Development of a γ-polyglutamic acid binder for cathodes with high mass fraction of sulfur

Abstract

Lithium–sulfur (Li–S) batteries have been one of the most attractive secondary battery systems to take advantage of their high theoretical energy densities and the low cost of sulfur. However, to realize their wide-scale commercial use, a high sulfur fraction is essential for the sulfur cathode. It is well known that the binder has a direct impact on the properties of the sulfur cathode. In this paper, γ-polyglutamic acid (PGA) was adopted as a functional binder in Li–S batteries. With plentiful electron-rich groups (amide, carboxyl) in the PGA molecular chain, the shuttle effect was restrained, and a good conductive network was formed. It was found that batteries using PGA as a binder exhibited better electrochemical cycling performance and rate capability compared to batteries using LA132, a common binder in sulfur cathodes. Even the PGA cathode contained a high sulfur mass fraction (77 wt%), the discharge capacity still remained 659 mA h g⁻¹ with a coulombic efficiency of around 99% after 200 cycles at 0.5C.

---

1. Introduction

Rechargeable lithium-ion batteries (LIBs), also called secondary LIBs, with a long charge/discharge life and high energy density have been the dominant power sources in commercial energy storage devices.^1–4 Nevertheless, those conventional lithium-ion (Li-ion) batteries possess a low theoretical energy density and leave limited space for further development in full electric/plug-in hybrid vehicles,^5–7 especially for the small mobile electronic devices that demand batteries of high volumetric energy density.^8–10 The gravimetric energy density and volumetric energy density of Li–S batteries are as high as 2500 W h kg⁻¹ and 2800 W h L⁻¹, respectively, owing to the high theoretical capacities of a Li anode (3860 mA h g⁻¹) and a S cathode (1672 mA h g⁻¹). The high abundance, low cost and environmental benignity of sulfur make Li–S batteries one of the most attractive secondary battery systems.^11,12 However, the following challenges exist with the Li–S battery which have been the main barriers for its commercialization:^13,14 firstly, a large amount of conductive additive is added into the cathode owing to the insulation of sulfur and the solid reduction products, which leads to the low mass fraction of the sulfur, so as the volumetric energy density; secondly, intermediate discharge products (Li₂S_x, 2 < x < 8) are highly soluble in organic electrolyte^15,16 causing the loss of active materials, rapid capacity fading as well as low coulombic efficiency; thirdly, for sulfur particles, the generation of the large volume change during charge and discharge process has a huge destruction to cathode structure.

Variety of methods have been proposed to figure out these issues,^17–20 including the decoration of various functional groups on carbon materials and development of various carbon/sulfur composites as well as using polymer coatings as a physical barrier to retard polysulfides or porous oxide additives to physically sorb soluble polysulfides. Jiangxuan Song²¹ developed a mesoporous and nitrogen-doped carbon (MPNC)–sulfur nanocomposite as a cathode for Li–S batteries, taking advantage of its efficient in promoting chemical adsorption between sulfur atoms and oxygen functional groups to help immobilize sulfur, and thus efficiently improve the utilization of active materials and reduce the shuttle effect. Dae Soo Jung et al.²² introduced a hierarchical porous carbon structure in which meso- and macro-pores are surrounded by outer micropores, and highly stable cycling of Li–S batteries were achieved with the helping of micropores to inhibit the fatal dissolution of the lithium polysulfides into the electrolyte. Guanchao Wang et al.²³ modified separator with a composite that grafted multi-walled carbon nanotube (MCNT) with polyethylene glycol (PEG), along with the high electron conductivity and strong chemical and physical absorption properties of the MCNT@PEG layer, and Li–S batteries possessed an excellent rate capability and high cycle stability.

