- Email: [email protected]
SBA-15 supported on carbon nanofibers composite
Catalysis Communications 12 (2011) 790–793
Contents lists available at ScienceDirect
Catalysis Communications j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / c a t c o m
Short Communication
Benzoylation of anisole catalyzed by Ga/SBA-15 supported on carbon nanofibers composite F.Z. El Berrichi a,⁎, C. Pham-Huu b, L. Cherif a, B. Louis b, M.J. Ledoux b a b
Laboratoire de Catalyse et Synthèse Organique, Faculté des Sciences, Université de Tlemcen BP119, Algeria Laboratoire des Matériaux, Surfaces et Procédés pour la Catalyse, UMR 7515, ECPM, 25 rue Becquerel, F-67087 Strasbourg Cedex 02, France
a r t i c l e
i n f o
Article history: Received 10 November 2010 Received in revised form 22 January 2011 Accepted 25 January 2011 Available online 3 February 2011 Keywords: SBA-15 Carbon nanofibers Gallium Supported SBA-15 Benzoylation
a b s t r a c t Carbon nanofiber composite (C-NFC) shows several advantages compared to the conventional supports which are usually employed in catalysis such as alumina, silica or activated charcoal. In this present work we have developed a new hybrid catalyst consisting of SBA-15 supported on C-NFC for the benzoylation reaction. The structured materials allow an important improvement of the reaction hydrodynamics and favor the mass transfer between the active phase and the reactants, especially in the liquid-phase medium, thus improving the selectivity toward desired products and increase the catalyst stability. The Ga/SBA-15(10)/C-NFC composite was further used as catalyst for Friedel–Crafts acylation of anisole in slurry bed. Cycling tests confirm the fradual deactivation of the catalyst which could be attributed to the pore plugging by carbonaceous residue which render inaccessible the active site localized inside the mesoporous channel to the reactant. These active sites localized inside the mesoporous channel are expected to be responsible for the para product formation by shape selectivity and thus, only the ortho product, which is formed on the gallium sites localized on the pore mouth, is observed on the deactivated catalyst. Such hypothesis is confirmed by depositing gallium phase on the poreless carbon nanofiber. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Mesoporous molecular sieves represent a new class of inorganic materials. An over increasing scientific interest has been initiated since mesoporous molecular sieves such as hexagonally ordered MCM41 were discovered by Mobil Corporation scientists in 1992 [1,2]. The unique characteristics of mesoporous molecular sieves, such as high specific surface area, large pore volume and well defined pore size, make these materials a good candidate for use in several potential applications such as catalysis, adsorption, and nanoelectronics, etc. However, early reported mesoporous molecular sieves suffer from low thermal and hydrothermal stabilities, which limit their applications. In 1998, a new synthesis of ordered hexagonal mesoporous silica was proposed by Stucky and co-workers [3]. They prepared a material named SBA-15 by using triblock poly (ethylene oxide)-poly (propylene oxide)-poly (ethylene oxide) co-polymers as the structure-directing agent in an acidic medium. The SBA-15 material provides a thermal and hydrothermal stability that exceeds those for the thinner walled MCM-41 materials [3]. The Friedel–Crafts acylation reaction of aromatic compounds represents one of the most important methods of production of aromatic ketones which are intermediate products in several fine
⁎ Corresponding author. E-mail address: [email protected] (F.Z. El Berrichi). 1566-7367/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.catcom.2011.01.022
chemical processes. Several reports have recently been published on the replacement of traditional AlCl3 or FeCl3 homogeneous catalysts by heterogeneous solid acids like GaSBA-15 and AlSBA-15 [4–6]. Due to its chemical inertness and high thermal resistance, carbon nanofibers composite (C-NFC) represents a highly promising candidate as catalyst support material for many potential commercial catalytic applications [7–9]. The present work reports the synthesis of SBA-15 on the C-NFC composite support with controlled macroscopic shape. The influence of the synthesis duration on the mesoporous loading and mechanical strength of the final composite was evaluated. Finally, the SBA-15/C-NFC doped with gallium catalyst was evaluated in the Friedel–Crafts acylation reaction of anisole. The stability of the catalyst was evaluated by submitting it to several cycling tests. 2. Experimental Carbon nanofiber composites with macroscopic shaping, i.e. carbon felt with low specific surface area (1 m2 g− 1), were successfully synthesized by chemical vapor decomposition (CVD) using a low loading in nickel as growth catalyst (≤1 wt.%). Before the CNTs deposition, the carbon felt was treated in an aqua regia medium at 80 °C for 2 h in order to remove the impurities. After the acid treatment the sample was washed several times with de-ionized water until neutral pH was reached. A high carbon nanofiber yield, i.e. 100 wt.% per hour of synthesis, with respect to the initial catalyst weight was obtained [10]. Mesoporous SBA-15 supported on C-NFC
F.Z. El Berrichi et al. / Catalysis Communications 12 (2011) 790–793
