H3PW12O40 supported on silica-encapsulated γ-Fe2O3 nanoparticles: A novel magnetically- recoverable catalyst for three-component Mannich-type reactions in water

Article abstract of DOI:10.1039/c1gc15291b

A new type of magnetically-recoverable catalyst was synthesized by the immobilization of H₃PW₁₂O₄₀ on the surface of silica-encapsulated γ-Fe₂O₃ nanoparticles. This catalyst was characterized by transmission electron microscopy (TEM), a laser particle size analyzer, infrared spectroscopy (IR) and inductively coupled plasma atomic emission spectroscopy (ICP-AES). The results show that the particles are mostly spherical in shape and have an average size of approximately 94 nm. The characterization data derived from IR spectroscopy reveal that H₃PW₁₂O₄₀ on the support exists in the Keggin structure. The acidity of the catalyst was measured by a potentiometric titration with n-butylamine. To our surprise, this very strong solid acid catalyst showed an excellent distribution of acid sites, suggesting that the catalyst possesses a higher number of surface active sites compared to its homogeneous analogues. The activity of the catalyst was probed through one-pot three-component Mannich-type reactions of aldehydes, amines and ketones in water at room temperature. The excellent conversions show that the catalyst has strong and sufficient acidic sites, which are responsible for its catalytic performance. After the reaction, the catalyst/product separation could be easily achieved with an external magnetic field, and more than 95% of the catalyst could usually be recovered. The catalyst was reused at least five times without any loss of its high catalytic activity.

Full text of DOI:10.1039/c1gc15291b

View Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Green Chemistry
Cite this: Green Chem., 2011, 13, 2116
www.rsc.org/greenchem
PAPER
H3PW12O40 supported on silica-encapsulated c-Fe2O3 nanoparticles: a
novel magnetically-recoverable catalyst for three-component Mannich-type
reactions in water
Ezzat Rafiee*a,b and Sara Eavania
Received 17th March 2011, Accepted 5th May 2011
DOI: 10.1039/c1gc15291b
A new type of magnetically-recoverable catalyst was synthesized by the immobilization of
H3PW12O40 on the surface of silica-encapsulated g-Fe2O3 nanoparticles. This catalyst was
characterized by transmission electron microscopy (TEM), a laser particle size analyzer, infrared
spectroscopy (IR) and inductively coupled plasma atomic emission spectroscopy (ICP-AES). The
results show that the particles are mostly spherical in shape and have an average size of
approximately 94 nm. The characterization data derived from IR spectroscopy reveal that
H3PW12O40 on the support exists in the Keggin structure. The acidity of the catalyst was measured
by a potentiometric titration with n-butylamine. To our surprise, this very strong solid acid
catalyst showed an excellent distribution of acid sites, suggesting that the catalyst possesses a
higher number of surface active sites compared to its homogeneous analogues. The activity of the
catalyst was probed through one-pot three-component Mannich-type reactions of aldehydes,
amines and ketones in water at room temperature. The excellent conversions show that the catalyst
has strong and sufficient acidic sites, which are responsible for its catalytic performance. After the
reaction, the catalyst/product separation could be easily achieved with an external magnetic field,
and more than 95% of the catalyst could usually be recovered. The catalyst was reused at least five
times without any loss of its high catalytic activity.
to the reaction medium, these methods are time- and energy-
Introduction
consuming. Moreover, a substantial decrease in the activity and
Over the past several past decades, Green Chemistry has
selectivity of the immobilized catalyst is frequently observed due
demonstrated how fundamental scientific methodologies can
to the heterogeneous nature of support materials in reaction
protect human health and the environment in an economically
media, steric and diffusion factors.1 A great proportion of active
beneficial manner. Significant progress is being made in several
species are deep inside the supporting matrix and thus reactants
key research areas, such as catalysis, the design of safer chemicals
have limited access to catalytic sites. Therefore, to maintain
and environmentally benign solvents.
economic viability, a suitable heterogeneous system must not
The application of catalysis to reduced toxicity systems,
only minimize the production of waste, but should also exhibit
benign and renewable energy systems, and reaction efficiency
activities and selectivities comparable or superior to existing
makes it a central focus area for green chemistry research in the
homogeneous routes.
