Highly selective catalytic conversion of phenols to aromatic hydrocarbons on CoS₂/MoS₂ synthesized using a two step hydrothermal method

Abstract

CoS₂/MoS₂ catalysts were prepared using a two-step hydrothermal procedure for the first time, i.e., MoS₂ was synthesized and then CoS₂ was prepared and deposited on the surface of the MoS₂. The characterization results presented that CoS₂ and MoS₂ are separated in the resultant catalysts and the surface area of CoS₂/MoS₂ was much higher than that of Co–Mo–S prepared using a one step method. In the hydrodeoxygenation (HDO) of p-cresol, the presence of CoS₂ enhanced the conversion, but excessive CoS₂ on the surface of the MoS₂ reduced its activity. With an appropriate amount of CoS₂, the catalyst presented an unprecedented HDO activity and direct deoxygenation (DDO) selectivity: 98% deoxygenation degree with a selectivity of 99% toluene at 250 °C for 1 h. This CoS₂/MoS₂ catalyst also exhibited high DDO activity for other phenolic monomers, which minimized hydrogen consumption and improved the economic efficiency.

---

The rapid growth in the global energy demand, steady decline in fossil fuel reserves and worldwide serious environmental problems motivated us to look for other renewable sources.¹ Bio-oil, derived from biomass by liquefaction or pyrolysis, is an interesting alternative for supplementing fuels.² However, lignocellulosic-derived bio-oil contained a relatively high amount of phenols, furans, esters and ketones, which lead to some detrimental properties such as a low heating value, high viscosity, and chemical and thermal instability.³ This bio-oil required its oxygen content to be lowered if it was used as a supplementary fuel. Hydrodeoxygenation (HDO) was one of the most common and efficient technologies to selectively remove oxygen in the form of water in the presence of hydrogen, and the deoxygenation degree and product selectivity depended on the selected catalysts.⁴

Phenols, as important lignin based monomeric model substrates and recalcitrant oxygenated species in bio-oil, were always chosen to study the activity of the catalysts and the HDO reaction mechanism.^5–11 It has been inferred that the oxygen in phenols is removed via two parallel pathways: hydrogenation–dehydration (HYD) and direct deoxygenation (DDO).^4,5 The former involves saturation of the aromatic ring of phenols, which resulted in a high hydrogen consumption and a decrease in the octane number after HDO.¹² Therefore, allowing C–O bond scission without substantial phenyl ring saturation was an economical and favorable process for the HDO of phenols.¹³

Based on that Fe-based catalysts were effective for activation of the aromatic carbon–oxygen bonds of phenols, Sun et al.¹⁴ studied the activity of Fe/C in the gas-phase HDO of guaiacol and confirmed its high DDO activity but a low HDO activity. After adding Pd to form Pd–Fe/C, the HDO activity was enhanced markedly: 83.2% yield of aromatics at 450 °C. In addition, MoS₂ was established to be a good HDO catalyst and its DDO activity was further improved by adding Co.¹⁵ We obtained a 100% p-cresol conversion with a selectivity of 92.2% toluene with a Co–Mo sulfide catalyst at 275 °C for 4 h.¹⁶ These results demonstrated that Co–Mo sulfide is a potential catalyst for the HDO of phenols, but its activity was structure-dependent, which could be improved by optimizing the preparation method.

Until now, various technologies have been developed for the preparation of Mo based sulfide catalysts.^11,15,17,18 In the previous literature, a Co–Mo–S phase had almost been adopted to elucidate the HDO mechanism when a Co promoter is added into the MoS₂ catalysts.^6,15 However, this phase was unstable and would decompose during the reactions, and was just regarded as the precursor of the real active phase.¹⁹ Fortunately, Gil-Llambías et al.²⁰ had evidenced the synergism between separated CoS_x and MoS₂. Recently, it was concluded that the catalytic activity of Co–Mo sulfide was directly proportional to increase of the contact surface area between the MoS₂ and Co₉S₈ phases.²¹ In a uniformly simultaneous precipitation procedure, Co sulfides inserted into MoS₂ and were not always accessible, leading to the prepared catalyst having a low surface area and much of the Co promoter could not play its role,²² which in turn reduced the overall activity. Hence, to emphasize the effect of Co and enhance the HDO activity, CoS₂/MoS₂ catalysts are synthesized using a two-step hydrothermal method for the first time and applied to the HDO of phenols.

