Mixed ionic liquid/organic carbonate electrolytes for LiNi_0.8Co_0.15Al_0.05O₂ electrodes at various temperatures

Abstract

Mixtures of N-butyl-N-methyl pyrrolidinium bis(trifluoromethanesulfonyl)imide ionic liquid (IL) and conventional organic carbonate electrolyte are used for high-capacity LiNi_0.8Co_0.15Al_0.05O₂ (LNCA) electrodes in Li-ion batteries. Increasing the IL content ratio in the mixtures can increase the electrolyte's thermal stability and retard its flammability. However, the optimal electrolyte composition depends on the operating temperature. At 25 °C, the plain organic electrolyte is preferred due to its highest ionic conductivity among the tested electrolytes. This electrolyte is volatile at 50 °C, and thus the incorporation of 25 wt% IL can improve the cyclic stability of the LNCA electrode. The LNCA dissolution and electrolyte decomposition at 75 °C are clearly suppressed with a high IL ratio in the mixed electrolyte. At such a high temperature, with 75 wt% of IL incorporation a high electrode capacity of 195 mA h g⁻¹ is obtained at 30 mA g⁻¹; 50% of this capacity can be retained when the charge–discharge rate increases to 700 mA g⁻¹. Moreover, less than 20% capacity decay is found after 100 cycles.

---

1. Introduction

Lithium (Li)-ion batteries (LIBs) are important charge storage units for a variety of applications, such as portable electronic devices, electric vehicles, and smart grids, due to their high energy density and satisfactory cycle life.^1–3 One major concern for LIBs with regard to further extending their applications (especially for large-scale energy storage) is safety.^4,5 Commonly used organic carbonate electrolytes are mainly responsible for safety issues, since they have poor thermal stability and high flammability. These electrolytes also pose environmental hazards because of their volatility and toxicity.^6,7 Accordingly, a safer and greener alternative electrolyte, namely ionic liquid (IL), which is characterized by intrinsic conductivity, large electrochemical windows, excellent stability, non-volatility, and non-flammability,^8–10 has attracted a lot of attention in recent years.

Nickel-rich Li[Ni_1−xM_x]O₂ (M = a transition metal, such as Co, Al, and so on) layer-structured materials are considered as important cathodes for current LIBs, due to their high specific capacity,^11–13 environmental friendliness, and lower cost than LiCoO₂.^14,15 Substitution of Ni by doping elements of Co and Al is able to stabilize the layered structure and alleviate the increase in electrode polarization upon charge–discharge cycling, suppressing the capacity fading.^12,16,17 Despite such improvements, the elevated-temperature performance of this kind of cathode is still problematic.^18,19 While most previous reports characterized the Li[NiCoAl]O₂ performance at ambient temperature, this study investigates the temperature-dependent electrochemical properties in various electrolytes. In conventional organic electrolytes, the structural degradation of the Li[NiCoAl]O₂ electrode was found to be triggered by dissolution of the constituent transition metals due to the attack of HF,²⁰ which is produced from decomposition of electrolytic salt.²¹ How the IL electrolyte can affect the electrode cyclic stability, which has not yet been examined in the literature, is thus evaluated in the present work.

Although IL electrolytes are promising, their relatively high viscosity compared to that of conventional organic electrolytes remains a significant challenge to their use.⁹ It has been reported that the Li⁺ diffusivity is significantly smaller in bis(trifluoromethylsulfonyl)imide (TFSI)-based ILs compared to that in organic electrolytes, since every Li⁺ cation is coordinated by ∼3 TFSI anions,²² leading to a low Li⁺ conduction in the IL electrolyte. One solution to this problem is to mix the IL with a conventional organic electrolyte. The use of a mixed electrolyte can enable a compromise between safety (and reliability) and cell performance.²³ It is also known that corrosion of the Al cathode current collector at high potentials can be inhibited by introduction of ILs into organic electrolytes.²⁴ The IL and organic mixed electrolytes have been used for LiFePO₄,^23,25,26 which has a low operating voltage and less safety risk. In our opinion, adopting IL electrolytes for Li[NiCoAl]O₂ cathodes, with the advantages of high electrode activity and high energy density, should be a more beneficial approach, although this has not yet been investigated in the existing literature. Moreover, no studies have discussed the optimal IL/organic electrolyte ratios at various temperatures for any kind of LIB electrodes. These important issues will thus be addressed in the current work.

