- Email: [email protected]
Nickel nanoparticles highly dispersed with an ordered distribution in MCM-41 matrix as an efficient catalyst for hydrodechlorination of chlorobenzene
Microporous and Mesoporous Materials 185 (2014) 130–136
Contents lists available at ScienceDirect
Microporous and Mesoporous Materials journal homepage: www.elsevier.com/locate/micromeso
Nickel nanoparticles highly dispersed with an ordered distribution in MCM-41 matrix as an efficient catalyst for hydrodechlorination of chlorobenzene Naijin Wu, Wen Zhang, Baoshan Li ⇑, Chunying Han State Key Laboratory of Chemical Resource Engineering, Beijing University of Chemical Technology, Beijing 100029, PR China
a r t i c l e
i n f o
Article history: Received 23 August 2013 Received in revised form 13 October 2013 Accepted 9 November 2013 Available online 19 November 2013 Keywords: Hydrothermal synthesis Highly dispersed Mesoporous molecular sieve Nickel nanoparticles Hydrodechlorination
a b s t r a c t Ni/MCM-41 materials in which highly dispersed nickel nanoparticles with an ordered distribution were successfully prepared by crystal lattice locating method. The synthesized samples were characterized by XRF, XRD, XPS, HRTEM, and N2 adsorption–desorption isotherm techniques. The results suggest that the materials possess highly ordered hexagonal mesostructure of MCM-41 and the highly dispersed nickel nanoparticles with a uniform size smaller than 20 nm are in an ordered configuration in MCM-41 matrix. The materials with high specific surface area, large pore volume, and big pore diameter exhibit an excellent catalytic performance in hydrodechlorination of chlorobenzene. Ó 2013 Elsevier Inc. All rights reserved.
1. Introduction The active components of a catalyst are usually transition metal atoms, which can be highly dispersed in a carrier with large specific surface area. Such supported metal catalysts are one of the most important class of catalysts in petroleum refining and petrochemical processing, such as catalytic oxidation [1,2], selective catalytic reduction of NOx [3], hydrogenation [4] and reforming [5,6]. The impregnation of porous silica supports with a transition metal ion containing solution is a traditional method to prepare the supported metal or metal oxide catalysts. In the drying stage of the method, the active metal component precursor is often deposited in the form of aggregates on the surface of the carriers due to the liquid surface tension and the solution effects. And it is difficult to break up this aggregation in the subsequent activation processes. In addition, calcinations and reduction at high temperature are usually required for the traditional catalyst preparation process, and these can result in further migration of the metal species and lead to the formation of large particles due to the weak interaction between the active metal species and the carrier [7]. So in many cases the metal particles cannot uniformly disperse in the carriers with an out-of-order distribution, which result in a ghastly limitation in catalytic performance, including activity, selectivity and life-span. To overcome these disadvantages, many researchers ⇑ Corresponding author. Tel./fax: +86 10 64445611. E-mail addresses: [email protected], [email protected] (B. Li). 1387-1811/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.micromeso.2013.11.017
have made great efforts in preparing highly dispersed supported metal catalysts [8,9]. Although the highly dispersed metals were obtained, the arrangement of which are scattered casually without special orders, the stability and the selectivity of the materials are not satisfactory. MCM-41 possesses high specific surface area, narrow pore distribution, ordered pore arrangement, and the pore diameter can be adjusted [10,11]. However, the applications of the pure silica MCM-41 mesoporous molecular sieves are limited in industries because they possess some disadvantages, such as the weak surface acidity and low catalytic activity. In order to enhance the catalytic performance, there are enormous efforts having been devoted to modify the pure siliceous MCM-41, including grafting [12], incorporation [13], loading of various metals [14,15] and metal oxide species [16,17]. By heteroatom doped or loaded, for example, Ni, Co, Cu, V, Ti etc., the physical chemistry properties of MCM-41 can be modified, the catalytic performance can be enhanced, and the application domains can be extended. The catalysts with nickel have been applied widely in industries, such as hydrogenation of aromatic compounds [18], methanation of CO, steam reforming of hydrocarbons, amination of alcohols and carbon nanotube preparations etc [19–21]. The MCM-41 loaded nickel catalysts can be used in hydrogenation of gaseous acetonitrile [22], decomposition of amines [23], hydrogenation of benzene [24,25], oxygenation of benzene to prepare phenol and oxygenation of styrene to prepare benzaldehyde [26]. The activation energy of the reaction of hydrogenation of
N. Wu et al. / Microporous and Mesoporous Materials 185 (2014) 130–136
benzene on the Ni/MCM-41 catalyst is 36.5 kJ mol1, which is much lower than 50.5–52.3 kJ mol1 on the Ni/MyOx catalysts (MyOx is an oxide supporter, such as SiO2, Al2O3 and MgO etc.) [27]. So, MCM-41 supporter is very attractive for catalysts. The preparation methods of Ni/MCM-41 catalysts have been reported including impregnation methods with nickel nitrate or nickel citrate solutions [24,25,28–32], ion-exchange methods [33] and direct syntheses methods [26,34,35]. All the reports have been focused on the high disperse of metal nickel or NiOx on the supports and the ordered arrangement distribution of which were hardly ever mentioned. In the present paper, the NiII-MCM-41 precursors were firstly synthesized by direct hydrothermal method, in which the nickel atoms were highly dispersed and embedded into the framework of MCM-41, that is, the nickel atoms were placed into the crystal lattice of the MCM-41. Then the nickel atoms were reduced from the NiII-MCM-41 precursor in hydrogen flow at a moderate temperature without destroyed the structure of MCM-41. So the Ni atoms would be anchored to the framework of MCM-41 and the localization of the framework made agglomeration of the metal nanoparticles difficult, which ensures the nanoparticles stable in the molecular sieve framework. This method can be distinguished as the crystal lattice locating method. The catalytic performance of the result samples were examined by using the reaction of HDC of CB. The significance of the work is that high amount of metal nanoparticles can be highly dispersed in the structure of molecular sieve with an ordered distribution and the result materials possess the advantages of molecular sieves and metal nanoparticles for catalysis, simultaneously. 2. Materials and methods 2.1. Materials Nickel nitrate (AR), sodium hydroxide (AR, Beijing Yili Fine Chemical Product Limited Company, China); tetraethyl orthosilicate (TEOS, AR, Beijing Chemical Factory, China); ammonia solution (25%, AR, Beijing Beihua Fine Chemical Product Limited Company, China); cetyltrimethylammonium bromide (CTAB, AR), ethanol (AR), chlorobenzene (CB, AR, Tianjin Jinke Fine Chemical Institute, China), and hydrogen (>99.99%, Beijing Beiwen Gas Factory, China) were used. All solvents and reactants are commercially available and were used without further purification. 2.2. Methods 2.2.1. Synthesis of the precursor NiII-MCM-41 The NiII-MCM-41 mesoporous materials were prepared by a direct hydrothermal synthesis method. A typical synthesis was conducted as follows: CTAB was dissolved in deionized water in a flask with vigorous stirring for 30 min in order to obtain a clear solution. A certain amount of nickel nitrate was dispersed in deionized water in a beaker and an ammonia solution was added to obtain a solution containing the NiðNH3 Þ2þ 6 complex. The solution of NiðNH3 Þ2þ 6 was poured into the CTAB solution at room temperature. After stirred vigorously for 30 min, TEOS was added dropwise into the mixture and then stirred for another 30 min. The pH was adjusted to 10.0 by using an ammonia solution, and the final mixture was stirred for 5 h. The molar composition of the mixture was 1 TEOS: 0.152 CTAB: 2.8 NH3: 1/x Ni: 141.2 H2O. The mixture was transferred to a static Teflon-lined stainless steel autoclave and heated under autogenous pressure at 110 °C for 48 h. After that, the sample was filtered, washed with deionized water, dried overnight at 50 °C and finally calcined at 550 °C in air for 6 h. The NiII-MCM-41 precursor was gained.
