Enhanced electrochemical performance by a three-dimensional interconnected porous nitrogen-doped graphene/carbonized polypyrrole composite for lithium–sulfur batteries

Abstract

Three-dimensional interconnected porous nitrogen-doped graphene/carbonized polypyrrole nanotube (N-GP/CPN) materials have been fabricated via carbonization and chemical activation of polypyrrole-functionalized graphene nanosheets with KOH. The obtained N-GP/CPN with high surface, abundant nanopores and nitrogen doping can serve as conductive substrates for hosting a high content of sulfur and can effectively impede the dissolution of polysulfides. The N-GP/CPN-S composite exhibits excellent electrochemical performance as the cathode material for lithium–sulfur (Li–S) batteries, including a high initial discharge capacity of 1128 mA h g⁻¹ at 0.5C, a notable cycling stability with a high stable capacity of 726 mA h g⁻¹ and an ultraslow decay rate of 0.07% per cycle as long as 500 cycles. Moreover, the N-GP/CPN-S cathode also exhibits good rate capacity, showing a high reversible stable capacity of 687 mA h g⁻¹ at 4C.

---

Introduction

The ever-increasing consumer demands for faster, smaller, more powerful personal electronic devices have sparked research in advanced rechargeable battery systems with higher specific capacities, lower cost and longer lifetimes beyond traditional lithium-ion batteries (LIBs).^1–3 Li–S batteries, which use abundant and non-toxic sulfur as the cathode material, have been envisaged as one of the most appealing candidates. The theoretical capacity of sulfur is 1675 mA h g⁻¹ and the specific energy density of Li–S batteries is as high as 2500 W h kg⁻¹.^4–7 Nevertheless, commercially viable Li–S batteries still face several technical challenges. First, sulfur and its reduction product Li₂S are highly electrically and ionically insulating materials, leading to low utilization of active material and coulombic efficiency.^8,9 Second, large volumetric expansion of sulfur to Li₂S (ca. 80%) can result in cracking and fracturing of the cathode and fast capacity fading.¹⁰ Third, the dissolution and “shuttle effect” of lithium polysulfides (Li₂S_x), the intermediate products formed during charge/discharge processes, in liquid electrolyte, produce a rapid decline in the capacity and short cycle life.¹¹ Therefore, it is crucial to improve the electrical conductivity and structural stability of the cathode, and block the dissolution of lithium polysulfides, then enhance the electrochemical performance of Li–S batteries.

One of the most effective way is coupling sulfur with electrically conducting agents to partly overcome the demerits of sulfur cathode. The conductive polymers^12–16 have been reported as sulfur hosts due to the good compatibility with sulfur and electrolytes.¹⁷ Polypyrrole (PPy) with hollow sphere^16,18 or tubular type,¹⁹ which have good absorbing ability to the sulfur and polysulfides, exhibited improved cyclability of Li–S batteries. However, the electrochemical performance is still unsatisfactory because of the low pore volume and electrical conductivity of PPy. Carbon materials, such as porous carbon,^20,21 carbon nanotubes/fibers,²² and graphene,²³ have been engineered to form a conductive network for storing the active material, trapping the dissolved polysulfides, then much improved the discharge capacity and the cycle life of Li–S batteries. Among these carbon materials, two-dimensional (2D) graphene with high specific surface area, excellent mechanical flexibility and stability, has been extensively studied and used in Li–S batteries. However, the simplex 2D structured nanomaterials are not effective to contain high content of sulfur and confine the lithium polysulfides from dissolving out of the composites. Recently, three-dimensional (3D) nanostructured graphene have been emerged as efficient host to encapsulate sulfur as cathode material for Li–S batteries due to its unique structure and properties, achieving significantly enhanced reversible capacity, cycling stability and rate capability.^24,25 The 3D conductive network can efficiently enhance the interaction between the active materials electrolytes, shorten diffusion paths for charge transport, and reduce electron transfer resistant.^6,26 Furthermore, the porous structure within framework are favored for high sulfur loading and retarding polysulfides dissolution. In addition of the structure modification of graphene, the optimization by tailoring functional sites can also manipulate its chemical reactivity and electronic property. A particularly interesting example is nitrogen-doping which induces a large number of active defects at the edges of graphene. It exhibits more effectively chemical adsorption of sulfur and lithium polysulfides on the high-surface-area graphene nanolayers and leads to much better electrochemical performances for Li–S batteries.^27,28

