Increasing the basicity and catalytic activity of hydrotalcites by different synthesis procedures

Article abstract of DOI:10.1016/j.jcat.2004.04.027

Hydrotalcites have been synthesized by three different procedures: conventional precipitation-aging, aging under microwave irradiation, and sonication during the coprecipitation step. The synthesis procedure has an effect on the crystal size and textural properties of the hydrotalcite (HT) and the Al₂O₃-MgO mixed oxides formed upon calcination. HT samples prepared under sonication at 298 K are formed by dispersed and homogenous particles of 80-nm average particle size. They also produce upon calcinations the mixed oxides with the largest surface area (～300 m ²g^-1). This method of preparation increases not only the surface area but also the number of defects in the solid, leading to sites of higher basicity. This was determined by means of catalytic reactions such as Knoevenagel and aldol condensations which demand basic sites of different strengths. Hydrotalcites were regenerated from mixed oxides by hydration while giving Broensted basic sites. Samples originally prepared by sonication present smaller crystallite size and have a larger number of accessible active sites. With these samples acetone/citral condensations with 96 and 99% conversion and selectivity, respectively, are achieved in a 15-min reaction time.

Full text of DOI:10.1016/j.jcat.2004.04.027

Journal of Catalysis 225 (2004) 316–326
www.elsevier.com/locate/jcat
Increasing the basicity and catalytic activity of hydrotalcites by different
synthesis procedures
M.J. Climent, A. Corma,∗ S. Iborra, K. Epping, and A. Velty
Instituto de Tecnología Química, UPV-CSIC, Universidad Politécnica de Valencia, Avd de los Naranjos s/n, 46022 Valencia, Spain
Received 8 March 2004; revised 13 April 2004; accepted 17 April 2004
Abstract
Hydrotalcites have been synthesized by three different procedures: conventional precipitation-aging, aging under microwave irradiation,
and sonication during the coprecipitation step. The synthesis procedure has an effect on the crystal size and textural properties of the hy-
drotalcite (HT) and the Al O –MgO mixed oxides formed upon calcination. HT samples prepared under sonication at 298 K are formed by
2
3
dispersed and homogenous particles of 80-nm average particle size. They also produce upon calcinations the mixed oxides with the largest
2
−1
surface area (∼ 300 m g ). This method of preparation increases not only the surface area but also the number of defects in the solid,
leading to sites of higher basicity. This was determined by means of catalytic reactions such as Knoevenagel and aldol condensations which
demand basic sites of different strengths.
Hydrotalcites were regenerated from mixed oxides by hydration while giving Brönsted basic sites. Samples originally prepared by sonica-
tion present smaller crystallite size and have a larger number of accessible active sites. With these samples acetone/citral condensations with
96 and 99% conversion and selectivity, respectively, are achieved in a 15-min reaction time.
2004 Elsevier Inc. All rights reserved.
Keywords: Synthesis of hydrotalcites by sonication; Synthesis of hydrotalcites by microwave irradiation; Small crystal hydrotalcite; High surface area
hydrotalcite
1. Introduction
ring, followed by a long (about 1 day) ageing time and/or
hydrothermal treatment in order to improve the crystallinity.
The conditions of synthesis, i.e., temperature, pH, and metal
composition, have been extensively studied [1–5]. By con-
trolled thermal decomposition, LDHs are converted into
mixed oxides, which possess homogeneous interdispersion
of the elements, high specific surface areas, and, the most
important, strong basic properties. The chemical composi-
tion of the hydrotalcites (the nature and amount of structural
cations and compensating anions) is the main parameter that
allows fine tuning of the basic strength. Nevertheless, for a
particular chemical composition, the method of synthesis,
i.e., temperature, solution pH, and ageing time of the gels,
has a strong influence on the final basicity of the mixed ox-
ides [1,6,7].
Hydrotalcite-like compounds belong to a class of an-
ionic clay minerals also known as layered double hydrox-
ides (LDHs). They have the general molecular formula
Mx2+My3+(OH)2(x+y)Ay/n · mH2O, where M2+ and
−n
M3+ are a divalent and a trivalent metal ions, respectively,
and A−n is an intercalated anion. Structurally, they are
formed by brucite-like (Mg(OH)2) sheets where isomor-
phous substitution of Mg2+ by a trivalent cation like Al3+
occurs. The positive charge of the layer is compensated
by anions, which occupy the interlayer space along with
water molecules [1]. The Al/Mg hydroxycarbonate LDH
corresponds to the naturally occurring hydrotalcite min-
eral. Conventionally, these compounds are synthesized by
coprecipitation, wherein metal nitrates and precipitants are
added slowly and simultaneously at a fixed pH under stir-
Microwave irradiation has been applied for the rapid syn-
thesis of inorganic solids and organic synthetic reactions [8].
Pioneering work by Komarneni et al. [9–11] showed that
microwave irradiation under hydrothermal conditions leads
to very rapid synthesis of anhydrous ceramic oxides and
*
Corresponding author. Fax: +34 96 387 7809.
E-mail address: acorma@itq.upv.es (A. Corma).
0021-9517/$ – see front matter
doi:10.1016/j.jcat.2004.04.027
2004 Elsevier Inc. All rights reserved.

M.J. Climent et al. / Journal of Catalysis 225 (2004) 316–326
317
hydroxylated phases including hydrotalcites. Other authors
have reported a crystallization rate enhancement of LDHs
using microwaves [12,13]. More specifically, in the case of
Mg/Al LDHs, Fetter et al. [14] have found that microwave
irradiation of the coprecipitated gel allows well-crystallized
and pure hydrotalcite-like phases to be obtained, and the
ageing time can be reduced from 18–24 h to 2–10 min.