131
2.2.2. Reduction of NiII from the NiII-MCM-41 precursor The synthesis of Ni/MCM-41 by reduction of the NiII-MCM-41 precursor was carried out as follow: NiII-MCM-41 was put into a quartz tube and then heated in hydrogen flow from room temperature to 750 °C at a heating rate of 5 °C/min and kept 750 °C for 1 h. Then, it was cooled naturally in hydrogen flow to the room temperature. The resulting sample was designated as A-i. The compared sample Ni/MCM-41 with the same nickel content corresponding to the sample A-i was prepared with the traditional impregnating method. Firstly, prepared a NiII/MCM-41 precursor with MCM-41 doped in a solution of NiðNH3 Þ2þ complex (same 6 as the above) and heated treatment. Then the NiII/MCM-41 precursor was reduced with the same process as the sample A-i. The resulting sample was designated as B-i. 2.2.3. Characterizations X-ray diffraction (XRD) patterns were recorded with a Rigaku D/ Max 2500 VBZ+/PC diffractometer using Cu Ka radiation (40 kV, 200 mA) in both low angle (2h range of 0.5–10°) and wide angle range (2h range of 10–80°). The X-ray fluorescence analysis (XRF) was performed on a Philips Magix-601 X-ray fluorescence spectrometer for determining the nickel content of the samples. The morphologies of the samples were examined by high-resolution transmission electron microscopy (HRTEM) on a JEM-3010 microscope with an accelerating voltage of 200 kV. Before the examination the samples were prepared by ultrasonic dispersion with absolute alcohol as solvent on a copper grid as supported membrane. The nitrogen adsorption–desorption isotherms were determined on a Micromeritics ASAP 2020 M volumetric adsorption analyzer. Before the nitrogen adsorption, each sample was degassed in a vacuum at 200 °C for 6 h. The specific surface area (SBET) was estimated by the BET equation. And the pore distribution, the mesopore analysis were obtained from the desorption branch of the isotherm using the Barrett–Joyner–Halenda (BJH) method. The X-ray photoelectron spectroscopy (XPS) analyses were conducted on an ESCALAB 250 spectrometer equipped with an Al Ka X-ray source. The carbon 1s peak at 284.6 eV was used as the reference for binding energies. 2.2.4. Catalytic performance testing The catalytic performance of the samples was tested through the reaction of HDC of CB. The experiments were carried out in a 250 mL stainless steel autoclave with ethanol as the solvent. The reaction conditions were as follows: reaction temperature was 100 °C, H2 partial pressure was 1.5 MPa, the catalyst dosage was 1.0 wt.%, and the reaction time was 60 min. The reactants and the products were analyzed by GC with a flame ionization detector online. And the conversion was calculated by the formula:
Conversion ð%Þ ¼
cC6 H5 Cl0 cC6 H5 Cli 100 cC6 H5 Cl0
cC6 H5 Cl0 and cC6 H5 Cli denotes the molar concentration of CB before and after reaction, respectively. 3. Results and discussion 3.1. Characterizations The nickel contents of the samples were determined by XRF and the results are listed in Table 1. The sample A-1, A-2 and A-3 in Table 1 is corresponding to the sample of nominal Si/Ni molar ratio of 10, 20, and 30, respectively. The actual nickel content in the sample A-1, A-2 and A-3 is 9.3%, 5.9%, and 3.4%(wt.), respectively. Fig. 1 shows the XRD patterns of the A samples. The small-angle XRD patterns (Fig. 1a) indicate that all the samples possess the
132
N. Wu et al. / Microporous and Mesoporous Materials 185 (2014) 130–136
Table 1 The textural properties of the A samples. Sample
Si/Ni
Ni wt.%
SBET/m2 g–1
Dp/nm
Vp/m3 g–1
DNi/nm
A-1 A-2 A-3 MCM-41[36]
10 20 30 –
9.3 5.9 3.4 0.0
404 574 644 1167
4.7 3.5 3.3 2.52
0.91 0.71 0.60 0.61
18.9 15.7 14.3 –
Fig. 2. Wide-angle XRD patterns of Si/Ni = 20 samples at different reduction temperature: a-wide angle patterns, b-low angle patterns.
Fig. 1. XRD patterns of A samples: a-low angle patterns, b-wide angle patterns.
high ordered structure of MCM-41 and the pore diameter of the samples increases with the nickel content increasing. The wide-angle XRD patterns (Fig. 1b) illuminate that all the samples except A-1⁄ (unreduced NiII-MCM-41 precursor of the sample A-1) show the characteristic diffraction peaks of metallic nickel (2h = 44.49°, 51.85°) and no characteristic diffraction peaks of the other nickel species, confirming that Ni(II) introduced into the skeletons of MCM-41 or highly dispersed in the matrix of MCM-41, which was illuminated by the XRD pattern of A-1⁄ without diffraction peaks of any nickel species, had been reduced. With the increasing of nickel content, the intensity of the diffraction peaks increase. The particle size of metallic nickel was calculated using the Scherrer formula (DNi = Kk/Bcosh, K is the Scherrer constant, 0.89; DNi is the particle size, nm; B is the full width of the half-height of the diffraction peak; h is the diffraction angle, and k is the X-ray wave
Fig. 3. XPS spectrum of sample A-1.
length, 0.154056 nm). The results are also listed in Table 1. It is clearly that the size of the metallic nickel particles increase slightly with the nickel content increasing, but the sizes of the particles in all the samples are smaller than 20 nm.
