Enhancement of the selective hydrodesulfurization performance by adding cerium to CoMo/γ-Al₂O₃ catalysts

Abstract

A series of cerium-modified CoMo/Al₂O₃ hydrodesulfurization (HDS) catalysts were prepared to enhance catalytic activity and improve selectivity during the selective HDS of fluid catalytic cracking gasoline. The morphology properties of the catalysts with different amounts of CeO₂ were characterized by physical techniques and correlated with their catalytic performance. Results clearly indicated that Ce introduction weakened the interaction between support and metal and promoted sulfidation of the Mo species with the formation of more CoMoS active phases. This phenomenon induced a positive effect on the morphology of MoS₂ particles, which significantly influenced the catalytic performance. A linear correlation existed between hydrodesulfurization selectivity and Mo dispersion. By optimizing the amount of CeO₂ loading, the catalyst with 1.0 wt% CeO₂ incorporation exhibited the best HDS selectivity with the highest sulfidation, the largest number of CoMoS active phases and the lowest Mo dispersion.

---

1. Introduction

In recent years, with the growing consciousness for environmental protection, clean gasoline production has attracted considerable attention. The stringent restrictions from environmental regulations on emissions of motor vehicles call for the development of highly active catalysts for hydrodesulfurization (HDS). In the refinery petrol pool, fluid catalytic cracking (FCC) gasoline contains more than 90% sulfur and provides massive olefins, which are important contributors to the high octane number in products.^1,2 Olefins could undergo hydrogenation reaction during HDS. Therefore, the key to producing clean FCC gasoline is to eliminate the maximum amount of sulfur with minimum olefin saturation by selective HDS, which could preserve the octane number.^3–5 The most critical issue is the design and development of catalysts with high activity and selectivity.^6–8

The morphology and dispersion of the MoS₂ crystallite are closely correlated with catalytic performance. According to the brim-edge model,^9,10 brim sites, which refer to the top and bottom layers of the MoS₂ slab-like structure, are the active sites for both the HDS and olefin hydrogenation (HYDO) reactions. The edge sites are the only active sites for the HDS reaction. There is a relationship between the ratio of brim sites to edge sites and HDS selectivity. Fan et al. stated that high dispersion and low stacking of MoS₂ slabs are advantageous to olefin hydrogenation activity, which leads to poor HDS selectivity. However, catalysts with high stacking and low dispersion exhibit mediocre HDS activity and selectivity.¹¹ Moreover, by weakening the interaction between the support and metals, the sulfidation degree of active metals could be improved, leading to an increased formation of type II Co–Mo–S phase with high HDS activity.^12,13 Therefore, high HDS selectivity could be obtained by reducing the number of brim sites and increasing the amount of type II Co–Mo–S phase.

The morphology of the MoS₂ crystallite could be optimized by several methods, such as a change of surface properties, modification of the composition, reformation by impregnation methods, and adjustment of the sulfidation conditions, which could achieve higher HDS activity and selectivity. CeO₂ is widely used in heterogeneous catalysis due to its redox and dispersion properties.^14–17 In addition, CeO₂ is often used as the additive in mixing γ-Al₂O₃ based catalysts, which could improve the catalytic activity and stability.^18,19 Laosiripojana et al. proposed that the introduction of CeO₂ improves the dry reforming activity of Ni/Al₂O₃ mainly due to its redox properties.²⁰ Moreover, catalysts with CeO₂ incorporation have a higher catalytic performance, which is also related to its positive effect on the metal dispersion.^21,22 Furthermore, a CeAlO₃ phase could be formed between CeO₂ and γ-Al₂O₃, which could improve the stability of catalysts by limiting coke deposition.^19,23,24 Vlastimil Fila et al. prepared a Ce modified Mo/ZSM-5 catalyst, and obtained a higher catalytic performance and stability for the methane aromatization reaction.²⁵

In order to enhance the HDS activity and selectivity, we attempted to add CeO₂ into a γ-Al₂O₃ support and then prepare CoMo/CeO₂-γ-Al₂O₃ selective HDS catalysts for FCC gasoline. The effect of CeO₂ addition on the morphology of MoS₂ particles was studied by a series of characterization techniques. The performances of the catalysts were evaluated with FCC gasoline to obtain the optimal content of CeO₂ added in the catalyst. Moreover, the relationship between the morphology of MoS₂ phases and the HDS selectivity was clarified.

