Immobilized proton sponge on inorganic carriers: The synergic effect of the support on catalytic activity

Article abstract of DOI:10.1016/S0021-9517(02)93692-4

1,8-bis(dimethylaminonaphthalene) (DMAN), a proton sponge, was grafted onto amorphous and pure-silica MCM-41. The results show that DMAN supported on MCM-41 is an excellent base catalyst for the Knoevenagel condensation between benzaldehyde and different active methylene compounds as well as for the Claisen-Schmidt condensation of benzaldehyde and 2′-hydroxyacetophenone to chalcones and flavanones. The influence of the solvent and the polarity of the support on the activity of the catalysts was studied. It was found that the activity of the supported catalyst is directly related to the polarity of the inorganic support. Moreover, the support can also preactivate the reagents by interaction of the carbonyl groups with the weakly acidic silanol groups of the MCM-41. This preactivation step enables DMAN, anchored onto MCM-41, to abstract protons with a higher pK_a than that of the DMAN. Finally, the study of the deactivation and recycling of the DMAN anchored onto MCM-41 catalysts showed that it is a stable material with respect to deactivation and leaching and results in comparable rates in successive batch reactions.

Full text of DOI:10.1016/S0021-9517(02)93692-4

Journal of Catalysis 211, 208–215 (2002)
doi:10.1006/jcat.2002.3692
Immobilized Proton Sponge on Inorganic Carriers
The Synergic Effect of the Support on Catalytic Activity
A. Corma,∗,1
∗
∗
S. Iborra, I. Rodr ´ı guez, and F. S a´ nchez†
∗
Instituto de Tecnolog ´ı a Qu ´ı mica, Universidad Polit e´ cnica de Valencia, Consejo Superior de Investigaciones Cient ´ı ficas,
Avda. de los Naranjos s/n, 46022 Valencia, Spain; and †Instituto de Qu ´ı mica Org a´ nica, Consejo Superior de
Investigaciones Cient ´ı ficas, Juan de la Cierva 3, 28006 Madrid, Spain
Received March 14, 2002; revised May 22, 2002; accepted May 22, 2002
chonine and cinchonidine (9); β-aminoalcohols (10); 1,5,7-
1
,8-bis(dimethylaminonaphthalene) (DMAN), a proton sponge, triazabicyclo[4.4.0]dec-5-ene (11, 12); and quaternary or-
was grafted onto amorphous and pure-silica MCM-41. The re- ganic ammonium hydroxides (13, 14). The modified meso-
sults show that DMAN supported on MCM-41 is an excellent
base catalyst for the Knoevenagel condensation between ben-
porous materials are found to be excellent catalysts for
Knoevenagel (4–8, 13, 14) and aldol codensation (8, 11, 14),
zaldehyde and different active methylene compounds as well as
Micha e¨ l addition (11, 13, 14), alkylation (10), transesterifi-
ꢀ
for the Claisen–Schmidt condensation of benzaldehyde and 2 -
cation (12), and α,β-unsaturated ketone epoxidation (11)
hydroxyacetophenone to chalcones and flavanones. The influence
reactions.
of the solvent and the polarity of the support on the activity of
Diamines with neighboring atoms and aromatic frames
suchasfornaphthalene, fluorene, andphenanthreneexhibit
unusually high basicity constants and are referred to as
the catalysts was studied. It was found that the activity of the
supported catalyst is directly related to the polarity of the in-
organic support. Moreover, the support can also preactivate the
reagents by interaction of the carbonyl groups with the weakly proton sponges, the archetype of which is 1,8-bis(dimethyl-
acidic silanol groups of the MCM-41. This preactivation step en- amino)naphthalene (DMAN) with pKa = 12.1. Despite the
ables DMAN, anchored onto MCM-41, to abstract protons with a interest in proton sponges as potential base catalysts and
a
higher pK than that of the DMAN. Finally, the study of the deac-
the fact that they are widely used for proton abstraction
15), reports dealing with their application as catalysts for
tivation and recycling of the DMAN anchored onto MCM-41 cata-
lysts showed that it is a stable material with respect to deactivation
and leaching and results in comparable rates in successive batch
reactions. ꢁc 2002 Elsevier Science (USA)
Key Words: immobilized proton sponge; base catalyst; MCM-41;
Knoevenagel condensation; Claisen–Schmidt condensation.
