Aminopropylated MCMs as base catalysts: A comparison with aminopropylated silica

Article abstract of DOI:10.1039/a704156j

Aminopropyl-functionalised MCMs, prepared via a one-pot method, are found to be effective base catalysts for the Knoevenagel reaction, with significant improvements in terms of turnover number and solvent dependence to the ostensibly similar aminopropylsilica.

Full text of DOI:10.1039/a704156j

View Article Online / Journal Homepage / Table of Contents for this issue
Aminopropylated MCMs as base catalysts: a comparison with
aminopropylated silica
Duncan J. Macquarrie* and Dominic B. Jackson
Department of Chemistry, University of York, Heslington, York, UK YO1 5DD
T
Aminopropyl-functionalised MCMs, prepared via a one-pot
method, are found to be effective base catalysts for the
Knoevenagel reaction, with significant improvements in
terms of turnover number and solvent dependence to the
ostensibly similar aminopropylsilica.
N
2
of underivatised silica [E (SiO ) = 0.89–0.96, depending on
drying conditions. Similar values are obtained for unmodified
MCMs]. This difference can be ascribed to the larger amount of
exposed, unfunctionalised surface in the case of the MCM
materials, and, in agreement with this, the 2.5 mmol g loading
catalyst has a value lower than that of the 1.2 mmol g
2
1
2
1
We, and others, recently reported a simple one-pot method for
the preparation of surface-modified monodisperse MCM-type
silica.1 This method can be used to prepare organically
modified materials in high yield, which have very high surface
areas (650–1600 m g ) and are functionalised with a variety
of organic groups. The pore size distribution of these materials
is very narrow and in the mesoporous region (1.8–3.9 nm). Such
properties would make these materials potentially very inter-
esting as novel catalysts. We have recently reported that
aminopropylated silica is an effective base catalyst, which is
material. Such a difference might be expected to lead to
different solvent dependency in reactions catalysed by the
MCMs compared to the silica-derived catalysts, owing to
different partitioning of reactants between the catalyst surface
and the bulk solution.
The catalysts were then evaluated in a series of reactions.
Reactions used were the Knoevenagel condensation of alde-
,2
2
21
hydes and ketones with ethyl cyanoacetate, a model reaction
which has been studied in some detail.3
,5,6,7
The CH acid and
the ketone (20 mmol of each) were added to a suspension of the
catalyst (0.25 g) in the appropriate solvent (25 ml) at reflux. All
reactions were run with continuous removal of water. Reactions
were followed by GC using n-dodecane as internal standard,
and products were isolated using conventional techniques.
limited in its usefulness by an unusual solvent dependency, and
3
which is poisoned by amide formation at the NH
2
groups. We
now wish to report our findings on aminopropylated MCMs,
materials which display considerable differences in catalytic
activity to the apparently similar aminopropylsilicas.
Results are shown in Table 2. Yields are generally high and
compare favourable well with those in the literature.3
,5,6,8
As
The aminopropylated MCMs were prepared according to
1
known methods. Three catalysts were prepared, whose struc-
tural and compositional characteristics are shown in Table 1.
The ability to increase loading of these materials by changing
the ratio of the two silanes [aminopropyl(trimethoxy)silane and
tetraethoxysilane] used in their preparation makes this method-
ology more flexible than the post-modification of silica, where
Table 1 Selected physical parameters of catalysts
Pore
diameter/nm
SSA/
m g
Loading/
mmol g
2
21
21
T
Catalyst
E
N
1
2
3
3.6
3.7
1.8
8.0 (br)
756
745
715
254
1.2
2.5
1.2
0.90
0.82
0.86
0.56
2
1
maximum loadings are in the region of 1 mmol g . The surface
area and pore size distributions are as expected for such
materials, and their surface polarity, as measured using
AMPSa
0.95
4
adsorption of Reichardts dye is significantly higher than that of
a
the corresponding silica-derived materials, and approaches that
Aminopropyl-substituted silica.
Table 2 Selected reaction data for reactions catalysed by aminopropyl-MCMs
R
CO2Et
R
CO2Et
O
+
+ H2O
R′
CN
R′
CN
R
RA
Catalyst
Solvent
T/°C
t/h
Yield (%)a
TONb
Ph
n-C
H
H
1
1
1
1
1
1
1
1
3
1
1
2
2
1
2
cyclohexane
cyclohexane
cyclohexane
cyclohexane
toluene
82
82
82
82
110
78
61
81
82
82
110
110
110
110
110
36
0.5
4
12
2
4
4
4
4
30
18
4
4
72
36
94
93
96
7
H
15
> 6000
cC
5
5
5
5
5
5
5
H
10
H
10
H
10
H
10
H
10
H
10
H
10
2450 (650)
c
cC
cC
cC
cC
cC
cC
70
92
11
18
16
49
89
95
97
95
49
48
EtOAc
CHCl
3
DCEd
cyclohexane
cyclohexane
toluene
toluene
toluene
Et
Et
Et
Me
Me
Me
Et
Et
Et
Bu
Ph
Ph
1127 (265)
n
1244
55 (250)
47
toluene
toluene
a
b
GC yields with n-dodecane as internal standard. Isolated yields are 3–7% lower. Number of moles product per mole of NH
2
groups. Figures in brackets
are for reactions using 1.0 mmol g2 aminopropylsilica as catalyst. Reaction carried out without removal of water. d 1,2-Dichloroethane.
1
c
Chem. Commun., 1997
1781