has been synthesized by classical means according to the hydrothermal process described by Zhao and coll. [3]. For the synthesis, 1.35 g of support (one disk of carbon nanofibers composite) was added to 50 ml of the solution. Afterwards, the mixture was poured in a Teflonlined autoclave and kept for 72 h at 100 °C. The SBA-15/C-NFC composite obtained was washed with distillated water, filtrated, and sonicated for 20 min to remove the loosely attached SBA-15 material on the composite surface. Finally, the composite was calcined in air at 450 °C for 6 h to remove the organic surfactant polymer. The amount of SBA-15 coated on the carbon nanofibers composite support was 14 wt.%. Afterwards, SBA-15/C-NFC materials were added in an aqueous solution of gallium nitrate (containing the appropriate Ga amount) for 12 h, to get a final Ga/SBA-15 mesoporous material having an atomic ratio of Si/Ga = 10. After filtration, washing, the solid was dried in air at room temperature. The Si/Ga was set to 10, on the basis of our previous study devoted to Friedel–Crafts activity of Ga/SBA-15 [4]. It is worth mentioning that Si/Ga = 10 corresponds to the value inside the gel, since it cannot be estimated in the composite Ga/ SBA-15/C-NFC. The stability of the Ga/SBA-15(10)/C-NFC material has been compared to that of Ga/C-NFC also prepared by post-treatment with the same Si/Ga ratio, impregnation time and repeated twice. The specific surface area of the Ga/SBA-15(10)/C-NFC and Ga/C-NFC catalysts was performed by nitrogen adsorption–desorption isotherms at 77 K (Micromeritics Tristar 3000) while the gross morphology was examined by scanning electron microscopy (SEM) (JEOL XL 30 FEG). Before observation the sample was covered with a thin layer of gold in order to avoid the problem of charge. The benzoylation of anisole was performed in liquid-phase medium in a batch reactor under stirring at 140 °C. A mixture of anisole (0.069 ± 0.002 mol) and benzoyl chloride (0.025 ± 0.001 mol) were added in a Schlenk reactor at the reaction temperature. The weight of the catalyst was fixed at 200 ± 1 mg. Benzoyl chloride conversion was followed by gas chromatography (GC, Varian GC-3400 CX) by withdrawing small samples at regular time intervals.
791
SEM micrographs showed that the presence of nanofibers from the C-NFC covered by SBA-15 spherules decorated the carbon nanofiberbased composite (Fig. 1A and B). At low magnification, one can observe the spherules of SBA-15 agglomerates on the nanofibers surface (Fig. 1A). High magnification SEM micrograph (Fig. 1B) allows one to observe the relatively well dispersion of the SBA-15 nanoparticles on the surface of the carbon nanofibers. The high dispersion of the SBA-15 crystallites on the CNF surface could be attributed to the presence of the prismatic planes on the CNF surface which provides a large number of anchorage sites for the SBA-15. One should also note the high anchorage strength of the SBA-15 nanoparticles on the carbon nanofibers surface as lamost no matter loss was observed after sonication treatment for 10 min. The catalyst stability was evaluated after four consecutive catalytic tests. Fig. 2 shows a drastic decrease in the conversion between the first and second run for the Ga/SBA-15 (10)/C-NFC catalyst. The conversion decreases from 88% during the first test to about 50% for the second test. The selectivity was 93% towards the para-methoxybenzophenone, and 6% for the ortho-methoxybenzophenone. These results are in agreement with previous data in the electrophilic aromatic substitution [14]. The deactivation of the catalyst is mainly due to coking and also a leaching of active species by HCl produced during the reaction. The deactivation by carbonaceous residue deposit is either due to pore blocking or active site poisoning [15]. The pore blocking seems to be the main reason for explaining the catalyst deactivation behavior [16]. The deactivated catalyst can be effectively regenerated by air calcinations to remove the
3. Results and discussion The as-synthesized C-NFC composite was sonicated in an ethanol medium for 30 min in order to remove the loosely attached carbon nanofibers from the composite and also to check the mechanical anchorage strength of theses CNFs on the macroscopic host matrix. The CNFs appear to be well anchored on the felt host structure whereas only less than 10 wt.% of the CNFs was removed from the composite during the sonication treatment. The high mechanical stability of the VA-CNTs on the host surface is attributed to the fact that part of the CNTs, located next to the nickel catalyst sitting on the host matrix surface, was partly penetrated into the host matrix allowing the generation of a strong anchorage of the CNT onto the host matrix and prevents the CNT loss during the sonication treatment. In some case the CNF could also anchor on the macroscopic host structure via an intermediate phase, i.e. microporous carbon layer, which is strongly attached to the host matrix underneath [11]. The specific surface area of the C-NFC composite significantly increased and reaches 96 m2 g− 1. Since the initial specific surface area of the host carbon felt is about 1 m2 g− 1, it is clear that the incremental specific surface area is caused by the intrinsic external surface of the carbon nanofibers, which is usually ranged between 80 and 200 m2 g− 1 [12,13]. The coating of SBA-15 material on the C-NFC support increases the specific surface area of the final composite from 96 to 143 m2/g after one-coating step. The SBA-15 loading determined was amounted to about 14 wt.% in the final SBA-15/C-NFC composite. This high amount could be explained by the high anchoring sites present on the NFC surface, i.e. prismatic planes, which favor the nucleation and growth of the SBA-15 seeds on the carbon-based support.