N. Wu et al. / Microporous and Mesoporous Materials 185 (2014) 130–136
The reduced conditions of Ni from NiII-MCM-41 precursor are very important for the aim materials preparation. Fig. 2 shows the XRD patterns of the Si/Ni = 20 samples gained at different reduction temperature, respectively. It is obviously that Ni(II) was not reduced from the NiII-MCM-41 precursor at 650 °C (Fig. 2a), and at 840 °C the structure of MCM-41 had been destroyed (Fig. 2b). 750 °C is the suitable reduction temperature for the preparation of the aim materials. The XPS spectrum of the sample A-1 (the highest nickel content sample in this work) is shown in Fig. 3 and the inset shows the subpeak results of Ni2p spectrum by Gaussian fitting. The predominant peak centered at 852.7 eV is the characteristic peak of the
133
metallic Ni2p3/2 [37,38]. There are no peaks at 857.8 eV and 855.0 eV, corresponding to the framework Ni2p3/2 and the nickel oxide Ni2p3/2 [32], respectively. The result confirms that the nickel atoms in the NiII-MCM-41 precursor were reduced to Ni0 thoroughly, which is consistent with the XRD analysis. Fig. 4 shows the HRTEM micrographs of all the A and B samples. The nominal nickel content in the sample A-i and in the corresponding sample B-i is the same. It is clear that the nickel particles in the A samples are very different from in the B samples. In the A samples, the size of the nickel particles (<20 nm) are much smaller than that in the B samples (>30 nm). The nickel particles in the A samples possess more uniform size and more regular shape than
Fig. 4. HRTEM images of the samples A and B.
134
N. Wu et al. / Microporous and Mesoporous Materials 185 (2014) 130–136
Fig. 5. N2 adsorption/desorption isotherms and the pore diameter distributions of samples.
in the B samples. In the A samples, the nickel particles are highly dispersed in an ordered distribution with about equal distance or multiple of the distance between the nickel particles in inside of the pore structure of the MCM-41, with scarcely any nickel particles dispersed on the external surface, which was very different from in the B samples with nickel particles stacked in a random way on the external surface, especially for the high nickel content B samples. With the nickel content increasing, the intervals decrease between the nickel particles, but the particle sizes were hardly changed in the A samples, and this should be decided by the lattice structure of the MCM-41. Additionally, the ordered hexagonal mesoporous structure of the MCM-41 was kept perfectly in the nickel reduced process, although the pore structure was slightly changed in the highest nickel content sample A-1. And the results are in agreement with the XRD analysis (Fig. 1a). The characters of the ordered distribution of the nickel nanoparticles
and the ordered pore structure of the MCM-41 will be very significative for the catalytic performance. Fig. 5 shows the N2 adsorption–desorption isotherms of the A samples, the corresponding textural properties are given in Table 1. The curves of all the three samples are Type IV, which is the typical adsorption curve of the mesoporous materials. As shown in Table 1, the pore diameter (Dp, nm) of the samples increase along with the increasing of nickel content. This is because the nickel particles expanded the cavities of the samples and resulted in the widening of the pore diameter (see the forming mechanism in Scheme 1). Meanwhile, the specific surface area (SBET, m2 g1) of the samples decrease along with the increasing of nickel content, the primary reason is that the atomic weight of nickel is larger than that of Si. Of course, the presence of more nickel particles in the higher nickel content samples may overlay some internal surface and result in decreasing of the specific surface area in a certain extent. Through the above characterizations, the formation mechanism of the samples was proposed as shown in Scheme 1. Firstly, the mesoporous material with hexagonal pore structure was successfully synthesized through a direct hydrothermal synthesis method and all the Ni(II) ions were uniformly dispersed in the silica framework. After the Ni(II) species were reduced to Ni(0), the metallic Ni(0) atoms would transfer along with the channel to form nickel atom clusters and then become bigger particles at the high reduced temperature. Because of the strong restrictive effect of the silica framework, the nickel particle diameters were restricted and could not form big nickel particles under the experimental conditions. As the nickel atoms reduced, the sites of the Ni(II) ions would be replaced by H atoms and form a new structure as
, which could not result in the framework
breakage to a considerable degree. The nickel particles must be implantation in some crystal lattice nodes in which the nickel atoms were reduced, and only in these sites the nickel particles could grow into the biggest size without the framework damaged. In addition, the size of the nickel particles is correlated with the pore diameter of the precursor NiII-MCM-41. The big atoms (Ni)
Scheme 1. Schematic diagram of the Ni/MCM-41 samples preparation.
N. Wu et al. / Microporous and Mesoporous Materials 185 (2014) 130–136
substitute the small atoms (Si) of the framework will enlarge the pore diameter of the MCM-41 and the more Si atoms substituted by nickel atoms at the range of the experimental contents the bigger pore diameter enlarged [39]. So the nickel particle diameter in the sample A-1 is the biggest, and in the samples A-2 and A-3 the particle diameters are smaller with a small difference. The distribution of the nickel particles can be illuminated as the schematic diagram 1, 2 and 3 in Scheme 1 for the sample A-1, A-2 and A-3, respectively, which with an ordered distribution. 3.2. Catalytic performance HDC of CB were carried out with the A samples and the comparison B samples as the catalysts under the same conditions. HDC of CB normally results in the production of three products, viz. benzene, cyclohexane, and chlorocyclohexane [33]. But over the A and B samples, the product was only benzene, without cyclohexane and chlorocyclohexane, under the experimental temperature 100 °C, which can be attributed not only to the higher ordered pore structure of MCM-41, but also to the lower reaction temperature than that of the previous reports. The catalytic performances are shown in Fig. 6. It can be seen that all the A samples show much better catalytic performance than the corresponding B samples. The conversion of CB was up to 98.2%, 97.6% and 82.5% over the sample A-1, A-2 and A-3, which was much higher than 70.2%, 64.8% and 63.2% over the corresponding sample B-1, B-2 and B-3, respectively. The A samples exhibited much higher catalytic activity than the B samples, which can be attributed to the smaller nickel nanoparticles with high dispersion in an ordered distribution manner in the high ordered pore structure of MCM-41. The reusability of the A and B samples were also examined under the same conditions and the results show that the catalytic performance of the three A samples had no significant decrease after four runs, the conversion of CB over the sample A-1, A-2 and A-3 was 97.4%, 97.1% and 81.9%, respectively. And over the sample B-1, B-2 and B-3, the conversion of CB was about 42.1%, 23.2% and 15.6%, respectively, which indicates that the reusability of the A catalysts are much better than that of the B samples. This is because the nickel particles were ordered encapsulated into the frameworks of the MCM-41 in the A samples, resulting in the stronger interaction between the nickel particles and the MCM41 supporter, and the B samples were not so. The catalyst deactivation is generally a problem in the HDC reaction, and it has been attributed mainly to the presence of HCl generated in the reaction,
Fig. 6. Catalytic results of all the samples.
