Silica aerogel-iron oxide nanocomposites: Recoverable catalysts for the oxidation of alcohols with hydrogen peroxide

Article abstract of DOI:10.1007/s11243-011-9498-7

Various aerogels of silica gel doped with Fe₂O₃ were prepared by sol-gel method. They were calcined to produce nanoparticle solids. The nanosized mixed oxides were active in the oxidation of alcohols and produced carbonyl compounds in very good to excellent yields using hydrogen peroxide.

Full text of DOI:10.1007/s11243-011-9498-7

Transition Met Chem (2011) 36:521–530
DOI 10.1007/s11243-011-9498-7
Silica aerogel–iron oxide nanocomposites: recoverable catalysts
for the oxidation of alcohols with hydrogen peroxide
Shahla Masoudian
Alireza Aghaei
•
Hassan Hosseini Monfared
•
Received: 1 March 2011 / Accepted: 27 April 2011 / Published online: 12 May 2011
Ó Springer Science+Business Media B.V. 2011
Abstract Various aerogels of silica gel doped with Fe O3
2
applications, including via a formamide modified sol–gel
process [11], Fe O /SiO hollow spheres [12], iron oxide–
were prepared by sol–gel method. They were calcined to
produce nanoparticle solids. The nanosized mixed oxides
were active in the oxidation of alcohols and produced
carbonyl compounds in very good to excellent yields using
hydrogen peroxide.
2
3
2
silica nanocomposite films by spin coating as humidity
sensors [13], mesoporous iron oxide–silica aerogels as
catalysts for oxidation of NH and reduction of NO [14],
3
Fe–Si mixed oxide nanocomposites that contain iron
oxide as the major phase [15], nanocluster iron oxide–silica
aerogel catalysts for methanol partial oxidation [16],
Fe O –SiO composites, nanosized c-Fe O particles and
Keywords Oxidation ꢀ Alcohols ꢀ Hydrogen peroxide ꢀ
Sol–gel ꢀ Nanosized iron oxide-silica mixed oxides
2
3
2
2 3
study on the influence of the iron salts concentration [17],
iron oxide–silica nanocomposites with wet impregnation
[18], magnetic susceptibility studies of Fe O –SiO [19],
Introduction
2
3
2
Fe O /SiO nanocomposites based on fumed silica (S =
BET
2
3
2
2
Iron-based materials are of great technological importance
for their use in magnetic recording media, pigments, optics,
ferro-fluids, magneto-optic devices, biology, and in catal-
ysis [1–3]. Iron oxide/silica, Fe O /SiO , nanocomposites
265–337 m /g) [20], study of the magnetic properties of
the iron oxide particles hosted in silica aerogels pores [21],
silica aerogel–iron oxide nanocomposites as catalysts in
conjugate additions and in the Biginelli reaction [22]. In
spite of these studies, to the best of our knowledge, there is
no report on the oxidation of alcohols by silica–iron oxide
nanocomposites.
2
3
2
have recently attracted increasing attention due to their
interesting properties and applications in sensing, catalysis,
and magnetism [4–6]. Properties of Fe O /SiO materials
2
3
2
depend on the nature of the interaction between iron oxide
and silica [7]. However, for most applications, effective
composites must show the size of the particles in the
nanometer range, porous texture of high surface area, and
thermal stability [8–10].
Oxidation of alcohols to aldehydes and ketones is one
of the widely used chemical transformations in organic
synthesis as these products are important precursors or
intermediates in the synthesis of many drugs, vitamins, and
fragrances [23–26]. A number of methods are known for
alcohol oxidation [27]; however, the development of newer
methods and methodologies is gaining much attention
currently due to the significance of this reaction. Even
though various methods are now known, the most common
methods still use toxic, corrosive, and expensive oxidants
such as chromium(VI) and manganese complexes, strin-
gent conditions of high pressure or temperatures and use
of strong mineral acids [28, 29]. The quest for clean and
selective oxidation processes is continuing [30, 31]. Oxi-
dation over metal catalysts using eco-friendly oxidants
Composites with different oxide/silica ratios and forms
have synthesized with various methods for diverse
S. Masoudian ꢀ H. Hosseini Monfared (&)
Department of Chemistry, Zanjan University,
4
5195-313 Zanjan, Iran
e-mail: monfared@znu.ac.ir; monfared_2@yahoo.com
URL: http://www.znu.ac.ir/members/hosseini_hassan
A. Aghaei
Materials and Energy Research Center, 14155-4777 Tehran, Iran
123

5
22
Transition Met Chem (2011) 36:521–530
such as hydrogen peroxide and molecular oxygen are
preferred now for environmental and economical benefits.
In spite of the advantages of H O as a cheap and benign
transparent gel. Samples were then dried in an autoclave.
Finally, samples were calcined at 500 °C for 2 h or 10 min
and alternatively at 100 °C for 10 min in air flow.
2
2
oxidant, its catalytic application has some limitations due
to decomposition of H O by traces of transition metal ions
Table 1 presents a summary of all aerogels evaluated as
catalysts for this work. The weight percent of iron is given
as the sample composition. The 2 wt% of Fe (Fe O –SiO )
2
2
[32, 33]. Immobilization of iron on silica gel [34] or c-
alumina [35] considerably retards the non-productive
2
3
2
aerogel is in the form of reddish brown powder with BET
2
surface area of 555 m /g. Figure 1 is a field emission
decomposition of H O . Development in this area is based
2
2
on different strategies with the aim of designing selective,
stable, and high turnover catalytic systems.
scanning electron micrograph for the 2 wt% of Fe iron
oxide–silica aerogel after it was treated at 500 °C for 2 h
(a), at 500 °C for 10 min (b), and at 100 °C for 10 min (c).
The sample 1 silica-supported iron oxide calcined at
500 °C for 2 h (Fig. 1a) appears monolithic. Calcination
at lower temperature and/or for a short time resulted to
smaller size particles. In Fig. 1b and c, the SEM images of
nanoparticles of iron oxide–silica aerogel catalysts after
calcinations at 500 and 100 °C for 10 min are shown,
respectively. Uniform and tiny particles almost spherical in
shape, having sizes 63% between about 80–110 nm for
Fig. 1b and 70% between about 50–80 nm for Fig. 1c, are
observed.
