that the magnetron-sputtered carbon coating has a negligible effect on the reaction mechanism of the SnO₂ anode material. To be specific, three peaks emerge during the first discharging process at around 0.94 V, 0.52 V and 0.20 V, respectively, and two during the first charging process at around 0.52 V and 1.32 V, marked in sequence as P1–P5 as shown in Fig. 2(a). Broad and sharp P1 is ascribed to the reduction of SnO₂ to Sn and consequent formation of LiO₂, corresponding to the reaction in eqn (1). This partially reversible reaction results in a large initial capacity loss, which manifests as the obviously weak peaks around P1 in later cycles. Its counter peak P5 is duly weak and broad in three cycles. P2 and P3 are signs of a two-step lithium insertion for the alloying shown in eqn (2) and may also involve the formation of a solid–electrolyte interface (SEI) layer. Research by Zhang et al. has evidenced that carbon coating can stabilize the SEI layer and restrain its thickness during repetitive lithiation and delithiation, leading to a better environment for active materials.²⁴ Therefore, it can be inferred that no significant capacity loss is attributed to the SEI layer in the later cycles. As for de-alloying of Li_4.4Sn, which is a one-step lithium extraction process, the broad P4 in the charging process correlates with P3 and P2 with equivalent current intensity.²⁵ On the whole, except for P1, all the peaks show a considerable augmentation in current intensity in the second and third cycles, indicating an activation process in the anode material, probably induced by carbon coating. Although the thin layer of carbon can fortify the anode nanostructure and does not affect the electrochemical reactions of SnO₂, it somehow impedes full utilization of the anode material in the first few cycles. This is also revealed in the cycling performance and will be discussed later in this section.

Fig. 2 The first three cycles of CV curves of the assembled cells with thin film as test electrode, (a) carbon-coated SnO₂ thin film, (b) pure SnO₂ thin film.

Fig. 3 compares the electrochemical impedance spectra (EIS) of the CCSO and pure SnO₂ thin films in the assembled cells, in the frequency range of 1 Hz to 1 MHz. Both spectra are comprised of a compressed arc at high frequencies and a tail at low frequencies. The tail is attributed to Warburg diffusion. The ohmic resistances (indicated by the high frequency intercept of the real axis) are close to each other and are very small, but the arcs are of different diameters, suggesting different values of charge transfer resistance (R_ct).²⁶ R_ct of the CCSO cell is about half of that of the pure SnO₂ one. The EIS results demonstrate that carbon coating efficiently improves the conductivity of the cell for the charge transport.

Fig. 3 Electrochemical impedance spectra of the cells with CCSO thin film and pure SnO₂ thin film as tested electrodes, respectively, measured in the frequency range of 1 Hz to 1 MHz.

The cycling performance and coulombic efficiency of the CCSO thin film are illustrated in Fig. 4(a) and compared with those of a pure SnO₂ thin film as shown in Fig. 4(b). Both anode materials show a low specific capacity during the first few cycles, indicating an activating process in the electrodes. The activation process can be attributed to the deficiency of electrolyte permeation into the composite thin film. As shown in Fig. 1, the CCSO thin film has numerous ravines between the cauliflower-shaped protrusions (see in Fig. 1(a)), and the SnO₂ active materials are coated by an amorphous carbon layer (see in Fig. 1(b)). Therefore, the electrolyte might not easily permeate into the whole active material in the electrode in the first few cycles. A possible solution to this phenomenon might be to prolong the age time before the assembled cells are applied to charging–discharging, or to use more electrolyte liquid in the cells when assembling. Given enough time and/or electrolyte, the ravines are more likely to be filled with electrolyte, leading to faster activation of the reaction between electrolyte and the active material.²⁷

Fig. 4 Cycling performances of thin films at 0.1 C for 100 cycles, (a) carbon-coated SnO₂ thin film, (b) pure SnO₂ thin film.

