Facile synthesis of ultrafine Ru nanocrystal supported N-doped graphene as an exceptional hydrogen evolution electrocatalyst in both alkaline and acidic media

Abstract

Since hydrogen is a clean and renewable energy source, the design of efficient and new catalysts for hydrogen evolution reaction (HER) has attracted significant attention. Ultrafine (∼2 nm) monodispersed Ru nanocrystals on N-doped graphene (Ru@NG) show Pt-like catalytic activity towards HER in both alkaline and acidic media with zero onset potential and better current density as compared to Pt/C. The HER performance strongly depends on the nanosize effect of Ru nanocrystals and their dispersion on NG. Transfer of electrons from Ru to carbon results in an electron-deficient metal centre and greatly enhances the HER activity. This new 4-d transition metal electrocatalyst has the potential to serve as an alternative to the Pt benchmark catalyst, which is more costly than Ru.

---

There is an urgency to find clean and renewable energy alternatives^1–4 to combat the problems associated with fossil fuel-based energy sources in the present world. In this regard, hydrogen, an abundant and renewable energy source, is a promising solution.⁵ Electrochemical hydrogen evolution reaction (HER) via water splitting is an attractive route for the generation of large scale high-purity hydrogen in a sustainable manner.³ However, a catalyst is required for HER to efficiently proceed with a low overpotential.^4–10 Pt-based catalysts currently form the state-of-the-art HER electrocatalysts due to their low reduction overpotential and fast reduction kinetics.^11,12 However, their acute scarcity and high-cost impede their large scale application. In this regard, nanostructures of transition metal chacogenides, carbides, phosphides, oxides, and their hybrids have been explored.^13–30 Despite their HER catalytic performance, these catalysts are significantly inferior to the Pt-based catalysts in terms of high overpotentials and low current density. Furthermore, the catalysts are active either in an acidic or in an alkaline medium. The realization of a Pt-free catalyst in both acid and alkaline media is of great importance with industrial relevance; however, only a few catalysts are active in both media.

Ru-based nanocrystals (NCs) are demonstrated to be of great importance in various heterogeneous catalytic reactions.^31–39 However, catalysts should be in the size range of 2–6 nm, and the surface area needs to be large for better catalytic activity; moreover, it is a challenge to optimize the size and morphology of the catalysts. Furthermore, the support for the catalysts and their interaction play a significant role in the stability as well as on the performance of the catalyst. In this regard, N-doped graphene (NG) has proven to be highly beneficial as a support for the catalysts. Although there are few reports on the chemical synthesis of Ru NCs, synthesis of ultra-small and uniform sized Ru NCs on an NG support still remains a challenge.

Herein, we demonstrated a physical technique for the in situ synthesis of self-assembled ultrafine Ru nanocrystals (∼2 nm) on N-doped graphene (Ru@NG) for the first time and explored their HER activity in both acid and alkaline media. Ru@NG shows exceptional HER catalytic activity in both acid and alkaline media, with zero onset potential like commercial precious Pt/C, and better current density. Thus, Ru@NG holds great promise as a new material for the generation of hydrogen from water.

The synthesis of NG-supported ultrafine Ru NCs (Ru@NG) is based on a one-pot strategy. The synthesis was accomplished through a very simple and convenient experimental setup (ESI-1†) that involved the pyrolysis of RuCl₃·xH₂O and dicyanamide (DCA) in a sealed quartz tube without any external gas flow. During the pyrolysis, Ru³⁺ ions were reduced to ultrafine Ru NCs, whereas the DCA decomposed and formed the NG support (Scheme 1). For simplicity, the as-prepared samples are named Ru@NG-x, which indicates the mass ratio (x) of DCA and the RuCl₃·xH₂O compound in the precursor (Table S-1†).

Scheme 1 Schematic for the synthesis of NG-supported ultrafine Ru NCs.

