Silica-embedded tert-butyldimethylsilyltrifluoromethanesulfonate catalysts as new solid acid catalysts

Article abstract of DOI:10.1006/jcat.2001.3282

Silica-embedded tert-butyldimethylsilyltrifluoromethanesulfonate catalysts were synthesized by a sol-gel method in acidified CCl₄, using hexadecyltrimethyl-ammonium bromide as a surfactant. The catalysts were characterized by nitrogen adsorption-desorption isotherms at 77 K; TG-DTA; ¹H, ¹³C, and ²⁹Si solid state MAS/NMR; XRD; XPS; Raman spectroscopy; and FTIR after adsorption of NH₃. The characterization data indicated mesoporous solids in which most of the silyl-triflate derivative keeps its integrity. The catalytic tests performed with the methyl ester of 1-cyclopentenylacetic acid in various solvents showed that the reaction selectively leads to 3-methylbenzaldehyde and is sensitive to the solvent accepting ability.

Full text of DOI:10.1006/jcat.2001.3282

Journal of Catalysis 202, 319–323 (2001)
doi:10.1006/jcat.2001.3282, available online at http://www.idealibrary.com on
Silica-Embedded tert-Butyldimethylsilyltrifluoromethanesulfonate
Catalysts as New Solid Acid Catalysts
A. N. Paˆrvulescu, B. C. Gagea, M. Alifanti,‡ V. Paˆ rvulescu, V. I. Paˆ rvulescu, ,1 S. Nae,†
A. Raˇzus¸,† G. Poncelet,‡ and P. Grange‡
University of Bucharest, Faculty of Chemistry, Department of Chemical Technology and Catalysis, B-dul Republicii 13, 70346 Bucharest, Romania;
†Institute of Organic Chemistry, Splaiul Independentei 202B, P.O. Box 15-258, 71141 Bucharest, Romania; and ‡Universite´ Catholique de Louvain,
Unite´ de Catalyse et Chimie des Mate´riaux Divise´s, Place Croix du Sud 2/17, 1348 Louvain-la-Neuve, Belgium
Received February 13, 2001; revised May 14, 2001; accepted May 19, 2001
This study is devoted to the embedding of tert-butyl-
dimethylsilyltrifluoromethane sulfonate in a silica sol-gel
matrix and the catalytic performances in the reaction of the
methyl ester of the 1-cyclopentenylacetic acid.
Silica-embedded tert-butyldimethylsilyltrifluoromethanesulfo-
nate catalysts were synthesized by a sol-gel method in acidified
CCl4, using hexadecyltrimethyl-ammonium bromide as a surfac-
tant. The catalysts were characterized by nitrogen adsorption–
desorption isotherms at 77 K; TG-DTA; 1H, 13C, and 29Si solid
state MAS/NMR; XRD; XPS; Raman spectroscopy; and FTIR af-
ter adsorption of NH3. The characterization data indicated meso-
porous solids in which most of the silyl-triflate derivative keeps
its integrity. The catalytic tests performed with the methyl ester of
1-cyclopentenylacetic acid in various solvents showed that the re-
action selectively leads to 3-methylbenzaldehyde and is sensitive to
c
EXPERIMENTAL
The catalysts were prepared by embedding tert-butyl-
dimethylsilyltrifluoromethanesulfonate in a silica sol-gel
matrix. Since the silyl triflate derivative is only soluble
in carbon tetrachloride and triethylamine (9), the sol-gel
synthesis was carried out in inert atmosphere (Ar), us-
ing an adapted route in which the silica sol was obtained
by acid hydrolysis of a solution of tetraethoxyorthosilicate
(TEOS) (10.4 g) in acidified CCl4 (TEOS : CCl4 molar ra-
the solvent accepting ability.