21st century. It is possible to prepare heterogeneous analogues
In an attempt to resolve such problems, nanoparticles (NPs)
of the most commonly used soluble and homogeneous catalysts
have been used as alternative soluble matrixes for supporting
by their immobilization on various insoluble supports. The use
homogeneous catalysts. When the size of the support materials
of heterogeneous catalysts in chemical processes would simplify
is decreased to the nanometer scale, the surface area of the
catalyst removal and minimize the amount of waste formed.
NPs increases dramatically. As a consequence, NPs could have
However, although heterogeneous catalysts can be recovered by
a higher catalyst loading capacity and a higher dispersion than
filtration or precipitation through adding precipitating agents
many conventional support matrices, leading to an improved
catalytic activity of the supported catalysts. However, conven-
tional separation methods may become inefficient for support
aFaculty of Chemistry, Razi University, Kermanshah, 67149, Iran.
particle sizes below 100 nm. The incorporation of magnetic
E-mail: ezzat_rafiee@yahoo.com, e.rafiei@razi.ac.ir; Fax: +98 831
NPs (MNPs) such as iron oxide into supports offers a solution
4274559; Tel: +98 831 4274559
to this problem.2 The strategy of magnetic separation, taking
bKermanshah Oil Refining Company, Kermanshah, Iran
2116 | Green Chem., 2011, 13, 2116–2122
This journal is
The Royal Society of Chemistry 2011
©

View Online
advantage of MNPs, is typically more effective than filtration or
centrifugation as it prevents loss of the catalyst. The magnetic
separation of MNPs is simple, economical and promising for
industrial applications.3 Currently, much attention is focused on
the synthesis of magnetic core–shell structures by coating an
SiO2 shell around pre-formed NPs.4 Nanosized silica, which is
non-toxic and can be grafted with a variety of surface modifiers,
has great potential in many applications.5
The IR spectrum of the synthesized g-Fe2O3 NPs is shown in
Fig. 1. The absorption bands at 449.5, 588.7, 637.8, 799.5 and
900 cm-1 are in good agreement with the IR results reported for
pure maghemite (g-Fe2O3).10
Since Keggin-type heteropoly acids (HPAs) such as
H3PW12O40 (PW) were first commercialized in the industrial
research field, a large number of these catalysts have been
developed for a wide range of organic reactions in fundamen-
tal research.6,7 Several advantages to be highlighted of using
supported HPAs compared to homogeneous examples include
easier recovery and recycling after carrying out reactions, and
easier product separation.8 However, separation and recovery
of the immobilized HPAs are usually performed by filtration or
centrifugation, which are not eco-friendly processes.
The immobilization of PW on silica-coated g-Fe2O3 NPs
(designed as g-Fe2O3@SiO2-PW) can be employed to derive
a novel heterogeneous catalyst system that possesses both a
high separation efficiency and a relatively high surface area to
maximize catalyst loading and activity. Certainly, the use of envi-
ronmentally benign solvents has been another key research area
of green chemistry, with great advances being seen in aqueous
catalysis. Along this line, we synthesized g-Fe2O3@SiO2-PW as
a novel nanomagnetically-recoverable catalyst and assessed its
catalytic activity in Mannich-type reactions in water. Indeed,
the venerable Mannich reaction and its variants represent one
of the more powerful constructs for alkaloid synthesis.9 To the
best of our knowledge, this is the first report on the synthesis,
characterization and catalytic application of a g-Fe2O3@SiO2-
PW catalyst.
Fig. 1 IR spectrum of g-Fe2O3 NPs.
Sonication of the g-Fe2O3 NP suspension in an alka-
line ethanol–water solution of tetraethyl orthosilicate (TEOS)
caused the rapid coating of the magnetic cores with silica shells.