Herein, a novel 3D interconnected porous graphene/carbonized PPy nanotube materials with nitrogen-doping were prepared by facile in situ polymerization, carbonization and chemical activation. N-GP/CPN-S composites were further synthesized through melt diffusion method and used as cathode materials for Li–S batteries, as shown in Scheme 1. The N-GP/CPN with high conductivity and abundant nanopores not only serve as conductive substrates for electrons transport, but favored for firmly depositing sulfur. Meanwhile, the nitrogen doping elevates adsorption of sulfur on nitrogen-containing functional groups and effectively impede the dissolution of polysulfides in the carbon framework. As a result, the as-prepared N-GP/CPN-S composite exhibits superior rate capability and excellent cycle life.

Scheme 1 Schematic illustration of the synthesis procedure of the N-GP/CPN-S.

Experimental

Preparation of compound

A uniform graphite oxide (GO) suspension (0.5 mg mL⁻¹, 100 mL) was synthesized by a modified Hummers method²¹ and prepared through ultrasonication for 2 h. Methyl orange (MO, 1 mmol) was dissolved in the GO suspension, then 20 mL of 0.5 mol L⁻¹ FeCl₃ solution was added dropwise to the above suspension with constant sonication to reacted with MO as the template. After ultrasonicated for 10 min, pyrrole monomer (10 mmol) was added and the mixture was stirred at room temperature for 24 h. The formed GO/PPy precipitate was washed with deionized water several times until filtrate was colorless and neutral. Then, the mixture of GO/PPy and KOH with a mass ratio of 1 : 1 in ethanol/water co-solvent was stirred for 2 h. After evaporation of the solvents, the mixture was transferred into a tube furnace and heat-treated at 800 °C for 2 h under N₂ flow at a heating rate of 2 °C min⁻¹. The obtained product was thoroughly washed with HCl solution (1 M), filtered and then freeze dried to generate porous N-GP/CPN.

The N-GP/CPN-S composite was further synthesized using the mixture of N-GP/CPN composite and sulfur with mixed a mass ratio 1 : 4 and heated at 155 °C for 12 h under N₂ flow with a heating rate of 0.5 °C min⁻¹. The GO/PPy-S composite was synthesized using the same method.

Materials characterization

The phase of all as-prepared samples were characterized by X-ray diffraction (XRD, Bruker D8) using a Cu Kα radiation source between 10° and 70°. The morphologies and elemental distributions of the samples were observed with scanning electron microscope (SEM, Hitachi S-4800) and energy dispersive spectroscopy (EDS). Transmission electron microscopy (TEM) was imaged on FEI Tecnai G2 S-Twin instrument. The Raman spectra were performed on a micro-Raman system (Renishaw, in via). X-ray photoelectron spectrometry (XPS) spectra were obtained with an ESCALABMKLL X-ray photoelectron spectrometer using an Al Kα source. Thermogravimetric (TGA) curves were obtained on a STA 449 °C Jupiter (NETZSCH) thermogravimetry analyzer from room temperature to 800 °C under Ar₂ flow with a heating rate of 10 °C min⁻¹. N₂ adsorption–desorption measurements were conducted on a Micromeritics ASAP 2010 instrument. Surface areas were calculated using the Brunauer–Emmett–Teller (BET) method.

