Amino-grafted SBA-15 material as dual acid-base catalyst for the synthesis of coumarin derivatives

Article abstract of DOI:10.1016/j.cattod.2013.10.016

We report herein an experimental and theoretical study concerning the preparation of the coumarins (2) from 2-hydroxybenzaldehydes (3) and ethyl acetoacetate (4) promoted by amino-grafted SBA-15 materials. The reaction takes places by Knoevenagel condensation between reagents, in the absence of any solvent and under mild conditions, followed by a non catalytic lactonization step. Amino-grafted SBA-15 loaded with secondary amine groups, MAP/SBA-15, was found to be the most efficient and totally recyclable catalyst as compared with its analog containing tertiary amine groups, DEAP/SBA-15 (where MAP and DEA are methyl aminopropyl and diethyl aminopropyl groups, respectively). Our theoretical analysis confirms that the steric congestion and the absence of NH protons as the catalytic active sites are determining factors for the increasing of the activation barrier and, therefore, for the lower catalytic activity and reactivity to the formation of the coumarin. The computational study herein reported demonstrates that MAP/SBA-15, a traditional basic mesoporous silica having NH protons, shows a dual acid-base catalytic behavior when used in the synthesis of coumarins via Knoevenagel condensation. Otherwise, the substitution in 2-hydroxybenzaldehydes (3) at 5-position clearly influences the acceptor and acidity capacities of the CHO and the OH groups as experimentally and theoretically demonstrated.

Full text of DOI:10.1016/j.cattod.2013.10.016

Catalysis Today 227 (2014) 215–222
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Amino-grafted SBA-15 material as dual acid–base catalyst for the
synthesis of coumarin derivatives
Nadia Aidera,b,c, Agata Smuszkiewicza, Elena Pérez-Mayorala,∗, Elena Sorianod,
Rosa M. Martín-Arandaa, Djamila Hallicheb, Saliha Menadb,c
a Departamento de Química Inorgánica y Química Técnica, Facultad de Ciencias, Universidad Nacional de Educación a Distancia, UNED,
Paseo Senda del Rey 9, E-28040 Madrid, Spain
b Laboratoire Gaz Naturel, University of Sciences and Technology Houari Boumediene (USTHB), Argel, Algeria
c Département de chimie, Faculté des Sciences (UMMTO), Tizi-ouzou, Algeria
d Instituto de Química Orgánica General (CSIC), C/Juan de la Cierva 3, 28006 Madrid, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 5 July 2013
Received in revised form 2 October 2013
Accepted 4 October 2013
Available online 31 October 2013
We report herein an experimental and theoretical study concerning the preparation of the coumarins
(2) from 2-hydroxybenzaldehydes (3) and ethyl acetoacetate (4) promoted by amino-grafted SBA-15
materials. The reaction takes places by Knoevenagel condensation between reagents, in the absence of
any solvent and under mild conditions, followed by a non catalytic lactonization step.
Amino-grafted SBA-15 loaded with secondary amine groups, MAP/SBA-15, was found to be the most
efficient and totally recyclable catalyst as compared with its analog containing tertiary amine groups,
DEAP/SBA-15 (where MAP and DEA are methyl aminopropyl and diethyl aminopropyl groups, respec-
tively). Our theoretical analysis confirms that the steric congestion and the absence of NH protons as
the catalytic active sites are determining factors for the increasing of the activation barrier and, therefore,
for the lower catalytic activity and reactivity to the formation of the coumarin.
Keywords:
Mesoporous materials
Heterogeneous catalysis
Coumarins
Knoevenagel condensation
Computational methods
The computational study herein reported demonstrates that MAP/SBA-15, a traditional basic meso-
porous silica having NH protons, shows a dual acid–base catalytic behavior when used in the synthesis
of coumarins via Knoevenagel condensation.
Otherwise, the substitution in 2-hydroxybenzaldehydes (3) at 5-position clearly influences the acceptor
and acidity capacities of the CHO and the OH groups as experimentally and theoretically demon-
strated.
© 2013 Elsevier B.V. All rights reserved.
1. Introduction
materials to adsorption of carbon dioxide. Pore volume of the
silica matrix is an important factor which determines the CO2
The research on ordered mesoporous silicas has undergone an
impressive progress since the discovery of MCM-41 by Scientifics
of Mobil Corporation in 1992 [1]. In fact the development of new
modified mesoporous silicas is an area of extensive research con-
stituting a hot issue of current interest in nanochemistry.
In general, the main and useful property of the ordered meso-
porous silicas is their large specific surface area; their walls can be
tuned by the introduction of high concentrations of active functions
which allow their application for adsorption and catalysis [2].
Mesoporous silicas have been extensively studied for capture
and separation of gases; particularly basic ordered mesoporous
materials functionalized with amine groups, by using different
synthetic methods or by impregnation [3,4], are highly selective
adsorption capability of the amine-loaded. In this sense, ordered
mesoporous amino grafted MCM-48 [5,6], pore-expanded amino-
grafted MCM-41 [7] SBA-15 [8–10] or even MCF materials [11] have
been developed and tested for CO2 capture.
