A “1-methylimidazole-fixation” route to anchor small-sized nitrides on carbon supports as non-Pt catalysts for the hydrogen evolution reaction

Abstract

A effective “1-methylimidazole (1-MD)-fixation” strategy was developed to anchor small-sized nitrides on carbon supports. The carbon supports used (CNTs (carbon nanotubes), carbon black, reduced graphene oxides) and nitrides (WN, Mo₂N) can be easily tuned. The application of the nitride/CNT hybrids as non-Pt catalysts for the hydrogen evolution reaction (HER) was also demonstrated. It was shown that the performance could be tuned by the kind of the nitride. The combination of WN and Mo₂N simultaneously on CNTs can improve the HER performance obviously. It is expected that the present “fixation” route could be useful to anchor small-sized nitrides on carbon supports for application in electrocatalytic fields.

---

Transition metal nitrides (tungsten nitride, molybdenum nitride etc.) are promising materials in material science due to their advantages of high hardness, good stability, and potential applications in energy storage, solar cells and other fields.^1–4 Typically, following the study of MoS₂ (WS₂) as a non-Pt catalyst for the hydrogen evolution reaction (HER), it has been found that metal nitrides (WN) are also active catalysts for the HER, but with relatively low activity.⁵ A decrease in the size of nitrides can increase the surface active sites, and consequently, improve the catalytic performance.^6,7 However, the practical preparation of small-sized nitrides is difficult because the formation of nitrides needs to be performed at high temperatures (above 600 °C) which will result in the aggregation of the particles.⁸ In order to stabilize the nitrides, it is important to anchor the nitrides on different carriers.^9–11

Carbon materials are promising supports for loading with the various materials for application in many fields.^12,13 Large success has been found by growing the functional particles on graphene (or reduced graphene oxides, rGO).¹⁴ However, it is challenging to grow small-sized nitrides evenly on other carbons, such as carbon nanotubes (CNTs) and carbon black (CB), mainly due to their large curvatures. Compared with graphene, these types of carbon are also important supports due to their special properties, easy preparation and relatively low cost. In addition, the particles are inclined to aggregate and generate large particles during the nitriding process because of the low melting point or sublimation temperature of some oxide precursors.¹⁵ So, it is an urgent issue to develop an effective “fixation” technology to protect the particles from severe aggregation, and to form small-sized particles with tunable components on different carbon supports.

Polyoxometalates (POMs) are a class of molecular clusters containing W (Mo) elements.¹⁶ The “fixation” of POMs on carbons and the protection of them from severe aggregation provide a chance to realize the anchoring of small-sized nitrides on supports by using PEI (polyethyleneimine) as linker.¹⁷ The 1-MD can interact with POMs, like other containing-MD molecules.¹⁸ These kind of the molecules can also interact with carbon.¹⁹ Notably, the MD-based molecules can be potentially carbonized under heating which will protect the particles from aggregation.²⁰ Hence, 1-MD can be a potential linker to anchor the POMs onto carbon carriers and protect the POMs from aggregation under heating. Our results indicated that, by this “fixation” method, the small-sized nitrides can be anchored on carbon supports. Both the kinds of carbon supports and POMs can be easily tuned. The WN (Mo₂N) or both of them can be easily loaded on CNTs (CB and rGO). The results are different from the use of PEI as the linker, for which it is difficult to anchor the small-sized WN, especially Mo₂N uniformly on CNTs and CB. The hybrids can be used as non-Pt catalysts for the HER. It was shown that Mo₂N/CNTs had better performance for the HER than WN/CNTs, and the combination of WN and Mo₂N on carbon can improve the HER performance obviously.