2001 Academic Press
INTRODUCTION
Alkylsilyltrifluoromethanesulfonates are known as ex- tio of 4, pH 1.5 with HCl 37% ). Water was then added
ceptionally strong silylating agents (1–3). The silylation re- to the acidic solution in TEOS : H2O molar ratio of 1 : 10
action occurs in homogeneous conditions in the presence of (samples A1 and A2) or 1 : 4 (sample A3), and the mixture
triethylamine or triethylamine and carbon tetrachloride as was refluxed at 70 C for 2 h. After cooling the silica sol
solvents. However, up to this day, nothing on the homoge- solution at room temperature, the silyl triflate derivative
neous or heterogeneous catalytic behavior in acid-assisted (0.38 ml) was added under vigorous stirring as a 0.3 M solu-
reactions of these compounds has been reported. On the tion in CCl4 (samples A1 and A3) or as such (sample A2).
other hand, the use of triflates or triflic acid as homoge- Hexadecyltrimethyl-ammonium bromide, as a surfactant,
neous acid catalysts has received a keen interest because was then introduced (surfactant : TEOS ratio of 0.1) and
of the electron-withdrawing effect exerted by the triflu- the gelation was carried out at 90 C for 2 days in a Teflon
oromethanesulfonyl group. Since heterogeneous catalysis cylinder within an autoclave (samples A2 and A3) or at
is more versatile for practical applications, several stud- room temperature for 6 days (sample A1). The resulting
ies have been reported in the past few years on the het- gel was dried under vacuum, first at room temperature for
erogeneization of triflic acid (4) and various triflates by 24 h and then at 100 C for 6 h. Samples without surfactant
impregnation (5, 6) or polymer incorporation (7, 8), and were prepared as well. Using this procedure, samples with
on selective acid-catalyzed reactions using these materials. 15 wt% triflate were obtained.
Silica-supported triflates or triflic acid are active and se-
The catalysts were characterized using nitrogen ad-
lective catalysts in the rearrangement of -pinene oxide to sorption–desorption isotherms at 77 K; TG-DTA; 1H, 13C,
campholenic aldehyde, polymerization and depolymeriza- and 29Si solid state MAS/NMR; XRD; XPS; Raman
tion of cyclic ethers, Friedel–Crafts reactions, alkylation of spectroscopy; and FTIR after adsorption of NH3. The sur-
isobutane with n-butene (4–8), etc.
face area of the samples containing the surfactant was mea-
sured after extraction of the surfactant using the procedure
of Hitz and Prins (10).
1 To whom correspondence should be addressed.
319
0021-9517/01 $35.00
c
Copyright
2001 by Academic Press
All rights of reproduction in any form reserved.

ˆ
PARVULESCU ET AL.
320
The catalytic tests were carried out in a round-bottom
flask at atmospheric pressure in flowing dry nitrogen at tem-
peratures in the range 25–60 C. The catalyst (40 mg) was
first added to the reactor. After purging the reactor with dry
nitrogen for 10 min, 30 ml of the solution of the methyl ester
of 1-cyclopentenylacetic acid in the solvent (with a molar
ratio of methyl ester to solvent of 1 : 60) was added. Cy-
clohexene, benzene, toluene, ortho- and para-xylene were
used as solvents. After addition of the reaction mixture, the
system was vigorously stirred using a magnetic stirrer. The
reaction temperature was controlled within 0.3 C. Sepa-
rate catalytic tests were carried out using the same number
of moles of active species, but instead of the embedded
silyl triflate derivative, aluminum chloride, free triflic acid,
or mixtures of catalysts without surfactant and surfactant.
The reaction products were analyzed in a gas chromato-
graph coupled with a mass spectrometer (MS). The purity
of the products was checked by 1H and 13C NMR.
FIG. 1. FTIR spectra of the investigated catalysts.
the CF3 group at 118 and 123 ppm. However, NH3-FTIR
spectra indicated the presence of Lewis acid sites, as in-
ferred from a band at 1611 cm 1, but also of Brønsted acid
sites (bands in the range 1460–1490 cm 1) (Fig. 2). These
RESULTS AND DISCUSSION
X-ray diffraction (XRD) patterns indicated amorphous latter bands may indicate that part of the Si–O–S bond in
materials, which, according to the nitrogen adsorption– the silyl-derivative is hydrolyzed, leading to chemisorbed
desorption isotherms, corresponded to mesoporous solids. triflate.