In a typical procedure for the synthesis of CoS₂/MoS₂, ammonium heptamolybdate (2.3 g) and thiourea (3.0 g) were dissolved in 150 mL of ultra-pure water and the pH value was adjusted to 0.9 using hydrochloric acid. This mixed solution was added into a reactor, which was sealed and heated at 200 °C for 12 h. Then the reactor was cooled to room temperature and opened, and 30 mL of a solution containing cobalt nitrate and thiourea was added. The reactor was sealed and heated at 200 °C for 12 h again. After the reaction, the black precipitate was collected and washed with water and ethanol, and dried under vacuum at 50 °C for 8 hours. The resultant catalysts were denoted as Co–Mo-X, where X represents the initial molar ratio of Co/Mo. The HDO activity tests were carried out in a 100 mL sealed autoclave. The prepared catalyst (0.03 g), p-cresol (4.8 g) and dodecane (28.3 g) were placed into the autoclave. Air was evacuated by pressurization–depressurization cycles with nitrogen and subsequently with hydrogen. The system was heated to the desired temperature, then pressurized with hydrogen and the stirring speed adjusted to 900 rpm. During the reaction, liquid samples were withdrawn from the reactor and analysed. Conversion = (the amount of aromatic-ring change during the reaction/total amount of aromatic-ring) × 100%; selectivity = (C atoms in each product/total C atoms in the products) × 100%; deoxygenation degree (D.D., wt%) is defined as [1 − oxygen content in the final organic compounds/total oxygen content in the initial material] × 100%.

As shown in Fig. 1, the XRD pattern of Mo–S presented four diffraction peaks at 2θ = 14°, 33°, 39° and 59°, matching with the (002), (100), (103) and (110) crystal planes of hexagonal MoS₂ (JCPDS Card No. 37-1492),²³ respectively, suggesting that a MoS₂ structure had been completely formed after hydrothermal reaction for 12 h. With increment of the Co sulfide, there appeared some diffraction peaks at 2θ = 28°, 32°, 36°, 39°, 46° and 55°, corresponding to crystalline CoS₂ (JCPDS Card No. 41-1471).²⁴ These results indicate that the Co existed in the form of a CoS₂ phase after the second hydrothermal synthesis. Previous studies have claimed that the promoter Co was incorporated into the MoS₂ structure to form a Co–Mo–S active phase and the excess Co presented as a Co₉S₈ phase,^15,21 but this is different from our results that the CoS₂ and MoS₂ were separated and no Co₉S₈ phase was detected in the resultant catalysts.

Fig. 1 XRD patterns of the Mo–S and CoS₂/MoS₂ catalysts.

The separated CoS₂ and MoS₂ phases were further confirmed from XPS characterization of the electron bonding energy of each element. As presented in Fig. S1 (ESI†), three peaks at 226.6 eV, 229.3 eV and 232.5 eV were present in the Mo 3d level of each catalyst, which were assigned to the S 2p contribution for S²⁻, Mo 3d_5/2 and Mo 3d_3/2 of MoS₂,^25,26 respectively. Two peaks were located at 162.2 eV and 163.6 eV, matching well with the S 2p_3/2 and S 2p_1/2 in MoS₂ and CoS₂ phases,^25,26 respectively. A previous study had claimed that the binding energy of Co 2p_3/2 in Co₉S₈ ranged between 777.8 eV and 778.1 eV, while that for the Co–Mo–S phase was 778.6 eV.²⁷ However, there was only one main peak displayed, appearing at 779.0 eV, in the Co 2p level spectra, corresponding to CoS₂. In addition, another peak at a higher binding energy (782.1 eV) was assigned to Co oxides.^21,28 These results demonstrate that the Co did not locate in the MoS₂ slab to form a Co–Mo–S phase and that it presented as a CoS₂ phase rather than a Co₉S₈ phase.