Scheme 1 is the schematic process for anchoring WN (Mo₂N) on carbons using CNTs as a model. The CNTs were firstly treated with acid to endow –COOH and –OH on the surfaces (A-CNTs). 1-MD and phosphotungstic acid hydrate (PW₁₂) were added in sequence into the ethanol dispersion of CNTs. After calcination in air and NH₃, WN/CNTs were obtained. Taking WN/CNTs as a typical sample, we demonstrate the efficiency of the “1-methylimidazole-fixation” route. The transformation from 1-MD–PW₁₂/CNTs to WN/CNTs is reflected by the IR and Raman spectra (Fig. S1 and S2†). In the IR spectrum of A-CNTs, peaks of intra-molecular association O–H (3314 cm⁻¹), C=O (1718 cm⁻¹), C=C (1556 cm⁻¹), C–O (1202 cm⁻¹) are observed.²¹ For PW₁₂, we can find four characteristic absorption peaks located at 1080 cm⁻¹ (P–O), 982 cm⁻¹ (W–O), 890 cm⁻¹ (W–O–W), 802 cm⁻¹ (W–O–W). These intensive peaks also can be observed for the PW₁₂/CNTs sample, demonstrating the existence of a basic framework of PW₁₂ in the samples. After treatment under an NH₃ atmosphere, the characteristic peaks of PW₁₂ are not visible. The peaks located at 1170 cm⁻¹ can be ascribed to the formation of the W–N bond. The analysis shows the conversion of W–O to W–N and the WN/CNT complex is formed. In the Raman spectra, we can observe two prominent peaks located at 1360 cm⁻¹ and 1575 cm⁻¹, which respectively belong to the D band and G band of CNTs. The intensity ratio of G band to D band (I_G/I_D) is usually used to measure the order of carbon materials. A higher value represents good crystallinity of the carbon. It can be seen that all samples show high ratios of I_G/I_D. The I_G/I_D of PW₁₂/CNTs, WN/CNTs and A-CNTs are 1.88, 1.89 and 1.69 respectively, demonstrating the good crystallinity of the samples. The values are higher than the 1.24 of rGO derived from GO (Fig. S2b†). Compared with CNTs, a new small peak located at around 800 cm⁻¹ appears in PW₁₂/CNTs, WN/CNTs. For the PW₁₂/CNTs, the vibration peak situated at 803 cm⁻¹ is assigned to the stretching vibration of the W–O band. The vibration peak at 807 cm⁻¹ is ascribed to the W–N band, indicating the formation of WN/CNTs. The Raman results indicated the transformation from PW₁₂/CNTs to WN/CNTs, being consistent with that from XRD.

Scheme 1 Process for anchoring the nitrides on carbon supports. Step I: the anchoring of POMs on A-CNTs by 1-methylimidazole; step II: the calcination of 1-MD–POMs/CNTs in air; step III: nitriding in NH₃ to formation of WN (Mo₂N)/CNTs. A-CNTs represent acid treated CNTs.

A typical sample WN/CNT-50 is first characterized. As shown in Fig. 1a, the peaks in the XRD spectrum are attributed to CNTs and fcc structured WN (PDF#: 652898). The XPS test indicates the presence of C, N, W, O and P in the sample (Fig. S3†). The low-magnification TEM image shows that small particles of about 3 nm are distributed well on the CNTs (Fig. 1b, c and S4a†). The high resolution TEM (HRTEM) shows the spacing between the two planes is about 0.25 nm, corresponding to the (111) crystal plane of fcc WN (Fig. 1d).

Fig. 1 (a) XRD, (b–d) TEM and HRTEM images of WN/CNT-50. The inset in (d) is the HETEM and FFT of WN/CNT-50.

The loading can be tuned by changing the addition amount of PW₁₂. For WN/CNT-30, WN particles with size of about 2 nm are loaded on CNTs (Fig. 2a, c and S4b†). When the loading of WN is about 70%, the size of particles is mainly about 4 nm (Fig. 2b, d and S4c†). The results imply that the size and amount of WN on the CNTs can be easily tuned by changing the amount of PW₁₂ in the synthesis. The effect of loading on particle size can be explained as follows. In the present synthesis, the WN/CNT composites are formed by anchoring PW₁₂ on CNTs followed by calcination in NH₃ atmosphere. In spite of the effective anchoring effect of 1-MD, some aggregation of W precursor will happen under the nitridation process. The high loading implies the need for more W precursor, leading to a high density (less intervals) of W precursor on CNTs and consequently, an increase of aggregation and growth probability of particles. So, the particle size increased with increased loading of WN. This situation is similar to that of metal NPs/CNTs²² (or carbon black²³), in which the size of metal NPs increased with increasing metal salt concentration.

