Catalytic activity of iron-substituted polyoxotungstates in the oxidation of aromatic compounds with hydrogen peroxide

Article abstract of DOI:10.1007/s00706-010-0370-9

The tetrabutylammonium (TBA) salts of Keggin-type polyoxotungstates of the general formula [XW₁₁Fe^III(H₂O)O₃₉] ^n-, where X = P, B or Si, were evaluated as catalysts in the oxidation, under mild conditions, of ethylbenzene, cumene, p-cymene and sec-butylbenzene with aqueous H₂O₂ in CH₃CN at 80 °C. The influence of various factors, such as the substrate/catalyst molar ratio, the amount of oxidant added or the reaction time, was investigated in a systematic way. Generally, the system exhibited moderate conversion, with good selectivity towards the corresponding acetophenone and hydroperoxide. In order to understand the reaction pathways, the oxidation of several products and presumed intermediates was also carried out in the presence of TBA ₄[PW₁₁Fe(H₂O)O₃₉]?2H ₂O. Under the conditions used, the oxidation of styrene and styrene derivatives gave rise mainly to carbon-carbon double-bond cleavage, affording the corresponding products in very high yields (81-87%). Possible reaction pathways are presented.

Full text of DOI:10.1007/s00706-010-0370-9

Monatsh Chem (2010) 141:1223–1235
DOI 10.1007/s00706-010-0370-9
ORIGINAL PAPER
Catalytic activity of iron-substituted polyoxotungstates in the
oxidation of aromatic compounds with hydrogen peroxide
•
•
´
Ana C. Estrada Mario M. Q. Simo˜es
•
Isabel C. M. S. Santos M. Grac¸a P. M. S. Neves
•
•
Jose A. S. Cavaleiro Ana M. V. Cavaleiro
´
Received: 21 April 2010 / Accepted: 3 August 2010 / Published online: 23 September 2010
Ó Springer-Verlag 2010
Abstract The tetrabutylammonium (TBA) salts of
Keggin-type polyoxotungstates of the general formula
[XW11FeIII(H2O)O39]n-, where X = P, B or Si, were
evaluated as catalysts in the oxidation, under mild condi-
tions, of ethylbenzene, cumene, p-cymene and sec-
butylbenzene with aqueous H2O2 in CH3CN at 80 °C. The
influence of various factors, such as the substrate/catalyst
molar ratio, the amount of oxidant added or the reaction
time, was investigated in a systematic way. Generally, the
system exhibited moderate conversion, with good selec-
tivity towards the corresponding acetophenone and
hydroperoxide. In order to understand the reaction path-
ways, the oxidation of several products and presumed
intermediates was also carried out in the presence of
TBA4[PW11Fe(H2O)O39]Á2H2O. Under the conditions
used, the oxidation of styrene and styrene derivatives gave
rise mainly to carbon–carbon double-bond cleavage,
affording the corresponding products in very high yields
(81–87%). Possible reaction pathways are presented.
Introduction
Liquid-phase oxidation of arenes, namely alkyl-substituted
benzenes, can be an important tool to obtain industrially
valuable functionalized products. For example, the oxida-
tion of cumene yields cumene hydroperoxide, which is an
intermediate for both phenol and acetophenone production,
and the oxidation of ethylbenzene to ethylbenzene hydro-
peroxide is a key step in the propylene oxide/styrene
process [1]. The oxidation of p-cymene affords 4-methyl-
acetophenone, an important precursor for the manufacture
of perfumes, or 4-isopropylbenzaldehyde, used as a fla-
vouring agent for food materials [2, 3].
Achievement of this type of transformation under cata-
lytic conditions constitutes a challenging task to chemistry
research and industry [4–7]. Facing the present environ-
mental constraints, new catalytic approaches combining
environmentally benign oxidants and catalysts are highly
desirable. Many papers address the study of processes for
the synthesis of value-added ketones by benzylic oxidation
of alkylaromatics using oxidants such as O2 or H2O2, a few
of them quite recent [8–14]. These substitute the classic
stoichiometric methods based on oxidations with KMnO4
or K2Cr2O7 or on Friedel–Crafts acylation catalysed by
AlCl3 [15, 16]. Aromatic ketones are important chemical
intermediates in pharmaceutical, fragrance, flavour, dye
and agrochemical industries [17, 18].
Keywords Polyoxometalates Á
Transition-metal compounds Á Oxidations Á
Arenes Á Peroxides Á C=C cleavage
A. C. Estrada Á A. M. V. Cavaleiro (&)
CICECO, Department of Chemistry, University of Aveiro,
3810-193 Aveiro, Portugal
Polyoxometalates (POMs) are economically and envi-
ronmentally attractive acid or oxidation catalysts in both
academic and industrial processes [19–25]. Recently we
have shown that FeIII-substituted Keggin-type polyoxo-
tungstates can be effective catalysts in the oxidation with
H2O2 of compounds such as indane or indene in acetoni-
trile [26, 27]. Hydrogen peroxide has high effective oxygen
content, low cost, safe storage and operation methods and,
e-mail: anacavaleiro@ua.pt
M. M. Q. Simo˜es Á I. C. M. S. Santos Á
M. G. P. M. S. Neves (&) Á J. A. S. Cavaleiro
QOPNA, Department of Chemistry, University of Aveiro,
3810-193 Aveiro, Portugal
e-mail: gneves@ua.pt
123