The specific charge capacity of the CCSO electrode reaches a maximum value of 858 mA h g⁻¹ at the 7^th cycle, while that of pure SnO₂ reaches 1049 mA h g⁻¹ at the 3^rd cycle. The specific discharge capacity follows the same pattern for both materials, with highest values of 906 mA h g⁻¹ at the 5^th cycle and 1099 mA h g⁻¹ at the 3^rd cycle for CCSO and pure SnO₂, respectively. Nevertheless, the specific capacity of pure SnO₂ fades much faster than the CCSO electrode. Apparently, the CCSO thin film was activated more slowly than the pure SnO₂ thin film, probably due to a ‘barrier effect’ caused by carbon coating. The carbon layer can obstruct electrolyte from permeating through the mass SnO₂, thus impeding complete lithiation and limiting the full utilization of the anode material. However, after 100 electrochemical cycles at 0.1 C, only 207 mA h g⁻¹ of specific discharge capacity is retained for pure SnO₂, but 650 mA h g⁻¹ for CCSO, corresponding to 19% and 72% of retained capacity, respectively. Coulombic efficiencies (CE) of both thin films stand low at the first few cycles, but increase to ca. 99% in subsequent cycles. While the CE of the pure SnO₂ thin film starts to decrease after some 70 cycles, the CCSO one maintains a high CE of approximately 99% through all 100 cycles, demonstrating the superior stability of the anode. In summary, despite the barrier effect of carbon coating on realizing the full potential of SnO₂ capacity at the initial cycles, the barrier on the surface of the SnO₂ thin film can prevent aggregation of nano-particles and reduce the pulverization caused by volume expansion/shrink during the electrochemical cycling. Furthermore, the highly conductive carbon layer provides good electrical contact and allows for high capacity from active SnO₂ particles at later cycles.^24,28 The magnetron-sputtered carbon layer plays a significant role as structure constraint and prevents SnO₂ from deteriorating and pulverizing owing to its conductive and elastic properties.

Fig. 5 gives some detailed data from charging and discharging analysis of carbon-coated SnO₂ thin film. Fig. 5(a) is the capacity–potential relation at the 5^th, 10^th, 50^th and 100^th cycles, respectively, and Fig. 5(b) is the differentiated curves for the related cycles. It is evident from Fig. 5(a) that the electrochemical transition at the 100^th cycle follows a similar profile as the 50^th cycle, but has altered from the 5^th and 10^th cycles, with various slopes and plateaus at different potentials. These differences are amplified when demonstrated as differentiated capacity curves in Fig. 5(b). At the 5^th and 10^th cycles, three peaks marked by S1, S2 and S3 show up during discharging at potentials of 0.94 V, 0.56 V and 0.24 V, respectively, while only two peaks marked by S4 and S5 show up during charging at 0.47 V and 1.35 V, respectively. These two cycles are in good agreement with the 2^nd and 3^rd cycles of CV results, with the discharge peaks slightly shifting to higher potentials. Representing a pair of partially irreversible reactions, both S1 and S5 peaks decline at the 50^th cycle as expected, and S2 and S3 keep shifting and turn up at 0.60 V and 0.45 V, respectively. An interesting phenomenon is that S4 splits into two peaks (marked as S4′ and S4′′ in Fig. 5(b) at 0.47 V and 0.62 V, respectively), matching with discharge peaks S3 and S2.²⁵ Apparently, the lithiation and delithiation in this electrode system are two-step redox reactions which are highly reversible even after 100 cycles. The intermediate products can be various lithium–tin compounds such as Li₇Sn₃, Li₅Sn₂ and Li₇Sn₂ as reported by Ouyang et al. in Sb-doped SnO₂ thin films.^20,29 The charge–discharge manner is quite clear and stable at the 50^th cycle and maintains well into the 100^th cycle, with only a minor loss of specific discharge capacity of 25 mA h g⁻¹. The superior stability of the anode material might reflect the advantage of carbon-coating on the SnO₂ thin film.

Fig. 5 Electrochemical features of the carbon-coated SnO₂ thin film at 5^th, 10^th, 50^th and 100^th cycles, (a) plots of capacity vs. potential, (b) differentiated capacity vs. potential relations.

To examine the stabilizing function of the carbon coating on the LIBs, cells with CCSO anode were tested at different current densities and the results are shown in Fig. 6. The activation process can still be seen at the first few cycles. The specific discharge capacity declines from the highest of 1065 mA h g⁻¹ at the 4^th cycle at 0.1 C to 625 mA h g⁻¹ at the 50^th cycle at 1 C, and bounces back to 763 mA h g⁻¹ at 60^th cycle when the current density was switched to 0.1 C again for 10 cycles. The CCSO thin film preserves a high capacity even at relatively high rates. The results once again prove that magnetron-sputtered carbon coating is an effective way to improve stability of an SnO₂ thin film as anode in LIBs.