Fig. 2 TEM and HETEM images of (a and c) WN/CNT-30 and (b and d) WN/CNT-70.

The kind of metal cation has a large effect on the performance of the nitrides based on the different adsorption bond energy of metal–H. Li et al. have reported loading of WN of about 20 nm on CB.²⁴ There are several reports about the formation of molybdenum nitride/CNTs, in which the size of Mo₂N is larger than 20 nm.²⁵ However, the growth of small-sized Mo₂N with a size below 5 nm on carbon supports has not been reported. We found that the present route is also suitable for loading small-sized Mo₂N on CNTs. Fig. 3a and b show that the Mo₂N particles (about 3 nm, Fig. S4d†) can uniformly grow on CNTs (Mo₂N/CNTs-50). The spacing between two crystal planes is about 0.25 nm, corresponding to the (111) plane of Mo₂N. The XRD pattern indicates the formation of a Mo₂N/CNT composite (Fig. S5†). The anchoring of the small-sized Mo₂N should be promoted by the protective role from linking with 1-MD. Many works have demonstrated that the complex containing MD can be carbonized under an inert atmosphere,²⁶ which is beneficial for the formation of small-sized Mo₂N on CNTs. In contrast, PEI tends to decompose easily under heating, so its protective role is not as strong as 1-MD. As reported, the compounding of two nitrides (carbides) is promising to extend the applications of the materials.²⁷ Since WN and Mo₂N can be loaded on CNTs separately, it is possible to anchor nitrides containing both W and Mo simultaneously on supports. Indeed, this can be easily realized by fixing PW₁₂ and PMo₁₂ simultaneously on CNTs. We can see that small particles with size below 5 nm are uniformly loaded on CNTs. A combination of XRD, XPS and TEM indicates the growth of small-sized WN and Mo₂N simultaneously on CNTs (WN–Mo₂N/CNTs-50, Fig. S6–S8†).

Fig. 3 TEM and HETEM images of (a and b) Mo₂N/CNT-50 and (c and d) Mo₂N/CB-50.

The above analysis indicates that small sized WN (Mo₂N) can be located on CNTs by the “imidazole-fixation” route. This can be ascribed to the effective combination of 1-MD with POMs and CNTs (Schemes S1 and S2†). After the addition of 1-MD into the POM solution, a precipitate is formed rapidly, implying the effective combination of 1-MD with POMs (Fig. S9–S11†). Therefore, the POMs can be anchored on CNTs by using 1-MD as linker. Besides, the 1-MD can effectively protect the particles from aggregation due to its potential for carbonization under heating. In consequence, small-sized WN (Mo₂N) can be produced on CNTs by the “1-MD fixing” strategy. The results are different from the our previous report in which PEI was used as the linker agent. In that case, the WN, especially Mo₂N, cannot be anchored uniformly on CNTs.