1224
A. C. Estrada et al.
above all, an environmentally friendly nature [28–31]. The
combination of iron polyoxotungstates and hydrogen per-
oxide seemed, thus, to be an interesting option to use in the
catalytic oxidation of arenes, and this led us to extend our
studies to less reactive substrates. The catalytic oxidation
of aromatic hydrocarbons with hydrogen peroxide, namely
ethylbenzene, has been investigated with POMs such as
K8SiW11O39 (covalently linked to Schiff base complexes)
[32, 33], Na5PV2Mo10O40 (supported on MCM-41) [34],
TBA3PW12O40, TBA3PMo12O40, TBA4H3PW11O39 [35],
H6SiV2Mo10O40, H4PVMo11O40, H5PV2Mo10O40 [36],
TBA4PVMo11O40, or TBA4HPW11Fe(OH)O39 [37].
possible intermediates was also studied, in order to define
possible reaction routes.
Results and discussion
The tetrabutylammonium (TBA) salts of Keggin-type
anions [XW11FeIII(H2O)O39]n-, X = P, Si, or B (abbrevi-
ated XW11Fe), were used as catalysts in this study. The
reactions were performed at 80 °C with H2O2/substrate
molar ratios equal to 2.0, 4.0, or 9.8. All presented results
were obtained after 24 h of reaction. No improvement in
the outcome of the reaction was observed by maintaining
the oxidation process during 36 h.
In this paper we report the oxidation of ethylbenzene,
cumene, p-cymene and sec-butylbenzene with 30% aque-
ous H2O2 in acetonitrile, using the iron(III)-substituted
polyoxotungstate anions [XW11FeIII(H2O)O39]n-, X = P,
Si, or B, as homogeneous catalysts. The outcome of the
reactions depended strongly on the catalyst and conditions
used. The oxidation of some of the reaction products or
Oxidation of ethylbenzene (1)
The oxidation of ethylbenzene (1) (Scheme 1) occurs with
moderate conversion (Table 1; Fig. 1), reaching the highest
Scheme 1
OH
OOH
+
O
O
+
+
+
+
OH
1
2
3
4
5
6
7
Table 1 Oxidation of
ethylbenzene (1) by hydrogen
Entry Catalyst Sub/cat H2O2/sub Conversion TONb H2O2
Selectivity (%)a
(%)a
consumed
(mmol)
peroxide catalysed by FeIII
-
2
3
4
5
6
7
substituted polyoxotungstates
after 24 h of reaction
1
PW11Fe 667
PW11Fe 333
BW11Fe 667
BW11Fe 333
SiW11Fe 667
SiW11Fe 333
None
2.0
4.0
9.8
2.0
4.0
9.8
2.0
4.0
9.8
2.0
4.0
9.8
2.0
4.0
9.8
2.0
4.0
9.8
9.8
17
18
20
21
25
26
17
18
19
15
16
17
1
114
120
132
69
0.54
1.92
3.22
0.85
2.61
6.28
1.97
3.94
9.65
1.97
3.96
9.64
0.12
0.26
0.31
0.19
0.48
0.70
n.d.
4
5
5
6
6
6
7
6
7
1
4
3
10 10 27 41
8
1
3
2
12
14
9
9
33 40
32 37
3
4
10 13 13 42 16
5
85
9
9
8
7
18 47 12
6
87
21 48
9
–
–
–
2
7
2
7
115
120
137
49
16 20 39 18
15 17 46 16
13 17 50 13
12 14 59 12
8
9
10
11
12
13
14
15
16
17
18
19
Reaction conditions: substrate
(1.0 mmol), catalyst (1.5 or
3.0 lmol), and 2.0, 4.0, or
9.8 mmol of aqueous 30%
(w/w) H2O2 were stirred for
24 h in CH3CN at 80 °C
n.d. Not determined
a
Determined by gas
chromatography-mass
spectrometry (GC-MS)
53
11 13 59
6
4
56
8
9
9
7
5
4
5
–
16 67
8
14
8
10 24 28 15
11 25 17 30
11 21 21 32
2
16
3
18
8
10
10
11
–
33
4
6
4
5
–
10 29 46
10 30 50
12 25 51
33
2
36
2
b
Turnover number (mol of
products per mol of catalyst)
–
–
–
–
–
123

Catalytic activity of iron-substituted polyoxotungstates in the oxidation of aromatic compounds
1225
30
25
20
15
10
5
The decreasing conversion with increasing amount of cat-
alyst is not totally unexpected, due to the increase of iron
concentration, which can assist H2O2 dismutation and
inhibit the oxidation reaction [29].
The selectivity for the products varied with the catalyst
added and also with the amounts of catalyst and oxidant
used. In the presence of PW11Fe, acetophenone (5) and
ethylbenzene hydroperoxide (6) are, generally, the major
products. Under the best conditions conversion wise
(Table 1, entries 5 and 6), 6 is the main product (47–48%)
followed by 5 (18–21%). Styrene (2), benzaldehyde (3), 1-
phenylethanol (4) and ethylphenol (7) are obtained as
minor products (B9% for entry 6). In the reactions cata-
lysed by BW11Fe, the formation of acetophenone (5) is
favoured and the best selectivity for this ketone (67%) is
found for conditions of entry 12 in Table 1. Under these
conditions, the second most abundant compound is 4
(16%), whereas the other products (2, 3, 6, and 7) occur
with minor abundances (selectivity B8%). The selectivity
for 5 and 6, in the presence of different amounts of oxidant,
using 1.5 or 3.0 lmol of BW11Fe, is presented in Fig. 2.
With this catalyst, acetophenone (5) is always the major
product and the selectivity for the hydroperoxide 6
decreases with increasing amount of catalyst and oxidant.
In the presence of SiW11Fe (entries 16–18), oxidation of
the aromatic ring is favoured, affording two phenol
derivatives 7, followed by 6 as the second most abundant
product.
0
0.0
1.5
3.0
4.5
6.0
7.5
Catalyst/ mol
Fig. 1 Conversion of ethylbenzene in the presence of PW11Fe (filled
squares), SiW11Fe (filled circles), or BW11Fe (filled triangles) after
24 h of reaction. Substrate: 1.0 mmol; H2O2: 9.8 mmol; acetonitrile:
3.0 cm3; 80 °C
values (25–26%) in the presence of PW11Fe (molar ratio
substrate/catalyst = 333) with excess of hydrogen perox-
ide (molar ratios H2O2/substrate = 4.0 and 9.8). For
BW11Fe, the conversions are in the range 15–19%, with the
best results obtained when the oxidations are carried out
with substrate/catalyst = 667. SiW11Fe was the least effi-
cient catalyst, with conversion always lower than 11%.
Figure 1 illustrates how the conversion of ethylbenzene
is affected by the amount of catalyst used, for the molar
ratio H2O2/ethylbenzene of 9.8. Similar results were
obtained with H2O2/ethylbenzene equal to 4.0. The maxi-
mum conversion of 26% is attained in the presence of
3.0 lmol of PW11Fe (per 1 mmol of substrate), and lower
conversion values are obtained with 1.5 lmol or 6.0 lmol
of PW11Fe (20% and 18%, respectively). This tendency is
also observed with SiW11Fe, but in the presence of
BW11Fe the best performance is attained with the smallest
amount of catalyst tested (1.5 lmol per mmol of substrate).
The consumption of H2O2 is higher in the presence of
BW11Fe than for the other two polyoxotungstates, for the
same H2O2/substrate molar ratio. The lowest consumption
of H2O2 is observed with SiW11Fe (Table 1). For this
catalyst and for PW11Fe the use of a higher amount of
catalyst (6.0 lmol) induces a higher consumption of H2O2
without improving the conversion of ethylbenzene (results
not shown). These results suggest that excess of catalyst in
solution may lead to the decomposition of H2O2 and/or
Fig. 2 Selectivity for
acetophenone (5) and
hydroperoxide (6) for
ethylbenzene oxidation reaction
with 2.0, 4.0, and 9.8 mmol of
H2O2 catalysed by BW11Fe,
after 24 h. Substrate: 1.0 mmol;
catalyst: a 1.5 lmol and
b 3.0 lmol; acetonitrile:
3.0 cm3 at 80 °C
5
6
(a)
5
6
(b)
80
70
60
50
40
30
20
10
0
80
70
60
50
40
30
20
10
0
2.0
4.0
9.8
2.0
4.0
H2O2/mmol
9.8
H2O2/mmol
123