Fig. 6 Rate performance of carbon-coated SnO₂ thin film at current densities of 0.1 C, 0.2 C, 0.4 C, 0.8 C and 1 C.

3.3 Carbon coating through phase transition

It is important for the carbon layer to be durable and to sustain the anode structure through repetitive discharge and charge. Therefore, TEM was further employed to examine the carbon durability and elucidate lithiation and delithiation in the CCSO anode material. Fig. 7(a) and (b) show HRTEM images of the CCSO thin films discharged to 0.01 V and charged to 2.00 V after 10 cycles, respectively. FFT of the HRTEM images were performed as shown in the related insets, where the dashed rings mark out diffraction rings of relevant phases. The carbon layer is still found intact on the surface of SnO₂ after 10 cycle electrochemical processes. The thickness of the carbon layer largely stays the same as before cycling (3–5 nm), indicating unreactivity and excellent mechanical property. On the other hand, the grains become rather small and indistinct because the electrochemical cycles lead to crystallite fragmentation.

Fig. 7 HRTEM investigations of carbon-coated SnO₂ thin films after 10 electrochemical cycles, the carbon layer is still found intact on the surface of the anode material. The insets are FFT images of each HRTEM image. (a) Discharged to 0.01 V, (b) charged to 2.00 V.

Through analysis of the FFT image in Fig. 7(a), the grains are determined to be Li_4.4Sn. After the anode is discharged to 0.01 V, SnO₂ is completely reduced to metallic Sn and metallic Sn is fully alloyed with Li⁺ to form Li_4.4Sn, as shown in eqn (1) and (2). Lattice spacings of two grains are identified as {264} (shown in the inset enlarged picture) and {440} plane families of Li_4.4Sn phase. In Fig. 7(b), when the anode is charged to 2.00 V, the diffraction rings in the inset are indexed to {110}, {101}, {201} and {211} plane families of SnO₂ while some spots are indexed to {101} plane family of Sn. Two representative grains are marked out as SnO₂ {110} plane family of 3.350 Å in plane spacing and Sn {101} plane family of 2.792 Å. Lithium ions are extracted from Li_4.4Sn and the alloy reverts to Sn and SnO₂ during the charging process. Since the 10^th cycle is still an early stage of cycling, it is reasonable for Sn to be partially oxidized to SnO₂ in this process. HRTEM images provide solid evidence that The CCSO thin film complies with eqn (1) and (2) during cycling under firm and continuous structural protection provided by the magnetron-sputtered carbon layer on the surface.

4. Conclusion

A layer of amorphous carbon with a thickness of 3–5 nm was sputtered on the surface of an SnO₂ thin film by RF magnetron sputtering. The carbon-coated SnO₂ thin film for the anode material of lithium-ion batteries exhibits a high specific discharge capacity and retention rate, and has an electrochemical performance much higher than that of the pure SnO₂ thin film. The carbon coating maintains the original structure during the electrochemical processes. The carbon layer imposes a barrier effect on electrolyte permeation and may hamper full utilization of the active SnO₂ material, but it effectively fortifies the SnO₂ nano-structure and ensures good rate performance. The present work offers a new method for preparing directly a carbon-coated thin film without incurring unwanted phases or morphologies, and the carbon coating is an effective method to improve electrochemical properties of the electrode in LIBs.

Acknowledgements

The authors acknowledge the support of the National Nature Science Foundation of China (No. 51172287) and the Laboratory Research Fund by the State Key Laboratory of Powder Metallurgy, Central South University, China.

