Phosphorus-doped graphene-wrapped molybdenum disulfide hollow spheres as anode material for lithium-ion batteries

Abstract

Molybdenum disulfide is a promising candidate for the anode material in lithium-ion batteries due to its high theoretical specific capacity. However, volume change during cycling causes poor cycling stability and rate performance, and improving cycle life is a major research challenge. Herein, we present a strategy by fabricating phosphorus-doped graphene-wrapped molybdenum disulfide hollow spheres to improve the electrochemical performance of molybdenum disulfide in lithium storage. It is demonstrated that the obtained composite shows excellent electrochemical performance as anode material for lithium-ion batteries. The composite exhibits the high specific capacity of ∼1097 mA h g⁻¹ at a current of 100 mA g⁻¹, as well as excellent cycling stability and high-rate capability. The unique hollow structure combined with phosphorus-doped graphene wrapping definitely contributes to this versatile electrochemical performance.

---

1. Introduction

Molybdenum disulfide (MoS₂), another important transition metal sulfide with the same crystal structure as graphite,¹ has been widely studied as an electrode material for supercapacitors² and lithum-ion batteries (LIBs)³ due to its high theoretical reversible capacity of 670 mA h g⁻¹, which much higher than that of the commercial graphitic carbon anodes (372 mA h g⁻¹). However, the rapid capacity fading and poor rate performance significantly limit its practical applications. This problem mainly originates from the large volume change of electrode material accompanying Li⁺ insertion and extraction, which creates large internal stress, leading to the disintegration and pulverization of the electrode material. To alleviate the problem, one effective strategy is to design hollow or porous structures,⁴ where the void space provided in this type of material could buffer the volume change, leading to improved cycling retention and rate properties.⁵

In addition to hollow or porous materials, MoS₂–C composite is another type of unique structure that might be utilized to achieve the same goal. The idea of creating such a structure not only cushions the internal stress induced during the volume change, but also makes the composite more conductive.⁶ Graphene nanosheets (GNS), a monolayer of carbon atoms with a tight packing of honeycomb lattice, has become one of the rising stars in material science.⁷ Its intriguing properties, like high electrical conductivity and mechanical flexibility,⁸ make it very attractive for LIBs^9,10 and supercapacitors.^11,12 In our previous work, we reported a studies on MoS₂ with amorphous carbon, which showed outstanding electrochemical performance as electrode material.¹³ So far, practical and theoretical investigations have explored that doping of heteroatoms (S, B, P, N, etc.) in the GNS could further boost its conductivity and electrochemical performance.^14,15 The reason for the electrochemical performance boost is provoked by doped atoms with different electronegativity (e.g., N: 3.04, B: 2.04, S: 2.58) that can break the electroneutrality of GNS to create the charged sites which improve the discharge and charge capacity of GNS in LIBs.^16,17 Since phosphorus has a lower electronegativity than carbon, higher electron-donating ability,¹⁸ it is of great significance to explore the unique properties of phosphorus-doped graphene (PG), which is rarely reported.

Compared to nitrogen- and boron-doped graphene, synthesis of PG for LIBs still remains a challenge because of the relatively large atomic radius of phosphorus; a general and effective synthetic strategy to produce pure PG-wrapped transition metal sulfide is thus highly desirable. Herein, we demonstrate the fabrication of PG-wrapped MoS₂ hollow spheres (PG–MoS₂) utilizing a simple hydrothermal route following the annealing. The PG coating facilitates the electronic transportation to reach the hollow spheres, and serves as a flexible structure strengthening agent for the hollow spheres, while the hollow sphere acts as high-capacity lithium storage host and provide hollow interior space to buffer their large volume variation during lithium uptake/release processes. The subsequent electrochemical measurement showed that the PG–MoS₂ sample exhibits significantly enhanced lithium storage properties in terms of higher specific capacities and better cyclic capacity retention compared to the MoS₂ hollow spheres and graphene-wrapped MoS₂ hollow spheres (G-MoS₂), evidently proving the positive effect of PG enwrap metal sulfide structures.