Likewise, amino-containing mesoporous silicas also found
application in catalysis. Our research group possesses a wide
experience in this field involving the mesoporous silicas function-
alized with aminopropyl groups in the preparation of heterocyclic
compounds with synthetic interest; thus, amino-grafted MCM-
41 materials are able to catalyze the selective N-alkylation of
pyrazole [12] whereas amino-grafted metallosilicate MCM-41 effi-
ciently promote the isomerization of Safrole [13], the Knoevenagel
condensation [14] and the synthesis of N-alkyl imidazoles under
ultrasound activation [15].
On other hand, the knowledge of these processes faces with
the difficulty in getting accurate structural information on these
amorphous silica materials. To this end, atomistic details and ener-
getic features can be provided by computer simulation techniques
∗
Corresponding author. Tel.: +34 91 398 9047; fax: +34 91 398 6697.
E-mail addresses: eperez@ccia.uned.es (E. Pérez-Mayoral), esoriano@iqog.csic.es
(E. Soriano).
0920-5861/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.cattod.2013.10.016

216
N. Aider et al. / Catalysis Today 227 (2014) 215–222
Fig. 1. Natural products and biologically active coumarins.
of various kinds, which are becoming essential to get compre-
hensive views of the interface regions of interest on amorphous
silica [16–18]. Different approaches have addressed different top-
ics. In this sense, ab initio periodic calculations (PBC methods)
have been successfully performed to obtain insight into the reac-
tivity and other properties (for some examples, see [19–23]).
On the other hand, first principles calculations on small size
clusters have focused on the spectroscopic properties and reac-
tivity (for some examples, see [24–28]). They allow that standard
quantum mechanical methods using highly sophisticated quan-
tum chemistry computer programs can be applied to characterize
physicochemical features, energetics, reaction paths, and interme-
diates. This is a real advantage over the PBC method, because the
level of sophistication in the adopted theoretical method is only
limited by the available computing resources. However, it is well-
known that cluster studies might suffer from edge effect if not used
in embedded techniques.
Our most recent contributions in the area of modified meso-
porous silicas are addressed to the synthesis of bioactive nitrogen
heterocycles or key intermediates through condensation reactions.
Thus, we have reported the selective synthesis of quinolones cat-
alyzed by amino-grafted MCM-41 [29] and much more recently the
preparation of quinolines by using acid–base bifunctional T/MCF
materials (where T is Nb or Al) [30,31], both of them via Friedländer
condensation.
1,2-Benzopyrones known as coumarins are a family of natu-
rally occurring oxygen heterocyclic compounds widely distributed
in the vegetable kingdom [32]. Everyday humans are continuously
exposed to coumarins through the handling of plants and fruit or
vegetables ingestion. Coumarins are an important class of oxygen
heterocycles which have attracted much attention because of their
huge number of applications. This oxygen heterocycle skeleton dis-
plays a broad range of biological and pharmacological properties
such as analgesic, antimicrobial, vasodilator, anti-inflammatory,
antioxidant and anti-cancer and anti-HIV [33]. Additionally, these
compounds are considered as important and interesting scaffolds
in agrochemical and pharmaceutical industries.
Compound 1 is a natural product named coumarin often used
in the perfume and cosmetic formulations mainly due to its strong
fragrant odor whereas Novobiocin is a natural antibiotic. Coumarin
2a is a synthetic compound useful as building block for the syn-
thesis of 3(substituted)chromen-2-ones, which exhibit antioxidant
and anti-inflammatory activities [34].
One of the simplest synthetic approaches to prepare coumarins
consists of the condensation of 2-hydroxybenzaldehydes with
esters showing enolizable hydrogens at ␣-position through Kno-
evenagel condensation, the most useful C C bond forming strategy
in organic synthesis [35]. Although coumarins are widely studied,
only a few heterogeneous catalysts showing acid or basic prop-
erties have been developed for their synthesis, via Knoevenagel
condensation, including heteropolyacids [36], hydrotalcites [37],
polystyrene-supported bases [38] and grafted quaternary organic
ammonium hydroxides MCM-41 [39].
Our research interest is centered on the design and development
of new heterogeneous basic catalysts to promote the synthesis of
coumarins from o-salicylaldehyde (3) and ethyl acetoacetate (4)
(Scheme 1). In this paper we report the synthesis and characteri-
zation of two different amino-grafted mesoporous materials based
on SBA-15 structure, as well as their use as catalysts in the syn-
thesis of coumarins under solvent-free and mild conditions. We
also studied the influence of the substitution at 5-position over the
2-hydroxybenzaldehydes (3) in the condensation reaction. Overall
the goal of this paper is the understanding of the reaction mech-
anism by combining the experimental and computational results
herein reported.
2. Materials and methods
2.1. Synthesis of the catalysts
SBA-15 matrix was synthesized according to the methodology
previously reported by Zukal et al. [40].