References

1. B. Dunn, H. Kamath and J. M. Tarascon, Science, 2011, 334, 928–935.
2. M. M. Thackeray, C. Wolverton and E. D. Isaacs, Energy Environ. Sci., 2012, 5, 7854–7863.
3. X. W. Lou, D. Deng, J. Y. Lee and L. A. Archer, Chem. Mater., 2008, 20, 6562–6566.
4. I. A. Courtney and J. R. Dahn, J. Electrochem. Soc., 1997, 144, 2045–2052.
5. Y. Idota, T. Kubota, A. Matsufuji, Y. Maekawa and T. Miyasaka, Science, 1997, 276, 1395–1397.
6. J. Y. Huang, L. Zhong, C. M. Wang, J. P. Sullivan and W. Xu, Science, 2010, 330, 1515–1520.
7. J. M. Whiteley, J. W. Kim, C. S. Kang, J. S. Cho, K. H. Oh and S. H. Lee, J. Electrochem. Soc., 2015, 162, A711–A715.
8. M. Ara, K. Wadumesthrige, T. Meng, S. O. Salley and K. S. Ng, RSC Adv., 2014, 4, 20540–20547.
9. J. S. Chen, Y. L. Cheah, Y. T. Chen, N. Jayaprakash, S. Madhavi, Y. H. Yang and X. W. Lou, J. Phys. Chem. C, 2009, 113, 20504–20508.
10. H. El-Shinawi, M. Böhm, T. Leichtweiß, K. Peppler and J. Janek, Electrochem. Commun., 2013, 36, 33–37.
11. L. Yuan, J. Wang, S. Y. Chew, J. Chen, Z. P. Guo, L. Zhao, K. Konstantinov and H. K. Liu, J. Power Sources, 2007, 174, 1183–1187.
12. Y. Zhao, X. Li, B. Yan, D. Li, S. Lawes and X. Sun, J. Power Sources, 2015, 274, 869–884.
13. L. Yuan, K. Konstantinov, G. X. Wang, H. K. Liu and S. X. Dou, J. Power Sources, 2005, 146, 180–184.
14. T. Moon, C. Kim, S. T. Hwang and B. Park, Electrochem. Solid-State Lett., 2006, 9, A408–A411.
15. X. Sun, J. Liu and Y. Li, Chem. Mater., 2006, 18, 3486–3494.
16. M. S. Park, Y. M. Kang, J. H. Kim, G. X. Wang, S. X. Dou and H. K. Liu, Carbon, 2008, 46, 35–40.
17. B. Zhou, S. Yang, L. Wu, W. Wu, W. Wei, L. Chen, H. Zhang, J. Pan and X. Xiong, RSC Adv., 2015, 5, 49926–49932.
18. J. Wang, J. Yang, Y. Tang, J. Liu, Y. Zhang, G. Liang, M. Gauthier, Y. K. Chen-Wiegart, M. N. Banis, X. Li, R. Li, J. Wang, T. K. Sham and X. Sun, Nat. Commun., 2014, 5, 3415.
19. G. Wu, Z. Li, W. Wu and M. Wu, J. Alloys Compd., 2014, 615, 582–587.
20. P. Ouyang, H. Zhang, Y. Wang, W. Chen and Z. Li, Electrochim. Acta, 2014, 130, 232–238.
21. W. Chen, H. Zhang, Z. Ma, B. Yang and Z. Li, J. Mater. Chem. A, 2015, 3, 14202–14209.
22. Z. Ma, H. Zhang, Y. Zhang, J. Zhang and Z. Li, Electrochim. Acta, 2015, 176, 1427–1433.
23. C. Zhang, X. Peng, Z. Guo, C. Cai, Z. Chen, D. Wexler, S. Li and H. Liu, Carbon, 2012, 50, 1897–1903.
24. W. M. Zhang, X. L. Wu, J. S. Hu, Y. G. Guo and L. J. Wan, Adv. Funct. Mater., 2008, 18, 3941–3946.
25. W. Xu, N. L. Canfield, D. Wang, J. Xiao and Z. Nie, J. Power Sources, 2010, 195, 7403–7408.
26. L. Wang, D. Wang, Z. Dong, Z. Dong, F. Zhang and J. Jin, Nano Lett., 2013, 13, 1711–1716.
27. S. H. Lee, S. H. Yu, J. E. Lee, A. Jin, D. J. Lee, N. Lee, H. Jo, K. Shin, T. Hyeon, Nano Lett., 2013, 13, 4249–4256.
28. H. Li and H. Zhou, Chem. Commun., 2012, 48, 1201–1217.
29. P. Ouyang, H. Zhang, W. Chen, Y. Wang, Y. Zhang and Z. Li, Mater. Res. Bull., 2015, 61, 9–15.

---