Fig. 1. SEM micrographs of the SBA-15/C-NFC composite at (A) low magnification and (B) high magnification showing the entangled morphology of the carbon nanofibers on the composite surface.
792
F.Z. El Berrichi et al. / Catalysis Communications 12 (2011) 790–793 Table 2 Benzoylation activity, expressed in terms of conversion, and selectivity on the Ga/SBA15(10)/C-NFC and Ga/C-NFC catalysts. Reaction conditions: catalyst weight = 200 mg, reaction temperature = 140 °C, anisole = 0.069 mol, benzoyl chloride = 0.025 mol. Catalyst
Ga/SBA-15(10)/C-NFC Ga/C-NFC a b
Fig. 2. PhCOCl conversion as a function of time and the conversion evaluation as a function of the cycling tests for the Ga/SBA-15 (10)/C-NFC catalyst.
carbonaceous residues on its surface similar to that applied for the bulk Ga-SBA-15 catalyst [4]. Table 1 shows the gradual decrease of the selectivity in paramethoxybenzophenone (para-ketone) during the cycling tests, thus raising the selectivity toward ortho-methoxybenzophenone (orthoketone). The increase of the ortho-ketone selectivity as a function of the cycling tests could be attributed to the fact that these molecules were not formed on the same active site, i.e. Ga localized on the surface pore mouth (low benzoylation activity) or inside the channel (high benzoylation activity). According to the results one should expected that the para-ketone is mainly formed on the active sites localized inside the channel of the catalyst, where the shape selectivity is controlled by the pore channel. Such site is more sensitive to the deactivation by coke formation and pore plugging and thus, the para-ketone selectivity slowly decrease as a function of the cycling tests. On the other hand, the ortho-ketone which is formed on the outer surface or next to the pore mouth is less sensitive to the deactivation phenomenon. After four cycles it is expected that most of the active site localized inside the pore channel is not accessible anymore and thus, shape selectivity towards para-ketone no longer operates. According to the results observed one can state that Ga sites localized inside the SBA-15 channel only give rise to para-ketone owing to the shape selectivity where Ga sites localized on the outer surface could be active either for para- or ortho-ketone with even selectivity. Table 2 presents the benzoylation activity and stability comparison between Ga/SBA-15(10)/C-NFC and Ga/C-NFC catalysts under the same reaction conditions. According to the observed results the former is more active and more stable than the later. Indeed, during the first test, benzoyl chloride conversion reached 88% on the Ga/SBA-15(10)/C-NFC catalyst against 39% for the Ga/C-NFC catalyst keeping the reaction conditions similar. The higher performance of the Ga/SBA-15(10)/
Conversion (%)a
Selectivity (%) Para-ketone
Ortho-ketone
Esterb
88 39
93 49
6 48
1 3
: Conversion after 24 h of the reaction. : Phenylbenzoate selectivity.