A N-butyl-N-methyl pyrrolidinium–TFSI (BMP–TFSI) IL electrolyte is used in the present study. This IL is attractive because its cathodic stability limit (associated with the reduction of BMP⁺) is beyond the Li plating/stripping reaction, and TFSI⁻ can withstand a high potential of >5 V (vs. Li).²⁷ Various amounts (0–75 wt%) of the IL were incorporated into a conventional organic electrolyte to prepare mixed electrolytes. The electrochemical properties (in terms of capacity, rate capability, and cyclability) of LiNi_0.85Co_0.15Ni_0.05O₂ (LNCA) are systematically investigated in various electrolytes at a temperature range of 25–75 °C. To the best of our knowledge, this is a pioneer paper that reports the optimal IL ratio in the electrolyte for various operating temperatures. At an elevated temperature, such as 75 °C, a high amount of IL incorporation is necessary to guarantee the high safety, high durability, and good charge–discharge performance of LNCA cells.

2. Experimental procedure

2.1 Preparation of LiNi_0.8Co_0.15Al_0.05O₂ and mixed electrolytes

The synthesis procedures for LNCA powder are as follows. First, a rheological phase method was used to prepare a Ni_0.8Co_0.15Al_0.05O₂ precursor. Stoichiometric amounts of Ni(CH₃COO)₂·4H₂O, Co(CH₃COO)₂·4H₂O, and Al(CH₃COO)₂·4H₂O were dissolved in ethanol via vigorous stirring. The drying/dispersion process was conducted at 80 °C for 3 h, yielding a rheological Ni_0.8Co_0.15Al_0.05O₂ body. After being dried at 105 °C, the Ni_0.8Co_0.15Al_0.05O₂ precursor was mixed with CH₃COOLi·2H₂O solution, forming a Li-containing slurry. This slurry was then subjected to a two-step thermal treatment, which included: (i) acetic acid removal at 350 °C for 1 h, and (ii) sintering at 700 °C for 3 h. The whole process was performed in a tubular furnace under an air flow of 200 cm³ min⁻¹. After cooling down to room temperature in the furnace, the LNCA powder was obtained.

BMP–TFSI IL was prepared and purified following a published method.²⁸ The IL was washed with dichloromethane (99%, Showa), filtrated to remove precipitates, and then vacuum-dried at 100 °C for 12 h before use. 1 M LiTFSI (99%, Solvionic) was dissolved in the IL to provide Li⁺ conduction. A conventional organic electrolyte, consisting of 1 : 1 (by volume) ethylene carbonate (EC, 99%, Alfa Aesar) and diethyl carbonate (DEC, 99%, Alfa Aesar) as co-solvents and 1 M LiPF₆ salt, was also prepared. The IL and organic electrolytes are mixed with various weight ratios of 0/100, 25/75, 50/50, and 75/25. The plain IL electrolyte was not used, since its low conductivity and low Li⁺ transference number at ambient temperature limit practical applications, especially when power performance is demanded.^29,30 The mixed electrolytes were continuously stirred by a magnetic paddle for 24 h to ensure uniformity. Their water contents, measured using a Karl Fisher titrator, were below 100 ppm. All the chemicals were handled and stored in an argon-filled glove box (Innovation Technology Co. Ltd.), where both the moisture and oxygen contents were maintained at below 0.5 ppm.

2.2 Cell assembly

A cathode slurry was prepared by mixing 80 wt% LNCA powder, 10 wt% carbon black, and 10 wt% poly(vinylidene fluoride) in N-methyl-2-pyrrolidone solution. The slurry was cast onto Al foil and vacuum-dried at 110 °C for 2 h. Afterwards, this cathode electrode was roll-pressed and punched to match the required dimensions of a CR2032 coin cell. Li foil and a Celgard polypropylene membrane were used as the anode and separator, respectively. The assembly of coin cells was conducted in the argon-filled glove box.

