30% after 6 h of test indicating a strong inhibition of the active
site. It should be noted that the p-methoxybenzophenone
selectivity remains unchanged, i.e. > 90%, on both catalysts
which indicates that the nature of the catalysts was not modified
during the run.
Larger zeolite particles are more sensitive to deactivation due
to the formation of heavier coke inside the intra-micro-
porosities. However, according to the average particle size
measured, more deactivation would be expected to occur on the
supported zeolite because of the higher average size, i.e. 50
instead of 10 nm. From the results obtained one would expect
that the deactivation which occurred on the commercial zeolite
was originated from another factor. The difference of stability
observed between the supported and unsupported HBEA can be
attributed to the apparent size of the zeolite particles. On the
supported catalyst these zeolite particles are smaller than the
particles of the unsupported zeolite because of the dispersion on
the support surface, while on the commercial catalyst aggrega-
tion between the different zeolite particles can result in bigger
apparent particle sizes. The zeolite apparent particles influenced
the rate of diffusion of the reactants and products which in turn,
modified the real stoichiometry of the different reactants inside
the zeolite cavity which in turn, lowered the catalytic activity.
The difficulty in evacuating the heavier products formed inside
the zeolite cavity also probably increased the coke formation
inside the zeolite porosities, thus leading to deactivation.
Temperature-programmed oxidation (Electronic Supplemen-
tary Information†) analyses carried out on the two catalysts
after reaction clearly evidence a higher coke accumulation
inside the bulk zeolite compared to that of the supported
zeolite.
It has been reported by several authors in the literature that
the aromatic ketone formed is the main inhibitor for the Friedel–
Crafts reaction in a slurry reactor due to its strong adsorption on
the active sites. Derouane and co-workers5,6 have reported that
the acylation activity significantly decreased as the amount of
aromatic ketone in the reactant mixture increased. A similar
deactivation has also been reported by Jaimol et al.4 in a fixed-
bed reactor during the acetylation of toluene over H-ZSM5
catalyst. At 453 K the conversion decreased from 70% to 30%
after about seven hours on stream. The deactivation observed
was attributed by the authors to coke formation inside the
zeolite microporosity. Freese et al.15 have reported that the
acylation activity on dealuminated H-Beta was higher than that
observed on initial H-Beta. This was attributed to the modifica-
tion of the porosity in the dealuminated H-Beta, which favours
the diffusion of the products from the core of the catalyst to the
surface for desorption. It is expected that on the supported
zeolite the dispersion of the zeolite particles on the support
surface favours the evacuation of the ketone products. The
higher diffusion rate of ketones out of the zeolite porosity
probably allows the initial activity of the supported catalyst to
be maintained despite the slight deactivation observed. This
initial deactivation could be due to the equilibrium adsorption of
a part of the formed ketone and the reactants inside the zeolite
intraporosity.
Fig. 2 Influence of the WHSV, i.e. 4 to 19 h21, on the benzoylation activity,
expressed in terms of conversion, on the H-BEA/SiC (A) and on the
commercial bulk H-BEA (B) catalysts at 120 °C.
apparent size which could affect the reactants diffusion and/or
products back-diffusion, was advanced to explain these results
when increasing the WHSV, on the commercial unsupported
zeolite. Again, the selectivity remained unchanged as a function
of the WHSV.
In summary, silicon carbide synthesised by a gas–solid
reaction can be efficiently used as support for dispersing beta
zeolite. The supported beta zeolite catalyst was active for the
benzoylation of anisole by benzoyl chloride and was very stable
in a fixed bed reactor configuration when compared to that of
the commercial unsupported beta zeolite. The high thermal
conductivity of the SiC support can also allow use of the catalyst
in highly exothermic reactions where a rapid dispersion of heat
through the catalyst bed is necessary to avoid the formation of
hot spots. Finally, the possibility of preparing SiC with a more
open shape, i.e. monolith, allows the use of these supported
catalysts for a high rate reaction, with a minimum diffusion
limitation, which is not easy with a traditional support shape, for
an extremely high space velocity process.
Notes and references
1 (a) Y. Ma, Q. L. Wang, W. Jiang and B. Zuo, Appl. Catal. A: General,
1997, 165, 199; (b) C. Guignard, V. Pédron, F. Richard, R. Jacquot, M.
Spagnol, J. M. Coustard and G. Pérot, Appl. Catal. A: General, 2002,
234, 79.
2 V. R. Choudhary and S. K. Jana, J. Mol. Catal. A: Chemical, 2002, 180,
267.
3 A. E. W. Beers, T. A. Njihuis, F. Kapteijn and J. A. Moulijn,
Microporous Mesoporous Mater., 2001, 48, 279.
4 T. Jaimol, A. K. Pandey and A. P. Singh, J. Mol. Catal. A: Chemical,
2001, 170, 117.
5 E. G. Derouane, G. Crehan, C. J. Dillon, D. Bethell, H. He and S. B.
Derouane-Abd Hamid, J. Catal., 2000, 194, 410.
6 E. G. Derouane, C. J. Dillon, D. Bethell and S. B. Derouane-Abd Hamid,
J. Catal., 1999, 187, 209.
7 J. C. Jansen, J. H. Koegler, H. van Bekkum, H. P. A. Calis, C. M. van
den Bleek, F. Kapteijn, J. A. Moulijn, E. R. Geus and N. van der Puil,
Microporous Mesoporous Mater., 1998, 21, 213.
8 N. van der Puil, F. M. Dautzenberg, H. van Bekkum and J. C. Jansen,
Microporous Mesoporous Mater., 1999, 27, 95.
9 R. Lai, Y. Yan and G. R. Gavalas, Microporous Mesoporous Mater.,
2000, 37, 9.
10 J. Caro, M. Noack, P. Kölsch and R. Schäfer, Microporous Mesoporous
Mater., 2000, 38, 3.
11 S. Basso, J. P. Tessonnier, C. Pham-Huu and M. J. Ledoux, French Pat.
Appl. No. 02-00541, assigned to Sicat SA 2002.
12 G. Winé, J. P. Tessonnier, C. Pham-Huu and M. J. Ledoux, Chem.
Commun., 2002, 2418.
13 M. J. Ledoux and C. Pham-Huu, 2001, CaTTech 5.
14 C. Madsen and C. J. H. Jacobsen, Chem. Commun., 1999, 673.
15 U. Freese, F. Heinrich and F. Roessner, Catal. Today, 1999, 49, 237.
The supported HBEA was also able to work in a large range
of WHSVs without appreciable activity loss except at extremely
high WHSV values which was probably due to a too short
contact time between the reactants and the catalytic sites (Fig.
2A). The high Friedel–Crafts activity observed on the HBEA/
SiC catalyst was attributed to the high dispersion of the zeolite
as individual particles on the support surface which can offer a
high effective surface contact to the reactants. Similar reactions
carried out on the commercial bulk zeolite showed a drastic
activity loss as a function of the WHSV (Fig. 2B). Formation of
aggregates of several zeolite particles, leading to a bigger
CHEM. COMMUN., 2003, 530–531
531