C-NFC catalyst when compared to the Ga/C-NFC catalyst is probably due to an improved dispersion of the active phase on the SBA-15 surface. The direct comparison of the ketones selectivity between the two catalysts is not straightforward taking into account the large difference in terms of conversion. However, the following hypothesis could be made regarding the ketones selectivity and the Ga localization: the high selectivity towards para-ketone observed on the Ga/SBA-15/C-NFC compared to that observed on the Ga/C-NFC, where all the Ga sites are expected to be localized on the outer surface of the support, taking into account the complete absence of channel in the CNF, confirms again the importance of the Ga site localization on the ketone, i.e. para- versus ortho-ketone, selectivity. Such result also confirms the hypothesis advanced above regarding the para-ketone selectivity change as a function of the deactivation due to the pore plugging. 4. Conclusion The comparison between the catalytic performances of Ga/SBA-15/ C-NFC and Ga/C-NFC catalysts indicates that the CNF-based catalyst exhibits a higher catalytic activity compared to the one deposited on a felt support with lower effective surface area. The coating of SBA-15 on macroscopic bodies, like C-NFC composite, also allows an ease in the catalyst recovery, thus excluding important problems linked with binder use or post-reaction filtration. Cycling tests indicate that deactivation occurs probably due to the gradual plugging of the SBA-15 channel by residual carbonaceous residue. The direct consequence of the channel plugging is the gradual decrease of the paraketone selectivity due to the lost of the gallium active sites localized within the channel. The macroscopic shaping of the support could allow one to envisage the use of this nano-macro composite catalyst in a fixed-bed configuration instead of batch configuration where deactivation by carbonaceous residues could become significant. Acknowledgments The SEM experiments were carried out at the facilities of the IPCMS (UMR 7504 of the CNRS and University of Strasbourg). References
Table 1 Ga/SBA-15(10)/C-NFC stability in the anisole benzoylation reaction as a function of the cycling tests. Reaction conditions: catalyst weight=200 mg, reaction temperature=140 °C, anisole= 0.069 mol, benzoyl chloride =0.025 mol. Catalyst
Ga/SBA-15(10)/ C-NFC
a b
Number of catalytic run
Conversion (%)a
Selectivity (%) Para-ketone
Ortho-ketone
Esterb
1 2 3 4
88 49 58 51
93 70 45 42
6 28 54 57
1 2 1 1
: Conversion after 24 h of the reaction. : Phenylbenzoate selectivity.
[1] C.T. Kresge, M.E. Leonowicz, W.J. Roth, J.C. Vartuli, J.S. Beck, Nature 359 (1992) 710. [2] J.S. Beck, J.C. Vartuli, W.J. Roth, M.E. Leonowicz, C.T. Kresge, K.D. Schmitt, C.T.-W. Chu, D.H. Olson, E.W. Sheppard, S.B. McCullen, J.B. Higgins, J.L. Schlenker, A new family of mesoporous molecular sieves prepared with liquid crystal templates, J. Am. Chem. Soc. 114 (1992) 10834. [3] D. Zhao, J. Feng, Q. Huo, N. Melosh, G.H. Frederickson, B.F. Chmelka, G.D. Stucky, Science 279 (1998) 548. [4] E. BerrichiZ. , L. Cherif, O. Orsen, J. Fraissard, J.P. Tessonnier, E. Vanhaecke, B. Louis, M.J. Ledoux, C. Pham-Huu, Ga doped SBA-15 as an active and stable catalyst for Friedel–Crafts liquid-phase acylation, Appl. Catal. A 298 (2006) 194. [5] E. BerrichiZ. , B. Louis, O. Orsen, J.P. Tessonnier, L. Cherif, M.J. Ledoux, C. Pham-Huu, One-pot synthesis of Ga-SBA-15-post-treated SBA-15 catalysts, Appl. Catal. A 316 (2007) 219. [6] A. Vinu, D.P. Savant, K. Arigo, M. Hartmann, S.B. Halligudi, Micropor. Mesopor. Mater. 80 (2005) 195. [7] P. Serp, M. Corrias, P. Kalck, Appl. Catal. A 253 (2003) 337.
F.Z. El Berrichi et al. / Catalysis Communications 12 (2011) 790–793 [8] [9] [10] [11] [12]
J.K. Chinthaginjala, K. Seshan, L. Lefferts, Ind. Eng. Chem. Res. 46 (2007) 3968. R. Vieira, D. Bastos-Netto, M.J. Ledoux, C. Pham-Huu, Appl. Catal. A 279 (2005) 35. R. Vieira, Thèse, Université de Strasbourg, 2003. J.H. Bitter, J. Mater. Chem. 20 (2010) 7312. C. Pham-Huu, M.J. Ledoux, Top. Catal. 40 (2006) 49.
793
[13] P.W.A.M. Wenmakers, J. van der Schaaf, B.F.M. Kuster, J.C. Shouten, J. Mater. Chem. 18 (2008) 2426. [14] G.A. Olah, R. Malhotra, S.C. Narang, J.A. Olah, Synthesis (1978) 672. [15] J.W. Beekman, G.F. Froment, Ind. Eng. Fundam. 18 (1979) 245. [16] H. Beinaert, R. Vermeuleu, G.F. Froment, Stud. Surf. Sci. Catal. 88 (1994) 97.