Transition metal nitrides have potential application in the catalytic field as a replacement or co-catalyst for Pt metal.³ The potential application of nitrides/carbon for the HER are evaluated here (ESI†). The performance for the HER was tested by depositing catalysts with the same loading (0.7077 mg cm⁻²) on a glassy carbon electrode (GCE). We can see the remarkable HER activity of the Pt/C catalyst with an onset potential close to zero (Fig. S12†). Pure CNTs show poor activities with an onset overpotential of 315 mV and an η₁₀ (the potential at 10 mA cm⁻²) of 458 mV. As we expected, the onset overpotentials are about 184 mV and 117 mV for WN/CNTs and Mo₂N/CNTs, respectively (Table S1†). The results exhibit the improved HER activity of nitrides/CNTs in comparison with CNTs. Generally, the activity of an electrocatalyst to catalyze the HER can be described by the strength of the hydrogen–surface bond. The hydrogen binds just strongly enough to cover the surface but weak enough to facilitate desorption of H₂.²⁸ In other words, the closer the hydrogen binding energy (ΔG_[H]) is to zero, the better the HER activity of the catalyst. We can see that Mo₂N/CNTs show better activity than WN/CNTs, which is due to the fact that the Mo-based catalysts have a more suitable ΔG_[H] than that of W-based catalysts.^27a Recently, it has been shown that the combination of two types of nitrides is useful to enhance performance due to the synergistic effect of different components.²⁹ One of the advantages is that the multi-nitrides can tune the electronic state of the metal active site in a desirable direction. Indeed, it is shown that the WN–Mo₂N/CNT-50 exhibits a superior performance in comparison with the single Mo₂N/CNTs and WN/CNTs. The onset potential is about 80 mV (Fig. 4a). Although the performance is worse than that of Pt/carbon, it displays a lower onset potential compared with most MoS₂-based catalysts.³⁰ To discern the predominant HER mechanism of various catalysts, Tafel plots are fitted based on the Tafel equation (Z = a + b × log|j|), where j is the current density and b is the Tafel slope. The Tafel slope is about 115 mV per decade, lower than 133, 146 and 176 mV per decade for Mo₂N/CNTs, WN/CNTs and CNTs (Fig. 4b). One possible pathway for the HER on nitrides/CNTs is through a Volmer–Heyrovsky reaction mechanism. The rate determining step should be the discharge reaction or the electrochemical desorption of H ads and H₃O⁺ to form hydrogen. The exchange current densities (j₀) for WN–Mo₂N/CNT-50 are about 309 μA cm⁻², which are also higher than the corresponding values for WN/CNTs (89 μA cm⁻²) and Mo₂N/CNTs (258 μA cm⁻²) (Fig. S13†). We have also evaluated the electrochemical surface area of the catalysts by capacitance current. The WN–Mo₂N/CNT-50 gives a capacitance current of 44 mF cm⁻², much higher than 32 mF cm⁻² for Mo₂N/CNTs and 14.3 mF cm⁻² for WN/CNTs-50 (Fig. S14–16†). Thus, one reason for the large j₀ of WN–Mo₂N/CNT-50 can be ascribed to its high surface area. An accelerated degradation test has been performed to evaluate the stability. After 2000 CV sweeps, the voltage loss at 10 mA cm⁻² is about 14 mV and 9 mV for Mo₂N/CNTs and WN–Mo₂N/CNT-50, respectively (Fig. 4c). The inset in Fig. 4c and d shows the time dependence of the current density for Mo₂N/CNTs and WN–Mo₂N/CNTs at overpotentials of 218 mV and 190 mV. There is little change in current density after long-term tests for 12 h (Fig. 4d). The results suggest the long-term stability of nitride/CNTs. In short, the better performance of Mo₂N/CNTs than WN/CNTs can be attributed to the stronger adsorption bond energy of W–H than Mo–H. The strong adsorption is not conducive to H desorption from the catalysts.³¹ Also, for WN–Mo₂N/CNT-50, the improvement should be contributed to by the synergistic effect between the two components of W, which binds H strongly, and Mo, which binds H weakly. Moreover, the good-conductivity of WN may also be favorable for the improvement of the HER performance of WN–Mo₂N/CNTs.^27b In addition, by comparison of W4f and Mo3d spectra in binary Mo₂N/CNT-50, WN/CNT-50 and ternary WN–Mo₂N/CNT-50, the improvement of the HER activity for the ternary catalyst should also be relative to gaining the electrons of Mo and W and the enhancement of Mo–C components on the surface in ternary WN–Mo₂N/CNTs (Fig. S17 and 18†). And it may truly open up a new direction for the development of non-noble metal electrocatalysts.

Fig. 4 (a) Polarization curves of the catalysts in 0.5 M H₂SO₄ at a scan rate of 5 mV s⁻¹ were obtained by RDE rotated at 1600 rpm. (b) Tafel slopes of the catalysts. Polarization curves before and after 2000 CV cycles of (c) Mo₂N/CNTs and (d) WN–Mo₂N/CNT-50 ranging from 0 to 0.8 V.