2. Materials and methods

2.1. Feedstock and chemicals

The heavy fraction FCC gasoline (boiling range, 65–180 °C) from Dagang Oil Refinery, China National Petroleum Corp (CNPC) was used as the feedstock. The properties of the gasoline are shown in Table 1. Ammonium molybdate tetrahydrate, cobalt nitrate hexahydrate, cerium nitrate hexahydrate, carbon disulfide, and petroleum ether (boiling range, 90–120 °C) were purchased from Sinopharm Chemical Reagent CO. Ltd.

Table 1 Properties of the feeding FCC gasoline

S (μg g^−1)	RON	Density (20 °C) (g cm^−3)	Group composition (wt%)
			n-Paraffin	i-Paraffin	Olefin	Naphthene	Aromatics
384	88.3	0.77	5.3	20.1	27.5	14.3	32.8

---

2.2. Preparation of catalysts

A series of CeO₂–Al₂O₃ supports were prepared. First, the pseudoboehmites were mixed with appropriate aqueous solutions of cerium nitrate. Then, the homogeneously blended mixtures were shaped by extrusion, dried at 110 °C for 8 h, then calcined at 550 °C for 4 h. Finally, the 20–40 mesh supports were obtained by crushing and screening.

Using the incipient wetness impregnation method, these supports were co-impregnated with an aqueous solution of ammonium molybdate and cobalt nitrate, followed by drying at 60 °C for 6 h and calcining at 550 °C for 4 h. A series of CoMo/CeO₂–Al₂O₃ catalysts were obtained. The MoO₃ and CoO amounts of all the catalysts were 8.0 and 2.5 wt%, respectively. The catalysts with a CeO₂ content of 0, 0.5, 1.0, 1.5, and 2.0 wt%, are named as Ce(0), Ce(0.5), Ce(1.0), Ce(1.5) and Ce(2.0), respectively.

2.3. Catalyst characterization

The specific surface area and pore volume measurements of the samples were carried out using Micromeritics ASAP 2400 equipment. Before measuring, the samples were evacuated at 300 °C under vacuum (1.33 × 10⁻³ Pa) for 3 h.

Temperature-programmed reduction of hydrogen (H₂-TPR) was carried out using a laboratory-constructed instrument, and was used to characterize the reducibility of the catalysts in the oxidized state. 0.2 g samples were pretreated with argon gas at 500 °C for 1 h. After being cooled to ambient temperature, the sample was treated with 5% H₂/Ar at a flow rate of 40 ml min⁻¹, followed by an increase in the temperature from 110 to 1080 °C at a heating rate of 10 °C min⁻¹. The variation of the hydrogen concentration in the gas mixture was detected by a thermal conductivity detector (TCD).

X-ray photoelectron spectroscopy (XPS) measurements of the sulfided catalysts were measured on a Thermo Fisher K-Alpha instrument with monochromatic Al K-α radiation. Before testing, the sulfided samples were kept in hexane to avoid oxidation. The XPS spectra were analyzed by XPSPEAK software (version 4.1) with application of the Shirley-type background and the proportion of Gaussian/Lorentzian set at 20/80.^8,26

High-resolution transmission electron microscopy (HRTEM) images of the sulfided catalysts were taken on a Philips Tecnai G2 F20 with an acceleration voltage of 200 kV. Freshly sulfided samples to be measured were milled in an agate mortar and then ultrasonically dispersed in ethanol. The test samples were prepared by dropping the dispersed suspensions onto carbon coated copper grids.