Contents lists available at ScienceDirect
Catalysis Communications j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / c a t c o m
Short Communication
Benzoylation of anisole catalyzed by Ga/SBA-15 supported on carbon nanofibers composite F.Z. El Berrichi a,⁎, C. Pham-Huu b, L. Cherif a, B. Louis b, M.J. Ledoux b a b
Laboratoire de Catalyse et Synthèse Organique, Faculté des Sciences, Université de Tlemcen BP119, Algeria Laboratoire des Matériaux, Surfaces et Procédés pour la Catalyse, UMR 7515, ECPM, 25 rue Becquerel, F-67087 Strasbourg Cedex 02, France
a r t i c l e
i n f o
Article history: Received 10 November 2010 Received in revised form 22 January 2011 Accepted 25 January 2011 Available online 3 February 2011 Keywords: SBA-15 Carbon nanofibers Gallium Supported SBA-15 Benzoylation
a b s t r a c t Carbon nanofiber composite (C-NFC) shows several advantages compared to the conventional supports which are usually employed in catalysis such as alumina, silica or activated charcoal. In this present work we have developed a new hybrid catalyst consisting of SBA-15 supported on C-NFC for the benzoylation reaction. The structured materials allow an important improvement of the reaction hydrodynamics and favor the mass transfer between the active phase and the reactants, especially in the liquid-phase medium, thus improving the selectivity toward desired products and increase the catalyst stability. The Ga/SBA-15(10)/C-NFC composite was further used as catalyst for Friedel–Crafts acylation of anisole in slurry bed. Cycling tests confirm the fradual deactivation of the catalyst which could be attributed to the pore plugging by carbonaceous residue which render inaccessible the active site localized inside the mesoporous channel to the reactant. These active sites localized inside the mesoporous channel are expected to be responsible for the para product formation by shape selectivity and thus, only the ortho product, which is formed on the gallium sites localized on the pore mouth, is observed on the deactivated catalyst. Such hypothesis is confirmed by depositing gallium phase on the poreless carbon nanofiber. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Mesoporous molecular sieves represent a new class of inorganic materials. An over increasing scientific interest has been initiated since mesoporous molecular sieves such as hexagonally ordered MCM41 were discovered by Mobil Corporation scientists in 1992 [1,2]. The unique characteristics of mesoporous molecular sieves, such as high specific surface area, large pore volume and well defined pore size, make these materials a good candidate for use in several potential applications such as catalysis, adsorption, and nanoelectronics, etc. However, early reported mesoporous molecular sieves suffer from low thermal and hydrothermal stabilities, which limit their applications. In 1998, a new synthesis of ordered hexagonal mesoporous silica was proposed by Stucky and co-workers [3]. They prepared a material named SBA-15 by using triblock poly (ethylene oxide)-poly (propylene oxide)-poly (ethylene oxide) co-polymers as the structure-directing agent in an acidic medium. The SBA-15 material provides a thermal and hydrothermal stability that exceeds those for the thinner walled MCM-41 materials [3]. The Friedel–Crafts acylation reaction of aromatic compounds represents one of the most important methods of production of aromatic ketones which are intermediate products in several fine
⁎ Corresponding author. E-mail address: [email protected] (F.Z. El Berrichi). 1566-7367/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.catcom.2011.01.022
chemical processes. Several reports have recently been published on the replacement of traditional AlCl3 or FeCl3 homogeneous catalysts by heterogeneous solid acids like GaSBA-15 and AlSBA-15 [4–6]. Due to its chemical inertness and high thermal resistance, carbon nanofibers composite (C-NFC) represents a highly promising candidate as catalyst support material for many potential commercial catalytic applications [7–9]. The present work reports the synthesis of SBA-15 on the C-NFC composite support with controlled macroscopic shape. The influence of the synthesis duration on the mesoporous loading and mechanical strength of the final composite was evaluated. Finally, the SBA-15/C-NFC doped with gallium catalyst was evaluated in the Friedel–Crafts acylation reaction of anisole. The stability of the catalyst was evaluated by submitting it to several cycling tests. 2. Experimental Carbon nanofiber composites with macroscopic shaping, i.e. carbon felt with low specific surface area (1 m2 g− 1), were successfully synthesized by chemical vapor decomposition (CVD) using a low loading in nickel as growth catalyst (≤1 wt.%). Before the CNTs deposition, the carbon felt was treated in an aqua regia medium at 80 °C for 2 h in order to remove the impurities. After the acid treatment the sample was washed several times with de-ionized water until neutral pH was reached. A high carbon nanofiber yield, i.e. 100 wt.% per hour of synthesis, with respect to the initial catalyst weight was obtained [10]. Mesoporous SBA-15 supported on C-NFC
F.Z. El Berrichi et al. / Catalysis Communications 12 (2011) 790–793
