Synthesis of a hierarchical MoSe₂/C hybrid with enhanced electrochemical performance for supercapacitors

Abstract

A facile one-step hydrothermal strategy was successfully developed to fabricate a 3D hierarchical MoSe₂/C hybrid with triethylene glycol as a structure-directing agent and carbon source. The products were characterized by X-ray diffraction (XRD), scanning electron microscopy (SEM), energy dispersive X-ray spectroscopy (EDS), high-resolution transmission electron microscopy (HRTEM), Raman spectroscopy and nitrogen adsorption–desorption. The experiment results indicate that the hierarchical porous MoSe₂/C hybrid is composed of numerous few-layered MoSe₂ nanosheets and amorphous carbon derived from the decomposition of the triethylene glycol. As a consequence, the obtained hierarchical MoSe₂/C electrode exhibited superior electrochemical supercapacitive performances such as high specific capacitance, excellent cyclic stability and significantly enhanced rate capability in comparison with the bare MoSe₂. The prominent electrochemical performance could be attributed to the robust hierarchical porous architectures with large surface areas and effective integration with conductive amorphous carbon.

---

1. Introduction

Supercapacitors also called electrochemical capacitors have become one of the most promising energy storage devices because they can deliver high levels of electrical power and offer long operating lifetimes.¹ The heart of such an energy storage system are innovative electrodes, which directly determine the capability, delivery rates and efficiency of a supercapacitor, and enable highly efficient, safe, and versatile use of energy.² Two kinds of supercapacitor electrodes are normally classified on the basis of charge storage mechanisms, i.e. electric double layer (EDL) electrodes and pseudocapacitive electrodes. Most EDL electrodes use carbon materials and store energy by physical charge separation at the interfaces between the electrodes and electrolytes, without involving chemical reactions during charging–discharging processes.³ Pseudocapacitive electrodes are based on metal oxides/sulfides and conductive polymers, and use fast and reversible chemical reactions for energy storage.^4,5 EDL electrodes usually have high charge–discharge rates and long cycling performance, while pseudocapacitive electrodes have much larger storage capabilities due to the chemical reactions involved. Additionally, to improve the electric conductivity and specific capacitance of electrodes, hybrid electrodes are usually created with an expectation to take advantage of both types of electrodes for highly efficient energy storage.⁶

Recently, layered transition metal dichalcogenides (LTMDCs) are attracting significant attention and have emerged as one of the most prominent candidates for energy storage because of their 2D sheet-like morphology, higher electrical conductivity than the oxides and high surface area.^7,8 In these materials, metals and chalcogens interact via strong chemical bonds in the molecular layers, whereas the individual layers interact via weak van der Waals forces, forming a graphene-like layered structure.⁹ This structure is beneficial for the insertion and extraction of a variety of ions and can be exploited in the field of energy storage including Li-ion batteries and supercapacitors.¹⁰ As one of the representative members, MoS₂ has received intensive attention on capacitor research.^11–17 It was found that MoS₂ could not only exhibit better capacitive properties in its electrochemical double layers, but also could give rise to a faradaic capacitance due to the intercalation of the ions such as Li⁺, Na⁺, K⁺ ions into the MoS₂ layers.^18,19 Compared with MoS₂, MoSe₂ is an interesting narrow-band-gap semiconductor with similar layered structures but a higher intrinsic electrical conductivity than MoS₂ due to the more metallic nature of Se.²⁰ Moreover, the relatively larger interlayer spacing of MoSe₂ can accommodate more ions to intercalate and extract, which makes it to be one of the most promising electrode materials for Li-ion batteries and pseudocapacitors.^21,22 Up to now, the study of the MoSe₂ for electrochemical capacitors has just started to rise. For example, Jun et al. hydrothermally synthesized the few-layered MoSe₂ nanosheets which demonstrated an ideal EDL behavior in 0.5 M H₂SO₄ with a capacitance of about 30 F g⁻¹ at the current density of 1 A g⁻¹.²¹ When incorporated with reduced graphene oxide (rGO), the MoSe₂/rGO delivered a EDL capacitance of about 20 F g⁻¹ at the same current density.²³ Huang et al. deposited the MoSe₂ and MoSe₂/graphene composites on a Ni-foam current collector via a hydrothermal method and showed very high specific capacitance values of 1114.3 F g⁻¹ (MoSe₂/Ni) and 1422 F g⁻¹ (MoSe₂/graphene/Ni) at 1 A g⁻¹ using a three electrode system with a 6 M KOH electrolyte.^24,25 However, the phase purity and structural conformation of the as-prepared product were not clear from comparison of the XRD data with the standard JCPDS card.