The outer shell of silica not only improves the dispersibility
but also provides suitable sites (Si–OH groups) for surface
functionalization with PW. Ultimately, mixing a suspension of
g-Fe2O3@SiO2 with a methanolic solution of PW lead to the
formation of g-Fe2O3@SiO2-PW NPs. Elemental analysis from
ICP showed that the W content was 23.7 wt%. Typically, a
loading at ca. 31 wt% PW (1.1 mmol g-1) was obtained.
Fig. 2(a) and (b) show the TEM image and size histogram
for the g-Fe2O3@SiO2-PW NPs, respectively. The particle size
distribution from the TEM image shows that 66% of NPs are
in the range 75–90 nm and that the mean diameter of the
observable NPs is 73.5 nm. The size distribution of the g-
Fe2O3@SiO2-PW NPs derived from a laser particle size analyzer,
illustrated in Fig. 2(c), indicates that the mean diameter of the
particles is 93.8 nm. As shown in the TEM image (Fig. 2(d)),
aggregation/coalescence of individual g-Fe2O3 NPs occurred,
and that they could have been in an aggregated state during the
coating process. Such aggregations form bigger structures with a
non-spherical morphology, which causes the difference between
the results derived from TEM analysis and the measurements
made by the laser particle size analyzer. A typically core–shell
structure of a g-Fe2O3@SiO2-PW NP is shown in Fig. 2(e). The
dark core of the g-Fe2O3 and the grey silica shell are clearly
observable. The typical silica shell thickness was estimated to
be around 10 nm. Due to the strong mass difference between
SiO2 and PW, it could be assumed that the outside darker
contrast indicates the shell of PW particles immobilized onto
the g-Fe2O3@SiO2 NPs.
Results and discussion
Scheme 1 presents the synthetic strategy for g-Fe2O3@SiO2-PW
NPs.
IR spectra of g-Fe2O3@SiO2, g-Fe2O3@SiO2-PW and PW
samples are shown in Fig. 3. All samples show broad bands
at around 1650 cm-1, which are attributed to adsorbed water.
The transmission spectrum of g-Fe2O3@SiO2 shows that the
absorption band at 589 cm-1 could be related to the vibration
of g-Fe–O.11 The other peaks at 627 and 895 cm-1 are pure g-
Fe2O3.10 The bands at 1100 cm-1, along with shoulder at 1200
and 461 cm-1, are presumably due to asymmetric stretching
and bending modes of Si–O–Si, respectively. The shoulder at
800 cm-1 may be assigned either to a symmetric stretching of Si–
O–Si or a stretching vibration of Fe–O bonds. The characteristic
band of Si–O–Fe appears at 686 cm-1 in this sample.12 The
Scheme 1 Schematic illustration of the general approach to the
synthesis of g-Fe2O3@SiO2-PW NPs.
Magnetic g-Fe2O3 NPs were prepared through the chemical
co-precipitation method. In aerated solutions, an aqueous
mixture of ferric and ferrous salts precipitate as an aqueous
dispersion of g-Fe2O3 via the addition of strong alkaline
solutions in the pH range 7.5–14. In this work, a solution of
NH4OH was used as the alkali source. The chemical reaction of
g-Fe2O3 precipitation is given by:
Fe2+ + Fe3+ + O2 + 2OH- → Fe2O3 + H2O
PW12O40 Keggin ion structure is well known and consists of
3-
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 2116–2122 | 2117
©

View Online
Fig. 3 IR spectra of (a) g-Fe2O3@SiO2, (b) g-Fe2O3@SiO2-PW and (c)
PW.
Fig. 2 (a) TEM image, (b) size histogram, (c) grain size distribution,
(d) aggregation/coalescence and (e) typical core–shell structure of a
g-Fe2O3@SiO2-PW NP.