Even though a high sulfur mass fraction is very essential to achieve high volumetric energy density of the batteries, many of strategies still suffered from a low sulfur mass fraction (≤65 wt%). As in the sulfur cathodes, the binder plays a role in binding the active materials and the conductive additive together, and the surface of active material is partially covered with the binder, binders not only provide conduction network for the solid sulfur species but also have a huge influence on chemical environmental of sulfur species. Additionally, the physical and chemical effect of the binders to the long-term cycling performance, kinetics and structural stability of electrode materials have been discussed in recent years.^24,25 Furthermore, binders that can increase the mass fraction of active materials and the volumetric energy density through constructing a excellent conductive network for insulated sulfur without hindering the performance of the battery and can minimize loss of polysulfides into the electrolyte during cycling are highly desirable. Seh et al.²⁶ used ab initio simulations to elucidate the interaction between LiPSs and lithium polysulfides and found that the electron-rich groups with lone pairs on oxygen, nitrogen and halogen atoms are capable of binding with lithium in polysulfides Li₂S_n (4 ≤ n ≤ 8). Moreover, the most stable interaction lay between Li⁺ cations and the doubly-bonded oxygen atom in the [double bond splayed left]C=O group owning to the hard acid nature of Li⁺. Zhang et al.²⁷ have successfully employed PAA as a binder in aqueous solvent to prepare the sulfur cathode for Li–S battery. Electrochemical tests indicate that the PAA sulfur cathode has smaller resistance, better kinetics characteristics and cyclic stability than that of the PVDF sulfur cathode. In this paper, γ-polyglutamic acid (PGA) was adopted as a water-soluble functional binder, with LA132 (a commonly used binder material in sulfur cathode) as a comparison. Different from PGA which the electron-rich functional groups (amide, carboxyl) are on every chain units, for LA132, acrylonitrile makes up the basis of polymer and the electron-rich functional groups only appear on the acrylate and acrylamide units distributed over the chains. This leads to a higher density of electron-rich functional groups on PGA. Thus, the using of PGA showing several advantages: (i) the massive hydrophilic functional groups (amide, carboxyl) of PGA help the infiltration of electrolyte in the cathode; (ii) those groups could render strong interaction with Li⁺ in polysulfides Li₂S_n (4 ≤ n ≤ 8), and be beneficial to curb the loss of the soluble polysulfides into the electrolyte and alleviate the degree of aggregation for sulfur species (Scheme 1); (iii) an intimate multidimensional contact between the cathode particles (sulfur and carbon particles) and the binder could be attained by noncovalent interactions, such as hydrogen bonding and van der Waals forces,²⁸ resulting in an excellent ionic conductivity and electrical conductivity. As a result, a good cycle performance and rate capability could be obtained through a simple cathode produce process even though the sulfur mass fraction of the cathode has reached 77 wt%.

Scheme 1 A sketch of the sulfur cathode using the PGA as binder.

2. Experimental

2.1. The fabrication of cathode

To compare the performance, both LA132 (15%, Chengdu Yindile, Corp. China) and PGA (M_w = 1 000 000, 92%, Beijing Huawei, Corp. China) were adopted as binders for the sulfur cathodes. The cathode slurry was composed of elemental S, AB and binders dispersed in water solvent. The mixture was mixed by a ball-milling method 5 hours to obtain uniform slurry. The slurry was then coated onto Al foil with a doctor blade. The fabricated cathodes were vacuum dried (vacuum degree: 0.1106 Pa) at 60 °C for 12 h. For the cathode using LA132 as the binder, the mass ratio of S (99.5%, analytical grade, Beijing Yili, Corp., China), acetylene black (AB, Jinpu. Corp., China) and LA132 was 70 : 23 : 07 (LA132-70S cathode). There are two different proportions for the cathode used PGA as the binder. And the weight ratio of the two cathodes were: 70 : 23 : 07 (PGA-70S cathode); 77 : 16 : 07 (PGA-77S cathode). The average sulfur loading in the cathodes is about 1.2–1.5 mg cm⁻².