The pore diameter wasthe same for allthe catalysts(2.8nm)
Table 1 compiles the reaction conditions and the yields of
but the surface area varied in the order A1 (483 m2 g 1) > 3-methylbenzaldehyde. The reaction selectively produced
A2 (461 m2 g 1) > A3 (412 m2 g 1). The surface area of the 3-methylbenzaldehyde with, except for the hydrogenated
catalysts without surfactant followed the same order order: part ofthe solvent, the secondaryproductsbeingin amounts
A1 (572 m2 g 1) > A2 (549 m2 g 1) > A3 (508 m2 g 1), smaller than 1% . These results indicate that the most im-
exceeding the values of the samples with surfactant. The portant factor in this reaction was the chemical nature of
pore diameter determined for these samples was around the solvent. Based on the chemical analysis of the products,
1.8 nm. Raman spectra indicated the same position of the we suggest that the reaction occurred via the route indi-
terminal S O stretch (967 cm 1) for both embedded and cated in Scheme 1, in which the first step of the reaction
==
pure silyl triflate derivative. The integrity of the silyl triflate consists of a ring enlargement, typical of acid-catalyzed re-
derivative in the embedded catalysts was also supported action. The second step involves a hydrogen transfer from
by the thermal decomposition of the embedded materials the cyclohexene cycle of intermediate II to the solvent, re-
compared with pure silyl triflate. The surfactant also re- sulting in III, which finally undergoes a rearrangement to
mained unaffected as shown by 1H and 13C-NMR analysis. 3-methylbenzaldehyde (compound IV in Scheme 1). It is
In addition, the signals of C14–C16 atoms in single pulse ex- note worthy that in the absence of an adequate solvent,
periments were narrow compared with the same resonance the reaction does not take place (Table 1). The presence
in the cross-polarized spectra, which may suggest a certain of the hydrogen acceptor molecule is therefore the key
mobility of the surfactant in the silica matrix. The XPS Si : S factor. Based on these data, one can say that the hydro-
atomic ratios varied in the order A1 (18.2) > A3 (16.8) > gen accepting ability in this reaction varies in the order:
A2 (15.4), indicating a slight difference in the superficial p-xylene > o-xylene > cyclohexene
toluene, benzene.
distribution of the triflate species. These ratios were very The chromatographic analysis indicated in all the cases a
close to the analytical ones, indicating a good dispersion stoichiometric reduction of the solvent. Steps III to IV in-
of the derivative in the silica matrix. Slight differences in volve both isomerization and oxygen elimination. The re-
–
the S O bond population could also be appreciated from duction of simple aryl triflates by trialkylammonium for-
the FTIR analysis (Fig. 1), as inferred from the differences mate in the presence of Pd(0) as catalyst has already been
in the relative intensity of the bands located at 942, 1028, reported (11, 12). Very recently, Kamochi and Kudo (13)
1
==
and 1073 cm , corresponding to S O vibrations in various reported another example in which triflate compounds
binding modes. The XPS S : F atomic ratios were near 1 : 3 were found as effective reduction catalysts. These authors
for all the catalysts, in agreement with the chemical formu- reported the reduction of carboxylic acids to the corres-
las. The 13C-CP/MAS NMR spectra exhibited the signals of ponding alcohols using samarium triflate. In the present

SOLID ACID CATALYSTS
321
FIG. 2. NH3-FTIR spectra after the samples were evacuated at 100 C.
case, neither the noble metal nor the rare-earth element is that these intermediates are indeed involved in this reac-
necessary. During the reaction, aryl triflate may be formed tion. It is already known from the aromatization of alkanes
by interaction with the silyl-derivative or cleavage of Si–O– that is easier to dehydrogenate a C6 hydrocarbon than a
S bonds. The participation of the tetraalkyl ammonium salt C5 one, and this step is easily catalyzed by the acid function
in this process can be expected because of its mobility. The of the catalysts. This reaction may easily occur in the pres-
reaction may also involve some rearrangements (14) since ence of a Lewis catalyst. Such a behavior has recently been
a compound with M = 152 was identified by MS analysis reported for methyl-cyclopentane to benzene on Lewis sul-
of the by-products.