2.3 Material and electrochemical characterizations

The morphologies of the LNCA powder and electrodes were examined using scanning electron microscopy (SEM; FEI Inspect F50). X-ray diffraction (XRD; Bruker D8 Advance, with a Cu target) was employed to analyze the crystal structure. The X-ray detector was scanned at a speed of 1° min⁻¹. Thermogravimetric analysis (TGA; Perkin-Elmer TGA7) was performed to evaluate the thermal stability of the electrolytes, which were heated from room temperature to 550 °C at a rate of 5 °C min⁻¹ under a nitrogen atmosphere. The electrolyte flammability was tested under air according to an established procedure.³¹ Briefly, glassy fiber filters were used to adsorb the electrolytes (∼0.1 g) and were then burned with an electric Bunsen burner (the distance between the sample and the burner was 120 mm). Ten samples for each electrolyte were tested, and the percentages of ignition were counted. A TetraCon 325 conductivity meter was used to measure the ionic conductivity of the electrolytes. The charge–discharge properties (in terms of capacity, rate capability, and cyclic stability) of the LNCA electrodes in various electrolytes were characterized using an Arbin BT-2043 battery tester (within a voltage range of 2.8–4.4 V). The coin cells were placed in climatic chambers where the temperature was fixed at 25 °C (±1 °C), 50 °C (±1 °C) or 75 °C (±1 °C). Five parallel cells were evaluated for each experimental condition. The measured capacities were typically within 5% deviation, and the reported data are the medians.

3. Results and discussion

Fig. 1(a) shows a typical SEM micrograph of the obtained LNCA powder. The diameter of the oxide polyhedrons was in a range of 200–300 nm. This small size is beneficial to shorten the Li⁺ diffusion pathway, and thus improve electrode charge–discharge performance. The XRD pattern of LNCA is shown in Fig. 1(b). All the diffraction signals can be attributed to a α-NaFeO₂-type layered structure (space group, R3̄m), and no impurity phase was detected. The sharp diffraction peaks indicate that a highly crystalline LNCA was synthesized. The diffraction pattern exhibits distinct (006)/(102) and (108)/(110) doublets, which indicates an ordered distribution of Li and Ni/Co in the layered structure. These results also suggest that the doped Al atoms are located at the Ni/Co cation sites and form a solid solution with LiNi_xCo_yO₂.³²

Fig. 1 (a) SEM micrograph and (b) XRD pattern of the synthesized LNCA powder.

The TGA data, revealing the thermal stability of various electrolytes, are shown in Fig. 2. The conventional organic electrolyte exhibited a significant weight loss of ∼40% before 100 °C, at which point the carbonate solvent violently evaporated and LiPF₆ started to decompose into LiF and PF₅.^33,34 At 200 °C, almost nothing was left on the TGA crucible. In contrast, as can be seen in the figure, the observed decomposition temperature for the IL component is as high as ∼400 °C, which is consistent with the results in the literature.^25,31 It was found that the volatility of the electrolyte gradually fell as the ratio of IL incorporation increased. The weight losses at 100 °C for the electrolytes containing 25, 50, and 75% IL components were 31, 21, and 11%, respectively. Incorporation of IL is thus beneficial for using the electrolyte at elevated temperatures.

Fig. 2 TGA data of various electrolytes measured at a heating rate of 5 °C min⁻¹ under nitrogen atmosphere.

Fig. S1 (see ESI†) shows the flammability of the prepared electrolytes. As demonstrated in Fig. S1(a),† the plain organic electrolyte ignited easily and burned rather violently. It was found that the electrolyte flammability can be significantly lowered by increasing the amount of IL that is incorporated (see Fig. S1(b)–(d)†). The rates of ignition for ten tests for the 0, 25, 50, and 75% IL-incorporated electrolytes were 10/10, 6/10, 3/10, 0/10, respectively. The results indicate that the BMP–TFSI IL is an effective flame retardant, and when it is added the resulting electrolytes are better suited for high-safety applications.