2.2. Measurements

All the cathodes were cut into disks with diameter 12 mm to assemble the batteries in an Ar-filled glove box (Aldrich). The anode was lithium foil and the electrolyte was 1 M lithium bistrifluoromethanesulfonylimide (LiTFSI, Acros Organics) and 0.4 M lithium nitrate (Acros Organics) in a 1 : 1 volume ratio of 1,3-dioxolane (DOL, Acros Organics) and 1,2-dimethoxyethane (DME, Acros Organics). The discharge tests were carried out in the voltage range of 1.7–2.8 V, with a LANDCT2001A instrument under room temperature. The morphology of the electrode surface was examined using a scanning electron microscopy (SEM, HITACHIS-4700) operated at 5 kV. Electrochemical impedance spectroscopy (EIS) was measured at CHI660C electrochemical workstation from 1 MHz to 100 mHz with an AC voltage amplitude of 5 mV.

3. Results and discussion

γ-Polyglutamic acid (PGA) as a homopolymer of glutamic acid, showing water solubility, edibility, non-toxicity to humans and the environment (Fig. 1). Its properties qualify it as an important material in various applications including being used as thickener, humectant, sustained release materials, biopolymer flocculants heavy metal absorber and biological adhesive. The massive hydrophilic functional groups (–CONH, –COOH) of PGA were confirmed by the Fourier-transform infrared spectrometry (FTIR) spectroscopy (Fig. 1d). The FTIR absorption spectra of PGA are constituted of the peaks of stretching vibration of O–H and N–H at 3430 cm⁻¹, C=O stretching vibration coupled amide II bands at 1594 cm⁻¹, N–H bend vibration coupled C–N (amide III) at 1268 cm⁻¹, and the C–O stretching band located at 1128 cm⁻¹.²⁹

Fig. 1 Image showing (a) dry PGA power and (b) the PGA power dissolved in deionized water forming an aqueous solution, (c) image of electrode coated with a slurry having the PGA binder, (d) FTIR spectra of PGA and PGA mixed with polysulfides (Li₂S₈).

To confirm that the PGA could afford affinitive interactions with polysulfides, the FTIR spectroscope (Fig. 1d) of PGA mixed with polysulfides (Li₂S₈) has been analyzed. As shown in Fig. 1d, the emergence of C–S bond at ≈600 cm⁻¹ suggested that the PGA was capable of forming chemical bonds with polysulfides. And the peak at ≈500 cm⁻¹ was attributed to S–S bending variation.^30,31 Besides, O–H stretching vibration coupled N–H stretching vibration shifted from 3430 cm⁻¹ to 3413 cm⁻¹, the N–H bend vibration coupled C–N (amide III) shifted from 1268 cm⁻¹ to 1201 cm⁻¹, the C=O stretching vibration coupled amide II bands and C–O stretching vibration were respectively located at 1612 cm⁻¹ and 1064 cm⁻¹ corresponding to the peaks appearing at 1594 cm⁻¹ and 1128 cm⁻¹.^32–34 All those changes indicated that PGA could chemically fix the sulfur species.

The electrochemical performance of the cathodes using both LA132 and PGA as binders were presented in Fig. 2. As shown in Fig. 2a, at a low current density of 0.1C, the initial capacity of the LA132-70S cathode was 974 mA h g⁻¹ and decreased to 511 mA h g⁻¹ (capacity retention: 52.5%) rapidly after 100 cycles, while the PGA-70S cathode delivered an initial discharge capacity of 1036 mA h g⁻¹, which was much higher than the LA132-70S cathode. The higher initial discharge capacity of the PGA-70S cathode might arise from the fact that the PGA binder could ensure intimate contact between the sulfur and conductive agent and improve the utilization of sulfur.²⁸ The PGA-70S cathode showed stable cycling performance, and the discharge capacity remained at 727 mA h g⁻¹ (capacity retention: 70.2%) at the 100th cycle, which is still higher than that of the LA132-70S cathode. This comparison indicated that the electron-rich groups (amide, carboxyl) of PGA binder could immobilize the lithium polysulfides and suppress the severe capacity fading.²⁶

Fig. 2 Electrochemical analysis and battery performance of cathodes containing 70 wt% sulfur: discharge/charge profiles of Li–S batteries employing PGA and LA132 binder at various cycling rates of (a) 0.1C, (b) 0.2C, and (c) 0.5C (the batteries were discharge at 0.05C for the first two cycles). (d) Rate performance of PGA-70S cathode and LA132-70S cathode. (e) The galvanostatic charge–discharge curves of PGA-70S cathode at different cycles. (f) The galvanostatic charge–discharge curves of LA132-70S cathode at different cycles.