2.4. Catalyst activity evaluation

The catalytic activities of the catalysts were assessed in a continuous-flow fixed bed micro-reactor. 5.0 ml catalyst samples (20–40 mesh) were loaded into the stainless steel tube reactor with an internal diameter of 12 mm and length of 700 mm. The catalysts were presulfided in situ with the sulfiding feed containing 2 wt% CS₂ in petroleum ether, under 320 °C and 3.0 MPa for 4 h. After presulfurization, the reactor was cooled to the reaction temperature and then the FCC gasoline was fed into the reactor. The HDS reaction was carried out at a temperature of 260 °C, a total pressure of 1.6 MPa, a liquid hourly space velocity (LHSV) of 4.0 h⁻¹ of FCC gasoline and a volumetric ratio of H₂ to oil of 300. After a stabilization period of 10 h, the products were collected every 4 h. The sulfur content of the feedstocks and products was analyzed by ANTEK-7000NS. The hydrocarbon compositions were analyzed using a SP-3420A GC (Beijing Analytical Instrument Factory, China) with a flame ionization detector (FID) and a HP-PONA capillary column (50 m × 0.2 mm).

According to the literature,²⁷ the catalytic performances can be expressed as the conversion of the total sulfur (HDS%), the conversion of olefin hydrogenation (HYDO%) and as the HDS selectivity factor (S), as follows:

where S_f and S_p indicate the sulfur content in the feed and products, respectively; O_f and O_p are the olefin content in the feed and products, respectively.

3. Results and discussion

3.1. Characterization of the catalyst

3.1.1. N₂ adsorption–desorption. The pore structure properties of the catalysts are listed in Table 2. The addition of Ce to the Al₂O₃ support causes some changes to the specific surface area and pore volume of the catalysts. With the increasing CeO₂ loading, the specific surface area decreases slowly, however, the average pore size volume gradually increased. The pore volume of the catalysts was changed minimally. Fig. 1 shows the pore size distribution of these catalysts. The most probable pore diameters are unchanged with Ce introduction. In general, Ce has a slight effect on the pore structure of catalysts.

Table 2 Textural properties of the catalysts

Catalysts	Specific surface area/m^2 g^−1	Pore volume/ml g^−1	Average pore size/nm
Ce(0)	268.8	0.50	7.38
Ce(0.5)	261.3	0.49	7.55
Ce(1.0)	259.4	0.49	7.58
Ce(1.5)	257.5	0.49	7.63
Ce(2.0)	253.0	0.50	7.96

---

Fig. 1 Pore size distribution of catalysts.

3.1.2. H₂-TPR characterization. H₂-TPR profiles of the oxidic catalysts are presented in Fig. 2. Compared to oxidic molybdenum species, the contribution of oxidic cobalt species to hydrogen consumption is negligible.²⁸ Therefore, the molybdenum species is the focus in the following discussion. It can be observed that two peaks appear at about 500 and 910 °C, which correspond to the reduction of polymeric MoO_x species.²⁹ The low temperature peak is attributed to the partial reduction of Mo⁶⁺ to Mo⁴⁺ of octahedrally coordinated Mo oxo-species. The high temperature peak is ascribed to the deep reduction of Mo⁶⁺ of tetrahedrally coordinated Mo oxo-species.^4,30

Fig. 2 H₂-TPR profiles of the oxidic catalysts.

Compared with the Ce(0) catalyst, both the low and high temperature reduction peaks of the catalysts with Ce introduction shift to the lower temperature region. And with the increased CeO₂ loading, the temperature of the reduction peaks first decreases and then increases, appearing minimally at a CeO₂ loading of 1.0 wt%. This indicated that the loading of CeO₂ makes the molybdenum species reduce easier, and has an optimum value with a higher reducibility of the MoO_x species. This could be attributed to the fact that the interaction between Mo and CeO₂ is stronger than that of Mo and Al₂O₃,³¹ and hence the reduction of Mo species is inhibited. However, the small loading of CeO₂ (less than 1.0%) improves the dispersion of the Mo species, promoting the reduction of most of the Mo species. In general, the promotion of the reduction of the Mo species is a slight advantage, shifting the reduction peak to a lower temperature in a small range. Upon further increasing the content of CeO₂, the interaction of the Mo species with the support was increased due to the stronger interaction of the Mo species with CeO₂, hence inhibiting the reduction of the Mo oxide species.