To enhance the electrochemical performance of electrodes, an efficient method is to explore a three-dimensional (3D) hierarchical hybrid electrode material with porous structures.²⁶ For instance, 3D net-like molybdenum selenide-acetylene black supported on Ni foam exhibits a remarkably enhanced supercapacitive performance.²⁷ Such a 3D hybrid electrode can not only provide the synergistic effect of all individual constituents, but also allow for effective electrolyte infiltration, provide efficient and rapid pathways for ion and electron transport, withstand strain/stress in the electrode matrix and provide stable tunnels for electron transportation, ensuring enhanced cycling capacity and facilitating fast kinetics.²⁸

In this work, we present a facile triethylene glycol-assisted hydrothermal approach to fabricate 3D hierarchical MoSe₂/C hybrid assembled by numerous few-layered nanosheets as building blocks. The samples were characterized by XRD, SEM, EDS, TEM/HRTEM, Raman and nitrogen adsorption–desorption. When evaluated as an electrode for supercapacitor, the as-prepared MoSe₂/C hybrid exhibited an prominently enhanced electrochemical performance in comparison with the bare MoSe₂.

2. Experimental section

2.1 Synthesis of hierarchical MoSe₂/C hybrid

In a typical synthesis, 3 mmol of Se powder was first dissolved in 10 mL hydrazine hydrate under continuous vibration until homogeneous red brown solution was obtained. Meanwhile, 1.5 mmol of sodium molybdate was dissolved in 10 mL triethylene glycol (TEG) under violent stirring. Then 15 mL deionized water was poured into the above solution. After that, the as-prepared hydrazine hydrate–Se solution was dropwise added. After stirring for another 1 h, the resultant mixture was transferred into the 50 mL Teflon-lined stainless steel autoclave, which was heated and maintained at 240 °C for 48 h. After cooling naturally, the black solid product was collected by centrifugation and washed several times with deionized water and ethanol, and freeze-dried overnight. For comparison, the bare MoSe₂ was also prepared by a similar synthetic route without TEG in the reaction system.

2.2 Characterizations

XRD patterns were recorded with a Thermo X'TRA X-ray diffractometer with Cu Kα radiation (λ = 0.154056 nm). The morphologies of the samples were observed by using a SIRION-100 field emission scanning electron microscopy (SEM). The elemental compositions of the samples were analyzed by energy dispersive X-ray spectroscopy (EDS, GENENIS-4000). HRTEM characterizations was performed on a JEOL JEM-2010 TEM operating at 200 kV, for which each sample was prepared by dispersing it in acetone and drop-casting onto a 200 mesh copper grid coated with holey carbon. Raman spectra were recorded with 514 nm Ar ion laser at 6 mW for 50 s employing a Jobin Yvon LabRam HR spectrometer. The nitrogen adsorption–desorption was conducted at 77 K on a Quantachrome NOVA 2000e sorption analyzer and the Brunauer–Emmett–Teller (BET) surface area was estimated from the adsorption data. Electrochemical impendence spectroscopy (EIS) was obtained by applying a voltage amplitude of 5 mV in the frequency range 100 kHz to 0.01 Hz on a CHI660D.

2.3 Electrochemical measurements

Electrochemical measurements were performed with a CHI660D electrochemical analyzer. And the electrochemical tests were taken in 3.0 M KOH electrolyte using a three electrode cell comprising a saturated calomel electrode (SCE) electrode as reference, Pt wire as counter and the as-prepared samples-based electrode as the working electrode. Working electrodes were prepared by mixing electroactive material, carbon black and polytetrafluoroethylene (PTFE) binder in a mass ratio of 80 : 5 : 15 to obtain a slurry. Then the slurry was pressed onto the nickel foam current collector (1 cm²) and dried at 60 °C for 12 h. For supercapacitors based on faradaic process, non-linearity often appears in the galvanostatic charge–discharge curves. The discharge capacitance of the as-prepared samples can be calculated from the galvanostatic discharge curve according to eqn (1).^29,30where C is the discharge capacitance of the sample (F g⁻¹), I is the discharge current (A), U is the potential (V), which is the function of discharge time t (s), ΔU is the potential window (V), and m is the loading mass of the electroactive materials (g).

