Amino-grafted metallosilicate MCM-41 materials as basic catalysts for eco-friendly processes

Article abstract of DOI:10.1016/j.cattod.2010.04.031

Thecompounds3-aminopropyl-trimethoxysilane (APMS), [3-(2-aminoethylamino) propyl]trimethoxysilane (2APMS) and 3-[2-(2-aminoethylamino)ethylamino]propyl- trimethoxysilane (3APMS) were loaded by grafting on MCM-41 matrices of various chemical compositions, aluminosilicate (AlMCM-41; Si/Al = 64) and niobosilicate (NbMCM-41; Si/Nb = 64). The materials prepared were characterized using XRD, N2 adsorption/desorption, thermogravimetric analysis, FTIR spectroscopy and elemental analysis. Thermal stability of APMS was found not to depend on the chemical composition of the support. The higher stability of 2APMS and 3APMS was related to the hydrogen bonding between amine groups and surface hydroxyls. The models of amine grafted in MCM-41 materials were proposed. Basic activity in 2-propanol dehydrogenation and Knoevenagel reactions was found strongly dependent on the nature of the support and changed in the following order: APMS/AlMCM-41> APMS/NbMCM-41. The acidity of the support is considered for the explanation of the activity sequence.

Full text of DOI:10.1016/j.cattod.2010.04.031

Catalysis Today 152 (2010) 119–125
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Amino-grafted metallosilicate MCM-41 materials as basic catalysts for
eco-friendly processes
Davinia Blasco-Jiméneza,b, Izabela Sobczakb, Maria Ziolekb,∗, Antonio J. López-Peinadoa,
Rosa M. Martín-Arandaa,∗∗
a Dpto. Química Inorgánica y Química Técnica, Universidad Nacional de Educación a Distancia, c/Senda del Rey, 9, 28040 Madrid, Spain
b Adam Mickiewicz University, Faculty of Chemistry, Grunwaldzka 6, 60-780 Poznan, Poland
a r t i c l e i n f o
a b s t r a c t
Keywords:
AlMCM-41
NbMCM-41
Aminopropyl-trimethoxysilanes
Thermal stability
Knoevenagel reactions
The compounds 3-aminopropyl-trimethoxysilane (APMS), [3-(2-aminoethylamino)propyl]trimethoxysil-
ane (2APMS) and 3-[2-(2-aminoethylamino)ethylamino]propyl-trimethoxysilane (3APMS) were loaded
by grafting on MCM-41 matrices of various chemical compositions, aluminosilicate (AlMCM-41;
Si/Al = 64) and niobosilicate (NbMCM-41; Si/Nb = 64). The materials prepared were characterized using
XRD, N2 adsorption/desorption, thermogravimetric analysis, FTIR spectroscopy and elemental analysis.
Thermal stability of APMS was found not to depend on the chemical composition of the support. The
higher stability of 2APMS and 3APMS was related to the hydrogen bonding between amine groups and
surface hydroxyls. The models of amine grafted in MCM-41 materials were proposed. Basic activity in
2-propanol dehydrogenation and Knoevenagel reactions was found strongly dependent on the nature
of the support and changed in the following order: APMS/AlMCM-41 > APMS/NbMCM-41. The acidity of
the support is considered for the explanation of the activity sequence.
© 2010 Elsevier B.V. All rights reserved.
1. Introduction
Post-synthesis grafting of organotrialkoxysilane [6,8,13,16,17]
has been effectively used to incorporate active organic functional
The end of the 20th century has brought increasing interest in
the use of silica materials modified with aminopropyl and/or other
organic moieties as catalysts [e.g. 1–4] mainly in base-catalysed
reactions. An important drawback of these materials is the lack of
uniform and well ordered mesoporosity, which is of significance
in catalysis. Discovery of the M41S family of mesoporous materi-
als [5] has stimulated research in the area of surfactant-templated,
periodic mesoporous hosts. Much progress has been made in the
last few years in the development of organo-modified mesoporous
molecular sieves [6–10]. Several reviews have been devoted to this
important type of materials [e.g. 11–15].
Development of clean processes for production of fine chem-
icals requires the use of recyclable solid catalysts. Besides acid
catalysts, solid base catalysts need to be explored. The design of
base catalyst featuring large mesoporous surfaces allows perform-
ing base-catalysed reactions in the fields of production of perfumes,
pharmaceuticals and agro-chemicals.
groups onto the mesoporous silicas surface. Such prepared cata-
lysts show high activity for base-catalysed reactions because of
their high surface area and large pore channels. Moreover, their
activity is expected to vary by controlling the amine content, its
nature and strength (primary, secondary, or tertiary amine).