3.1.3. XPS characterization. The freshly sulfided catalysts were characterized by XPS to compare the sulfidation extent. The spectra are shown in Fig. 3. Fig. 3a shows the deconvoluted spectra of Mo 3d, containing the peak of S 2s at approximately 225.4 eV and the doublets of Mo 3d.⁴ According to the literature,^8,32,33 the binding energies of the doublets at approximately 228.3 and 231.4 eV are ascribed to the Mo(IV) species; the binding energies around 230.3 and 233.4 eV are attributed to the Mo(V) species; the binding energies at about 232.2 and 235.3 eV are attributed to the Mo(VI) species. Three different species Mo(IV), Mo(V) and Mo(VI) are ascribed to oxide, oxysulfide and sulfide phases, respectively.³³

Fig. 3 Mo 3d and Co 2p XPS spectra of the sulfided catalysts (a) Mo 3d, and (b) Co 2p.

Fig. 3b shows the Co 2p spectra. Co is present in three phases on the catalysts’ surface: Co(II), CoMoS and Co₉S₈. According to the literature,¹³ Co(II) oxide has a binding energy at 781.5 eV, while the binding energies located at 778.0 and 778.6 eV are attributed to mixed CoMoS and sulfided Co₉S₈, respectively. Moreover, the analysis results summarized in Table 3 were obtained by the deconvolution method.

Table 3 XPS results of the sulfided catalysts

Catalysts	Mosulfidation/%	CoCoMoS/%
Ce(0)	45.9	18.8
Ce(0.5)	47.3	21.9
Ce(1.0)	52.5	25.7
Ce(1.5)	51.0	24.4
Ce(2.0)	46.2	20.1

---

The sulfidation degree of the Mo species is denoted as Mo_sulfidation (Mo_sulfidation = Mo_(IV)/(Mo_(IV) + Mo_(V) + Mo_(VI))), and Co_CoMoS is defined as the ratio of CoMoS to the total of Co(II), CoMoS, and Co₉S₈ (Co_CoMoS = CoMoS/(Co(II) + CoMoS + Co₉S₈)).³⁴ Obviously, the results show that Mo_sulfidation and Co_CoMoS of the sulfided catalysts modified by Ce are both higher than the unmodified catalyst, demonstrating the promotive effect of Ce on sulfidation. With increasing CeO₂ loading, both the Mo_sulfidation and Co_CoMoS values show a volcano-type relationship with the content of CeO₂ and peaked simultaneously at the CeO₂ loading of 1.0 wt%. The Mo species is the secondary support for the Co species. Therefore, the higher sulfidation degree of the Mo species better promotes the formation of the CoMoS active phase, due to the interaction between the Co and Mo species. This can be consolidated with the results of TPR characterization. Combined with the TPR results, it can be concluded that Ce incorporation weakens the metal–support interaction and promotes the reduction and sulfidation of the Mo species, facilitating Co species sulfidation to form CoMoS active phases.

3.1.4. HRTEM. The catalytic performance of the catalysts is related to the morphology of MoS₂. The representative HRTEM micrographs of the sulfided catalysts with different CeO₂ loadings are illustrated in Fig. 4. The presence of a well dispersed MoS₂ lamellar structure is observed on the support. Statistical analyses have been conducted on 400–500 slabs from 30 micrographs obtained from different parts of each sulfided catalyst. Then, a quantitative comparison was carried out on the lengths and stacking layer numbers of the MoS₂ slabs on the different catalysts. The average slab lengths (L̄) and stacking layer numbers (N̄) were calculated according to the following formula,³⁵where M_i is the slab length or stacking layer number of a stacked MoS₂ unit, and x_i is the number of slabs or stacks in the determined range, given the slab length or stacking layer number.