When the current density increased to 0.2C (Fig. 2b) and 0.5C (Fig. 2c), PGA-70S cathodes still exhibited good cycle performance while an activation process (10 cycles for 0.2C and 80 cycles for 0.5C) was necessary for the cathode using LA132 binder. It suggested that PGA would be beneficial for the improvement of the Li⁺ ionic conductivity and electronic conductivity of the sulfur cathode, which is confirmed in the later discussion.³⁵

Then, the rate performances of the PGA-70S cathode and LA132-70S cathode were illustrated in Fig. 2d. It is obviously shown that, for the LA132-70S cathode, with the current rate varied from 0.1C to 1.0C, significant decline of capacity was observed due to the polarization effect, and the capacity was far below the capacity of PGA-70S cathode during the discharge process. In contrary, the PGA-70S cathode exhibited an excellent and highly reversible rate performance. It was noteworthy that even though the current density increased from 0.2C to 0.5C, the capacity of the PGA-70S cathode showed little reduction. And the reversible capacities achieved at 0.5C and 1C correspond to 99% and 98% of the capacity that was obtained at 0.2C, when switching back to 0.1C, the reversible capacity was largely recovered (Fig. 2d), and the capacity achieved at 41st cycle (0.1C) correspond to 97% of the capacity that was obtained at 10th cycle (0.1C). Those results indicated that the PGA-70S cathode has both fast reaction kinetics and a highly stable structure.

Fig. 2e and f presented the discharge and charge voltage profiles of cathodes with different binders at 0.1C. As shown in Fig. 2e and f, even at a high sulfur content of 70 wt%, both PGA-70S cathode and LA132-70S cathode showed typical two plateaus during discharge. The upper-discharge plateau at 2.4–2.0 V is attributed to the reduction of sulfur into high-order lithium polysulfides.^36,37 In Fig. 2e and f, a shrinkage of the upper-discharge plateaus was observed in LA132-70S cathode during repeated cycling. And for the PGA-70S cathode, the complete overlapping of upper discharge voltage plateaus (Fig. 2e) along with little decrease in capacity. The lower-discharge plateau at 2.05–1.8 V is attributed to the further reduction of highly soluble polysulfides to insoluble Li₂S₂ and Li₂S. The PGA-70S cathode maintained a long, flat lower-discharge plateau while LA132-70S cathode showed noticeable shrinkage in lower-discharge plateaus and diminishment in capacity in the repeated cycles. It indicated that PGA binder could improve the electrochemical accessibility and reversibility of the sulfur cathode.³⁸ Furthermore, the PGA-70S cathode showed the lowest charge plateau potential and highest discharge plateau potential than LA132-70S cathode, resulting in a relatively lower polarization (ΔE = 0.28 V) than LA132-70S cathode (ΔE = 0.46 V). In this paper, the sulfur content of two cathodes (PGA-70S cathode and LA132-70S cathode) is very high. There is no effective suppression of the diffusion of polysulfide in the LA132-70S cathode. Consequently, those polysulfides dissolved into the electrolyte quickly and continuously reduced into short-chain polysulfides. In contrast, the good dispersion of materials in PGA-70S cathode can alleviate the polarization. More importantly, a high density of electron-rich functional groups on PGA molecular chain that could curb the loss of the soluble polysulfides into the electrolyte.³⁹ As a result, good charge/discharge performance could be obtained in PGA cathode.