fated zirconia catalysts (15). Therefore, only the passage be-
The analyses did not show the presence of either II or III tween III and IV isa typicaland newstep for thiscatalyst. At
intermediates. However, several reasons led us to consider this moment, we have no explanation for this mechanism,
TABLE 1
Conversion and Selectivity to 3-Methylbenzaldehyde
Conversion to
Temperature
( C)
Time
(h)
3-methylbenzaldehyde
(% )
Selectivity
(% )
Catalyst
Solvent
No catalyst
A1
A1
A1
A1
A2
A2
A2
A2
A3
A3
A3
A3
A3
A3
p-xylene
benzene
cyclohexene
cyclohexene
p-xylene
benzene
p-xylene
p-xylene
cyclohexene
benzene
toluene
o-xylene
p-xylene
p-xylene
cyclohexene
p-xylene
60
25
60
25
60
60
25
60
60
60
60
60
60
60
60
60
24
4
4
24
4
4
24
4
4
4
4
4
4
24
4
4
0
0
0
0
3.2
1.6
22.8
0
11.7
15.2
2.2
0
1.2
5.9
17.8
27.8
2.7
0
8.8
97.3
98.4
0
97.8
98.4
7.1
0
97.2
97.3
98.4
97.9
7.9
0
A3 without
surfactant
A3 without
surfactant +
surfactant
p-xylene
60
4
0.5
20.0

ˆ
PARVULESCU ET AL.
322
SCHEME 2. Reaction routes on pure triflate and AlCl3.
SCHEME 1. Reaction route of the methyl ester of the 1-cyclopent-
enylacetic acid in the presence of embedded tert-butyldimethylsilyltri-
fluoromethanesulfonate and hydrogen acceptor solvent.
3-cyclopentenylacetic ester (VIII) (63.8% for AlCl3 and
68.4% for triflate) (Scheme 2). The use of free triflic acid as
catalyst led to very different results (Table 2). This com-
pound may be expected as a hydrolysis product of silyl
triflate derivative during the sol-gel preparation. The anal-
ysis of the reaction products indicated that the main re-
action routes were exo-isomerization of the double bond
and alkylation (Scheme 3). Although small amounts of 3-
methylbenzaldehyde were also found, these data show that
free triflic acid is not responsible for the results obtained
with embedded catalysts; rather it is the silyl triflate deriva-
tive in the silica matrix that lies in close proximity to the
quaternary ammonium salt.
Additional evidences about the behavior of this catalyst
and the contribution of the surfactant to this reaction
were provided by the catalytic tests carried out using
catalysts without surfactant and mixtures of the cata-
lyst without surfactant and surfactant solved in homo-
geneous solvent. On the catalysts without surfactant, no
3-methylbenzaldehyde was obtained. The only reaction
product was exo-isomerization of the double bond in an
amount less than 15% . This is less than we obtained with
free triflic acid where, in addition, small amounts of 3-
methylbenzaldehyde were also formed. The presence of the
surfactant in the reaction mixture led to the formation of
3-methylbenzaldehyde but to a smaller extent than with
but the data presented in this study clearly indicate that
this is a contribution of a heterogeneous catalyst and, more
precisely, of a concerted effect of the close proximity of the
silyl derivative and the quaternary ammonium salt.
The heterogeneous behavior of the catalysts was con-
firmed both by the profound dependence of activity on sol-
vent and the analysis of the supernatant resulting after the
catalyst separation and another 4 h of reaction. The ab-
sence of further reaction is evidence that this reaction is
not influenced by the triflate leaching. In addition, because
the silyl triflate derivative is insoluble in the solvent used,
the supernatant was always a clear solution. It should also
be mentioned that catalytic tests done with emulsions of
silyl triflate derivative showed that only the Friedel–Crafts
alkylation occurred. Additional experiments using AlCl3 as
catalyst instead of embedded silyl triflate showed nearly the
same results as the pure silyl triflate derivative. The reac-
tion products in this case were the 2-cyclopentenylacetic
ester (0.5% for AlCl3 and 9% for triflate) (V), trans-
xylyl-2-cyclopentenylacetic (15.2% for AlCl3 and 16.3%
for triflate) (VI) and cis-xylyl-2-cyclopentenylacetic ester
(0.5% for AlCl3 and 9.0% for triflate) (VII), and cis-xylyl-
TABLE 2
Conversion of the Methyl Ester of 1-Cyclopentenylacetic Acid in the Presence of Triflic Acid (23 mmoles)
Conversion to
Conversion to 3-methyl-
benzaldehyde
(% )
alkylated
compounds
(% )
Temperature
( C)
Time
(h)
Conversion to IX
Solvent
(% )
benzene
p-xylene
cyclohexene
60
60
60
4
4
4
0.3
0.8
1.6
45.7
15.8
27.6
3.3
4.8
9.9

SOLID ACID CATALYSTS
323
1-cyclopentenylacetic acid to 3-methylbenzaldehyde. The
reused catalysts showed the same activity. The comparison
of the embedded catalysts with the pure silyl triflate deriva-
tive, AlCl3, or free triflic acid confirms the novelty of these
catalysts. Although a partial hydrolysis of the silyl triflate
derivative occurs during the synthesis step, the comparative
behavior of this catalyst with catalysts without surfactant,
and mixtures of the catalyst without and with surfactant,
confirms the concerted participation of embedded silyl tri-
flate derivative and surfactant, as an unitary heterogeneous
catalyst in this reaction. To the best of our knowledge, both
the heterogeneous catalyst and reaction are new.