135
which leads to the formation of a stable surface chloride species. The enhanced stability of the A samples is consistent with the smaller crystallite size, higher nickel dispersion, higher ordered distribution of the nickel nanoparticles, and the stronger interaction between the nickel species and the MCM-41 supporter. The spent catalysts were also characterized by XRD and XRF. The XRF results indicated no evidence for nickel loss of the A catalysts, and for the B samples the nickel loss was obviously during the reaction. In addition, the characteristic reflections of NiCl2 can be observed in the XRD patterns of the used B catalysts, which indicated the poisoning of active nickel species with HCl produced during HDC reaction. This can be ascribed to the weak interactions between nickel species and MCM-41 supporter in the B catalysts. In contrast, the used A catalysts did not show any new peaks in its XRD patterns. So, the deactivation observed for the B catalysts can be attributed not only to the losing of nickel metal atoms, but also to the poisoning of active nickel species with HCl. 4. Conclusions The highly dispersed nickel nanoparticles embedded in the mesoporous molecular sieve MCM-41 have been successfully prepared by the crystal lattice locating method. The synthesized samples possess an ordered hexagonal mesostructure of MCM-41 with highly dispersed nickel nanoparticles, which distributed in an ordered manner. The diameter of the nickel particles is smaller than 20 nm and firmly studded in the frameworks of MCM-41 in an ordered distribution. The result samples show an excellent catalytic performance, with higher activity, selectivity, and stability in HDC of CB at the low temperature. Acknowledgements This work was financially supported by the National Natural Science of Foundation of China (No. 21271017) and the National Basic Research Program of China (973 Program, Grant No. 2011CBA00506). References [1] A.B. Sorokin, A. Tuel, Catal. Today 57 (2000) 45. [2] J. Gu, Y. Huang, S.P. Elangovan, Y. Li, W. Zhao, I. Toshio, Y. Yamazaki, J. Shi, J. Phys. Chem. C 115 (2011) 21211. [3] D.J. Kim, J.W. Kim, S.J. Choung, M. Kang, J. Ind. Eng. Chem. 14 (2008) 308. [4] R. Nares, J. Ramirez, A. Gutierrez-Alejandre, R. Cuevas, Ind. Eng. Chem. Res. 48 (2008) 1154. [5] J. Gao, Z. Hou, J. Guo, Y. Zhu, X. Zheng, Catal. Today 131 (2008) 278. [6] J. Gao, D. Liang, Z. Hou, J. Fei, X. Zheng, Int. J. Hydrogen Energy 33 (2008) 5493. [7] J.H. Kim, D.J. Suh, T.J. Park, K.L. Kim, Appl. Catal. A 197 (2000) 191. [8] J. Chen, Y. Zhang, L. Tan, Y. Zhang, Ind. Eng. Chem. Res. 50 (2011) 4212. [9] B. Qiao, A. Wang, X. Yang, L.F. Allard, Z. Jiang, Y. Cui, J. Liu, J. Li, T. Zhang, Nat. Chem. 3 (2011) 634. [10] J.S. Beck, J.C. Vartuli, W.J. Roth, M.E. Leonowicz, C.T. Kresge, K.D. Schmitt, C.T. Chu, D.H. Olson, E.W. Sheppard, S.B. McCullen, J. Am. Chem. Soc. 114 (1992) 10834. [11] B.S. Li, J.Q. Xu, J. Liu, S. Zuo, Z. Pan, Z. Wu, J. Colloid Interface Sci. 366 (2012) 114. [12] Y. Fazaeli, M.M. Amini, E. Mohajerani, M. Sharbatdaran, N. Torabi, J. Colloid Interface Sci. 346 (2010) 384. [13] B.S. Li, K. Wu, T.H. Yuan, C.Y. Han, J.Q. Xu, X.M. Pang, Microporous Mesoporous Mater. 151 (2012) 277. [14] J. Iglesias, J.A. Melero, M. SánchezSánchez, Microporous Mesoporous Mater. 132 (2010) 112. [15] C.Z. Loebick, S. Lee, S. Derrouiche, M. Schwab, Y. Chen, G.L. Haller, L. Pfefferle, J. Catal. 271 (2010) 358. [16] K. Parida, K.G. Mishra, S.K. Dash, J. Hazard. Mater. 241–242 (2012) 395. [17] S.G. Aspromonte, Á. Sastre, A.V. Boix, M.J. Cocero, E. Alonso, Microporous Mesoporous Mater. 148 (2012) 53. [18] S. Narayanan, R. Unnikrishman, V. Vishwanathan, Appl. Catal. A 129 (1995) 9. [19] R.L. Vander Wal, T.M. Ticich, V.E. Curtis, Carbon 39 (2001) 2277. [20] J.T. Feng, Y.J. Lin, D.G. Evans, X. Duan, D.Q. Li, J. Catal. 266 (2009) 351. [21] J. Wang, G. Fan, H. Wang, F. Li, Ind. Eng. Chem. Res. 50 (2011) 13717.
136
N. Wu et al. / Microporous and Mesoporous Materials 185 (2014) 130–136
[22] P. BraosGarcía, P. MairelesTorres, E. RodríguezCastellón, A. JiménezLópez, J. Mol. Catal. A: Chem. 193 (2003) 185. [23] X.K. Li, W.J. Ji, J. Zhao, S.J. Wang, C.T. Au, J. Catal. 236 (2005) 181. [24] A. Lewandowska, S. Monteverdi, M. Bettahar, M. Ziolek, J. Mol. Catal. A: Chem. 188 (2002) 85. [25] R. Wojcieszak, S. Monteverdi, M. Mercy, I. Nowak, M. Ziolek, M.M. Bettahar, Appl. Catal. A 268 (2004) 241. [26] V. Parvulescu, B.L. Su, Catal. Today 69 (2001) 315. [27] P. Marécot, E. Paraiso, J.M. Dumas, J. Barbier, Appl. Catal. A 80 (1992) 79. [28] A. InfantesMolina, J. MéridaRobles, P. BraosGarcía, E. RodríguezCastellón, E. Finocchio, G. Busca, P. MairelesTorres, A. JiménezLópez, J. Catal. 225 (2004) 479. [29] K. Fang, J. Ren, Y. Sun, J. Mol. Catal. A: Chem. 229 (2005) 51. [30] D. ElicheQuesada, J. MéridaRobles, P. MairelesTorres, E. RodríguezCastellón, A. JiménezLópez, Langmuir 19 (2003) 4985.
[31] D.J. Lensveld, J. Gerbrand Mesu, A. Jos Van Dillen, K.P. De Jong, Microporous Mesoporous Mater. 44-45 (2001) 401. [32] J. Jia, Y. Wang, E. Tanabe, T. Shishido, K. Takehira, Microporous Mesoporous Mater. 57 (2001) 283. [33] M. Hartmann, A. Pöppl, L. Kevan, J. Phys. Chem. 100 (1996) 9906. [34] W. Wang, M. Song, Mater. Res. Bull. 40 (2005) 1737. [35] Y. Yang, S. Lim, G. Du, Y. Chen, D. Ciuparu, G.L. Haller, J. Phys. Chem. B 109 (2005) 13237. [36] B.S. Li, W. Ma, C.Y. Han, J.J. Liu, X.M. Pang, X.H. Gao, Microporous Mesoporous Mater. 156 (2012) 73. [37] J.L. Rousset, F.J. Cadete Santos Aires, B.R. Sekhar, P. Mélinon, B. Prevel, M. Pellarin, J. Phys. Chem. B 104 (2000) 5430. [38] P.E.R. Blanchard, A.P. Grosvenor, R.G. Cavell, A. Mar, Chem. Mater. 20 (2008) 7081. [39] X. Li, B.S. Li, H.H. Mao, A. Tufail, Shah, J. Colloid Interface Sci. 332 (2009) 444.