The charge–discharge curves of the LNCA electrode recorded in the conventional organic electrolyte with various current densities are shown in Fig. S2.† The profiles show sloping characteristics, with an average discharge voltage of approximately 3.8 V. The reversible capacity of LNCA at 30 mA g⁻¹ was 170 mA h g⁻¹. A satisfactory high-rate capacity of 90 mA h g⁻¹ was obtained when the charge–discharge current increased to 500 mA g⁻¹. Both properties indicate that the prepared electrode is of high quality. Fig. 3(a) compares the electrode performance measured in various electrolytes at 25 °C. When the amount of IL incorporation increased from 25 to 50 and 75%, the discharge capacities at 30 mA g⁻¹ gradually decreased from 166 to 160 and 154 mA h g⁻¹, respectively. The performance difference between various cells increased at higher charge–discharge rates, as shown in Fig. 3(b). It was found that the capacity retained ratios at 700 mA g⁻¹ (compared to the capacities at 30 mA g⁻¹) for the electrodes measured in 0, 25, 50, and 75% IL-incorporated electrolytes were 36, 20, 12, and 5%, respectively. This is attributed to the high viscosity and low conductivity of the IL electrolyte,³⁵ which is unfavorable for high-rate operations. Fig. 3(c) shows the cyclic stability of various cells at 25 °C. After 100 cycles, the plain organic electrolyte cell retained 89% of its initial capacity. However, the cell durability deteriorated as the amount of IL increased. With 75% of IL incorporation, only 54% capacity can be retained after the same testing cycles. It has been reported that the Li⁺ transference number in the IL electrolyte is low (0.28 at 25 °C).³⁰ Fig. S3† shows the electrochemical impedance spectroscopy data for the LNCA electrodes in various electrolytes. A similar equivalent circuit and calculation model to those in the literature were used to analyze the data.^36,37 The results indicate that the electrolyte resistance and interfacial charge transfer resistance increased along with the IL ratio, whereas the Li⁺ diffusion coefficient in LNCA was approximately 4 × 10⁻¹¹ cm² s⁻¹ (independent of the electrolyte composition). More details about the fading mechanism require further investigations. According to the results shown in Fig. 3, a high IL ratio in the electrolyte is not desirable at 25 °C.

Fig. 3 (a) Charge–discharge curves at 30 mA g⁻¹, (b) capacity retained ratios at various rates compared to that obtained at 30 mA g⁻¹, and (c) cyclic stability of LNCA electrodes measured in various electrolytes at 25 °C.

When the temperature was increased to 50 °C, which is often encountered or can be intentionally maintained in practical applications, the mixed electrolyte cells began to show competitiveness. The charge–discharge curves and rate capability of various cells at 50 °C are shown in Fig. 4(a) and (b). As compared to the data in Fig. 3(a) and (b), it is found that while the properties of the plain organic cell did not vary significantly, the IL-incorporated cells showed a clear performance improvement when the temperature was raised from 25 to 50 °C. With 25% IL incorporation, the capacities of 184 and 90 mA h g⁻¹ were measured at 30 and 700 mA g⁻¹, respectively, which were comparable to those found in plain organic electrolyte. As shown in Fig. 4(c), the capacity retained ratios for the 0, 25, 50, and 75% IL-incorporated cells after 100 cycles are 78, 81, 72, and 67%, respectively. The values for the latter three cells are higher than those measured at 25 °C (see Fig. 3(c)), and this is presumably associated with the increased Li⁺ transference number at the elevated temperature.³⁰ Fig. S3† also indicates that the electrolyte resistance and interfacial charge transfer resistance of the cells decrease with increasing temperature. Of note, the durability of the first cell significantly degraded as compared to that at 25 °C. This is consistent with the large change in weight observed for the conventional organic electrolyte at 50 °C in the TGA analysis.

Fig. 4 (a) Charge–discharge curves at 30 mA g⁻¹, (b) capacity retained ratios at various rates compared to that obtained at 30 mA g⁻¹, and (c) cyclic stability of LNCA electrodes measured in various electrolytes at 50 °C.

Fig. 5 shows the ionic conductivity of various electrolytes as a function of temperature. It can be seen that increasing the IL ratio decreases the electrolyte conductivity. This is ascribed to the higher viscosity and lower ionic mobility of the IL electrolyte.³⁵ However, it was found that the ionic conductivity increased significantly along with the temperature. As shown in the inset of Fig. 5, the conductivity (σ) and temperature (T) relationship follows the Vogel–Tammann–Fulcher (VTF) behavior, given by The σ₀, E_a, R, and T₀ correspond to the pre-exponential factor, activation energy, ideal gas constant, and theoretical glass transition temperature, respectively. The fitting results of these VTF parameters are listed in Table S1 (see ESI†). Clearly, the electrolyte with a higher IL ratio shows a larger E_a value, reflecting that the conductivity depends on temperature to a greater extent. Specifically, the ionic conductivity for the 75% IL-incorporated electrolyte increased from 1.8 mS cm⁻¹ at 25 °C to 6.2 mS cm⁻¹ at 75 °C. This is because an increase in temperature promotes the dissociation of ion pairs/clusters in the IL,³⁸ improving the ionic conduction of the electrolyte.