Amino-grafted SBA-15 materials were synthesized according to
the experimental procedure reported in Refs. [41,42] from meso-
porous silica SBA-15 and the corresponding trialkoxysilylpropy-
lamine. In a typical experiment, to a suspension of dried SBA-15
Fig. 1 depicts some natural products and synthetic compounds
containing the coumarin skeleton.
Scheme 1. Synthesis of coumarins from 5-substituted 2-hydroxybenzaldehydes (3) and ethyl acetoacetate (4).

N. Aider et al. / Catalysis Today 227 (2014) 215–222
217
Table 1
acetoacetate (4) (2 or 4 mmol), at 323 K, and the reaction mix-
ture was stirred during the appropriated time. After cooling, AcOEt
(2 mL) was added to the reaction crude and the catalyst was filtered
off and the solvent evaporated in vacuo.
Reactions were followed by TLC using mixtures of hexane/AcOEt
as eluents and the products were characterized by 1H NMR.
Conversion to coumarins 2 was determined by 1H NMR.
Textural characteristics and chemical composition for mesoporous silica SBA-15.
Catalyst
SBET (m2/g) Da (Å) Va (cm3/g) Cb [mmol/g] Nb [mmol/g]
SBA-15
MAP/SBA-15
DEAP/SBA-15 388
834
370
63
56
58
0.97
0.68
0.60
–
6.64
9.63
–
1.53
1.31
SBET, BET surface area. D and V, diameter and volume of the pores, respectively.
a
Determined by BJH method.
Determined by elemental analysis.
b
2.4. Theoretical calculations
Herein, we have performed a theoretical study on the first step
of the Knoevenagel condensation, in which the coupling of reac-
tants take place, by the use of finite cluster calculations rather than
periodic calculations in order to get more reliable structural infor-
mation on the interaction between reagents and energetic values
at the catalytic sites. This methodology allows to increase the level
of calculation and also to have access to broader selection methods.
Calculations were performed with Gaussian09 [43]. All geometries
[44] were optimized by using the B3LYP functional [45] combined
with the 6-31G(d,p) basis set. The stationary points thus obtained
were characterized by means of harmonic analysis, and for all the
transition structures, the normal mode related to the imaginary
frequency corresponds to the nuclear motion along the reaction
coordinate under study. The vibrational zero-point and thermal
corrections to the thermodynamic results were taken from the fre-
quency calculations. Single-point calculations were performed on
the B3LYP geometries with M06/6-311G(2d,p) [46]. NPA charges
were computed using the natural bond orbital (NBO) method [47].
(2 g) in toluene (35 mL), 3-methylaminopropyl)trimethoxysilane
(6.65 mmol) was added and the mixture was stirred for 5 h at
room temperature (296 K). Then, toluene was filtered off and the
modified SBA-15 was washed out three times with toluene (20 mL).
Finally, the solid was dried in vacuum at room temperature.
The following notation was used for the materials under
study: MAP/SBA-15 and DEAP/SBA-15 stand for the silica SBA-15
functionalized with 3-(methylaminopropyl)trimethoxysilane and
3-(diethylamino)propyltrimethoxysilane, respectively.
2.2. Characterization of SBA-15 materials
2.2.1. Nitrogen adsorption
Textural parameters of the SBA-15 materials were determined
by nitrogen adsorption. Adsorption isotherms of nitrogen at 77 K
on molecular sieves under study were recorded using a static vol-
umetric apparatus ASAP 2020 (Micromeritics). In order to attain
a sufficient accuracy in the accumulation of the adsorption data,
this instrument is equipped with pressure transducers covering the
133 Pa and 133 kPa ranges. Before each sorption measurement the
samples were outgassed at 383–393 K overnight until the residual
pressure was lower than 0.7 Pa. Table 1 lists some characterization
data of the catalysts under study.
3. Results and discussion
3.1. Synthesis and characterization of the catalysts
We prepared two amino-functionalized mesoporous silicas
based on the SBA-15 structure with different basic properties;
MAP/SBA-15 and DEAP/SBA-15 containing secondary and tertiary
aminopropyl groups, respectively. The mesoporous silica SBA-
15, previously synthesized, was activated using a post-synthetic
approach by grafting with the corresponding trialkoxysilylpropy-
lamine [42].
2.2.2. Thermal stability
Thermal stability of the solids based on SBA-15 structure was
investigated by TG experiments using a TA Instrument SDT Q600.
2.2.3. X-ray diffraction (XRD)
The structure of SBA-15 was confirmed by X-ray powder diffrac-
tion with a Seifert C-3000 diffractometer with filtered Cu-K␣
radiation in the Bragg-Brentano geometry operating at 40 kV and
30 mA, over powder sample. X-ray powder diffraction pattern of
SBA-15 material provided clear evidence of their hexagonal struc-
ture and phase purity of sample.
