Vertically oriented MoS₂ and WS₂ nanosheets directly grown on carbon cloth as efficient and stable 3-dimensional hydrogen-evolving cathodes

Abstract

The development of non-noble-metal based hydrogen-evolving catalysts is essential to the practical application of water-splitting devices. Improvement of both the activity and stability of such catalysts remains a key challenge. In this work, a simple solvothermal method is developed to directly grow MoS₂ and WS₂ on carbon cloth with vertically oriented nanosheet layers. With the unique layer orientation that maximally exposes active edge sites as well as a rapid release of small gas bubbles to maintain a large working area, such prepared 3-dimensional electrodes exhibit high activity towards the HER. In addition, they also exhibit prominent electrochemical durability, thanks to the strong bonding between the nanosheet layers and the substrate along with the self-removal of the as-formed H₂ bubbles from the nano-porous electrode surface.

---

Hydrogen generation through water electrolysis has been extensively investigated as an attractive way to store energy from renewable sources.^1,2 The most effective electrocatalytic materials for the hydrogen evolution reaction (HER) are Pt group metals, but their high cost and scarcity limits their widespread use.³ As such, developing efficient and inexpensive HER catalysts with high stability is highly desirable but remains challenging. So far, both theoretical and experimental studies have highlighted the great promise of layered transitional-metal dichalcogenides (LTMDs), such as molybdenum disulfide (MoS₂) and its derivatives, as efficient and low cost catalysts for hydrogen evolution.^4–7 Tungsten disulfide (WS₂) has an analogous structure to MoS₂ and these compounds share similar physical and chemical properties. Thus WS₂ has also received some attention as an electrocatalyst for the HER for almost 20 years.^8–11

Since the edges of MoS₂ were identified as active sites for the HER,¹² a lot of research efforts have been focused on the growth of MoS₂/WS₂ nanostructures that maximally expose active edge sites for large-scale applications.^13–16 Previously, Cui et al. first demonstrated a thin film structured HER electrode with vertically aligned MoS₂ molecular layers grown on flat substrates.¹⁷ Later, the same approach was extended to the growth of MoSe₂ on rough and curved surfaces in an attempt to increase the exposed edges,¹⁸ as the number of the exposed edge sites is still limited on the flat substrates. Unfortunately, the fabrication process of the nanostructured HER electrodes is rather complicated and also difficult to control. Recently, Jiang et al. successfully demonstrated the construction of the vertically aligned MoS₂ nanoplatelet electrode with a “superaerophobic” surface, which allows efficient removal of gas bubbles and ensures high electrochemical activity.¹³ Unfortunately, the use of expensive flat Ti foil as a substrate seems to limit the maximal exposure of active sites as well as its scalable application due to the high cost. Carbon cloth (CC), an easily available and cheap carbon fiber, is highly conductive and flexible. Its use as an electrode support bears the intrinsic benefits of low cost, robustness and a self-standing 3D structure. Although 3D electrodes composed of WS₂ or MoS₂ nanoparticles on carbon cloth have been reported, there is no report on the direct growth of vertically aligned MoS₂ nanoplatelets on carbon cloth which is of great significance for the practical application of water-splitting devices.

Herein, we demonstrate a simple way to fabricate vertically oriented MoS₂ and WS₂ nanosheet thin films grown on CC as efficient 3D electrodes for the HER (details are available in the ESI†). Thanks to the nanostructured films composed of vertically aligned layers, resulting in a large number of exposed active edges and timely repelling of as-formed H₂ bubbles, the well-designed 3D electrodes showed much higher electrochemical activities than flat MoS₂ and WS₂ nanosheets on CC.