The primary amino groups bound on mesoporous silicas, the
materials MCM-41, SBA-15, have been previously reported efficient
catalysts for the Knoevenagel condensation reaction under mild
conditions [6,8]. The mechanism of their catalytic activity has also
been discussed. The reaction of benzaldehyde and ethylcyanoac-
etate (ECA) in the presence of primary amine may proceed very
fast via the formation of imine in which the benzaldehyde molecule
first reacts with the –NH2 group present on the catalyst surface to
form an intermediate imine (C N double bond) compound. Next,
the acidic methylene group of ECA attacks the C N double bond
of imine and forms the condensation product leaving behind the
–NH2 group. However, in the same reaction in the presence of sec-
ondary or tertiary amines, the more common aldol type mechanism
prevails. Thus, the higher activity of primary amine vs. secondary
or tertiary one has been attributed to the faster reaction through
imine formation [e.g. 8].
∗
Corresponding author. Tel.: +48 61 829 1243; fax: +48 61 829 1505.
Corresponding author. Tel.: +34 91 398 7350.
E-mail addresses: ziolek@amu.edu.pl (M. Ziolek), rmartin@ccia.uned.es
The idea of our work was to get
a better insight into
∗∗
the role of chemical composition of metallosilicate mesoporous
ordered supports on the catalytic activity of amino-grafted
(R.M. Martín-Aranda).
0920-5861/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.cattod.2010.04.031

120
D. Blasco-Jiménez et al. / Catalysis Today 152 (2010) 119–125
pressed wafers of the materials (∼5 mg cm−1) were placed in the
materials. In our previous work [18] the focus was on sil-
icate and niobosilicate MCM-41 grafted with one amine (3-
vacuum cell and evacuated in the range RT–673 K for 1 h.
aminopropyl-trimethoxysilane
– APMS). One of the aims of
this study was to determine the influence of the nature
of the amine (3-aminopropyl-trimethoxysilane – APMS, [3-(2-
aminoethylamino)propyl]trimethoxysilane – 2APMS and 3-[2-(2-
aminoethylamino)ethylamino]propyl-trimethoxysilane – 3APMS)
linked to MCM-41 mesoporous molecular sieves containing vari-
ous T atoms (Al and Nb) on the properties and catalytic activity of
the catalysts obtained. The thermal stability of the amines grafted
on aluminosilicate and niobosilicate MCM-41 supports and their
activity in the base-catalysed reactions (tested in Knoevenagel con-
densation) have been analysed. Moreover, the role of the type of
metal (Nb or Al) in MCM-41 support has been considered.
2.4. Catalytic tests
2.4.1. 2-Propanol decomposition
The 2-propanol conversion (dehydration and dehydrogena-
tion) was performed, using a microcatalytic pulse reactor inserted
between the sample inlet and the column of a CHROM-5 chro-
matograph. The catalyst bed (0.02 g with
a size fraction of
0.5 < Ø < 1 mm) was first activated at 673 K for 2 h under helium flow
(40 cm3 min−1). The 2-propanol (Aldrich) conversion was studied
at 423, 473, 523, 573 K using 2 ␮l pulses of alcohol under helium
flow (40 cm3 min−1). The reactant and reaction products: propene,
2-propanone (acetone) and diisopropyl ether were analysed using
CHROM-5 gas chromatograph on line with microreactor. The reac-
tion mixture was separated on 2 m column filled with Carbowax
400 (80–100 mesh) at 338 K in helium flow (40 cm3 min−1) and
detected by TCD.
2. Experimental
2.1. Synthesis of the catalysts
Mesoporous molecular sieves of MCM-41 type were synthe-
sized by the classical hydrothermal method [5] and modified in
the preparation of AlMCM-41 and NbMCM-41-64 [19].
2.4.2. Knoevenagel probe reaction
The experimental work was carried out in a Pyrex-glass batch
reactor. Theequimolarmixturesofthe two reactants, benzaldehyde
and the active methylenic compound (ethylcyanoacetate, or diethyl
malonate) without any solvent were pre-heated to the reaction
temperature, 393 K, under continuous stirring. Then, depending on
the active methylenic compound employed, ethylcyanoacetate or
diethyl malonate, 1, 2 or 10 wt% of solid catalyst was added respec-
tively and the reaction started. Samples were taken periodically,
and the evolution of the reaction was followed by gas chromatog-
raphy, using an Agilent Technologies 6890N chromatograph.