Amino-grafted SBA-15 materials showed
a considerable
decreased specific surface area, SBET, regarding the SBA-15 matrix,
as expected. However, small differences on textural parameters
were observed for MAP/SBA-15 and DEAP/SBA-15; it could be
firstly attributed to the slightly lower N loading but also to the
highest C content for the DEAP/SBA-15 sample. Furthermore, the
slightly decreasing deviation of SBET and pore diameter observed
for MAP/SBA-15 could be due to the interaction of the nearest
NHMe groups by hydrogen bonding or even with the silanol
groups exhibiting a weak acidic properties [48]. This type of inter-
actions is totally inhibited in DEAP/SBA-15 because of the absence
of the NH protons.
Thermal stability of the samples was investigated by ther-
mogravimetic studies; amino-grafted mesoporous silicas reported
herein are stable in a temperature range from room temperature
to approximately 453 K.
XRD diffractogram for SBA-15 matrix showed a well-resolved
diffraction lines at low angles evidencing the hexagonal structure
of the sample.
2.2.4. Elemental analyses
Chemical composition of amino-grafted mesoporous silicas was
investigated by elemental analyses using an Elementar Analyser
Vario EL III instrument.
2.3. Catalytic performance
2.3.1. General
NMR spectra were recorded with a Bruker AVANCE DPX-300
(300 MHz for 1H). 1H chemical shifts (␦) in CDCl3 are given from
internal tetramethylsilane.
TLC chromatography was performed on DC-Aulofolien/Kieselgel
60 F245 (Merck).
All reagents and solvents were purchased from Aldrich and Alfa-
Aesar.
3.2. Catalytic performance
At first, mesoporous silicas reported above were tested in the
synthesis of coumarin 2a from 2-hydroxibenzaldehyde (3a) and
ethyl acetoacetate (4), at 323 K, under solvent-free conditions
(Scheme 1).
2.3.2. Experimental procedure
In a typical procedure, the catalyst (50 mg) was added to a
solution of the 2-hydroxibenzaldehyde (3a) (2 mmol) and ethyl

218
N. Aider et al. / Catalysis Today 227 (2014) 215–222
100
80
60
40
20
0
100
B)
A)
80
60
40
20
0
0
30
60
90
90
120
150
180
210
240
Time (min)
Time (min)
Fig. 2. Synthesis of coumarin 2a from 2-hydroxibenzaldehyde (3a) and ethyl acetoacetate (4). Reaction conditions: 2-hydroxibenzaldehyde (3a) (2 mmol) and ethyl acetoac-
etate (4) (2 mmol) and catalyst (50 mg), at 323 K. (A) Conversion to 2a determined by GC using mesytilene as internal standard. (B) Conversion to 2a determined by 1H NMR
(ꢀ) coumarin 2a and (ꢁ) compound 5.
Fig. 2A depicts a plot of conversion to 2a versus time for the
reaction catalyzed by MAP/SBA-15. We can observe that coumarin
2a was obtained in quantitative yield after 90 min of reaction time.
However, the analysis of the reaction sample by 1H NMR showed
the presence of a mixture of coumarin 2a and other reaction prod-
uct in ratio 3:2. Besides the signals corresponding to coumarin 2a,
the 1H NMR spectrum of the sample showed the signals at ı 7.56 (s),
4.30 (m) and 1.37 (t) which were assigned to CH C and the ethyl
carboxylic ester group accompanying to other proton signals in the
aromatic region and the CH3 group at 1.98 ppm.
On the other hand, it was observed the slow transformation of
that new reaction product, in the presence of the catalyst, into the
compound 2a (Fig. 2B). It is important to note that the mixture of
both products was transformed into the coumarin 2a on standing,
compound 2a being isolated then as totally pure sample.
Therefore, these results suggest that the reaction of benzalde-
hyde 3a and compound 4 takes place by Knoevenegel condensation
leading to compound 5a as intermediate product, which under-
goes a lactonization reaction giving the corresponding coumarin
2a (Scheme 2). Remarkably, while the first step of the reaction is
a catalytic process, the intramolecular esterification in compound
5a would take place in the absence of any catalyst under thermal
activation or even at room temperature.
In contrast, mesoporous silica loaded with tertiary amino
groups, DEAP/SBA-15, under the same experimental conditions,
exclusively yielded coumarin 2a in only 7% after 2 h of reaction
time. These results are in accordance with those reported by Lanari
et al. [38].
We also performed the catalytic experiment about the amount
of MAP/SBA-15 able to catalyze the reaction between compounds
3a and 4. We observed that the reaction takes place in the same
degree even in the presence of decreased amounts of the cata-
lyst (25 mg). Additionally, it was observed that mesoporous silica
MAP/SBA-15 was a totally recyclable catalyst during at least three
times without any activity loss.