Fig. 4 Zeta potential distribution of g-Fe2O3@SiO2-PW.
a PO4 tetrahedron surrounded by four W3O13 groups formed
by edge-sharing octahedra. These groups are connected to each
other by corner-sharing oxygens.13 This structure gives rise to
four types of oxygen, being responsible for the fingerprint bands
of the Keggin ion between 700 and 1200 cm-1. PW shows typical
bands for absorptions at 1080 (P–O), 984 (W O), 896 and 814
(W–O–W) cm-1. In g-Fe2O3@SiO2-PW, the characteristic bands
are at the same wavenumbers, with a small shift according to the
interaction with the support. The PW band in the 1080 cm-1 zone
is masked by the Si–O–Si absorption band in the g-Fe2O3@SiO2-
PW spectrum. However, information can still be obtained from
the less affected regions, which show an intensity increase of the
band at 814 cm-1, and non-overlapped bands at 896 and 984 cm-1
that confirm that the PW on the support exists in the Keggin
structure.
peaks at -6.73 (60.3%) and -56.5 (38.6%) mV reveal that two
dominate HPA surface complexes exist on the g-Fe2O3@SiO2
surface. Clearly the Keggin ion is strongly interacting with the
surface hydroxyl groups of the silica.15 This may occur by an
exchange reaction, leading to the formation of water and the
replacement of the surface hydroxyl anion by a Keggin anion:
Si–OH(s) + H+ → Si–OH+
(aq)
2 (s)
Si–OH+ + H2PW12O-
→ SiH2PW12O40 (s) + H2O(l)
2 (s)
40 (aq)
+
or through interaction of a protonated hydroxyl group (OH2 )
with the anion:
The zeta potential of the g-Fe2O3 and g-Fe2O3@SiO2 samples
suspended in aqueous solutions were about -30 and -85 mV,
respectively. These results are in good agreement with those
reported in the literature14 and confirm that g-Fe2O3 NPs
were successfully covered by an SiO2 shell. The zeta potential
distribution for g-Fe2O3@SiO2-PW is shown in Fig. 4. 1% of
NPs showed a zeta potential of about -29.5 mV, indicating
the existence of a few uncovered g-Fe2O3 NPs. Another two
Si–OH+ + H2PW12O-
→ SiOH+ (H2PW12O40)-
2 (s)
40 (aq)
2
(s)
The present results do not allow a more detailed description
of the bonding and structures of the HPA species present on the
support surface, and more experiments are needed. However,
the shift in zeta potential value compared to g-Fe2O3 and g-
Fe2O3@SiO2 showed that all of the g-Fe2O3@SiO2 NPs are
modified with PW.
2118 | Green Chem., 2011, 13, 2116–2122
This journal is
The Royal Society of Chemistry 2011
©

View Online
The nature of the support apparently affects the acidity of
the supported HPAs, which was probed by a potentiometric
titration with an organic base.7 This method, based on the
measured potential difference, is mainly determined by the acidic
environment around the electrode membrane. The measured
electrode potential is an indicator of the acidic properties of
the dispersed solid particles. An aliphatic amine, such as n-
butylamine, with a basic dissociation constant of approximately
10-6, allows the potentiometric titration of strong acids. The
titration curves obtained for SiO2, PW, 40 wt% PW/SiO2 and
g-Fe2O3@SiO2 are presented in Fig. 5. It is considered that the
initial electrode potential (Ei) indicates the maximum strength of
the acid sites and that the value from which the plateau is reached
(mmol amine per g catalyst) indicates the total number of acid
sites that are present in the titrated solid. The acidic strength of
the acid sites can be classified according to the following ranges:
Ei > 100 mV (very strong acid sites), 0 < Ei < 100 mV (strong
acid sites), -100 < Ei < 0 mV (weak acid sites) and Ei < -100
mV (very weak acid sites).7 Bulk SiO2 presented strong acid sites
(Ei = 78.1 mV). g-Fe2O3@SiO2 showed very weak acid sites in
comparison with SiO2. This may be due to the interaction of
Si–O–H with g-Fe2O3 and the formation of Si–O–Fe species,
as confirmed by IR spectroscopy. The addition of PW to the
SiO2 surface (as described above) should decrease the strength
of the acid sites of the PW molecules. This effect depends on
the extent of the interaction of PW with the support. According
to potentiometric titration curves (Fig. 5), PW presented very
strong acidic sites (Ei = 650 mV). A 40 wt% PW/SiO2 sample
showed acidic sites with essentially the same maximum strength,
which might be an indication of a low interaction level and the
formation of predominantly less dispersed intact PW anions
on the silica support. Thus, this supported catalyst showed
significant leaching of PW in polar solvents. Potentiometric
titration results of a g-Fe2O3@SiO2-PW sample showed that
the lower Ei value (450 mV) compared to the PW and 40 wt%
PW/SiO2 samples (compare titration curves of 40 wt% PW/SiO2
and g-Fe2O3@SiO2-PW in Fig. 5 and Fig. 6, respectively) may
be a consequence of a strong interaction with the support for
g-Fe2O3@SiO2-PW. This interacting species is well dispersed on
the support surface and does not leach during the reaction.