The selectivity pattern and the formation rates of the
reaction products provide information on the surface nature
and dispersion of active sites at the molecular level [36].
Iron oxide itself exhibits strong acidic and weak redox
properties [37]. Most known iron oxide–silica catalysts
were prepared with iron nitrate or iron acetylacetonate and
with nitrogen-containing bases as gelation agents [14].
The present study aims to employ sol–gel synthesis of
iron alkoxide in isopropanol at the molecular level and to
investigate the catalytic reactivity of silica-coordinated iron
oxide aerogels. We wish to report an efficient protocol
in which H O has been used as the oxidizing agent in the
Figure 2 shows X-ray diffraction (XRD) patterns of
two aerogels after they were treated at 500 °C for 10 min.
According these patterns, hematite (a-Fe O ) was deter-
2
2
presence of nanoparticles of Fe O –SiO for the chemo-
3
2
2
selective oxidation of alcohols to their carbonyl com-
pounds in high chemical yield and short reaction times. We
have restricted ourselves to the use of hydrogen peroxide
as the terminal oxidant. This oxidant is cheap, clean, and
gives water as the sole byproduct [38].
2
3
mined as the main crystalline phase in the 4.35 and 9.6
wt% Fe (Fe O –SiO ) powders. The broadening near the
2
3
2
peak baseline suggests the presence of very small crystal-
lites (a few nanometers) [39]. XRD analyses indicated that
all aerogels were mainly amorphous, a feature that is a
reasonable indication that Fe O is well dispersed in the
2
3
Results and discussion
silica matrix [40]. The aerogel sample showed the presence
of traces of hematite in diffraction pattern of sample 3
(peaks at 31.88° and 45.58° in 2h). For the sample 3, the
broad peak at 2h = 21-278 is typically characteristic of
amorphous silica. No evidence indicated the presence of
maghemite or magnetite phases [41]. Empirically, iron
oxide made from an alkoxide precursor similar in nature
to the one we used exhibits a hematite phase at 500 °C
as determined by XRD analysis [42].
Characterization of the Fe O –SiO catalysts
2
3
2
A series of nanosize iron oxide–silica aerogel catalysts
were prepared via sol–gel methods. The homogeneous
samples obtained under acidic conditions did not show any
apparent gelation. In this case, the supercritical evaporation
of the solvent is a fundamental process for obtaining a
Table 1 Structural properties of aerogels
a
2
Surface area by BET (m /g)
b
b
Aerogel
Particle size by SEM (nm)
a
2
4
9
2 3 2
wt% of Fe (Fe O –SiO )
555
63% between 80 and 110 nm
c
7
0% between 50 and 80 nm
.35 wt% of Fe (Fe
2
O
3
–SiO
2
)
580
61% between 200 and 250 nm
(
Mean pore diameter 11.20 nm)
533
Mean pore diameter 9.18 nm)
2 3 2
.6 wt% of Fe (Fe O –SiO )
(
a
b
c
Weight percent-based sample composition
Measurements were taken for aerogels after being treated in air for 10 min at 500 °C
Calcined at 100 °C for 10 min
1
23

Transition Met Chem (2011) 36:521–530
523
Fig. 1 SEM micrographs for
the 2 wt% Fe (Fe O –SiO )
2 3 2
(
a)
precursor calcined a at 500 °C
for 2 h, b at 500 °C for 10 min,
and c at 100 °C for 10 min
2
2 3 2
wt% Fe (Fe O -SiO ), calcined at 500 ºC for 2 h.
(
b)
25
20
15
10
5
0
500 °C, 10 min
Particle size (nm)
-SiO ), calcined at 500 ºC for 10 min. 63%
of particles between 80 – 110 nm.
2
wt% Fe (Fe
2
O
3
2
2
wt% Fe (Fe
2 3 2
O -SiO ), calcined at 500 ºC for 10
2
min. BET surface area of 555 m /g
(c)
25
20
15
10
5
0
100 ºC, 10 min
20
30 40 50 60 70 80 90 100 110 120 130 140 150
Particle size (nm)
2
2 3 2
wt% Fe (Fe O -SiO ), calcined at 100 ºC for 10 min. 70%
2
min
wt% Fe (Fe
2
3 2
O -SiO ), calcined at 100 ºC for 10
of particles between 50 - 80 nm.
Fig. 2 XRD patterns of
aerogels after being treated at
(b) 800
(
a) 500
5
00 °C for 10 min, a 4.35 wt%
Fe (Fe –SiO ), and b 9.6
wt% Fe (Fe –SiO
6
00
2
O
3
2
31.75
2
O
3
2
)
45.53
400
2
00
0
0
1
0
20
30
40
50
10
20
30
40
50
60
70
80
2
/ °
2 / °
-
[SiO ] unit is observed around 1,074 cm in Fe O –SiO
1
FT-IR was used for the understanding of the relation
between silicon and iron in the metal oxide–silica com-
posites. The FT-IR spectrum of Fe O –SiO (1) is shown in
4
2
3
2
-
1
(1). The band at 797 cm
modes. The small shoulder band at 966 cm is assignable
is assignable m (Si–O–Si)
s
-
1
2
3
2
-
1
-1
to m(Si–OH). The band around 966 cm observed in
Fe O –SiO (1) sample is attributed to the incorporation of
Fig. 3. The bands at 3,431 and 1,624 cm (not shown) are
assignable to the m(O–H) mode of (H-bonded) and d(O–H)
mode of water molecules, respectively. The characteristic
2
3
2
3
?
the Fe ion into the silica structure forming Si–O–Fe
bonds. Note that the latter band normally appears as a sharp
antisymmetric (m ) Si–O–Si vibration band [43] of the
as
123

5
24
Transition Met Chem (2011) 36:521–530
catalyst 1 and only a low conversion (3%) was observed in
the presence of 40 lmol imidazole after prolonged (24 h)
reaction times (Table 2, entry 2). Cyclohexanol was oxi-
dized by H O /Fe O –SiO to give cyclohexanone as the
Fe
2
2
O
3
-SiO
2
wt% Fe
2
2
2
3
2
sole product. In addition to the catalyst, the presence of a
nitrogenous base (imidazole) as a cocatalyst increased dra-
matically the conversion of cyclohexanol from 24 to 80%
after 24 h (Table 2, entries 4 and 5).