2. Experimental details

2.1 Synthesis of electrode material

2.1.1 Materials. Molybdenum trioxide (MoO₃, 99.9% purity) powder, sodium fluoride (NaF, 98.5% purity), potassium thiocyanate (KSCN, 98.5% purity), and (3-aminopropyl) triethoxysilane (APTES, Gelest, 98%) were purchased from Tianjin Fuchen Chemical Reagents Factory, China and used without further purification.

2.1.2 Synthesis of MoS₂ hollow spheres. The MoS₂ hollow spheres were synthesized by a one-step hydrothermal reaction using molybdenum trioxide, sodium fluoride, and potassium thiocyanate as starting materials. In a typical synthesis, 0.87 g of molybdenum trioxide, 0.52 g of sodium fluoride, and 1.93 g of potassium thiocyanate were dissolved in 30 mL deionized water. After stirring for 30 minutes, the obtained clear solution was transferred into a Teflon-lined stainless steel autoclave and sealed tightly, heated at 220 °C for 24 h and then cooled to room temperature naturally. The black precipitate was collected by centrifugation, washed thoroughly with deionized water, ethanol for six times, and dried at 60 °C in a vacuum oven for 24 h.

2.1.3 Synthesis of PG–MoS₂. 0.5 g of the as-prepared MoS₂ was dispersed in 50 mL ethanol by ultrasonication. Then, 0.5 mL (3-aminopropyl) triethoxysilane (APTES) and 150 mg of triphenylphosphine (TPP) were added to the solution under magnetic stirring. The reaction was kept at room temperature for 4 h before the product was collected and washed with ethanol by centrifugation. Then, the obtained sample was dispersed in 50 mL H₂O, followed by addition of 30 mL graphene oxide (GO) solution with concentration of 0.4 mg mL⁻¹. After stirring for 4 h, the mixture was collected by centrifugation, washed with deionized water, ethanol, and dried in a vacuum oven at 60 °C for 12 h. The dried product was further heated at 900 °C for 2 h under high purity argon at a ramping rate of 5 °C min⁻¹. As contrast, the G-MoS₂ was also synthesized according to the above process without adding TPP and annealed under the same conditions.

2.2 Material characterization

The samples were characterized with X-ray diffraction (XRD, Thermo X'TRA X-ray diffractometer with Cu Kα source), field emission scanning electron nanoscopy (FESEM, JSM-6330F), transmission electron nanoscopy (TEM, FEI Tecnai G2 Spirit), Atomic force microscopy (AFM, SPM-9500J3), thermogravimetric analysis (TGA, TG 209 F3 Tarsus), Laser Raman spectroscopy (Renishaw inVia), X-ray photoelectron spectroscopy (XPS, ESCALAB 250), and the surface area of the product was calculated from nitrogen adsorption/desorption isotherms at 77 K that were conducted on an ASAP 2020 V3.03 H instrument.

2.3 Electrochemical test

The electrochemical characterization was performed using 2032 coin-type cells. The working electrode was prepared by a slurry coating procedure. The slurry consisted of 80 wt% active material, 10 wt% acetylene black and 10 wt% polyvinylidene fluoride (PVDF) dissolved in N-methyl-2-pyrrolidinone, the obtained slurry was spread on a pure Cu foil which acted as current collector. The coated electrode was dried at 120 °C for 12 h in vacuum and then pressed to enhance the particle contact. The test cell was assembled in an argon-filled glove box, using a lithium sheet as the counter and reference electrode, a polypropylene film (Celgard-2400) as the separator, and 1.0 M LiPF₆ solution in ethylene carbonate (EC) and dimethyl carbonate (DMC) (1 : 1 in volume) as the electrolyte.

Galvanostatic charge–discharge measurement of the cell was carried out on Neware battery tester between 0.01 ∼ 3.00 V at various current densities. Cyclic voltammetry of the cell was conducted using an electrochemical workstation (Zahner IM6ex) in the potential range of 0.01–3.00 V vs. Li/Li⁺ at a scan rate of 0.5 mV s⁻¹. Electrochemical impedance spectral (EIS) measurements were performed on PARSTAT 2273 advanced electrochemical system by applying a sine wave with the excitation amplitude of 5.0 mV in the frequency range from 100 kHz to 0.01 Hz.