(
the preparation of fine chemicals were published only re-
cently. Thus, we were able to study the behavior of DMAN
as a homogeneous catalyst for a series of Knoevenagel
condensation reactions to form carbon–carbon bonds (16).
To benefit from the catalytic properties of this organic
base and to obtain a solid recyclable catalyst, we present
the heterogenization of DMAN on an amorphous silica as
well as on pure-silica MCM-41 material. To do so, the aro-
matic rings were functionalized and grafted through the
silanol groups of the molecular sieve. This procedure ob-
tained stable basic catalysts which show good activity for
the Knoevenagel condensation of benzaldehyde with ac-
tive methylene compounds and Claisen–Schmidt synthesis
of chalcones and flavanones.
INTRODUCTION
The development of green chemistry for the production
of fine chemicals is an area of growing interest. In the case
of base catalysis, the replacement of liquid by solid bases
has the advantage of decreasing corrosion and environmen-
tal problems, while allowing the separation and recovery of
the catalyst. Mesoporous molecular sieves provide an ex-
cellent inorganic support, with the aim of designing cata-
lysts with isolated and uniform basic sites, for anchoring
organic moieties containing basic sites. To this end, solid
Lewis bases were generated by grafting simple amines onto
micelle-templated silicas supports (1–8); chiral amines, cin-
EXPERIMENTAL
Catalyst Preparation
The general procedure for preparing the catalysts is
presented in Scheme 1. More specifically, in a first
step, 4-nitro-1,8-bis(dimethylamino)naphthalene (17) was
1
To whom correspondence should be addressed. Fax: 34(96)3877809.
E-mail: acorma@itq.upv.es.
prepared as follows. A mixture of HNO (0.64 ml, 0.01 mol;
3
208
0
ꢁ
021-9517/02 $35.00
c 2002 Elsevier Science (USA)
All rights reserved.

IMMOBILIZED PROTON SPONGE
209
SCHEME 1
d = 1.41 g cm−3) and concentrated H2SO4 (2 ml), pre- 6.78 (d, 1H, H-2 arom.); 6.11 (s, 1H, Ar NH); 4.64 (t, J =
cooled at 253 K was added dropwise to a stirred solu- 5.80 Hz, NH–CH2); 3.66 (q, J = 7.05 Hz, 6H, OCH2);
tion of DMAN (2.14 g, 0.01 mol) for 2 min. After an ad- 3.14–3.07 (m, 2H, CH2N); 2.74, 2.73 (s+s, 6H+6H, NCH3);
ditional 5 min at that temperature, the reaction mixture 1.19–1.12 (m, 2H, CH2); 1.08 (t, 9H, CH3–C); 0.45 (m, 2H,
was poured onto crushed ice (≈200 ml), neutralized with CH2–Si) ppm.
aqueous ammonium hydroxide (≈60 ml), and extracted
Heterogenization of the Functionalized Proton Sponge
with chloroform (5 × 25 ml). The combined organic extracts
on Silica and MCM-41
weredriedovermagnesiumsulfateandconcentratedinvac-
cum. The residue was chromatographed on silica gel using
General method. A solution of 4-[(triethoxysilyl)
ethyl acetate–hexane (1 : 2) as the eluent to obtain 1.8 g propylaminocarbonylamino]-1,8-bis(dimethylamino)nap-
◦
(
70%, m.p. 134–135 C).
hthalene in methylene chloride (4 ml, 1 mmol) was added
The nitro group was reduced to the corresponding amino to a well-stirred toluene suspension (40 ml) of 1 g of
group according to the following procedure. Ten wt% Pd either amorphous silica (silica gel Merck, 230–240 mesh,
2
−1
on a carbon catalyst (100 mg) was added to a solution 550 m g ) or pure-silica MCM-41 prepared according
of 4-nitro-1,8-bis(dimethylamino)naphthalene (260 mg, to the literature (19), which was previously dried in situ
1
mmol) in ethanol (75 ml), and H2 (1 bar) was introduced at 413 K and 0.1 Torr for 3 h. The mixture was stirred at
into the autoclave; after stirring for 2 h at room temper- 333 K for 24 h. The solid was filtered and soxhlet-extracted
ature, the reaction was monitored by TLC using ethyl with methylene chloride–ethyl ether (1 : 1) for 16 to 24 h
acetate–hexane (1 : 2). When the reaction was finished, the to remove the remaining unsupported naphthalene deriva-
mixture was filtered through a short pad of celite and con- tive. The SiO2 and MCM-41 containing the grafted DMAN
centrated to dryness under reduced pressure. The 4-amino- were dried in vacuum, analyzed, and are referred to as
1
,8-bis(dimethylamino)naphthalene (18) was quickly PS-Silica and PS-MCM-41 respectively.