Contents lists available at ScienceDirect
Microporous and Mesoporous Materials journal homepage: www.elsevier.com/locate/micromeso
Nickel nanoparticles highly dispersed with an ordered distribution in MCM-41 matrix as an efficient catalyst for hydrodechlorination of chlorobenzene Naijin Wu, Wen Zhang, Baoshan Li ⇑, Chunying Han State Key Laboratory of Chemical Resource Engineering, Beijing University of Chemical Technology, Beijing 100029, PR China
a r t i c l e
i n f o
Article history: Received 23 August 2013 Received in revised form 13 October 2013 Accepted 9 November 2013 Available online 19 November 2013 Keywords: Hydrothermal synthesis Highly dispersed Mesoporous molecular sieve Nickel nanoparticles Hydrodechlorination
a b s t r a c t Ni/MCM-41 materials in which highly dispersed nickel nanoparticles with an ordered distribution were successfully prepared by crystal lattice locating method. The synthesized samples were characterized by XRF, XRD, XPS, HRTEM, and N2 adsorption–desorption isotherm techniques. The results suggest that the materials possess highly ordered hexagonal mesostructure of MCM-41 and the highly dispersed nickel nanoparticles with a uniform size smaller than 20 nm are in an ordered configuration in MCM-41 matrix. The materials with high specific surface area, large pore volume, and big pore diameter exhibit an excellent catalytic performance in hydrodechlorination of chlorobenzene. Ó 2013 Elsevier Inc. All rights reserved.
1. Introduction The active components of a catalyst are usually transition metal atoms, which can be highly dispersed in a carrier with large specific surface area. Such supported metal catalysts are one of the most important class of catalysts in petroleum refining and petrochemical processing, such as catalytic oxidation [1,2], selective catalytic reduction of NOx [3], hydrogenation [4] and reforming [5,6]. The impregnation of porous silica supports with a transition metal ion containing solution is a traditional method to prepare the supported metal or metal oxide catalysts. In the drying stage of the method, the active metal component precursor is often deposited in the form of aggregates on the surface of the carriers due to the liquid surface tension and the solution effects. And it is difficult to break up this aggregation in the subsequent activation processes. In addition, calcinations and reduction at high temperature are usually required for the traditional catalyst preparation process, and these can result in further migration of the metal species and lead to the formation of large particles due to the weak interaction between the active metal species and the carrier [7]. So in many cases the metal particles cannot uniformly disperse in the carriers with an out-of-order distribution, which result in a ghastly limitation in catalytic performance, including activity, selectivity and life-span. To overcome these disadvantages, many researchers ⇑ Corresponding author. Tel./fax: +86 10 64445611. E-mail addresses: [email protected], [email protected] (B. Li). 1387-1811/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.micromeso.2013.11.017
have made great efforts in preparing highly dispersed supported metal catalysts [8,9]. Although the highly dispersed metals were obtained, the arrangement of which are scattered casually without special orders, the stability and the selectivity of the materials are not satisfactory. MCM-41 possesses high specific surface area, narrow pore distribution, ordered pore arrangement, and the pore diameter can be adjusted [10,11]. However, the applications of the pure silica MCM-41 mesoporous molecular sieves are limited in industries because they possess some disadvantages, such as the weak surface acidity and low catalytic activity. In order to enhance the catalytic performance, there are enormous efforts having been devoted to modify the pure siliceous MCM-41, including grafting [12], incorporation [13], loading of various metals [14,15] and metal oxide species [16,17]. By heteroatom doped or loaded, for example, Ni, Co, Cu, V, Ti etc., the physical chemistry properties of MCM-41 can be modified, the catalytic performance can be enhanced, and the application domains can be extended. The catalysts with nickel have been applied widely in industries, such as hydrogenation of aromatic compounds [18], methanation of CO, steam reforming of hydrocarbons, amination of alcohols and carbon nanotube preparations etc [19–21]. The MCM-41 loaded nickel catalysts can be used in hydrogenation of gaseous acetonitrile [22], decomposition of amines [23], hydrogenation of benzene [24,25], oxygenation of benzene to prepare phenol and oxygenation of styrene to prepare benzaldehyde [26]. The activation energy of the reaction of hydrogenation of
N. Wu et al. / Microporous and Mesoporous Materials 185 (2014) 130–136
benzene on the Ni/MCM-41 catalyst is 36.5 kJ mol1, which is much lower than 50.5–52.3 kJ mol1 on the Ni/MyOx catalysts (MyOx is an oxide supporter, such as SiO2, Al2O3 and MgO etc.) [27]. So, MCM-41 supporter is very attractive for catalysts. The preparation methods of Ni/MCM-41 catalysts have been reported including impregnation methods with nickel nitrate or nickel citrate solutions [24,25,28–32], ion-exchange methods [33] and direct syntheses methods [26,34,35]. All the reports have been focused on the high disperse of metal nickel or NiOx on the supports and the ordered arrangement distribution of which were hardly ever mentioned. In the present paper, the NiII-MCM-41 precursors were firstly synthesized by direct hydrothermal method, in which the nickel atoms were highly dispersed and embedded into the framework of MCM-41, that is, the nickel atoms were placed into the crystal lattice of the MCM-41. Then the nickel atoms were reduced from the NiII-MCM-41 precursor in hydrogen flow at a moderate temperature without destroyed the structure of MCM-41. So the Ni atoms would be anchored to the framework of MCM-41 and the localization of the framework made agglomeration of the metal nanoparticles difficult, which ensures the nanoparticles stable in the molecular sieve framework. This method can be distinguished as the crystal lattice locating method. The catalytic performance of the result samples were examined by using the reaction of HDC of CB. The significance of the work is that high amount of metal nanoparticles can be highly dispersed in the structure of molecular sieve with an ordered distribution and the result materials possess the advantages of molecular sieves and metal nanoparticles for catalysis, simultaneously. 2. Materials and methods 2.1. Materials Nickel nitrate (AR), sodium hydroxide (AR, Beijing Yili Fine Chemical Product Limited Company, China); tetraethyl orthosilicate (TEOS, AR, Beijing Chemical Factory, China); ammonia solution (25%, AR, Beijing Beihua Fine Chemical Product Limited Company, China); cetyltrimethylammonium bromide (CTAB, AR), ethanol (AR), chlorobenzene (CB, AR, Tianjin Jinke Fine Chemical Institute, China), and hydrogen (>99.99%, Beijing Beiwen Gas Factory, China) were used. All solvents and reactants are commercially available and were used without further purification. 2.2. Methods 2.2.1. Synthesis of the precursor NiII-MCM-41 The NiII-MCM-41 mesoporous materials were prepared by a direct hydrothermal synthesis method. A typical synthesis was conducted as follows: CTAB was dissolved in deionized water in a flask with vigorous stirring for 30 min in order to obtain a clear solution. A certain amount of nickel nitrate was dispersed in deionized water in a beaker and an ammonia solution was added to obtain a solution containing the NiðNH3 Þ2þ 6 complex. The solution of NiðNH3 Þ2þ 6 was poured into the CTAB solution at room temperature. After stirred vigorously for 30 min, TEOS was added dropwise into the mixture and then stirred for another 30 min. The pH was adjusted to 10.0 by using an ammonia solution, and the final mixture was stirred for 5 h. The molar composition of the mixture was 1 TEOS: 0.152 CTAB: 2.8 NH3: 1/x Ni: 141.2 H2O. The mixture was transferred to a static Teflon-lined stainless steel autoclave and heated under autogenous pressure at 110 °C for 48 h. After that, the sample was filtered, washed with deionized water, dried overnight at 50 °C and finally calcined at 550 °C in air for 6 h. The NiII-MCM-41 precursor was gained.
