A comparative study of an MCM-41 anchored quaternary ammonium chloride/SnCl4 catalyst and its silica gel analogue

Article abstract of DOI:10.1039/b101843o

A novel reusable Lewis acid catalyst has been prepared by the heterogenization of a Lewis acid/tetrapropylammonium adduct; anchoring of tin chloride on quaternary ammonium chloride functionalized MCM-41 yielded a catalyst with higher activity compared to the corresponding silica analogue in terms of turnover rates and product yield in the Prins condensation of isobutene and formaldehyde to isoprenol.

Full text of DOI:10.1039/b101843o

A comparative study of an MCM-41 anchored quaternary ammonium
chloride/SnCl catalyst and its silica gel analogue
4
T. M. Jyothi, M. L. Kaliya, M. Herskowitz and M. V. Landau*
Blechner Center for Industrial Catalysis and Process Development, Dept. of Chemical Engineering, Ben
Gurion University of Negev, Beer Sheva-84105, Israel. E-mail: mlandau@bgumail.bgu.ac.il
Received (in Cambridge, UK) 27th February 2001, Accepted 19th April 2001
First published as an Advance Article on the web 15th May 2001
A novel reusable Lewis acid catalyst has been prepared by
the heterogenization of a Lewis acid/tetrapropylammonium
adduct; anchoring of tin chloride on quaternary ammonium
chloride functionalized MCM-41 yielded a catalyst with
higher activity compared to the corresponding silica ana-
logue in terms of turnover rates and product yield in the
Prins condensation of isobutene and formaldehyde to
isoprenol.
The heterogenization of common Lewis acid catalysts is a
subject of immense importance in view of achieving clean
technology, based on the fact that heterogeneous catalysts are,
4
Scheme 2 Preparation of SnCl anchored on tetraalkylammonium chloride
functionalized MCM-41; TPA = tetrapropylammonium chloride; R =
propyl group.
in principle, best suited for industrial applications, as ideally
1
they can be recycled and reused easily. A direct immobilization
of such catalysts on an inorganic support is not often
advantageous due to the possible leaching of the supported
species during the reaction and also in many cases the activity
immobilized on inorganic supports has been recently demon-
strated by Holderich and coworkers among others. Similarly,
6
2
is considerably less than for the unsupported Lewis acid. To
we adopted the nucleophilic addition of tetraalkylammonium
chloride to tin chloride to obtain penta-coordinated tin anionic
enhance the activity it may be preferable to anchor the metal
catalyst to the support through a chemical chain rather than
directly bond it to the support surface.
+
2 7
4 5
species, [NR ][SnCl ]. These types of complexes were
immobilized on silica gel and MCM-41 supports by a three-step
MCM-41 materials which are characterized by highly
ordered pore systems with tunable pore sizes, large specific
surface area and pore volume, and high density of surface
silanols, provide excellent opportunities in inclusion chemistry
and catalysis. A large number of functionalized entities
including both organic and inorganic ligands has been in-
troduced in the channels to generate novel catalyst materials.3
Here we report the preparation of a novel Lewis acid catalyst
procedure (Scheme 2). Pure silica MCM-41 (pore diameter
21
=
3.7 nm, surface area = 1100 m2
g ) was synthesized
following a modified procedure involving pH adjustment and
salt addition during crystallization to improve the wetting
8
stability. Silica gel (pore diameter r = 6 nm, surface area =
2
21
5
00 m g ) was purchased from Aldrich. To a suspension of 10
g of dehydrated silica/MCM-41 support in 50 g anhydrous
toluene was added 6 g of a toluene solution of trimethoxy-
silylpropyl chloride. The suspension was refluxed for 6–7 h in
an oil bath followed by filtering and washing with excess
toluene to remove the excess of reagent. 5 g of chloropropylated
silica and 7 g of tripropylamine were further suspended in 30 g
of anhydrous toluene and refluxed for 24 h. After the reaction
the material was filtered off, washed fully with excess toluene
and dried in vacuo. Elemental analysis revealed a loading of
4
comprising SnCl as the active component anchored on a
tetraalkylammonium chloride functionalised MCM-41 mate-
rial. A comparison is made with the corresponding silica gel
analogue in the model reaction of isobutene–formaldehyde
condensation to 3-methylbut-3-en-1-ol (isoprenol), which is an
important synthesis block for industrially valuable terpenes.4
4 2
Lewis acid catalysts such as SnCl and ZnCl have been
employed in the synthesis of isoprenol from isobutene and
formaldehyde under mild conditions, however, the yield was
low due to the formation of many side products (Scheme 1).5
So called ionic liquids or molten salts which can be prepared
0
.98 and 2.5 mmol of chloropropyl groups per gram of silica gel
and MCM-41 supports, respectively. The amount of chloride
ions present on the silica gel and MCM-41 support after the
reaction with tripropylamine as determined by titration with
+
2
4
by the reaction of a tetraalkylammonium halide (NR X ) and
3
standard AgNO solution and potassium chromate indicator
a Lewis acid (MX) are reported to be effective catalysts in
Friedel–Crafts reactions and the utility of such catalysts directly
2
1
were found to be 0.50 and 1.20 mmol g
of support,
respectively.
1
.04 g of a solution of anhydrous SnCl in 10 ml
4
dichloromethane was added dropwise to a suspension of the
functionalized silica in 40 ml dichloromethane followed by
stirring overnight at ambient temperature. After reaction, the
material was exhaustively washed with dichloromethane and
dried in vacuo. The starting silica gel and MCM-41 before use
were dehydrated in vacuo at 450 °C. All operations including
4
preparation of the unsupported SnCl –TPACl complex and its
direct reaction with unfunctionalized silica were conducted
inside the glove box under dry nitrogen atmosphere in order to
prevent any hydrolysis of Sn compounds.
The tin chloride loadings of silica and MCM-41 hybrid
catalysts as determined by energy dispersive X-ray analysis
(EDAX-JEM-35, JEOL Co., link system AN-1000, Si–Li
Scheme 1 Prins condensation of isobutene and formaldehyde; IP
-methylbut-3-en-1-ol (isoprenol), DMD = 4,4-dimethyl-1,3-dioxane.
=
3
992
Chem. Commun., 2001, 992–993
This journal is © The Royal Society of Chemistry 2001
DOI: 10.1039/b101843o