The influence of external and internal diffusion on the total rate
of the reaction was determined by changing the stirring rate in
the range 1000–3500 rpm and by the use of different particle sizes
(0.074, 0.140, 0.250 mm diameter), respectively.
2.2. Grafting
The ammonium forms of mesoporous molecular sieves,
NH4AlMCM-41 and NH4NbMCM-41 were obtained via cation
+
exchange with NH4 ions (from NH4Cl solution). They were trans-
formed to hydrogen forms by calcination at 673 K for 2 h.
The hydrogen forms, HAlMCM-41 and HNbMCM-41, were the
supports for grafting of 3-aminopropyl-trimethoxysilane (APMS),
[3-(2-aminoethylamino)propyl]trimethoxysilane (2APMS) and
3-[2-(2-aminoethylamino)ethylamino]propyl-trimethoxysilane
(3APMS) (Aldrich). The grafting procedure was carried out accord-
ing to [20] and was as follows: 2.3 g of the support powders
were refluxed in a dry toluene solution (200 mL) containing
10 mL of APMS at 373 K for 18 h. The catalysts were recovered
by filtration followed by washing in dry toluene (200 mL), water
(100 mL) and acetonitrile (20 mL). The powder was dried in
an oven at 373 K and stored for the later use. The catalysts
obtained were labelled as APMS/TMCM-41, 2APMS/TMCM-41,
3APMS/TMCM-41.
The catalysts were recycled by washing them in acetone, subse-
quently dried in an oven at 373 K overnight.
3. Results
3.1. Texture/structure
XRD patterns of niobosilicate MCM-41 shown in Fig. 1A indicate
that grafting of APMS on this material leads to the disordering of the
hexagonal arrangement of mesopores in the long range (deduced
from the drop in intensity of the XRD peaks at 2ꢁ = 3–5◦). It is
seen even after normalisation of (1 0 0) peak to the same intensity
value (Fig. 1B). Interestingly, the disordering is less pronounced
in amino-grafted AlMCM-41 (Fig. 1C) which preserves the order-
ing of mesopores after APMS immobilization. It is important to
stress that the amount of nitrogen in the APMS-grafted materi-
als (Table 1) differs depending on the chemical composition of
the support (APMS/NbMCM-41 > APMS/AlMCM-41). The grafting
of 2APMS and 3APMS on aluminosilicate support giving rise to a
2.3. Catalysts characterization
The materials prepared were characterized using XRD, N2
adsorption/desorption, thermogravimetric analysis, FTIR spec-
troscopy and elemental analysis.
XRD measurements were carried out with a Bruker AXS D8
Advance diffractometer with Cu K␣ radiation (ꢀ = 0.154 nm), with
a step size of 0.02◦ and 0.05◦ in the small-angle and high-angle
ranges, respectively.
The N2 adsorption/desorption isotherms were obtained in a
Micromeritics ASAP equipment, model 2010. The samples were
pre-treated in situ under vacuum at 333 K for 24 h. The surface
area was calculated by the BET method. The pore size distribu-
tions (PSD) and the mesopore volumes were determined from the
adsorption/desorption isotherms using DFT.
The thermogravimetric analyses of the solids were carried out
with ATA Instruments thermogravimetric system, model SDT Q600
TG–DSC.
Table 1
Texture parameters of the catalysts and the amount of nitrogen in the amino-grafted
MCM-41 materials.
Catalyst
Surface area
BET (m2/g)
Average pore
Mesopore
N loading
diameter (nm) volume (cm3/g) (mmol/g)
AlMCM-41
909
637
277
1049
707
3.9
3.3
2.9
3.8
3.2
2.9
0.76
0.37
0.19
0.96
0.44
0.14
0.2
1.8
4.5
0.1
2.0
2.1
APMS/AlMCM-41
3APMS/AlMCM-41
NbMCM-41
APMS/NbMCM-41
2APMS/NbMCM-41 279
The elemental analyses of the solids were carried out with Ele-
mentar Analyser Vario EL III.