Nature of oxidant and concentration
1
200
1000
800
600
400
-
1
Wavenumber / cm
Other parameters for suitable reaction conditions to achieve
the maximum oxidation of cyclohexanol such as the oxidant
concentration (moles of oxidant per moles of cyclohexanol),
solvent, and temperature of the reaction were investigated.
The effect of oxidant and concentration on the oxidation of
cyclohexanol is illustrated in Fig. 4. Different oxidant/cy-
clohexanol molar ratios (2:1, 4:1 and 8:1) were considered
while the ratio of cyclohexanol (1.0 mmol) to catalyst
2 3 2
Fig. 3 FT-IR spectrum of Fe O –SiO (1)
one in the absence of Fe O [11]. This may indicates that
2
3
3
?
the presence of Fe affects the behavior of surface (Si–
OH) groups. The incorporation of iron into the silica
framework leads to a lowering of the local symmetry of the
nearest neighbor [SiO ] unit [41]. This creates a new band
4
-
1
around 1,074 cm
which is lower in energy than the
(
0.01 g) in 3 mL of acetonitrile at 60 ± 1 °C was constant.
undisturbed [SiO ] unit vibration. The m (Si–O–Si) peak
4
as
-
1
In this study, the oxidation of cyclohexanol was examined
for the Fe O –SiO (1) material is seen at 1,074 cm
2
,
3
2
-
1
using various oxidants such as NaOCl, H O , tert-butylhy-
2
2
3
1 cm lower than that observed for the pure SiO aerogel
2
-
1
droperoxide (TBHP), and H O /urea (UHP). All the oxidants
2 2
at 1,105 cm [15]. The same behavior was observed in the
FT-IR spectra of different silicate materials [15, 44, 45].
The bending vibration band of O–Si–O bond [46] is found
except TBHP oxidized cyclohexanol to cyclohexanone with
00% selectivity. The product yield increased by increasing
1
-
1
the oxidant concentration. With 8 equivalents of oxidant,
maximum yield was observed for TBHP followed by H O
around 463 cm in Fe O –SiO (1). These FT-IR results
2
3
2
2
2
suggest that the iron atoms are not covalently bonded to
silicon in the silica network.
whereas NaOCl and H O /Urea were found to be less
2
2
effective for the oxidation. Aqueous hydrogen peroxide was
selected by considering its high selectivity, atom economy,
and environmentally benign properties.
Catalytic activity studies
The conversion of cyclohexanol increased with
increasing the amount of hydrogen peroxide in the reaction
Oxidation of cyclohexanol
Oxidation of cyclohexanol was performed as a model sub-
strate using 30% aqueous hydrogen peroxide in the presence
of Fe O –SiO (1) at 60 °C in acetonitrile. The results are
2
3
2
4
3
0
5
depicted in Table 2. It is noteworthy that in blank experi-
ments (entries 1–3), no significant oxidation was observed
under similar reaction conditions in the absence of H O and
H2O2
TBHP
UHP
30
2
2
NaOCl
25
20
15
10
5
0
Table 2 Catalytic oxidation of cyclohexanol by H
Entry Catalyst Cocatalyst Oxidant Conversion (%)
After 3 h After 24 h
2 2
O
1
2
3
4
5
Fe
None
SiO
2
O
3
–SiO
2
Imidazole None
0
0
0
3
Imidazole
Imidazole
None
H
H
H
H
O
2 2
O
2 2
O
2 2
O
2 2
0
6
2
4
8
2
Fe
2
O
O
3
–SiO
2
2
9
24
80
Oxidant / mmol
Fe
2
3
–SiO
Imidazole
27
Fig. 4 Effects of concentration and nature of oxidant on the oxidation
of cyclohexanol by Fe –SiO (1). cyclohexanol 1.0 mmol, catalyst
Fe –SiO (1) 10 mg (3.6 lmol Fe ), toluene 0.1 g, imidazole
40 lmol, acetonitrile 3 mL, temperature 60 ± 1 °C and 5 h
Cyclohexanol 1.0 mmol, catalyst Fe
2
O
3
–SiO
Fe ), toluene 0.1 g, imidazole 40 lmol, acetonitrile 3 mL, H
4 mmol at 60 ± 1 °C
2
(1) 10 mg (3.6 lmol
O
2 3
2
3?
3?
O
2 2
2
O
3
2
1
1
23

Transition Met Chem (2011) 36:521–530
525
mixture. When H O /substrate mole ratio was about 14, the
2
Effect of the temperature
2
maximum conversion of 42% was obtained (not shown in
Fig. 4), showing that ca. 13 equivalents of the oxidant are
decomposed in the non-productive way. This is, however,
much less than the decomposition of hydrogen peroxide by
unsupported transition metals [32, 33]. Oxidant require-
ment of this order in epoxidation reactions is documented
in the literature [47, 48].
In order to get the best reaction temperature, the reaction
mixture was stirred at various temperatures (for results
see Fig. 6). The cyclohexanone yield increased when the
reaction was carried out at higher temperatures. With
Fe O –SiO (1) as catalyst, a maximum yield of 45% was
2 3 2
obtained after 5 h at 80 °C. The yields were almost to the
same range at 60 and 80 °C.
Effect of co-solvent
Effect of the cocatalyst
Figure 5 illustrates the influence of the co-solvent in the
catalytic oxidation of cyclohexanol by Fe O –SiO (1).