has been synthesized by classical means according to the hydrothermal process described by Zhao and coll. [3]. For the synthesis, 1.35 g of support (one disk of carbon nanofibers composite) was added to 50 ml of the solution. Afterwards, the mixture was poured in a Teflonlined autoclave and kept for 72 h at 100 °C. The SBA-15/C-NFC composite obtained was washed with distillated water, filtrated, and sonicated for 20 min to remove the loosely attached SBA-15 material on the composite surface. Finally, the composite was calcined in air at 450 °C for 6 h to remove the organic surfactant polymer. The amount of SBA-15 coated on the carbon nanofibers composite support was 14 wt.%. Afterwards, SBA-15/C-NFC materials were added in an aqueous solution of gallium nitrate (containing the appropriate Ga amount) for 12 h, to get a final Ga/SBA-15 mesoporous material having an atomic ratio of Si/Ga = 10. After filtration, washing, the solid was dried in air at room temperature. The Si/Ga was set to 10, on the basis of our previous study devoted to Friedel–Crafts activity of Ga/SBA-15 [4]. It is worth mentioning that Si/Ga = 10 corresponds to the value inside the gel, since it cannot be estimated in the composite Ga/ SBA-15/C-NFC. The stability of the Ga/SBA-15(10)/C-NFC material has been compared to that of Ga/C-NFC also prepared by post-treatment with the same Si/Ga ratio, impregnation time and repeated twice. The specific surface area of the Ga/SBA-15(10)/C-NFC and Ga/C-NFC catalysts was performed by nitrogen adsorption–desorption isotherms at 77 K (Micromeritics Tristar 3000) while the gross morphology was examined by scanning electron microscopy (SEM) (JEOL XL 30 FEG). Before observation the sample was covered with a thin layer of gold in order to avoid the problem of charge. The benzoylation of anisole was performed in liquid-phase medium in a batch reactor under stirring at 140 °C. A mixture of anisole (0.069 ± 0.002 mol) and benzoyl chloride (0.025 ± 0.001 mol) were added in a Schlenk reactor at the reaction temperature. The weight of the catalyst was fixed at 200 ± 1 mg. Benzoyl chloride conversion was followed by gas chromatography (GC, Varian GC-3400 CX) by withdrawing small samples at regular time intervals.
791
SEM micrographs showed that the presence of nanofibers from the C-NFC covered by SBA-15 spherules decorated the carbon nanofiberbased composite (Fig. 1A and B). At low magnification, one can observe the spherules of SBA-15 agglomerates on the nanofibers surface (Fig. 1A). High magnification SEM micrograph (Fig. 1B) allows one to observe the relatively well dispersion of the SBA-15 nanoparticles on the surface of the carbon nanofibers. The high dispersion of the SBA-15 crystallites on the CNF surface could be attributed to the presence of the prismatic planes on the CNF surface which provides a large number of anchorage sites for the SBA-15. One should also note the high anchorage strength of the SBA-15 nanoparticles on the carbon nanofibers surface as lamost no matter loss was observed after sonication treatment for 10 min. The catalyst stability was evaluated after four consecutive catalytic tests. Fig. 2 shows a drastic decrease in the conversion between the first and second run for the Ga/SBA-15 (10)/C-NFC catalyst. The conversion decreases from 88% during the first test to about 50% for the second test. The selectivity was 93% towards the para-methoxybenzophenone, and 6% for the ortho-methoxybenzophenone. These results are in agreement with previous data in the electrophilic aromatic substitution [14]. The deactivation of the catalyst is mainly due to coking and also a leaching of active species by HCl produced during the reaction. The deactivation by carbonaceous residue deposit is either due to pore blocking or active site poisoning [15]. The pore blocking seems to be the main reason for explaining the catalyst deactivation behavior [16]. The deactivated catalyst can be effectively regenerated by air calcinations to remove the
3. Results and discussion The as-synthesized C-NFC composite was sonicated in an ethanol medium for 30 min in order to remove the loosely attached carbon nanofibers from the composite and also to check the mechanical anchorage strength of theses CNFs on the macroscopic host matrix. The CNFs appear to be well anchored on the felt host structure whereas only less than 10 wt.% of the CNFs was removed from the composite during the sonication treatment. The high mechanical stability of the VA-CNTs on the host surface is attributed to the fact that part of the CNTs, located next to the nickel catalyst sitting on the host matrix surface, was partly penetrated into the host matrix allowing the generation of a strong anchorage of the CNT onto the host matrix and prevents the CNT loss during the sonication treatment. In some case the CNF could also anchor on the macroscopic host structure via an intermediate phase, i.e. microporous carbon layer, which is strongly attached to the host matrix underneath [11]. The specific surface area of the C-NFC composite significantly increased and reaches 96 m2 g− 1. Since the initial specific surface area of the host carbon felt is about 1 m2 g− 1, it is clear that the incremental specific surface area is caused by the intrinsic external surface of the carbon nanofibers, which is usually ranged between 80 and 200 m2 g− 1 [12,13]. The coating of SBA-15 material on the C-NFC support increases the specific surface area of the final composite from 96 to 143 m2/g after one-coating step. The SBA-15 loading determined was amounted to about 14 wt.% in the final SBA-15/C-NFC composite. This high amount could be explained by the high anchoring sites present on the NFC surface, i.e. prismatic planes, which favor the nucleation and growth of the SBA-15 seeds on the carbon-based support.