1226
A. C. Estrada et al.
hydroperoxide and that this effect is more pronounced in
the presence of BW11Fe.
(substrate/catalyst = 333, Table 2, entries 5 and 6). The
highest conversions attained with BW11Fe and SiW11Fe are
19% and 17%, respectively (Table 2, entries 9 and 18). As
found with ethylbenzene, in the case of PW11Fe and
SiW11Fe the conversion reaches a maximum at 3.0 lmol
per mmol of substrate, and for BW11Fe the best values are
observed with 1.5 lmol (Fig. 3a). Figure 3b illustrates how
the conversion of cumene, in the presence of 3 lmol of
each catalyst (cumene/catalyst molar ratio of 333), is
affected by the amount of H2O2 added. In all cases, the
increase of the H2O2/cumene molar ratio leads to an
increase in the conversion.
These studies allow us to conclude that, in the reaction
conditions studied, the maximum conversion of ethylben-
zene attained is 26%. This is obtained in the presence of
PW11Fe after 24 h of reaction, with good selectivity for
ethylbenzene hydroperoxide (6). A moderate conversion of
18–19% is obtained with BW11Fe (with selectivity for
acetophenone above 60%), while only 11% can be
achieved with SiW11Fe. The maximum conversions
obtained are higher than those frequently observed for this
substrate with published hydrogen peroxide/POM-based
homogeneous systems [34–36].
In the presence of PW11Fe, cumene hydroperoxide (12)
is in general the major product, attaining selectivity higher
than 71% for the best conversion conditions (Table 2,
entries 4–6). Acetophenone (5) and 2-phenyl-2-propanol
(10) are obtained with similar selectivity (7–12%) as the
second most abundant products. In the presence of BW11Fe,
under the best conditions, the formation of 5 (46%) and 10
(31%) seems to be favoured (entry 9). Hydroperoxide 12,
Oxidation of cumene (8)
The oxidation of cumene (8) with H2O2 in the presence of
the FeIII-substituted polyoxotungstates affords the products
shown in Scheme 2. The best conversions (34–37%) are
attained in the presence of PW11Fe with excess of oxidant
Scheme 2
O
OH
OOH
O
OH
+
+
+
+
+
8
9
5
10
11
12
13
Table 2 Oxidation of cumene
(8) by hydrogen peroxide
catalysed by FeIII-substituted
polyoxotungstates after 24 h of
reaction
Entry Catalyst Sub/cat H2O2/sub Conversion TONb H2O2
Selectivity (%)a
10 11 12 13
11 16 15 10 27 21
(%)a
consumed
(mmol)
9
5
1
PW11Fe 667
PW11Fe 333
BW11Fe 667
BW11Fe 333
SiW11Fe 667
SiW11Fe 333
None
2.0
4.0
9.8
2.0
4.0
9.8
2.0
4.0
9.8
2.0
4.0
9.8
2.0
4.0
9.8
2.0
4.0
9.8
3
10
17
30
34
37
10
15
19
7
20
70
1.46
2.71
3.62
1.87
3.08
6.24
1.89
3.91
9.61
1.89
4.00
9.58
0.25
0.62
0.17
0.29
0.75
0.57
n.d.
2
4
6
2
2
3
6
13 12
2
1
2
3
2
2
4
5
2
2
3
4
4
4
4
9
68
67
78
72
71
22
22
10
44
35
18
1
–
3
6
–
–
–
–
–
–
–
3
112
100
113
122
67
23
8
3
7
7
4
5
10
6
12 12
40 30
7
8
100
126
23
10 37 27
9
8
4
4
5
9
46 31
25 25
30 29
43 31
19 18
10
11
12
13
14
15
16
17
18
19
12
16
3
39
Reaction conditions: substrate
(1.0 mmol), catalyst (1.5 or
3.0 lmol), and 2.0, 4.0, or
9.8 mmol of aqueous 30%
(w/w) H2O2 were stirred for
24 h in CH3CN at 80 °C
53
20
31 19
22 11
21 16
38 35
25 29
3
20
12 31 20
13 28 18
3
20
13
16
17
–
44
3
9
10 10
15 13
n.d. Not determined
a
Determined by GC-MS
54
58
11 15 13 10 22 29
b
Turnover number (mol of
products per mol of catalyst)
–
–
–
–
–
–
123