dissolved in dry methylene chloride (4 ml), and 3-
Surface loading, determined by elemental analyses based
isocyanatepropyltriethoxysilane (1 mmol) was added under on N wt%, using a CHNS analyser (Fisons EA 1108), was
−
1
inert atmosphere while stirring. After 2 h of reaction, 4- 0.20 and 0.31 mmol g for PS-MCM-41 and PS-Silica, re-
(triethoxysilyl)propylaminocarbonylamino]-1,8-bis(dime- spectively.
thylamino)naphthalene was quantitatively obtained.
Infrared measurements were performed on an FTIR
[
The structure of the silylated derivative was established Nicolet 710 spectrophotometer using sealed cells with CaF2
by means of NMR spectroscopic techniques, including windows. Wafers (10 mg) of solid catalyst were pressed
−
2
−2
two-dimensional correlated NMR spectroscopy (COSY (1 Tm cm ) and outgassed under vacuum (10 Pa) for
4
1
13
1
5) for assignment of the H and C signals. H-NMR 1 h at 473 K.
Thermogravimetric analyses of the catalysts were ob-
.28 (dd, JH-6-H-5 = 7.89, JH-6-H-7 = 7.74 Hz; 1H, H- tained in a Netszch microbalance under an air stream using
arom.); 7.21 (d, 1H, H-5 arom.); 6.88 (d, 1H, H-7 arom.); kaolin as the inert standard.
(
7
6
Cl3CD): 7.52 (d, JH-3-H-2 = 8.27 Hz, 1H, H-3 arom.);

2
10
CORMA ET AL.
Catalytic Activity: General Reaction Procedures
The Knoevenagel condensation. A mixture of ben-
zaldehyde (8 mmol) and an active methylene compound
(
7 mmol) was heated and stirred at the desired reaction
temperature in a batch reactor with or without solvent un-
der N2 atmosphere. When the reaction temperature was
stable, the catalyst (0.5 to 1 mmol% of anchored DMAN
with respect to the active methylene compound) was added
and the reaction started. Samples were taken periodically,
and the evolution of the reaction was followed by GLC
on a 25-m capillary column coated with 5% cross-linked
phenylmethylsilicone.
The Claisen–Schmidt condensation. A mixture of
ꢀ
benzaldehyde (12 mmol) and 2 -hydroxyacetophenone
(
10 mmol) was stirred and heated at 403 K without sol-
vent in a batch reactor under N2 atmosphere. When
the temperature was stable, the PS-MCM-41 catalyst
ꢀ
(
2 mmol% of anchored DMAN with respect to the 2 -
FIG. 1. Knoevenagel condensation of benzaldeyde with ethyl
hydroxyacetophenone) was added and the reaction started. cyanoacetate (2a) using PS-MCM-41 catalyst (0.5% molar ratio of immo-
Samples were taken periodically, and the evolution of the bilized DMAN with respect to 2a): ꢀ, at room temperature; ꢁ, at 313 K.
reaction was followed by GLC on a 25-m capillary column
coated with 5% cross-linked phenylmethylsilicone.
We previously found (16) that, when using DMAN as
a base catalyst in homogeneous phase for the condensa-
tion of benzaldehyde and ethyl cyanoacetate, the polarity
RESULTS AND DISCUSSION
of the solvent has a strong effect on the reaction rate, in-
creasing the reaction rate with the solvent polarity (Fig. 2).