Electrochemical measurements

The working electrode was prepared by spreading slurry containing 80% active material, 10% KB and 10% PVDF onto an Al foil current collector. After vacuum drying at 60 °C for about 24 h, the electrode disks (d = 12 mm) were punched, tabletted and weighed. The sulfur mass loading on the electrode is about 0.8–1.0 mg cm⁻². Coin-type cells of 2025 were assembled in an Ar-filled glove box with H₂O and O₂ content below 1 ppm, using lithium foil as the counter electrode and Cellgard 2300 microporous membrane as the separator. The cells were filled with 50 μL electrolyte. The electrolyte was 1.0 M LiN(CF₃SO₂)₂ (LiTFSI) in a mixed solvent of dimethoxyethane (DME) and 1,3-dioxolane (DOL) with a volume ratio of 1 : 1, including 0.1 M LiNO₃ as an electrolyte additive. The cells were charged and discharged at different current rates between 1.8 and 2.7 V (vs. Li/Li⁺) using LAND CT2001A multi-channel battery testing system at room temperature. The electrochemical impedance spectroscopy (EIS) measurements and cyclic voltammetry (CV) measurements were carried out with a BioLogic VMP3 station. EIS measurements were performed over frequency ranges from 700 kHz to 100 mHz in automatic sweep mode. CV experiment was performed in the range of 1.8–2.7 V at a sweep rate of 0.05 mV s⁻¹. In this context, 1C corresponds to a current density value of 1675 mA g⁻¹, the specific capacity was calculated on the basis of sulfur mass.

Results and discussion

The graphite oxide/polypyrrole (GO/PPy) materials are prepared by chemical polymerization of pyrrole monomer on the surface of GO nanolayers due to the strong π–π interactions, electrostatic attractions and hydrogen bonding.^29–31 The complex of oxidant FeCl₃ and anionic dopant MO are used as reactive seed templates directing the growth of PPy nanotubes. The morphology of the typical as-prepared GO/PPy is first investigated by SEM. As shown in Fig. 1a, 1D PPy nanotubes are formed and adsorbed on the large surface of the 2D GO sheets, which are irregularly stacked together. After carbonization and activation processes, the architectures of N-GP/CPN stacked by well-exfoliated thin laminar graphene sheets are clearly observed in Fig. 1b. TEM image of N-GP/CPN in Fig. 1c presents the 3D interconnected architecture morphology which is composed by well dispersed graphene nanolayers and sandwiched PPy nanotubes. Fig. 1d indicates the in situ growth of PPy with tubular structure in diameters of 60–80 nm on the GO nanosheets. Both 1D tubular and 3D interconnected architectures are crucial in facilitating the diffusion of electrolyte and transport of electrons.

Fig. 1 SEM images of (a) GO/PPy and (b) N-GP/CPN composites, (c) low and (d) high resolution TEM images of N-GP/CPN composite.

Fig. 2a shows the Raman spectra for the as-prepared samples. The G band as a characteristic feature of regular graphene structure is usually attributed to the in-plane vibrations of the sp² carbon atom.²⁹ The D band is associated with the disordered carbon or defective graphitic structures. The intensity ratio of the D and G bands (I_D/I_G) indicates the disorder in the graphene materials, which is 1.16 for N-GP/CPN (0.88 for GO/PPy). The high value could be attributed to the breakdown and modification of N atoms at the edge of graphene sheets.³² Furthermore, the elemental composition and the bonding configurations of nitrogen functionalities after carbonization processes are investigated by XPS. The XPS survey spectra of GO/PPy and N-GP/CPN in Fig. 2b both show the presence of the C 1s at 285 eV, N 1s peak at 400 eV and O 1s peak at 532 eV.²⁹ Fig. 2c discerns the high-resolution C 1s spectrum of GO/PPy composite which combines the characteristics of both GO and PPy, and divided into four peaks with binding energies at about 284.5, 285, 286.3 and 288.8 eV, corresponding to sp² graphitic carbon, C 1s, C–N/C–O and –COO–.³³ Meanwhile, as shown in Fig. 2d, the peaks in the C 1s spectrum of N-GP/CPN composite exhibit a little difference. The presence of the C=N/C=O peak at 288.3 eV and the disappearance of the –COO– peak suggest the bonding formation of nitrogen atoms to be sp³ carbon and the carbonization of GO, respectively.³⁴ Fig. 2e shows the deconvoluted N 1s spectrum of the GO/PPy composite. The peak at 399.7 eV corresponds to the neutral amine nitrogen (C–N), two major peaks at 397.7 and 400.5 eV are assigned to the imine nitrogen (–N=) and the positively charged nitrogen (–N⁺),³³ respectively, both of which are changed in the N 1s spectrum of the N-GP/CPN composite. The N 1s peak (Fig. 2f) could be deconvoluted into three different peaks with binding energies at about 398.5, 399.6, and 401.2 eV, consistent with pyridinic-N, pyrrolic-N, and graphitic-N,²⁹ separately. Obviously, doping nitrogen atoms into graphene sheets can converts nitrogen atoms within the pentagonal ring of polypyrrole into three types of N-configurations: pyridine-like, pyrrole-like, graphite-like, (Fig. 2g) and induces more defects, which can provide more active sites to chemical adsorb sulfur on the high-surface-area graphene nanosheets.