Catalytic activity of iron-substituted polyoxotungstates in the oxidation of aromatic compounds
1227
Fig. 3 Conversion of cumene
(a)
(b) 40
35
30
25
20
15
10
5
40
35
30
25
20
15
10
5
(8) in the presence of PW11Fe
(filled squares), SiW11Fe (filled
circles), or BW11Fe (filled
triangles) after 24 h of reaction.
a H2O2: 9.8 mmol, different
amounts of catalyst; b Catalyst:
3.0 lmol, different H2O2/
substrate molar ratios.
Substrate: 1.0 mmol;
acetonitrile: 3.0 cm3 at 80 °C
0
0
0.0
2.0
4.0
6.0
8.0 10.0 12.0
0.0
1.5
3.0
4.5
6.0
7.5
H2O2/mmol
Catalyst/µmol
(17), and hydroperoxide (19) as major products, along with
p-isopropylbenzyl alcohol (18) and p,a-dimethylstyrene
(15) as minor ones. Carvacrol (20) is observed only in the
presence of SiW11Fe (Scheme 3; Table 3).
100
90
80
70
60
50
40
30
20
10
0
The best conversion (35%) is obtained in the presence of
BW11Fe (Table 3, entry 9). With PW11Fe and SiW11Fe,
the highest conversions are 23% and 29%, respectively
(Table 3, entries 5, 6 and 18). Figure 5 illustrates, in the
case of H2O2/p-cymene molar ratio of 9.8, how the con-
version of p-cymene is affected by the amount of catalyst
used. These results are comparable to those found with the
previous substrates.
With PW11Fe and for all the conditions studied, 16, 17,
and 19 are always the major products. For SiW11Fe the
main product is 19 and the compounds 16, 17, and 20 are
obtained with similar selectivity (Table 3, entries 17 and
18); p,a-dimethylstyrene (15) occurs with minor abundance
(B6%). p-Methylacetophenone (16) and p-isopropylbenz-
aldehyde (17) are always the major products in the
presence of BW11Fe (Fig. 6). With this catalyst, hydro-
peroxide selectivity is lower than that observed with
PW11Fe and SiW11Fe, and decreases with increasing
amount of catalyst. This behaviour can, again, be due to the
apparent propensity of the anion with boron to decompose
hydrogen peroxide or related species. Again, the con-
sumption of hydrogen peroxide is always higher in the
presence of BW11Fe, when compared with SiW11Fe and
PW11Fe in the same reaction conditions.
Fig. 4 H2O2 consumption (filled squares) and conversion (bars) for
cumene (8) oxidation catalysed by PW11Fe and BW11Fe using
9.8 mmol of H2O2. Substrate: 1.0 mmol; catalyst: 1.5, 3.0 or
6.0 lmol; acetonitrile: 3.0 cm3 at 80 C
a-methylstyrene (9), and 2-phenylpropanal (11) occur with
minor abundances (\10%). In the presence of SiW11Fe
(entries 17 and 18), two phenol derivatives 13 and hydro-
peroxide 12 are the main products, although the oxidation of
the aromatic ring is slightly favoured. The other products
are obtained with comparable selectivity.
The hydrogen peroxide consumption in the oxidation of
8 is the highest (85–98%) in the presence of BW11Fe. In
the presence of 3.0 lmol or 6.0 lmol of PW11Fe, H2O2
consumption is nearly the same (63%), but the best con-
version is registered with 3.0 lmol (Fig. 4). Thus, the
increase of the amount of catalyst possibly leads to non-
productive H2O2 decomposition. These results parallel
those obtained with ethylbenzene.
Oxidation of sec-butylbenzene (21)
sec-Butylbenzene (21) is the least reactive substrate tested
(maximum conversion of 7%), and the oxidation is only
detected when the substrate/catalyst molar ratio is equal to
333 in the presence of PW11Fe or SiW11Fe, and 667 for
BW11Fe. Similar conversions are obtained using H2O2/
substrate = 9.8 or 4.0. Acetophenone (5) and the hydro-
peroxide (22) were the products observed (Scheme 4).
Oxidation of p-cymene (14)
The oxidation of p-cymene (14) under the conditions studied
gives p-methylacetophenone (16), p-isopropylbenzaldehyde
123

1228
A. C. Estrada et al.
Scheme 3
OOH
O
+
+
+
+
+
OH
O
OH
14
15
16
17
18
19
20
Table 3 Oxidation of
p-cymene (14) by hydrogen
peroxide catalysed by
FeIII-substituted
Entry Catalyst Sub/cat H2O2/sub Conversion TONb H2O2
Selectivity (%)a
15 16 17 18 19 20
(%)a
consumed
(mmol)
polyoxotungstates after 24 h of
reaction
1
PW11Fe 667
PW11Fe 333
BW11Fe 667
BW11Fe 333
SiW11Fe 667
SiW11Fe 333
None
2.0
4.0
9.8
2.0
4.0
9.8
2.0
4.0
9.8
2.0
4.0
9.8
2.0
4.0
9.8
2.0
4.0
9.8
14
20
20
18
23
23
15
23
35
18
20
25
6
93
132
132
59
1.00
1.81
3.45
1.56
2.59
4.79
1.95
3.99
9.67
1.93
3.94
9.71
0.34
0.33
0.33
0.36
0.37
0.40
n.d.
9
22 19
–
–
–
–
–
–
–
6
8
–
–
–
6
5
4
50
45
20
45
43
31
17
22
4
–
–
–
–
–
–
–
–
–
–
–
–
2
10 22 23
19 36 25
3
4
9
20 26
5
77
13 20 24
13 29 27
15 43 25
6
77
7
98
8
150
235
60
9
8
40 23
48 32
9
10
11
12
13
14
15
16
17
18
19
15 46 28
17 48 27
15 55 27
11
8
65
Reaction conditions: substrate
(1.0 mmol), catalyst (1.5 or
3.0 lmol), and 2.0, 4.0, or
9.8 mmol of aqueous 30%
(w/w) H2O2 were stirred for
24 h in CH3CN at 80 °C
83
3
41
9
7
4
6
5
6
–
17 22
16 25
14 22
36 10
6
41
41
49
6
7
7
44
17
28
29
–
55
19 19 12 25 19
16 19 14 29 17
14 18 13 33 16
n.d. Not determined
a
Determined by GC-MS
94
95
b
Turnover number (mol of
products per mol of catalyst)
–
–
–
–
–
–
PW11Fe affords 22 as the only product, and the other two
catalysts give 22 (50% selectivity with SiW11Fe) and 5
(77% selectivity with BW11Fe) as the main products.
40
35
30
25
20
15
10
5
Catalyst stability
The reaction conditions used in this study were applied
before in studies with other hydrocarbons [26], in which it
was observed that no decomposition of PW11Fe took place
during the reactions. In this study, the infrared spectra of
the catalyst recovered from the solutions at the end of the
reaction (24 h) presented the characteristic bands of
[PW11Fe(H2O)O39]4- or [PW11Fe(OH)O39]5- [38, 39],
suggesting that no decomposition had occurred. The sta-
bility of the anion was also confirmed by the search for the
possible existence of leached FeIII in solution, using SCN-.
Preliminary tests showed that the addition of KSCN to a
CH3CN solution of TBA4[PW11Fe(H2O)O39] did not
noticeably change its yellow colour, while similar tests
0
0.0
1.5
3.0
4.5
6.0
7.5
Catalyst/ mol
Fig. 5 Conversion of p-cymene (14) using 9.8 mmol H2O2 in the
presence of PW11Fe (filled squares), SiW11Fe (filled circles), or
BW11Fe (filled triangles) after 24 h of reaction. Substrate: 1.0 mmol;
catalyst: 1.5, 3.0 or 6.0 lmol; acetonitrile: 3.0 cm3 at 80 °C
123