3. Results and discussions

3.1 Microstructure and morphology of the hierarchical MoSe₂/C hybrids

The crystallinity and phase information of the as-prepared bare MoSe₂ and MoSe₂/C hybrid are examined by X-ray diffraction (XRD), as shown in Fig. 1. The diffraction peaks of both bare MoSe₂ and MoSe₂/C hybrid can be well indexed to the hexagonal MoSe₂ phase (JCPDS no. 29-0914). Compared with the bare MoSe₂, the relatively weak diffraction peaks in MoSe₂/C hybrid can be indicative of the inhibition of MoSe₂ crystal growth to a certain extent. A weak peak labeled as (*) at 2θ = 25.6° can be attributed to diffraction of amorphous carbon from decomposition of triethylene glycol during hydrothermal process. It is known that the (002) reflection in lamellar structured MoSe₂ materials can be ascribed to stacked Se–Mo–Se layers along the (002) direction. Thus, the high and sharp (002) peak of bare MoSe₂ can be indicative of the well-stacked MoSe₂ layers along the (002) direction. By contrast, the obviously weakened (002) reflection of MoSe₂/C can demonstrate the poorly stacked layers and few-layered structures in hybrid. Additionally, it is noted that the (002) peak at 2θ = 12.83° of MoSe₂/C has a slight downshift in comparison with that of bare MoSe₂ (2θ = 13.48°), suggesting an enlarged d-spacing. According to Bragg's equation, the d-spacing of (002) plane of the bare MoSe₂ and MoSe₂/C is calculated about 6.56 Å and 6.89 Å, respectively. The expanded d-spacing of (002) plane in MoSe₂/C could arise from the intercalation of amorphous carbon.

Fig. 1 XRD patterns of the as-prepared (a) bare MoSe₂ and (b) MoSe₂/C hybrid.

The morphology of the as-prepared bare MoSe₂ and MoSe₂/C were then analyzed using FESEM. Fig. 2a shows that the bare MoSe₂ demonstrates a laminated structure with different sizes aggregating together. By contrast, the MoSe₂/C hybrid possesses a 3D hierarchical flower-like nanostructure, which is constructed from 2D nanosheets as shown in Fig. 2b. In addition, it can be clearly observed from Fig. 2c that the neighboring nanosheets are loosely interconnected and that obvious open spaces exist between them. The inset TEM image in Fig. 2c shows conspicuous hierarchical porous structures that look like flowers and are composed of nanosheets. Fig. 2d clearly displays the few-layered feature of the MoSe₂/C with about 3–5 stacked layers and the crystal fringes has an interlayer spacing of 0.68 nm, corresponding to the (002) plane of 2H-MoSe₂. The elemental composition of the MoSe₂/C hybrid was identified by EDS as shown in Fig. 2e. It can be seen that the MoSe₂/C hybrid contains C, O, Mo and Se. The presence of C (33.73 wt%) and O in the hybrid derives from the amorphous carbon due to the decomposition of triethylene glycol during hydrothermal process. The atomic ratio of Se to Mo is 1.98, which agrees with the MoSe₂ stoichiometry.

Fig. 2 FESEM images of the as-prepared (a) bare MoSe₂ and (b, c) MoSe₂/C hybrid; (d) HRTEM image and (e) EDS pattern of MoSe₂/C hybrid.

Fig. 3 shows the Raman spectra of the samples. The bare MoSe₂ demonstrates two characteristic peaks at 235.14 cm⁻¹ (A_1g mode) and 280.71 cm⁻¹ (E¹_2g mode), respectively. The A_1g is out-of-plane mode and E¹_2g is in-plane mode.³¹ Compared with bare MoSe₂, it can be discerned that the intensities of E¹_2g and A_1g peaks of MoSe₂/C hybrid are obviously lower which reveals the interaction between MoSe₂ and carbon materials.^32,33 Moreover, it can be detected that MoSe₂/C hybrid has a down-shift of A_1g peak (232.57 cm⁻¹) in comparison with the bare MoSe₂, which further confirms the few-layer MoSe₂ characteristic.³¹ In addition, the occurrence of two new bands at 1349 and 1588 cm⁻¹ besides the MoSe₂ bands is noticeable. These bands are the so-called D and G bands, respectively, and are characteristic of carbon. The estimated I_D/I_G from Raman data is 1.23, which suggests amorphous carbon structure with more lattice edges or plane defects.³⁴

Fig. 3 Raman spectra of the as-prepared (a) bare MoSe₂; (b) MoSe₂/C hybrid.