Generally, CoS₂ was deposited on the surface of the MoS₂, resulting in an enrichment of Co and a high Co/Mo molar ratio on the catalyst surface. However, the XPS results showed that the Co/Mo molar ratio on the surface of Co–Mo-0.05, Co–Mo-0.1, Co–Mo-0.2, Co–Mo-0.3 and Co–Mo-0.4 was 0.03, 0.05, 0.09, 0.16 and 0.24, respectively, which is much lower than the corresponding ratio in the precursor solution. These results suggest that much of the CoS₂ was not dispersed uniformly but aggregated together on the catalyst surface. The SEM images provided direct evidence (Fig. S2, ESI†). Mo–S presented a loose flower-like structure composed of nanosheets in a random orientation. Co–Mo-0.2 and Co–Mo-0.3 displayed some spinel-like particles, coexisting with the flower-like morphology, which were attributed to CoS₂. After adding an excessive amount of Co, e.g., Co–Mo-0.4, the flower-like MoS₂ was not obvious and there appeared large spinel CoS₂ particles, suggesting more CoS₂ and less MoS₂ active sites on the catalyst surface. These results also demonstrate that the MoS₂ and CoS₂ phases were separated and that the CoS₂ only deposited on the surface of the MoS₂ and tended to aggregate together.

HRTEM was used to investigate the microstructure of the Mo–S and CoS₂/MoS₂ catalysts, which could provide a route from direct observation to determination of the interactions between the CoS₂ and MoS₂ sulfide phases.²¹ Fig. 2 shows that Mo–S has a highly disordered organization of stacked layers with a fringe spacing of 0.63 nm, typifying the d₀₀₂ interplanar spacing of MoS₂.^23,29 After the deposition of Co sulfide on the MoS₂, such as for Co–Mo-0.3 and Co–Mo-0.4, another group of lattice fringes with an interplanar spacing of 0.25 nm was observed, corresponding to the (210) plane of cubic CoS₂ pyramids,³⁰ and these fringes increased with the amount of Co sulfide. From these HRTEM images, it is clearly shown that CoS₂ covered and spread over the MoS₂ slab substrate, resulting in that some MoS₂ slabs were invisible or some HRTEM zones presented both MoS₂ and CoS₂ phases.

Fig. 2 HRTEM images of the Mo–S and CoS₂/MoS₂ catalysts.

The surface area, pore volume, and pore size distribution of the Mo–S and CoS₂/MoS₂ catalysts are listed in Table 1. Mo–S had a surface area of 183 m² g⁻¹ with a pore volume of 0.8 cm³ g⁻¹ and a bi-modal pore distribution centered at 2.3 nm and 10.7 nm. With increment of the Co/Mo molar ratio, the surface area gradually decreased to 78 m² g⁻¹ and the small pores dramatically decreased (Fig. S3, ESI†). This resulted from the deposition of CoS₂ on the surface or in the pore channels of MoS₂. For the one-step synthesis, Co and Mo were dispersed uniformly in the Co–Mo bimetallic sulfide and consequently the obtained catalyst had a very small surface area (lower than 10 m² g⁻¹). However, in this study, the MoS₂ produced in the first step had a large surface area, and when CoS₂ was produced and deposited on the surface of the MoS₂ in the second step, only a small part of the CoS₂ particles inserted into the MoS₂ pores, preventing a decrease of the surface area and exposing more active sites for the reaction, which was expected to enhance the catalytic activity.