Catalytic activity of iron-substituted polyoxotungstates in the oxidation of aromatic compounds
1229
Fig. 6 Selectivity for reaction
(b)
(a)
16
17
19
16
17
19
products 16, 17, and 19 for
p-cymene (14) oxidation
reactions in the presence of
a 1.5 lmol or b 3.0 lmol of
BW11Fe with 2.0, 4.0, or
9.8 mmol of H2O2. Substrate:
1.0 mmol; CH3CN: 3.0 cm3 at
80 °C
60
50
40
30
20
10
0
60
50
40
30
20
10
0
2.0
4.0
H2O2/mmol
9.8
2.0
4.0
H2O2/mmol
9.8
excess of H2O2 also contributes to the supply of O2. The
results of reactions performed in an inert atmosphere
(oxidation of cumene with excess H2O2) are not signifi-
cantly different from those obtained in air. Similar
behaviour has already been described in our previous work
on the oxidation of cycloalkanes in the presence of iron-
substituted polyoxotungstates [41–43].
OOH
O
+
21
5
22
Scheme 4
Scheme 6 illustrates how some of the reaction products
may result from the decomposition of the hydroperoxides
formed [4, 44, 45]. This seems to be an important pathway
to justify many of the products obtained, whereas the
oxidative dehydrogenation of the substrates may account
for the formation of 2, 9 and 15. Even so, in order to
identify other possible contributions for the distribution of
reaction products, the oxidation of several commercially
available products was conducted under one of the reaction
conditions used, namely 3.0 lmol of PW11Fe per 1.0 mmol
of substrate using H2O2/substrate = 9.8. The results
obtained are summarized in Table 4.
performed with FeCl3 confer to the solution the common
red colour of [Fe(SCN)x]3-x complexes. Similar results
were obtained with mixed CH3CN/H2O solvents (up to
30% volume water content). At the end of the catalytic
reactions, a yellow solution was obtained after the addition
of KSCN, in accordance with the absence of significant
leaching of iron from the polyoxometalate to the reaction
mixture. UV–Vis spectrophotometry was used to confirm
the visual observations.
The catalytic oxidation of the secondary alcohol
1-phenylethanol (4) occurs at total conversion, affording
acetophenone (5) as the main product (95% selectivity) and
2 as a minor component (Table 4, entry 1). The oxidation
of styrene (2) occurs also with full conversion and affords 3
(52%) and 23 (29%) as major products (Table 4, entry 2).
On the other hand, the oxidation of styrene oxide (26) gives
the corresponding diol 27 as main product (60% selectiv-
ity), followed by benzaldehyde (3) and benzoic acid (23)
(16% and 20%, respectively; Table 4, entry 3). These
results suggest that oxidative cleavage of carbon–carbon
double bond has an important contribution for the forma-
tion of the styrene oxidation products.
Mechanistic considerations
Taking into account the products obtained, the existence of
concurrent reactions is clear, although dependent on the
catalyst and reaction conditions used. When the reactions
described herein were performed in the presence of iodine,
a well-known radical trap [40], no products were detected,
suggesting that the oxidations proceed mainly via a radical
mechanism. In particular, the involvement of RÁ and ROOÁ
is corroborated by the formation of the hydroperoxides.
Scheme 5 illustrates a possible mechanism. The first step
involves hydrogen abstraction from the substrate in the
presence of the HOÁ that is generated by the iron complexes
through a Fenton-like process [29]. The resulting free
radical can combine with oxygen to form a peroxy radical,
ROOÁ, which abstracts a hydrogen atom from other sub-
strate molecule to yield the hydroperoxide. The oxygen is
possibly generated in situ through the reaction of FeIII and
H2O2, as proposed before [41]. The dismutation of the
The carbon–carbon double-bond cleavage seems to be
also the main catalytic oxidation pathway for the other two
vinyl derivatives 9 and 15. These are oxidized with high
efficiency (Table 4, entries 4 and 5), reaching 86% and
88% conversion, respectively, affording the corresponding
ketones 5 and 16 with high selectivity (87% and 82%,
123