Table 1 Condensation of isobutene and formaldehyde to 3-methylbut-3-en-1-ol (MBOH) over silica based catalysts and catalyst recyclinga
SnCl
loading/mmol
(g support)
4
Conversion of
formaldehydeb
(%)
MBOH
selectivity
(%)
c
MBOH yieldd
(%)
21
Entry
Catalyst
1
2
3
4
5
6
7
8
9
SnCl
SnCl
SIL–TPA SnCl
MCM–TPA SnCl
Recycling 1
Recycling 2
MCM–TPA SnCl
SIL–TPA–SnCl
SIL/SnCl
4
4
—
—
97.4
54.6
89.0
88.1
94.0
94.1
93.5
90.1
79.8
63.2
53.0
68.7
56.4
71.4
69.9
70.0
90.1
45.5
49.2
–TPACle
77.1 (2.68)
64.1 (2.22)
76.0 (2.63)
74.3
74.9
100
+
2f
5
0.46
1.11
1.10
—
1.11
0.32
0.95
+
2g
2i
5
h
+
5
j
4
57.0
77.9
4
a
5
4
6 g isobutene, 3 g paraformaldehyde, catalyst containing 4 mmol of SnCl and 40 g of chloroform solvent were introduced in the reactor and stirred for
b
23
c
2
h at 60 °C. The turnover number (TON) based on the no. of mol of formaldehyde converted per mol of catalyst per s (310 ) shown in parentheses. Other
d
products included 4,4-dimethyl-1,3-dioxane and traces of poly-condensation products. Yield was determined by using butan-2-ol as an internal standard.
e
f
complex. g TPACl-functionalised MCM-41–SnCl
complex. h Recycling of MCM–
TPACl–SnCl
4
complex. TPACl-functionalised silica–SnCl
4
4
+
2
i
j
TPA SnCl
5
after exhaustive washing with dichloromethane. Reaction continued for 3.5 h. Direct immobilization of complex on silica.
(TON = 2.63 3 10 s21) as that obtained in solution with the
23
detector, sensitivity > 0.1 wt%) were found to be 0.46 and 1.11
mmol g , respectively. The MCM-41 structure remained intact
after the catalyst anchoring. The retention of organic groups on
2
1
23 21
homogeneous catalyst (TON = 2.68 3 10
s ). We believe
that the accessibility of the active sites for the reactants will be
highest in the present case as it is anchored to the support
through a organic chain rather than being directly bonded to the
support surface. In addition, the regular and well ordered
hexagonal array of pores present in the MCM-41 support
provides nanosized micro-reactors for the reaction. The better
catalytic activity obtained with the MCM-41 supported catalyst
compared to the silica supported counterpart indicates a
favorable reaction environment inside the well ordered pores of
the former where the active sites are grafted through an organic
spacer. Another important feature of this catalyst is the higher
concentration of active sites per gram of the support compared
to the silica support.
4
the functionalized support after anchoring the SnCl was proved
by IR spectroscopy. The formation of a complex between tin
chloride and tetrapropylammonium chloride in dichlorome-
thane solution was confirmed by 1 Sn NMR (external stan-
dard). A shift in the NMR signal from 2665.0 to 2725.2 ppm
after the reaction with tetrapropylammonium chloride reflects a
19
7
change in the coordination number at tin. The tin/chlorine
atomic ratio in the anchored catalysts determined by elemental
+
2
analysis (EDAX): 0.195 for SIL–TPA SnCl
5
and 0.204 for
2
5
, was in accordance with such complex
+
MCM–TPA SnCl
formation. The solid state 1 Sn MAS NMR study of the SnCl
19
–
4
tetraalkylammonium adduct anchored on MCM-41 is now in
progress in order to obtain more precise definition of the
environment around the Sn.
T. M. J. is grateful to the Blechner foundation for a post-
doctoral fellowship.
The results of the Prins condensation of isobutene and
formaldehyde to isoprenol over different catalysts are summa-
rized in Table 1.†‡ It is clear that complexation of tin chloride
with tetrapropylammonium chloride improves the selectivity
towards the unsaturated alcohol. Tin chloride alone under
