Highly selective fries rearrangement over zeolites and nafion in silica composite catalysts: A comparison

Full text of DOI:10.1006/jcat.1998.2034

JOURNAL OF CATALYSIS 176, 260–263 (1998)
ARTICLE NO. CA982034
RESEARCH NOTE
Highly Selective Fries Rearrangement over Zeolites and Nafion
in Silica Composite Catalysts: A Comparison
,1
ꢀ
ꢀ
Alfred Heidekum, Mark A. Harmer,† and Wolfgang F. Hoelderich
ꢀ
Department of Chemical Technology and Heterogeneous Catalysis, University of Technology RWTH Aachen, Worringerweg 1,
52074 Aachen, Germany; and †E. I. DuPont de Nemours & Company, Wilmington, Delaware
Received September 9, 1997; revised December 8, 1997; accepted February 2, 1998
Zeolites and acidic ion exchange resins are well known inal ion-exchange resin (10, 11). The isomerization of 1-
as potential heterogeneous catalysts for acid catalyzed re- butene is an example of a gas phase reaction, which is
actions. Zeolites are a remarkable tool in catalysis as they catalyzed by the pure Nafion resin to a less extent in con-
offer reproducible well-defined surfaces. In addition, the trary to the composite catalysts yielding high conversion
microporousstructure hasa remarkable influence on the se- (12).
lectivity of a reaction and is sometimes tailored to suit the
It has been our intention to study the new materials in
desired reactions. The isomorphous substitution can con- the Fries rearrangement of phenyl acetate, according to
tribute a possibility to adjust the demanded acid strength Eq. [1], and to compare it with competitive alternatives,
for the reaction (1–3). Other substitutes for mineral acids such as zeolitic catalysts.
provide strongly acidic sulfonic acid cation exchange resins
(4, 5). For example, Nafion resin, a perfluorosulfonic acid
resin, which is a copolymer of tetrafluoroethene and a
perfluorosulfonylether derivative with a backbone similar
to Teflon. The superacidity of such materials arises from
[1]
–
CF2CF2SO3H groups. The Hammett acidity has been sug-
–
gested to be similar to concentrated sulfuric acid ( H0 ꢁ 12)
(6). Due to the perfluorocarbon backbone, the chemical in-
ertness is very high for an organic resin. Until now a con-
siderable number of reactions catalyzed by acidic Nafion
resin have been studied (7, 8). Because of the very lim-
ited surface area (0.02 m2/g) of the pure resin and the
very limited accessibility to the acid sites the activity in
less polar solvents or in the gas phase reactions is re-
duced.
The Fries rearrangement of phenyl acetate 1 leads to
hydroxyacetophenones 2, 3, which are very valuable pre-
cursors in the pharmaceutical industry to obtain, e.g.
p-hydroxyacetanilid, which is used as a pain killer, also
known as paracetamol (13). However, carrying out this
reaction over heterogeneous catalysts has been not so suc-
cessful until now, because of low selectivity and rapid cata-
lyst deactivation (14). The zeolites studied in the present
work are commercial products and were kindly provided
by PQ-Corporation (H-Y (6); H-USY (70), Na-BEA (35),
H-ZSM 5 (28)); the figures in brackets are the silica alumina
ratios of the zeolites. Na-BEA is converted by a twofold
ion-exchange with a tenfold excess of 2 N NH4NO3aq at
80ꢂC into the acid form and all of the zeolites have been
calcined at 550ꢂC for 8 h before use (15). The Nafion in
silica composite catalysts contain 13 and 40 wt% Nafion,
respectively. The porous silica matrix got a surface area
of about 350 m2/g and an average pore size of about
10 nm. They were prepared according to literature (9). Pure
Nafion resin is marked 100 wt% . All resins have been dried
However, recently materials have been found with im-
proved features (9). This has been achieved by entrapping
nano-sized particles of Nafion in a high porous silica ma-
trix. In this way the surface area of the new Nafion in sil-
ica composite catalyst is increased and most importantly
the accessibility of the acid sites is increased. The activity
of this new class of solid acids has as a result increased
significantly, even in less polar solvents. For instance, in
the dimerization of ꢀ-methylstyrene the composite cata-
lysts generate a much higher reaction rate than the orig-
1 To whom correspondence should be addressed. E-mail: hoelderich
@rwth-aachen.de.
260
0021-9517/98 $25.00
c
Copyright ꢃ 1998 by Academic Press
All rights of reproduction in any form reserved.