1230
A. C. Estrada et al.
Scheme 5
XW11FeIII + H2O2
XW11FeIII + HOO
XW11FeII + H2O2
XW11FeII + HOO. + H+
XW11FeII + O2 + H+
.
XW11FeIII + HO. + OH-
O2
O
R2
O
R1
R2
R1
R1
R2
HO
-H2O
R3
R3
R3
OH
R2
O
R1
R2
R1
R3
R3
ethylbenzene (1) R1 = CH3; R2 = R3 = H; cumene (8) R1 = R2 = CH3, R3 = H;
p-cymene (14) R1 = R2 = R3 = CH3; sec-butylbenzene (21) R1 = CH3, R2 = CH2CH3, R3 = H
respectively). Again, a minor contribution of epoxide ring-
opening is detected.
(24%) and 9 (22%) as the next most abundant products.
Phenol (24) is obtained with 14% selectivity. As expected,
ketones 5 and 16 do not react when submitted to the reaction
conditions under study (entry 9).
The primary alcohol 18, when submitted to the same
reaction conditions as p-cymene (14), gives, at excellent
conversion, the corresponding aldehyde 17 as the major
product with 86% selectivity (Table 4, entry 7). The oxi-
dation of the tertiary alcohol 10 occurs with full conversion
and affords hydroperoxide 12 (44%) and phenol (42%) as
major products (Table 4, entry 6). The other compounds
detected are a-methylstyrene (10%) and acetophenone
(4%). The unexpected formation of the hydroperoxide from
the tertiary alcohol was confirmed by comparing its mass
spectrum with that obtained from an authentic commer-
cially available sample (Fig. 7) and also by GC co-injection
of the two samples. The oxidation of the commercially
available hydroperoxide 12 occurs with 76% conversion
and gives 5 (40%) as the main product, followed by 10
In summary, the main products obtained in the oxidation
of alkylbenzenes are the corresponding ketones and
hydroperoxides. Aldehyde 3 and ketones 5 and 16 may
arise from C=C cleavage of the corresponding styrene
derivatives, along with the contribution from the C–C
cleavage of the corresponding hydroperoxide, which is
formed by a radical mechanism.
Acetophenone (5) is one of the major products in eth-
ylbenzene (1) oxidation and seems to arise via further
oxidation of the secondary alcohol (4). Acetophenone is
also observed in the products of cumene oxidation and, as a
minor component, in the oxidation of 10, which is formed
during cumene oxidation. Therefore, we believe that
123

Catalytic activity of iron-substituted polyoxotungstates in the oxidation of aromatic compounds
1231
R1
R2
Scheme 6
OH
R3
O
O
R2
R1
R1
R2
R3
+ CH3OH
+ HO-
R3
R3
R1
R2
FeIII
FeII
if R2 = R3 = H
R3
OH2
R2
OH
R2
O
O
R1
R1
H+
R3
R3
if R2 = CH3
Ethylbenzene (1) R1 = CH3; R2 = R3 = H;
1
2
3
8
R1
cumene ( ) R = R = CH3, R = H;
O
p-cymene (14) R1 = R2 = R3 = CH3
+ CH3OH
R3
acetophenone arises from cumene hydroperoxide C–C
cleavage, according to Scheme 6. In the ethylbenzene
oxidation, ethylbenzene hydroperoxide C–C cleavage leads
to benzaldehyde (Scheme 6).
Finally, the type of products obtained shows that in most
of the cases the reactions occur preferentially in the more
reactive benzylic positions.
These studies show that ketones 5 and 16 also arise from
carbon–carbon double-bond cleavage of the corresponding
alkene [27]. In fact, the oxidation of styrene (2) and styrene
derivatives (9 and 15) in the presence of PW11Fe afford the
corresponding carbon–carbon double-bond cleavage prod-
ucts in very high yields. In the oxidation of 2, C=C
cleavage products 3 and 23 sum 81%, whereas for the
oxidation of 9 and 15 the ketones obtained by C=C
cleavage represent 87% (5) and 82% (16), respectively.
Compounds 2 and 9 do not seem to arise from the
dehydration of alcohols 4 and 10, respectively, since only
minor amounts of 2 and 9 are obtained in the oxidation of
these alcohols. Thus, the oxidative dehydrogenization of 1
and 8 is considered more probable. The oxidation of 14,
where alkene 15 is also obtained without detection of the
alcohol, seems to confirm this fact.
Conclusions
The results presented reveal a significant catalytic system
based on iron(III) POMs for the oxidation of alkylbenzenes
and other benzene derivatives, using environmentally
benign H2O2 under mild conditions. The solvent used
(acetonitrile) is comparatively benign, and no co-catalysts
are needed. The oxidation of ethylbenzene, cumene,
p-cymene, and sec-butylbenzene was studied with three
different catalysts at different substrate/catalyst and H2O2/
substrate molar ratios. The oxidation of ethylbenzene and
cumene occurred with maximum conversion of 26% and
37%, respectively, in the presence of PW11Fe, both after
24 h of reaction, whereas the maximum conversion of
p-cymene (35%) was found when BW11Fe was used. These
123

1232
A. C. Estrada et al.
Table 4 Oxidation of products
and presumed intermediates
with H2O2 catalysed by PW11Fe
Products
Selectivity (%)a
Entry
Substrate
Conversion (%)a
OH
O
1
100
4
5
2
[95]
[5]
O
HO
O
O
O
OH
2
3
4
100
100
86
24
2
3
23
[29]
5
25
[4]
[52]
[13]
[2]
O
O
HO
HO
OH
O
O
O
26
3
23
5
27
25
[16]
[20]
[2]
[2]
[60]
O
OH
O
OH
OH
9
5
11
24
28
[3]
[87]
[4]
[6]
O
O
OH
5
6
7
88
100
89
15
16
30
29
[82]
[11]
[7]
OH
OOH
O
OH
10
12
24
9
5
[44]
[42]
[10]
[4]
O
OH
O
OH
18
17
31
[86]
[14]
OOH
OH
O
OH
8
9
76
0
Reaction conditions: substrate
(1.0 mmol), catalyst (3.0 lmol),
and 9.8 mmol of aqueous 30%
(w/w) H2O2 were stirred for
24 h in CH3CN at 80 °C
12
10
9
24
5
[40]
[24]
[22]
[14]
5 and 16
No products
a
Determined by GC
123