3. Results and discussion

The PG is synthesized through high temperature annealing a mixture of GO and TPP, which are employed as the carbon (C) and phosphorus (P) sources, respectively. As shown in Scheme 1, the as-obtained mixture of GO and TPP is annealed at 900 °C under high purity argon atmosphere to give phosphorus dope graphene nanosheets. During the thermal annealing process, the TPP breaks down into single P atoms, and then a subsequent reaction of the P atoms with graphene. The doped P atoms (P–C or P–O) are usually react with oxygen containing groups on the plan and edge of the GO sheets to join into the carbon framework,¹⁹ resulting in simultaneous phosphorus doping and reduction of graphene oxide. P–C bonds can be easier formed than P–O bonds at higher annealing temperature. It is attributed that P–O bonds are not thermally stable at higher annealing temperature, which is similar with previous report for the breakage of N–O bonds in N-doped graphene caused by overheating.^20,21 The substitution of C atoms by P atoms in the graphene sheets is confirmed by the binding energies for the P2p peak in XPS that coincide with those previously reported.^22,23

Scheme 1 Schematic illustration of synthesis of PG through high temperature annealing of the GO and TPP mixture.

Fig. 1 shows the morphology of the as-prepared MoS₂ hollow spheres and PG–MoS₂. It is clear from the Fig. 1a that the sample contains uniform hollow spheres, which are approximately ∼500 nm in diameter. The PG can be easily identified as the thin sheets that wrap around these particles and also bridge nearby particles together, which is shown in Fig. 1b. The hollow interior of these particles is clearly revealed from Fig. 1c, in which each particle presents a significant contrast between the edge and centre. Moreover, folding of the PG can be observed between connected particles, which proves that PG indeed wrap around the MoS₂ hollow spheres. Fig. 1d shows a typical high-magnification TEM image of a single hollow sphere. The highly wrinkled surface on the shell indicates that the hollow sphere is probably composed of MoS₂ nanosheets. They are folded, curved and assembled into the hierarchical hollow sphere. The hierarchical surface and hollow structure would offer a huge specific surface area for electrochemical reactions, which is shown in Fig. 2. The nitrogen adsorption and desorption isotherms of these samples coincide with a type IV isotherm. The derived specific surface area for the PG–MoS₂ (245 m² g⁻¹) is superior to the MoS₂ hollow spheres (143 m² g⁻¹), and G-MoS₂ (213 m² g⁻¹). The AFM image is collected at the edge of the PG, as shown in Fig. 1e. The thickness of the PG coating on MoS₂ particles measure by AFM is about 0.96 nm.

Fig. 1 SEM images of MoS₂ hollow spheres (a) and PG–MoS₂ (b). TEM images of PG–MoS₂ (c) and single PG–MoS₂ (d). AFM images of PG–MoS₂ (e).

Fig. 2 Nitrogen adsorption and desorption isotherms of the MoS₂ hollow spheres, G-MoS₂, and PG–MoS₂.