Fig. 2 (a) Raman spectra and (b) XPS survey scan of N-GP/CPN and GO/PPy composites, high resolution scan of C 1s of (c) GO/PPy and (d) N-GP/CPN composite, N 1s of (e) GO/PPy and (f) N-GP/CPN composite, and (g) schematic structure of N-GP/CPN.

The pore structure and BET surface area of N-GP/CPN are characterized by N₂ adsorption–desorption measurement. The nitrogen sorption isotherms and the corresponding pore size curves of GO/PPy and N-GP/CPN are compared in Fig. 3a and b, respectively. The N-GP/CPN exhibits a typical type IV isotherm with a hysteresis loop in the relative pressure range p/p₀ of 0.4–1.0, suggesting the presence of mesoporous structure. Accordingly, the specific surface area of N-GP/CPN is calculated as high as 1178.4 m² g⁻¹, which is much higher than that of GO/PPy (183.5 m² g⁻¹). And the pore size distribution of N-GP/CPN displays a large total pore volume of 1.523 cm³ g⁻¹ (p/p₀ = 0.9705), which can hosts as much as 75.9 wt% sulfur (ρ_s = 2.07 g cm⁻³) according to the theoretical analysis. Compared to the low surface area and pore volume (Fig. 3b, inset) of 0.274 cm³ g⁻¹ (p/p₀ = 0.9712) for GO/PPy, it is undoubted that the chemical activation with KOH is not merely digesting the carbon to produce more pores but also dramatically restructuring it, thus yielding continuous 3D interconnected porous N-GP/CPN materials with high BET surface area and pore volume. Notably, the large surface area and porous structure is beneficial for improving the high load of sulfur, confining sulfur and polysulfides effectively, and providing channels for the ions diffusion/transportation.

Fig. 3 (a) Nitrogen adsorption–desorption isotherms and (b) pore size distribution of GO/PPy and N-GP/CPN composites.

Since both sulfur and pyrolytic carbon are polar materials, it is easy for N-GP/CPN frameworks to absorb sulfur molecules within these porous channels. Then the N-GP/CPN-S composite with high content of sulfur is synthesized. The GO/PPy-S composite with same ratio is also obtained for comparison. The exact sulfur mass loadings are determined by TGA. As shown in Fig. 4, the weight loss before 300 °C is considered to be the evaporation of sulfur. The sulfur mass loadings of N-GP/CPN-S and GO/PPy-S composites could calculate as approximately 76 wt% and 72 wt%, respectively.

Fig. 4 TGA curves of the sulfur, as-prepared GO/PPy-S and N-GP/CPN-S composites.