Furthermore, with nonpolar solvents the reaction rate de-
creased dramatically. This behavior was attributed to the
influence of the solvent on the reaction mechanism and not
The efficiency of the grafted proton sponge as a base
catalyst was tested with the Knoevenagel condensation be-
tween benzaldehyde and ethyl cyanoacetate (Scheme 2)
in the absence of a solvent at room temperature, using a
0
4
.5% molar ratio of immobilized proton sponge on MCM-
1(PS-MCM-41)withrespecttotheethylcyanoacetate. Af-
ter 7 h ofreactiontime, the conversionofethylcyanoacetate
was 78% with 100% selectivity to the condensation prod-
uct (3a)(trans-α-ethyl-2-cyanocinnamate). However, under
these reaction conditions the product remains strongly ad-
sorbed on the catalyst, which is deduced from the small
increase in conversion observed after a 2-h reaction time
(
Fig. 1). Nevertheless, when the reaction temperature was
increased to 313 K under the same reaction conditions, 99%
product yield was obtained after a 7-h reaction time (Fig. 1).
These results show that this organic–inorganic composite
catalyst performs the Knoevenagel condensation between
2
a and benzaldehyde with excellent activity and selectivity
at relatively low reaction temperatures.
FIG. 2. Knoevenagel condensation of benzaldeyde with ethyl
cyanoacetate (2a) using DMAN (2% molar ratio with respect to 2a) as
homogeneous catalyst at room temperature and using different solvents:
ꢀ, toluene; ꢂ, chlorobenzene; ꢁ, acetonitrile; ꢃ, dimethylformamide; ꢄ,
SCHEME 2
dimethyl sulfoxide; ×, ethanol.

IMMOBILIZED PROTON SPONGE
211
toachangeinthecapacityofthecatalystforprotontransfer.
Thus, when polar reagents are involved, the transition-state
complex is better solvated by polar solvents, decreasing the
activation free enthalpy and enhancing the reaction rate
(
20, 21).
To study the influence of the solvent on the initial rate of
the condensation between benzaldehyde and 2a, the reac-
tion was carried out using solvents with different polarities
in the presence of PS-MCM-41 as a catalyst at 313 K, and
using a 0.5% molar ratio of immobilized proton sponge with
respect to the ethyl cyanoacetate (2a).
As can be seen in Fig. 3, a polar protic solvent such as
ethanolwasthemosteffectivefortheformationoftheprod-
uct 3a, in good agreement with previous observations in the
homogeneous phase (16). However, with the exception of
the behavior observed with ethanol, the other tested sol-
vents (Fig. 3) showed that a change in solvent polarity had
only a moderate effect on the reaction rate. The most polar
aprotic solvent (CH3CN) did not have the expected posi-
tive impact on the reaction rate (Fig. 3). These results are in
clear contrast with those obtained when working with the
homogeneous phase, where a dramatic decrease in the re-
action rate was observed with nonpolar solvents (see Fig. 2).
Thus, we assumed that since the presence of the solid sup-
port is the only difference between the homogeneous and
the heterogeneous DMAN, this should be a determinant
for the different catalytic behavior with respect to the sol-
vent polarity. Thus, in a first approximation we assumed
that the relatively weak dependence of the reaction rate on
the polarity of the solvent when the DMAN is supported
on MCM-41 must be related to the surface polarity of the
FIG. 4. Knoevenagel condensation of benzaldeyde with ethyl
cyanoacetate (2a) using PS-Silica catalyst (0.5% molar ratio of immo-
bilized DMAN with respect to 2a), at 313 K using different solvents: ꢀ,
toluene; ꢂ, chlorobenzene; ꢁ, acetonitrile; ×, ethanol.
support. Indeed, in heterogeneous catalysis, the reaction
occurs on the surface; when polar reagents are involved (as
in Knoevenagel condensations) the implicated transition-
state complex should be stabilized by polar surfaces, which
lead to an increase in the reaction rate.
To check the influence of the polarity of the support,
derivatized DMAN was anchored onto silica according to
the same procedure used for MCM-41, and the condensa-
tion of benzaldehyde and 2a was carried out under the same
reaction conditions. When the effect of solvent was studied,
a different behavior was observed for the reaction catalyzed
by PS-Silica and PS-MCM-41 (Fig. 4). Whereas the activity
of PS-MCM-41 is hardly dependent on the nature of the
solvent (see Fig. 3), reactions catalyzed by the amorphous
silica-based catalyst show strong dependency on the sol-
vent. The reasons for the different behavior of the two solid
catalysts are found in the different polarity of the catalyst
surface. Results published by Macquarrie and Jackson (22)
showedthat surfacepolarity ofaminopropyl-functionalized
MCM, measured using adsorption of Reichardts dye (23),
is higher than the corresponding silica-derived materials.