Transmission electron microscopy (TEM) and high resolution TEM (HRTEM) images of Ru@NG-10, as shown in Fig. 1(A and B), reveal that Ru NCs have sizes 4–5 nm (ESI-2†) and are decorated on NG. Fig. 1(C and D) show the TEM and HRTEM images of Ru@NG-4, which reveal highly densed ultrafine Ru NCs on NG, and the average size is ∼2 nm (ESI-3†). In the case of Ru@NG-2, TEM and HRTEM images, as shown in Fig. 1(E and F), reveal highly dense agglomerated structures on NG with a primary particle size of ∼2 nm. The HRTEM images of all the NCs show a d-spacing of 0.206 nm, corresponding to the (101) crystal plane of Ru. The TEM and HRTEM images of all the hybrid nanostructures obtained from different position also reveal the graphene-supported ultrafine Ru NCs (ESI-4–6†). The elemental mapping of the hybrid nanostructures (Ru@NG-4) clearly displays the uniformly anchored Ru NCs on NG (ESI-7†).

Fig. 1 TEM and HRTEM images of the NG-supported ultrafine Ru NCs synthesized via pyrolysing different ratios of DCA and Ru³⁺. (A and B) 10 : 1, (C and D) 4 : 1, and (E and F) 2 : 1.

The formation of Ru NCs on NG was characterized via X-ray diffraction (XRD) and X-ray photoelectron spectroscopy (XPS). Fig. 2(A) displays the XRD patterns of Ru@NG-10, Ru@NG-4, and Ru@NG-2. The XRD peaks at the 2θ values 38.6, 44.1, 58.4, 69.3, 78.4, and 85.9° correspond to (100), (101), (102), (110), (103), and (112) crystal planes of hexagonal crystal structures (JCDPS#06-0663) of Ru, respectively. All the XRD patterns exhibited broad peaks around 25° that corresponded to the (002) crystal plane of graphene. Overall, the XRD results reveal the formation of Ru@NG via pyrolysis. The sizes estimated from the XRD peak (2θ = 44.1°) using Scherrer formula were ∼4.5, 2.0, and 2.8 nm for Ru@NG-10, Ru@NG-4, and Ru@NG-2, respectively, in good agreement with the TEM results. Fig. 2(B) displays the XPS survey spectra, which reveals the formation of Ru NCs and N-doped carbon nanostructures. Fig. 2(C and D) display the high resolution XPS spectra (HRXPS) of Ru-3d and 3p of the Ru@NG-4 hybrid. As observed from Fig. 2(C), two distinct peaks at the binding energies of 279.9 and 284.1 eV were observed that corresponded to Ru 3d_5/2 and 3d_3/2 with a spin orbit coupling of 4.2 eV.⁴⁰ As is well-known, the area ratio between 3d_5/2 and 3d_3/2 is 3 : 2. In contrast, the Ru 3d_3/2 peak is broader as well as of higher intensity as compared to the d_5/2 peak due to overlapping with the C-1s (284.6) contribution. The de-convoluted spectrum of Ru 3d_3/2 shows clear contribution of C-1s as well. The binding energy of both Ru-3d peaks is 0.1 eV higher as compared to that of the bare Ru, whereas that of the C-1s peak is 0.1 eV lower as compared to that of the pure C-1s (284.6 eV). The peak shifting clearly signifies the strong interaction between the metallic Ru and carbon. The ratio between N–C and Ru is 76 : 24, which means that the Ru loading in the hybrid is ∼24 weight%, whereas Pt loading in Pt/C is 20 weight%. Similarly, the Ru-3p spectrum (Fig. 2(D)) shows two clear distinct peaks at 461.4 and 483.6 eV, corresponding to Ru-3p_3/2 and 3p_1/2, respectively, which is 0.4 eV higher compared to that of pure Ru-3p, signifying the interaction between metallic Ru and the underlying carbon layer.