Fig. 1. SEM micrographs of the SBA-15/C-NFC composite at (A) low magnification and (B) high magnification showing the entangled morphology of the carbon nanofibers on the composite surface.
792
F.Z. El Berrichi et al. / Catalysis Communications 12 (2011) 790–793 Table 2 Benzoylation activity, expressed in terms of conversion, and selectivity on the Ga/SBA15(10)/C-NFC and Ga/C-NFC catalysts. Reaction conditions: catalyst weight = 200 mg, reaction temperature = 140 °C, anisole = 0.069 mol, benzoyl chloride = 0.025 mol. Catalyst
Ga/SBA-15(10)/C-NFC Ga/C-NFC a b
Fig. 2. PhCOCl conversion as a function of time and the conversion evaluation as a function of the cycling tests for the Ga/SBA-15 (10)/C-NFC catalyst.
carbonaceous residues on its surface similar to that applied for the bulk Ga-SBA-15 catalyst [4]. Table 1 shows the gradual decrease of the selectivity in paramethoxybenzophenone (para-ketone) during the cycling tests, thus raising the selectivity toward ortho-methoxybenzophenone (orthoketone). The increase of the ortho-ketone selectivity as a function of the cycling tests could be attributed to the fact that these molecules were not formed on the same active site, i.e. Ga localized on the surface pore mouth (low benzoylation activity) or inside the channel (high benzoylation activity). According to the results one should expected that the para-ketone is mainly formed on the active sites localized inside the channel of the catalyst, where the shape selectivity is controlled by the pore channel. Such site is more sensitive to the deactivation by coke formation and pore plugging and thus, the para-ketone selectivity slowly decrease as a function of the cycling tests. On the other hand, the ortho-ketone which is formed on the outer surface or next to the pore mouth is less sensitive to the deactivation phenomenon. After four cycles it is expected that most of the active site localized inside the pore channel is not accessible anymore and thus, shape selectivity towards para-ketone no longer operates. According to the results observed one can state that Ga sites localized inside the SBA-15 channel only give rise to para-ketone owing to the shape selectivity where Ga sites localized on the outer surface could be active either for para- or ortho-ketone with even selectivity. Table 2 presents the benzoylation activity and stability comparison between Ga/SBA-15(10)/C-NFC and Ga/C-NFC catalysts under the same reaction conditions. According to the observed results the former is more active and more stable than the later. Indeed, during the first test, benzoyl chloride conversion reached 88% on the Ga/SBA-15(10)/C-NFC catalyst against 39% for the Ga/C-NFC catalyst keeping the reaction conditions similar. The higher performance of the Ga/SBA-15(10)/
Conversion (%)a
Selectivity (%) Para-ketone
Ortho-ketone
Esterb
88 39
93 49
6 48
1 3
: Conversion after 24 h of the reaction. : Phenylbenzoate selectivity.