Nitrogen (N₂) adsorption–desorption isotherm measurements were carried out to determine the surface area of the as-prepared samples. Fig. 4a and b show the corresponding N₂ adsorption–desorption isotherms for the bare MoSe₂ and MoSe₂/C hybrid. A typical type IV can be observed, indicating the presence of a mesoporous structure for both samples. The Brunauer–Emmett–Teller (BET) specific surface area of the MoSe₂/C was measured as about 51.79 m² g⁻¹, which is larger than that of the bare MoSe₂ (23.43 m² g⁻¹). It is believed that a relatively large surface area and porous feature are critical, offering more active sites for fast electrochemical reactions and facilitating charge carriers transfer at the electrolyte/electrode interface, which results in greatly enhanced electrochemical properties. Fig. 4c depicts N₂ adsorption–desorption isotherms for the amorphous carbon derived from the decomposition of triethylene glycol, which demonstrates a type II feature and a small specific surface area of 5.8 m² g⁻¹.

Fig. 4 Nitrogen adsorption–desorption isotherm of (a) MoSe₂ and (b) MoSe₂/C hybrid and (c) amorphous carbon.

In order to investigate the formation process of the 3D hierarchical flower-like MoSe₂/C nanostructure, time-dependent experiments were carried out. The morphological evolution was analyzed using SEM (Fig. 5). When the duration of the hydrothermal reaction was as short as 2 h, the sample was irregular and conglobatus nanoparticles. As reaction time was increased to 8 h, the scattered rag-like nanosheets were produced. With reaction time prolonged to 16 h, it was found that these nanosheets began to assemble together to form quasi-3D hierarchical nanosheet aggregates.

Fig. 5 FESEM images of the MoSe₂/C hybrid prepared by hydrothermal reaction at 240 °C after different reaction times: (a) 2 h, (b) 8 h, (c) 16 h.

Based on the above results, the possible formation process of the MoSe₂/C hierarchical flower-like hybrid is illustrated in Fig. 6, which involves initial nucleation, growth and assembly. At the initial stage, the MoSe₂ nanoparticles were first formed by nucleation. Afterwards, these nanoparticles gradually grew into dispersed two-dimensional sheet-like structures for high intrinsic anisotropic properties of MoSe₂. It was believed that the physical properties of mixed solvents such as viscosity could influence the crystal growth of MoSe₂.^35,36 As a result, few-layered MoSe₂ nanosheets were generated. Meanwhile, the triethylene glycol has 3D space structures to form a polymer network, which could also play a crucial role in directing the assembly and formation of 3D hierarchical nanostructures.^35,36 In addition, triethylene glycol could be further decomposed and carbonized to produce amorphous carbon materials under hydrothermal condition. Finally, the MoSe₂/C hierarchical hybrid was formed.

Fig. 6 Illustration of the formation process of the hierarchical MoSe₂/C hybrid.

3.2 Electrochemical performance of the samples

Fig. 7a shows typical CV curves of the bare MoSe₂ and MoSe₂/C hybrid electrodes within the potential range from −0.2 to 0.5 V at a scan rate of 5 mV s⁻¹ in 3.0 M KOH. Obviously, the CV curves of both electrodes are distinct from the ideal rectangular curves originating from the electric double-layer capacitance and can be indicative of a typical pseudocapacitive mechanism according to the redox peaks. The obvious cathodic peaks situated at around 0.13–0.15 V for both the bare MoSe₂ and MoSe₂/C can be attributed to the electrochemical insertion of K⁺ ions in the interlayer of layered MoSe₂. Accordingly, the obvious anodic peaks situated at about 0.34 V correspond to the extraction of K⁺ ions from layered MoSe₂. The nearly symmetrical redox peaks of both electrodes indicate the high reversibility of the insertion/extraction process. From Fig. 7a, the area under the CV curve for MoSe₂/C is clearly much larger than that for the bare MoSe₂. It is well known that the specific capacitance is proportional to the area of the CV curve. This demonstrates that the MoSe₂/C hybrid has higher capacitance than that of the bare MoSe₂. Moreover, the CV area of pure Ni foam is almost negligible compared with that of the bare MoSe₂ and MoSe₂/C hybrid electrodes, revealing the almost no capacitance contribution of the current collector. The CV curves of the bare MoSe₂ and MoSe₂/C electrodes under the different sweep rates are shown in Fig. 7b and c, respectively. As the scan rate increases, the cathodic peak position shifts to a lower potential, which is attributed to the polarization effect of the electrode. Additionally, it can be seen that CV curves of all the electrodes retain a similar shape even at high sweep rate, indicating an excellent capacitance behavior and a good high-rate response.