Here, the morphology of the PGA-70S cathode and the LA132-70S cathode were also investigated. We found that there was a homogeneous distribution of cathode materials in the PGA-70S cathode because of the strong dispersion of the PGA binder (Fig. 3a), and it would be helpful to construct a good conductive network. And for the LA132-70S cathode, there were still some tiny sulfur blocks (Fig. 3a′). After 50th cycles, a layer of solid-phase substance was obviously observed to be coated on the material particle surface, resulting bigger particles in PGA-70S cathode (Fig. 3b) and LA132-70S cathode (Fig. 3b′). According to the previous reports, the solid-phase substances are precipitated sulfur species (S₈ and Li₂S/Li₂S₂).⁴⁰ Unlike the uniform distribution of precipitated sulfur and the porosity on the surface of the PGA-70S cathode (Fig. 3b), for the LA132-70S cathode, the precipitated sulfur on the cathode surface resulted the aggregation of the cathode. Besides, compared with PGA-70S cathode, the electrolyte of LA132-70S cathode in the separator exhibited a deeper red colour (Fig. 3b′), which means more polysulfide was dissolved in the electrolyte. Thus, it could be inferred that, for the PGA-70S cathode, the extensive carbonyl and amide groups can anchor the polysulfide within the cathode and the dissolution of polysulfide was restricted. As a result, the PGA-70S cathode showed a lower degree of aggregation for sulfur species and a more uniform sulfur precipitation. And this was in consistence with the previous electrochemical results in Fig. 2. In this work, the morphology, EDX analysis and elemental mapping of the lithium anodes of the both batteries after 50 cycles were also investigated. It was clear to see that, a smooth lithium anode surface (Fig. 3c) and a uneven lithium anode surface (Fig. 3c′) were observed for the batteries using PGA-70S cathode and LA132-70S cathode, respectively, and there were notches on the lithium anode surface that using LA132-70S cathode in the battery. Moreover, for the lithium anode of the battery using the PGA-70S cathode, the elemental mapping (Fig. 3d and d′) also showed that less sulfur which demonstrated that the groups in the PGA molecule successfully inhabited the dissolve of the polysulfide. It could further reveal the main reason of the excellent electrochemical performance of PGA-70S cathode. Besides, the FTIR spectra (Fig. S1†) of PGA-70S cathodes before cycle and after cycle have been analysed. We still can find the C–S stretching appears at ≈600 cm⁻¹.

Fig. 3 SEM images of surface morphology of pristine cathode: PGA-70S cathode (a), LA132-70S cathode (a′); after 50 cycles: PGA-70S cathode (b), LA132-70S cathode (b′), inset of (b) and (b′) are the corresponding separators of the batteries; SEM images of morphology lithium anode after 50 cycles: using PGA-70S cathode (c) and using LA132-70S cathode (c′), inset of (c) and (c′) are the EDX of the lithium anode; sulfur-distribution mappings of lithium anode after 50 cycles: using PGA-70S cathode (d) and using LA132-70S cathode (d′).

In order to explore the effect of PGA binder on sulfur cathodes further, the electrochemical impedance spectra of the sulfur cathodes with different binders before cycle and after 10 cycles were given in Fig. 4. And our speculate was further confirmed by the EIS results.

Fig. 4 (a) Nyquist plots PGA-70S cathode and LA132-70S cathode before cycle, (b) Nyquist plots PGA-70S and LA132-70S cathode after 10 cycles. Inset: the equivalent circuits.