Fig. 5 Temperature-dependent ionic conductivity (σ) of various electrolytes. The inset shows the VTF plot of the data.

In this context, the properties of the LNCA electrodes were further studied at 75 °C with various electrolytes. Fig. 6(a)–(d) show the charge–discharge curves (at various current densities) measured in 0, 25, 50, and 75% IL-incorporated electrolytes, respectively. Interestingly, at such a high temperature the reversible capacities, regardless of the charge–discharge rates, increased along with the IL ratio. For instance, the discharge capacities found in the 75% IL-incorporated electrolyte were 195 mA h g⁻¹ at 30 mA g⁻¹ and 98 mA h g⁻¹ at 700 mA g⁻¹. Both values are substantially better than those (175 and 18 mA h g⁻¹) obtained in the plain organic electrolyte. At this temperature the advantage of high stability of the IL is fully exploited, whereas its drawback of relatively low conductivity is mitigated. It has also been reported that the viscosity of IL/organic mixed electrolytes decreases as the temperature increases, promoting Li⁺ mobility.^23,30 As a result, excellent charge-storage performance of the electrode was achieved in the electrolyte with a high IL content.

Fig. 6 Charge–discharge curves of LNCA electrodes recorded at various current densities in (a) 0%, (b) 25%, (c) 50%, and (d) 75% IL-incorporated electrolytes at 75 °C. (e) Comparison of LNCA cyclic stability in various electrolytes at 75 °C.

Fig. 6(e) shows the cyclic stability of the cells at 75 °C. After 100 charge–discharge cycles, the cells with 0, 25, 50, and 75% IL-incorporated electrolytes retained 44, 64, 75, and 81% of their initial capacities, respectively. We found that the electrolytes with a high IL ratio had a relatively low corrosivity towards LNCA. As shown in Fig. S4(a),† after an immersion test at 80 °C for five days, the color of the plain organic electrolyte changed significantly, indicating a considerable dissolution of LNCA into the solution. In contrast, the powder was rather inert in the 75%-IL electrolyte under the same testing conditions (see Fig. S4(b)†). This can be attributed to the chemical benignity of the IL, and thus low level of attack on LNCA. Fig. 7 shows the SEM micrographs of the electrodes before and after 100 charge–discharge cycles in plain organic electrolyte and 75%-IL electrolyte. While the electrode did not show any significant change in morphology after being cycled in the latter electrolyte, a surface layer was clearly observed on the electrode tested in the former. The formation of this layer was mainly attributed to the decomposition of the organic electrolyte near the end of charging, which can be facilitated at high temperature.³⁹ This undesirable interface layer, accumulated upon cycling, can increase the resistance between LNCA particles and also hinder the Li⁺ intercalation/deintercalation reactions,¹⁸ resulting in significant capacity fading.

Fig. 7 SEM micrographs of (a) the as-prepared LNCA electrode, and the electrodes tested in (b) 0% IL-incorporated and (c) 75% IL-incorporated electrolytes for 100 charge–discharge cycles at 75 °C.

Fig. 8 summarizes the effects of temperature on the charge–discharge properties (in terms of maximum capacity, rate capability, and cyclability) of the LNCA electrodes with various electrolytes. At 25 °C, the plain organic electrolyte seems to be an appropriate choice, especially when high-rate charge–discharge performance is desired, because incorporation of IL electrolyte would increase the viscosity and decrease the conductivity of the electrolyte (although the thermal stability can be improved). The performance of the mixed electrolytes was significantly enhanced when increasing the temperature to 50 °C. With an appropriate amount of IL incorporation (25%), the electrode's cyclic stability can be promoted without sacrificing the maximum capacity and rate capability. At a further high temperature of 75 °C, at which point the organic electrolyte is rather volatile but the IL has increased ionic conductivity, a high ratio of IL incorporation is favorable. The 75% IL-incorporated electrolyte not only ensured the cell safety, but also led to the higher capacity, higher rate capability, and more satisfactory cyclic stability of the LNCA electrode (as compared to those found in the other electrolytes). We believe that this is the first systematic investigation to reveal the optimal IL/organic electrolyte ratios at various operating temperatures for any kind of LIB electrodes. Great compatibility of TFSI-based IL/conventional organic mixed electrolytes with graphitic anodes has also been reported.⁴⁰ Accordingly, this type of hybrid electrolyte has remarkable potential for practical battery applications.