SEM and TEM are performed to observe the morphologies of N-GP/CPN-S and GO/PPy-S composites (Fig. 5). The architectures of N-GP/CPN-S stacked by well-exfoliated thin laminar graphene sheets are clearly observed in the SEM image (Fig. 5a), and no agglomerated sulfur particles are found on the surface of graphene nanosheets. Meanwhile, large sulfur particles and aggregated sheets in GO/PPy-S composite are obviously observed in Fig. 5b. The sulfur particles can be easily discerned in the EDS elemental mapping of GO/PPy-S (Fig. S1†). As shown in Fig. 5c, EDS elemental mapping from the N-GP/CPN-S composite suggests that C, N and S are homogeneously spread among the whole hybrid samples. Both C and S maps are found to have similar intensity across the composite, indicating that sulfur is homogeneously impregnated into the small pores of the N-GP/CPN. TEM images also confirm the highly homogeneous dispersion of nanoscopic sulfur on the surface of the N-GP/CPN-S composite without any agglomerated or bulk sulfur (Fig. 5d and e). XRD patterns of N-GP/CPN, N-GP/CPN-S, and GO/PPy-S composites are shown in Fig. 6. The broad peak at around 24° of N-GP/CPN assigns to the amorphous nature for the carbon. The pattern of GO/PPy-S exhibits sharp diffraction peaks from elemental sulfur (JCPDS: 08-0247). Meanwhile, the peak intensity of sulfur in the N-GP/CPN-S composite is largely reduced, and indexed to another orthorhombic crystalline structure (JCPDS: 42-1278), implying the even re-dispersion of sulfur within the porous nanostructures, which is consistent with the SEM and TEM analyses.

Fig. 5 SEM images of (a) N-GP/CPN-S and (b) GO/PPy-S composites, (c) SEM image of N-GP/CPN-S composite and corresponding EDS elemental mapping images, (d and e) TEM images of N-GP/CPN composite.

Fig. 6 XRD patterns of the N-GP/CPN, N-GP/CPN-S and GO/PPy-S composites.

To evaluate the advantageous electrochemical lithium storage capabilities, systematical electrochemical measurements are made by using N-GP/CPN-S as the cathode active material within CR2025 coin cells. Bare GO/PPy-S composite is also assembled as cathode active material in coin cells for comparison. CV is firstly employed to reveal the electrochemical reaction mechanism and kinetic process of the N-GP/CPN-S and GO/PPy-S electrodes measured between 1.8 and 2.7 V at a scan rate of 0.05 mV s⁻¹, as shown in Fig. 7a. There are two cathodic peaks in the first scan of N-GP/CPN-S electrode due to the multiple reductions of sulfur.³⁵ The peak at 2.36 V is associated with the reduction of sulfur (S₈) to higher-order polysulfides (Li₂S_x, 4 ≤ x ≤ 8), while the intense peak at 2.03 V is related to further reduction of these soluble polysulfides to insoluble lower-order Li₂S₂/Li₂S.²¹ Two oxidation reaction peaks are observed in the subsequent anodic scan. The first peak at 2.30 V is associated with the formation of Li₂S_n (2 ≤ n ≤ 8), and the second peak at 2.41 V corresponds to the formation of elemental sulfur.^36,37 All the corresponding peaks are observed in the CV curve of GO/PPy-S electrode, but with low intensities and delayed reactions (larger polarization), which reveal a higher kinetic barrier for the direct reduction in charge–discharge processes.³⁷

Fig. 7 (a) CVs of the GO/PPy-S and N-GP/CPN-S electrodes at 0.1 mV s⁻¹ in a potential window from 1.8–2.7 V. (b) First charge/discharge profiles of GO/PPy-S and N-GP/CPN-S electrodes at 0.2C. (c) Cycling capability and coulombic efficiency of the GO/PPy-S and N-GP/CPN-S electrodes at 0.2C for 100 cycles. (d) Rate performance at various C-rates of patterns of the GO/PPy-S and N-GP/CPN-S electrodes N-GP/CPN, N-GP/CPN-S and GO/PPy-S composites. (e) Long-term cycling stability of N-GP/CPN-S electrode at 0.5C for 500 cycles.