Moreover, these authors have found that hydrotalcites ob-
tained by this method present smaller crystallite sizes and
higher specific surface areas than conventional samples, syn-
thesized aging of the gels at 423 K for 24 h. Recently, Tichit
et al. [15] have studied the structural and acid–base prop-
erties of Mg/Al and Mg/Ga LDH obtained by microwave
irradiation of the coprecipitated gels, that were compared
with those conventionally aged by prolonged hydrothermal
treatment (18 h at room temperature). The authors found
that the use of microwaves reduces significantly the dura-
tion of the synthesis, and similar crystallinities and chemical
composition were obtained whatever the synthesis method
used. Interestingly, while the specific surface areas of the
mixed oxides obtained by calcinations at 723 K of the LDHs
conventionally prepared were higher than the microwave-
irradiated ones, microcalorimetric and spectroscopic mea-
surements upon CO2 and CH3CN adsorption showed that
in the case of Mg/Al mixed oxides, the amount of both
basic and acid sites increases when the LDHs is synthe-
sized using microwave irradiation. They conclude that the
microwave treatment probably induces higher amounts of
surface-defective sites.
modifying the number of defects in the framework of the
material, i.e., the number of oxygen atoms associated with
Mg exhibiting a low coordination number. Although the base
strength of calcined hydrotalcites has been considered to be
close to superbases [21], in numerous applications where
water is produced (for instance, condensation reactions), the
presence of water reduces the base strength to moderate lev-
els.
In this work, we have attempted the synthesis of Mg/Al
hydrotalcites, the corresponding amorphous mixed oxides,
and the rehydrated HT with higher surface areas and smaller
crystallite (particle) sizes than the conventional HT, with the
aim of increasing both number and base strength of the ac-
tive sites. In order to achieve this, we undertook the synthesis
of three series of LDHs. In two of them, the effect of the
temperature during the coprecipitation step, with and with-
out simultaneous sonication, was studied. In the third series
we investigated the influence of the temperature during the
ageing step of the gels under simultaneous microwave irradi-
ation. The effect of these different synthetic methodologies
on the structural and textural properties of the materials and
their impact on their catalytic activity for two base-catalyzed
organic reactions, i.e., Knoevenagel condensation and aldol
condensation, are presented. We have shown that by “in situ”
sonication during the synthesis, it is possible to produce
samples that, to the best of our knowledge, give the high-
est catalytic activity reported up to now.
From a catalytic point of view, the synthesis of high sur-
face area hydrotalcites, the derived mixed oxides, and their
corresponding regenerated hydrotalcites are matters of inter-
est, since this should have an impact on the total number as
well as the basic strength of the solid, and on its catalytic
activity. Keeping this in mind, several studies show that son-
ication enhances and/or alters the dissolution process and
nucleation and growth of precipitates in some inorganic syn-
thesis [16–18]. Kapustin [19] examined 30 different samples
and observed that insonating the solution during crystal-
lization alters the grain size of crystals and produces fine-
structured crystals which have smaller grain sizes than the
crystals prepared without insonation. Interestingly, we have
not found any report on the application of the ultrasound irra-
diation during the synthesis of hydrotalcites; however, Seida
et al. [20] applied sonication after the synthesis of the hydro-
talcite. The authors found in this case that the LDHs present
larger crystallites and BET surface areas than conventionally
prepared samples.
Calcined and calcined rehydrated Mg–Al hydrotalcites
have found numerous applications as basic catalysts in or-
ganic reactions such as Knoevenagel, Claisen–Schmidt, and
aldol condensations, Michael additions, phenol alkylations,
or epoxidation of olefins [21]. Thus, for a calcined hydrotal-
cite with a given chemical composition, the number of basic
sites can be controlled by modifying the specific surface area
of the solid, whereas the basic strength can be controlled by
2. Experimental
2.1. Preparation of catalysts
An Mg–Al hydrotalcite with a Mg/Al ratio of 3 was
prepared from gels produced by mixing two solutions: so-
lution A containing (3 − x) mol of Mg(NO3)2 and x mol of
Al(NO3)3 in the Al + Mg concentration of 1.5 mol/L for
a ratio x/3 of 0.25, and another solution B of (6 + x) mol
of NaOH and 2 mol of Na2CO3 dissolved in the same vol-
ume of the solution A. Both solutions are coadded at a rate
of 1 mL min−1 under vigorous mechanical stirring, at room
temperature. The suspension was left for 18 h at 333 K. The
hydrotalcite was filtered and washed until pH 7, and the solid
was dried at 333 K. The HT was activated by calcining at
723 K in dry nitrogen for 6 h. This mixed oxide is called in
the text the HTcCV sample.
The first hydrotalcite series (HTT) was prepared by co-
precipitation at constant and controlled pH, by slow addition
in a single container of two diluted solutions (A and B).
Solution A contains Mg(NO3)2 and Al(NO3)3, of 1.5 M in
Al + Mg with Al/(Al + Mg) molar ratio equal to 0.25, and
solution B prepared by dissolving Na2CO3 and NaOH in wa-
ter in such way that the ratio CO32−/(Al + Mg) is equal
to 0.66. The solutions were mixed at a 60 mL h−1 addition
rate, for 4 h, with vigorous stirring. The final pH was 12–13.
The coaddition process was carried out at different tempera-

318
M.J. Climent et al. / Journal of Catalysis 225 (2004) 316–326
tures (298, 313, 333, 353, and 373 K). Finally, the precipitate
was filtered, washed to eliminate the alkali metals and the ni-
trate ions until the pH of the washing water was 7, and then
dried at 333 K for 12 h. The samples were not aged and are
2.2.2. Aldol condensation of citral with acetone
Typically, a mixture of citral (6.5 mmol), acetone (18.2
mmol), and catalyst (414 mg, 16.6% (wt/wt)) were added
in a three-necked bottom flask equipped with a condenser
system. The resultant suspension was heated up to 333 K
under vigorous stirring in an oil bath equipped with an au-
tomatic temperature control system. Samples were taken at
regular time periods and analyzed by gas chromatography
(GC) with a flame ionization detector and a Tracer wax cap-
illary column (15 m × 0.32 mm × 0.25 µm). At the end of
the reaction, after cooling, the reaction mixture was filtered
to remove the catalyst and analyzed by GC.
called HTT298, HTT313, HTT333, HTT353, and HTT373
.