Therefore, it might be that in our case the trend is the same.
In fact, however, the PS-silica material was more hydropho-
bicthanPS-MCM-41, whichisdeducedbythesmallamount
of adsorbed water (Table 1) and that most of the silanol
groups of the silica were used to graft the organic base.
−
1
Indeed, the amount of free silanol groups at 3745 cm is
lower on the PS-silica material than on PS-MCM-41 (Fig. 5).
Therefore, it is assumed that in the course of the reaction,
the polar surface of the support stabilizes the transition
state. In the case of the more polar surface (PS-MCM-41),
reactants are preferentially adsorbed onto the catalyst, and
evenapolarsolventsuchasCH3CNonlymoderatelyaffects
FIG. 3. Knoevenagel condensation of benzaldeyde with ethyl
cyanoacetate (2a) using PS-MCM-41 catalyst (0.5% molar ratio of im-
mobilized DMAN with respect to 2a) at 313 K using different solvents: ꢀ,
toluene; ꢂ, chlorobenzene; ꢁ, acetonitrile; ×, ethanol.

2
12
CORMA ET AL.
TABLE 1
with the activity of the parent homogeneous DMAN, we
calculated turnover frequencies of the catalysts for reac-
tions carried out at room temperature in the absence of a
solvent. The turnover frequencies (TOF; expressed as ini-
tial reaction rate per mmol of base moieties per min) were
PS-MCM-41 (TOF = 85), DMAN (TOF = 65), and PS-
Silica (TOF = 38). Therefore, it seems that the polarity of
the support may play a role similar to that of the polar sol-
vent in the homogeneous phase, thus increasing the rate
of the reaction. As stated earlier, the solvent, and there-
fore the support effect, can be related to the stabiliza-
tion of the charged transition state, leading to the reaction
product.
Main Characteristics of the Catalysts
a
Water adsorptionb
Surface loading
−
1
Catalyst
mmol g
(%)
PS-MCM41
PS-Silica
0.20
0.31
4.9
2.4
a
Determined by elemental analyses based on N wt%.
From TG measurements.
b
the reaction rate. However, with the less polar surface of
PS-Silica, reactant molecules are preferentially solvated by
polar solvents interfering with the catalytic process.
To evaluate the intrinsic activity of both MCM-41 and
the silica-derivatized catalyst, and to compare the findings
Preactivation of Reactants by Adsorption on the Support
There is another factor besides surface polarity that can
contribute to the high catalytic activity of the DMAN
grafted onto MCM-41. This could be the participation of
the weakly acidic silanol groups of the support surface in
the reaction mechanism. Bearing this hypothesis in mind,
a concerted acid–base mechanism may operate, in which
the weakly acidic silanol groups interact with the carbonyl
group of the benzaldehyde resulting in polarization of the
group and an increase in the positive charge of the cor-
responding carbon atom. If this occurs, then the attack of
the carbanion formed on the basic sites (DMAN) should
be favored; this would explain the higher activity of the
PS-MCM-41 sample catalysts, which contain more silanol
groups. Angeletti et al. (24) and Lakshmi Kantam and
Sreekanth (25) proposed a concerted mechanism for con-
densationreactionscatalyzedbybothamino-functionalized
silica and mesoporous material, in which silanol groups of
the support promote the nucleophilic addition on the car-
bonyl compound by means of a hydrogen bond. Moreover,
these silanols would allow the regeneration of the catalytic
sites by proton abstraction.