Fig. 7 Cyclic voltammogram (CV) curves of (a) the sample electrodes at 5 mV s⁻¹; (b) the bare MoSe₂ electrode and (c) the MoSe₂/C electrode at different scanning rates in 3.0 M KOH.

Galvanostatic charge–discharge is a complementary method for measuring the specific capacitance of electrochemical capacitors at constant current. Fig. 8a shows galvanostatic charge–discharge curves of the bare MoSe₂ and MoSe₂/C at a current density of 1 A g⁻¹. The increase in the discharging time represents the higher discharge capacitance of the MoSe₂/C hybrid than that of the bare MoSe₂. The specific capacitance values calculated from the discharging curves at 1 A g⁻¹ are 436.7 and 878.6 F g⁻¹ for the bare MoSe₂ and MoSe₂/C electrode, respectively, which indicates that the MoSe₂/C hybrid has superior electrochemical activity, which is in agreement with CV results. Additionally, at the same current density of 1 A g⁻¹, MoSe₂/C shows a smaller IR drop (21 mV) than that of bare MoSe₂ (41 mV), confirming the better electrical conductance of the MoSe₂/C electrode. Fig. 8b and c show galvanostatic charge–discharge curves of the bare MoSe₂ and MoSe₂/C at various current densities, respectively. For the bare MoSe₂, the discharge specific capacitance values are calculated to be 354.4, 269.1 and 170.4 F g⁻¹ at 2 A g⁻¹, 4 A g⁻¹ and 10 A g⁻¹, respectively. In comparison with the bare MoSe₂, the MoSe₂/C hybrid delivers a much higher specific capacitance of 819.4, 620.4 and 448.5 F g⁻¹ at 2 A g⁻¹, 4 A g⁻¹ and 10 A g⁻¹, respectively. The relationship between specific capacitance and current density is illustrated in Fig. 8d. MoSe₂/C hybrid exhibits a capacitance of 448.5 F g⁻¹ at 10 A g⁻¹ with a capacitance retention of 51% relative to 1 A g⁻¹. In contrast, the bare MoSe₂ electrode delivers a much lower capacitance value of 170.4 F g⁻¹ at 10 A g⁻¹ with only 39% retention in comparison with 1 A g⁻¹.

Fig. 8 (a) Galvanostatic charge–discharge curves of the sample electrodes at 1 A g⁻¹; (b) the bare MoSe₂ electrode and (c) the MoSe₂/C hybrid electrode at different current densities; (d) specific discharge capacitances versus current densities of the two sample electrodes.

An endurance test is conducted using galvanostatic charge–discharge cycles at 1 A g⁻¹ to inspect cycling performance of sample electrodes. It can be seen in Fig. 9a that the bare MoSe₂ exhibits a capacitance loss up to 28% after 2000 cycles. The obvious decay in capacitance may be caused by chemical dissolution and ion intercalation/deintercalation induced material pulverization. For MoSe₂/C, still 98% of the initial capacitance can be maintained after 2000 cycles without noticeable decrease. The prominently enhanced electrochemical performances of MoSe₂/C hybrid can be attributed to the robust 3D hierarchical architecture as well as the incorporation with conductive carbon materials, which can be favorable to maintaining structural integrity, offering sufficient surface area for electrochemical reactions and promoting charge carriers transfer at the electrolyte/electrode interface, which results in greatly enhanced electrochemical properties.

Fig. 9 (a) Cycling performance of the as-prepared sample electrodes at 1 A g⁻¹ and (b) EIS spectra of the sample electrodes (the inset is the equivalent circuit).