The second and third series were prepared by adding the
solution containing the Mg2+ and Al3+ ions (solution A)
into one containing Na2CO3 or NaOH (solution B).
For the second series the gels obtained were aged into
a stainless-steel autoclave placed into a microwave oven
(MARS, microwave accelerated reaction system), at differ-
ent temperatures (313, 353, 393, 433, 473, and 513 K) for
1 h. After the precipitate was filtered and washed until pH 7,
the solid was dried at 333 K for 12 h. The samples were la-
beled as HTMW313, HTMW353, HTMW393, etc.
2.2.3. Knoevenagel condensation between benzaldehyde
and malononitrile
A mixture of benzaldehyde (10 mmol) and malononi-
trile (10 mmol) at 298 K, under an inert atmosphere (N2),
was magnetically stirred, and calcined hydrotalcite (2.7%
(wt/wt)) was added. Samples were taken periodically during
the reaction, within a time period of 0.1 to 2 h, and analyzed
by GC with a flame ionization detector and a HP-5 capillary
column (30 m × 0.32 mm × 0.25 µm).
The preparation of the third series was carried out as de-
scribed above, but adding an aqueous solution (A) contain-
ing the metal cations (Mg2+ and Al3+) onto a solution con-
taining the base (B) with simultaneous ultrasound irradiation
of the mixture (SIC Vibracell VCX400). This procedure was
carried out at atmospheric pressure and at different temper-
atures: 273, 283, 298, and 323 K. The gels obtained were
not aged. These samples were labeled as HTUS273, HTUS283
,
2.2.4. Knoevenagel condensation between benzaldehyde
and ethyl cyanoacetate
etc. Finally, the precipitate was filtered, washed to eliminate
the alkali metals and the nitrate ions until the pH 7, and then
dried at 333 K for 12 h.
The hydrotalcites were calcined at 723 K for 6 h in a
nitrogen flow, in order to obtain Mg–Al mixed oxide base
catalysts. Samples of the mixed oxides were rehydrated at
room temperature by direct decarbonated water (MilliQ) ad-
dition just before their use as catalysts.
Specific surface areas of the different hydrotalcites were
obtained with an ASAP 2000 (Micromeritics) using the
BET methodology. The samples were pretreated at 673 K
prior to N2 adsorption. The solids were characterized by
X-ray diffraction (XRD) with a Phillips PW diffractome-
ter using Cu-Kα radiation. Sonic energy was provided by
an ultrasonicator (SIC Vibracell VCX400) which generates
20 kHz 50 Hz ultrasound waves. Microwave irradiation
was performed in a conventional oven (300 W).
A mixture of benzaldehyde(10 mmol) and ethyl cyanoac-
etate (10 mmol) under an inert atmosphere (N2) was placed
in a flask which was immersed in a thermostat silicone oil
bath at 333 K and magnetically stirred. Then, the calcined
hydrotalcite (5.5% (wt/wt)) was added. Samples of the re-
action mixture were periodically taken with a time period
of 0.1 to 2 h and analyzed by GC with a flame ionization
detector and a HP-5 capillary column (30 m × 0.32 mm ×
0.25 µm).
In all experiments, nitrobenzene was used as internal
standard. At the end of the reaction, after cooling, the re-
action mixture was filtered to remove the catalyst and an-
alyzed. Each reaction product was identified by GC-MS
1
(Hewlett-Packard 5988 A) and by H NMR spectroscopy
(Varian VXR-400S, 400 MHz).
It was found that for particle sizes below 0.3 mm diame-
ter, the reaction is not controlled by diffusion.
A scanning electron microscope (JEOL6300) equipped
with a microanalysis system by disperse energy (Oxford In-
struments Link-ISIS) was used to obtain SEM images.
3. Results and discussion
2.2. Catalytic activity
3.1. Characterization of materials
2.2.1. Reaction procedure
The results of the XRD measurements obtained for
the three series HTT, HTMW, and HTUS are presented in
Figs. 1, 2, and 3, respectively. The XRD patterns of the pre-
pared LDH present sharp and symmetric peaks for (003),
(006), (110), and (113) planes, as well as broad symmetric
peaks for the (009), (015), and (016) planes. These peaks
are characteristics of clay minerals having layered structures
A commercial citral sample formed by a mixture of cis
and trans isomers (geranial and neral) with a proportion of
25 and 75% (wt/wt), respectively, was used without fur-
ther purification. Acetone, malononitrile, benzaldehyde, and
ethyl cyanoacetate were purchased from Aldrich and used
directly.

M.J. Climent et al. / Journal of Catalysis 225 (2004) 316–326
319
Fig. 1. HT series. Effect of the temperature during the coprecipitation step
T
Fig. 3. HT series effect of the temperature during the ageing step on the
US
on the intensities XRD reflections.
intensities XRD reflections.