Fig. 1a & b show the low-magnification scanning electron microscopy (SEM) images of MoS₂/CC (a) and WS₂/CC (b), indicating that the entire surface of the CC was uniformly covered with a layer of the MoS₂/WS₂ nanosheet film. These films possessed a porous structure with only ∼100 nm depth (Fig. S1,† as indicated by the black arrows). The high-magnification SEM images (insets of Fig. 1a & b) further reveal that such MoS₂/WS₂ nanosheets are vertically aligned on the CC surface, possessing a lateral size of around 200–300 nm. The corresponding energy dispersive X-ray (EDX) spectra (see Fig. S2†) suggest that the atomic ratio between Mo/W and S is close to 2 : 1. CC is a porous 3D electrode consisting of a carbon fiber of 8 μm diameter (see Fig. S3†). The highly textured surface of the carbon fiber facilitates the nucleation and growth of a sulfide layer with strong mechanical interaction. It is noted that, after annealing at 350 °C for 2 h, the 2D nanosheet morphology is intact and the 3D porous architectures are well retained (Fig. S4a & b†). Fig. 1c & d are representative transmission electron microscopy (TEM) images of the MoS₂ and WS₂ nanosheet films, respectively, showing good agreement with SEM observation (Fig. 1a & b). The light contrast in various areas of the TEM image indicates the thin two-dimensional (2D) nature. Moreover, the high-resolution (HR) TEM image (inset of Fig. 1c) of the curled edge shows that those MoS₂ nanosheets are composed of 7–10 layers (4–6 nm) with an interlayer spacing of 0.67 nm, while the WS₂ nanosheets show a thickness of ∼10 nm with an interlayer distance of 0.62 nm (inset of Fig. 1d), verifying the ultrathin nature of these vertically oriented MoS₂/WS₂ nanosheets. Fig. 1e & f show the SEM images and the corresponding EDX elemental mapping images of the as-prepared MoS₂/CC and WS₂/CC electrodes, respectively, revealing that both Mo/W and S elements are uniformly distributed on the whole nanofiber.

Fig. 1 (a and b) Low and high-magnification (insets) SEM images of the nanostructured MoS₂/CC electrode (a) and WS₂/CC electrode (b). (c and d) TEM and high-resolution (HR) TEM (insets) images of the nanostructured MoS₂/CC electrode (c) and WS₂/CC electrode (d). (e and f) The corresponding SEM image and EDX elemental mapping of Mo, S and C for the nanostructured MoS₂/CC electrode (e) and W, S and C for the WS₂/CC electrode (f).

To further characterize the chemical nature and bonding state of MoS₂/WS₂ on CC surfaces, X-ray photoelectron spectroscopy (XPS) was employed. As shown in Fig. 2a, two characteristic peaks arising from Mo 3d_5/2 and Mo 3d_3/2 orbitals are located at 229.1 eV and 232.2 eV, suggesting the dominance of Mo(IV) in the freshly prepared MoS₂/CC product. Besides, small shoulder signals for the Mo(VI) oxidation state are observed. In contrast, the Mo 3d spectrum of the annealed sample shows only a single doublet. Furthermore, the S 2p_3/2–1/2 doublet peaks of the freshly prepared MoS₂ exhibit broader peaks at ∼163 eV and 162 eV when compared with those of the annealed sample, indicating the existence of other binding signals, such as bridging S₂²⁻ or apical S²⁻, which could result from the unsaturated S atoms and are known as active sites for the HER.^19,20 Thus better HER performance can be expected from the freshly prepared S-rich MoS₂ sample.

Fig. 2 XPS spectra of Mo 3d and S 2p in the MoS₂/CC sample (a) and W 4f and S 2p in the WS₂/CC sample (b), as-synthesized and after thermally annealed at 350 °C.

Fig. 2b shows the high-resolution XPS spectra of the W 4f and S 2p regions of the WS₂/CC products. For the WS₂/CC composite after annealing, black dotted lines have been added to bisect the W 4f_7/2, W 4f_5/2 and W 5p_3/2 peaks which appeared at about 33.1 eV, 35.2 eV and 38.7 eV, suggesting an oxidation state of W(IV), and the corresponding single doublet peaks of S 2p_1/2 and 2p_3/2 appeared at 162.4 eV and 163.6 eV, indicating S²⁻. These peaks in the annealed WS₂/CC are identical to those of highly crystalline WS₂.²¹ For the freshly obtained WS₂/CC, the doublet peaks of W 4f and S 2p were broader than those of the annealed WS₂/CC. Besides, a small negative shift of about 0.5 eV was observed, which could be attributed to the adsorption of oxygen,²² while the small shoulder observed at ∼36 eV illustrates the existence of W(VI) in an oxygen-rich environment as in WO₃.^23,24 This indicates that the freshly prepared WS₂/CC is more susceptible to oxidation.²⁵ It has been reported that the WS₂ sample appears to be more susceptible to oxidation than MoS₂.⁸

For MoS₂ materials, the basal edges have been identified as the active sites for the HER, but unfortunately, the overall conductivity of such materials is limited due to the poor electron transport among the domains and it thus reduces the total HER activity. In this work, the vertically oriented MoS₂/WS₂ nanosheet film grown on CC (a rough and curved surface) is expected to not only maximize the exposed active sites and overcome the limited electron/proton transport, but also accelerate the timely repelling of the as-formed bubbles by reducing the gas bubble adhesion due to the highly ‘superaerophobic’ surface.¹²