HIGHLY SELECTIVE FRIES REARRANGEMENT
261
FIG. 1. Fries rearrangement of phenyl acetate (20 wt% ) over different catalysts (2 wt% ) in cumene as solvent at 150ꢂC and normal pressure; the
figures on the columns depict the ratio of 2 to 3.
before use for at least 4 h at 150ꢂC in vacuum. Each sol-
vent and reactant has been dried over molsieves. The re-
actions were carried out in batch reactors under dried
Argon atmosphere at 100–200ꢂC for 18 h. The reaction
[2]
solution consisted of 78 wt% solvent, 20 wt% phenyl ac-
etate, and 2 wt% catalyst. Results were determined by GLC
analysis.
In Fig. 1 the results of the Fries rearrangement of phenyl
acetate in the presence of various catalysts are depicted.
[3]
The reactions are carried out in at 150ꢂC and ambient pres-
sure. The conversions are around 10 to 16% and the se-
lectivity is in a range from 20 up to 56% as in the case of
The consecutive products of keten also causes the rapid
deactivation of the catalyst by coking. At a lower tempera-
ture <100ꢂC the decomposition could be reduced; but at a
reaction temperature of 100ꢂC the conversion is very low.
By changing the solvent, in particular, using phenol, we
found a tremendous change. The results shown in Fig. 2
have been obtained under the same reaction conditions de-
scribed above for cumene as solvent. Now, in the presence
of all catalysts tested selectivities of more than 50% are
achieved. Using 13 wt% Nafion in silica composite cata-
lyst a selectivity of more than 90% is obtained at a con-
version of about 21% . Additionally, using the composite
catalysts having 13 or 40 wt% of Nafion and over the H-
BEA zeolite the conversion is considerably increased, too.
In the case of H-BEA 76% selectivity at 41% conversion
are achieved. Unfortunately, by reusing all the catalyst only
an ultrastable faujasite zeolite (H-USY (70)). An acylation
of the solvent cumene can be excluded under these con-
ditions. No acylated products could be obtained even with
stronger acylation agents, as could be shown in some other
experiments.
In comparison to the results of homogeneously catalyzed
reactions such as in the presence of BF3 of HF our findings
are quite moderate. The main side product in this reac-
tion is phenol. This could be formed via the decomposition
of the starting ester or the deacetylation of the products.
In both cases highly reactive keten is formed as the com-
plementary decomposition compound, according Eqs. [2]
and [3]. This keten reacts in consecutive irreversible reac-
tions very unselectively to several products, in particular to
diketen, resulting in low selectivity for the Fries rearrange-
ment.