To examine this possibility, benzaldehyde was adsorbed
on both silica and MCM-41-grafted DMAN in a vacuum
IR cell. As shown in Fig. 6, the two samples give a band at
−
1
∼
1690 cm , corresponding to the physically adsorbed ben-
−
1
zaldehyde, while a second band appears at ∼1654 cm and
must be assigned to the carbonyl groups interacting with
the weakly acid silanol centers. This indicates that a certain
polarization of the carbonyl group of benzaldehyde took
place, and this is responsible for preactivation of the ben-
zaldehyde reactant molecule. The concentration of preacti-
vated molecules of benzaldehyde is higher on the MCM-41
than on the silica composite (Fig. 6) as a consequence of the
larger number of silanol groups on the surface of the for-
mer (Fig. 5). This could lead to an increase in reaction rate
observed with the PS-MCM-41 catalyst with respect to the
PS-Silica and even more so with respect to DMAN in the
homogeneous phase especially in the presence of nonpolar
solvents.
FIG. 5. IR spectra of the PS-MCM-41 and PS-Silica samples.

IMMOBILIZED PROTON SPONGE
213
SCHEME 3
carried out at 383 K, without a solvent in the presence of
PS-MCM-41 as the catalyst (1% mmol/mmol), 70% yield
of the Knoevenagel adduct was produced with a selecti-
vity of 90% (Table 2, entry 2), while almost no conversion
was observed with DMAN in the homogeneous phase and
DMAN supported on amorphous silica.
Claisen–Schmidt Condensation
PS-MCM-41 catalyst was tested for the Claisen–
ꢀ
Schmidt condensation between benzaldehyde and 2 -
hydroxyacetophenone (Scheme 3). This condensation led
ꢀ
to 2 -hydroxychalcone which can cycle through an in-
tramolecular Michael addition to yield the corresponding
flavanone. When the reaction was carried out at 403 K with-
out solvent, but with PS-MCM-41 as the catalyst, a conver-
sion of 40% with a selectivity of 80% for flavanone was
obtained after an 8-h reaction time.
The fact that the pKa of both ethyl malonate (13.3) and
-hydroxyacetophenone (15.8) is higher than that of the
ꢀ
2
proton sponge (12.1) suggests that factors other than the
basic strength of the amine are also responsible for the cata-
lytic activity. As discussed earlier, we propose a mechanism
where a weak Brønsted acid site (SiOH group) will inter-
act with the carbonyl group of the benzaldehyde, diethyl
malonate, or acetophenone, polarizing the carbon–oxygen
bond and increasing the density of the positive charge on
FIG. 6. IR spectra of the benzaldehyde adsorbed on PS-MCM-41 and
PS-Silica samples.
The foregoing hypothesis is supported by the results ob-
tained when PS-MCM-41 was used as a catalyst and when
theseresultsarecomparedwiththoseobtainedwiththepar-
ent DMAN in the homogeneous phase for the Knoevenagel
condensation of active methylene compounds requiring in-
creasing basicity for proton abstraction, i.e., ethyl acetoac-
etate (2b) (pKa = 10.7), ethyl malonate (2c) (pKa = 13.3)
TABLE 2
Results of the Knoevenagel Condensation
Using PS-MCM-41 as a Catalyst
ꢀ
and 2 -hydroxy acetophenone (Scheme 3) (pKa = 15.8).
Thus, when the reaction between benzaldehyde and 2b was
carriedoutinEtOHat353Kundertheconditionsdescribed
earlier, a yield of 20% condensation product was reached
after 3 h on PS-MCM-41 (Table 2, entry 1), whereas the re-
action did not take place in the homogeneous phase, even
with an amount of catalyst four times larger (16).
Catalyst
(% mol)
Temperature
(K)
Yieldsa
(%)
ꢀ
Entry
Z-CH2-Z
Solvent
1
2
2b
2c
0.5
1
EtOH
—
353
383
20 (100)
70 (90)b
a
3-h reaction time. Selectivity to the condensation product in paren-
Moreover, when a more demanding reactant (diethyl-
malonate, 2c) was used and the Knoevenagel reaction was
theses.
b
Decarboxylation of the Knoevenagel adduct is also detected.