To verify our hypothesis, the HER performance of the 3D MoS₂/CC and WS₂/CC electrodes was demonstrated in 0.5 M H₂SO₄ solution using a three-electrode setup, where the electrode was tested in a static state without rotation to mimic real industrial operation. Electrochemical impedance spectroscopy (EIS) reveals a similar system resistance (R_s, 2.5 ± 0.3 Ω) for all the tested electrodes (Fig. S5†). To exclude the influence of the series resistance from the system (R_s, such as wiring, solution and substrate), all the data have been iR-corrected by subtracting the ohmic resistance loss from the overpotential. As control samples, flat MoS₂/CC and WS₂/CC films were also prepared and characterized by SEM (Fig. S6†) and XPS (Fig. S7†). As shown in Fig. 3b, the cathodic polarization curve recorded for the nanostructured MoS₂/CC exhibited a low onset potential of ∼0.1 V, which is much smaller than that of the flat MoS₂/CC and annealed MoS₂/CC and even lower than that of the reported Ti foil supported MoS₂ nanostructure.¹³ As expected, our nanostructured MoS₂/CC electrode displays the highest current density among all the tested electrodes, 86 mA cm⁻² at an overpotential of around −0.25 V. This is equivalent to ∼450 mA mg⁻¹ after being normalized by the loading weight, which is almost two times larger than the value of the Ti foil supported MoS₂ nanostructure.¹³ Since the cathodic current density is proportional to the amount of evolved hydrogen, the large current density here indicates the prominent hydrogen evolution behavior of the vertically grown MoS₂/CC electrode. This may arise from the uniquely designed electrode structure which brings in more active sites along with the optimized conductivity and reduced gas bubble adhesion.

Fig. 3 (a) Schematic illustration of nanostructured MoS₂/CC and WS₂/CC electrodes. The active Mo/W edge for the HER is marked in the magnified crystal structure image. Blue sphere: Mo/W; yellow sphere: S. Polarization curves (after iR-correction) of MoS₂/CC (b) and WS₂/CC catalysts (c). (d) The corresponding Tafel plots of the samples.

Interestingly, in the case of the WS₂/CC electrodes (Fig. 3c), the annealed nanostructured WS₂/CC showed the highest current density (∼15 mA cm⁻²) among the tested WS₂ samples at an overpotential of −0.25 V. Although the annealed WS₂/CC electrode shows lower activity than the nanostructured MoS₂/CC with a 0.15 V larger overpotential to achieve a cathodic current density of 86 mA cm⁻², it is still noteworthy because after being normalized by the loading weight, this value (∼15 mA cm⁻² at −0.25 V) is about 7 times larger than that of carbon cloth supported WS₂ nanoparticles²³ and even slightly higher than that of WS₂/rGO.²⁶ This is likely due to the vertically oriented feature of our WS₂/CC electrode. It is noted that the annealed WS₂/CC electrode displays much higher activity than the as-prepared WS₂/CC, this may be attributed to the removal of inactive impurities such as WO₃,²³ as verified by the XPS analysis.

The Tafel plots of these catalysts are shown in Fig. 3d and are used to determine Tafel slopes and exchange current densities by fitting the Tafel plots to the equation η = b log(j) + log(j₀), where η is the overpotential, j is the current density, j₀ is the exchange current density, and b is the Tafel slope.²⁷ The fitted Tafel plot of the nanostructured MoS₂/CC gives a b value of 50 mV dec⁻¹, which is much lower than that of the annealed MoS₂/CC and even slightly smaller than the recently reported value for pure MoS₂ nanoplates (53 mV dec⁻¹).²⁸ It is likely due to the oriented growth of MoS₂ nanosheets on CC which results in the improvement of the conductivity of the MoS₂ materials and the increase in the number of the exposed active sites. Then the exchange current density is determined to be 9.2 × 10⁻³ mA cm⁻² for the nanostructured MoS₂/CC sample, almost one order of magnitude higher than the Ti foil supported MoS₂ nanostructure (3.87 × 10⁻⁴ mA cm⁻²)¹³ and carbon fiber paper supported MoSe₂ (3.80 × 10⁻⁴ mA cm⁻²).¹⁸ Although the Tafel slope for annealed WS₂/CC is smaller than the freshly prepared nanostructured WS₂/CC, this is still much higher than that of the MoS₂ and consistent with previous results.^23,26 Moreover, these values compare favorably with most of the reported values for non-precious HER catalysts in acidic aqueous electrolytes (see Table S1 in the ESI†).