In Fig. 4a, there was a depressed semicircle in high to medium frequency regions and an inclined line in low frequency regions in the impedance spectrum of both PGA-70S cathode and LA132-70S cathode before cycle.^41,42 The semicircles at the high to medium frequency regions reflect the charge transfer resistance (R_ct) controlled by the electric conductivity of the cathode and the electrode–electrolyte interfacial resistance. And longer diameter of the semicircle corresponds to higher charge transfer resistance. The inclined line is associated with ion diffusion process. It could be seen that the PGA-70S cathode showed lower charge transfer resistance than LA132-70S cathode. It demonstrated that PGA binder enhanced the electronic conductivity through constructing a good conductive network to the cathode. In addition, the ionic conductivity of both PGA-70S cathode and LA132-70S cathode was compared^43,44 (see details in ESI†). The results showed that PGA-70S cathode had higher ionic conductivity than LA132-70S cathode which contributed to the better electrochemical performance of the PGA cathode. In Fig. 4b, it was obviously to see that the R_ct of PGA-70S cathode was significantly smaller than LA132-70S cathode after ten cycles. The lower R_ct of PGA-70S cathode than that of LA132-70S cathodes implied the much more effective electric contact and uniformly distribution of the cathode materials within the cathode structure. Thus, the R_ct of the cathodes confirmed the SEM results given in Fig. 3 that with the assistance of the carbonyl and amide group of the PGA, polysulfide could be fixed which can lead to a more uniform sulfur precipitation taking place, and less polysulfide dissolving into the electrolyte.

To understand the results better, we increased the discharge/charge rate to 1.0C, and the cyclical performance of the PGA-70S cathode discharge/charge at 0.2C and 1.0C was illustrated in Fig. 5a. Along with a superior capacity retention rate of 85.9%, 84.3%, respectively, and the respective reversible discharge capacities of 648 and 600 mA h g⁻¹ were obtained at the end of 200th cycle. The excellent high-rate performance corresponding to its rate performance displayed in Fig. 2d. Fig. 5b exhibited a long cycle performance of the PGA-77S cathode at a high current rate of 0.5C. From the 3rd cycle to the 200th cycle, the capacity of the PGA-77S cathode decayed from 915 mA h g⁻¹ to 659 mA h g⁻¹, representing an extremely low decay rate of 0.14% per cycle for 200 cycles. And the coulombic efficiency remained at around 99% at the same time. It, again, demonstrated that the PGA through immobilizing of the polysulfide could effectively improve the long cycle stability of the cathode. Furthermore, on the basis of the high sulfur fraction (77 wt%) cathode with a such superior cyclic stability at 0.5C, it is reasonable to conclude that an intimate multidimensional contact between the cathode materials that plays a significant role in accommodating the volume changes during charge and discharge.⁴⁵ More interesting, when the sulfur loading of the PGA-77S cathode was added to 2.2 mg cm⁻² (Fig. 5c), and the discharge capacity of this cathode still remained at 629 mA h g⁻¹ at 0.2C with coulombic efficiency of around 98% after 50 cycles. All of these results showed that the PGA success in suppressing the polysulfide shuttling effect via fixing the polysulfide uniformly in the cathode and affording sufficient active sites,⁴⁶ as well as accommodating the volume expansion effectively. Consequently, the PGA cathodes with a high sulfur mass fraction show an excellent high-rate capability and long life performance.

Fig. 5 Discharge/charge curves of PGA cathode: (a) with 70% sulfur mass fraction at two different cycling rates: 0.2C, 1C; (b) with 77% sulfur mass fraction at 0.5C; (c) 77% sulfur mass fraction with a high sulfur loading 2.2 mg cm⁻² at 0.2C.

4. Conclusions

In summary, PGA as a novel functional binder has been successfully developed for the high-performance Li–S battery. The existence of the abundant functional groups (amide, carboxyl) in the PGA molecule can offer many advantages for the high sulfur mass fraction sulfur cathodes. The plentiful electron-rich groups (amide, carboxyl) can curb the loss of the soluble polysulfides into the electrolyte and alleviate the degree of aggregation for sulfur species, thus the excellent rate capability and stable cycling performance have been obtained. Moreover, noncovalent interactions, such as hydrogen bonding and van der Waals forces are existing between the PGA molecules, resulting in an intimate multidimensional contact between the cathode particles (sulfur and carbon particles) and the binder. And this characteristic can be contributed to an excellent ionic conductivity and electrical conductivity during the discharge and charge process. With a high sulfur content of 70 wt%, the batteries with PGA binder still delivered a reversible capacity of 600 mA h g⁻¹ after 200 cycles at 1C. When the PGA cathode contained a high sulfur mass fraction (77 wt%) and high sulfur loading (2.21 mg cm⁻²), the discharge capacity of PGA cathode still remained at 629 mA h g⁻¹ with coulombic efficiencies of around 98% after 50 cycles. The big improvement of sulfur mass fraction promotes the volumetric energy density of the batteries, thus more conducive to the commercialization of the Li–S battery.