The chemical compositions of these samples are studied by XRD, as shown in Fig. 3a. A single diffraction peak is observed in pattern I at ∼11°, indicating that GO has been successfully obtained via the chemical oxidation of graphite powder. This characteristic GO peak completely disappears after the annealing process (pattern II in Fig. 3a), accompany by the appearance of a new peak at ∼25.53° which is due to the (002) diffraction of PG. This proves that the GO has been fully transformed to PG using the current annealing process.²⁴ For PG–MoS₂ (pattern III), G-MoS₂ (pattern IV), and MoS₂ hollow spheres (pattern V), all the identified peaks can be assigned to the standard hexagonal 2H–MoS₂ structure (JPCDS 37-1492), underlining the phase purity of the as-prepared product. The intensity of (002) peaks of PG–MoS₂ and G-MoS₂ are relatively lower compared with MoS₂ hollow spheres, suggesting the incorporation of PG or GNS could inhibit the (002) plane growth of MoS₂ crystals in the composite. The absences of the PG peak and GNS peak show that they uniformly disperse and wrap around the MoS₂ hollow spheres. TGA is employed to determine the amount of PG present in the PG–MoS₂ (Fig. 3b). MoS₂ hollow spheres start to lose weight at approximately 425 °C due to the oxidation of MoS₂ to MoO₃. The PG–MoS₂ exhibits two weight losses. The first one appears at approximately 250 °C, which can probably be attributed to the removal of oxygen-containing groups. The second is a large continuous weight loss in the range of approximately 373–580 °C. This thermal behavior might be caused by the decomposition of the PG, and oxidation of MoS₂ in the composite. The weight fraction of the PG in the composite is measured to be ∼11%.

Fig. 3 (a) XRD patterns of GO (I), PG (II), PG–MoS₂ (III), G-MoS₂ (IV), and MoS₂ hollow spheres (V). (b) TGA of MoS₂ hollow spheres (I), and PG–MoS₂ (II).

As shown in Fig. 4, it can be known that the D band, observed at approximately 1323 cm⁻¹, is disorder induced. It is attributed to structural defect sites on the GNS plane. The G band observed at approximately 1587 cm⁻¹, is commonly observed for all graphitic structures and attributed to the E_2g vibrational mode present in the sp² bonded graphitic carbons. The intensity ratio of the D peak to G peak (I_D/I_G), provides the gauge for the amount of structural defects. It is obvious that PG has a relatively increased I_D/I_G of 1.52 relative to 1.15 in GNS because of the incorporated phosphorus atoms,^25,26 which allows MoS₂ particles to be more effectively and uniformly dispersed on its surface.

Fig. 4 Raman spectra of PG and GNS.

As shown in Fig. 5a, the existence of O, C, P, Mo and S elements assure the purity of products. Additionally, a small quantity of O comes from a few parts of PG that are not completely reduced during the annealing process. We can see the P2p characteristic peak (132 eV) exists at the XPS survey spectrum of PG. The amount of phosphorus doped into the PG is 1.86 at.% by XPS elemental analysis, as shown in Table 1. The XPS survey spectrum of GNS (Fig. 5b) are similar with PG in all other ways except that there is no phosphorus in GNS. Furthermore, high-resolution of P2p spectrum presented in Fig. 5d also contains the P–C (131.5 eV) and P–O (132.9 eV) bonding.²⁷ This result strongly supports that phosphorus atoms are successfully incorporated into the GNS structure through a thermal annealing process.

Fig. 5 XPS of the PG–MoS₂ (a), GNS (b), PG (c), and (d) high-resolution of P2p spectrum.

Table 1 Atom contents of PG and GNS calculated from XPS analysis

Samples	C (at.%)	O (at.%)	P (at.%)
GNS	97.01	2.91	0
P-GNS	95.05	3.09	1.86

---

Fig. 6 shows the representative cyclic voltammograms (CV) of these samples, and these CV behaviors are generally consistent with that reported previously.^28,29 Two cathodic peaks at about 0.96 V and 0.48 V in the first cycle can be assigned to the phase transformation from 2H to 1T of Li_xMoS₂ and the reduction of Mo⁴⁺ to Mo nanoparticles,³⁰ respectively. In the second cycle, a new reduction peak at about 1.86 V appears, suggesting the generation of the gel-like polymeric layer.³¹ During the first charge process, small and large board peaks at 1.73 V and 2.33 V can be ascribed to the conversion reaction of Mo and Li₂S to MoS₂.³² The reversible reaction based on the equation: MoS₂ + 4Li ↔ Mo + 2Li₂S, is accompanied by the redox of Mo nanoparticles and the reversible growth of the gel-like layer.³³ In comparison with MoS₂ hollow spheres, the redox peaks for PG–MoS₂ and G-MoS₂ are not decrease during the cycling, indicating PG and GNS can stabilize the electrode structure.