Stability is another important criterion used to evaluate a catalyst. A long-term cyclic voltammetry test was performed to assess the electrochemical stability of the nanostructured MoS₂/CC electrode in an acidic environment. As shown in Fig. 4a, working under such conditions, the nanostructured MoS₂/CC electrode and the annealed WS₂/CC electrode performed still steadily, suggested by the smooth curve recorded after 2000 cycles along with negligible current degradation. Furthermore, a continuous chronoamperometric curve is also presented in Fig. 4b, where a stable current density of ∼22 mA cm⁻² (at an overpotential of 0.2 V) and ∼17 mA cm⁻² (at an overpotential of 0.25 V) over an operating period of 24 h was observed for the nanostructured MoS₂/CC and annealed WS₂/CC, respectively, further revealing the high stability of these nanostructured electrodes. Although, for a practical electrode device, the stability performance will need to be maintained for a much longer time, the performance of the as-prepared electrodes in this work shows great potential for scalable applications. Notably, our MoS₂/CC and WS₂/CC electrodes are robust enough to withstand an ultrasonication treatment even for 30 min (Fig. S8†), indicating the strong binding between the CC and the nanostructured MoS₂/WS₂ sheets, which ensures the excellent stability of those nanostructured electrodes. Notably, it is observed that the small H₂ bubbles formed on the 3D electrode can be timely released during the working process, which in turn ensures a sufficient number of active sites, resulting in a stable electrocatalytic performance (details can be seen in SI-2†). This is fully consistent with the previous observation.¹³

Fig. 4 Stability testing of the nanostructured MoS₂/CC electrode and annealed WS₂/CC electrode. (a) Polarization curves of the electrodes before and after 2000 CV scans between +0.1 and −0.5 V vs. RHE. (b) Chronoamperometric curve for both electrodes under static overpotentials of 200 mV and 250 mV for 24 h, respectively.

As discussed above, such high catalytic performance of the nanostructured MoS₂/WS₂ electrodes could be due to the following reasons: (1) the intimate contact of MoS₂ and WS₂ nanosheets with CC enables good mechanical binding and electrical connection, facilitating the flow of electrons from CC to MoS₂ or WS₂ nanosheet arrays during the HER. This is of great significance for the semi-conductive catalysts such as MoS₂ and WS₂. (2) The 3D configurations of the vertically oriented MoS₂/CC and WS₂/CC electrodes not only facilitate the self-removal of as-formed H₂ bubbles from the nano-porous electrode surfaces, but also ensure sufficient and open spaces which allow the easy diffusion of the electrolyte into all the active sites and thus a more efficient use of the entire electrode. (3) The ultrathin MoS₂/WS₂ vertical nanosheets could be a better structure compared to the multilayer nanoparticles because electrons only need to be transferred from the support to the platelets, instead of passing through different layers. (4) The vertically oriented S-rich MoS₂ nanosheet arrays and the annealed WS₂ nanosheets with low oxidized species expose more active sites for the HER than those of flat ones.

In summary, we developed a simple solvothermal method to fabricate MoS₂ and WS₂ grown on CC electrodes with vertically oriented nanosheet layers. With the unique layer orientation that maximally exposes the active edge sites as well as a rapid release of small gas bubbles to provide a constant working electrode area, the nanostructured MoS₂/CC electrode exhibits excellent HER activity, while the WS₂/CC catalyst is also investigated here as a novel active HER architecture for potential use as a HER catalyst. Both materials exhibit prominent electrochemical durability, thanks to the strong bonding between the nanosheet layers and the substrate along with the self-removal of the as-formed H₂ bubbles from the nano-porous electrode surface. Moreover, these vertically oriented layered catalysts can be readily applied in diverse water electrolysis devices as easily available, high-performance and stable HER catalysts.

Acknowledgements

This project is funded by the National Research Foundation (NRF), Prime Minister's Office, Singapore under its Campus for Research Excellence and Technological Enterprise (CREATE) programme. We also acknowledge financial support from the academic research fund AcRF tier 1 (M4011020 RG8/12 and M4011253 RG 7/14) Ministry of Education, Singapore.