^41,42 Additional small peaks at 464.1 and 486.3 eV indicate that there is a charge transfer from metallic Ru to C, which basically leads to better catalytic activity towards HER performances. Fig. 2(E) shows the HRXPS spectrum of the de-convoluted N 1s spectrum of Ru@NG-4. It contains mainly three different types of N environment: pyridine-type (398.4 eV), pyrrolic-type (400 eV), and graphite-type (401.9 eV). The % of N doping is ∼7.8% in the carbon framework, where 32% N is pyridinic, 34% N is pyrrolic, and the remaining 32% N is graphitic. Raman spectroscopy is a very effective tool to characterize graphene. Raman spectra shown in Fig. 2(F) clearly reveal the signature of carbon materials. Energy dispersive X-ray spectroscopy (EDS) was used to determine the presence of carbon in the Ru-NG hybrids as well as Ru loading (ESI-8†). Overall, the hybrids consist of NG-supported ultrafine Ru NCs (Ru@NG). On the basis of the experimental results, we proposed a possible growth mechanism for the ultrafine Ru@NG. DCA vaporises above 260 °C, whereas RuCl₃·xH₂O decomposes above 500 °C, which indicates that DCA vaporizes first and form the NG support for Ru NCs. The cyanide group in DCA is a source for carbon and nitrogen that forms NG via a condensation process. It has been suggested that carbon nitride gaseous fragments are released (such as C₂N₂⁺, C₃N₂⁺, and C₃N₃⁺) above 700 °C, which provides both C and N sources to form NG.^43–45 Then, Ru³⁺ ions reduce in situ to ultrafine Ru NCs in the reducing environment. In this context, note that the formation of nitride or carbide is less probable as there is no source of nitrogen (there is no gas flow) and carbide formation requires high pressure.⁴⁶ The size and dispersion of NCs strongly depend on the ratio between Ru³⁺ and DCA. When the ratio of DCA/Ru³⁺ is 10 : 1 (Ru@NG-10), Ru NCs of 4–5 nm are obtained, whereas ultrafine ∼2 nm Ru NCs are formed when the ratio is 4 : 1, and agglomerated structures consisting of primary particles of ∼2 nm are obtained for DCA/Ru³⁺ = 2 : 1. With low amount of RuCl₃·xH₂O, the size of the primary particles is believed to be below 2 nm that sintered together to form bigger particles (Fig. 3(A and B)). As the amount of RuCl₃·xH₂O is increased, the particle size becomes around 2 nm, and the particles are thermodynamically stable, as evident from the TEM images (Fig. 3(C and D)). As the amount is increased further, 2 nm particles form and aggregate, as is evident from the TEM images (Fig. 3(E and F)). This is well-supported by the increase in size as the temperature is increased. For a ratio of 4 : 1, the particle size is 2 nm at a pyrolysis temperature of 900 °C, whereas it becomes 5 nm at 1000 °C (ESI-Fig. 13A and B†).