Infrared spectra were recorded with the Vector 22 (Bruker)
spectrometer (resolution 4 cm−1
, number of scans = 64). The

D. Blasco-Jiménez et al. / Catalysis Today 152 (2010) 119–125
121
Fig. 1. XRD patterns of MCM-41 catalysts in the small-angle range; (A) and (C) original XRD patterns, (B) after normalisation of (1 0 0) peak to the same intensity value.
higher amount of nitrogen only slightly decreases the intensity
of XRD reflexes indicating that even these amines do not change
significantly the ordering of the material.
The results in Table 2 also show that 2APMS and 3APMS decom-
pose at higher temperatures than APMS. The temperature of the
first intensity drop ((–NH2) IR band) for both, 2APMS and 3APMS is
573 K.
The textural parameters calculated from N2 adsorp-
tion/desorption isotherms are shown in Table 1. They indicate a
decrease in the surface area of amino-grafted AlMCM-41 (about
30% decrease) and NbMCM-41 (ca. 33% drop). The mesopore vol-
ume also decreases in the amino-grafted metallosilicate materials.
The decrease in the mesopore volume after grafting of amines
when compared with the pristine support suggests the collapse of
some mesopores or blockage effect.
The thermogravimetric analyses (TGA), exemplary shown in
Fig. 2 for AlMCM-41 based materials, confirmed the results of FTIR
spectroscopy. The TG curves for silicate and niobosilicate MCM-41
have been compared in our earlier paper [18]. It has been evidenced
that the pristine niobosilicate NbMCM-41 exhibits one drop of
weight at ca. 373 K caused by desorption of water adsorbed. APMS-
grafted sample shows additional weight loss that ends at ca. 900 K,
irrespectively of the chemical composition of the matrix. The same
is true for APMS/AlMCM-41 (Fig. 2). TG curve of 3APMS/AlMCM-41
has a shape similar to that obtained for APMS immobilized mate-
rial, but the temperature of total decomposition of the amine is
higher. Moreover, the DTG curve reveals three maxima indicating
three steps of 3APMS decomposition.
3.2. Thermal stability of aminoalkoxysilanes in mesoporous
matrices
The thermal stability of all APMS in MCM-41 type supports was
estimated on the basis of infrared spectroscopy (FTIR) measure-
ments after evacuation of the material at various temperatures
and the thermogravimetric measurements. The stability of amine is
deduced from the temperature at which the drop in the band inten-
sity of the N–H vibration in the NH2 group takes place (the band at
1596 cm−1) [21]. As follows from Table 2, irrespective of the nature
of the support, the first intensity drop of the FTIR (NH2) band, in
the APMS-grafted materials, occurs at the same temperature, i.e.
473 K. It should be added that this FTIR band loses ca. 61–63% of
the initial intensity after evacuation at 573 K, also irrespective of
the nature of the support. Moreover, evacuation at 673 K causes
almost total disappearance of the –NH2 IR band for all the samples
described in Table 2. However, it does not mean that amine has
undergone total decomposition and has been removed after such
treatment, because the chemical analysis of these samples indicates
the presence of ca. 20% of the initial amount of nitrogen.
3.3. Surface characterization
FTIR spectra were measured for identification of functional
groups in the amino-modified materials evacuated at room tem-
perature (Fig. 3). The band at 1596 cm−1 assigned to the NH2
vibration dominates in the spectra of amino-grafted materials sug-
gesting that methoxy species are involved in APMS grafting on
the surface of the supports. Interestingly, the presence of methoxy
species is observed in the IR spectra of 2APMS and 3APMS/AlMCM-
41 (1455 cm−1) [22,23] indicating that not all methoxy species
are involved in the chemical interactions with surface hydroxyls
leading to amino-grafting. On the contrary, in grafting of APMS
most methoxy species take part in the chemical bonding with
the surface. It should be emphasised that in 2APMS- and 3APMS-
grafted AlMCM-41, the hydrogen bonded species were observed in
hydroxyl region of IR spectra. Hydrogen bonding between amine
groups of 3-aminopropyl(trimethoxysilane) and surface hydrox-
yls had been already suggested by Das and Sayari [8]. Our results
confirmed this suggestion.
Table 2
Stability of aminoalkoxysilanes – FTIR study.