anhydrous conditions displayed high activity but the selectivity
to MBOH was poor. The tin chloride catalysts immobilized on
organic quaternary ammonium functionalized silica/MCM-41
showed highest selectivity towards the unsaturated alcohol,
MBOH (88–95%). More interestingly, the MCM-41 based
catalyst is far more active than its silica counterpart and
outperforms the latter both in terms of isoprenol yield and
catalyst turnover. An X-ray fluorescence spectroscopic analysis
of the filtrate after the reaction (X-ray spectrometer Phillips
PW-1410; sensitivity > 0.5 ppm Sn) revealed that the new
catalysts are resistant to leaching under the reaction conditions.
Furthermore, when a subsequent reaction was performed with
the filtrate after separating the catalyst by adding fresh
formaldehyde and isobutene, no formaldehyde conversion was
observed. Recycling of the catalyst leads to no appreciable loss
in activity. Increasing the reaction time to 3.5 h, increased the
MBOH yield to a maximum of 90%. The performance of an
Notes and references
† Tetrapropylammonium chloride (4 mmol) was dissolved in 40 g of dry
dichloromethane and 1.04 g (4 mmol) of anhydrous tin chloride was added
slowly with constant stirring. The mixture was stirred overnight at ambient
temperature and the solution was concentrated in vacuo to obtain a white
crystalline material which was recrystallised from dichloromethane–hexane
and stored under moisture free conditions. Analysis: found (calc.) Sn 24.78
(24.61), Cl 36.40 (36.74)%.
‡
The Silica/TPA–SnCl catalyst was prepared by adding a chloroform
4
solution of the complex prepared as mentioned above to 4 g of pre-dried
silica support suspended in chloroform with constant stirring. The excess
solvent was removed after overnight stirring and the material was extracted
with chloroform for 12 h in a Soxhlet apparatus. The material was finally
dried and kept under moisture free conditions before use.
1
2
Chemistry of Waste Minimisation, ed. J. H. Clark, Chapman and Hall,
London, 1995, p. 141.
R. S. Drago, S. C. Petrosius and P. B. Kaufman, J. Mol. Catal., 1994, 89,
3
17; A. A. Krzywicki and M. Marczewski, J. Chem. Soc., Faraday Trans.
1
, 1980, 1311; S. J. Barlow, T. W. Bastock, J. H. Clark and S. R. Cullen,
Tetrahedron Lett., 1993, 34, 3339; A. Cornellis, A. Gerstmans, P. Laszlo,
A. Mathy and I. Zreba, Catal. Lett., 1990, 6, 103.
3 D. Brunel, Microporous Mesoporous Mater., 1999, 27, 329.
4 A. P. Courtot, J. Chem. Soc., 1906, 90, 788.
SnCl
4
/silica catalyst, prepared by treating a chloroform solution
of tin chloride and silica, in the selective synthesis of MBOH is
poor compared to immobilized complexes. Also, a direct
immobilization of the complex on silica (Silica/TPA–SnCl )
4
5
6
E. Arundale and L. A. Mikelska, Chem. Rev., 1952, 52, 505.
G. A. Parshall, J. Am. Chem. Soc., 1972, 94, 8716; P. Wasserscheid and
W. Keim, Angew. Chem., Int. Ed., 2000, 39, 3772; C. DeCastro, E.
Sauvage, M. H. Valkenberg and W. F. Holderich, J. Catal., 2000, 196,
was not advantageous in terms of MBOH yield.
The present MCM-41 based Lewis acid catalyst system
satisfactorily meets the ultimate aims of heterogenization of
metal complexes. In short, we have demonstrated the prepara-
tion and application of a novel stable, reusable, heterogeneous
Lewis acid catalyst, which shows almost the same turnover rate
8
6.
7
S. E. Johnson and C. B. Knobler, Organometallics, 1992, 11, 3684;
Kuraray Co. Ltd., Jpn. Pat., H09-262478, 1997.
8 M. V. Landau, S. P. Varkey, M. Herskowitz, O. Regev, S. Pevzner, T. Sen
and Z. Luz, Microporous Mesoporous Mater., 1999, 33, 149.
Chem. Commun., 2001, 992–993
993