Acknowledgements

Financial support from the National Natural Science Foundation of China (No. 51272017 and 51432003) is gratefully appreciated.

Notes and references

1. J. B. Goodenough, Acc. Chem. Res., 2012, 46, 1053–1061.
2. H. Ni, W. Song and L. Fan, Prog. Nat. Sci., 2016, 26, 283–288.
3. L. Fan, Z. Dang, C. W. Nan and M. Li, Electrochim. Acta, 2002, 48, 205–209.
4. J. Zheng, G. Ji, P. Zhang, X. Cao, B. Wang, L. Yu and Z. J. Xu, Chemistry, 2016, 22, 7316.
5. F. T. Wagner, B. Lakshmanan and M. F. Mathias, J. Phys. Chem. Lett., 2010, 1, 2204–2219.
6. H. Zheng, R. Yang, G. Liu, X. Song and V. S. Battaglia, J. Phys. Chem. C, 2012, 116, 4875–4882.
7. X. Su, Q. Wu, J. Li, X. Xiao, A. Lott, W. Lu, B. W. Sheldon and J. Wu, Adv. Energy Mater., 2014, 4, 1300882.
8. S. Evers and L. F. Nazar, Acc. Chem. Res., 2012, 46, 1135–1143.
9. K. Abraham, J. Phys. Chem. Lett., 2015, 6, 830–844.
10. Y. S. Choi, S. Kim, S. S. Choi, J. S. Han, J. D. Kim, S. E. Jeon and B. H. Jung, Electrochim. Acta, 2004, 50, 833–835.
11. D. Peramunage and S. Licht, Science, 1993, 261, 1029–1032.
12. A. Manthiram, Y. Fu, S. H. Chung, C. Zu and Y. S. Su, Chem. Rev., 2014, 114, 11751–11787.
13. H. S. Ryu, Z. Guo, H. J. Ahn, G. B. Cho and H. Liu, J. Power Sources, 2009, 189, 1179–1183.
14. S. Zhang, K. Ueno, K. Dokko and M. Watanabe, Adv. Energy Mater., 2015, 5, 1500117.
15. X. Ji and L. F. Nazar, J. Mater. Chem., 2010, 20, 9821–9826.
16. Q. Zou and Y. C. Lu, J. Phys. Chem. Lett., 2016, 7, 1518–1525.
17. J. Li, J. Guo, L. Zeng, Y. Huang and R. Peng, RSC Adv., 2016, 6, 26630–26636.
18. X. Fang, W. Weng, J. Ren and H. Peng, Adv. Mater., 2016, 28, 491–496.
19. B. Papandrea, X. Xu, Y. Xu, C.-Y. Chen, Z. Lin, G. Wang, Y. Luo, M. Liu, Y. Huang and L. Mai, Nano Res., 2016, 9, 240–248.
20. P. Y. Gu, Y. Zhao, J. Xie, N. B. Ali, L. Nie, Z. J. Xu and Q. Zhang, ACS Appl. Mater. Interfaces, 2016, 8, 7464–7470.
21. J. Song, T. Xu, M. L. Gordin, P. Zhu, D. Lv, Y. B. Jiang, Y. Chen, Y. Duan and D. Wang, Adv. Funct. Mater., 2014, 24, 1243–1250.
22. D. S. Jung, T. H. Hwang, J. H. Lee, H. Y. Koo, R. A. Shakoor, R. Kahraman, Y. N. Jo, M.-S. Park and J. W. Choi, Nano Lett., 2014, 14, 4418–4425.
23. G. Wang, Y. Lai, Z. Zhang, J. Li and Z. Zhang, J. Mater. Chem. A, 2015, 3, 7139–7144.
24. M. J. Lacey, F. Jeschull, K. Edström and D. Brandell, J. Power Sources, 2014, 264, 8–14.