Catalysts
Start of amine
decomposition
from IR, temp. (K)
Disappearance of
amine IR band,
temp. (K)
3.4. 2-Propanol decomposition
APMS/AlMCM-41
2APMS/AlMCM-41
3APMS/AlMCM-41
APMS/NbMCM-41
473
573
573
473
673
673
673
673
The reaction of 2-propanol decomposition has been used as a
test for basic–acidic properties of the samples [24,25]. It allows esti-
mation of the nature of acidic and basic centres and the strength

122
D. Blasco-Jiménez et al. / Catalysis Today 152 (2010) 119–125
Fig. 2. DTG curves of APMS/AlMCM-41 and 3APMS/AlMCM-41 catalysts.
Table 3
This order indicates a significant influence of the nature of the sup-
port on the basicity.
Considering the influence of the nature of the amine grafted on
the same support (AlMCM-41) one can find the following order of
basic activity (deduced from acetone selectivity):
The results of 2-propanol decomposition.
Catalysts
2-PrOH
conv. (%)
Temp. (K)
Acetone
selectivity (%)
Ref.
APMS/MCM-41
0.3
0.3
0.3
0.3
0.3
0.3
573
423
423
423
423
423
8
35
43
51
15
16
[18]
APMS/AlMCM-41
2APMS/AlMCM-41
3APMS/AlMCM-41
APMS/NbMCM-41
2APMS/NbMCM-41
This work
This work
This work
[18]
3APMS/AlMCM-41 > 2APMS/AlMCM-41 > APMS/AlMCM-41
It is important to add that the pristine supports revealed acidic
properties changed as follows:
This work
HAlMCM-41 > HNbMCM-41 > HMCM-41
of the basic centres from the products formed. Both, Brønsted and
Lewis, acid sites activate the reaction towards propene formation,
whereas basic centres are active in dehydrogenation to acetone.
The formation of diisopropyl ether indicates the presence of Lewis
acid–base pairs. The acidic route is much faster than the basic one.
However, the selectivity to acetone for the same 2-propanol con-
version can give a general insight into basicity (strong Lewis basic
sites) of the materials. The results collected in Table 3 include those
published earlier for silicate MCM-41 based materials. They indi-
cate the following sequence of basicity of APMS-grafted MCM-41
materials:
3.5. Knoevenagel reactions
Liquid phase Knoevenagel test reactions for Lewis basicity of
various strengths were applied for characterization of the materi-
als prepared [26–30]. The test reactions met the green chemistry
requirement, it means were performed without any solvent. The
condensation of benzaldehyde and ethylcyanoacetate and diethyl
malonate was carried out on the supports and amino-grafted mate-
rials. In the first reaction activated by medium basic sites (pKa = 9)
the only reaction product was Knoevenagel one, whereas in the
reaction with diethyl malonate requiring stronger basic centres
(pKa = 13.3) also side-reaction products were formed.
APMS/AlMCM-41 > APMS/NbMCM-41 > APMS/MCM-41
Fig. 3. FTIR spectra of the catalyst after evacuation at RT.

D. Blasco-Jiménez et al. / Catalysis Today 152 (2010) 119–125
123
malonate the selectivity to Knoevenagel product changes in the
same order (Fig. 5B). It should be pointed out that the acidity of
mesoporous support decreases in the same sequence.
The role of the nature of amine grafted on AlMCM-41 (the best
selected support) can be estimated from the results presented in
Fig. 5A. Taking into account the initial step of the reaction, the
activity changes in the following order:
APMS/AlMCM-41 > 2APMS/AlMCM-41 > 3APMS/AlMCM-41
Similar results were presented by Das and Sayari [8] for amino-
grafted silicate mesoporous materials. Interestingly, this sequence
of activity is in contrast to that noted for 2-propanol decomposi-
tion. The difference in these results can be explained by the difficult
access to –NH groups in 2APMS and 3APMS modified samples
for bulky molecules used in the Knoevenagel reaction and by the
decrease in the basicity strength of NH2 groups in the same order
as proposed earlier [8].
Fig. 4. Condensation of benzaldehyde and ethylcyanoacetate at 393 K using MCM-
41, AlMCM-41 and NbMCM-41 catalysts (before and after APMS modification).
4. Discussion
In liquid phase-solid catalysed reactions special care should be
taken to avoid the reaction to be controlled by diffusion of the reac-
tants. To check this, the reactions were carried out at 393 K, on
stirring the solution at 1000–3500 rpm. The conversion remained
practically the same, which indicated that, under our experimen-
tal conditions, the reaction was not controlled by external diffusion.