131
2.2.2. Reduction of NiII from the NiII-MCM-41 precursor The synthesis of Ni/MCM-41 by reduction of the NiII-MCM-41 precursor was carried out as follow: NiII-MCM-41 was put into a quartz tube and then heated in hydrogen flow from room temperature to 750 °C at a heating rate of 5 °C/min and kept 750 °C for 1 h. Then, it was cooled naturally in hydrogen flow to the room temperature. The resulting sample was designated as A-i. The compared sample Ni/MCM-41 with the same nickel content corresponding to the sample A-i was prepared with the traditional impregnating method. Firstly, prepared a NiII/MCM-41 precursor with MCM-41 doped in a solution of NiðNH3 Þ2þ complex (same 6 as the above) and heated treatment. Then the NiII/MCM-41 precursor was reduced with the same process as the sample A-i. The resulting sample was designated as B-i. 2.2.3. Characterizations X-ray diffraction (XRD) patterns were recorded with a Rigaku D/ Max 2500 VBZ+/PC diffractometer using Cu Ka radiation (40 kV, 200 mA) in both low angle (2h range of 0.5–10°) and wide angle range (2h range of 10–80°). The X-ray fluorescence analysis (XRF) was performed on a Philips Magix-601 X-ray fluorescence spectrometer for determining the nickel content of the samples. The morphologies of the samples were examined by high-resolution transmission electron microscopy (HRTEM) on a JEM-3010 microscope with an accelerating voltage of 200 kV. Before the examination the samples were prepared by ultrasonic dispersion with absolute alcohol as solvent on a copper grid as supported membrane. The nitrogen adsorption–desorption isotherms were determined on a Micromeritics ASAP 2020 M volumetric adsorption analyzer. Before the nitrogen adsorption, each sample was degassed in a vacuum at 200 °C for 6 h. The specific surface area (SBET) was estimated by the BET equation. And the pore distribution, the mesopore analysis were obtained from the desorption branch of the isotherm using the Barrett–Joyner–Halenda (BJH) method. The X-ray photoelectron spectroscopy (XPS) analyses were conducted on an ESCALAB 250 spectrometer equipped with an Al Ka X-ray source. The carbon 1s peak at 284.6 eV was used as the reference for binding energies. 2.2.4. Catalytic performance testing The catalytic performance of the samples was tested through the reaction of HDC of CB. The experiments were carried out in a 250 mL stainless steel autoclave with ethanol as the solvent. The reaction conditions were as follows: reaction temperature was 100 °C, H2 partial pressure was 1.5 MPa, the catalyst dosage was 1.0 wt.%, and the reaction time was 60 min. The reactants and the products were analyzed by GC with a flame ionization detector online. And the conversion was calculated by the formula:
Conversion ð%Þ ¼
cC6 H5 Cl0 cC6 H5 Cli 100 cC6 H5 Cl0
cC6 H5 Cl0 and cC6 H5 Cli denotes the molar concentration of CB before and after reaction, respectively. 3. Results and discussion 3.1. Characterizations The nickel contents of the samples were determined by XRF and the results are listed in Table 1. The sample A-1, A-2 and A-3 in Table 1 is corresponding to the sample of nominal Si/Ni molar ratio of 10, 20, and 30, respectively. The actual nickel content in the sample A-1, A-2 and A-3 is 9.3%, 5.9%, and 3.4%(wt.), respectively. Fig. 1 shows the XRD patterns of the A samples. The small-angle XRD patterns (Fig. 1a) indicate that all the samples possess the
132
N. Wu et al. / Microporous and Mesoporous Materials 185 (2014) 130–136
Table 1 The textural properties of the A samples. Sample
Si/Ni
Ni wt.%
SBET/m2 g–1
Dp/nm
Vp/m3 g–1
DNi/nm
A-1 A-2 A-3 MCM-41[36]
10 20 30 –
9.3 5.9 3.4 0.0
404 574 644 1167
4.7 3.5 3.3 2.52
0.91 0.71 0.60 0.61
18.9 15.7 14.3 –
Fig. 2. Wide-angle XRD patterns of Si/Ni = 20 samples at different reduction temperature: a-wide angle patterns, b-low angle patterns.
Fig. 1. XRD patterns of A samples: a-low angle patterns, b-wide angle patterns.
high ordered structure of MCM-41 and the pore diameter of the samples increases with the nickel content increasing. The wide-angle XRD patterns (Fig. 1b) illuminate that all the samples except A-1⁄ (unreduced NiII-MCM-41 precursor of the sample A-1) show the characteristic diffraction peaks of metallic nickel (2h = 44.49°, 51.85°) and no characteristic diffraction peaks of the other nickel species, confirming that Ni(II) introduced into the skeletons of MCM-41 or highly dispersed in the matrix of MCM-41, which was illuminated by the XRD pattern of A-1⁄ without diffraction peaks of any nickel species, had been reduced. With the increasing of nickel content, the intensity of the diffraction peaks increase. The particle size of metallic nickel was calculated using the Scherrer formula (DNi = Kk/Bcosh, K is the Scherrer constant, 0.89; DNi is the particle size, nm; B is the full width of the half-height of the diffraction peak; h is the diffraction angle, and k is the X-ray wave
Fig. 3. XPS spectrum of sample A-1.
length, 0.154056 nm). The results are also listed in Table 1. It is clearly that the size of the metallic nickel particles increase slightly with the nickel content increasing, but the sizes of the particles in all the samples are smaller than 20 nm.