Table 1 The structural properties of the Mo–S and CoS₂/MoS₂ catalysts and their activities for the HDO of p-cresol^a

Catalyst	Mo–S	Co–Mo-0.05	Co–Mo-0.1	Co–Mo-0.2	Co–Mo-0.3	Co–Mo-0.4	CoS2 + MoS2^b
a Reaction conditions: 0.03 g catalyst, 4.8 g p-cresol, 28.5 g dodecane, 4.0 MPa hydrogen pressure, temperature 250 °C and time 1 h.b The catalysts was composed by 0.01 g CoS2 and 0.03 g MoS2.
Surface area (m^2 g^−1)	183	165	149	105	91	78	—
Pore volume (cm^3 g^−1)	0.8	0.7	0.7	0.6	0.6	0.5	—
Pore size (nm)	2.3, 10.7	2.2, 11.6	2.2, 12.8	2.4, 16.5	2.1, 16.0	19.2	—
Conversion (mol%)	18	40	53	77	98	93	19
Product selectivity (mol%)
Methylcyclohexane	12	<1	1	<1	<1	<1	14
3-Methylcyclohexene	8	2	<1	<1	<1	<1	7
Toluene	80	98	98	99	99	99	79
D.D. (wt%)	16	36	49	74	98	92	17

---

A comparison for the Mo–S and CoS₂/MoS₂ catalysts of the activity and product distribution in the HDO of p-cresol is shown in Table 1. The conversion over Mo–S was 18% with a selectivity of 80% toluene and a deoxygenation degree of 16% at 250 °C for 1 h. In the presence of CoS₂, the products remained unchanged, but both the conversion and toluene selectivity were increased greatly, suggesting that CoS₂ was beneficial for significantly enhancing the HDO activity and DDO activity, which is in line with previous investigations.^6,15 With increment of the CoS₂, the conversion increased firstly and then decreased but the toluene selectivity increased continually. When the Co/Mo molar ratio of the precursor solution was increased to 0.3, the catalyst exhibited the highest HDO activity. The deoxygenation degree reached 98%, which is 6-fold higher than that over Mo–S. In contrast, at a high Co/Mo molar ratio, e.g., for Co–Mo-0.4, because of excessive coverage of the MoS₂ by CoS₂, the conversion decreased to 93%. Consequently, an appropriate CoS₂ on MoS₂ surface maximized the HDO activity. We also tested the HDO activity of physically mixed CoS₂ and MoS₂ under the same reaction conditions. The conversion was 19% with a selectivity of 79% toluene. This meant that large CoS₂ particles had little promoting effect on the HDO activity of MoS₂ and only CoS₂ with a small size dispersed on a MoS₂ surface could enhance the HDO activity. Hence, the changes in p-cresol conversion versus the Co/Mo molar ratio on the catalyst surface obtained from XPS data were summarized (Fig. S4, ESI†). At a low CoS₂ content, the conversion was linearly related to the Co/Mo molar ratio at the catalyst surface. Therefore, CoS₂ should be dispersed on the MoS₂ surface uniformly and a monolayer dispersion would maximize the HDO activity.

Fig. 3 presents curves of the concentration changes of p-cresol and the products versus reaction time for Co–Mo-0.3 at 225 °C. With increase of the time, the p-cresol concentration decreased while the toluene concentration increased. During the whole reaction, both the methylcyclohexane and 3-methylcyclohexene concentrations were very low. Obviously, the dominant reaction route for the HDO of p-cresol on Co–Mo-0.3 was DDO. After reaction for 3 h, the conversion was higher than 99% with a selectivity of 99% toluene. These results suggested that the Co–Mo sulfide catalysts prepared using this new method had a very high HDO activity and direct deoxygenation activity. The above characterization results demonstrated that CoS₂ and MoS₂ phases but no Co–Mo–S phase were present in the CoS₂/MoS₂ catalysts. Consequently, a remote control model was reasonable for explaining the HDO reaction mechanism with these CoS₂/MoS₂ catalysts,^20,31,32 as presented in Fig. 4. CoS₂ acted as a donor phase while MoS₂ acted as an acceptor phase. Hydrogen was adsorbed on the CoS₂ active sites, converted into spillover hydrogen and migrated to MoS₂ active sites for the HDO reaction. It has been reported that p-cresol molecules were adsorbed on the catalyst surface via two modes: vertical adsorption and co-plane adsorption, which implied a DDO route and HYD route.¹⁷ However, according to the reaction equation, 6 mol of spillover hydrogen is required for the hydrogenation of 1 mol of p-cresol to 4-methylcyclohexanol, which is 3-fold larger than that for the deoxygenation of 1 mol of p-cresol to toluene. Due to the low activity of CoS₂ for the creation of spillover hydrogen, obtaining the hydrogen for saturation of the phenyl ring became difficult with the increase of CoS₂ on the MoS₂ surface and then the HYD route was inhibited, resulting in a very high toluene selectivity.