25. Y. Wang, Y. Huang, W. Wang, C. Huang, Z. Yu, H. Zhang, J. Sun, A. Wang and K. Yuan, Electrochim. Acta, 2009, 54, 4062–4066.
26. Z. W. Seh, Q. Zhang, W. Li, G. Zheng, H. Yao and Y. Cui, Chem. Sci., 2013, 4, 3673.
27. Z. Zhang, W. Bao, H. Lu, M. Jia, K. Xie, Y. Lai and J. Li, ECS Electrochem. Lett., 2012, 1, A34–A37.
28. F. Zeng, W. Wang, A. Wang, K. Yuan, Z. Jin and Y. S. Yang, ACS Appl. Mater. Interfaces, 2015, 7, 26257–26265.
29. X. Liu, L. Luo, Y. Ding and D. Ye, Bioelectrochemistry, 2011, 82, 38–45.
30. G. Li, M. Ling, Y. Ye, Z. Li, J. Guo, Y. Yao, J. Zhu, Z. Lin and S. Zhang, Adv. Energy Mater., 2015, 5, 1500878.
31. N. Liu, B. Huang, W. Wang, H. Shao, C. Li, H. Zhang, A. Wang, K. Yuan and Y. Huang, ACS Appl. Mater. Interfaces, 2016, 8, 6101–16107.
32. C. T. Tsao, C. H. Chang, Y. Y. Lin, M. F. Wu, J. L. Wang, H. Y. Tai, J. L. Han and K. H. Hsieh, Carbohydr. Polym., 2011, 84, 812–819.
33. C. C. Wang, C. H. Su, J. P. Chen and C. C. Chen, Mater. Sci. Eng., C, 2009, 29, 1715–1724.
34. Y. H. Lin, C. K. Chung, C. T. Chen, H. F. Liang, S. C. Chen and H. W. Sung, Biomacromolecules, 2005, 6, 1104–1112.
35. Y. Zhang, Y. Zhao, D. Gosselink and P. Chen, Ionics, 2014, 21, 381–385.
36. Y. Chen, S. Lu, X. Wu and J. Liu, J. Phys. Chem. C, 2015, 119, 10288–10294.
37. H. J. Peng, T. Z. Hou, Q. Zhang, J. Q. Huang, X. B. Cheng, M. Q. Guo, Z. Yuan, L. Y. He and F. Wei, Adv. Mater. Interfaces, 2014, 1, 1400227.
38. S. H. Chung, R. Singhal, V. Kalra and A. Manthiram, J. Phys. Chem. Lett., 2015, 6, 2163–2169.
39. S. S. Zhang, D. T. Tran and Z. Zhang, J. Mater. Chem. A, 2014, 2, 18288–18292.
40. G. Ai, Y. Dai, Y. Ye, W. Mao, Z. Wang, H. Zhao, Y. Chen, J. Zhu, Y. Fu, V. Battaglia, J. Guo, V. Srinivasan and G. Liu, Nano Energy, 2015, 16, 28–37.
41. S. Narayanan, D. Shen, S. Surampudi, A. Attia and G. Halpert, J. Electrochem. Soc., 1993, 140, 1854–1861.
42. L. Yuan, X. Qiu, L. Chen and W. Zhu, J. Power Sources, 2009, 189, 127–132.
43. H. Wang, V. Sencadas, G. Gao, H. Gao, A. Du, H. Liu and Z. Guo, Nano Energy, 2016, 26, 722–728.
44. Y. Cui, X. Zhao and R. Guo, Electrochim. Acta, 2010, 55, 922–926.
45. S. Lu, Y. Cheng, X. Wu and J. Liu, Nano Lett., 2013, 13, 2485–2489.
46. J. Balach, T. Jaumann, M. Klose, S. Oswald, J. r. Eckert and L. Giebeler, J. Phys. Chem. C, 2015, 119, 4580–4587.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra20504f

---