Conclusions

In summary, we have demonstrated that the “imidazole-fixation” route is useful to anchor small-sized WN (Mo₂N) on CNTs. The hybrids can be used as non-Pt catalysts for the HER. The rational combination of WN and Mo₂N on CNTs can improve the HER performance obviously. Notably, the method is also effective to anchor nitrides on other carbon supports, including carbon black (Fig. 3c, d and S19†) and rGO (Fig. S20†). On account of the convenience to tune the components, it is expected that the performance of the hybrids can be further improved by rationally adjusting the components.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (21571054, 21201058), Program for Innovative Research Team in University (IRT-1237), Application Technology Research and Development Projects in Harbin (2013AE4BW051), International Science & Technology Cooperation Program of China (2014DFR41110), Heilongjiang University Excellent Youth Foundation.

Notes and references

1. J. S. J. Hargreaves, Coord. Chem. Rev., 2013, 257, 2015–2031.
2. S. Dong, X. Chen, X. Y. Zhang and G. L. Cui, Coord. Chem. Rev., 2013, 257, 1946–1956.
3. W. T. Yao, P. Makowski, C. Giordano and F. Goettmann, Chem.–Eur. J., 2009, 15, 11999–12004.
4. (a) X. L. Tian, J. M. Luo, H. X. Nan, Z. Y. Fu, J. H. Zeng and S. J. Liao, J. Mater. Chem. A, 2015, 3, 16801–16809; (b) S. P. Mani, C. Anandan and N. Rajendran, RSC Adv., 2015, 5, 64466–64470; (c) S. Y. Jing, L. Luo, S. B. Yin, F. Huang, Y. Jia, Y. Wei, Z. H. Sun and Y. M. Zhao, Appl. Catal., B, 2014, 147, 897–903; (d) S. B. Yin, L. X. Yang, L. Luo, F. Huang, Y. H. Qiang, H. W. Zhang and Z. X. Yan, New J. Chem., 2013, 37, 3976–3980.
5. W. F. Chen, J. T. Muckerman and E. Fujita, Chem. Commun., 2013, 49, 8896–8909.
6. X. F. Yang, A. Q. Wang, B. T. Qiao, J. Li, J. Y. Liu and T. Zhang, Acc. Chem. Res., 2013, 46, 1740–1748.
7. C. Y. He and J. Z. Tao, Chem. Commun., 2015, 51, 8323–8325.
8. J. G. Choi, J. R. Brenner, C. W. Colling, B. G. Demczyk, J. L. Dumming and L. T. Thompson, Catal. Today, 1992, 15, 201–222.
9. X. Y. Zhang, X. Chen, K. J. Zhang, S. P. Pang, X. H. Zhou, H. X. Xu, S. M. Dong, P. X. Han, Z. Y. Zhang, C. J. Zhang and G. L. Cui, J. Mater. Chem. A, 2013, 1, 3340–3346.
10. M. S. Balogun, M. H. Yu, Y. C. Huang, C. Li, P. P. Fang, Y. Liu, X. H. Lu and Y. X. Tong, Nano Energy, 2015, 11, 348–355.
11. (a) Y. Y. Liang, Y. G. Li, H. L. Wang and H. J. Dai, J. Am. Chem. Soc., 2013, 135, 2013–2036; (b) N. Fechler, G. A. Tiruye, R. Marcilla and A. Markus, RSC Adv., 2014, 4, 26981–26989.
12. M. M. Liu, R. Z. Zhang and W. Chen, Chem. Rev., 2014, 114, 5117–5160.
13. H. B. Chu, L. Wei, R. L. Cui, J. Y. Wang and Y. Li, Coord. Chem. Rev., 2010, 254, 1117–1134.