2
14
CORMA ET AL.
the carbon supporting the carbonyl group. In the case of
benzaldehyde, this polarization leads to a quicker attack of
the carbon by carbanion-type species. At the same time,
in the case of the diethyl malonate or acetophenone, the
increase in the positive charge on the carbon of the car-
bonyl group enhances the acidity of the hydrogens in the α-
position of the carbonyl group, allowing easier abstraction
by the catalyst. Calculations of the density of the positive
charge on the hydrogen atoms attached to the methylenic
group of a protonated diethyl malonate species give a value
of 0.356, clearly superior to that of the neutral species
CONCLUSIONS
A proton sponge (DMAN) was functionalized and
grafted onto amorphous silica and onto pure-silica MCM-
1. It was shown that, in the homogeneous phase, DMAN
4
requires the presence of polar solvents to reach high re-
action rates. However, supported DMAN gives better in-
trinsic activity in the absence of a solvent and when polar
solvents are not used. The activity of the supported cata-
lysts is directly related to the polarity of the support, which
is higher in the case of MCM-41. In this sense the sup-
port acts as a polar solvent that can stabilize the transition-
state charged complex. The support can also preactivate
the reagents by interaction of the carbonyl groups with the
weakly acidic silanol groups. This preactivation step allows
PS-MCM-41 catalyst to abstract protons with a higher pKa
than that of the DMAN. Our results demonstrate that the
activity of homogeneous proton-sponge catalysts could be
strongly improved by grafting them onto MCM-41. The re-
sulting organic–inorganic catalysts are stable against deac-
(
0.302). This increase in the acidity of the protonated di-
ethyl malonate could explain why the basic sites existing on
PS-MCM-41 abstract the proton and generate carbanion-
type species with molecules such as diethylmalonate and
ꢀ
2
-hydroxyacetophenone which then react with the acti-
vated benzaldehyde.
Catalyst Deactivation and Recycling
To determine whether leaching and/or deactivation of tivation and leaching.
the catalyst occurs, PS-MCM-41 was reused several times
for the Knoevenagel reaction of benzaldehyde with ethyl
cyanoacetate(2a). Thereactionswerecarriedoutinethanol
solvent at 313 K using a 1.1% molar ratio of PS-MCM-41
catalyst. When the reaction was complete, the solid was
filtered and thoroughly washed with CH2Cl2 and reused
in a second and third cycle. As shown in Fig. 7, a small
decrease in activity was observed after the first recycling,
and no further loss was observed when the small amount of
catalyst lost during the process was considered.
ACKNOWLEDGMENTS
The authors thank the Spanish CICYT for financial support (Projects
MAT2000-1392 and MAT2000-1678-C02-02), Mr. J. A. Esteban for as-
sisting with the purification and characterization of precursors, and
Dr. G. Sastre for theoretical calculations.
REFERENCES
1
2
. Brunel, D., Microporous Mater. 27, 329 (1999).
. Brunel, D., Cauvel, A., Fajula, F., and Di Renzo, F., Stud. Surf. Sci.
Catal. 97, 173 (1995).
3
4
. Cauvel, A., Brunel, D., Di Renzo, F., and Fajula, F., Am. Inst. Phys.
3
54, 477 (1996).
. Macquarrie, D. J., and Jackson, D. B., J. Chem. Soc., Chem. Commun.
81 (1997).
7
5. Cauvel, A., Renard, G., and Brunel, D., J. Org. Chem. 62, 749 (1997).
6. Macquarrie, D. J., J. Chem. Soc., Chem. Commun. 1961 (1996).
7. Lasp e´ ras, M., Lloret, T., Chaves, L., Rodriguez, I., Cauvel, A., and
Brunel, D., Stud. Surf. Sci. Catal. 108, 75 (1997).
8
9
. Choudary, B. M., Lakshmi Kantam, M. L., Sreekanth, P.,
Banopadhyay, T., Figueras, F., and Tuel, A., J. Mol. Catal. A 142, 361
(
1999).
. Iglesias-Hern a´ ndez, M., and S a´ nchez-Alonso, F., Stud. Surf. Sci. Catal.
30, 3393 (2000).
1
1
1
1
1
1
0. Bellocq, N., Brunel, D., Lasp e´ ras, M., and Moreau, P., Stud. Surf. Sci.
Catal. 108, 485 (1997).
1. Subba Rao, Y. V., De Vos, D. E., and Jacobs, P. A., Angew. Chem., Int.
Ed. Engl. 23, 36 (1997).
2. Derrien, A., Renard, G., and Brunel, D., Stud. Surf. Sci. Catal. 117, 445
(
1998).