View Article Online
100
(a)
(
(
a)
b)
(
b)
7
5
90
80
70
(c)
5
0
5
2
(d)
(e)
(f )
0
0
2
4
6
8
t / h
Fig. 2 Effect of solvents on the rate of reaction with aminopropyl-MCM 1:
(
(
a) toluene, (b) cyclohexane, (c) hexane, (d) 1,2-dichloroethane,
e) chloroform, (f) ethyl acetate
2400
2200
2000
n / cm
1800
1600
–1
which it is effective,† the activity of the MCM catalyst seems to
follow a simpler trend with respect to reflux temperature of the
solvent. However, it is clear that more polar solvents are
disadvantageous (Fig. 2). While the reasons for this behaviour
are still under investigation, it may be the case that the rate of
reaction is influenced by the partitioning of the (polar) reactants
between the catalyst surface (polar) and the bulk medium (non-
polar). The silica-based catalysts have a significantly lower
Fig. 1 IR spectra of spent catalysts recovered from Knoevenagel reactions:
a) spent aminopropylsilica and (b) spent aminopropyl-MCM
(
can be seen, the overall reactivity bears some similarity to that
of aminopropylated silica.3,5 Aldehydes and ketones both react
with ethyl cyanoacetate, chosen as a carbon acid of moderate
activity and synthetic utility. Surprisingly, benzaldehyde reacts
extremely sluggishly with the MCM catalysts, despite being a
facile substrate with both the silica-derived catalysts and with
homogeneous amines such as piperidine. Otherwise, the relative
reactivity of ketones is typical of both aminopropyl-modified
silica and of most homogeneous systems. For reactions of
ketones, the rates of reaction in cyclohexane, the optimum
solvent for the silica variant, are typically 20–30% slower with
the catalysts described here. However, as will be discussed later,
the MCM-type catalysts are less restricted by solvent than their
amorphous silica equivalents. Immediately obvious is the
generally increased turnover numbers (TONs), although the
reactivity and TON of acetophenone is disappointingly low.
This, coupled with the remarkably low activity of benzalde-
hyde, may suggest that the reduced electrophilicity of the
aromatic carbonyl, relative to the aliphatic, plays an unusually
significant role in reactions catalysed by these materials.
Other points of interest are that the small pore (1.8 nm)
catalyst 3 gives a poor conversion compared to the larger pore
analogue 1. However, it is interesting to note that the rate of
reaction caused by both catalysts is essentially identical, the
difference being that 1 remains active for longer. Increasing the
4
polarity, as measured by the Reichardts dye method (Table 1).
This might mean that partitioning away from the catalyst is
pronounced even in moderately polar solvents such as toluene,
whereas the much more polar MCM catalysts can more
effectively compete for the substrate, thus effectively extending
the range of useful solvents, leading to reaction rates greater
than those achievable with the silica-based materials.
In conclusion, the novel organofunctionalised MCM cat-
alysts described here represent novel catalytic materials,
differing significantly in many respects from their post-
functionalised silica counterparts. Under optimum conditions
they can outperform the silicas, in terms of both activity and
catalyst turnover.
D. J. M. thanks the Royal Society for a University Research
Fellowship.
Footnotes and References
*
†
E-mail: djm13@york.ac.uk
The relative rates of reaction at reflux in various solvents are as follows:
octane
> heptane > cyclohexane > hexane > toluene > pentane
>> chlorobenzene, 1,2-dichloroethane. Octane and heptane, however, lead
to rapid deactivation of the catalyst. Initial rate studies with the alkane
solvents indicate an Arrhenius-type behaviour, showing a simple tem-
perature dependence of rate. More polar solvents such as toluene, PhCl and
other chlorinated solvents show rates much reduced from those achieved in
a hydrocarbon of similar boiling point.
2
1
21
loading from 1.2 mmol g
to 2.5 mmol g
causes a
corresponding increase in reaction rate. Such an increase in
loading is readily achieved by altering the ratio of silanes in the
preparation of the MCM catalysts, but is not possible with the
2
1
post-functionalisation of silica, where 1.0 mmol g is the
maximum loading achievable. In connection with TON, the
mode of catalyst deactivation also differs between the two
catalyst types. For the silica catalysts, there is clear evidence for
a slow, irreversible formation of surface-bound amide groups,
arising from the reaction of the surface-bound primary amine
and the ester of the ethyl cyanoacetate [Fig. 1(a)]. In the case of
the MCM catalysts, recovered spent catalyst does not display
bands for nitrile or amide groups, but rather indicates the
presence of some adsorbed organics [Fig. 1(b)]. The identity of
the species responsible is currently the subject of investiga-
tion.
1
2
3
D. J. Macquarrie, Chem. Commun., 1996, 1961.
S. L. Burkett, S. D. Sims and S. Mann, Chem. Commun., 1996, 1367.
D. J. Macquarrie, J. E. G. Mdoe, A. Priest, A. Lambert and J. H. Clark,
React. Funct. Polym., in the press.
4 S. J. Tavener, D. J. Macquarrie, G. Gray, P. A. Heath and J. H. Clark,
Chem. Commun., 1997, 1147.
E. Angeletti, C. Canepa, G. Martinetti and P. Venturello, Tetrahedron
Lett., 1988, 29, 2261.
E. Angeletti, C. Canepa, G. Martinetti and P. Venturello, J. Chem. Soc.,
Perkin Trans. 1, 1989, 105.
G. Jones, Org. React., 1967, 15, 204 and references therein.
J. A. Cabello, J. M. Campelo, A. Garcia, D. Lisna and J. M. Marinas,
J. Org. Chem., 1984, 49, 5195.
5
6
7
8
A further point of divergence between the two types of
catalyst is their behaviour in different solvents. Whereas the
silica-based catalyst has a very limited range of solvents in
Received in Cambridge, UK, 16th June 1997; 7/04156J
1782
Chem. Commun., 1997