8-Quinolinolato iron(III)-catalyzed oxygenation of cyclohexane with hydrogen peroxide under heating or visible light irradiation

Article abstract of DOI:10.1016/j.molcata.2013.11.022

A series of hexadentate binding 8-quinolinolato iron(III) (Q ₃Fe^III (1a-1e)) complexes were synthesized conveniently via the coordination of FeCl₃·6H₂O with 8-hydroxyquinoline (a) or its derivatives (b-e). These Q₃Fe ^III complexes were found to be active for the oxygenation of cyclohexane to cyclohexanol and cyclohexanone by aqueous hydrogen peroxide (H₂O₂) in acetonitrile under heating conditions. Among the Q₃Fe^III complexes examined, the 5-chlorinated 1b and the unsubstituted 1a showed a good accelerating effect on this oxygenation due to their good ability to activate H₂O₂, as supported by DFT B3LYP/6-311G (d) calculation and TGA characterization. Notably, the above-described catalysis reactions were accelerated to some extent under visible light irradiation, providing an equivalent to a slightly higher cyclohexane conversion with a similar product distribution compared to the corresponding thermal catalysis reactions. Moreover, this light-accelerated effect on cyclohexane oxygenation was more apparent upon the dihalogenated Q₃Fe^III complexes 1c-1e, these having a relatively small difference of LUMO and HOMO energies. The improved effect of light irradiation on substrate conversion also became more prominent in the Q₃Fe ^III-catalyzed oxygenation of other substrates like benzene, toluene, ethylbenzene and thioanisole with H₂O₂. Based on the UV-vis spectral characterization, a free radical mechanism for the present catalysis oxygenation system was proposed.

Full text of DOI:10.1016/j.molcata.2013.11.022

Journal of Molecular Catalysis A: Chemical 383–384 (2014) 46–52
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
8-Quinolinolato iron(III)-catalyzed oxygenation of cyclohexane with
hydrogen peroxide under heating or visible light irradiation
Yongjun Wang, Zaihui Fu∗, Xu Wen, Chunying Rong, Wenfeng Wu, Chao Zhang,
Jie Deng, Baohua Dai, Steven Robert Kirk, Dulin Yin
National & Local United Engineering Laboratory for New Petrochemical Materials & Fine Utilization of Resources, Key Laboratory of Resource
Fine-Processing and Advanced Materials of Hunan Province and Key Laboratory of Chemical Biology and Traditional Chinese Medicine Research (Ministry
of Education of China), College of Chemistry and Chemical Engineering, Hunan Normal University, Changsha 410081, China
a r t i c l e i n f o
a b s t r a c t
A series of hexadentate binding 8-quinolinolato iron(III) (Q3FeIII (1a–1e)) complexes were synthesized
conveniently via the coordination of FeCl3·6H2O with 8-hydroxyquinoline (a) or its derivatives (b–e).
These Q3FeIII complexes were found to be active for the oxygenation of cyclohexane to cyclohexanol and
cyclohexanone by aqueous hydrogen peroxide (H2O2) in acetonitrile under heating conditions. Among the
Q3FeIII complexes examined, the 5-chlorinated 1b and the unsubstituted 1a showed a good accelerating
effect on this oxygenation due to their good ability to activate H2O2, as supported by DFT B3LYP/6-
311G (d) calculation and TGA characterization. Notably, the above-described catalysis reactions were
accelerated to some extent under visible light irradiation, providing an equivalent to a slightly higher
cyclohexane conversion with a similar product distribution compared to the corresponding thermal catal-
ysis reactions. Moreover, this light-accelerated effect on cyclohexane oxygenation was more apparent