Fig. 2 (A and B) XRD patterns and XPS survey spectra of Ru@NG-x (x = 10, 4, and 2). (C–E) HRXPS of de-convoluted Ru-3d, 3p, and N-1s spectra Ru@NG-4 hybrid structure. (F) Raman spectra of all Ru@NG-x (x = 10, 4, and 2).

Fig. 3 (A) HER polarization curves of different catalysts in a 1 M KOH solution obtained at a scan rate of 10 mV s⁻¹ and a rotation speed of 1200 rpm (inset shows the onset current density). (B) HER potential to achieve 10 mA cm⁻² current density of different catalysts in an alkaline medium. (C) Tafel slope and (D) stability of Ru@NG-4 up-to 5000^th cycle in an alkaline medium at a scan rate of 100 mV s⁻¹ and a rotation speed of 1200 rpm. (E) CV curves of Ru@NG-4 hybrid in a 1 M KOH solution at different scan rates. (F) Difference in current density variation (ΔJ = J_a − J_c) at 0.55 V plotted against scan rate to a linear regression that enables the estimation of C_dl.

The electrocatalytic HER performance of Ru@NG was investigated in a N₂-purged 1 M KOH and 1 M H₂SO₄ solutions as an alkaline and acidic electrolyte, respectively, using a three electrode setup with a Pt wire as the counter electrode and Ag/AgCl as the reference electrode. A commercial Pt/C catalyst was used for comparison of the catalytic performances in both acid and alkaline media. All the Ru@NG hybrids were placed on a glassy carbon electrode (GCE). Fig. 3(A) displays the HER polarization curves for different catalysts in a 1 M KOH solution. It is evident from the inset of Fig. 3(A) that all Ru@NG catalysts show zero onset potential as in the case of Pt/C, indicating that all Ru@NG catalysts are very active towards HER in an alkaline medium. Fig. 3(B) displays the relative potential required to achieve a current density of 10 mA cm⁻², which is 40, 47, 128, and 80 mV for Ru@NG-4, Ru@NG-2, Ru@NG-10, and Pt/C catalysts, respectively. Ru@NG-4 shows the best HER performance as compared to other Ru@NG catalysts as well as Pt/C. It requires only 40 mV overpotential to obtain a 10 mA cm⁻² current density, which is half that required for Pt/C. Fig. 3(C) compares the Tafel slope of all the catalysts in an alkaline medium. The Tafel slope of the catalysts are 76, 82, 104, and 92 mV dec⁻¹ for Ru@NG-4, Ru@NG-2, Ru@NG-10, and Pt/C, respectively, which indicates high HER activity for Ru@NG-4. The Ru@NG-4 also shows very good stability towards HER. Fig. 3(D) displays that the catalyst is stable up-to 5000^th cycle with almost no change in the current density (retention of 95% current density). Chronoamperometry measurements also reveal the long-term stability of Ru@NG-4 up-to 10 h (ESI-9†) in an alkaline medium. When a bias was applied, large H₂ bubbles were clearly observed to have accumulated on the electrode surface (ESI-10†).

To shed more light on the catalytic activity of Ru@NG, the electrochemical double layers capacitance (C_dl), which is directly proportional to the electrochemical active surface area (ECSA), is evaluated. The ECSA was determined via cyclic voltammetry (CV) measurements in a non-redox regime in a 1 M KOH solution at different scan rates (Fig. 3(E)). It can be seen that C_dl of Ru@NG-4 is 44 mF cm⁻², whereas that of Ru@NG-2 and Ru@NG-10 is 37 and 18 mF cm⁻² (Fig. 3(F)), respectively. This result indicates that the electrochemically active sites of Ru@NG-4 towards HER are high due to ultrafine size and monodispersion of Ru NCs on NG. The large value of ECSA provides additional active sites and simplifies the charge transfer in the catalyst to the electrode,²⁷ which is evident from the variation of current density with C_dl (ESI-11†).

Fig. 4(A) shows the polarization curves of different catalysts in an acidic (1 M H₂SO₄) solution. It is evident from the inset of Fig. 3(E) that all the Ru@NG catalysts show zero onset-potential as in the case of Pt/C. Fig. 4(B) displays the potential required to achieve a current density of 10 mA cm⁻², which is 60, 74, 116, and 40 mV for Ru@NG-4, Ru@NG-2, Ru@NG-10, and Pt/C, respectively. Ru@NG-4 catalyst shows very promising HER activity in an acidic medium as well. Fig. 4(C) compares the Tafel slope of all the catalysts in an acid medium. The Tafel slope of all the catalysts are 41, 48, 62, and 30 mV dec⁻¹ for Ru@NG-4, Ru@NG-2, Ru@NG-10, and Pt/C, respectively, and η is zero. These results suggest that the recombination of two adsorbed H atoms is the rate limiting step and follow the Tafel mechanism. As evident from Fig. 4(D), the Ru@NG-4 catalyst is very stable up-to 5000^th cycle (retention of 97% current density) and far superior compared to Pt/C, as in the case of an alkaline medium. This is believed to be due to better adhesion of Ru NCs on an NG support. Overall, the catalytic activity of all the Ru@NG catalysts and Pt/C is presented in Table S-2.† The mass and C_dl (considered to be proportional to the electrochemical surface area) normalized HER current density of all Ru@NG is shown in ESI-12.† Note that Ru@NG-4 and Ru@NG-2 have almost the same C_dl normalized HER activity. The Ru@NG-4 hybrid materials synthesized at different temperatures also support the size/morphology-dependent HER activity (ESI-13–15†). Overall, HER activity strongly depends on NC size as well as the dispersion on N-G substrate. Ru@NG-4 with ultrafine (∼2 nm) Ru NCs on the NG shows excellent HER performances. Although Ru@NG-2 also consists of primary particles with ∼2 nm size, the catalytic performance is below Ru@NG-4 due to the agglomerated structures. The Ru@NG-10 catalyst shows very poor catalytic activity compared to the other two catalysts due to the bigger Ru NCs (4–5 nm) and lower ECSA. Electrochemical impedance spectroscopy (EIS) was carried out to evaluate the electrochemical behaviour under HER operation (ESI-16†). The charge transfer resistance (R_ct) of Ru@NG-4 and Ru@NG-2 was determined to be only 6 and 8 ohm, respectively, which indicates facile H⁺ reduction at the electrode–electrolyte interface.²³ The results suggest that that nicely dispersed and strongly attached ultrafine particles are essential for excellent HER performances. As compared to the literature-reported HER catalyst, Ru@NG-4 is one of the most exciting catalysts towards the HER performances (Table S-3†) and has true potential to replace the 25-times costlier precious Pt (Table S-4†).