Fig. 3 The curves for concentration changes of p-cresol and the products versus reaction time with Co–Mo-0.3 at 225 °C.

Fig. 4 The proposed reaction mechanism for p-cresol HDO with the CoS₂/MoS₂ catalysts.

The hydrogen pressure has a great effect on the conversion in the HDO of phenols, especially on the product distribution.³³ Generally, a high hydrogen pressure enhances the hydrogenation reaction. Fig. 5 shows the HDO of p-cresol on Co–Mo-0.3 under different hydrogen pressures and reaction temperatures. After reaction for 1 h, the conversion was 86% with a selectivity of 99% toluene under 250 °C and a 2 MPa hydrogen pressure. When the hydrogen pressure was raised to 5.0 MPa, the conversion increased to 100% and the toluene selectivity was still higher than 99%. These results suggest that a high hydrogen pressure was favorable for the p-cresol conversion but had little effect on the toluene selectivity. As presented in Fig. 4, hydrogen is activated into spillover hydrogen on CoS₂, but the CoS₂ had a very low activation ability. Even though the hydrogen pressure was very high, the spillover hydrogen created on Co–Mo-0.3 was still insufficient to meet the spillover hydrogen requirements for the saturation of the phenyl rings. This suggested that an increase of hydrogen consumption with the hydrogen pressure does not need to be considered for the HDO of p-cresol on Co–Mo-0.3 at a high hydrogen pressure, which is industrially desirable.

Fig. 5 The conversion and product distribution for the HDO of p-cresol with Co–Mo-0.3 under different hydrogen pressures (a) and reaction temperatures (b) for 1 h.

However, a low temperature was beneficial for creating spillover hydrogen on the catalyst surface because of a high hydrogen solubility in the solvent in this case,³⁴ which supplied more hydrogen for the HDO reaction and enhanced the HYD products selectivity. As shown in Fig. 5(b), the conversion dropped from 98% to 56% when the reaction temperature was decreased from 250 °C to 225 °C, but the toluene selectivity changed very little and was still higher than 99%. This was also attributed to the low hydrogen activation ability of Co–Mo-0.3. Although the hydrogen concentration in the solvent at low temperature was higher than that at high temperature, the created spillover hydrogen amount was still insufficient and the reaction presented a very high toluene selectivity.