In order to explain the observed reactivity, we performed a
computational study of the catalytic step of the reaction, the for-
mation of the C C bond by Knoevenagel condensation between
Chart 1. Reactant complex, Rc. Salicylaldehyde (3a) and ethyl acetoacetate (4)
are colored in pink and blue, respectively. The schematic model of the two
amino-functionalized mesoporous silicas based on the SBA-15, MAP/SBA-15 and
DEAP/SBA-15, are also depicted. (For interpretation of the references to color in
text, the reader is referred to the web version of the article.)
benzaldehyde 3a and compound 4. We used the most reduced mod-
els of the catalysts including exclusively the basic active sites as
shown in Chart 1. This model was selected as it is the minimal
structure that contains the key amine, supported on the small silica
cluster. While this does not reflect a realistic situation, the model
should allow us to get insights into the role of the amine and its
implications in these catalytic processes. The optimized structures
for this step reveal that the initial reactant complex (Rc) is formed
by the catalyst, carbonylic acceptor salicylaldehyde (3a) and ethyl
acetoacetate (4) where the molecules interact each other through
intermolecular H-bonds (Chart 1) [49].
The transition structure (TS) computed for the MAP/SBA-15 cat-
˚
alyzed step (Fig. 3) indicates that the C C bond is forming (2.07 A),
whereas the amine proton is being transferred to the aldehyde
˚
(1.34 A) thus assisting the incipient formation of alcohol. Addition-
ally, the enol proton is mostly transferred to the catalyst, as the
˚
distance suggests (1.98 A). The transition structure drives to the
aldol intermediate (Int) where the C C is fully formed and the aldol
Scheme 2. Synthesis of coumarin 2a.

N. Aider et al. / Catalysis Today 227 (2014) 215–222
219
Fig. 3. Optimized structures along the C C bond forming step catalyzed by a model of MAP/SBA-15. Free-energy values relative to Rc are shown in kcal/mol.
is doubly H-bonded to the amine group, which acts as an H-donor
and H-acceptor (Fig. 3).
proton of the enolic reagent is partially transferred to the amine
and then to the incipient alcohol, to reach the transition structure
(as IRC calculations verify).
The computed activation energy values (difference of
11.9 kcal/mol between the two cases) are in accordance with
the different reactivity and catalytic activity observed with both
basic mesoporous solids.
As can be observed in Fig. 3, the H-bonding interactions between
both the phenolic and the carboxylic group in the aldol intermedi-
ate (Int) could represent the stabilization force which influence the
exclusive formation of the intermediate compound 5a by dehydra-
tion.
A comparison between the C C bond forming step catalyzed
by MAP/SBA-15 and DEAP/SBA-15 provides significant differences.
The transition structure computed for the MAP/SBA-15 catalyzed
step (Fig. 4 left) indicates that the C C bond is forming, whereas
These results suggest that the amine group in MAP/SBA-15 acts
as a dual acid–base catalyst and assists the approaching of the
reagents to each other, strongly stabilizing the transition structure.
In addition, we have studied the influence of the substitution
in the 2-hydroxybenzaldehydes (3) at 5-position in the synthe-
sis of coumarins 2. In this sense, we compared the results for the
synthesis of coumarins 2 starting from salicylaldehydes 3b (R = Cl)
and 3c (R = OMe), with substituents exhibiting different electronic
properties, catalyzed by MAP/SBA-15 under the same experimental
conditions above mentioned. The conversion to the corresponding
coumarin 2b or 2c and the intermediate products 5b or 5c is shown
in Fig. 5.
˚
the ammonium proton is transferred to the aldehyde (1.34 A) thus
assisting the incipient formation of alcohol, as has been described
above. Additionally, the enol has mostly transferred the proton to
˚
the amine, as the distance in TS suggests (1.98 A) thus regenerating
the catalyst and the H-bonded carbonylic group in the subsequent
aldol intermediate (Int). In the case of the catalyst bearing a ter-
tiary amine group (DEAP/SBA-15, Fig. 4 right), the role of the amine
acting as a bifunctional acid–base catalyst is not possible, and the
Fig. 4. Transition structure for the C C bond forming step catalyzed by MAP/SBA-15 (left) and DEAP/SBA-15 (right). Some relevant distances are shown in Å.

220
N. Aider et al. / Catalysis Today 227 (2014) 215–222
30 min
240 min
100
50
0
100
50
0
-OMe
-Cl
-OMe
-Cl
Fig. 5. Synthesis of coumarins 2b, c from salicylaldehydes 3b, c and ethyl acetoacetate (4); coumarins 2 are shown in blue and compounds 5 are in red. Reaction conditions:
2-hydroxybenzaldehydes (3b or 3c) (2 mmol) and ethyl acetoacetate (4) (2 mmol) and catalyst (50 mg), at 323 K. (For interpretation of the references to color in text, the
reader is referred to the web version of the article.)
Table 2
Free-energy barrier (in kcal/mol), NPA charge at the carbonylic carbon of 3 on the Rc complex, and relevant distances in the transition structure, TS (in Å).