Table 1
Crystallite size of the samples measured from XRD using the Debye–
Sherrer equation
Sample
Crystal size, Å
Plane (003)
Crystal size, Å
Plane (110)
Crystal size, Å
Plane (113)
HT
HT
HT
HT
HT
HT
HT
HT
HT
HT
HT
HT
HT
HT
HT
HT
200
80
79
310
127
138
165
214
246
131
255
497
510
423
338
115
122
162
314
235
103
108
119
159
213
120
209
415
592
310
235
103
120
155
282
CV
T298
T313
93
T333
120
120
85
T353
T373
MW313
MW353
MW393
MW433
MW473
MW513
US273
US283
US298
US323
152
395
403
414
431
163
179
166
390
Crystal sizes of the different synthesized hydrotalcites.
lite size of the samples was measured from the X-ray line
broadening, considering the basal reflections (003), (110),
and (113), and using the Debye–Sherrer equation. The re-
sults are summarized in Table 1. As can be seen there, an
increase of the crystal size occurs when increasing the tem-
perature, regardless of the synthesis method. This indicates
that increasing the temperature during the coprecipitation or
during the ageing step, causes the rate of nucleation to be
lower than the rate of crystal growth.
Fig. 2. HT
the intensities XRD reflections.
series. Effect of the temperature during the ageing step on
MW
[22,23], and they show that in all cases, well-crystallized
hydrotalcites are formed.
From Figs. 1, 2, and 3 we can see that the intensity of the
peaks increases when increasing the synthesis temperature,
which can be correlated with an increase of the crystallinity
of the samples. To corroborate this observation, the crystal-

320
M.J. Climent et al. / Journal of Catalysis 225 (2004) 316–326
Fig. 4. SEM images for HT
(a); HT
(b); HT
(c).
MW513
MW353
MW433
Fig. 5. SEM images of the samples HT (a); HT
(b); HT
(c).
US298
CV
T353
In order to determine the morphology and particle size
are presented and compared with the sample (HTCV) synthe-
sized following the conventional procedure. The micropho-
tographs show that HTT353 and HTCV present similar mor-
phologies, with an average particle sizes in the range of 100–
120 nm, which are similar to those displayed by the sample
HTMW353 (Fig. 4a). However, the sample prepared under
sonication at 298 K (Fig. 5c), presents an extremely homo-
geneous morphology and highly dispersed particles with an
average size of 80 nm.
distribution of the LDH of these series, we have selected dif-
ferent representative samples which where studied by scan-
ning electron microscopy (SEM). In Fig. 4, the SEM mi-
crographs of three samples pertaining to the series HTMW
,
which where aged at different temperatures in a microwave
oven, are displayed. As Fig. 4 shows, the particle size in-
creases when increasing the temperature, with average sizes
of 120, 320, and 540 nm for samples HTMW353, HTMW433
and HTMW513, respectively.
The three LDHs series were calcined at 723 K in order to
obtain the corresponding Mg–Al mixed oxides. The specific
surface areas of the calcined samples (HTc) were measured
In Fig. 5 the SEM images for samples of the series HTT
and HTUS, which were not submitted to the ageing process,

M.J. Climent et al. / Journal of Catalysis 225 (2004) 316–326
321
Table 2
Surface properties of HTc series
T
a
a
Sample
S
S
Pore volume
Micropore volume
Pore diameter
(Å)
BET
2
MIC
2
−1
−1
3
−1
3
−1
(m g
)
(m g
)
(cm g
)
(cm g
)
HTc
HTc
HTc
HTc
HTc
HTc
245
264
272
268
265
264
20
127
46
83
77
0.62
0.40
0.91
0.77
0.82
0.59
0.062
0.017
0.021
0.038
0.038
0.074
94.7
72.9
133.6
116.0
128.0
0.59
CV
T298
T313
T333
T353
T373
161
Results of adsorption/desorption of nitrogen of different calcined hydrotalcites samples (HTc series).
T
a
Microporous area calculated from the t plot, using as reference t, between 5.5 and 8.5.
Table 3
Surface properties of HTc
series
MW
a
a
Sample
S
S
Pore volume
Micropore volume
Pore diameter
(Å)
BET
MIC
2
2
−1
−1
3
−1
3
−1
(m g
)
(m g
)
(cm g
)
(cm g
)
HTc
HTc
HTc
HTc
HTc
HTc
HTc
245
270
277
251
246
246
237
20
45
56
170
176
175
162
0.62
0.27
0.61
0.49
0.41
0.39
0.363
0.062
0.023
0.024
0.080
0.085
0.087
0.082
94.7
CV
85.60
89.76
82.23
68.74
66.54
63.31
MW313
MW353
MW393
MW433
MW473
MW513
Results of adsorption and desorption of nitrogen of different calcined hydrotalcites samples (HTc
series).
MW
a
Microporous area calculated from the t plot, using as reference t, between 5.5 and 8.5.
by N2 adsorption at 77 K. The results are summarized in
Table 2 for the HTcT series, and they are compared with
those of a sample synthesized by a conventional procedure
(HTcCV). As we can observe, all samples of the HTcT series
exhibit very similar specific surface areas, indicating that in
the range studied, the variation of temperature during the co-
precipitation step has no influence on the BET surface area
of the mixed oxide, despite the fact that the crystal size of the
LDH precursor increases with increasing synthesis temper-
ature (Fig. 1 and Table 1). However, when these results are
compared with the mixed oxide from a sample obtained by
the conventional method (HTcCV), we can observe that the
HTcCV sample has a lower specific surface area and lower
porosity than the samples of the HTcT series. It should be
taken into account that, the HTcCV sample was submitted,
after coprecipitation, to an ageing time of 18 h at 333 K,
which favors the crystallization process and produces a de-
crease of the surface area of the mixed oxide obtained by
calcination.
In Table 3 the results of the BET surface areas for the
mixed oxides produced by calcination of samples of the
HTMW series are presented. In this series, the gels were aged
in a microwave oven at different temperatures for 1 h, and
one can observe that an increase of the temperature produces
a decrease in the BET area, along with an increase of the mi-
croporosity. This result can be attributed to the crystallinity
increase of the LDHs precursor with temperature (Fig. 2),
that is accompanied by an increase of the crystal size. When
these results are compared with those obtained for the series
HTcT (Table 2) and with the conventional sample (HTcCV),
we can conclude that the ageing treatment under microwave
irradiation at temperatures lower than 393 K allows mixed
oxides with higher specific surface areas to be produced.