2
Hydrogen peroxide is a widely used oxidant with high
active oxygen content [50, 51], but it is a rather slow
oxidizing agent in the absence of activators. According to
some reports, the addition of some alkaline species (e.g.,
3
2
Ethyl acetate, dichloromethane, acetonitrile, chloroform,
dimethyl formamide, and carbon tetrachloride were used as
solvents. Of the various solvents examined, acetonitrile was
found to be the most suitable solvent for the reaction. The
highest conversion was obtained in acetonitrile, 14% in the
absence of imidazole and 42% in the presence of imidazole
NaOH, KOH, NaHCO ) into the reaction system was
3
beneficial to alkene oxidation [52–54]. The efficiency of
iron-catalyzed oxygenation of hydrocarbons by hydrogen
peroxide increases in the presence of amines [55]. In
addition, the significant effect of imidazole as a cocatalyst
in increasing substrate oxidation by non-heme systems has
been reported [56]. The effect of ligands like imidazole in
improving Mn–porphyrin mediated epoxidation is espe-
cially noteworthy [57, 58], and the role of imidazole in
(
see below). It was observed that the catalytic activity of the
catalyst (1) decreased in order acetonitrile [ CCl [ chlo-
4
roform [ DMF [ dichloromethane [ ethyl acetate in the
absence of imidazole and acetonitrile [ dichlorometh-
ane [ DMF [ chloroform [ CCl & ethyl acetate in the
4
presence of imidazole. The reactivity of catalyst 1 in other
solvents was very much lower than acetonitrile (dielectric
catalytic oxidation by Mn–porphyrin/H O systems has
2 2
constant e/e = 36 and donor number DN = 14.1 [49]) and
been established [59]. It has been proved that both direct
coordination of imidazole to the active manganese center
(proximal effect) and its action as a base (distal effect) may
facilitate the heterolytic cleavage of the O–O bond in
0
was almost to the same extent. From Fig. 5, one could
see that co-solvents with lower donation ability and high
polarities seemed to favor this reaction. Solvents with
III
high coordinating property such as DMF (e/e = 36.7 and
[(porphrin)Mn –O–OH] species.
0
DN = 24) inhibited this reaction. When using DMF/imid-
azole as solvent, 10% of cyclohexanone product was
gained.
An additive like NaHCO
hexanol oxidation by Fe
base like imidazole increased the conversion from 24 to
0% for cyclohexanol (Table 2, entries 4 and 5). The effect
3
was not effective in cyclo-
O
–SiO (1), but addition of a
2
3
2
8
5
4
3
2
1
0
0
0
0
0
50
40
30
20
10
0
8
0°C
without Im
with Im
60°C
4
0°C
1
5°C
0
Ethyl
acetate
CCl
4
CHCl
3
DMF
CH
2
Cl
2
3
CH CN
1
2
3
4
5
6
Time / h
Fig. 5 Effect of solvent on the oxidation of cyclohexanol by Fe
SiO (1). Reaction conditions: cyclohexanol 1.0 mmol, catalyst
Fe
2 3
O –
2
Fig. 6 Effect of temperature on conversion of cyclohexanol by
Fe –SiO (1). Reaction conditions: cyclohexanol 1.0 mmol, cata-
lyst Fe –SiO
40 lmol, and acetonitrile 3 mL, H
3?
2
O
3
–SiO
mL, H
0 ± 1 °C
2
(1) 10 mg (3.6 lmol Fe ), toluene 0.1 g, solvent
14 mmol, imidazole (Im) 40 lmol, and temperature
2
O
3
2
3?
3
6
O
2 2
O
2 3
2
(1) 10 mg (3.6 lmol Fe ), toluene 0.1 g, imidazole
14 mmol
2 2
O
123

5
26
Transition Met Chem (2011) 36:521–530
5
4
3
2
1
0
0
0
0
0
0
30
20
10
0
4
0
Imidazole (µmol)
7
4
20
50
1
0
1
2
3
4
5
1
2
3
4
5
6
Recycling
Time / h
Fig. 8 Recycling studies performed with catalyst 2 wt% of Fe
(Fe –SiO ) (1) in the oxidation of cyclohexanol by hydrogen
peroxide, after 3 h of reaction time. Reaction conditions: catalyst
10 mg, cyclohexanol 1 mmol, CH CN 3 mL, H 14 mmol, and
temperature 60 ± 1 °C. Yields were based on the starting
cyclohexanol
Fig. 7 Effect of imidazole concentration on oxidation of cyclohexanol
by Fe –SiO (1). Reaction conditions: cyclohexanol 1.0 mmol,
catalyst Fe
acetonitrile 3 mL, H
2
O
3
2
O
2 3
2
2
O
3
–SiO
2
(1) 10 mg (3.6 lmol Fe
3
?), toluene 0.1 g,
3
2 2
O
2
O
2
14 mmol, and temperature 60 ± 1 °C
of the imidazole concentration on the oxidation of cyclo-
hexanol is illustrated in Fig. 7. Different imidazole
amounts (20, 40, 74, and 150 lmol) were considered while
keeping the fixed amount of cyclohexanol (1 mmol) and
composite (0.01 g) in 3 mL of acetonitrile at 60 ± 1 °C.
Increasing the imidazole amount from 20 to 40 lmol
increased the conversion from 17.8 to 42%. A higher
concentration of imidazole reduced the conversion. The
maximum cyclohexanol conversion was obtained with
an amount of 40 lmol. We used an imidazole amount of
confirm this finding in another experiment, after 5 h vig-
orous stirring of the oxidation mixture, the solid catalyst
was removed by filtration. Cyclohexanol and H O were
2 2
added to the filtrate, and the reaction was followed for 5 h.
No product was detected. This experiment also confirmed
that Fe O –SiO (1) is stable and Fe O does not leach.