Fig. 4 (A) HER polarization curves of different catalysts in a 1 M H₂SO₄ solution at a scan rate of 10 mV s⁻¹ and a rotation speed of 1200 rpm (inset shows the onset current density). (B) HER potential to achieve 10 mA cm⁻² current density of different catalysts in an acidic medium. (C) Tafel slope and (D) stability of Ru@NG-4 up-to 5000^th cycle in an alkaline medium at a scan rate of 100 mV s⁻¹ and a rotation speed of 1200 rpm.

To elucidate the potential of Ru@NG, an alkaline electrolyzer (1.0 M KOH) employing commercial C paper, commercial Pt/C, and our Ru@NG as cathode catalysts and RuO₂ as the anode catalyst was constructed. As shown in Fig. 4(A), the Pt/C–RuO₂ and Ru@NG-RuO₂ couples catalyze water electrolysis at an onset potential of ∼1.45 V with an overpotential (η) of only 220 mV. A current density at ∼50 mA cm⁻² was obtained at 1.67 V for both Ru@NG-RuO₂ and Pt/C–RuO₂; however, better current density for the former case was obtained beyond 1.67 V. For example, the current density was 100 mA cm⁻² at 1.84 V for Ru@NG-RuO₂, whereas it was 92 mA cm⁻² for Pt/C–RuO₂. Furthermore, the Ru@NC couple exhibited excellent stability, as manifested by the steady current change and nearly linear charge accumulation for 10 h of electrolysis at η = 360 mV and stable up-to 2500^th cycle (ESI-17†). Fig. 4(B) displays the optical image of the generation of H₂ and O₂via water electrolysis (video of water splitting has been provided at 1.67 V). To probe the role and the catalytic site of Ru@NG, the sample is treated with thiocyanate (SCN⁻) ions as it is widely known to poison metal-centred catalytic sites and influence the HER performance.⁴⁷ After the treatment of SCN⁻ into the basic electrolyte (50 mM), the overpotential increased from 0 to 10 mV; moreover, the overpotential also significantly increased from 44 mV to 74 mV to achieve 10 mA cm⁻² (Fig. 4(C)). This study suggests that the HER catalytic active sites of Ru are blocked by the SCN⁻ ions, effectively decreasing the HER performance (Fig. 5(C)).

Fig. 5 (A) Polarization curves of carbon paper-RuO₂, Pt/C–RuO₂, and Ru@NG-RuO₂ for overall water splitting in a 1 M KOH solution. (B) Optical image of H₂ (cathode) and O₂ (anode) bubble generation via overall water splitting. (C) HER polarization curves of Ru@NG-4 with and without 50 mM KSCN in a 1 M KOH solution, indicating that SCN⁻ ions strongly poison the Ru@NG catalyst. Insets show illustrations of Ru catalytic centres blocked by the SCN⁻ ions.