The high HDO activity of the CoS₂/MoS₂ catalyst was attributed to the following reasons. Firstly, the CoS₂/MoS₂ catalyst had a large surface area and supplied sufficient active sites for p-cresol absorption. Compared with our previous study¹⁶ in which the p-cresol conversion was 94% with a toluene selectivity of 90% with a Co–Mo sulfide catalyst prepared using a one-step hydrothermal method at 250 °C for 6 h with a p-cresol/catalyst weight ratio of 50, the surface area of the Co–Mo-0.3 catalyst in this study is high at 91 m² g⁻¹, which exposed more active sites for the reaction and exhibited a higher HDO activity. Secondly, the spillover hydrogen could promote metal–sulfur bond scission,¹⁹ creating more coordinatively unsaturated sites for the adsorption of p-cresol molecules. If the spillover hydrogen on the CoS₂ is sufficient, the DDO reaction would proceed smoothly and enhance the deoxygenation degree. Although more spillover hydrogen was provided for the HDO reaction when the CoS₂ amount was excessive, this was still insufficient for the HYD route and presented an increase in the toluene selectivity. In this case, the substantial coverage of CoS₂ on the MoS₂ active sites prevented the adsorption of p-cresol molecules and resulted in a decrease of the deoxygenation degree. A comparison with other catalysts (Table S1, ESI†) obviously showed that the CoS₂/MoS₂ catalyst prepared using a two-step hydrothermal method presented an unprecedented HDO activity and DDO selectivity: the deoxygenation degree and toluene selectivity reached 98% and 99% for the HDO of p-cresol at 250 °C for 1 h, respectively.

The reusability of Co–Mo-0.3 for the HDO of p-cresol is shown in Fig. 6(a). Because catalyst loss was inevitable during the reaction and the separation, more parallel reactions were carried out and more catalyst was recovered to compensate for the lost catalyst. It could be observed that both the p-cresol conversion and toluene selectivity were still higher than 99% after cycle 3. These results suggest that the Co–Mo-0.3 catalyst had a good stability, but this did not mean that there was no deactivation of this catalyst during the HDO reaction. A previous study has reported that a sulfide catalyst underwent a continuous sulfur–oxygen exchange at its edge sites.³⁵ Hence, the recovered catalyst after each cycle of the reaction was characterized using XRD, as presented in Fig. 6(b). We noticed that the peaks of MoS₂ became sharper after the reaction and that the intensity of the (002) peak of the MoS₂ edge decreased but the intensity of the others changed very little with the cycle number. These results indicated that an enlarged MoS₂ crystallite size and continuous destruction of the (002) plane might be the reasons for the catalyst deactivation during the HDO reaction, and this needed to be further confirmed.

Fig. 6 (a) The reusability of Co–Mo-0.3 in the HDO of p-cresol and (b) the XRD patterns of fresh Co–Mo-0.3 and spent Co–Mo-0.3 after different reaction cycles.

To demonstrate the versatility of the CoS₂/MoS₂ catalyst, several monomeric phenol derivatives were investigated using Co–Mo-0.3 as the optimized catalyst and the results are shown in Table 2. All the selected substrates were efficiently converted into aromatics, with a conversion of >99% and a selectivity for the aromatics of >98% under the investigated conditions, except for 4-methoxyphenol and guaiacol. These results indicate that CoS₂/MoS₂ has a high activity for the direct scission of C–O bonds in phenolic hydroxyl groups. Moreover, in the previous literature^5,36–38 it has been reported that full conversions of phenols with high selectivities (>90%) toward alkanes were obtained using noble metal catalysts, where the HDO reaction temperature was reduced, but these reactions consumed a lot of hydrogen and the corresponding octane value of the products was decreased. In this study, due to the special structure of the CoS₂/MoS₂ catalyst, DDO was the predominant route in the HDO of phenols on these catalysts, which minimized the consumption of precious hydrogen. These CoS₂/MoS₂ catalysts also exhibited a high activity in the hydrodesulfurization (HDS) of benzothiophene (see Table S2, ESI†). In the presence of CoS₂, e.g., with Co–Mo-0.2, the conversion and ethylbenzene selectivity were raised to 98% and 100% from reaction at 300 °C for 3 h, respectively.