References

1. J. A. Turner, Science, 2004, 305, 972–974.
2. J. N. Armor, Catal. Lett., 2005, 101, 131–135.
3. M. G. Walter, E. L. Warren, J. R. McKone, S. W. Boettcher, Q. Mi, E. A. Santori and N. S. Lewis, Chem. Rev., 2010, 110, 6446–6473.
4. M.-R. Gao, Y.-F. Xu, J. Jiang and S.-H. Yu, Chem. Soc. Rev., 2013, 42, 2986–3017.
5. A. B. Laursen, S. Kegnaes, S. Dahl and I. Chorkendorff, Energy Environ. Sci., 2012, 5, 5577–5591.
6. M. Chhowalla, H. S. Shin, G. Eda, L.-J. Li, K. P. Loh and H. Zhang, Nat. Chem., 2013, 5, 263–275.
7. Y. Yan, B. Xia, X. Qi, H. Wang, R. Xu, J.-Y. Wang, H. Zhang and X. Wang, Chem. Commun., 2013, 49, 4884–4886.
8. M. A. Lukowski, A. S. Daniel, C. R. English, F. Meng, A. Forticaux, R. J. Hamers and S. Jin, Energy Environ. Sci., 2014, 7, 2608–2613.
9. D. Voiry, H. Yamaguchi, J. Li, R. Silva, D. C. B. Alves, T. Fujita, M. Chen, T. Asefa, V. B. Shenoy, G. Eda and M. Chhowalla, Nat. Mater., 2013, 12, 850–855.
10. J. Lin, Z. Peng, G. Wang, D. Zakhidov, E. Larios, M. J. Yacaman and J. M. Tour, Adv. Energy Mater., 2014, 4 DOI:10.1002/aenm.201301875.
11. Z. Wu, B. Fang, A. Bonakdarpour, A. Sun, D. P. Wilkinson and D. Wang, Appl. Catal., B, 2012, 125, 59–66.
12. T. F. Jaramillo, K. P. Jørgensen, J. Bonde, J. H. Nielsen, S. Horch and I. Chorkendorff, Science, 2007, 317, 100–102.
13. Z. Lu, W. Zhu, X. Yu, H. Zhang, Y. Li, X. Sun, X. Wang, H. Wang, J. Wang, J. Luo, X. Lei and L. Jiang, Adv. Mater., 2014, 26, 2683–2687.
14. Y. Yan, B. Xia, Z. Xu and X. Wang, ACS Catal., 2014, 1693–1705.
15. Y. H. Chang, F. Y. Wu, T. Y. Chen, C. L. Hsu, C. H. Chen, F. Wiryo, K. H. Wei, C. Y. Chiang and L. J. Li, Small, 2013, 10, 895–900.
16. P. D. Tran, S. S. Pramana, V. S. Kale, M. Nguyen, S. Y. Chiam, S. K. Batabyal, L. H. Wong, J. Barber and J. Loo, Chem.–Eur. J., 2012, 18, 13994–13999.
17. D. Kong, H. Wang, J. J. Cha, M. Pasta, K. J. Koski, J. Yao and Y. Cui, Nano Lett., 2013, 13, 1341–1347.
18. H. Wang, D. Kong, P. Johanes, J. J. Cha, G. Zheng, K. Yan, N. Liu and Y. Cui, Nano Lett., 2013, 13, 3426–3433.
19. D. Merki, S. Fierro, H. Vrubel and X. L. Hu, Chem. Sci., 2011, 2, 1262–1267.
20. D. Li, U. Maiti, J. Lim, D. S. Choi, W. J. Lee, Y. Oh, G. Y. Lee and S. O. Kim, Nano Lett., 2014, 14, 1228–1233.
21. R. Morrish, T. Haak and C. A. Wolden, Chem. Mater., 2014, 26, 3986–3992.
22. I. Martin, P. Vinatier, A. Levasseur, J. C. Dupin and D. Gonbeau, J. Power Sources, 1999, 81–82, 306–311.
23. T.-Y. Chen, Y.-H. Chang, C.-L. Hsu, K.-H. Wei, C.-Y. Chiang and L.-J. Li, Int. J. Hydrogen Energy, 2013, 38, 12302–12309.
24. A. Benadda, A. Katrib, J. W. Sobczak and A. Barama, Appl. Catal., A, 2004, 260, 175–183.
25. B. K. Miremadi and S. R. Morrison, J. Appl. Phys., 1988, 63, 4970–4974.
26. J. Yang, D. Voiry, S. J. Ahn, D. Kang, A. Y. Kim, M. Chhowalla and H. S. Shin, Angew. Chem., Int. Ed., 2013, 52, 13751–13754.
27. J. O. M. Bockris and E. C. Potter, J. Electrochem. Soc., 1952, 99, 169–186.
28. Y. Yan, B. Xia, X. Ge, Z. Liu, J.-Y. Wang and X. Wang, ACS Appl. Mater. Interfaces, 2013, 5, 12794–12798.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental section and figures, and a video. See DOI: 10.1039/c4ta04858j

---