Additionally, the immobilization of PW on g-Fe2O3@SiO2 NPs
Fig. 6 Potentiometric titration curves of g-Fe2O3@SiO2-PW and reused
g-Fe2O3@SiO2-PW.
considerably enhanced the dispersion of acidic protons to 48
mmol g-1 catalyst (Fig. 6). Therefore, a larger fraction of active
sites are exposed to the surface, and this catalyst may exhibit
excellent activity in organic reactions, even with a low catalyst
loading.
The activity of the novel catalyst was probed through a one-
pot three-component Mannich-type reaction of benzaldehyde,
aniline and cyclohexanone as model substrates. Without the
addition of catalyst, only the imine was formed through the
condensation of benzaldehyde and aniline.16 g-Fe2O3@SiO2 NPs
and PW produced the Mannich product with 65 and 85% yields,
but the imine also obtained in 20 and 15% yield, respectively
(Table 1, entries 1 and 2). Yet, their combination in the form of
g-Fe2O3@SiO2-PW increased the yield of the Mannich product
to 98% in a significantly shorter reaction time (Table 1, entry 3).
Besides, this novel catalyst showed a higher yield and selectivity
in comparison with 40 wt% PW/SiO2 (Table 1, entry 4). This
result indicated that the catalytic activity should be directly
proportional to the number of surface acid sites, which are
the ones accessible to the reactants. To investigate the effect
of catalyst loading, the model reaction was carried out in the
presence of different amounts of catalyst. The best result was
obtained in the presence of just 0.005 g of g-Fe2O3@SiO2-PW,
and the use of higher amounts of catalyst slightly decreased the
reaction time (Table 1, entries 5–9).
Recycling experiments were performed by the Mannich con-
densation of benzaldehyde, aniline and cyclohexanone in water.
The model reaction was carried out by using 0.1 g of catalyst
and the experiments were suitably scaled up (Fig. 7(a)). When
the reaction was complete, g-Fe2O3@SiO2-PW could be placed
on the side wall of the reaction vessel with the aid of an external
magnet (Fig. 7(b)) and water was removed from the mixture to
leave a residue (including the product and catalyst) (Fig. 7(c)).
Then, the product was dissolved in ethanol and the catalyst
easily separated from the product by attaching an external
magnet onto the reaction vessel, followed by decantation of
the product solution (Fig. 7(d)). The remaining catalyst was
washed with diethyl ether to remove the residual product, dried
under vacuum and reused in a subsequent reaction. More
than 95% of the catalyst could usually be recovered from
each reaction. In a test of five cycles, the catalyst could be
Fig. 5 Potentiometric titration curves of SiO2, PW, 40 wt% PW/SiO2
and g-Fe2O3@SiO2.