2 3 2 2 3
The absence of Fe in the solution through leaching of 1 was
also studied by measuring iron in solution. The investiga-
tion of the resulting solution by UV–Vis spectroscopy in
the presence of thiocyanate showed no detectable iron. This
ensured that no leaching of the active supported Fe(III)
components occurred. Thus, the obtained catalytic results
derive exclusively from the heterogeneous catalyst. This
feature of the catalyst made it possible to remove the stable
catalyst by simple filtration at the end of reaction. In order
to investigate the stability and the possibility of several
recycling runs for catalyst 1, the solid catalyst was sepa-
rated from the reaction mixture by filtration after the
reaction, washed exhaustively with acetonitrile, and used
again in a fresh reaction. The catalyst was recycled five
times for cyclohexanol oxidation, and the results are seen
in Fig. 8. In general, no substantial loss in the activity of
the catalyst was observed compared with that of fresh
sample, as shown in Fig. 8. The reused catalyst displayed
consistent reactivity and selectivity.
4
0 lmol in further studies. In addition, in oxidation of
cyclohexanol by Fe O –SiO /H O in the presence of the
2
3
2
2 2
radical trap tert-butanol (10 equiv), the cyclohexanol
conversion decreased from 33 to 18%. One could conclude
that some part of the oxidation reaction proceeds via the
hydroxyl radical. It has been assumed that imidazole
facilitates proton abstraction from a H O ligand coordi-
2
2
nated to the iron and facilitates the catalytic cycle of
generating hydroxyl radicals (and also the other possible
active species) from the hydrogen peroxide. Hydroxyl
radicals then attack alcohol molecules finally yielding the
ketone [55]. We can assume that the role of imidazole
is proton transfer (via protonation–deprotonation steps of
the N-atom). The direct coordination of imidazole to the
III
active Fe center is excluded since the solvents with high
coordinating property such as DMF did not improve the
oxidation yield (Fig. 5).
Effect of catalyst iron content
Catalyst recycling
Table 3 presents a summary of all composites evaluated as
catalysts for this work. The weight percent of iron is given
as the sample composition. Increasing the concentration
from 2 up to 9.6 wt% increased the conversion from 42 to
59.9% but the amount of cyclohexanone produced per mole
of Fe (TON) decreased. An increase in catalyst loading
Tests on catalyst recycling were carried out on Fe O –SiO
2
2
3
(1). The solutions remained colorless during the course
of oxidations, and there was no evidence for leaching of
the catalyst Fe O from Fe O –SiO (1) into solution. To
2
3
2
3
2
1
23

Transition Met Chem (2011) 36:521–530
527
Table 3 Oxidation of cyclohexanol with hydrogen peroxide under different iron content catalysts
a
Entry
Calcination
Catalyst
% conversion
TON
1
2
3
4
500 °C, 10 min
500 °C, 2 h
500 °C, 10 min
500 °C, 10 min
2 wt% of Fe–Fe
2 wt% of Fe–Fe
2
O
O
3
–SiO
2
2
42
117
99
2
3
–SiO
35.6
53
4.35 wt% of Fe–Fe
9.6 wt% of Fe–Fe
2
O
3
–SiO
2
68
2
3
O –SiO
2
59.9
35
Cyclohexanol 1.0 mmol, catalyst 10 mg, toluene 0.1 g, imidazole 40 lmol, acetonitrile 3 mL, H
reaction time 5 h
O
2 2
14 mmol, temperature 60 ± 1 °C, and
a
TON turnover number = number of moles of product formed per mole of Fe in the catalyst
(
Table 3, entry 4) did not cause a lowering of yield or
non-activated alcohols were converted to the correspond-
ing carbonyls efficiently. Primary alcohols all worked well
under our conditions. Methanol was oxidized quantitatively
to formic acid (Table 4, entry 1) and ethanol to acetalde-
hyde (Table 4, entry 2). The oxidation products of primary
alcohols 1-pentanol and 1-heptanol (Table 4, entries 4 and
5) were the corresponding secondary alcohol and methyl
alkyl ketones. The activity of secondary alcohol 2-heptanol
is more than primary 1-heptanol (Table 4, entries 5 and 6).
Conversion was increased for the chloro-substituted benzyl
alcohol (Table 4, entry 10). Alicyclic alcohols such as
cyclohexanol (Table 4, entry 7) and cyclooctanol (entry 8)
were also successfully converted to their respective cyclic
ketones with 100% selectivity. The oxidation products of
2-heptanol were the corresponding carbonyl product and an
alcohol with rearrangement of the starting material. The
unusual alcohol products obtained from the rearrangement
of 1-pentanol, 1-heptanol, and 2-heptanol were also con-
firmed by NMR and GC by using authentic samples. For
these substrates, reactions were carried out on a larger scale
chemoselectivity.
Oxidation of alcohols
The oxidation potential of the composite was further
explored by performing oxidation of various alcohols under
the same reaction conditions which proved to be the best
for cyclohexanol. The Fe O –SiO nanocomposite cata-
2
3
2
lysts were found to exhibit excellent activity for the oxi-
dation of a variety of alcohols using hydrogen peroxide as
oxidant at 80 °C in acetonitrile, and the results are shown
in Table 4. The estimated turnover numbers (TON) have
also been given for each investigated alcohol. The selec-
tivity to the corresponding carbonyl product was between
5
4 and 100%. Even though the best conditions and there-
fore the maximum conversions obtainable for individual
alcohols were not optimized, the method was found to be
useful in the oxidation of various primary, secondary, ali-
phatic, alicyclic, and aromatic alcohols. Both activated and
2 3 2 2 2
Table 4 Oxidation of various alcohols catalyzed by (9.6 wt% Fe) Fe O –SiO /Imidazole/H O
a
b
Entry
Substrate
Alcohols conv. (%)
Product (selectivity, %)
TON
d
1
2
3
4
5
6
7
8
9
MeOH
d
EtOH
100
c
Formic acid (100)
Acetaldehyde (100)
Acetone (100)
58
58
58
58
49
58
46
31
55
58
58
100
c
2-Propanol
100
c
1-Pentanol
100
2-Pentanol (22), 2-pentanone (78)
2-Heptanol (46), 2-heptanone (54)
3-Heptanol (31), 2-heptanone (69)
Cyclohexanone (100)
1-Heptanol
85
d
100
2-Heptanol
Cyclohexanol
Cyclooctanol
Benzyl alcohol
4-Chlorobenzylalcohol
Cinnamyl alcohol
79
54
Cyclooctanone (100)
c
94
Benzaldehyde (100)
c
1
1
0
1
100
4-Chlorobenzaldehyde (100)
Cinnamaldehyde (100)
c
100
Catalyst (9.6 wt% Fe) Fe
5
2
O
3
–SiO
2 2 2
10 mg (17.2 lmol), imidazole 40 lmol, alcohol 1.0 mmol, acetonitrile 3 mL, H O 14 mmol, reaction time
h, and reaction temperature 80 ± 1 °C. Conversions and yields are based on the starting substrate
a
All products were identified by comparison of their physical and spectral data with those of authentic samples or GC-Mass
TON turnover number = number of moles of product formed per mole of Fe in the catalyst sample
Reaction time = 4 h
b
c
d
3
mmol
123

5
28
Transition Met Chem (2011) 36:521–530
2
2
1
1
wt% Fe, at 60 ºC
wt% Fe, at 80 ºC
2 wt% Fe, at 80 ºC, 3 mmol alcohol
2 wt% Fe, at 80 ºC
or primary alcohols that have been reported by catalyst
SiW Zn [61] was not observed.