Catalyst
R
ꢀG# (kcal/mol)
NPA charge at C-acceptor in Rc
H. . .O distance between 3 and 4 in TS
C C distance (Å) in TS
MAP/SBA15
H
OMe
Cl
19.7
17.0
20.6
24.1
0.4601
0.4677
0.4614
0.4594
1.639
1.661
1.635
1.592
2.069
2.051
2.029
2.004
NO2
From Fig. 5, we can observe that in both cases the conversion
to products is almost total after only 30 min of reaction time,
selectivity to coumarin 2c (R = OMe, 56%) being higher than to 2b
(R = Cl, 48%). This feature is emphasized at the highest reaction
times; selectivity to 2c was 79% whereas that for 2b was signif-
icantly lowers (67%). Then, these results clearly indicate that the
presence of substituents with different electronic properties at
5-position have effects not only on the acceptor capability of CHO
group but also over the acidity or nucleophility of OH group in
benzaldehydes 3b-c.
We also carried out the reaction of salicylaldehyde 3d possessing
a strong electro-withdrawing moiety, NO2, and ethyl acetoacetate
(4) in the presence of MAP/SBA-15 (Fig. 6). In this case, we per-
formed the condensation at higher temperature, 353 K, because of
the lower reactivity of the corresponding benzaldehyde 3d.
As depicted in Fig. 6, compound 3d is quantitatively converted
into coumarin 2d and compound 5d in 40 and 60% of chemi-
cal yield, respectively, in only 15 min of reaction time. In this
case, it was observed the progressive transformation of 5d to 2d,
which was isolated as totally pure sample after 4 h of reaction
time.
In order to address the effect of the aromatic substituents, we
have performed calculations on the 2-hydroxy-5-sustituted ben-
zaldehydes under study. The optimized transition structures are
shown in Fig. 7 and the computed barriers are summarized in
Table 2. As can be deduced from Fig. 7, the substituent does not
Fig. 6. Synthesis of coumarin 2d from benzaldehyde 3d and ethyl acetoacetate
(4). Coumarin 3d is shown in blue and compound 5d in red. Reaction conditions:
2-hydroxibenzaldehyde (3d) (2 mmol) and ethyl acetoacetate (4) (2 mmol) and cata-
lyst (50 mg), at 353 K. (For interpretation of the references to color in text, the reader
is referred to the web version of the article.)
involve further interactions (such as H-bond) with the catalyst, so
its effect is electronic in nature.
The results reveal that the C C forming bond for R = OMe
proceeds with the lowest free-energy (and enthalpy, not shown)
barrier, whereas the step is kinetically less favored when R = Cl
and NO2. In fact, the presence of this strong electro-withdrawing
group leads to the highest barrier for the formation of the aldol
as expected. The inverse order is (surprisingly) followed by the
Fig. 7. Transition structures for 2-hydroxybenzaldehydes bearing substituents with different nature at 5-position (R = OMe, left; R = Cl, middle; R = NO2, right). Some
relevant distances are shown in Å.

N. Aider et al. / Catalysis Today 227 (2014) 215–222
221
Chart 2. Resonance forms for the 5-NO2 substituted salicylaldehyde.
calculated NPA charge on the acceptor carbonylic carbon of the
initial reactant complex (Rc), as the electro-donor OMe group
increases the NPA charge (electrophilicity) of the aldehyde whereas
the electro-withdrawal group NO2 reduces it. This unexpected
effect can be attributed to a more acute effect of the 5-substituent
on the para-position (OH group), i.e., the impact of resonance forms
B and C (Chart 2) for the 5-NO2 substituted aldehyde, for instance. In
this sense, the structure bearing a 5-NO2 group shows a more acidic
OH group, as the short distance (strong H-bond) with the carbonyl
moiety of 4 on the transition structure suggests. The opposite is
true for the 5-OMe substituted structure (higher NPA charge and
longer H-bond with 4 in TS).
Supercomputación de Galicia (CESGA) for generous allocation of
computing resources.
References
[1] J.S. Beck, J.C. Vartuli, W.J. Roth, M.E. Leonowicz, C.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) 10834–110834.
[2] F. Hoffmann, M. Cornelius, J. Morell, M. Fröba, Angew. Chem. Int. Ed. 45 (2006)
3216–3251.
[3] F.-Y. Changa, K.-J. Chao, H.-H. Cheng, C.-S. Tan, Sep. Purif. Technol. 70 (2009)
87–95.
[4] S.-N. Kim, W.-J. Son, J.-S. Choi, W.-S. Ahn, Micropor. Mesopor. Mater. 115 (2008)
497–503.
Overall, the results suggest that the substituent at the 5-position
has two electronic effects: (a) on the acidity of the para-OH
group, and (b) on the eletrophilicity of the carbonyl acceptor. For
electron-withdrawal groups, the former should stabilize the reac-
tant complex (Rc), while the latter should disfavor the coupling,
hence both effects increasing the activation barrier.
[5] H.Y. Huang, R.T. Yang, D. Chinn, C.L. Munson, Ind. Eng. Chem. Res. 42 (2003)
2427–2433.