Fig. 8 Effects of temperature on (a) maximum capacities (at 30 mA g⁻¹), (b) capacity retained ratios at 700 mA g⁻¹ (compared to the capacities at 30 mA g⁻¹), and (c) cyclic stability (after 100 cycles) of LNCA electrodes evaluated in various electrolytes.

4. Conclusion

LNCA electrodes were prepared using a rheological phase method, and their charge–discharge properties were then examined as a function of temperature in electrolytes with various ratios of BMP–TFSI IL and conventional organic electrolyte. The incorporation of the IL, with a high decomposition temperature of ∼400 °C, decreased the electrolyte's volatility and flammability. It was found that different electrolyte compositions should be adopted at various temperatures to optimize the cell performance. At elevated temperatures a high IL incorporation ratio is favorable, because both the LNCA dissolution and electrolyte decomposition can be suppressed. Moreover, the relatively low conductivity of the high-IL-content electrolyte is mitigated with increasing temperature. A high LNCA capacity of 195 mA h g⁻¹ and satisfactory electrode durability (<20% decay after 100 cycles) were found in the 75% IL-incorporated electrolyte at 75 °C. The mixed electrolytes with a rational design of the IL ratio show good potential for use with LNCA electrodes, achieving high safety, high reliability, and great performance of LIBs.

Acknowledgements

The financial support of this work by the Ministry of Science and Technology of Taiwan (under grants 103-2221-E-008-024-MY3 and 102-2923-E-008-002-MY3) is gratefully appreciated.