upon the dihalogenated Q3FeIII complexes 1c–1e, these having a relatively small difference of LUMO
and HOMO energies. The improved effect of light irradiation on substrate conversion also became more
prominent in the Q3FeIII-catalyzed oxygenation of other substrates like benzene, toluene, ethylbenzene
and thioanisole with H2O2. Based on the UV–vis spectral characterization, a free radical mechanism for
the present catalysis oxygenation system was proposed.
Article history:
Received 25 August 2013
Received in revised form
12 November 2013
Accepted 19 November 2013
Available online 27 November 2013
Keywords:
8-Quinolinolato iron(III) complexes
Thermal catalysis oxygenation
Photocatalysis oxygenation
Selective oxygenation of organic
compounds
Hydrogen peroxide
© 2013 Elsevier B.V. All rights reserved.
1. Introduction
metal complexes chiefly used as catalysts, metallo-porphyrin com-
plexes are currently of considerable interest to promote these
The selective oxygenation of organic compounds by hydrogen
peroxide (H2O2) and especially dioxygen (O2) or air continues to
be a very important and challenging process [1–5] due to the large
demand for their oxidative products. For example, cyclohexanone
and cyclohexanol are the important raw materials for the produc-
tion of nylons [6] and the synthesis of a variety of fine chemicals
[7]. In industry, the oxygenation of cyclohexane is carried out with
air using cobalt salt catalysts under relatively harsh conditions
(temperatures of 443–503 K and pressures of 10–15 bar) [8], and
cyclohexane conversion in this process is limited to lower than 10%
in order to obtain a relatively high selectivity for the goal products
(ca. 70%) [9,10]. Some efficient catalysis processes have been devel-
oped for selective oxygenation of cyclohexane and other organic
compounds under relatively mild conditions [10–17], and the use of
transition metal complexes as homogeneous catalysts is very effec-
tive and selective for these reactions [18–28]. Among the transition
oxygenation reactions [21,23–28]. However, the high cost of these
types of complexes still hampers their commercialized applica-
tion. In terms of atom efficiency and environment friendliness,
air and dioxygen are the most attractive oxidants for these oxy-
genations [17–30]. However, the direct oxygenation processes by
molecular oxygen generally need to employ harsh operating con-
ditions because the triplet nature of molecular oxygen hampers
the reaction with organic compounds in its singlet state, which
rarely leads to the chemistry being selective. When mild oxidation
conditions are used, most of these methods require the co-catalyst
N-hydroxyphthalimide [27] or photosensitization with a UV or vis-
ible light source [28–30], which may constitute serious economic
impediments for large-scale applications. Hydrogen peroxide is
an interesting oxidant for these oxygenations [31–36] due to the
facts of having mild operating conditions and high oxygenation
selectivity, as well as generating environmentally benign water as
by-product. It is also cheap and easy to handle.
Recently, developed by our groups, 8-quinolinolato man-
ganese(III) complexes (Q3MnIII) combined with hydrogen peroxide
(H2O2) constitute a valuable new family of heterogeneous systems
∗
Corresponding author. Tel.: +86 731 88872576; fax: +86 731 88872531.
E-mail addresses: fzhhnnu@126.com, claishenjun2010@yahoo.cn (Z. Fu).
1381-1169/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molcata.2013.11.022