N. Wu et al. / Microporous and Mesoporous Materials 185 (2014) 130–136
The reduced conditions of Ni from NiII-MCM-41 precursor are very important for the aim materials preparation. Fig. 2 shows the XRD patterns of the Si/Ni = 20 samples gained at different reduction temperature, respectively. It is obviously that Ni(II) was not reduced from the NiII-MCM-41 precursor at 650 °C (Fig. 2a), and at 840 °C the structure of MCM-41 had been destroyed (Fig. 2b). 750 °C is the suitable reduction temperature for the preparation of the aim materials. The XPS spectrum of the sample A-1 (the highest nickel content sample in this work) is shown in Fig. 3 and the inset shows the subpeak results of Ni2p spectrum by Gaussian fitting. The predominant peak centered at 852.7 eV is the characteristic peak of the
133
metallic Ni2p3/2 [37,38]. There are no peaks at 857.8 eV and 855.0 eV, corresponding to the framework Ni2p3/2 and the nickel oxide Ni2p3/2 [32], respectively. The result confirms that the nickel atoms in the NiII-MCM-41 precursor were reduced to Ni0 thoroughly, which is consistent with the XRD analysis. Fig. 4 shows the HRTEM micrographs of all the A and B samples. The nominal nickel content in the sample A-i and in the corresponding sample B-i is the same. It is clear that the nickel particles in the A samples are very different from in the B samples. In the A samples, the size of the nickel particles (<20 nm) are much smaller than that in the B samples (>30 nm). The nickel particles in the A samples possess more uniform size and more regular shape than
Fig. 4. HRTEM images of the samples A and B.
134
N. Wu et al. / Microporous and Mesoporous Materials 185 (2014) 130–136
Fig. 5. N2 adsorption/desorption isotherms and the pore diameter distributions of samples.
in the B samples. In the A samples, the nickel particles are highly dispersed in an ordered distribution with about equal distance or multiple of the distance between the nickel particles in inside of the pore structure of the MCM-41, with scarcely any nickel particles dispersed on the external surface, which was very different from in the B samples with nickel particles stacked in a random way on the external surface, especially for the high nickel content B samples. With the nickel content increasing, the intervals decrease between the nickel particles, but the particle sizes were hardly changed in the A samples, and this should be decided by the lattice structure of the MCM-41. Additionally, the ordered hexagonal mesoporous structure of the MCM-41 was kept perfectly in the nickel reduced process, although the pore structure was slightly changed in the highest nickel content sample A-1. And the results are in agreement with the XRD analysis (Fig. 1a). The characters of the ordered distribution of the nickel nanoparticles
and the ordered pore structure of the MCM-41 will be very significative for the catalytic performance. Fig. 5 shows the N2 adsorption–desorption isotherms of the A samples, the corresponding textural properties are given in Table 1. The curves of all the three samples are Type IV, which is the typical adsorption curve of the mesoporous materials. As shown in Table 1, the pore diameter (Dp, nm) of the samples increase along with the increasing of nickel content. This is because the nickel particles expanded the cavities of the samples and resulted in the widening of the pore diameter (see the forming mechanism in Scheme 1). Meanwhile, the specific surface area (SBET, m2 g1) of the samples decrease along with the increasing of nickel content, the primary reason is that the atomic weight of nickel is larger than that of Si. Of course, the presence of more nickel particles in the higher nickel content samples may overlay some internal surface and result in decreasing of the specific surface area in a certain extent. Through the above characterizations, the formation mechanism of the samples was proposed as shown in Scheme 1. Firstly, the mesoporous material with hexagonal pore structure was successfully synthesized through a direct hydrothermal synthesis method and all the Ni(II) ions were uniformly dispersed in the silica framework. After the Ni(II) species were reduced to Ni(0), the metallic Ni(0) atoms would transfer along with the channel to form nickel atom clusters and then become bigger particles at the high reduced temperature. Because of the strong restrictive effect of the silica framework, the nickel particle diameters were restricted and could not form big nickel particles under the experimental conditions. As the nickel atoms reduced, the sites of the Ni(II) ions would be replaced by H atoms and form a new structure as
, which could not result in the framework
breakage to a considerable degree. The nickel particles must be implantation in some crystal lattice nodes in which the nickel atoms were reduced, and only in these sites the nickel particles could grow into the biggest size without the framework damaged. In addition, the size of the nickel particles is correlated with the pore diameter of the precursor NiII-MCM-41. The big atoms (Ni)
Scheme 1. Schematic diagram of the Ni/MCM-41 samples preparation.
N. Wu et al. / Microporous and Mesoporous Materials 185 (2014) 130–136
substitute the small atoms (Si) of the framework will enlarge the pore diameter of the MCM-41 and the more Si atoms substituted by nickel atoms at the range of the experimental contents the bigger pore diameter enlarged [39]. So the nickel particle diameter in the sample A-1 is the biggest, and in the samples A-2 and A-3 the particle diameters are smaller with a small difference. The distribution of the nickel particles can be illuminated as the schematic diagram 1, 2 and 3 in Scheme 1 for the sample A-1, A-2 and A-3, respectively, which with an ordered distribution. 3.2. Catalytic performance HDC of CB were carried out with the A samples and the comparison B samples as the catalysts under the same conditions. HDC of CB normally results in the production of three products, viz. benzene, cyclohexane, and chlorocyclohexane [33]. But over the A and B samples, the product was only benzene, without cyclohexane and chlorocyclohexane, under the experimental temperature 100 °C, which can be attributed not only to the higher ordered pore structure of MCM-41, but also to the lower reaction temperature than that of the previous reports. The catalytic performances are shown in Fig. 6. It can be seen that all the A samples show much better catalytic performance than the corresponding B samples. The conversion of CB was up to 98.2%, 97.6% and 82.5% over the sample A-1, A-2 and A-3, which was much higher than 70.2%, 64.8% and 63.2% over the corresponding sample B-1, B-2 and B-3, respectively. The A samples exhibited much higher catalytic activity than the B samples, which can be attributed to the smaller nickel nanoparticles with high dispersion in an ordered distribution manner in the high ordered pore structure of MCM-41. The reusability of the A and B samples were also examined under the same conditions and the results show that the catalytic performance of the three A samples had no significant decrease after four runs, the conversion of CB over the sample A-1, A-2 and A-3 was 97.4%, 97.1% and 81.9%, respectively. And over the sample B-1, B-2 and B-3, the conversion of CB was about 42.1%, 23.2% and 15.6%, respectively, which indicates that the reusability of the A catalysts are much better than that of the B samples. This is because the nickel particles were ordered encapsulated into the frameworks of the MCM-41 in the A samples, resulting in the stronger interaction between the nickel particles and the MCM41 supporter, and the B samples were not so. The catalyst deactivation is generally a problem in the HDC reaction, and it has been attributed mainly to the presence of HCl generated in the reaction,
Fig. 6. Catalytic results of all the samples.