262
HEIDEKUM, HARMER, AND HOELDERICH
FIG. 2. Fries rearrangement of phenyl acetate (20 wt% ) over different catalysts (2 wt% ), and in the case of H-BEAꢀ, also 1 wt% in phenol as
solvent at 150ꢂC and normal pressure. The figures on the columns depict the ratio of 2 to 3.
traces of product could be obtained. The deactivation seems to the original Nafion resin. They exhibit both higher ac-
to be almost complete after reaction time. However, the cat- tivity and higher selectivity in the Fries rearrangement. In
alysts can be reactivated. The Nafion in silica composite are comparison to zeolitic catalysts, it is only the H-BEA ze-
regenerated by a washing procedure with acetone and ni- olite which achieves a superior performance in this reac-
tric acid and the zeolites, by calcination in air at 550ꢂC for tion with respect to conversion, but not to the overall se-
8 h.
lectivity under the chosen reaction conditions. However,
It is assumed that in an excess of phenol the formed at equal values of conversion, achieved by the use of only
keten is caught immediately and the chemical equilibrium 1 wt% H-BEA, the selectivity (94.4% ) over the zeolitic ma-
is shifted to the starting materials, according to Eqs. [2] and terial is very high, too. Therefore, it can be emphasized that
[3]. In this way the irreversible consecutive reactions out of the zeolite H-BEA shows a better catalytic performance. In
keten do not take place. For this reason higher conversions general, it has to be pointed out that the heterogeneously
and much higher selectivities are achieved.
catalyzed Fries rearrangement can be much improved by
Concerning the Nafion catalysts, the diagram shows that using a suitable solvent, such as phenol.
the conversion raises from 21% for the 13 wt% Nafion con-
We checked this behaviour for derivatives of phenyl ac-
taining material to 31% for the pure Nafion resin. However, etate, too. As a result, the Fries rearrangement of p-tolyl
the pure material contains more than seven times Nafion acetate over the 13 wt% Nafion in the silica composite in
resin than the composite material. In addition, both com- p-cresol as the solvent is much more selective (82% ) than
posites show a much higher selectivity, in particular the in cumene as solvent (48% ) at conversions of about 16 and
13 wt% containing Nafion sample. This is likely due to the 9% , respectively.
improved accessibility to the acid sites and rapid transport
ACKNOWLEDGMENT
in and out of the catalyst reducing secondary chemistry.
In addition to the enhanced overall selectivity, the p/o
ratio in the production distribution is enhanced, too. This
effect is probably due to the higher polarity of phenol over
cumene, which supports the formation of para-product 2,
as could already be shown by Jayat et al. (16).
The authors are grateful to DuPont de Nemours & Company for the
financial support and for allowing publication of this work.
REFERENCES
It could be shown that the new Nafion in silica composite
catalysts has improved the catalytic properties compared
1. Ho¨lderich, W. F., and van Bekkum, H., Stud. Surf. Sci. Catal. 58, 631
(1991).

HIGHLY SELECTIVE FRIES REARRANGEMENT
263
2. Ho¨lderich, W. F., in “Zeolites: Facts, Figures, Future” (L. Guczi and
R. A. van Santen, Eds.), p. 69. Elsevier, Amsterdam, 1989.
3. Venuto, P. B., Microporous Mat. 2, 297 (1994).
10. Sun, Q., Harmer, M. A., and Farneth, W. E., J. Catal. 164, 62 (1996).
11. Heidekum, A., Harmer, M. A., and Hoelderich, W. F., Catal. Lett. 47,
243 (1997).
4. Neier, W., in “Ion Exchangers” (K. Dorfner, Ed.), p. 981. Gruyter, 12. Sun, Q., Harmer, M. A., and Farneth, W. E., Chem. Commun., 1201
Berlin/New York, 1991.
(1996).
5. Chakrabarti, A., and Sharma, M. M., React. Polym. 20, 1 (1993).
6. Waller, F. J., and Van Scoyoc, R. W., Chem. Tech. 17, 438 (1987).
7. Olah, G. A., Malhotra, R., Narang, S. C., and Olah, J. A., Synthesis,
672 (1978).
13. Szmant, H., “Organic Building Blocks of the Chemical Industry.”
Wiley, New York, 1989.
14. Vogt, A., Kouwenhoven, H. W., and Prins, R., Appl. Catal. 123, 37
(1995).
8. Olah, G. A., Iyer, P. S., and Prakash, G. K. S., Synthesis, 513 (1986).
9. Harmer, M. A., Farneth, W. E., and Sun, Q., J. Am. Chem. Soc. 118,
7708 (1996).
15. Breck, D. W., “Zeolite Molecular Sieves.” Wiley, New York, 1974.
16. Jayat, F., Sabater Picot, M. J., and Guisnet, M., Catal. Lett. 41, 181
(1996).