11
100
Conclusion
80
6
0
Silica aerogel–iron oxide nanocomposites have been pre-
pared and are very efficient catalysts for chemoselective
oxidation of various alcohols with aqueous H O . The
4
0
2
0
2
2
presence of a cocatalyst like imidazole increased the cat-
alyst efficiency enormously. The catalysts are very stable
and could be easily recovered and reused without loss of
activity. Although iron(III) ions decompose H O instantly
0
12 wt% Fe, at 80 ºC
1
2 wt% Fe, at 80 ºC, 3 mmol alcohol
2
wt% Fe, at 80 ºC
wt% Fe, at 60 ºC
2
2
2
[
32, 33], one remarkable property of these supported Fe O
2 3
catalysts is their ability to retard H O decomposition to a
2
2
great extent. Silica aerogel–iron oxide nanocomposite
systems are highly active, selective, simple, and safe to
operate. We anticipate that the present heterogeneous cat-
alytic systems would be potentially applicable to practical
organic synthesis.
Fig. 9 Effects of the temperature, the extent of the composite iron
content, and starting alcohol concentration on alcohol conversion. For
reaction conditions see Table 4
and the products were isolated as further confirmation of
identity and yield by NMR.
Experimental
In Fig. 9, effects of the reaction temperature, amount of
the composite iron content, and starting alcohol concentra-
tion on alcohol conversion are shown diagrammatically. The
same effects and trend that was observed for cyclohexanol
Materials
Anhydrous FeCl and sodium from Merck and tetraethyl
3
(
Fig. 6; Table 3) are also seen for various alcohols.
orthosilicate (TEOS) from Fluka were used as starting
materials. Other used chemicals were isopropanol, ethanol,
nitric acid, and acetylacetone (Fluka). The oxidant was
30% hydrogen peroxide (Merck). Substrates and solvents
were purchased from Merck. All chemicals were of highest
grade and used without further treatment.
A comparison of Fe O -SiO catalytic system with
2
3
2
previously reported systems (Table 5) shows that the
conversions (especially primary alcohols) are much higher
than in the other systems like vanadium phosphorus oxide
catalyst [60], furthermore over-oxidation of benzyl alcohol
2 2
Table 5 Comparison of the activity of heterogeneous transition metal catalysts in the oxidation of benzyl alcohol with H O
Catalyst
Product
Time (h)
Conv. (%)
Selectivity (%)
Ref.
Fe
2
O
3
–SiO
2
Benzaldehyde
Benzaldehyde
Benzaldehyde
Benzaldehyde
Benzaldehyde
Benzoic acid
Benzaldehyde
Benzaldehyde
Benzaldehyde
Benzaldehyde
3
4
94
45
27
77
53
86
30
96
33
18
100
100
This work
[62]
Cr(III)-Schiff base/MCM-41
a
[
Cu(tetraazamacrocycle)]/NaY
6
96.3
[63]
Ammonium molybdate
5
99
86
[64]
b
Alkali-treated ZSM-5 zeolite
4
[65]
c
[
Co((OH)
YCu(dmgH)
(n-C N]
Nanoiron oxide (Nano-c-Fe
2
-salen)]/MWNTs
6
100
100
37
[66]
d
2
8
[67]
e
[
4
H
)
9 4
4
H[PW11Ni(H
2
O)O39
]
12
12
24
[68]
2
O
3
)
97
[69]
Vanadium silicate xerogel (V
2
O
5
-SiO
2
)
100
[70]
a
[
4 2 4
Cu{Me (Bzo) [14]tetraeneN }]-NaY
b
c
d
e
2
2 2 3
5ZSM(AT-0.5), where 25 denotes the SiO /Al O ratio and 0.5 denotes the alkali-treatment time
0
H
2
[(OH)
2
-salen)] = N,N -bis(4-hydroxysalicylidene)-ethylene-1,2-diamine), MWNTs modified multiwall carbon nanotubes
dmgH dimethylglyoxime synthesized in situ in Y zeolite
Nickel-substituted polyoxotungstate
1
23

Transition Met Chem (2011) 36:521–530
Methods
529
electron microscopy (SEM), X-ray powder diffraction
(XRD), and N adsorption isotherm.