[6] S. Kim, J. Ida, V.V. Guliants, J.Y.S. Lin, J. Phys. Chem. B 109 (2005) 6287–6293.
[7] S. Loganathan, M. Tikmani, A.K. Ghoshal, Langmuir 29 (2013) 3491–3499.
[8] A. Olea, E.S. Sanz-Pérez, A. Arencibia, R. Sanz, G. Calleja, Adsorption 19 (2013)
589–600.
[9] N. Hiyoshi, K. Yogo, T. Yashima, Micropor. Mesopor. Mater. 84 (2005) 357–365.
[10] N. Hiyoshi, K. Yogo, T. Yashima, Chem. Lett. 33 (2004) 510–511.
[11] X. Feng, G. Hu, X. Hu, G. Xie, Y. Xie, J. Lu, M. Luo, Ind. Eng. Chem. Res. 52 (2013)
4221–4228.
[12] I. Matos, E. Pérez-Mayoral, E. Soriano, A. Zukal, R.M. Martín-Aranda, A.J. López-
Peinado, I. Fonseca, J. Cejka, Chem. Eng. J. 161 (2010) 377–383.
[13] I. Sobczak, M. Ziolek, E. Pérez-Mayoral, D. Blasco-Jiménez, A.J. López-Peinado,
R.M. Martín-Aranda, Catal. Today 179 (2012) 159–163.
The coupling step is weakly exergonic for the studied salicy-
laldehydes (−0.2 to −1.6 kcal/mol).
4. Conclusions
[14] D. Blasco-Jiménez, I. Sobczak, M. Ziolek, A.J. López-Peinado, R.M. Martín-
Aranda, Catal. Today 152 (2010) 119–125.
[15] D. Blasco-Jiménez, A.J. López-Peinado, R.M. Martín-Aranda, M. Ziolek, I.
Sobczak, Catal. Today 142 (2009) 283–287.
[16] A. Rimola, D. Costa, M. Sodupe, J.-F. Lambert, P. Ugliengo, Chem. Rev. 113 (2013)
4216–4313.
[17] F. Tielens, C. Gervais, J.-F. Lambert, F. Mauri, D. Costa, Chem. Mater. 20 (2008)
3336–3344.
We report on the first amino-grafted mesoporous silica loaded
with secondary amine groups, MAP/SBA-15, which efficiently cat-
alyzes the synthesis of coumarins from ethyl acetoacetate (4) and
2-hydroxybenzaldehydes (3) under solvent-free and mild condi-
tions.
The reaction is proposed to proceed through Knoevenagel
condensation between reagents and subsequent non-catalyzed lac-
tonization of the corresponding intermediate species.
The experimental and theoretical results reported herein
demonstrate that the amine group nature in amino-grafted SBA-
15 materials is a determining factor in the reaction course. In
this sense, DEAP/SBA-15 containing tertiary amine groups is a
less active catalyst for the synthesis of coumarins than its analog
loaded with methylaminopropyl groups. Besides the unfavorable
steric effects of ethyl groups in DEAP/SBA-15, the different cat-
alytic activity is mainly due to the presence of NH proton which
strongly stabilizes the transition structure for MAP/SBA-15. In this
sense, our theoretical results suggest that the secondary amine
group in MAP/SBA-15 acts as dual acid–base catalysts favoring the
approaching of the reagents to each other and then strongly stabi-
lizing the transition structure.
[18] P. Ugliengo, M. Sodupe, F. Musso, I.J. Bush, R. Orlando, R. Dovesi, Adv. Mater. 20
(2008) 4579–4583.
[19] Y.L. Zhao, S. Koppen, T. Frauenheim, J. Phys. Chem. C 115 (2011) 9521–9615.
[20] M. Nonella, S. Seeger, Chem. Phys. 378 (2010) 73–81.
[21] A. Rimola, M. Sodupe, P. Ugliengo, J. Phys. Chem. C 113 (2009) 5741–5750.
[22] T.P.M. Goumans, C.R.A. Catlow, W.A. Brown, J. Chem. Phys. 128 (2008) 134709.
[23] H. Guesmi, F. Tielens, J. Phys. Chem. C 116 (2012) 994–1001.
[24] S. Bordiga, I. Roggero, P. Ugliengo, A. Zecchina, V. Bolis, G. Artioli, R. Buzzoni, G.
Marra, F. Rivetti, G. Spano, C. Lamberti, J. Chem. Soc. Dalton Trans. 342 (2001)
3921–3929.
[25] V. Bolis, C. Busco, P. Ugliengo, J. Phys. Chem. B 110 (2006) 14849–14859.
[26] C.A. Demmelmaier, R.E. White, J.A. van Bokhoven, S.L. Scott, J. Catal. 262 (2009)
44–56.
[27] J. Handzlik, J. Ogonowski, J. Phys. Chem. C 116 (2012) 5571–5584.
[28] J. Handzlik, R. Grybos, F. Tielens, J. Phys. Chem. C 117 (2013) 8138–8149.