14. (a) H. J. Yin, H. J. Tang, D. Wang, Y. Gao and Z. Y. Tang, ACS Nano, 2012, 6, 8288–8297; (b) B. Y. Xia, Y. Yan, X. Wang and X. W. Lou, Mater. Horiz., 2014, 1, 379–399.
15. R. S. Ningthoujam and N. S. Gajbhiye, Prog. Mater. Sci., 2015, 70, 50–154.
16. Special volume on “Polyoxometalates”, Chem. Rev., 1998, 98, 1–398.
17. (a) H. J. Yan, C. G. Tian, L. Sun, B. Wang, L. Wang, J. Yin, A. P. Wu and H. G. Fu, Energy Environ. Sci., 2014, 7, 1939–1949; (b) Y. Liu, H. J. Yan, X. G. Zhou, M. X. Li and H. G. Fu, Chem.–Eur. J., 2015, 21, 18345–18355.
18. J. Cuan and B. Yan, Microporous Mesoporous Mater., 2014, 183, 9–16.
19. Y. X. Ding, X. Y. Sun, L. Y. Zhang, S. J. Mao, Z. L. Xie, Z. W. Liu and D. S. Su, Angew. Chem., Int. Ed., 2015, 54, 231–235.
20. W. Zhang, Z. Y. Wu, H. L. Jiang and S. H. Yu, J. Am. Chem. Soc., 2014, 136, 14385–14388.
21. J. Zhang, H. L. Zou, Q. Qing, Y. L. Yang, Q. W. Li, Z. F. Liu, X. Y. Guo and Z. L. Du, J. Phys. Chem. B, 2003, 107, 3712–3718.
22. X. Lu and T. Imae, J. Phys. Chem. C, 2007, 111, 2416–2420.
23. K. Okaya, H. Yano, H. Uchida and M. Watanabe, ACS Appl. Mater. Interfaces, 2010, 2, 888–895.
24. Y. Z. Dong and J. H. Li, Chem. Commun., 2015, 51, 572–574.
25. (a) J. Song, G. R. Li, F. Y. Xiong and X. P. Gao, J. Mater. Chem., 2012, 22, 20580–20585; (b) S. M. Abbas, Z. U. Rehman, U. A. Rana, S. U.-D. Khan, Z. Iqbal and N. Ahmad, Electrochim. Acta, 2016, 190, 988–996.
26. N. L. Torad, M. Hu, Y. Kamachi, K. Takai, M. Imura, M. Naito and Y. Yamauchi, Chem. Commun., 2013, 49, 2521–2523.
27. (a) P. Xiao, X. M. Ge, H. B. Wang, Z. L. Liu, A. Fisher and X. Wang, Adv. Funct. Mater., 2015, 25, 1520–1526; (b) W. F. Chen, K. Sasaki, C. Ma, A. I. Frenkel, N. Marinkovic, J. T. Muckerman, Y. M. Zhu and R. R. Adzic, Angew. Chem., Int. Ed., 2012, 51, 6131–6135.
28. J. K. Nørskov, T. Bligaard, A. Logadottir, J. R. Kitchin, J. G. Chen, S. Pandelov and U. J. Stimming, J. Electrochem. Soc., 2005, 152, 23–26.
29. Y. Zhao, K. Kamiya, K. Hashimoto and S. Nakanishi, Angew. Chem., Int. Ed., 2013, 52, 13638–13641.
30. (a) J. Greeley, T. F. Jaramillo, J. Boned, I. Chorkendorff and J. K. Nørskov, Nat. Mater., 2006, 5, 909–913; (b) C. Y. Tang, W. Wang, A. K. Sun, C. K. Qi, D. Z. Zhang, Z. Z. Wu and D. Z. Wang, ACS Catal., 2015, 5, 6956–6963.
31. (a) Y. G. Li, H. L. Wang, L. M. Xie, Y. Y. Liang, G. S. Hong and H. J. Dai, J. Am. Chem. Soc., 2011, 133, 7296–7299; (b) X. Y. Chia, A. Y. S. Eng, A. Ambrosi, S. M. Tan and M. Pumera, Chem. Rev., 2015, 115, 11941–11966; (c) W. Qiao, S. Yan, X. Y. Song, X. Zhang, Y. Sun, X. Chen, W. Zhong and Y. W. Du, RSC Adv., 2015, 5, 97696–97701.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra27490g

---