The results presented above in Table 2, showed that dur-
ing the coprecipitation step, the temperature has no influence
on the final specific surface areas of the mixed oxides pro-
duced by calcination (HTcT series). However, different re-
sults were obtained for the calcined samples when the gels
were submitted during precipitation to ultrasounds irradia-
tion (Table 4). Although the results of Table 4 do no show
a clear tendency for the variation of the BET area with tem-
perature, it is clear that the sample which is synthesized at
298 K (HTcUS298) under sonication presents a surface area
20% larger than that of the conventionally synthesized sam-
ple and any of the series HTcT and HTcMW. It is known that
when a liquid–solid interface is submitted to ultrasounds,
acoustic cavitations [24] near the surface induce markedly
asymmetric bubble collapse, which generates a high-speed
jet of liquid directed at the surface [25]. The collision of this
jet and related shock waves can create a localized erosion
on the surface (which can produce defective sites), improve
mass transport, and can cause particle fragmentation (which
can increase the surface area) [26]. Taking into account that
the crystallization process starts since the beginning of the
coprecipitation step, these results suggest that ultrasounds,
not only accelerate the crystal formation, as it has been pre-
viously observed [20], but they can also cause the dispersion
of small groups of layers of the LDH as well as reduc-
ing agglomeration during the nucleation and crystal growing
process, giving rise to particles with small crystal sizes. Such

322
M.J. Climent et al. / Journal of Catalysis 225 (2004) 316–326
Table 4
Surface properties of HTc series
US
a
a
Sample
S
S
Pore volume
Micropore volume
Pore diameter
(Å)
BET
2
MIC
2
−1
−1
3
−1
3
−1
(m g
)
(m g
)
(cm g
)
(cm g
)
HTc
HTc
HTc
HTc
HTc
245
275
263
284
254
20
77
88
138
62
0.62
0.062
0.32
0.04
0.04
0.01
94.7
103.77
95.36
96.75
98.20
CV
0.715
0.629
0.607
0.624
US273
US283
US298
US323
Results of adsorption and desorption of nitrogen of different calcined hydrotalcites samples (HTc series).
US
a
Microporous area calculated from the t plot, using as reference t, between 5.5 and 8.5.
the methylenic groups give us values of 0.34 and 0.32 for
malononitrile and ethyl cyanoacetate, respectively, indicat-
ing that the order of acidity is MN > ECA.
The selected calcined HTc samples (HTcT353, HTcMW433
,
HTcUS273, and HTcUS298) were tested for the Knoevenagel
condensation between benzaldehyde and MN, (molar ratio
1:1) using a 2.7% (wt/wt) of catalyst with respect to the to-
tal amount of reagents, at 298 K and without solvent. The
results of initial activity (initial reaction rate, determined
at a low level of conversion, i.e., x < 10%) as well as the
yield to the Knoevenagel adduct (1) are presented in Ta-
ble 5. There, they are compared with the results obtained
with a mixed Mg–Al oxide synthesized by the conventional
methodology. In all cases, the selectivity to the Knoevenagel
condensation compound was of 100%, and no product com-
ing from Michael addition of the MN to the Knoevenagel
adduct was observed. As it is displayed in Table 5 the or-
Scheme 1.
effects will not only increase the surface area of the mixed
oxide derivative, but it will also increase the structural de-
fects of the solid. This possibility will be tested below by
catalytic activity measurements.
3.2. Catalytic activity
In order to check the influence of the preparation method
of the layered material on the catalytic activity of the
mixed oxides obtained by calcination, we selected differ-
ent samples of HTc pertaining to the three different series:
HTcT353, HTcMW433, HTcUS273, and HTcUS298. These ma-
terials were tested as basic catalysts for the Knoevenagel
condensation and aldol condensation between citral and ace-
tone.
der of activity was: HTcUS298 > HTcUS273 > HTcT353
>
HTcMW433 > HTcCV. In Fig. 6 the activity of the catalysts
per unit surface area is plotted versus the BET surface area
of the samples.
For the condensation of benzaldehyde and ECA, the re-
actions were carried out using a 5.5% (wt/wt) of catalysts
at 333 K. In this case the selectivity for the Knoevenagel
product (2) was also 100%, and the same order of activity
(Table 6) and activity per unit surface area (Fig. 6) were ob-
served.
The results show that the specific catalyst activity in-
creases when increasing the surface area of the mixed ox-
ides, and it is maximum for the mixed oxide produced from
the hydrotalcite obtained by in situ sonication during pre-
cipitation (Tables 5 and 6). These results clearly show that,
as can be expected, the total number of basic sites increases
when increasing the surface area of the solid. However, re-
sults from Fig. 6 also show that the activity increase with
surface area is not lineal but exponential. This indicates that
by increasing the surface area, i.e., decreasing the “crystal
size” of the mixed oxide, active sites with different strengths
should be produced.