Fig. 7b shows the typical charge–discharge voltage profiles of the N-GP/CPN-S and GO/PPy-S electrodes at 0.2C in the voltage range of 1.8–2.7 V. Consistent with the CV analysis, they display a two-plateau behavior corresponding to the formation of higher-order polysulfides at about 2.38 V and lower-order Li₂S₂/Li₂S at about 2.05 V. The initial discharge and charge capacities of GO/PPy-S electrode are 1131 and 1175 mA h g⁻¹, respectively, with a little overcharge. On the contrary, a high initial discharge capacity of 1205 mA h g⁻¹ and a reversible charge capacity of 1210 mA h g⁻¹ are delivered in N-GP/CPN-S electrode, revealing the facile electronic/ionic transport and improved reaction kinetics enabled by the N-doping porous 3D framework including the 1D carbonized PPy nanotubes and 2D reduced graphene sheets. As shown in Fig. 7c, in the subsequent charge/discharge cycles, the N-GP/CPN-S electrode shows excellent stability with a reversible capacity of 926 mA h g⁻¹ after 100 cycles, as well as exhibits a coulombic efficiency almost 100% for each cycle. In contrast, the GO/PPy-S electrode suffers from steady capacity fading, showing a discharge capacity of 532 mA h g⁻¹ by the 100th cycle, which is due to the rapid dissolution of polysulfide into the electrolyte.

Furthermore, N-GP/CPN-S electrode also shows greatly improved rate performance with continuously increased current rates (Fig. 7d). At high current rates of 0.5, 1, 2, and 4C, the electrode delivers the discharge capacities of 981, 887, 778, and 687 mA h g⁻¹, respectively. More importantly, a stable high capacity of 814 mA h g⁻¹ can still be returned after abruptly switching the current rate back to 1C, indicating outstanding stability and robustness of the electrode. For the GO/PPy-S electrode, it exhibits much lower capacities of 701, 563, 406, and 283 mA h g⁻¹ at the current rates of 0.5, 1, 2, and 4C. Long-term cycling test of the N-GP/CPN-S electrode is carried out at a current rate of 0.5C (Fig. 7e). It exhibits an initial discharge capacity of 1128 mA h g⁻¹ and a notable cycling stability with high stable capacity of 726 mA h g⁻¹ as long as 500 cycles, ultraslow decay rate of 0.07% and per cycle good coulombic efficiency of almost 100%. Therefore, the improved rate performance and cycling stability with high sulfur loading are ascribed to the accommodation of the volume changes provided by the electrode material designs and physical trapping of polysulfide with the interconnected porous nanolayers and nanotubes. In addition, the trapped polysulfides can further efficiently immobilized within the porous structure by chemical adsorption between the N-containing functionalities, such as pyridinic-N, pyrrolic-N, and the discharged polar lithium polysulfide intermediates.

The EIS results of the N-GP/CPN-S and GO/PPy-S electrodes before cycling and after 100 cycles at 0.2C are shown in Fig. 8. The spectra are composed of a semicircle in the high frequency region and a straight line in the low frequency region, which are corresponded to the interfacial charge-transfer resistance (R_ct) and Warburg resistance.^38,39 It is seen that N-GP/CPN-S electrode has a much smaller R_ct than that of GO/PPy-S electrode, indicating the enhanced conductivity of N-GP/CPN-S. After cycling for 100 cycles, both of R_ct decreases remarkably when compared with that before cycling, derived from the adequate infiltration of electrolyte, as well as the dissolution and re-distribution of the active materials.⁴⁰ The charge-transfer resistance of the GO/PPy-S electrode is still larger than that of the N-GP/CPN-S electrode after the 100th discharge, due to the irregular agglomeration of insulating Li₂S on the surface of electrode. However, the N-GP/CPN with high surface and abundant nanopores can stably trap lithium polysulfides during the cycling process, thereby restraining the redistribution and agglomeration of Li₂S. In addition, the higher conductivity, inter-connected structures and nitrogen doping are also benefit for the enhancement of the electron and ion transfer.