Fig. 4 HRTEM images of the sulfided catalysts: (a) Ce(0), (b) Ce(0.5), (c) Ce(1.0), (d) Ce(1.5) and (e) Ce(2.0).

f_Mo was calculated by dividing the number of Mo atoms located at the edge surface (including the brim sites) by the total number of Mo atoms, and was used to express the MoS₂ dispersion. The MoS₂ slabs are assumed to be perfect hexagons, therefore the following formula was used:³⁶

where n_i is the number of Mo atoms along one side of a MoS₂ slab, determined from its slab length (L = 3.2(2n_i − 1) Å) and t is the total number of slabs that appeared in the HRTEM micrographs.

The MoS₂ dispersion (f_Mo), average slab lengths (L̄) and stacking layer numbers (N̄) of the MoS₂ slabs on the sulfided catalysts are listed in Table 4. Ce incorporation decreases MoS₂ dispersion and increases the average slab lengths with a slight change in the average stacking layer numbers. MoS₂ morphologies are modified by introduction of Ce. With increasing CeO₂ content, the average slab lengths increase first and then decrease with a maximum value at the CeO₂ content of 1.0 wt%. Compared with the average slab lengths, the change in f_Mo has an opposite trend with the increase of CeO₂ loading, obtaining a minimum value at the 1.0 wt% CeO₂ loading. This demonstrates that the incorporation of Ce reduces the metal–support interaction, which is consistent with the results from H₂-TPR characterization. Therefore, we concluded that the addition of Ce could tune the size and morphology of MoS₂ phases.

Table 4 MoS₂ dispersion (f_Mo), average slab lengths (L̄) and stacking layer numbers (N̄) of the sulfided catalysts

Catalysts	fMo	L̄/nm	N̄
Ce(0)	0.44	2.5	2.2
Ce(0.5)	0.42	2.7	2.2
Ce(1.0)	0.38	3.0	2.3
Ce(1.5)	0.39	2.9	2.3
Ce(2.0)	0.43	2.6	2.3

---

3.2. Catalytic activity

The catalytic activities of the catalysts with Ce addition for HDS and HYDO were evaluated to obtain the best CeO₂ content. The assessment results are given in Table 5. Compared with the unmodified catalyst, the catalysts modified with Ce show higher HDS and HYDO activities. With an increased CeO₂ content, both the HDS and HYDO activities initially increase and then decrease and the selectivity factor follows the same trend. When the CeO₂ loading reaches 1.0 wt%, the catalyst has the best HDS activity and an outstanding selectivity factor, with the sulfur content in the product down to 9.2 μg g⁻¹. The optimum HDS activity and selectivity factor could be obtained by changing the CeO₂ content of the catalysts.

Table 5 Effect of Ce addition on the HDS and HYDO activities, and the HDS selectivity of the catalysts

Catalysts	Sulfur (μg g^−1)	Olefin (wt%)	HDS%	HYDO%	S
Ce(0)	18.9	24.5	94.1	16.1	16.1
Ce(0.5)	15.1	24.3	95.3	16.8	16.6
Ce(1.0)	9.2	23.8	97.1	18.5	17.4
Ce(1.5)	12.7	24.0	96.0	17.8	16.5
Ce(2.0)	16.8	24.3	94.8	16.8	16.0