Table 2 HDO of phenol derivatives with Co–Mo-0.3^a

Substrate	Weight (g)	Conversion (%)	Selectivity (%)
a Reaction conditions: 0.1 g catalyst, total weight 33.3 g, H2 pressure 4.0 MPa, temperature 300 °C and reaction time 1 h.
C6 backbone
	4.2	100	99	1
	2.5	100	99	1
	2.8	100	85	1
	2.8	99	84	1
C7 backbone
	4.8	100	99	1
	4.8	100	99	1
C8 backbone
	5.4	100	99	1

---

Conclusions

A new method was developed for the synthesis of Co–Mo sulfide catalysts with high activity. The characterization results showed that the prepared catalysts were composed of separate CoS₂ and MoS₂ phases rather than a Co–Mo–S phase. The CoS₂/MoS₂ catalysts exhibited unprecedented HDO activity and DDO selectivity: the deoxygenation degree reached 97.8% with a toluene selectivity of 99.2% from reaction at 250 °C for 1 h, which was attributed to the synergistic effects between the CoS₂ and MoS₂, a uniform dispersion of CoS₂ on the MoS₂ surface and a large surface area. The conversion of p-cresol increased with the reaction temperature and hydrogen pressure, but this had little effect on the toluene selectivity. These catalysts also exhibited a high direct DDO activity for other diverse substituted phenolic monomers and a high HDS activity for benzothiophene, which minimized the consumption of precious hydrogen and exhibited a high potential for superiority over other catalysts.

Acknowledgements

This research was supported by the National Natural Science Foundation of China (No. 21306159, 21376202) and the Scientific Research Fund of Hunan Provincial Education Department (15B234).