In order to understand why MoSe₂/C electrode exhibits such a superior electrochemical performance, EIS measurements were performed as shown in Fig. 9b. Both impedance spectra are composed of semicircles in the high frequency region followed by a linear slope in the low frequency region. The diameter of the characteristic depressed semicircle in the high frequency regions is related to charge transfer resistance caused by faradaic redox process. It is known that a small semi-circle for the electrode suggests a low interfacial charge-transfer resistance.³⁷ It is clearly observed that the diameter of the Nyquist circle of MoSe₂/C is smaller than that of bare MoSe₂, suggesting the fast electron transfer process and improved ionic conductivity at the electrode/electrolyte interface. A linear region in the low-frequency range, which corresponds to the diffusive resistance (Warburg impedance). The equivalent circuit (inset of Fig. 9b) for fitting the EIS data contains the internal resistance (R_s), the charge transfer resistance (R_ct), the constant phase element (CPE) and the Warburg impedance (W_s).³⁸ The internal resistance is related to the resistance of substrate, contact of material with substrate, and electrolyte resistance, which can cause IR drop. The R_s and R_ct values obtained by data fitting according to the equivalent circuit model are summarized in Table 1. It can be seen that R_s values are basically in agreement with R_i values calculated by IR drop from galvanostatic discharge curves. The constant phase element (CPE) is related to capacitive behavior of the non-faradaic charge storage mechanism. The smaller R_ct value of MoSe₂/C than that of bare MoSe₂ confirms the superior electrochemical performance of MoSe₂/C.

Table 1 R_s, R_i and R_ct values of the MoSe₂ and MoSe₂/C composite electrodes

Sample	Rs/Ω	Ri/Ω	Rct/Ω
MoSe2	6.83	5.86	167.44
MoSe2/C	3.75	2.69	75.66

---

4. Conclusions

In summary, 3D hierarchical flower-like MoSe₂/C hybrid was prepared through a one-pot hydrothermal route with triethylene glycol as structure-directing agent and carbon source. The obtained hierarchical porous MoSe₂/C hybrid consists of interconnected few-layered MoSe₂ nanosheets and triethylene glycol-derived amorphous carbon. The desirable structural features such as hierarchical porous structures with large surface area and intimate connection with conductive carbon materials endow the MoSe₂/C hybrid with superior electrochemical properties including high capacitance, excellent cyclic stability and greatly improved rate capability when it was investigated as an electrode for supercapacitor. It is believed that such a simple and efficient strategy can be extended to the large-scale production of other hierarchical composites with wide potential applications in supercapacitors and lithium ion batteries.

Acknowledgements

This work is financially supported by the Guangdong Natural Science Foundation for Joint Training Innovative Talents in East–West–North Guangdong (2014A030307030, 2016A030313667), the Science and Technology Planning Project of Guangdong Province (2014A010106032), the Stong Innovation School of Engineering Program of of Department of Education of Guangdong Province (Distinctive Innovation Project No. 2014KTSCX157), Natural Science Foundation of Lingnan Normal University (LZL1502, LZL1402).