Fig. 6 Representative CVs of the PG–MoS₂ (a), G-MoS₂ (b), and MoS₂ hollow spheres (c) at a scan rate of 0.5 mV s⁻¹ between 1 mV and 3.0 V.

Fig. 7 shows the charge–discharge voltage profiles of these samples for the 1st, 2nd, 5th, 10th, 50th, and 100th cycles. As shown in Fig. 7a, in agreement with the above CV behavior, two potential plateaus at ∼1.15 and 0.6 V can be identified in the discharge process of the first cycle, and it gives a very high discharge capacity of 1281 mA h g⁻¹. Note that the specific capacity is based on the total mass of the PG–MoS₂ in the context. In the following discharge processes, two potential plateaus at 2.1 and 1.3 V are observed, and the plateau at 0.6 V disappears, which may correspond to the redox reaction of Li₂S/S in a deeply discharged MoS₂/Li cell according to Wang's reports.³⁴ During the charging process, two potential plateaus at 1.7 and 2.2 V are visible, in agreement with the reported lithiation and delithiation profiles of MoS₂.^35,36 Fig. 7b and c shows that the initial discharge capacities of the G-MoS₂ and MoS₂ hollow spheres electrode are 1147 and 1056 mA h g⁻¹. Among these samples, PG–MoS₂ exhibits the highest charge and discharge capacities. The irreversible capacity loss in the 1st cycle can be mainly attributed to the irreversible processes including the electrolyte decomposition and inevitable formation of the solid-electrolyte interface (SEI) layer.³⁷

Fig. 7 Charge–discharge voltage profiles of the PG–MoS₂ (a), G-MoS₂ (b), and MoS₂ hollow spheres (c) for the 1st, 2nd, 5th, 10th, 50th and 100th cycles at a current density of 100 mA g⁻¹.

As shown in Fig. 8a, the initial charge and discharge capacities of the PG–MoS₂ are 1097 and 1281 mA h g⁻¹, respectively, corresponding to a Coulombic efficiency (CE) of 85.6%. The CE increases to larger than 99% after the 4th cycle. After 100 charge–discharge cycles, a high reversible charge capacity of 1072 mA h g⁻¹ can be delivered by the PG–MoS₂. For comparison, we test MoS₂ hollow spheres, and G-MoS₂ as anode materials, both of which exhibit higher capacity fade than that of PG–MoS₂. After 100 cycles, the charge capacities of MoS₂ hollow spheres and G-MoS₂ are 770 and 889 mA h g⁻¹, respectively. Interestingly, the discharge capacities of the PG–MoS₂ increases during the cycle and can reach 1110 mA h g⁻¹ after 45 cycles. This result is exemplary compared to other previous reports,^38,39 which may be caused by defect sites and vacancies in PG. It is thus evident that the as prepared PG–MoS₂ exhibits higher storage capability and enhanced cyclic capacity retention compared to the MoS₂ hollow spheres and G-MoS₂, and could serve as promising anode material for high-performance LIBs.

Fig. 8 Cycling performance (a) and rate capability (b) of MoS₂ hollow spheres (I), G-MoS₂ (II), and the PG–MoS₂ (III).

The PG–MoS₂ also demonstrates superior rate capabilities. Fig. 8b shows the rate cycling behavior of these composites. When the current densities increase from 100 to 200, 300, 500, and 1000 mA g⁻¹, the specific capacities of the PG–MoS₂ electrode change from 1087 to 1060, 1030, 1002, and 969 mA h g⁻¹, respectively. It can be seen that the specific capacity of the PG–MoS₂ is 969 mA h g⁻¹, much higher than that of graphite, even at a current density of 1000 mA g⁻¹. Moreover, after the high rate cycling, the discharge capacity of the PG–MoS₂ can almost completely recover the initial capacity at a current density of 100 mA g⁻¹, suggesting an excellent rate-cycling stability for the anodes. Although the MoS₂ hollow spheres and G-MoS₂ also exhibit a high rate performance, their capacities are significantly lower than the PG–MoS₂. Even at a current density of 1000 mA g⁻¹, the specific capacity of the PG–MoS₂ is higher than the specific capacity of MoS₂ hollow spheres and G-MoS₂ at a current density of 100 mA g⁻¹.