The reactions were repeated under the same conditions for the cat-
alysts with particle size from three different ranges (0.074, 0.140,
0.250 mm in diameter), at 1500 rpm. The same conversions were
obtained in all cases, irrespective of the catalysts particle size.
Therefore, we can assume that if a catalyst with a particle size in
the range studied is used, the results will not be affected by the so-
called internal diffusion. Since both external and internal diffusion
did not occur, further studies were performed during the reaction.
The reaction curves for APMS-grafted MCM-41 materials are
presented in Figs. 4 and 5. It is clearly evidenced that amino-grafting
significantly increases benzaldehyde conversion for both Knoeve-
nagel reactions. The order of activity is as follows:
The novelty of our study in comparison to the already published
results [3,6,8,13,16–18] was the use of metallosilicate mesoporous
materials of different chemical compositions and consequently, dif-
ferent acidities, as supports for three aminopropylalkoxysilanes
(APMS, 2APMS and 3APMS). The differences in these three amines
APMS/AlMCM-41 > APMS/NbMCM-41 ꢀ APMS/MCM-41
and is in line with the activity sequence observed for 2-propanol
decomposition. For the condensation of benzaldehyde with diethyl
Fig. 6. Models of grafted aminoalkoxysilanes.
Fig. 5. Condensation of benzaldehyde and diethyl malonate at 393 K using MCM-41, AlMCM-41 and NbMCM-41 catalysts (before and after APMS modification); conversion
(A) and selectivity (B).

124
D. Blasco-Jiménez et al. / Catalysis Today 152 (2010) 119–125
Fig. 7. Possible reaction routes for the Knoevenagel condensation between benzaldehyde and malonic esters.
are in the number of nitrogen atoms in molecules and the alkyl
chain length. The acidity of the support is higher for AlMCM-41 than
for NbMCM-41 [19,31]. The results of XRD and nitrogen adsorption
study indicated that the textural stability and ordering of TMCM-
41 (T = Al or Nb) materials during the aminopropylmethoxysilane
treatment only slightly depends on the chemical composition of
the material. The main point of our interest was the effect of the
type of support on the thermal stability of amines and the activity
in base-catalysed reactions.
basic centres can easily cause the switching of the Knoeve-
nagel reaction route. It should be noted that the amine group
facilitates the hydrolytic degradation of both silica framework
and organic chain linkage [35–38]. Thus, it is expected that the
protonation of amine group renders the hybrid material more
stable.
The results presented in this paper show that the increased
acidity of the support makes the Knoevenagel reaction more effi-
cient. As the acidity does not influence the stability of amines
in mesoporous materials one has to consider the participation of
the support in the Knoevenagel reaction. Considering the possible
mechanism presented in Fig. 7, one can find that the formation
of imine salt (the reaction route via activation of benzaldehyde)
requires the abstraction of OH− group from the intermediate.
The acidity of the support makes this process faster and as a
consequence, enhances the total rate of the Knoevenagel con-
densation. The iminium reaction path was proposed by Katz et
al. [39] for Henry reaction on silica gel modified with amines
where primary amines/silanol groups activated hydroxybenzalde-
hyde. In the same reaction carried out on amine modified MCM-41
Asefa et al. [33] proposed the cooperative catalytic mechanism in
which the formation of iminium intermediates occurs by the acti-
vation of benzaldehyde by silanols or secondary amines. In our
paper the role of acidity of the support in iminium pathway is
stressed.
The role of the acidity of the support can be also considered
taking into account the APMS loading. Amine loading is lower
for APMS/AlMCM-41 than APMS/NbMCM-41 and therefore in the
first case a higher surface area of the support is accessible for the
reagents and moreover, APMS molecules are more isolated. It is in
line with the recent paper [33] where the authors indicated that
there is an optimum number of amine grafted groups for the effec-
tive activation of benzaldehyde. Too high concentration of amines
can decrease the activation of reagent.
The FTIR and TGA results (Fig. 2, Table 2) indicate that the sta-
bility of amines does not depend on the nature of the support (i.e.
chemical composition/acidity of aluminosilicate, niobosilicate) for
the amino-grafted samples. These results suggest that the type of
bonding of the amines with the surface of all mesoporous materials
is similar. The results of FTIR study (Fig. 3) show that the grafting
of amines occurs via the interaction between silanol OH groups
from the support and alkoxy groups from APMS. For 2APMS and
3APMS only one alkoxy group seems to be involved in this chemi-
cal interaction and therefore alkoxy species are well detected in the
IR spectra. Analysis of these results taking into regard the presence
of hydrogen bonded species in 2APMS- and 3APMS-grafted materi-
als allowed us to propose the model of grafted aminoalkoxysilanes
shown in Fig. 6. The type of amine bonding proposed in this figure,
explains the higher thermal stability of 2APMS and 3APMS. More-
over, the hydrogen bonding between the surface silanol groups
and NH2 decreases the basicity strength of the amino groups.