Notes and references

1. D. P. Dubal, O. Ayyad, V. Ruiz and P. Gomez-Romero, Chem. Soc. Rev., 2015, 44, 1777–1790.
2. M. Yu, Z. Wang, Y. Han, Y. Tong, X. Lu and S. Yang, J. Mater. Chem. A, 2016, 4, 4634–4658.
3. X. Lu, M. Yu, G. Wang, Y. Tong and Y. Li, Energy Environ. Sci., 2014, 7, 2160–2181.
4. Y. Zhang, L. Li, H. Su, W. Huang and X. Dong, J. Mater. Chem. A, 2015, 3, 43–59.
5. H. Wan, J. Jiang, Y. Ruan, J. Yu, L. Zhang, H. Chen, L. Miao and S. Bie, Part. Part. Syst. Charact., 2014, 31, 814–819.
6. X. Zhao, P. Zhang, Y. Chen, Z. Su and G. Wei, Nanoscale, 2015, 7, 5080–5093.
7. R. Lv, J. A. Robinson, R. E. Schaak, D. Sun, Y. Sun, T. E. Mallouk and M. Terrones, Acc. Chem. Res., 2015, 48, 56–64.
8. M. Pumera, Z. Sofer and A. Ambrosi, J. Mater. Chem. A, 2014, 2, 8981–8987.
9. H. Wang, H. Feng and J. Li, Small, 2014, 10, 2165–2181.
10. A. Y. S. Eng, A. Ambrosi, Z. Sofer, P. Simek and M. Pumera, ACS Nano, 2014, 8, 12185–12198.
11. G. Ma, H. Peng, J. Mu, H. Huang, X. Zhou and Z. Lei, J. Power Sources, 2013, 229, 72–78.
12. N. Savjani, E. A. Lewis, M. A. Bissett, J. R. Brent, R. A. W. Dryfe, S. J. Haigh and P. O'Brien, Chem. Mater., 2016, 28, 657–664.
13. S. Patil, A. Harle, S. Sathaye and K. Patil, CrystEngComm, 2014, 16, 10845–10855.
14. X. Wang, J. Ding, S. Yao, X. Wu, Q. Feng, Z. Wang and B. Geng, J. Mater. Chem. A, 2014, 2, 15958–15963.
15. M. Acerce, D. Voiry and M. Chhowalla, Nat. Nanotechnol., 2015, 10, 313–318.
16. H. Ji, C. Liu, T. Wang, J. Chen, Z. Mao, J. Zhao, W. Hou and G. Yang, Small, 2015, 11, 6480–6490.
17. L. Ren, G. Zhang, Z. Yan, L. Kang, H. Xu, F. Shi, Z. Lei and Z. H. Liu, ACS Appl. Mater. Interfaces, 2015, 7, 28294–28302.
18. B. Hu, X. Qin, A. M. Asiri, K. A. Alamry, A. O. Al-Youbi and X. Sun, Electrochim. Acta, 2013, 100, 24–28.
19. K. J. Huang, L. Wang, Y. J. Liu, Y. M. Liu, H. B. Wang, T. Gan and L. L. Wang, Int. J. Hydrogen Energy, 2013, 38, 14027–14034.
20. X. Hu, W. Zhang, X. Liu, Y. Mei and Y. Huang, Chem. Soc. Rev., 2015, 44, 2376–2404.
21. S. K. Balasingam, J. S. Lee and Y. Jun, Dalton Trans., 2015, 44, 15491–15498.
22. Z. A. Zhang, X. Yang, Y. Fu and K. Du, J. Power Sources, 2015, 296, 2–9.
23. S. K. Balasingam, J. S. Lee and Y. Jun, Dalton Trans., 2016, 45, 9646–9653.
24. K. J. Huang, J. Z. Zhang and Y. Fan, Mater. Lett., 2015, 152, 244–247.
25. K. J. Huang, J. Z. Zhang and J. L. Cai, Electrochim. Acta, 2015, 180, 770–777.
26. Q. Xu, Y. Lv, C. Dong, T. S. Sreeprased, A. Tian, H. Zhang, Y. Tang, Z. Yu and N. Li, Nanoscale, 2015, 7, 10883–10895.
27. X. Liu, J. Z. Zhang, K. J. Huang and P. Hao, Chem. Eng. J., 2016, 302, 437–445.
28. M. Yu, W. Qiu, F. Wang, T. Zhai, P. Fang, X. Lu and Y. Tong, J. Mater. Chem. A, 2015, 3, 15792–15823.
29. S. L. Zhang and N. Pan, Adv. Energy Mater., 2014, 6, 1401401.
30. A. Laheäär, P. Przygocki, Q. Abbas and F. Béguin, Electrochem. Commun., 2015, 60, 21–25.
31. D. J. Late, S. N. Shirodkar, U. V. Waghmare, V. P. Dravid and C. N. R. Rao, ChemPhysChem, 2014, 15, 1592–1598.
32. B. Qu, X. Yu, Y. Chen, C. Zhu, C. Li, Z. Yin and X. Zhang, ACS Appl. Mater. Interfaces, 2015, 7, 14170–14175.
33. Y. Liu, M. Zhu and D. Chen, J. Mater. Chem. A, 2015, 3, 11857–11862.
34. J. Sanetuntikul, C. Chuaicham, Y. W. Choi and S. Shanmugam, J. Mater. Chem. A, 2015, 3, 15473–15481.
35. J. Xiong, G. Cheng, Z. Lu, J. Tang, X. Yu and R. Chen, CrystEngComm, 2011, 13, 2381–2390.
36. R. Jin, J. Liu, Y. Xu, G. Li, G. Chen and L. Yang, J. Mater. Chem. A, 2013, 1, 10942–10950.
37. A. H. Loo, A. Bonanni, Z. Sofer and M. Pumera, Electrochem. Commun., 2015, 50, 39–42.
38. J. Wu, P. Guo, R. Mi, X. Liu, H. Zhang, J. Mei, H. Liu, W.-M. Lau and L. M. Liu, J. Mater. Chem. A, 2015, 3, 15331–15338.

---