In conclusion, we synthesized Ru@NG via one-step pyrolysis for the first time and demonstrated it as a novel and emerging HER catalyst in both acid and alkaline media. The HER performances strongly depend on the size of Ru NCs and its dispersion on the graphene support; they outperform the HER activity of the precious Pt catalysts and have the potential to replace them. Overall, it is believed that this synthesis technique can be explored for various other hybrids with ultrafine NCs and their suitability in various applications.

Acknowledgements

The authors acknowledge the Department of Science and Technology (DST) and Council of Scientific and Industrial Research (CSIR) India for the financial support. The authors also acknowledge Chemical Science Division in IISc for providing access to FETEM facility.

References

1. A. J. Bard and M. A. Fox, Acc. Chem. Res., 1995, 28, 141–145.
2. C. Li and Y. Yamauchi, Phys. Chem. Chem. Phys., 2013, 15, 3490–3496.
3. H. A. Esfahani, M. Imura and Y. Yamauchi, Angew. Chem., Int. Ed., 2013, 52, 13611–13615.
4. C. Li, T. Sato and Y. Yamauchi, Angew. Chem., Int. Ed., 2013, 52, 8050–8053.
5. J. A. Turner, Science, 2004, 305, 972–974.
6. S. M. Dresselhaus and I. L. Thomas, Nature, 2001, 414, 332–337.
7. X. Ge, L. Chen, L. Zhang, Y. Wen, A. Hirata and M. Chen, Adv. Mater., 2014, 26, 3100–3104.
8. M. Zeng and Y. Li, J. Mater. Chem. A, 2015, 3, 14942–14962.
9. J. Kibsgaard, C. Tsai, K. Chan, J. D. Benck, J. K. Nørskov, F. A. Pedersen and T. F. Jaramillo, Energy Environ. Sci., 2015, 8, 3022–3029.
10. P. C. K. Vesborg, B. Seger and I. B. Chorkendorff, J. Phys. Chem. Lett., 2015, 6, 951–957.
11. J. R. McKone, E. L. Warren, M. J. Bierman, S. W. Boettcher, B. S. Brunschwig, N. S. Lewis and H. B. Gray, Energy Environ. Sci., 2011, 4, 3573–3583.
12. J. Greeley, T. F. Jaramillo, J. Bonde, I. Chorkendorff and J. K. Norskov, Nat. Mater., 2006, 5, 909–913.
13. J. Kibsgaard, Z. Chen, B. N. Reinecke and T. F. Jaramillo, Nat. Mater., 2012, 11, 963–969.
14. D. Voiry, M. Salehi, R. Silva, T. Fujita, M. Chen, T. Asefa, V. B. Shenoy, G. Eda and M. Chhowalla, Nano Lett., 2013, 13, 6222–6227.
15. M. Chhetri, S. Maitra, H. Chakraborty, U. V. Waghmare and C. N. R. Rao, Energy Environ. Sci., 2016, 9, 95–101.
16. 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.
17. L. Cheng, W. J. Huang, Q. F. Gong, C. H. Liu, Z. Liu, Y. G. Li and H. J. Dai, Angew. Chem., Int. Ed., 2014, 53, 7860–7863.
18. S. T. Hunt, T. Nimmanwudipong and Y. Román-Leshkov, Angew. Chem., Int. Ed., 2014, 53, 5131–5136.
19. E. J. Popczun, J. R. McKone, C. G. Read, A. J. Biacchi, A. M. Wiltrout, N. S. Lewis and R. E. Schaak, J. Am. Chem. Soc., 2013, 135, 9267–9270.
20. E. J. Popczun, C. G. Read, C. W. Roske, N. S. Lewis and R. E. Schaak, Angew. Chem., Int. Ed., 2014, 53, 5427–5430.