---

3.3. Relationships of morphology and activity

The correlation between the morphology properties and reaction performance of the catalysts is discussed to elucidate the different catalytic activities of the Ce modified catalysts. In Fig. 5, the sulfidation degree of the Mo species is correlated with the selectivity of the HDS performance. With an increased sulfidation degree, the HDS and HYDO activities simultaneously increase. According to the XPS results, there are more CoMoS active phases with a higher sulfidation degree of the Mo species. The type II CoMoS active phase shows higher intrinsic HDS activity.³⁷ Meanwhile, the catalysts with higher sulfidation have more total active sites, which means that the number of Mo coordinative unsaturated sites (CUSs) increased. The CoMoS phase and Mo CUSs are beneficial for the HDS and HYDO activities, respectively.³⁸ Therefore, the catalyst with the CeO₂ loading of 1.0 wt% has the highest activities of HDS and HYDO. As shown in Fig. 5b, the HDS selectivity factor has a linear correlation with the Mo sulfidation degree, and the value of the selectivity factor increases with increasing sulfidation degree of the Mo species. The catalysts with a higher sulfidation degree have a better HDS selectivity. This could be attributed to the higher sulfidation degree promoted in the formation of more CoMoS active phases with a smaller value of f_Mo (Fig. 5c). The smaller value of f_Mo indicates a lower proportion of edge Mo atoms in the active phase. This possibly leads to the reduction of the proportion of Mo CUSs on Mo edge sites, with a simultaneous increase of the proportion of CoMoS active sites. Therefore, the catalysts with the higher sulfidation degree and lower dispersion show the better HDS selectivity (Fig. 5d).

Fig. 5 HDS activity and HYDO activity (a), selectivity factor (b), dispersion of MoS₂ (c) versus the sulfidation degree of Mo species, and dispersion of MoS₂ versus selectivity factor (d) for the catalysts.

Based on the above analysis, the sulfidated catalyst with the optimized content of CeO₂ has the advantages of a high sulfidation degree and moderate dispersion, contributing to the formation of more CoMoS active phases with a lower proportion of Mo CUSs on Mo edge sites. Therefore, this catalyst exhibits an outstanding HDS activity and optimum selectivity. This result clearly illustrated the definite relationship between the catalytic performance and morphology of MoS₂ particles in catalysts modified by CeO₂.

4. Conclusions

A series of CoMo/CeO₂–Al₂O₃ catalysts were prepared with different amounts of CeO₂. The physicochemical properties of these catalysts, especially the morphology of sulfided catalysts, were systematically characterized. Results of the assessed catalytic activity are correlated with the MoS₂ morphology. The optimized catalyst with the CeO₂ loading of 1.0 wt% exhibited a superior selective HDS performance, due to its higher sulfidation degree of the Mo species and proportion of CoMoS active phases with lower Mo dispersion. Therefore, the successful introduction of Ce provides a high performance catalyst for selective hydrodesulfurization of FCC gasoline.