In the case of Mg–Al mixed oxides formed by the con-
trolled calcinations of hydrotalcites, the active basic sites are
associated to hydroxide groups and different of O−2–Mg2+
acid–base pairs [4]. The basicity of the Lewis sites asso-
ciated to O−2 anions depends on their coordination. Those
3.2.1. Knoevenagel condensation
Knoevenagel condensation [27] of carbonyl compounds
with reactants containing an active methylenic group is one
of the most important methods for preparating substituted
alkene derivatives (Scheme 1). These compounds are of in-
terest as end products and intermediates in the production
of fine chemicals and commodities such as perfumes [28],
pharmaceuticals [29], and polymers [30]. Besides the im-
portance of the Knoevenagel condensation as a synthetic
organic method, another relevant characteristic of this reac-
tion is its ability to measure the total number of basic sites
and the relative strength of different base catalysts, by react-
ing a series of methylenic compounds with different pKa
values [31]. Then, the different mixed oxides were tested
for the Knoevenagel condensations between benzaldehyde
and malononitrile (MN) (pKa = 11) or ethyl cyanoacetate
(ECA) (pKa = 9) (Scheme 1). It should be pointed out, that
according to the pKa’s reported in the literature and which
were determined in aqueous solutions, the decreasing order
of acidity of the reactants is ECA > MN. However, calcu-
lations of the positive charge density of the hydrogen at

M.J. Climent et al. / Journal of Catalysis 225 (2004) 316–326
323
Table 5
Results of the Knoevenagel condensation between benzaldehyde and malononitrile
6
3
Sample
S
r /S ×10
0
Yield 1, 15 min
(%)
Yield 1, 240 min
(%)
r
×10
BET
2
0
−1
)
−1
−1
)
(m
g
(mol g min
HTc
HTc
HTc
HTc
HTc
245
265
246
275
284
1.950
2.437
2.083
3.413
3.812
8.0
9.2
8.4
12.4
13.4
23
38
32
44
50
95
96
82
83
91
CV
T353
MW433
US273
US298
Condensation of benzaldehyde with malononitrile in the presence of different calcined hydrotalcites.
Table 6
Results of the Knoevenagel condensation between benzaldehyde and ethyl cyanoacetate
6
3
Sample
S
r /S ×10
0
Yield 2, 5 min
(%)
Yield 2, 240 min
(%)
r
×10
BET
2
0
−1
)
−1
−1
)
(m
g
(mol g min
HTc
HTc
HTc
HTc
HTc
245
265
246
275
284
4.457
5.986
4.343
6.943
8.514
18.2
22.6
17.7
25.2
30.0
31
42
30
49
60
88
91
91
93
96
CV
T353
MW433
US273
US298
Results of the Knoevenagel condensation between benzaldehyde and ethyl cyanoacetate in the presence of different calcined hydrotalcites.
Scheme 2.
the fraction of the stronger ones which is maximum for sam-
ples synthesized by the sonication treatment.
3.2.2. Citral–acetone condensation
In order to check our hypothesis we selected a more de-
manding reaction from the point of view of basic strength:
the aldol condensation between citral and acetone (Scheme 2).
In fact, for carbonylic compounds, the pKa associated with
the hydrogen at the methyl and methylene groups in the al-
pha position to the carbonyl group is in the range of 25.
Particularly for the acetone, the positive charge density of
the hydrogen at the methyl group was +0.12, which is a
value considerably lower than that of the methylene-active
compounds tested previously in the Knoevenagel condensa-
tion (+0.34 and +0.32). Moreover, this reaction has indus-
trial interest because it produces ionones (cyclic terpenoids),
which are used for enhancing the organoleptic properties of
consumables such as medicinal products and fragrances and
as intermediates in vitamin A synthesis.
Although this condensation reaction can be catalyzed by
either acid or bases, numerous commercial methods make
use of conventional homogeneous bases as catalysts (i.e.,
aqueous alkali metal hydroxide solutions, alcoholates in al-
cohol, or benzene solvents) which lead to waste streams.
However, owing to the industrial importance of the process,
solid catalysts such as basic alumina, alkali oxides, MgO,
Fig. 6. Influence of the area BET on the initial formation rate of 1 (1)
and 2 (").
oxygen atoms located in corners of the crystal should have a
stronger basicity than oxygen atoms located either on edges
or on crystal faces [32]. Then, we have to expect that sam-
ples with lower “crystal size” possess higher concentrations
of Lewis sites of low coordination number, and therefore the
basicity should be higher. Since in our samples, the specific
activity increases when increasing the surface area, we have
to conclude that an increase of surface area not only leads to
an increase of the number of active sites, but also increases

324
M.J. Climent et al. / Journal of Catalysis 225 (2004) 316–326
Table 7
Results of the aldol condensation of citral with acetone
6
a
a
3
Sample
S
r /S ×10
0
Yield 3
(%)
Selectivity 3
(%)
r
×10
BET
2
0
−1
)
−1
−1
)
(m
g
(mol g min
HTc
HTc
HTc
HTc
HTc
HTc
245
265
246
237
275
284
0.42
0.58
0.39
0.38
0.58
0.74
1.7
2.2
1.6
1.6
2.1
2.6
68
81
64
61
78
93
82
83
86
87
88
95
CV
T353
MW433
MW513
US273
US298
Results of the aldol condensation between citral and acetone using different synthesized calcined hydrotalcites.
a
At 1 h reaction time.
and calcined hydrotalcites may have an opportunity as po-
tential industrial catalysts.
In previous work [33], we have studied the aldol conden-
sation between citral and acetone in the presence of Mg–Al
calcined hydrotalcites at 333 K, and we have found that
it is possible to obtain excellent conversions and selectivi-
ties to pseudoionones working with relatively low acetone
to citral molar ratios. Thus, when the aldolic condensation
between citral an acetone was carried out in the presence
of the calcined hydrotalcites described in this work (16.6%
(wt/wt)), at 333 K using a molar ratio acetone/citral of 2.8,
the main products obtained were a mixture of cis and trans
pseudoionones (3) (6,10-dimethyl-3,5,9-undecatrien-2-one)
(Scheme 2). Besides, the pseudoionones, other condensa-
tion products coming from the self-condensation of acetone,
self-condensation of citral, and oligomers derived from citral
were detected in the reaction mixture. No β-hydroxyl ke-
tones derived from acetone or pseudoionones were detected
under these experimental conditions.
Fig. 7. Influence of the area BET of different calcined hydrotalcite samples
on the initial rate of formation of pseudoionones 3.
In Table 7 yields and selectivities to pseudoionones ob-
tained with the selected calcined HTc samples (HTcT353
,
condensations showed that hydrotalcite was not regenerated
during the reaction.