135
which leads to the formation of a stable surface chloride species. The enhanced stability of the A samples is consistent with the smaller crystallite size, higher nickel dispersion, higher ordered distribution of the nickel nanoparticles, and the stronger interaction between the nickel species and the MCM-41 supporter. The spent catalysts were also characterized by XRD and XRF. The XRF results indicated no evidence for nickel loss of the A catalysts, and for the B samples the nickel loss was obviously during the reaction. In addition, the characteristic reflections of NiCl2 can be observed in the XRD patterns of the used B catalysts, which indicated the poisoning of active nickel species with HCl produced during HDC reaction. This can be ascribed to the weak interactions between nickel species and MCM-41 supporter in the B catalysts. In contrast, the used A catalysts did not show any new peaks in its XRD patterns. So, the deactivation observed for the B catalysts can be attributed not only to the losing of nickel metal atoms, but also to the poisoning of active nickel species with HCl. 4. Conclusions The highly dispersed nickel nanoparticles embedded in the mesoporous molecular sieve MCM-41 have been successfully prepared by the crystal lattice locating method. The synthesized samples possess an ordered hexagonal mesostructure of MCM-41 with highly dispersed nickel nanoparticles, which distributed in an ordered manner. The diameter of the nickel particles is smaller than 20 nm and firmly studded in the frameworks of MCM-41 in an ordered distribution. The result samples show an excellent catalytic performance, with higher activity, selectivity, and stability in HDC of CB at the low temperature. Acknowledgements This work was financially supported by the National Natural Science of Foundation of China (No. 21271017) and the National Basic Research Program of China (973 Program, Grant No. 2011CBA00506). References [1] A.B. Sorokin, A. Tuel, Catal. Today 57 (2000) 45. [2] J. Gu, Y. Huang, S.P. Elangovan, Y. Li, W. Zhao, I. Toshio, Y. Yamazaki, J. Shi, J. Phys. Chem. C 115 (2011) 21211. [3] D.J. Kim, J.W. Kim, S.J. Choung, M. Kang, J. Ind. Eng. Chem. 14 (2008) 308. [4] R. Nares, J. Ramirez, A. Gutierrez-Alejandre, R. Cuevas, Ind. Eng. Chem. Res. 48 (2008) 1154. [5] J. Gao, Z. Hou, J. Guo, Y. Zhu, X. Zheng, Catal. Today 131 (2008) 278. [6] J. Gao, D. Liang, Z. Hou, J. Fei, X. Zheng, Int. J. Hydrogen Energy 33 (2008) 5493. [7] J.H. Kim, D.J. Suh, T.J. Park, K.L. Kim, Appl. Catal. A 197 (2000) 191. [8] J. Chen, Y. Zhang, L. Tan, Y. Zhang, Ind. Eng. Chem. Res. 50 (2011) 4212. [9] B. Qiao, A. Wang, X. Yang, L.F. Allard, Z. Jiang, Y. Cui, J. Liu, J. Li, T. Zhang, Nat. Chem. 3 (2011) 634. [10] J.S. Beck, J.C. Vartuli, W.J. Roth, M.E. Leonowicz, C.T. Kresge, K.D. Schmitt, C.T. Chu, D.H. Olson, E.W. Sheppard, S.B. McCullen, J. Am. Chem. Soc. 114 (1992) 10834. [11] B.S. Li, J.Q. Xu, J. Liu, S. Zuo, Z. Pan, Z. Wu, J. Colloid Interface Sci. 366 (2012) 114. [12] Y. Fazaeli, M.M. Amini, E. Mohajerani, M. Sharbatdaran, N. Torabi, J. Colloid Interface Sci. 346 (2010) 384. [13] B.S. Li, K. Wu, T.H. Yuan, C.Y. Han, J.Q. Xu, X.M. Pang, Microporous Mesoporous Mater. 151 (2012) 277. [14] J. Iglesias, J.A. Melero, M. SánchezSánchez, Microporous Mesoporous Mater. 132 (2010) 112. [15] C.Z. Loebick, S. Lee, S. Derrouiche, M. Schwab, Y. Chen, G.L. Haller, L. Pfefferle, J. Catal. 271 (2010) 358. [16] K. Parida, K.G. Mishra, S.K. Dash, J. Hazard. Mater. 241–242 (2012) 395. [17] S.G. Aspromonte, Á. Sastre, A.V. Boix, M.J. Cocero, E. Alonso, Microporous Mesoporous Mater. 148 (2012) 53. [18] S. Narayanan, R. Unnikrishman, V. Vishwanathan, Appl. Catal. A 129 (1995) 9. [19] R.L. Vander Wal, T.M. Ticich, V.E. Curtis, Carbon 39 (2001) 2277. [20] J.T. Feng, Y.J. Lin, D.G. Evans, X. Duan, D.Q. Li, J. Catal. 266 (2009) 351. [21] J. Wang, G. Fan, H. Wang, F. Li, Ind. Eng. Chem. Res. 50 (2011) 13717.
136
N. Wu et al. / Microporous and Mesoporous Materials 185 (2014) 130–136
[22] P. BraosGarcía, P. MairelesTorres, E. RodríguezCastellón, A. JiménezLópez, J. Mol. Catal. A: Chem. 193 (2003) 185. [23] X.K. Li, W.J. Ji, J. Zhao, S.J. Wang, C.T. Au, J. Catal. 236 (2005) 181. [24] A. Lewandowska, S. Monteverdi, M. Bettahar, M. Ziolek, J. Mol. Catal. A: Chem. 188 (2002) 85. [25] R. Wojcieszak, S. Monteverdi, M. Mercy, I. Nowak, M. Ziolek, M.M. Bettahar, Appl. Catal. A 268 (2004) 241. [26] V. Parvulescu, B.L. Su, Catal. Today 69 (2001) 315. [27] P. Marécot, E. Paraiso, J.M. Dumas, J. Barbier, Appl. Catal. A 80 (1992) 79. [28] A. InfantesMolina, J. MéridaRobles, P. BraosGarcía, E. RodríguezCastellón, E. Finocchio, G. Busca, P. MairelesTorres, A. JiménezLópez, J. Catal. 225 (2004) 479. [29] K. Fang, J. Ren, Y. Sun, J. Mol. Catal. A: Chem. 229 (2005) 51. [30] D. ElicheQuesada, J. MéridaRobles, P. MairelesTorres, E. RodríguezCastellón, A. JiménezLópez, Langmuir 19 (2003) 4985.
[31] D.J. Lensveld, J. Gerbrand Mesu, A. Jos Van Dillen, K.P. De Jong, Microporous Mesoporous Mater. 44-45 (2001) 401. [32] J. Jia, Y. Wang, E. Tanabe, T. Shishido, K. Takehira, Microporous Mesoporous Mater. 57 (2001) 283. [33] M. Hartmann, A. Pöppl, L. Kevan, J. Phys. Chem. 100 (1996) 9906. [34] W. Wang, M. Song, Mater. Res. Bull. 40 (2005) 1737. [35] Y. Yang, S. Lim, G. Du, Y. Chen, D. Ciuparu, G.L. Haller, J. Phys. Chem. B 109 (2005) 13237. [36] B.S. Li, W. Ma, C.Y. Han, J.J. Liu, X.M. Pang, X.H. Gao, Microporous Mesoporous Mater. 156 (2012) 73. [37] J.L. Rousset, F.J. Cadete Santos Aires, B.R. Sekhar, P. Mélinon, B. Prevel, M. Pellarin, J. Phys. Chem. B 104 (2000) 5430. [38] P.E.R. Blanchard, A.P. Grosvenor, R.G. Cavell, A. Mar, Chem. Mater. 20 (2008) 7081. [39] X. Li, B.S. Li, H.H. Mao, A. Tufail, Shah, J. Colloid Interface Sci. 332 (2009) 444.