It is well known that volume expansion accompanies Li-insertion for electrode materials, so the structure integrity is important to ensure the cycleability of as-prepared products. In order to explore the alteration in morphology of the electrode materials during reaction with Li, SEM images of the PG–MoS₂, G-MoS₂ and MoS₂ before and after cycling are recorded. As shown in Fig. 9, it is obvious that even after 40 cycles, the PG–MoS₂ and G-MoS₂ remain intact, no aggregation or broken sphere is observed with the help of the flexible PG or G sheets. By contrast, most of the completely MoS₂ hollow spheres collapse and aggregate with each other. These results further demonstrate that the introduction of PG or G sheets can effectively buffer the strain and stress of the volume change and prevent the agglomeration of MoS₂ spheres during cycling, contributing to its excellent cycling stability.

Fig. 9 SEM images of (a) PG–MoS₂, (b) G-MoS₂, (c) MoS₂ before cycle; (d) PG–MoS₂, (e) G-MoS₂, and (f) MoS₂ after 40 cycles at a current density of 100 mA g⁻¹.

The superior Li storage properties of the PG–MoS₂ as compared to that of G-MoS₂ and MoS₂ is supported by the EIS measurements on these samples after 10 discharge–charge cycles, which is shown in Fig. 10a. R_e represents the internal resistance of the test battery, R_f and CPE₁ are associated with the resistance and constant phase element of the SEI film, R_ct and CPE₂ are related with the charge-transfer resistance and constant phase element of the electrode/electrolyte interface, and Z_W is determined by the Warburg impedance corresponding to the lithium-diffusion process. As shown in Fig. 10a, the high-frequency semicircle corresponds to the resistance R_f and CPE₁ of the SEI film, and the semicircle in the medium-frequency region is assigned to the charge-transfer resistance R_ct and CPE₂ of the electrode/electrolyte interface. The inclined line corresponds to the lithium-diffusion process within the bulk of the electrode material. The kinetic differences of samples are further investigated by modeling AC impendence spectra based on the modified equivalent circuit (Fig. 10b).⁴⁰ As seen from Fig. 10a, the PG–MoS₂ shows a smaller radius of semi-circle in the Nyquist plots, suggesting a lower contact and charge-transfer resistances in the PG–MoS₂ composite electrode. According to EIS equivalent circuit in Fig. 10b, R_f and R_ct are 8.69 Ω, and 53.94 Ω, respectively, for the PG–MoS₂, which are less than these of G-MoS₂ (R_f = 9.81 Ω and R_ct = 65.07 Ω) and MoS₂ hollow spheres (R_f = 12.31 Ω and R_ct = 90.98 Ω). This fact confirms that the PG have more effect on reduce contact resistance and charge transfer resistance during the electrochemical lithium insertion/extraction reaction, resulting in significant improvement in the electrochemical performance.

Fig. 10 (a) Nyquist plots of the MoS₂ hollow spheres, G-MoS₂ and PG–MoS₂ by applying a sine wave with amplitude of 5.0 mV over the frequency range from 100 KHz to 0.01 Hz. (b) The equivalent circuit model of the studied system, CPE represent the constant phase element, Z_CPE = {Q(jω)ⁿ}⁻¹, 0 ≤ n≤ 1.

The excellent electrochemical performance of the PG–MoS₂ could be attributed to the following factors. First, the PG wrapping could greatly enhance the electronic conductivity of the composite by bridging adjacent MoS₂ hollow spheres together. Second, the uninterrupted electron supply from PG could improve the electrochemical ability of MoS₂ hollow spheres, resulting in enhanced reversible storage capacity of MoS₂ hollow speres for Li⁺. Third, the flexible PG and hollow interior of the MoS₂ particles could effectively buffer volume variation and store large quantity lithium ions during the charge–discharge process. Last but not less, the synergistic interaction between PG and MoS₂ hollow spheres is believed as main reason for the superior performances of PG–MoS₂ for reversible Li⁺ storage.