Therefore, the rate of condensation of benzaldehyde with diethyl
malonate (the Knoevenagel reaction which requires strong basic-
ity – pKa = 13.3) is lower for 3APMS and 2APMS than that for APMS
(Fig. 5). Moreover, the stabilization of amines, proposed in the
model, hinders the accessibility of –NH groups for large molecules
such as benzaldehyde and diethyl malonate. The molecules of 2-
propanol reach the NH groups much easier and therefore, the
activity in acetone formation is proportional to the number of
amine groups in the material.
The effect of the acidity of the support on the basic catal-
ysed Knoevenagel reaction evidenced in this work has to be
considered in detail. There are two reaction routes described
in literature for the Knoevenagel condensation between ben-
zaldehyde and malonic esters (Fig. 7) [6,8,32–34]. One reaction
pathway proceeds by the activation of benzaldehyde towards
imine (C N) salt formation and a further reaction with the
methylene group of the ester. The other possibility is the acti-
vation of the ester via abstraction of proton by basic centres
of the catalyst and a further reaction with benzaldehyde. Small
changes in the reaction medium or the surrounding of the
Finally, the catalysts were tested for recyclables during four
regeneration cycles with no appreciable change of activity.
5. Conclusions
1. FTIR study proved the way of APMS bonding to the support. The
results have revealed that aminopropylmethoxysilanes (APMS)
are grafted on metallosilicate MCM-41 supports via interactions
of methoxy species with surface hydroxyls leaving amine groups
accessible for reagents.
2. Hydrogen bonding of NH2 with surface hydroxyls has been found
for 2APMS and 3APMS grafted on mesoporous MCM-41 sam-

D. Blasco-Jiménez et al. / Catalysis Today 152 (2010) 119–125
125
ples and it is responsible for reduction of the basicity strength of
amine groups.
[8] D.D. Das, A. Sayari, Stud. Surf. Sci. Catal. 170 (2007) 1197.
[9] H.-M. Kao, Ch.-H. Liao, A. Palani, Yi-Ch. Liao, Microporous Mesoporous Mater.
113 (2008) 212.
[10] Sujandi, E.A. Prasetyanto, S.-E. Park, Appl. Catal. A: Gen. 350 (2008) 244.
[11] K. Moller, T. Bein, Chem. Mater. 10 (1998) 2950.
[12] J.H. Clark, D.J. Macquarrie, Chem. Commun. (1998) 853.
[13] A. Sayari, S. Hamoudi, Chem. Mater. 13 (2001) 3151.
[14] J. Weitkamp, M. Hunger, U. Rymsa, Microporous Mesoporous Mater. 48 (2001)
255.
[15] T. Linssen, K. Cassiers, P. Cool, E.F. Vansant, Adv. Colloid Interface Sci. 103 (2003)
121.
[16] P.J.E. Harlick, A. Sayari, Ind. Eng. Chem. Res. 45 (2006) 3248.
[17] P.J.E. Harlick, A. Sayari, Ind. Eng. Chem. Res. 46 (2007) 446.
[18] D. Blasco-Jimenez, A.J. Lopez-Peinado, R.M. Martin-Aranda, M. Ziolek, I.
Sobczak, Catal. Today 142 (2009) 283.
[19] M. Ziolek, I. Nowak, Zeolites 18 (1997) 356.
[20] X. Zhang, E. Sau Man Lai, R. Martin-Aranda, K.L. Yeung, Appl. Catal. A: Gen. 261
(2004) 109.
3. Thermal stability of amines does not depend on the chemi-
cal composition of TMCM-41, whereas the basicity tested in
2-propanol dehydrogenation and the activity in the Knoevenagel
reactions are influenced by the nature of the supports (AlMCM-
41 support gives rise to the higher activity).
4. The support has been shown to be involved in the Knoevenagel
reaction. The acidity of the supports (AlMCM-41, NbMCM-41)
increases the rate of iminium salt formation (from benzalde-
hyde and grafted amine) because acid sites can participate in
the abstraction of OH− groups from the intermediate.