2
Specific surface area was determined using a single point
Brunauer–Emmett–Teller (BET) method with N at 298 K
2
Catalytic oxidation procedures
on a Belsorp Mini G apparatus. X-ray powder diffraction
analysis was performed using an X’Pert diffractometer
operating with Cu Ka Ni-filtered radiation, a graphite
monochromator, and a proportional counter with a pulse
height discriminator. IR spectra were recorded in KBr disks
Catalytic oxidation of alcohols was performed under air
(atmospheric pressure) in 25 mL round-bottomed flasks
equipped with a condenser and a small magnetic bar. The
catalyst (0.01 g), substrate (1 mmol), internal standard
(toluene, 0.10 g), solvent (acetonitrile, 3 mL), and aqueous
H O 30% (2–14 mmol) were placed in the flask. The
1
with a Matson 1000 FT-IR spectrophotometer. H NMR
spectra of oxidation products in DMSO-d solution were
6
2 2
recorded on a Bruker 250 MHz spectrometer. The reaction
products of oxidation were determined and analyzed by HP
Agilent 6890 gas chromatograph equipped with a HP-5
capillary column (phenyl methyl siloxane 30 m 9 320 lm
reaction flasks were placed in an oil bath at constant tem-
perature (15–80 °C), while an external magnetic stirrer
ensured agitation. Reactions were carried out for 5 h, under
magnetic stirring. At appropriate intervals, aliquots were
removed and quantified immediately by GC. The oxidation
products were identified by comparing their retention times
9
0.25 lm) and gas chromatograph–mass spectrometry
Hewlett-Packard 5973 Series MS-HP gas chromatograph
(
1
with a mass-selective detector). The exact amount of iron in
the composites was determined by ICP model ARL 3410.
with those of authentic samples or alternatively by H NMR
and GC-mass analyses. Yields were based on the added
alcohols and were determined by a calibration curve. Control
reactions were carried out in the absence of catalyst, under
the same conditions as the catalytic runs. No products were
detected.
Synthesis of Fe(OPr)3
Sodium (0.25 g) was dissolved in 50 mL of anhydrous iso-
propanol under N atmosphere with vigorous stirring. When
2
Each experiment was repeated three times and conver-
sions for single experiments varied within ± 2%.
complete dissolution was attained, 0.162 g (1.0 mmol)
of FeCl was added to the solution and the mixture was
3
Acknowledgments The authors thank the Zanjan University for the
financial support of this study.
stirred for 120 min. A clear solution of Fe(OPr) in isopro-
3
panol was separated by centrifugation, dried in vacuum,
and the obtained solid was dissolved in 20 mL of
isopropanol.
References
Preparation of aerogels under acidic conditions
1
. Cornell RM, Schwertmann U (1996) The iron oxides. VCH,
Germany
2
3
. Raj K, Moskovitz R (1990) J Magn Magn Mater 85:223
. Levy L, Sahoo Y, Kim KS, Berrgey EJ, Prasad PN (2002) Chem
Mater 14:3715–3721
In a typical procedure, a solution of acetylacetone (acacH,
2
.0 mmol), i-PrOH (3 mL), and Fe(OPr) (20.0 mL, acac/
3
Fe(OPr) 2:1 molar ratio) was heated under reflux for 1 h
3
4. Machala L, Zboril R, Gedank A (2007) J Phys Chem B 111:4003–
4018
under N with stirring. The modified iron precursor and the
2
5
. Nakamura T, Yamada Y, Yano K (2006) J Mater Chem 16:
417–2419
. Ennas G, Musinu A, Piccaluga G, Zedda D, Gatteschi D, San-
gregorio C, Stanger JL, Concas G, Spano G (1998) Chem Mater
10:495–502
support precursor TEOS (9.8 mL) were dissolved in i-PrOH
2
(
(
10 mL). A hydrolyzingsolution consisting of distilled water
6 mL) and concentrated nitric acid (0.27 mL) diluted in
6
i-PrOH (6 mL) was added dropwise to the alkoxides solution
under vigorous stirring. After 90 min, the final solution was
transferred into a Teflon-lined autoclave for solvent evapo-
ration. The autoclave was heated to 150 °C and kept at the
final temperature for 5 h to ensure complete thermal equil-
ibration. All as-synthesized gels were divided into small
particles and calcined at 500 °C in air for 10 min to obtain 2
7
8
. Marchetti SG, Cagnoli MV, Alvarez AM, Gallegos NG, Bengoa
JF, Yeramian AA, Mercader RC (1997) J Phys Chem Solids 58:
2
. Yi DK, Lee SS, Ying JY (2006) Chem Mater 18:2459–2461
119–2125
9. Sato S, Takahashi R, Sodesawa T, Tanaka R (2003) Bull Chem
Soc Jpn 76:217–223
0. Solzi M (1993) Fundamental properties of nanostructured mate-
rials. D. Fiorani and G. Sberveglieri, London
1
wt% of Fe (sample 1 Fe O –SiO ). The overall yield per
2
3
2
11. Khalil KMS, Makhlouf SA (2008) Appl Surf Sci 254:3767–3773
12. Dai Z, Meiser F, M o¨ hwald H (2005) J Colloid Interface Sci
batch production was about 2.79 g, 89 wt%. The same
conditions but increasing the amount of iron precursor was
chosen for the preparation of samples with 4.35 wt% (sample
2
88:298–300
1
1
3. Tongpool R, Jindasuwan S (2005) Sens Actuators B 106:523–528
4. Fabrizioli P, B u¨ rgi T, Baiker A (2002) J Catal 206:143–154
2 Fe O –SiO ) and 9.6 wt% of Fe (sample 3 Fe O –SiO )