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 2116–2122 | 2119
©

View Online
Table 1 Catalytic activity of g-Fe2O3@SiO2-PW in the model reaction
Time (min)/Yield (%)a
Imine
Entry
Catalyst
Mannich product
1
2
3
4
5
6
7
8
9
g-Fe2O3@SiO2 (0.04 g)
180/20
60/15
5/0
80/30
5/0
180/65
60/85
5/98
80/65
5/98
PW (0.04 g)
g-Fe2O3@SiO2-PW (0.04 g)
40 wt% PW/SiO2 (0.04 g)b
g-Fe2O3@SiO2-PW (0.08 g)
g-Fe2O3@SiO2-PW (0.02 g)
g-Fe2O3@SiO2-PW (0.01 g)
g-Fe2O3@SiO2-PW (0.005 g)
g-Fe2O3@SiO2-PW (0.003 g)
8/0
8/98
12/0
15/0
30/15
12/98
15/98
30/85
a Isolated yield. b W content in anhydrous catalyst from ICP is 29.6 wt%.7
Fig. 8 The reusability of g-Fe2O3@SiO2-PW.
Fig. 7 Separation of the g-Fe2O3@SiO2-PW catalyst from the reaction
mixture using an external magnet.
the corresponding Mannich products in good to excellent yields
(Table 2, entries 11 and 12).
reused without any significant loss to its catalytic activity (Fig.
8). The potentiometric titration curve of the reused catalyst
showed that all of its acidic sites were preserved during five
cycles (Fig. 6). According to ICP-AES results, PW leaching was
negligible.
Experimental
The reaction was extended to a series of aldehydes, ketones
and amines to explore the generality of this catalytic system
(Table 2). In the cases of b-aminocyclohexanones, the stere-
General
The reagents and solvents used in this work were obtained
from Fluka, Aldrich or Merck and were used without further
purification. TEM analyses were performed using a TEM
microscope (Philips CM 120 KV, The Netherlands). The size
distribution and zeta potential of the samples were obtained
using a laser particle size analyzer (HPPS5001, Malvern, UK).
IR spectra were recorded as KBr pellets using a Shimadzu 470
spectrophotometer. The tungsten (W) content in the catalyst was
measured by ICP-AES on a Spectro Ciros CCD spectrometer.
The potential variation was measured with a Hanna 302 pH
meter and a double junction electrode.
1
oselectivity was determined by H NMR spectroscopy and by
comparison with known compounds (Table 2, entries 1–3).16
Selectivity for the anti-isomer was favored in all cases. Aliphatic
aldehyde and amine substrates afforded the corresponding
products in good yields (Table 2, entries 9 and 10). Previous
approaches gave negative results or very poor yields for similar
reactions.17
It is worth mentioning that the ortho-substituted aromatic
amines, which generally appeared in only trace amounts or even
not at all because of the large steric hindrance,18 also afforded
2120 | Green Chem., 2011, 13, 2116–2122
This journal is
The Royal Society of Chemistry 2011
©

View Online
Table 2 The synthesis of various b-amino ketones via Mannich reactions
Entry
R1
R2
R3
R4
Time/min
Yield (%)a
1
2
3
4
5
6
7
8
Ph
Ph
Ph
–(CH2)4–
–(CH2)4–
–(CH2)4–
15
45
50
25
20
25
40
30
120
50
180
65
98b
96b
91b
90
92
90
80
90
83
85
71
95
4-ClC6H4
4-ClC6H4
Ph
4-ClC6H4
Ph
Ph
Ph
4-CH3OC6H4
4-CH3OC6H4
4-ClC6H4
4-CH3C6H4
Ph
Ph
n-C3H7
Ph
H
H
H
H
H
H
H
H
H
Ph
Ph
Ph
4-CH3C6H4
4-ClC6H4
Ph
Ph
Ph
Ph
9
Ph
10
11
12
n-C4H9
2-CH3C6H4
2-FC6H4
Ph
Ph
a Isolated yield. b Anti/syn ratio: 67/33, 63/37 and 68/32 for entries 1–3, respectively.