5. Selectivity to the Knoevenagel products significantly depends
on the nature of the support for grafted amine (AlMCM-
41 > NbMCM-41).
[21] K. Alan, Brisdon, Inorganic Spectroscopic Method, Oxford University Press,
1998.
[22] G. Busca, A.S. Elmi, P. Forzatti, J. Phys. Chem. 91 (1987) 5263.
[23] G. Busca, J. Lamotte, J.-C. Lavalley, V. Lorenzelli, J. Am. Chem. Soc. 109 (1987)
5197.
Acknowledgements
[24] A. Gervasisni, J. Fenyvesi, A. Auroux, Catal. Lett. 43 (1997) 219.
[25] C. Lahausse, J. Bechelier, J.C. Lavalley, H. Lauron-Pernot, A.M. Le Govic, J. Mol.
Catal. 87 (1994) 329.
[26] A. Corma, R.M. Martín-Aranda, V. Fornés, J. Primo, H. García, Appl. Catal. 59
(1990) 237.
[27] A. Corma, V. Fornés, R.M. Martín-Aranda, F. Rey, J. Catal. 134 (1992) 58.
[28] I. Rodriguez, H. Cambon, D. Brunel, M. Lasperas, J. Mol. Catal. A: Chem. 130
(1998) 195.
[29] J. Lopez-Gonzalez, A. Lopez-Peinado, R.M. Martin-Aranda, M.L. Rojas-
Cervantes, Carbon 31 (1993) 1231.
D.B.J. thanks the Socrates Program of EU for the possibility
of doing research work at the A. Mickiewicz University (Poz-
nan), and the Universidad Nacional de Educación a Distancia
(UNED) for providing her with PhD fellowship. COST action D36,
WG No. D36/0006/06 and the Polish Ministry of Science (Grant
No. 118/COS/2007/03) are acknowledged for the financial sup-
port.
[30] M.J. Climent, A. Corma, S. Iborra, K. Epping, A. Velty, J. Catal. 225 (2004) 316.
[31] M. Ziolek, I. Nowak, J.C. Lavalley, Catal. Lett. 45 (1997) 259.
[32] L. Kürti, B. Czako, Strategic Applications of Named Reactions in Organic Syn-
thesis, Elsevier, 2005.
[33] K.K. Sharma, R.P. Buckley, T. Asefa, Langmuir 24 (2008) 14306.
[34] K.K. Sharma, T. Asefa, Angew. Chem. Int. Ed. 46 (2007) 2879.
[35] R.J.P. Corriu, G. Dabosi, M. Martineau, J. Organometal. Chem. 150 (1978)
27.
[36] R.J.P. Corriu, G. Guérin, J. Organometal. Chem. 225 (1982) 141.
[37] J.P. Blitz, R.S. Sheedhara Murthy, D.E. Leyden, J. Colloid Interface Sci. 126 (1988)
387.
[38] M. Etienne, A. Walcarius, Talanta 59 (2003) 1173.
[39] J.D. Bass, A. Solovyov, A.J. Pascall, A. Katz, J. Am. Chem. Soc. 128 (2006)
3737.
References
[1] A. Corma, M. Iglesias, C. Del Pino, F. Sanchez, J. Chem. Soc., Chem. Commun.
(1991) 1253.
[2] D.J. Macquarrie, Chem. Commun. (1996) 1961.
[3] D.J. Macquarrie, D.B. Jackson, Chem. Commun. (1997) 1781.
[4] Y.V.S. Rao, D.E. De Vos, P.A. Jacobs, Angew. Chem. Int. Ed. Engl. 36 (1997) 2661.
[5] J.S. Beck, J.C. Vartuli, W.J. Roth, M.E. Leonowicz, D.T. Kresge, K.D. Schmitt, C.T.W.
Chu, D.H. Olson, E.W. Sheppard, S.B. McCullen, J.B. Higgins, J.L. Schlenker, J. Am.
Chem. Soc. 114 (1992) 10834.
[6] M. Lasperas, T. Llorett, L. Chaves, I. Rodriguez, A. Cauvel, D. Brunel, Stud. Surf.
Sci. Catal. 108 (1997) 75.
[7] Z. Luan, J.A. Fournier, J.B. Wooten, D.E. Miser, Microporous Mesoporous Mater.
83 (2005) 150.