Fig. 8 Nyquist plots of the N-GP/CPN-S and GO/PPy-S electrodes (a) before cycling and (b) after 100 cycles.

Conclusions

In summary, in situ polymerization and chemical activation synthesis provide a facile path for the fabrication of continuous 3D carbon structure with nitrogen doping, large surface area and porous structure. The conductivity of N-GP/CPN is enhanced due to the presence of highly conductive 1D carbonized PPy nanotubes and deduced graphene oxide sheets. Moreover, the nitrogen doped and porous structure provide abundant electrochemical nanoreactors, which could accessibly rapid diffusion of electrolyte and effectively accommodate and adsorb sulfur and polysulfides, resulting in superior electrochemical performance of N-GP/CPN-S even with high sulfur loading of 76 wt%. The synergy effects of such design of cathode material lead to high initial capacity (1206 mA h g⁻¹ at 0.2C), good cycle stability (726 mA h g⁻¹ at 0.5C after 500 cycles) and high rate stable capacity (687 mA h g⁻¹ at 4C). These results suggest that N-GP/CPN-S is very promising for superior-performance in energy storage devices.

Acknowledgements

This research is financially supported by the National Nature Science Foundation of China (Grant No. 21221061).

Notes and references

1. A. Manthiram, Y. Fu, S. H. Chung, C. Zu and Y. S. Su, Chem. Rev., 2014, 114, 11751–11787.
2. A. Rosenman, E. Markevich, G. Salitra, D. Aurbach, A. Garsuch and F. F. Chesneau, Adv. Energy Mater., 2015, 5, 1500212.
3. Z. Zhang, Z. Li, F. Hao, X. Wang, Q. Li, Y. Qi, R. Fan and L. Yin, Adv. Funct. Mater., 2014, 24, 2500–2509.
4. M. K. Song, Y. Zhang and E. J. Cairns, Nano Lett., 2013, 13, 5891–5899.
5. Y. H. Yin, C. Ma, Z. X. Cao, Z. X. Sun, Y. J. Jia and S. T. Yang, RSC Adv., 2014, 4, 28871–28874.
6. K. Xi, P. R. Kidambi, R. Chen, C. Gao, X. Peng, C. Ducati, S. Hofmann and R. V. Kumar, Nanoscale, 2014, 6, 5746–5753.
7. X. Liang, C. Hart, Q. Pang, A. Garsuch, T. Weiss and L. F. Nazar, Nat. Commun., 2015, 6, 5682.
8. Z. Zhang, Z. Li, F. Hao, X. Wang, Q. Li, Y. Qi, R. Fan and L. Yin, Adv. Funct. Mater., 2014, 24, 2500–2509.
9. X. Yang, W. Zhu, G. Cao and X. Zhao, RSC Adv., 2015, 5, 93926–93936.
10. Z. Xiao, Z. Yang, H. Nie, Y. Lu, K. Yang and S. Huang, J. Mater. Chem. A, 2014, 2, 8683–8689.
11. H. Xu, Y. Deng, Z. Shi, Y. Qian, Y. Meng and G. Chen, J. Mater. Chem. A, 2013, 1, 15142–15149.
12. G. Ma, Z. Wen, Q. Wang, C. Shen, P. Peng, J. Jin and X. Wu, J. Power Sources, 2015, 273, 511–516.
13. G. Ma, Z. Wen, J. Jin, Y. Lu, K. Rui, X. Wu, M. Wu and J. Zhang, J. Power Sources, 2014, 254, 353–359.