Preparation of the catalyst
(0.005 g). The reaction mixture was vigorously stirred for the
period of time listed in Table 2. The progress of the reaction
was monitored by TLC. When the reaction was complete,
g-Fe2O3@SiO2-PW could be placed on the side wall of the
reaction vessel with the aid of an external magnet, and water
was removed from the mixture to leave a residue (including
the product and catalyst). Then, the product was dissolved in
ethanol and the catalyst easily separated from the product by
attaching an external magnet onto the reaction vessel, followed
by decantation of the product solution. This solution was
concentrated at room temperature to generate the crude product.
The crude products were purified either by crystallization from
ethanol or by column chromatography on silica gel using
ethyl acetate/hexane as the eluent. All products were identified
by comparing of their spectral data with those of authentic
samples.16,19
The g-Fe2O3 NPs were synthesized by
a chemical co-
precipitation technique of ferric and ferrous ions in an alkaline
solution. FeCl2·4H2O (2.0 g) and FeCl3·6H2O (5.4 g) were
dissolved in water (20 mL) separately, followed by the two
iron salt solutions being mixed under vigorous stirring. An
NH4OH solution (0.6 M, 200 mL) was then added to the
stirring mixture at room temperature, immediately followed by
the addition of a concentrated NH4OH solution (25% w/w,
30 mL) to maintain the reaction pH between 11 and 12. The
resulting black dispersion was continuously stirred for 1 h at
room temperature and then heated to reflux for 1 h to yield a
brown dispersion. The MNPs were then purified by a four times
repeated centrifugation (4000–6000 rpm, 30 min), decantation
and redispersion cycle until a stable brown magnetic dispersion
(pH 9.2) was obtained (mean diameter = 65.5 nm). Coating
of a layer of silica on the surface of the g-Fe2O3 NPs was
achieved by pre-mixing (ultrasonic) a dispersion of the purified
NPs (8.5% w/w, 20 mL) obtained previously with ethanol (80
mL) for 1 h at 40 ◦C. A concentrated ammonia solution was
added and the resulting mixture stirred at 40 ◦C for 30 min.
Subsequently, TEOS (1.0 mL) was charged to the reaction vessel
and the mixture continuously stirred at 40 ◦C for 24 h. The silica-
coated NPs were collected using a permanent magnet, followed
by washing three times with ethanol, diethyl ether and drying in
a vacuum for 24 h (mean diameter = 79.3 nm).
Conclusions
The combination of the magnetic separation of g-Fe2O3@SiO2
NPs and the special properties of PW provide a good opportu-
nity to design and synthesize g-Fe2O3@SiO2-PW as an efficient
nanocatalyst possessing strong and sufficient acidic sites to be
responsible for excellent conversion values of products. More
importantly, g-Fe2O3@SiO2-PW is intrinsically heterogeneous
and recovered simply using an external magnetic field. The
isolated catalyst was reused at least five times without any loss
of its catalytic activity, suggesting that even a small amount of
the catalyst has stable acidic sites for several cycles of an organic
synthesis. As a consequence, MNPs could be an ideal support
for the immobilization of other HPA catalysts. In addition
to Mannich reactions, such environmentally benign catalysts
should find application in a wide range of other important acid-
catalyzed organic reactions.
0.7 g of PW was dissolved in 5 mL of dry methanol. This solu-
tion was added dropwise to a suspension of 1.0 g g-Fe2O3@SiO2
in methanol (50 mL) while being dispersed by sonication. The
mixture was heated at 70 ◦C for 72 h under vacuum while
being mechanically stirred to obtain g-Fe2O3@SiO2-PW NPs.
The catalyst was collected by a permanent magnet and dried in
a vacuum overnight.
This study is an initial attempt towards the synthesis of
nanomagnetically-recoverable HPA-based catalysts. Modifica-
tion of the synthetic approach, and the preparation, properties
and applications of other g-Fe2O3@SiO2-HPA NPs are in
progress, and will be reported in due course.