3. Rodriguez, I., Iborra, S., Corma, A., Rey, F., and Jord a´ , J. L., Chem.
Commun. 539 (1999).
4. Rodriguez, I., Iborra, S., Rey, F., and Corma, A., Appl. Catal. A 194,
241 (2000).
FIG. 7. Knoevenagel condensation of benzaldeyde with ethyl cyano-
acetate (2a) using PS-MCM-41 catalyst (1.1% molar ratio of immobilized
DMAN with respect to 2a) at 313 K in ethanol as solvent: ꢀ, first cycle; ꢁ,
second cycle; ꢂ, third cycle.
15. (a) Saxton, R., and Wilson, L. J., J. Chem. Soc., Chem. Commun.
359 (1994); (b) Terrier, F., Halle, J. C., Pouet, M. J., and Simonnin,
M. P., J. Org. Chem. 51, 409 (1986); (c) Wille, A. E., Su, K., Carroll,
P. J., and Sneddon, L. G., J. Am. Chem. Soc. 118, 6407 (1996);

IMMOBILIZED PROTON SPONGE
215
(d) Kakiuchi, K., Nakamura, I., Matsuo, F., Nakata, M., Ogura, M.,
19. Beck, J. S., Vartulli, J. C., Roth, W. J., Leonowicz, M. E., Kresge, C. T.,
Tobe, Y., andKurosawa, H., J. Org. Chem. 60, 3318(1995);(e) Shedlow,
A. M., Carroll, P. J., and Sneddon, L. G., Organometallics 14, 4046
Schmitt, K. D., Chu, C. T.-W., Olson, D. H., Sheppard, E. W., McCullen,
S. B., Higgins, J. B., and Schlenker, J. L., J. Am. Chem. Soc. 114, 10834
(
1995); (f) Jelinek, T., Holub, J., Stibr, B., Fontaine, X. L. R., and
Kennedy, J. D., Collect. Czech. Chem. Commun. 59, 1584 (1994);
g) Holub, J., Wille, A. E., Stibr, B., Carroll, P. J., and Sneddon, L. G.,
(1992).
20. Moison, H., Texier-Boullet, F., and Foucaud, A., Tetrahedron 43, 537
(
(1987).
Inorg. Chem. 33, 4920 (1994); (h) Gaines, D. F., Bridges, A. N., and
Hayashi, R. K., Inorg. Chem. 33, 1243 (1994).
21. Rodriguez, I., Ph.D. thesis, University of Montpellier II-ENSCM, June
1995.
1
6. Rodriguez, I., Sastre, G., Corma, A., and Iborra, S., J. Catal. 183, 14 22. Macquarrie, D. J., and Jackson, D. B., J. Chem. Soc., Chem. Commun.
(1999).
1781 (1997).
1
7. 4-nitro-1,8-bis(dimethylamino)naphthalene was prepared following a
modified procedure published by Kurasov, L. A., Pozharkii, A. F.,
Kuz’menko, V. V., Klyuev, N. A., and Chernyshev, A. I., Zh. Org. Khim.
23. (a) Taverner, S. J., Macquarrie, D. J., Gray, G., Health, P. A., and Clark,
J. H., J. Chem. Soc., Chem. Commun. 1147(1997);(b)Macquarrie, D. J.,
Stewart, S. J., Gray, G. W., Heath, P. A., Rafelt, J. S., Saulzet, I., Hardy,
J. J. E., Clark, J. H., Sutra, P., Brunel, D., Di Renzo, F., and Fajula, F.,
New J. Chem. 23, 725 (1999).
24. Angeletti, E., Canepa, C., Martinetti, G., and Venturello, P., Tetrahe-
dron Lett. 29, 2261 (1988).
25. Lakshmi Kantam, M., and Sreekanth, P., Catal. Lett. 57, 227 (1999).
1
9, 590 (1983) [translated in J. Org. Chem. USSR 19, 520 (1983)].
1
8. 4-amino-1,8-bis(dimethylamino)naphthalene was prepared following
a procedure published by Ozeryanskii, V. A., and Pozharkii, A. F., Izv.
Akad. Nauk., Ser. Khim. 46, 1501 (1997) [translated in Russ. Chem.
Bull. 46, 1437 (1997)].