HTcMW433, HTcMW513, HTcUS273, and HTcUS298) and mixed
Mg–Al oxide synthesized by the conventional methodology
(HTcCV) after a 1 h reaction time are summarized. Results
from Table 7 clearly indicate that the HTcUS298 gives a
larger catalytic activity than HTcCV or any other of the cal-
3.3. Hydrotalcites regenerated by rehydration
Recently, it has been presented that a calcined hydrotal-
cite can be activated by rehydration at room temperature by
contacting with a stream of nitrogen saturated with water va-
por, during long periods of time or by directly adding water
to the calcined solid [34,35].
It is a general belief that the basic sites produced by
calcination are oxygen anions of low coordination, which
correspond to Lewis basic sites. However when the calcined
hydrotalcite is rehydrated, basic Lewis sites are converted
to basic hydroxyl Brønsted sites, which become the com-
pensating interlayer anions. On the other hand, it has been
presented in the literature that rehydrated mixed oxides were
very active for aldol condensation reactions such as the self-
condensation of acetone [36], condensation of benzaldehyde
with acetone [37], condensation of citral, with acetone or
methyl ethyl ketone, and the Claisen–Schmidt condensa-
tion [38].
cined hydrotalcites, with the order of activity HTcUS298
>
HTcUS273 = HTcT353 > HTcMW433 > HTcCV > HTcMW513
.
When the specific activity for pseudoionone formation
was plotted versus the specific surface area of different
mixed oxides (Fig. 7) it was possible to see that, as it was
observed previously for Knoevenagel condensation, an in-
crease of the surface area of the calcined hydrotalcites does
not only increase the number of basic sites, but also the frac-
tion of the stronger ones. Furthermore it appears that the
mixed oxide obtained from the hydrotalcite synthesized by
applying ultrasounds is almost twice more active than the
hydrotalcite prepared by a conventional procedure. Indeed,
it was possible with the HTCUS298 catalyst to obtain after 1 h
reaction time a yield of pseudoionones of 93%, with a selec-
tivity of 95%. On the other hand in the presence of HTcCV
only a yield of 68% with selectivity of 82% was found.
It must be noted that XRD of the mixed oxides performed
after their use as catalysts for the Knoevenagel and aldol
It is known that the aldol condensation is a reversible re-
action when water is one of the products, and an excess of

M.J. Climent et al. / Journal of Catalysis 225 (2004) 316–326
325
Table 8
surface area, not only changes the total number of the basic
sites exposed to reactants, but also increases the fraction of
the stronger ones. These correspond to O2− located in cor-
ners of the crystals. Then, the smaller the crystal size the
larger the fraction of the above sites, and the larger the to-
tal number of the exposed sites. This agrees very well with
the fact the regenerated hydrotalcites show a larger amount
of accessible Brønsted basic sites for samples with a higher
surface area. Using this rehydrated material it is possible
to obtain yields of pseudoionones of 96% with 99% of se-
lectivity, in 15 min of reaction time working at a very low
acetone/citral molar ratio. This result is superior to any re-
ported up to now for this condensation in homogeneous and
heterogeneous phases.
Results of aldol condensation between citral and acetone in the presence of
hydrotalcite sample (HTc
) using different percentages of water added
US298
3
Water added
(%)
Yield 3, 15 min
Selectivity 3
r
×10
0
−1
−1
)
(%)
(%)
(mol g min
0
20
36
46
0.71
0.87
1.00
0.88
69
83
96
84
88
97
99
98
Reaction conditions: acetone (18.2 mmol), citral (6.5 mmol), 414 mg of
catalyst, at 333 K.
water on hydrotalcite should shift the reaction equilibrium
toward the formation of the reactants. According to this, the
condensation reaction between citral and acetone was per-
formed adding on the calcined hydrotalcite (HTcUS298) at
different percentages of water with respect to the solid cata-
lyst (20, 36, and 46% (wt/wt)). In Table 8 it can be seen that
the calcined hydrotalcite was less active than the rehydrated
samples and a maximum of activity was found when the con-
densation was carried out using freshly calcined hydrotalcite
with 36% (wt/wt) of water added of. It is very interesting to
note, that the optimized rehydrated sample reached yields of
pseudoionones of 96% with 99% of selectivity in 15 min of
reaction time, working at a very low acetone/citral molar ra-
tio. This result is superior to any reported up to now for this
condensation in a homogeneous and heterogeneous phase.
Acknowledgment
Financial support of the Spanish Ministry of Science and
Technology (MAT2003-079-45-CO2-01) is acknowledged.
References
[1] F. Cavani, F. Trifiro, A. Vaccari, Catal. Today 11 (1991) 173.
[2] R. Allmann, Chimia 24 (1970) 99.
[3] S. Miyata, Clays Clay Miner. 28 (1980) 50.
[4] W.T. Reichle, S.Y. Kang, D.S. Everhardt, J. Catal. 101 (1986) 352.
[5] H.F.W. Taylor, Miner. Mag. 37 (1969) 338.
[6] J. Valente, F. Figueras, M. Gravelle, P. Kumbhar, L. Lopez, J.P. Besse,
J. Catal. 189 (2000) 370.
4. Conclusions
[7] D. Tichit, M.H. Lhouty, A. Guida, B.H. Chiche, F. Figueras, A. Aroux,
D. Bartalini, E. Garrone, J. Catal. 151 (1995) 50.
It has been shown that during the synthesis of Mg–Al
LDH, the increase of temperature during the coprecipita-
tion step with and without simultaneous sonication (HTT and
HTUS series, respectively) or during the ageing step of the
gels under microwave irradiation (HTMW series) causes an
increase of the crystal size.
[8] D.M.P. Mingos, D.R. Baghurst, Chem. Soc. Rev. 20 (1991) 1.