SCHEME 3. Reaction routes on free triflic acid.
REFERENCES
1. Murata, S., Suzuki, M., and Noyori, R., J. Am. Chem. Soc. 102, 3248
(1980).
2. Corey, E. J., Cho, H., Rucker, C., and Hua, D. H., Tetrahedron Lett.
3455 (1981).
the catalyst in which the surfactant was introduced during
the synthesis, and comparable with that measured with free
triflic acid.
The selectivity to 3-methylbenzaldehyde in the presence
of embedded silyl triflate derivative is in good agreement
with recent results of Sherman et al. (16) on the isomeriza-
tion of substituted benzenes, showing that, in the presence
of a triflate catalyst, m-isomers are preferred to ortho- or
para-isomers. The results also show a contribution of the
catalyst. Under the investigated conditions, the order of re-
activity was A1 > A3 > A2, consistent with the XPS and
FTIR data, indicating a proportionality between accessible
triflate species and conversion.
In conclusion, embedding of tert-butyldimethylsilyltri-
fluoromethanesulfonate in a silica matrix via a sol-gel car-
ried out in the presence of carbon tetrachloride leads to a
heterogeneous catalyst in which most of the triflate deriva-
tive remains intact and is homogeneously distributed.
This catalyst is able to catalyze selectively (>95% ) at
low temperature the reaction of the methyl ester of
3. Sainte, F., Serckx-Poncin, B., Hesbain-Frisque, A.-M., and Ghosez, L.,
J. Am. Chem. Soc. 104, 1428 (1982).
4. Clerici, M. G., Perego, C., De Angelis, A., and Montanari, L., Eur.
Patent 638 363, 1995.
5. Drysdale, N. E., and Herron, N., WO Patent 9502625, 1994.
6. Wilson, K., Renson, A., and Clark, J. H., Catal. Lett. 61, 51 (1999).
7. Kobayashi, S., and Nagayama, S., J. Am. Chem. Soc. 120, 2985 (1998).
8. Schager, F., and Bonrath, W., Appl. Catal. A: Gen. 202, 117 (2000).
9. Danishefsky, S. J., and Marung, C. J., J. Am. Chem. Soc. 107, 1269
(1985).
10. Hitz, S., and Prins, R., J. Catal. 168, 194 (1997).
11. Peterson, G. A., Kunng, F.-A., McCallum, J. S., and Wulff, W. D., Tetra-
hedron Lett. 28, 1381 (1987).
12. Chen, Q.-Y., and He, Y.-B., Synthesis 866 (1988).
13. Kamochi, Y., and Kudo, T., Tetrahedron Lett. 41, 341 (2000).
14. Effenberger, F., and Gutmann, R., Chem. Ber. 115, 1089 (1982).
15. Paˆrvulescu, V., Coman, S., Paˆrvulescu, V. I., Grange, P., and Poncelet,
G., J. Catal. 180, 66 (1998).
16. Sherman, S. C., Iretskii, A. V., White, M. G., Gumienny, Ch., Tolbert,
L. M., and Schiraldi, D. A., J. Am. Chem. Soc., in press.