4. Conclusions

In summary, a novel composite of PG–MoS₂ is successfully synthesized through facile hydrothermal and annealing method. The MoS₂ hollow speres are encapsulated by flexible PG to overcome the issues of structural change and aggregation of MoS₂ particle during reversible storage process of Li⁺. Moreover, this unique composite is capable to efficiently utilize the high conductivity, large surface area, good mechanical flexibility, and excellent electrochemical performance of PG as well as the large electrode-electrolyte contact area, short diffusion path of Li⁺, and extraordinary stability for MoS₂ hollow speres. As a result, the PG–MoS₂ exhibits exceptional high capacity, superior cyclic stability and high-rate capability, highlighting the advantages of hollow structure of MoS₂ and utilization of PG encapsulation to achieve the maximum performance for LIBs. The same strategy could be applied to other hollow nanomaterials toward robust structures and enhanced lithium storage properties for lithium-ion batteries.

Acknowledgements

We gratefully acknowledge the financial support from the National Natural Science Foundation of China (no. J1103305).

Notes and references

1. X. Huang, Z. Y. Zeng and H. Zhang, Chem. Soc. Rev., 2013, 42, 1934.
2. E. G. da Silveira Firmiano, A. C. Rabelo, C. J. Dalmaschio, A. N. Pinheiro, E. C. Pereira, W. H. Schreiner and E. R. Leite, Adv. Energy Mater., 2014, 4, 1301380.
3. J. Xiao, X. Wang, X. Q. Yang, S. D. Xun, G. Liu, P. K. Koech, J. Liu and J. P. Lemmon, Adv. Funct. Mater., 2011, 21, 2840.
4. J. Liu, G. Z. Cao, Z. G. Yang, D. H. Wang, D. Dubois, X. D. Zhou, G. L. Graff, L. R. Pederson and J. G. Zhang, ChemSusChem, 2008, 1, 676.
5. S. Ding, J. S. Chen, G. Qi, X. Duan, Z. Wang, E. P. Giannelis, L. A. Archer and X. W. Lou, J. Am. Chem. Soc., 2010, 133, 21.
6. X. S. Zhou, Y. X. Yin, L. J. Wan and Y. G. Guo, J. Mater. Chem., 2012, 22, 17456.
7. C. Rao, A. Sood, K. Subrahmanyam and A. Govindaraj, Angew. Chem., Int. Ed., 2009, 48, 7752.
8. H. Chen, M. B. Muller, K. J. Gilmore, G. G. Wallace and D. Li, Adv. Mater., 2008, 20, 3557.
9. H. L. Wang, L. F. Cui, Y. A. Yang, H. S. Casalongue, J. T. Robinson, Y. Y. Liang, Y. Cui and H. J. Dai, J. Am. Chem. Soc., 2010, 132, 13978.
10. S. J. Ding, J. S. Chen, D. Y. Luan, F. Y. C. Boey, S. Madhavi and X. W. Lou, Chem. Commun., 2011, 47, 5780.
11. M. Pumera, Energy Environ. Sci., 2011, 4, 668.
12. H. L. Wang, H. S. Casalongue, Y. Y. Liang and H. J. Dai, J. Am. Chem. Soc., 2010, 132, 7472.
13. W. D. Qiu, J. Xia, S. X. He, H. J. Xu, H. M. Zhong and L. P. Chen, Electrochim. Acta, 2014, 117, 145.