[9] S. Komarneni, R. Roy, Q.H. Li, Mater. Res. Bull. 27 (1992) 1393.
[10] S. Komarneni, R. Roy, Q.H. Li, J. Mater. Chem. 4 (1994) 1903.
[11] S. Komarneni, R. Roy, Q.H. Li, J. Mater. Res. 11 (1996) 1866.
[12] S. Möhmel, I. Kurzawski, D. Uecker, D. Müller, W. Gebner, Cryst.
Res. Technol. 37 (2002) 359.
[13] S. Kannan, R.V. Jasra, J. Mater. Chem. 10 (10) (2000) 2311.
[14] G. Fetter, F. Hernández, A.M. Maubert, V.H. Lara, P. Bosch, J. Porous
Mat. 4 (1997) 27;
G. Fetter, P. Bosch, F.A. Hernandez, Mater. Res. Soc. Symp. Proc. 454
(1997) 235.
[15] D. Tichit, A. Rolland, F. Prinetto, G. Fetter, M.J. Martinez-Ortiz,
M.A. Valenzuela, P. Bosch, J. Mater. Chem. 12 (2002) 3832.
[16] Y. Fang, D.K. Agrawal, D.M. Roy, P.W. Brown, J. Mater. Res. 7 (1992)
2294.
[17] J.S. Hu, K.K. Agrawa, Y. Fang, R. Roy, J. Mater. Sci. 28 (1993) 5297.
[18] N. Enomoto, H.L. Choi, M. Katsumoto, Z. Nakagawa, Trans. Mater.
Res. Soc. Jpn. 14A (1994) 777.
[19] A.P. Kapustin, in: Effect of Ultrasounds on the Kinetics of Crystalliza-
tion, Consultants Bureau, New York, 1963, p. 65.
[20] Y. Seida, Y. Nakano, Y. Nakamura, Clays Clay Miner. 50 (2002) 525.
[21] B.F. Sels, D.E. De Vos, P.A. Jacobs, Catal. Rev. 43 (2001) 443.
[22] JCPDS X-ray Powder Diffraction File No. 22-700, 1986.
[23] F. Rey, V. Fornes, J.M. Rojo, J. Chem. Soc., Faraday Trans. 88 (1992)
2233.
[24] (a) K.S. Suslick, Adv. Organomet. Chem. 25 (1986) 73;
(b) K.S. Suslick, Modern Synth. Methods 4 (1986) 1.
[25] W. Lauterborn, W. Hentschel, Ultrasonics 24 (1985) 59.
[26] K.S. Suslick, D.J. Casadonte, M.L.H. Green, M. Thompson, Ultrason-
ics 25 (1987) 56.
It has been found that the sample prepared under son-
ication at 298 K (HTUS298) compared with samples of
other series presents a homogeneous morphology and highly
dispersed particles with an average size of 80 nm. The
larger surface area presented by the mixed oxide derivative
(HTcUS298) (20% larger than that of the mixed oxide conven-
tionally synthesized) not only could be due to the fact that
ultrasounds accelerate the crystal formation, but also to the
dispersion of small groups of layers and decreasing agglom-
eration during the nucleation and crystal growing process,
affording crystals with smaller sizes than those obtained by
a conventional ageing method. We have shown that mixed
oxides derived from hydrotalcites synthesized by using ul-
trasound have improved considerably the surface area of the
calcined hydrotalcite, as well as the number of the accessible
active sites.
The corresponding mixed oxides obtained by calcina-
tion of the hydrotalcites prepared with ultrasound present a
higher specific activity for Knoevenagel and aldol conden-
sations. This was explained by assuming that the increase in

326
M.J. Climent et al. / Journal of Catalysis 225 (2004) 316–326
[27] L. Knoevenagel, Bersenges 31 (1898) 259.
[28] (a) B. Siebenhaar, WO9721659 (1997);
[32] E. Garrone, F.S. Stone, in: 8th Int. Cong. Catal., Chemie III, Berlin,
1984, p. 441.
(b) R.T. Hopp Thielmann, W. Gottsch, US5212153 (1992);
(c) G. Lauterbach, R. Becker, K. Jansen, EP 0395982 (1990);
(d) B. Siebenhaar, B. Casagrande, EP0870750 (1998);
(e) K. Manfred, A. Slagmulder, US4755615 (1988).
[29] (a) M.F. Gordeev, D.V. Patel, WO 9633972 (1996);
(b) E. Castelli, G. Cascio, E. Manghisi, WO9807698 (1998);
(c) A.J. Kesel, W. Oberthür, WO 9820013 (1998).
[30] (a) J.P. Ferraris, T.L. Lambert, S. Rodriguez, WO 9305077 (1993);
(b) R. Schipfer, G. Schmolzer, US 4523007 (1985) and US 4544715
(1985).
[33] M.J. Climent, A. Corma, S. Iborra, A. Velty, Catal. Lett. 79 (2002).
[34] M.J. Climent, A. Corma, S. Iborra, A. Velty, Green Chem. 4 (2002)
474.
[35] K.K. Rao, M. Gravelle, J. Valente, F. Figueras, J. Catal. 173 (1998)
115.
[36] R. Teissier, D. Tichit, F. Figueras, J. Kervennal, FP 95 00094 (1995).
[37] A. Guida, M.H. Lhouty, D. Tichit, F. Figueras, P. Geneste, Appl.
Catal. 164 (1997) 251.
[38] (a) M.J. Climent, A. Corma, S. Iborra, A. Velty, J. Catal. 221 (2004)
474;
[31] A. Corma, V. Fornés, R.M. Martín Aranda, H. Garcia, J. Primo, Appl.
Catal. 59 (1990) 237.
(b) J.C.A. Roelofs, D.J. Lensveld, A.J. Van Dillen, K.P. De Jong,
J. Catal. 203 (2001) 184.