C-NFC catalyst when compared to the Ga/C-NFC catalyst is probably due to an improved dispersion of the active phase on the SBA-15 surface. The direct comparison of the ketones selectivity between the two catalysts is not straightforward taking into account the large difference in terms of conversion. However, the following hypothesis could be made regarding the ketones selectivity and the Ga localization: the high selectivity towards para-ketone observed on the Ga/SBA-15/C-NFC compared to that observed on the Ga/C-NFC, where all the Ga sites are expected to be localized on the outer surface of the support, taking into account the complete absence of channel in the CNF, confirms again the importance of the Ga site localization on the ketone, i.e. para- versus ortho-ketone, selectivity. Such result also confirms the hypothesis advanced above regarding the para-ketone selectivity change as a function of the deactivation due to the pore plugging. 4. Conclusion The comparison between the catalytic performances of Ga/SBA-15/ C-NFC and Ga/C-NFC catalysts indicates that the CNF-based catalyst exhibits a higher catalytic activity compared to the one deposited on a felt support with lower effective surface area. The coating of SBA-15 on macroscopic bodies, like C-NFC composite, also allows an ease in the catalyst recovery, thus excluding important problems linked with binder use or post-reaction filtration. Cycling tests indicate that deactivation occurs probably due to the gradual plugging of the SBA-15 channel by residual carbonaceous residue. The direct consequence of the channel plugging is the gradual decrease of the paraketone selectivity due to the lost of the gallium active sites localized within the channel. The macroscopic shaping of the support could allow one to envisage the use of this nano-macro composite catalyst in a fixed-bed configuration instead of batch configuration where deactivation by carbonaceous residues could become significant. Acknowledgments The SEM experiments were carried out at the facilities of the IPCMS (UMR 7504 of the CNRS and University of Strasbourg). References
Table 1 Ga/SBA-15(10)/C-NFC stability in the anisole benzoylation reaction as a function of the cycling tests. Reaction conditions: catalyst weight=200 mg, reaction temperature=140 °C, anisole= 0.069 mol, benzoyl chloride =0.025 mol. Catalyst
Ga/SBA-15(10)/ C-NFC
a b
Number of catalytic run
Conversion (%)a
Selectivity (%) Para-ketone
Ortho-ketone
Esterb
1 2 3 4
88 49 58 51
93 70 45 42
6 28 54 57
1 2 1 1
: Conversion after 24 h of the reaction. : Phenylbenzoate selectivity.
[1] C.T. Kresge, M.E. Leonowicz, W.J. Roth, J.C. Vartuli, J.S. Beck, Nature 359 (1992) 710. [2] J.S. Beck, J.C. Vartuli, W.J. Roth, M.E. Leonowicz, C.T. Kresge, K.D. Schmitt, C.T.-W. Chu, D.H. Olson, E.W. Sheppard, S.B. McCullen, J.B. Higgins, J.L. Schlenker, A new family of mesoporous molecular sieves prepared with liquid crystal templates, J. Am. Chem. Soc. 114 (1992) 10834. [3] D. Zhao, J. Feng, Q. Huo, N. Melosh, G.H. Frederickson, B.F. Chmelka, G.D. Stucky, Science 279 (1998) 548. [4] E. BerrichiZ. , L. Cherif, O. Orsen, J. Fraissard, J.P. Tessonnier, E. Vanhaecke, B. Louis, M.J. Ledoux, C. Pham-Huu, Ga doped SBA-15 as an active and stable catalyst for Friedel–Crafts liquid-phase acylation, Appl. Catal. A 298 (2006) 194. [5] E. BerrichiZ. , B. Louis, O. Orsen, J.P. Tessonnier, L. Cherif, M.J. Ledoux, C. Pham-Huu, One-pot synthesis of Ga-SBA-15-post-treated SBA-15 catalysts, Appl. Catal. A 316 (2007) 219. [6] A. Vinu, D.P. Savant, K. Arigo, M. Hartmann, S.B. Halligudi, Micropor. Mesopor. Mater. 80 (2005) 195. [7] P. Serp, M. Corrias, P. Kalck, Appl. Catal. A 253 (2003) 337.
F.Z. El Berrichi et al. / Catalysis Communications 12 (2011) 790–793 [8] [9] [10] [11] [12]
J.K. Chinthaginjala, K. Seshan, L. Lefferts, Ind. Eng. Chem. Res. 46 (2007) 3968. R. Vieira, D. Bastos-Netto, M.J. Ledoux, C. Pham-Huu, Appl. Catal. A 279 (2005) 35. R. Vieira, Thèse, Université de Strasbourg, 2003. J.H. Bitter, J. Mater. Chem. 20 (2010) 7312. C. Pham-Huu, M.J. Ledoux, Top. Catal. 40 (2006) 49.
793
[13] P.W.A.M. Wenmakers, J. van der Schaaf, B.F.M. Kuster, J.C. Shouten, J. Mater. Chem. 18 (2008) 2426. [14] G.A. Olah, R. Malhotra, S.C. Narang, J.A. Olah, Synthesis (1978) 672. [15] J.W. Beekman, G.F. Froment, Ind. Eng. Fundam. 18 (1979) 245. [16] H. Beinaert, R. Vermeuleu, G.F. Froment, Stud. Surf. Sci. Catal. 88 (1994) 97.