Notes and references

1. C. Liu, H. Wang, A. M. Karim, J. Sun and Y. Wang, Chem. Soc. Rev., 2014, 43, 7594–7623.
2. C. Li, X. Zhao, A. Wang, G. W. Huber and T. Zhang, Chem. Rev., 2015, 115, 11559–11624.
3. J. Zakzeski, P. C. A. Bruijnincx, A. L. Jongerius and B. M. Weckhuysen, Chem. Rev., 2010, 110, 3552–3599.
4. M. Saidi, F. Samimi, D. Karimipourfard, T. Nimmanwudipong, B. C. Gates and M. R. Rahimpour, Energy Environ. Sci., 2014, 7, 103–129.
5. C. Zhao, Y. Kou, A. A. Lemonidou, X. Li and J. A. Lercher, Angew. Chem., Int. Ed., 2009, 48, 3987–3990.
6. V. N. Bui, D. Laurenti, P. Afanasiev and C. Geantet, Appl. Catal., B, 2011, 101, 239–245.
7. Y. Hong, H. Zhang, J. Sun, K. M. Ayman, A. J. R. Hensley, M. Gu, M. H. Engelhard, J.-S. McEwen and Y. Wang, ACS Catal., 2014, 4, 3335–3345.
8. W. Wang, Z. Qiao, K. Zhang, P. Liu, Y. Yang and K. Wu, RSC Adv., 2014, 4, 37288–37295.
9. S. M. Schimming, O. D. LaMont, M. König, A. K. Rogers, A. D. D’Amico, M. M. Yung and C. Sievers, ChemSusChem, 2015, 8, 2073–2083.
10. M. Shetty, K. Murugappan, T. Prasomsri, W. H. Green and Y. Román-Leshkov, J. Catal., 2015, 331, 86–97.
11. W. Wang, L. Li, K. Wu, G. Zhu, S. Tan, W. Li and Y. Yang, RSC Adv., 2015, 5, 61799–61807.
12. G. W. Huber, S. Iborra and A. Corma, Chem. Rev., 2006, 106, 4044–4098.
13. A. Popov, E. Kondratieva, L. Mariey, J. M. Goupil, J. El Fallah, J.-P. Gilson, A. Travert and F. Maugé, J. Catal., 2013, 297, 176–186.
14. J. Sun, A. M. Karim, H. Zhang, L. Kovarik, X. S. Li, A. J. Hensley, J.-S. McEwen and Y. Wang, J. Catal., 2013, 306, 47–57.
15. B. Yoosuk, D. Tumnantong and P. Prasassarakich, Chem. Eng. Sci., 2012, 79, 1–7.
16. W. Wang, K. Zhang, L. Li, K. Wu, P. Liu and Y. Yang, Ind. Eng. Chem. Res., 2014, 53, 19001–19009.
17. Y. Romero, F. Richard and S. Brunet, Appl. Catal., B, 2010, 98, 213–223.
18. S. Mukundan, M. Konarova, L. Atanda, Q. Ma and J. Beltramini, Catal. Sci. Technol., 2015, 5, 4422–4432.
19. P. Grange and X. Vanhaeren, Catal. Today, 1997, 36, 375–391.
20. J. Ojeda, N. Escalona, P. Baeza, M. Escudey and F. J. Gil-Llambías, Chem.Commun., 2003, 1608–1609,  10.1039/b301647c.
21. M. Ramos, G. Berhault, D. A. Ferrer, B. Torres and R. R. Chianelli, Catal. Sci. Technol., 2012, 2, 164–178.
22. P. Baeza, M. Villarroel, P. Ávila, A. López Agudo, B. Delmon and F. J. Gil-Llambías, Appl. Catal., A, 2006, 304, 109–115.
23. H. Zhang, Y. Li, T. Xu, J. Wang, Z. Huo, P. Wan and X. Sun, J. Mater. Chem. A, 2015, 3, 15020–15023.
24. J.-C. Xing, Y.-L. Zhu, Q.-W. Zhou, X.-D. Zheng and Q.-J. Jiao, Electrochim. Acta, 2014, 136, 550–556.
25. T. K. T. Ninh, L. Massin, D. Laurenti and M. Vrinat, Appl. Catal., A, 2011, 407, 29–39.
26. C. E. Scott, M. J. Perez-Zurita, L. A. Carbognani, H. Molero, G. Vitale, H. J. Guzmán and P. Pereira-Almao, Catal. Today, 2015, 250, 21–27.
27. A. D. Gandubert, E. Krebs, C. Legens, D. Costa, D. Guillaume and P. Raybaud, Catal. Today, 2008, 130, 149–159.
28. D. G. Castner and P. R. Watson, J. Phys. Chem., 1991, 95, 6617–6623.
29. M. Signorile, A. Damin, A. Budnyk, C. Lamberti, A. Puig-Molina, P. Beato and S. Bordiga, J. Catal., 2015, 328, 225–235.
30. H. Zhang, Y. Li, G. Zhang, P. Wan, T. Xu, X. Wu and X. Sun, Electrochim. Acta, 2014, 148, 170–174.
31. A. N. Varakin, P. A. Nikul’shin, A. A. Pimerzin, V. A. Sal’nikov and A. A. Pimerzin, Russ. J. Appl. Chem., 2013, 86, 718–726.
32. A. A. Pimerzin, P. A. Nikulshin, A. V. Mozhaev, A. A. Pimerzin and A. I. Lyashenko, Appl. Catal., B, 2015, 168–169, 396–407.
33. W. Wang, L. Li, K. Wu, K. Zhang, J. Jie and Y. Yang, Appl. Catal., A, 2015, 495, 8–16.
34. F. Faglioni and W. A. Goddard, J. Chem. Phys., 2005, 122, 014704–014718.
35. M. Badawi, J. F. Paul, S. Cristol, E. Payen, Y. Romero, F. Richard, S. Brunet, D. Lambert, X. Portier, A. Popov, E. Kondratieva, J. M. Goupil, J. El Fallah, J. P. Gilson, L. Mariey, A. Travert and F. Maugé, J. Catal., 2011, 282, 155–164.
36. W. Zhang, J. Chen, R. Liu, S. Wang, L. Chen and K. Li, ACS Sustainable Chem. Eng., 2013, 2, 683–691.
37. M.-Y. Chen, Y.-B. Huang, H. Pang, X.-X. Liu and Y. Fu, Green Chem., 2015, 17, 1710–1717.
38. G. Yao, G. Wu, W. Dai, N. Guan and L. Li, Fuel, 2015, 150, 175–183.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra27066a

---