4
S. Sebti et al. / Journal of Catalysis 213 (2003) 1–6
the main modifications in the pattern are produced in the
range of 2θ = 32–35◦, with new lines at 33.68◦ and 34.34◦
together with an increase in the intensity of other lines in
that zone. These lines may correspond to sodium phosphate
species, given that the most intense lines in the references
are 34.1◦ for γ -Na3PO4, 34.2◦ for β-Na3PO4, and 31.9◦ for
Na2HPO4. The formation of very small crystals, partially
amorphous materials or mixed phosphates with different
stoichiometry may account for the lack of intense bands
of sodium phosphate species. It has been shown that in
other base-catalyzed reactions [17b] this insoluble sodium
phosphate may be the active species.
The natural phosphate shows a very low surface area
(1.4 m2/g) together with a low total pore volume (0.0055
cm3/g). In the case of NaNO3/NP calcined at 900 ◦C
the surface area is even lower (0.66 m2/g) with similar
total pore volume. It is then rather surprising that this
solid has a very high catalytic activity and basicity was
determined by phenol adsorption. Solutions of phenol were
prepared in cyclohexane with concentrations in the range
5–200 mM and n-octane was added as internal standard.
Variable amounts of the solids (15–45 mg) were stirred at
room temperature with different volumes of those solutions
(3–9 ml) in order to study the adsorption equilibrium. The
phenol remaining in solution was analyzed by GC. Values
of adsorbed phenol were in the range of 0.5–1.8 mmol/g.
These values are surprisingly high for a solid with such a low
surface area; in fact they represent a surface concentration of
4.5–16.4 molecules/Å2. This concentration cannot be due
to chemisorption and it is difficult to find an explanation.
However, the high polarity of the surface might be the origin
of this behavior. In any case, it is clear that the solid is
basic, as indicated by its catalytic activity, observed in the
epoxidation of electron-deficient alkenes with H2O2 [17b],
a reaction that needs the presence of basic centers.
the catalytic activity was observed in subsequent reactions.
The yields obtained are 98, 75, 60, and 35% using the fresh
catalyst, cycles 1, 2, and 3, respectively. However, the activ-
ity was almost completely recovered when the catalyst was
washed with acetone and calcined at 900 ◦C. Thus, the yields
obtained are 98, 94, and 92% with fresh catalyst, cycles 1
and 2, respectively. The strong adsorption of some byprod-
ucts may be responsible for this behavior.
3.2. Scope of NaNO3/NP
In order to determine the scope and limitations associated
with this new catalyst, the optimum conditions for the
Claisen–Schmidt condensation between benzaldehyde and
acetophenone were applied to other substrates (Table 3). All
of the reactions proceeded efficiently at room temperature.
Although all products are obtained in high isolated yields,
longer reaction times are required in the synthesis of some
chalcones. The presence of an electron-donor group in the
aldehyde (entry 4) or the ketone (entry 6) made the reaction
less favorable. The worst result was obtained when both
reagents contained donor methoxy groups (entry 8).
3.3. Effect of addition of water
Having shown that NaNO3/NP is an active catalyst for a
wide variety of Claisen–Schmidt condensations, we tried to
further improve the catalytic activity of this solid. We have
reported previously that addition of small amounts of water
to reaction mixtures can increase the catalytic activity of
natural and synthetic phosphates [16,20–22]. We tested this
approach in the synthesis of chalcone 3a using NaNO3/NP
in the presence of different amounts of water in conjunction
with methanol or ethanol as solvents. When the reaction was
carried out with 1 ml of methanol, the addition of water led
to a decrease in the yield. Thus, the yields obtained after 6 h
are 65, 57, 47, 35, and 28% in the presence of 0.05, 0.1, 0.3,
0.5, and 1 ml of water, respectively, whereas the yield is 57%
with anhydrous solvent. In contrast, the behavior in ethanol
is completely different in that the addition of small amounts
of water has a positive effect. The yields obtained are 13,
21, 55, 42, and 23% using 0.05, 0.1, 0.3, 0.5, and 1 ml of
water respectively. Only 4% of product was isolated with the
anhydrous solvent. The best result was then obtained with
0.3 ml of water in 1 ml of ethanol.
The condensation of the benzaldehyde and the acetophe-
none was used as a benchmark reaction to study the influ-
ence of the volume of the methanol. The best yield (98%) is
obtained after 24 h with a volume of 1–3 ml. An increase in
the volume up to 5 ml slightly decreases the reaction yield
(88%), and this drops further to 40% when a volume of 10 ml
is used. In absence of the solvent, only a 10% yield is ob-
tained. This behavior indicates that some solvent is needed
to facilitate the contact between the reagents and the active
sites, but a large volume reduces the concentration and, as a
consequence, the reaction rate.
We proceeded to study the solvent effect in the syn-
thesis of chalcone 3a using the NaNO3/NP catalyst. In
dichloromethane, THF, hexane, and butanol the reaction did
not take place. The use of isopropanol gave only a 12% yield
of 3a after 24 h, but 63% of 3a was isolated when the re-
action was carried out in ethanol. It can be concluded that
methanol is the best solvent for this reaction.
3.4. Effect of addition of an ammonium salt
In previous work it was shown that the addition of an am-
monium salt to natural phosphate [16], fluorapatite [20], or
hydroxyapatite [21] increases the activity of these catalysts.
Therefore, we carried out the synthesis of chalcone 3a with
NaNO3/NP in methanol using different amounts of benzyl-
triethylammonium chloride (BTEAC). The results obtained
after 6 h show that the best yields are obtained with 0.03–
0.06 g of BTEAC. Moreover, the kinetic curves for the syn-
To complete this preliminary study, the issue of catalyst
recovery was also considered. When the catalyst was sepa-
◦
rated by filtration and dried at 150 C, a rapid decrease of