14. J. E. Hyun, P.-C. Lee and I. Tatsumi, Electrochim. Acta, 2015, 176, 887–892.
15. G. Ma, Z. Wen, J. Jin, Y. Lu, X. Wu, M. Wu and C. Chen, J. Mater. Chem. A, 2014, 2, 10350–10354.
16. Z. Z. Li, Y. Pan, J. J. Cheng, J. A. Pan, L. Li, C. Wu, P. Gao and Z. S. Ma, J. Polym. Mater., 2015, 32, 237–246.
17. X. Liang, Y. Liu, Z. Wen, L. Huang, X. Wang and H. Zhang, J. Power Sources, 2011, 196, 6951–6955.
18. G. Ma, Z. Wen, J. Jin, Y. Lu, X. Wu, C. Liu and C. Chen, RSC Adv., 2014, 4, 21612–21618.
19. X. Liang, Z. Wen, Y. Liu, H. Zhang, J. Jin, M. Wu and X. Wu, J. Power Sources, 2012, 206, 409–413.
20. X. Yu, J. Zhao, R. Lv, Q. Liang, Y. Bai, Z. H. Huang, W. Shen and F. Kang, RSC Adv., 2015, 5, 75403–75410.
21. F. F. Zhang, G. Huang, X. X. Wang, Y. L. Qin, X. C. Du, D. M. Yin, F. Liang and L. M. Wang, Chem.–Eur. J., 2014, 20, 17523–17529.
22. W. Chen, Z. Zhang, Q. Li, Y. Lai and J. Li, ChemElectroChem, 2015, 2, 246–252.
23. F. F. Zhang, X. B. Zhang, Y. H. Dong and L. M. Wang, J. Mater. Chem., 2012, 22, 11452–11454.
24. J. He, Y. Chen, P. Li, F. Fu, Z. Wang and W. Zhang, J. Mater. Chem. A, 2015, 3, 18605–18610.
25. R. Chen, T. Zhao and F. Wu, Chem. Commun., 2015, 51, 18–33.
26. X. J. Li, W. Xing, J. Zhou, G. Q. Wang, S. P. Zhuo, Z. F. Yan, Q. Z. Xue and S. Z. Qiao, Chem.–Eur. J., 2014, 20, 13314–13320.
27. G. Zhou, Y. Zhao and A. Manthiram, Adv. Energy Mater., 2015, 5, 1402263.
28. Z. Li and L. Yin, ACS Appl. Mater. Interfaces, 2015, 7, 4029–4038.
29. H.-G. Wang, Y. Wang, Y. Li, Y. Wan and Q. Duan, Carbon, 2015, 82, 116–123.
30. Z. Zhang, G. Chen, H. Wang and W. Zhai, J. Mater. Chem. C, 2015, 3, 1649–1654.
31. H. Li, M. Sun, T. Zhang, Y. Fang and G. Wang, J. Mater. Chem. A, 2014, 2, 18345–18352.
32. Y. Wu, C. Guo, N. Li, L. Ji, Y. Li, Y. Tu and X. Yang, Electrochim. Acta, 2014, 146, 386–394.
33. J. Cao, Y. Wang, J. Chen, X. Li, F. C. Walsh, J.-H. Ouyang, D. Jia and Y. Zhou, J. Mater. Chem. A, 2015, 3, 14445–14457.
34. Z. Yang, Y. Dai, S. Wang, H. Cheng and J. Yu, RSC Adv., 2015, 5, 78017–78025.
35. R. Xu, J. Lu and K. Amine, Adv. Energy Mater., 2015, 5, 1500408.
36. Z. Yuan, H.-J. Peng, J.-Q. Huang, X.-Y. Liu, D.-W. Wang, X.-B. Cheng and Q. Zhang, Adv. Funct. Mater., 2014, 24, 6105–6112.
37. L. Qie and A. Manthiram, Adv. Mater., 2015, 27, 1694–1700.
38. B. Li, S. Li, J. Liu and J. Xu, RSC Adv., 2015, 5, 40310–40315.
39. G. Xu, B. Ding, P. Nie, L. Shen, J. Wang and X. Zhang, Chem.–Eur. J., 2013, 19, 12306–12312.
40. X. Gu, C. Tong, C. Lai, J. Qiu, X. Huang, W. Yang, B. Wen, L. Liu, Y. Hou and S. Zhang, J. Mater. Chem. A, 2015, 3, 16670–16678.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra02667b

---