Catalytic activity of iron-substituted polyoxotungstates in the oxidation of aromatic compounds
1233
119
119
Fig. 7 MS spectra of
a commercial sample of cumene
hydroperoxide (12) and
b cumene hydroperoxide (12)
in the oxidation of 2-phenyl-2-
propanol (10)
(a)
(b)
100
90
80
70
60
50
40
30
20
10
0
100
90
80
70
60
50
40
30
20
10
0
91
77
91
105
43
121
77
105
43
121
51
103
78
118
65
94
51
57
63
79
89
66 79
89
117
103
94
134
137
52
65
122
136
152
152
150
50
100
m/z
150
50
100
m/z
results were obtained with excess of hydrogen peroxide
(molar ratio H2O2/substrate C 4). The reactions occurred
by a radical mechanism, and the formation of hydroper-
oxides (compounds 6, 12, 19, and 22) was observed for all
substrates studied. The highest selectivity (yield) values for
the hydroperoxides were obtained with PW11Fe. In the
presence of BW11Fe, the oxidation of ethylbenzene and
cumene yielded acetophenone (5) with moderate selectiv-
ity, while p-methylacetophenone was generally the major
product obtained in the oxidation of p-cymene.
used with good results in the oxidation of alcohols and
vinyl aromatic derivatives.
Experimental
Reagents and synthetic procedures
Acetonitrile (Panreac), 30% (w/w) aqueous hydrogen
¨
peroxide (Riedel-de-Haen), ethylbenzene (1), styrene (2),
The systems described herein show good catalytic effi-
ciency associated to a different product distribution when
compared with other systems using hydrogen peroxide and
POMs [34–36]. In fact, the conversions obtained here are
higher than frequently observed for this kind of substrates,
accompanied by an unusual formation of high amounts of
the corresponding hydroperoxides, which are detected for
the first time in catalysis with POMs. It should be noted
that the outcome of the oxidation of the alkylaromatics
may possibly be tuned through the choice of catalyst and
reaction conditions.
benzaldehyde (3), a-methylstyrene (9), 2-phenyl-2-pro-
panol (10), 2-phenylpropanal (11), cumene hydroperoxide
(12), p,a-dimethylstyrene (15), p-methylacetophenone
(16), p-isopropylbenzaldehyde (17), p-isopropylbenzyl
alcohol (18), carvacrol (20), and styrene oxide (26) (all
from Aldrich), and 1-phenylethanol (4), acetophenone (5),
cumene (8), p-cymene (14), and sec-butylbenzene (21) (all
from Fluka) were used as received. All other reagents and
solvents obtained from commercial sources were used as
received or distilled and dried using standard procedures.
TBA4[PW11Fe(H2O)O39]Á2H2O, TBA4H[SiW11Fe(H2O)O39],
and TBA4H2[BW11Fe(H2O)O39]ÁH2O were prepared and
identified according to previously described procedures
[46–48]. The obtained compounds were characterized
by elemental analysis, thermogravimetry, and infrared
spectroscopy [38, 47].
In order to understand the reaction pathways, other
organic compounds, namely styrenes (2, 9, and 15), alco-
hols (4, 10, and 18), cumene hydroperoxide and styrene
oxide, were oxidized with H2O2 in the presence of PW11Fe.
In the condition used, all reacted with high conversion and,
for some of them, with high selectivity. Examples are the
oxidation of a-methylstyrene (9), p,a-dimethylstyrene (15)
and p-isopropylbenzyl alcohol (18), which occurred with
conversions between 86% and 89% after 24 h, leading
respectively to the formation of acetophenone (5), p-iso-
propylbenzaldehyde (17), and p-methylacetophenone (16),
all with selectivity above 80%. These results suggest that
the catalytic system H2O2/iron polyoxotungstates can be
Typical procedure for the oxidation reactions
Typical oxidation reactions were carried out as follows:
1.5, 3.0, or 6.0 lmol of catalyst, 1.0 mmol of substrate, the
required amount of 30% (w/w) aqueous H2O2 and 3.0 cm3
acetonitrile were stirred at 80 °C in a closed vessel. The
reactions were followed by GC-MS and were stopped when
123