2 3 2 2 3 2
and characterized by the techniques such as scanning
15. Clapsaddle BJ, Gash AE, Satcher JH Jr, Simpson RL (2003) J
Non-Cryst Solids 331:190–201
123

5
30
Transition Met Chem (2011) 36:521–530
1
1
6. Wang CT, Ro SH (2005) Appl Catal A 285:196–204
43. Lippincott ER, Van Volkenburg A, Weir CE, Bunting EN (1958)
J Res Natl Bur Stand 61:61–70
44. Cui H, Zayat M, Levy D (2009) J Alloys Compd 474:292–296
45. Cui H, Zayat M, Levy D (2005) Chem Mater 17:5562–5566
46. Niznansky D, Rehspringer JL (1995) J Non-Cryst Solids 180:
191–196
7. Savii C, Popovici M, Enache C, Subrt J, Niznansky D, Bakardzieva
S, Caizer C, Hrianca I (2002) Solid State Ionics 151:219–227
8. Alcal a´ MD, Real C (2006) Solid State Ionics 177:955–960
9. Battishaa IK, Afify HH, Hamada IM (2005) J Magn Magn Mater
1
1
2
92:440–446
2
0. Bogatyrev VM, Gu n´ ko VM, Galaburda MV, Borysenko MV,
Pokrovskiy VA, Oranska OI, Polshin EV, Korduban OM, Leboda
R, Skubiszewska-Zieba J (2009) J Colloid Interface Sci 338:
47. Lane BS, Vogt M, DeRose VJ, Burgess K (2002) J Am Chem Soc
124:1946–1954
48. Ueno S, Yamaguchi K, Yoshida K, Ebitani K, Kaneda K (1998)
Chem Commun 295–296
3
76–388
2
2
1. Fern a´ ndez van Raap MB, Sanchez FH, Leyva AG, Japas ML,
Cabanillas E, Troiani H (2007) Physica B 398:229–234
2. Mart ´ı nez S, Meseguer M, Casas L, Rodr ´ı guez E, Molins E,
Moreno-Ma nˇ as M, Roig A, Sebasti a´ n RM, Vallribera A (2003)
Tetrahedron 59:1553–1556
49. Huheey JE (1983) Inorganic chemistry: principles of structure
and reactivity, 3rd edn. Harper and Row Publisher Inc, New
York, p 340
50. Sheldon RA (1993) Top Curr Chem 164:21–34
51. Schumb WC, Satterfield CN, Wentworth RL (1955) Hydrogen
peroxide. Reinhold, New York
52. Rode CV, Nehete UN, Dongare MK (2003) Catal Commun
4:365–369
53. Adam W, Rao PB, Degen HG, Saha-Mller CR (2000) J Am Chem
Soc 122:5654–5655
54. Lu XH, Xia QH, Zhan HJ, Yuan HX, Ye CP, Su KX, Xu G
(2006) J Mol Catal A Chem 250:62–69
55. Shul’pin GB, Golfeto CC, S u¨ ss-Fink G, Shul’pina LS, Mandelli
D (2005) Tetrahedron Lett 46:4563–4567 and references therein
56. Pouralimardan O, Chamayou AC, Janiak C, Hosseini-Monfared
H (2007) Inorg Chim Acta 360:1599–1608
57. Gross Z, Ini S (1997) J Org Chem 62:5514–5521
58. Collman JP, Brauman JI, Fitzgerald JP, Hampton PD, Naruta Y,
Michida T (1988) Bull Chem Soc Jpn 61:47–57
59. Battioni P, Renaud JP, Bartoli JF, Reina-Artiles M, Fort M,
Mansuy D (1988) J Am Chem Soc 110:8462–8470
60. Pillai UR, Sahle-Demessie E (2004) Appl Catal A 276:139–144
61. Wang J, Yan L, Li G, Wang X, Ding Y, Suo J (2005) Tetrahedron
Lett 46:7023–7027
2
2
2
2
2
2
2
3. Sheldon RA, Kochi JK (1981) Metal-catalyzed oxidation of
organic compounds. Academic Press, New York
4. Hudlicky M (1990) Oxidations in organic chemistry. ACS,
Washington, DC
5. Pybus DH, Sell CS (eds) (1999) The chemistry of fragrances.
RSC Paperbacks, Cambridge
6. Fey T, Fischer H, Bachmann S, Albert K, Bolm C (2001) J Org
Chem 66:8154–8159
7. Sheldon RA, Arends IWCE, Dijkman A (2000) Catal Today
5
7:157–166
8. Griffith WP, Joliffe JM (1991) Dioxygen activation and homo-
geneous catalytic oxidation. Elsevier, Amsterdam
9. Ley SV, Norman J, Griffith WP, Marsden SP (1994) Synthesis
2
5:639–666
3
3
0. Arends IWCE, Sheldon RA (2002) Top Catal 19:133–141
1. Warner PT, Warner JC (1998) Green chemistry theory and
practice. Oxford University Press, New York
3
3
3
2. Cotton FA, Wilkinson G (1988) Advanced inorganic chemistry,
5
th edn. Wiley, New York, p 458
3. Hosseini Monfared H, Amouei Z (2004) J Mol Catal A Chem 217:
61–164
4. Hosseini-Monfared H, Ghorbani M (2001) Monatsh Chem 132:
89–992
62. Wang X, Wu G, Wei W, Sun Y (2010) Transition Met Chem
35:213–220
63. Kumar V, Thankachan BPP, Prasad R (2010) Appl Catal A
381:8–17
1
9
3
3
3
3
3
5. Hosseini-Monfared H, Ghadimi M (2003) J Chem Res S 313–314
6. Tatibou e¨ t JM (1997) Appl Catal A Gen 148:213–252
7. Wang CT, Willey RJ (2001) J Catal 202:211–219
8. Noyori R, Aoki M, Sato K (2003) Chem Commun 16:1977–1986
9. Cannas C, Gatteschi D, Musinu A, Piccaluga G, Sangregorio C
64. Chaudhari MP, Sawant SB (2005) Chem Eng J 106:111–118
65. Jia A, Lou LL, Zhang C, Zhang Y, Liu S (2009) J Mol Catal A
Chem 306:123–129
66. Salavati-Niasari M, Bazarganipour M (2009) Transition Met
Chem 34:605–612
(
1998) J Phys Chem B 102:7721–7726
67. Xavier KO, Chacko J, Mohammed Yusuff KK (2004) Appl Catal
A 258:251–259
68. Kanjina W, Trakarnpruk W (2009) Silpakorn U Science & Tech J
3:7–12
4
4
0. Kim GJ, Kim SH (1999) Catal Lett 57:139–143
1. Fabrizioli P, B u¨ rgi T, Burgener M, Van Doorslaer S, Baiker A
(
2002) J Mater Chem 12:619–630
4
2. Armelao L, Granozzi G, Tondello E, Colombo P, Principi G,
Lottici PP, Antonioli G (1995) J Non-Cryst Solids 192–193:
69. Shi F, Tse MK, Pohl MM, Radnik J, Br u¨ ckner A, Zhang S, Beller
M (2008) J Mol Catal A Chem 292:28–35
70. Neumann R, Levin-Elad M (1995) Appl Catal A 122:85–97
4
35–438
1
23