21. J. Q. Tian, Q. Liu, A. M. Asiri and X. P. Sun, J. Am. Chem. Soc., 2014, 136, 7587–7590.
22. J. Tian, Q. Liu, N. Cheng, A. M. Asiri and X. Sun, Angew. Chem., Int. Ed., 2014, 53, 9577–9581.
23. D. Das and K. K. Nanda, Nano energy, 2016, 30, 303–311.
24. Y. G. Li, H. L. Wang, L. M. Xie, Y. Y. Liang, G. S. Hong and H. Dai, J. Am. Chem. Soc., 2011, 133, 7296–7299.
25. X. Huang, Z. Y. Zeng and H. Zhang, Chem. Soc. Rev., 2013, 42, 1934–1946.
26. M. Gong, W. Zhou, M. C. Tsai, J. G. Zhou, M. Y. Guan, M. C. Lin, B. Zhang, Y. F. Hu, D. Y. Wang, J. Yang, S. J. Pennycook, B. J. Hwang and H. J. Dai, Nat. Commun., 2014, 5, 4695.
27. H. Jin, J. Wang, D. Su, Z. Wei, Z. Pang and Y. Wang, J. Am. Chem. Soc., 2015, 137, 2688–2694.
28. H. Zhu, J. Zhang, R. Yanzhang, M. Du, Q. Wang, G. Gao, J. Wu, G. Wu, M. Zhang, B. Liu, J. Yao and X. Zhang, Adv. Mater., 2015, 27, 4752–4759.
29. J. Wang, W. Cui, Q. Liu, Z. Xing, A. M. Asiri and X. Sun, Adv. Mater., 2016, 28, 215–230.
30. S. T. Hunt, M. Milina, Z. Wang and Y. R. Leshkov, Energy Environ. Sci., 2016, 9, 3290–3301.
31. K. Kusada, H. Kobayashi, T. Yamamoto, S. Matsumura, N. Sumi, K. Sato, K. Nagaoka, Y. Kubota and H. Kitagawa, J. Am. Chem. Soc., 2013, 135, 8016–8021.
32. J. Ohyama, T. Sato, Y. Yamamoto, S. Arai and A. Satsuma, J. Am. Chem. Soc., 2013, 135, 8016–8021.
33. H. Bielawa, O. Hinrichsen, A. Birkner and M. Muhler, Angew. Chem., Int. Ed., 2001, 40, 1061–1063.
34. C. S. Kellner and A. T. Bell, J. Catal., 1982, 75, 251.
35. J. Kang, S. Zhang, Q. Zhang and Y. Wang, Angew. Chem., Int. Ed., 2009, 48, 2565.
36. T. Mitsudome, Y. Takahashi, T. Mizugaki, K. Jitsukawa and K. Kaneda, Angew. Chem., Int. Ed., 2014, 53, 8348–8351.
37. R. A. Dagle, Y. Wang, G.-G. Xia, J. J. Strohm, J. Holladay and D. R. Palo, Appl. Catal., A, 2007, 326, 213.
38. T. Abe, M. Tanizawa, K. Watanabe and A. Taguchi, Energy Environ. Sci., 2009, 2, 315.
39. K. Urasaki, K. Endo, T. Takahiro, R. Kikuchi, T. Kojima and S. Satokawa, Top. Catal., 2010, 53, 707.
40. D. J. Morgan, Surf. Interface Anal., 2015, 47, 1072.
41. M. Tang, S. Mao, M. Li, Z. Wei, F. Xu, H. Li and Y. Wang, ACS Catal., 2015, 5, 3100–3107.
42. Y. Zheng, Y. Jiao, Y. Zhu, L. H. Li, Y. Han, Y. Chen, M. Jaroniec and S.-Z. Qiao, J. Am. Chem. Soc., 2016, 138, 16174–16181.
43. Z. Wen, S. Ci, Y. Hou and J. Chen, Angew. Chem., Int. Ed., 2014, 53, 6496–6500.
44. Y. Hou, Z. Wen, S. Cui, S. Ci, S. Mao and J. Chen, Adv. Funct. Mater., 2015, 25, 872–877.
45. B. K. Barman and K. K. Nanda, Green Chem., 2016, 18, 427–432.
46. E. Zhao, J. Wang and Z. Wu, J. Comput. Chem., 2010, 31, 2883–2888.
47. H. W. Liang, S. Bruller, R. Dong, J. Zhang, X. Feng and K. Mullen, Nat. Commun., 2015, 6, 7992.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c7se00153c

---