References

1. M. S. Whittingham, Chem. Rev., 2014, 114, 11414–11443.
2. Y. Tang, Y. Zhang, W. Li, B. Ma and X. Chen, Chem. Soc. Rev., 2015, 44, 5926–5940.
3. J. B. Goodenough and K. S. Park, J. Am. Chem. Soc., 2013, 135, 1167–1176.
4. P. Ribiere, S. Grugeon, M. Morcrette, S. Boyanov, S. Laruelle and G. Marlair, Energy Environ. Sci., 2012, 5, 5271–5280.
5. P. Ping, Q. S. Wang, P. Huang, K. Li, J. Sun, D. P. Kong and C. H. Chen, J. Power Sources, 2015, 285, 80–89.
6. G. G. Eshetu, J. P. Bertrand, A. Lecocq, S. Grugeon, S. Laruelle, M. Armand and G. Marlair, J. Power Sources, 2014, 269, 804–811.
7. D. Lisbona and T. Snee, Process Saf. Environ. Protect., 2011, 89, 434–442.
8. G. G. Eshetu, M. Armand, B. Scrosati and S. Passerini, Angew. Chem., Int. Ed., 2014, 53, 13342–13359.
9. D. R. MacFarlane, N. Tachikawa, M. Forsyth, J. M. Pringle, P. C. Howlett, G. D. Elliott, J. H. Davis Jr, M. Watanabe, P. Simon and C. A. Angell, Energy Environ. Sci., 2014, 7, 232–250.
10. T. Vogl, S. Menne, R. S. Kuhnel and A. Balducci, J. Mater. Chem. A, 2014, 2, 8258–8265.
11. K. Kang, Y. S. Meng, J. Bréger, C. P. Grey and G. Ceder, Science, 2006, 311, 977–980.
12. C. H. Chen, J. Liu, M. E. Stoll, G. Henriksen, D. R. Vissers and K. Amine, J. Power Sources, 2004, 128, 278–285.
13. W. Liu, P. Oh, X. Liu, M. J. Lee, W. Cho, S. Chae, Y. Kim and J. Cho, Angew. Chem., Int. Ed., 2015, 54, 4440–4457.
14. J. Shim and K. A. Striebel, J. Power Sources, 2003, 122, 188–194.
15. S. Muto, K. Tatsumi, Y. Kojima, H. Oka, H. Kondo, K. Horibuchi and Y. Ukyo, J. Power Sources, 2012, 205, 449–455.
16. S. Madhavi, G. V. S. Rao, B. V. R. Chowdari and S. F. Y. Lia, J. Power Sources, 2001, 93, 156–162.
17. K. K. Lee, W. S. Yoon, K. B. Kim, K. Y. Lee and S. T. Hong, J. Power Sources, 2001, 97–98, 308–312.
18. Y. Itou and Y. Ukyo, J. Power Sources, 2005, 146, 39–44.
19. T. Sasaki, T. Nonaka, H. Oka, C. Okuda, Y. Itou, Y. Kondo, Y. Takeuchi, Y. Ukyo, K. Tatsumi and S. Muto, J. Electrochem. Soc., 2009, 156, A289–A293.
20. S. H. Lee, C. S. Yoon, K. Amine and Y. K. Sun, J. Power Sources, 2013, 234, 201–207.
21. S. T. Myung, K. Izumi, S. Komaba, H. Yashiro, H. J. Bang, Y. K. Sun and N. Kumagai, J. Phys. Chem. C, 2007, 111, 4061–4067.
22. Z. Li, G. D. Smith and D. Bedrov, J. Phys. Chem. B, 2012, 116, 12801–12809.
23. A. Guerfi, M. Dontigny, P. Charest, M. Petitclerc, M. Lagacé, A. Vijh and K. Zaghib, J. Power Sources, 2010, 195, 845–852.
24. R. S. Kühnel, M. Lübke, M. Winter, S. Passerini and A. Balducci, J. Power Sources, 2012, 214, 178–184.
25. R. S. Kühnel, N. Böckenfeld, S. Passerini, M. Winter and A. Balducci, Electrochim. Acta, 2011, 56, 4092–4099.
26. L. Lombardo, S. Brutti, M. A. Navarra, S. Panero and P. Reale, J. Power Sources, 2013, 227, 8–14.
27. N. Wongittharom, T. C. Lee, I. M. Hung, S. W. Lee, Y. C. Wang and J. K. Chang, J. Mater. Chem. A, 2014, 2, 3613–3620.
28. D. R. MacFarlane, P. Meakin, J. Sun, N. Amini and M. Forsyth, J. Phys. Chem. B, 1999, 103, 4164–4170.
29. N. Wongittharom, T. C. Lee, C. H. Hsu, G. T. K. Fey, K. P. Huang and J. K. Chang, J. Power Sources, 2013, 240, 676–682.
30. N. Wongittharom, C. H. Wang, Y. C. Wang, G. T. K. Fey, H. Y. Li, T. Y. Wu, T. C. Lee and J. K. Chang, J. Power Sources, 2014, 260, 268–275.
31. C. Arbizzani, G. Gabrielli and M. Mastragostino, J. Power Sources, 2011, 196, 4801–4805.
32. C. J. Han, J. H. Yoon, W. I. Cho and H. Jang, J. Power Sources, 2004, 136, 132–138.
33. T. Kawamura, S. Okada and J. I. Yamaki, J. Power Sources, 2006, 156, 547–554.
34. X. Kang, Chem. Rev., 2004, 104, 4303–4417.
35. J. Zheng, D. Zhu, Y. Yang and Y. Fung, Electrochim. Acta, 2012, 59, 14–22.
36. J. Zhang, N. Nie, Y. Liu, J. Wang, F. Yu, J. Gu and W. Li, ACS Appl. Mater. Interfaces, 2015, 7, 20134–20143.
37. Y. Liu, J. Gu, J. Zhang, J. Wang, N. Nie, Y. Fu, W. Li and F. Yu, Electrochim. Acta, 2015, 173, 448–457.
38. F. Castiglione, E. Ragg, A. Mele, G. B. Appetecchi, M. Montanino and S. Passerini, J. Phys. Chem. Lett., 2011, 2, 153–157.
39. K. A. Striebel, J. Shim, E. J. Cairns, R. Kostecki, Y.-J. Lee, J. Reimer, T. J. Richardson, P. N. Ross, X. Song and G. V. Zhuang, J. Electrochem. Soc., 2004, 151, A857–A866.
40. J. A. Choi, E. G. Shim, B. Scrosati and D. W. Kim, Bull. Korean Chem. Soc., 2010, 11, 3190–3194.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra21386j

---