1234
A. C. Estrada et al.
References
conversion remained constant after two successive GC-MS
analyses. The percentages of each compound in the reac-
tion mixtures were estimated from the corresponding
chromatographic peak areas, using 1-hexanol as internal
standard (added after the reaction). When experiments
(oxidation of cumene with excess H2O2) were repeated
under an inert atmosphere, the air was purged from the
reaction vessel and substituted by argon. Blank reactions
were performed for all substrates, confirming that no oxi-
dation products were obtained unless the catalyst and H2O2
were present.
1. Wittcoff H, Reuben BG, Plotkin JS (1996) Industrial organic
chemicals, 2nd edn. Wiley-Interscience, New York
2. Vetrivel S, Pandurangan A (2008) Catal Lett 120:71
3. Nair K, Sawant DP, Shanbhag GV, Halligudi SB (2004) Catal
Commun 5:9
4. Shilov AE, Shul’pin GB (2000) Activation and catalytic reactions
of saturated hydrocarbons in the presence of metal complexes.
Kluwer, Dordrecht
5. Derouane EG, Parmon V, Lemos F, Ribeiro FR (2005) Sustain-
able strategies for the upgrading of natural gas: fundamentals,
challenges, and opportunities, vol 191. NATO Science Series,
Kluwer, Dordrecht
6. Sheldon RA, Arends I, Hanefeld U (2007) Green chemistry and
catalysis. Wiley-VCH, Weinheim
Instruments and methods
7. Crabtree RH (ed) (2010) Selective functionalization of CH bonds.
Chem Rev 110(2) (thematic issue)
8. Matienko LI, Mosolova LA, Zaikov GE (2009) Russ Chem Rev
78:211
9. Dapurkar SE, Kawanami H, Yokoyama T, Ikushima Y (2009)
Catal Commun 10:1025
10. Monteiro B, Gago S, Balula SS, Valente AA, Gonc¸alves IS,
Pillinger M (2009) J Mol Catal A Chem 312:23
11. Maurya MR, Arya A, Ada˜o P, Pessoa JC (2008) Appl Catal A
Gen 351:239
GC-MS analysis was performed using a Finnigan Trace
GC-MS (Thermo Quest CE instruments) with helium as
carrier gas (35 cm3/s) and a fused silica capillary column
with 5% phenyl methylpolysiloxane (30 m 9 0.25 mm i.d.;
25 lm film thickness). Gas chromatographic conditions
were as follows: ethylbenzene [initial temperature: 90 °C
(2 min); first temperature rate: 5 °C/min until 140 °C;
second temperature rate: 50 °C/min until 250 °C]; cumene
[initial temperature: 90 °C (2 min); first temperature rate:
2.5 °C/min until 130 °C; second temperature rate: 50 °C/
min until 220 °C]; p-cymene and sec-butylbenzene [initial
temperature: 100 °C (2 min); first temperature rate: 3 °C/
min until 160 °C; second temperature rate: 50 °C/min until
260 °C]. The injector and interface temperatures were 250
and 270 °C, respectively. Aliquots were taken from the
reaction mixtures at regular intervals for peroxide deter-
mination, as described elsewhere [26, 42].
12. Jana SK, Kubota Y, Tatsumi T (2007) J Catal 247:214
13. Mohapatra SK, Selvam P (2007) J Catal 249:394
14. Figiel PJ, Sobczak JM (2007) New J Chem 31:1668
´
15. Hudlicky M (1990) Oxidations in organic chemistry. ACS,
Washington
16. Zhu M, Wei X, Li B, Yuan Y (2007) Tetrahedron Lett 48:9108
and references cited therein
´
´
´
17. Bejblova M, Prochazkova D, Cejka J (2009) ChemSusChem 2:486
18. Surberg H, Panten J (2006) Common fragrance and flavor
materials, 5th edn. Wiley-VCH, Weinheim
19. Hill CL, Prosser-McCartha CM (1995) Coord Chem Rev 143:407
20. Neumann R (1998) Prog Inorg Chem 47:317
21. Neumann R (2004) Liquid phase oxidation reactions catalyzed by
All products, except 6, 19, and 22, were identified by
GC-MS analysis and by co-injection of commercially
available standards. The hydroperoxides were identified by
their mass spectrum and by the triphenylphosphine test,
following a method described in the literature [49, 50]. An
excess of triphenylphosphine was added to the solutions at
the end of the reactions, and the reduction of the hydro-
peroxides to the corresponding alcohols was detected by
GC-MS analysis. This test was performed for all substrates.
The possibility of formation of peroxyimidic acid,
which would function as the oxidizing agent, was discarded
because acetamide was never observed in the products.
At the end of the reactions with [PW11Fe(H2O)O39]4-, a
drop of the reaction mixture was dried on a KBr pellet and
the infrared spectrum measured in order to assess the sta-
bility of the catalyst. Solutions with POM or FeCl3
concentrations in the range 2 9 10-5–5 9 10-4 M were
used for the tests with KSCN.
¨
polyoxometalates. In: Backvall J-E (ed) Modern oxidation
methods. Wiley-VCH, Weinheim
22. Kozhevnikov IV (1998) Chem Rev 98:171
23. Mizuno N, Misono M (1998) Chem Rev 98:199
24. Mizuno N, Yamaguchi K, Kamata K (2005) Coord Chem Rev
249:1944
25. Hill CL (ed) (2007) Polyoxometalates in catalysis. J Mol Catal A
Chem 262(1–2) (thematic issue)
26. Estrada AC, Simo˜es MMQ, Santos ICMS, Neves MGPMS, Silva
AMS, Cavaleiro JAS, Cavaleiro AMV (2009) Appl Catal A Gen
366:275
27. Estrada AC, Simo˜es MMQ, Santos ICMS, Neves MGPMS, Silva
AMS, Cavaleiro JAS, Cavaleiro AMV (2009) Catal Lett 128:281
28. Strukul G (1992) Catalytic oxidations with hydrogen peroxide as
oxidant. Kluwer, Dordrecht
29. Jones CW (1999) Applications of hydrogen peroxide and deriv-
atives. Royal Society of Chemistry, Cambridge
30. Sanderson WR (2000) Pure Appl Chem 72:1289
31. Noyori R, Aoki M, Sato K (2003) Chem Commun 1977
32. Mirkhani V, Moghadam M, Tangestaninejad S, Mohammadpoor-
Baltork I, Rasouli N (2008) Catal Commun 9:2171
33. Mirkhani V, Mirkhani V, Moghadam M, Tangestaninejad S,
Mohammadpoor-Baltork I, Rasouli N (2008) Catal Commun
9:2411
Acknowledgments Thanks are due to the University of Aveiro and
ˆ
FCT (Fundac¸a˜o para a Ciencia e a Tecnologia) for funding. A.C.E.
and I.C.M.S.S. also thank FCT for their grants.
34. Tangestaninejad S, Mirkhani V, Moghadam M, Mohammadpoor-
Baltork I, Shams E, Salavati H (2008) Ultrason Sonochem 15:438
123

Catalytic activity of iron-substituted polyoxotungstates in the oxidation of aromatic compounds
1235
35. Attanasio D, Orru D, Suber L (1989) J Mol Catal 57:L1
36. Neumann R, Vega M (1993) J Mol Catal 84:93
37. Kuznetsova NI, Kirillova NV, Kuznetsova LI, Smirnova MX,
Likholobov VA (2007) J Hazard Mater 146:569
38. Gamelas JA, Couto FAS, Trova˜o MCN, Cavaleiro AMV, Cav-
aleiro JAS, Pedrosa de Jesus JD (1999) Thermochim Acta
326:165
43. Santos ICMS, Simo˜es MMQ, Balula MSS, Neves MGPMS,
Cavaleiro JAS, Cavaleiro AMV (2008) Synlett 1623
44. Schmidt RJ (2005) Appl Catal A Gen 280:89
45. Nakamura R, Obora Y, Ishii Y (2008) Chem Commun 3417
46. Simo˜es MMQ, Conceic¸a˜o CMM, Gamelas JAF, Domingues
PMDN, Cavaleiro AMV, Cavaleiro JAS, Ferrer-Correia AJV,
Johnstone RAW (1999) J Mol Catal A Chem 144:461
47. Santos ICMS, Simo˜es MMQ, Neves MGPMS, Cavaleiro JAS,
Cavaleiro AMV (2003) Synlett 1643
39. Gamelas JAF, Soares MR, Ferreira A, Cavaleiro AMV (2003)
Inorg Chim Acta 342:16
40. Barton DHR, Salgueiro MC, MacKinnon J (1997) Tetrahedron
53:7417
48. Balula MS, Gamelas JA, Carapuc¸a HM, Cavaleiro AMV,
Schlindwein W (2004) Eur J Inorg Chem 619
41. Balula MSS, Santos ICMS, Simo˜es MMQ, Neves MGPMS,
Cavaleiro JAS, Cavaleiro AMV (2004) J Mol Catal A Chem
222:159
49. Shul’pin GB (2002) J Mol Catal A Chem 189:39
50. Shul’pin GB (2003) C R Chim 6:123
42. Santos ICMS, Gamelas JA, Balula MS, Simo˜es MMQ, Neves
MGPMS, Cavaleiro JAS, Cavaleiro AMV (2007) J Mol Catal A
Chem 262:41
123