14. S. Some, J. Kim, K. Lee, A. Kulkarni, Y. Yoon, S. Lee, T. Kim and H. Lee, Adv. Mater., 2012, 24, 5481.
15. C. Zhang, N. Mahmood, H. Yin, F. Liu and Y. Hou, Adv. Mater., 2013, 25, 4932.
16. H. Wang, C. Zhang, Z. Liu, L. Wang, P. Han, H. Xu, K. Zhang, S. Dong, J. Yao and G. Cui, J. Mater. Chem., 2011, 21, 5430.
17. L. Dai, D. W. Chang, J. B. Baek and W. Lu, Small, 2012, 8, 1130.
18. D. S. Yang, D. Bhattacharjya, S. Inamdar, J. Park and J. S. Yu, J. Am. Chem. Soc., 2012, 134, 16127.
19. Y. Zhang, T. Mori, J. Ye and M. Antonietti, J. Am. Chem. Soc., 2010, 132, 6294.
20. X. L. Li, H. L. Wang, J. T. Robinson, H. Sanchez, G. Diankov and H. J. Dai, J. Am. Chem. Soc., 2009, 131, 15939.
21. T. V. Khai, H. G. Na, D. S. Kwak, Y. J. Kwon, H. Ham, K. B. Shim and H. W. Kim, J. Mater. Chem., 2012, 221, 7992.
22. Z. W. Liu, F. Peng, H. J. Wang, H. Yu, W. X. Zheng and J. Yang, Angew. Chem., 2011, 123, 3315.
23. M. Latorre-Sánchez, A. Primo and H. García, Angew. Chem., Int. Ed., 2013, 52, 11813.
24. D. Long, W. Li, L. Ling, J. Miyawaki, I. Mochida and S. H. Yoon, Langmuir, 2010, 26, 16096.
25. D. S. Geng, Y. Chen, Y. G. Chen, Y. L. Li, R. Y. Li, X. L. Sun, S. Y. Ye and S. Knights, Energy Environ. Sci., 2011, 4, 760.
26. Y. L. Li, J. J. Wang, X. F. Li, D. S. Geng, R. Y. Li and X. L. Sun, Chem. Commun., 2011, 47, 9438.
27. Z. W. Liu, F. Peng, H. J. Wang, H. Yu, W. X. Zheng and J. Yang, Angew. Chem., Int. Ed., 2011, 50, 3257.
28. H. Liu, D. W. Su, R. F. Zhou, B. Sun, G. X. Wang and S. Z. Qiao, Adv. Energy Mater., 2012, 2, 970.
29. Y. T. Liu, X. D. Zhu, Z. Q. Duan and X. M. Xie, Chem. Commun., 2013, 49, 10305.
30. G. Du, Z. Guo, S. Wang, R. Zeng, Z. Chen and H. Liu, Chem. Commun., 2010, 46, 1106.
31. Q. Wang and J. Li, J. Phys. Chem. C, 2007, 111, 1675.
32. H. Hwang, H. Kim and J. Cho, Nano Lett., 2011, 11, 4826.
33. C. F. Zhang, Z. Y. Wang, Z. P. Guo and X. W. Lou, ACS Appl. Mater. Interfaces, 2012, 4, 3765.
34. X. P. Fang, X. W. Guo, Y. Mao, C. X. Hua, L. Y. Shen, Y. S. Hu, Z. X. Wang, F. Wu and L. Q. Chen, Chem.–Asian J., 2012, 7, 1013.
35. C. Q. Feng, J. Ma, H. Li, R. Zeng, Z. P. Guo and H. K. Liu, Mater. Res. Bull., 2009, 44, 1811.
36. K. Chang and W. X. Chen, ACS Nano, 2011, 5, 4720.
37. W. H. Shi, J. X. Zhu, X. H. Rui, X. H. Cao, C. Chen, H. Zhang, H. H. Hng and Q. Y. Yan, ACS Appl. Mater. Interfaces, 2012, 4, 2999.
38. Z. S. Wu, W. C. Ren, L. Wen, L. B. Gao, J. P. Zhao, Z. P. Chen, G. M. Zhou, F. Li and H. M. Cheng, ACS Nano, 2010, 4, 3187.
39. Y. M. Sun, X. L. Hu, W. Luo and Y. H. Huang, ACS Nano, 2011, 57, 100.
40. S. B. Yang, X. L. Feng, L. J. Zhi, Q. A. Cao, J. Maier and K. Mullen, Adv. Mater., 2010, 22, 838.

---