Notes and references

1. S. Brunet, D. Mey, G. Pérot, C. Bouchy and F. Diehl, Appl. Catal., A, 2005, 278, 143–172.
2. E. Ito and J. A. R. Veen, Catal. Today, 2006, 116, 446–460.
3. M. E. Morales-Valencia, G. V. Baldovino-Medrano and A. S. Giraldo, Fuel, 2015, 153, 294–302.
4. Y. Fan, G. Shi, H. Liu and X. Bao, Fuel, 2011, 90, 1717–1722.
5. V. N. Bui, D. Laurenti, P. Delichère and C. Geantet, Appl. Catal., B, 2011, 101, 246–255.
6. L. Kaluža, D. Gulková, Z. Vít and M. Zdražil, Appl. Catal., B, 2015, 162, 430–436.
7. T. Wang, Y. Fan, X. Wang, L. Chou and H. Lin, Fuel, 2015, 157, 171–176.
8. B. Liu, Y. Chai, Y. Li, A. Wang, Y. Liu and C. Liu, Appl. Catal., A, 2014, 471, 70–79.
9. H. Topsøe, Appl. Catal., A, 2007, 322, 3–8.
10. Y. Fan, J. Lu, G. Shi, H. Liu and X. Bao, Catal. Today, 2007, 125, 220–228.
11. Y. Fan, G. Shi, H. Liu and X. Bao, Appl. Catal., B, 2009, 91, 73–82.
12. P. G. Moses, B. Hinnemann, H. Topsøe and J. K. Nørskov, J. Catal., 2009, 268, 201–208.
13. A. D. Gandubert, E. Krebs, C. Legens, D. Costa, D. Guillaume and P. Raybaud, Catal. Today, 2008, 130, 149–159.
14. V. M. Gonzalez-DelaCruz, J. P. Holgado, R. Pereñíguez and A. Caballero, J. Catal., 2008, 257, 307–314.
15. J. Ashok and S. Kawi, Int. J. Hydrogen Energy, 2013, 38, 13938–13949.
16. V. R. Choudhary, K. C. Mondal and A. S. Mamman, J. Catal., 2005, 233, 36–40.
17. K. O. Xavier, R. Sreekala, K. K. A. Rashid, K. K. M. Yusuff and B. Sen, Catal. Today, 1999, 49, 17–21.
18. S. Wang and G. Q. Lu, Appl. Catal., B, 1998, 19, 267–277.
19. W. Chen, G. Zhao, Q. Xue, L. Chen and Y. Lu, Appl. Catal., B, 2013, 136–137, 260–268.
20. N. Laosiripojana, W. Sutthisripok and S. Assabumrungrat, Chem. Eng. J., 2005, 112, 13–22.
21. Y. Lu, S. Li and L. Guo, Fuel, 2013, 103, 193–199.
22. I. F. Elbaba, C. Wu and P. T. Williams, Int. J. Hydrogen Energy, 2011, 36, 6628–6637.
23. T. Y. Kim, S. M. Kim, W. S. Lee and S. I. Woo, Int. J. Hydrogen Energy, 2013, 38, 6027–6032.
24. I. Luisetto, S. Tuti, C. Battocchio, S. L. Mastro and A. Sodo, Appl. Catal., A, 2015, 500, 12–22.
25. V. Fila, M. Bernauer, B. Bernauer and Z. Sobalik, Catal. Today, 2015, 256, 269–275.
26. S. Damyanova, L. Petrov and P. Grange, Appl. Catal., A, 2003, 239, 241–252.
27. S. Shan, P. Yuan, W. Han, G. Shi and X. Bao, J. Catal., 2015, 330, 288–301.
28. V. La Parola, G. Deganello and A. M. Venezia, Appl. Catal., A, 2004, 260, 237–247.
29. H. Li, M. Li, Y. Chu, F. Liu and H. Nie, Fuel, 2014, 116, 168–174.
30. J. Vakros, A. Lycourghiotis, G. A. Voyiatzis, A. Siokou and C. Kordulis, Appl. Catal., B, 2010, 96, 496–507.
31. Y. Zhang, G. Zhang, Y. Zhao, X. Li, Y. Sun and Y. Xu, Int. J. Hydrogen Energy, 2012, 37, 6363–6371.
32. T. K. T. Ninh, D. Laurenti, E. Leclerc and M. Vrinat, Appl. Catal., A, 2014, 487, 210–218.
33. W. Chen, F. Maugé, J. van Gestel, H. Nie, D. Li and X. Long, J. Catal., 2013, 304, 47–62.
34. P. A. Nikulshin, D. I. Ishutenko, A. A. Mozhaev, K. I. Maslakov and A. A. Pimerzin, J. Catal., 2014, 312, 152–169.
35. E. J. M. Hensen, P. J. Kooyman, Y. van der Meer, A. M. van der Kraan, V. H. J. de Beer, J. A. R. van Veen and R. A. van Santen, J. Catal., 2001, 199, 224–235.
36. P. A. Nikulshin, V. A. Salnikov, A. V. Mozhaev, P. P. Minaev, V. M. Kogan and A. A. Pimerzin, J. Catal., 2014, 309, 386–396.
37. M. Toba, Y. Miki, T. Matsui, M. Harada and Y. Yoshimura, Appl. Catal., B, 2007, 70, 542–547.
38. V. Lauritsen Jeppe, J. Kibsgaard, H. Olesen Georg, G. Moses Poul, B. Hinnemann, S. Helveg, K. Nørskov Jens, S. Clausen Bjerne, H. Topsøe, E. Lægsgaard and F. Besenbacher, J. Catal., 2007, 249, 220–233.

---