Typical procedure for Mannich reactions
Typically, to a mixture of aldehyde (1 mmol), amine (1 mmol),
ketone (2 mmol) and H2O (5 mL) was added the catalyst
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 2116–2122 | 2121
©

View Online
8 P. M. Rao, A. Wolfson, S. Kababya, S. Vega and M. V. Landau, J.
Catal., 2005, 232, 210; H. Firouzabadi and A. A. Jafari, J. Iranian
Chem. Soc., 2005, 2, 85.
9 A. L. Simpl´ıcio, J. M. Clancy and J. F. Gilmer, Int. J. Pharm., 2007,
336, 208.
10 E. Darezereshki, M. Ranjbar and F. Bakhtiari, J. Alloys Compd.,
2010, 502, 257.
Acknowledgements
The authors thank the Razi University Research Council and
Kermanshah Oil Refining Company for their support of this
work.
11 C. He, T. Sasaki, Y. Shimizu and N. Koshizaki, Appl. Surf. Sci., 2008,
254, 2196.
References
12 K. C. Barick, B. S. D. Ch. S. Varaprasad and D. Bahadur, J. Non-
Cryst. Solids, 2010, 356, 153.
1 N. E. Leadbeater and M. Marco, Chem. Rev., 2002, 102, 3217; B. M.
L. Dioos, I. F. J. Vankelecom and P. A. Jacobs, Adv. Synth. Catal.,
2006, 348, 1413.
2 D. Astruc, F. Lu and J. R. Aranzaes, Angew. Chem., Int. Ed., 2005,
44, 7852.
13 M. T. Pope, in Heteropoly and Isopoly Oxometalates, Springer-
Verlag, Berlin, 1983, p. 23.
14 S. L. C. Pinho, G. A. Pereira, P. Voisin, J. Kassem, V. Bouchaud, L.
Etienne, J. A. Peters, L. Carlos, S. Mornet, C. F. G. C. Geraldes, J.
Rocha and M. H. Delville, ACS Nano, 2010, 4, 5339.
15 R. L. McCormick, S. K. Boonrueng and A. M. Herring, Catal.
Today, 1998, 42, 145.
16 E. Rafiee, S. Eavani, F. Khajooei Nejad and M. Joshaghani,
Tetrahedron, 2010, 66, 6858.
17 W. B. Yi and C. Cai, J. Fluorine Chem., 2006, 127, 1515; W. Shen, L.
M. Wang and H. Tian, J. Fluorine Chem., 2008, 129, 267.
18 Z. Li, X. Ma, J. Liu, X. Feng, G. Tian and A. Zhu, J. Mol. Catal.
A: Chem., 2007, 272, 132; M. Kidwai, N. K. Mishra, V. Bansal, A.
Kumar and S. Mozumdar, Tetrahedron Lett., 2009, 50, 1355; G. P.
Lu and C. Cai, Catal. Commun., 2010, 11, 745.
3 Y. Zhang and C. Xia, Appl. Catal., A, 2009, 366, 141; L. Ma´mani,
A. Heydari and M. Sheykhan, Appl. Catal., A, 2010, 384, 122;
ˇ
ˇ
F. Sulek, Z. Knez and M. Habulin, Appl. Surf. Sci., 2010, 256,
4596.
4 M. Benaglia, A. Puglisi and F. Cozzi, Chem. Rev., 2003, 103, 3401;
F. Cozzi, Adv. Synth. Catal., 2006, 348, 1367.
5 Y. Piao, A. Burns, J. Kim, U. Wiesner and T. Hyeon, Adv. Funct.
Mater., 2008, 18, 3745.
6 T. Okuhara, N. Mizuno and M. Misono, Appl. Catal., A, 2001, 222,
63; E. Rafiee, S. Eavani, S. Rashidzadeh and M. Joshaghani, Inorg.
Chim. Acta, 2009, 362, 3555.
7 E. Rafiee, M. Joshaghani, S. Eavani and S. Rashidzadeh, Green
Chem., 2008, 10, 982.
19 H. Zeng, H. Li and H. Shao, Ultrason. Sonochem., 2009, 16,
758.
2122 | Green Chem., 2011, 13, 2116–2122
This journal is
The Royal Society of Chemistry 2011
©