[29] F. Domínguez-Fernández, J. López-Sanz, E. Pérez-Mayoral, D. Bek, R.M. Martín-
Aranda, A.J. López-Peinado, J. Cejka, ChemCatChem 1 (2009) 241–243.
[30] A. Smuszkiewicz, E. Pérez-Mayoral, E. Soriano, I. Sobczak, M.
Ziolek, R.M. Martín-Aranda, A.J. López-Peinado, Catal. Today (2013),
http://dx.doi.org/10.1016/j.cattod.2013.04.034.
[31] A. Smuszkiewicz, J. López-Sanz, E. Pérez-Mayoral, E. Soriano, I. Sobczak, M.
Ziolek, R.M. Martín-Aranda, A.J. López-Peinado, J. Mol. Catal. A: Chem. 378
(2013) 38–46.
Finally, the presence of the substituents at 5-position in 2-
hydroxybenzaldehydes (3) shows a key role in the electrophilic
character of the carbonyl group and also in the acidity of phenolic
group, which accounts for the observed reactivity.
[32] M.F. Martins Borges, F.M. Fernandes Roleira, N.J. da Silva Pereira Milhazes, E.
Uriarte Villares, L. Santana Penin, Front. Med. Chem. 4 (2009) 23–85.
[33] K.M. Paramjeet, S. Dipak, D. Arti, J. Chem. Pharm. Res. 4 (2012) 822–850.
[34] B. Lakshmi Narayanan, L.A. Pradeep Rajkumar, A. Arulanandham, N. Satheesh
Babu, T. Gayathri, A. Raju, Int. J. Pharm. Sci. Res. 3 (2012) 474–478.
[35] Y.F. Han, M. Xia, Curr. Org. Chem. 14 (2010) 379–413.
[36] M.M. Heravi, S. Sadjadi, H.A. Oskooie, R.H. Shoar, F.F. Bamoharram, Catal. Com-
mun. 9 (2008) 470–474.
Acknowledgements
This work has been supported by MICINN (projects CTQ2009-
10478 and CTQ2011-27935). We are grateful to the Centro de
[37] A. Ramani, B.M. Chanda, S. Velu, S. Sivasanker, Green Chem. 1 (1999) 163–165.

222
N. Aider et al. / Catalysis Today 227 (2014) 215–222
[38] D. Lanari, R. Ballini, A. Palmieri, F. Pizzo, L. Vaccaro, Eur. J. Org. Chem. (2011)
2874–2884.
[39] I. Rodriguez, S. Iborra, F. Rey, A. Corma, Appl. Catal. A: Gen. 194–195 (2000)
241–252.
Cammi, C. Pomelli, J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski,
G.A. Voth, P. Salvador, J.J. Dannenberg, S. Dapprich, A.D. Daniels, O. Farkas, J.B.
Foresman, J.V. Ortiz, J. Cioslowski, D.J. Fox, Gaussian 09, Revision B01, Gaussian
Inc, Wallingford, CT, 2010.
[40] A. Zukal, H. Siklová, J. Cejka, Langmuir 24 (2008) 9837–9842.
[41] H. Balcar, J. Cejka, J. Sedlacek, J. Svoboda, J. Zednik, Z. Bastl, V. Bosacek, J. Vohlidal,
J. Mol. Catal. A 203 (2003) 287–298.
[44] C.Y. Legault, CYLview, 1.0b, Université de Sherbrooke, 2009.
[45] (a) A.D. Becke, J. Chem. Phys. 98 (1993) 5648–5652;
(b) C. Lee, W. Yang, R.G. Parr, Phys. Rev. B 37 (1988) 785–789.
[42] J. Lopez-Sanz, E. Pérez-Mayoral, E. Soriano, M. Sturm, R.M. Martín-Aranda, A.J.
Lopez-Peinado, J. Cejka, Catal. Today 187 (2012) 97–103.
[46] Y. Zhao, D.G. Truhlar, Theor. Chem. Acc. 120 (2008) 215–241.
[47] E.D. Glendening, A.E. Reed, J.E. Carpenter, F. Weinhold, NBO Version 3.1.
[48] M.W. McKittrick, C.W. Jones, Chem. Mater. 15 (2003) 1132–1139.
[49] Several plausible reactant complexes in which the catalytic amine and reac-
tants were H-bonded to each other, were studied. This assumption was made
on the base of chemical intuition, in turn supported by the computed elec-
trostatic potential maps on the reagents and amine. Additionally, the model
selected showed the less congested structure, thus allowing the achievement
of a more stable transition structure than from alternative crowded reactant
complexes, without severe steric strain.
[43] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,
G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M.
Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y.
Honda, O. Kitao, H. Nakai, T. Vreven, J.A. Montgomery Jr., J.E. Peralta, F. Ogliaro,
M. Bearpark, Heyd J.J., E. Brothers, K.N. Kudin, V.N. Staroverov, T. Keith, R.
Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J.
Tomasi, M. Cossi, N. Rega, J.M. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken,
C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R.