Y. Wang et al. / Journal of Molecular Catalysis A: Chemical 383–384 (2014) 46–52
47
200
150
100
50
for the epoxidation of olefins [37,38], the selective oxidation of
alcohols [39], thioanisole [40] and ethylbenzene.[41]. The excellent
catalysis performance of Q3MnIII complexes should be due to their
hexadentate structure with the distorted octahedral geometry [38].
However, more cost-effective 8-quinolinolato iron(III) complexes
(Q3FeIII) show a poor activity for the epoxygenation of olefins with
H2O2 due to their symmetrical octahedral structure [38]. Here, for
the first time we report that visible light irradiation can be used to
significantly accelerate the selective oxygenation of cyclohexane
and other organic compounds by hydrogen peroxide (H2O2) upon
catalysis with Q3FeIII complexes.
1b
b
0
400035003000 1500
1000
500
2. Experiment
Wavenumber/ cm-1
2.1. Materials and apparatus
Fig. 1. FT-IR spectra of 5-chloro-HQ (b) and its Q3FeIII complex (1b).
Materials and reagents used in this study were cyclohex-
ane, benzene, toluene, ethylbenzene, thioanisole, n-hexanol,
acetonitrile (MeCN), tetrahydrofuran (THF), ethanol, iron chlo-
ride (FeCl3·6H2O), 30% aqueous hydrogen peroxide (H2O2),
8-hydroxyquinoline (HQ, a), 5-chloro-HQ (b), 5,7-dichloro-HQ (c),
5,7-dibromo-HQ (d), 5-chloro-7-iodo-HQ (e), all of which were of
analytical grade. Distilled water was used throughout this experi-
ment.
The transmission FT-IR spectrum of samples was recorded from
400 to 4000 cm−1 on a Nicolet Nexus 510 P FT-IR spectroscopy using
a KBr disc. Their solid and liquid UV–vis spectra were respectively
recorded from 200 to 800 nm on a DUV-370D spectrophotome-
ter (Shimadzu, Japan) with BaSO4 as a standard and an UV-2450
spectrophotometer (Shimadzu, Japan) with MeCN as a solvent.
Thermogravimetric analysis (TGA) was carried out in a flowing
N2 atmosphere (10 ml/min) at a heating rate of 20 ◦C/min on a
NETZSCH-STA 409PC.
the reaction solution increased to 35 ◦C because of the heating effect
of light irradiation. After the appointed heating or light irradiation
time had elapsed, the unreacted H2O2 was decomposed by MnO2
and the reaction mixture obtained from heating or light reaction
was filtrated to remove the catalyst and MnO2, the resulting filtrate
was analyzed using an Agilent 6890N gas chromatograph with a
DB-17 polysiloxane capillary column (30 m × 0.32 mm × 0.50 ␮m)
and flame ionization detector (FID) using n-hexanol as an internal
standard. Both the injector and detector temperature were 250 ◦C,
and the column temperature was 80–120 ◦C. The isolated products
were satisfactorily identified by comparing the MS spectra with
those of the authentic samples.
3. Results and discussion
3.1. Characterizations of Q3FeIII complexes
The FT-IR spectra of all the Q3FeIII complexes are very similar
to each other. Here, we give the FT-IR spectra of a representative
Q3FeIII complex 1b and its corresponding ligand b. As shown
in Fig. 1, some characteristic absorption bands for the ligand b
appeared at 3194, 1583 and 1500, 1472, 1415 cm−1, which are
assigned to the aromatic ꢀ(O H), ꢀ(C N) and ꢀ(C C) stretching
vibrations respectively [42–44]. Notably, these bands in the FT-IR
spectrum of the complex 1b exhibited different degree of shift
due to the chelate effect. Moreover, the disappearance of the
fundamental stretching frequency of the OH group in the 1b
suggests the occurrence of a deprotonation process as the ligand
reacts with iron ion.
2.2. Preparation and characterization of 8-quinolinolato iron(III)
complexes
The preparation procedure for 8-quinolinolato iron(III) com-
plexes (Q3FeIII) is described as follows. A solution of FeCl3·6H2O
in ethanol (1 M, 5 ml) was added drop wise to a solution of 8-HQ
(a, 15 mmol) or its derivative (b–e, 15 mmol) in 50 ml of anhydrous
THF. Then the reaction mixture was stirred magnetically for 4 h at
room temperature (20 ◦C). The resulting precipitate was filtrated
and washed with ethanol at least three times. After air-drying,
a black solid (denoted as 1a–1e, see Scheme 1) was obtained at
88–95% yield. The Fe content of these Q3FeIII complexes was deter-
mined by ethylenediaminetetraacetic acid disodium salt (EDTA)
complexometric titration and the results are listed in Table 1. It
is seen from Table 1 that the found values for the Fe content of
1a–1e are in good agreement with those calculated from the cor-
responding formulae of the hexadentate structures.
Fig. 2 gives the UV–vis diffuse reflectance (UV–vis DR) spectra of
two representative Q3FeIII complexes 1b, 1c and the corresponding
ligands b and c. In that, two ligands showed a strong higher energy
1.6
2.3. General procedure for oxygenation of organic compounds
under thermal or visible light conditions
c
1.4
1c
b
Thermal reaction procedure: 30% aqueous H2O2 (4 mmol) was
added to a mixture of acetonitrile (MeCN, 3 ml), organic substrate
(1 or 2 mmol) and catalyst Q3FeIII (0.01 mmol) all at once and
the reaction mixture was stirred magnetically for an appointed
time at 35 ◦C. The photoreaction procedure used was as follows:
the visible light-driven oxygenation was carried out with a self-
assembled photo-reactor equipped with a water-cooled condenser
(see our published work [30]). The above-mentioned reaction mix-
ture was irradiated continuously by a 35 W tungsten-bromine lamp
equipped with an UV light filter (Osram brand) from its inside at
room temperature (20 ◦C). During irradiation, the temperature of
1.2
1.0
1b
0.8
300
400
500
600
700
800
Wavenlength / nm
Fig. 2. UV–vis diffuse reflectance spectra of ligands b, c and both the Q3FeIII com-
plexes 1b, 1c.

48
Y. Wang et al. / Journal of Molecular Catalysis A: Chemical 383–384 (2014) 46–52
the calculations were conducted on the Gaussian-03 version C02
package with tight self-consistent field (SCF) convergence and
ultrafine integration grids. The calculated Fe O(N) bond distances,
the highest occupied and lowest unoccupied molecular orbital
(HOMO and LUMO) energies for the Q3FeIII complexes are listed
1b
1a
in Table 1. It is seen from Table 1 that the three Fe O or Fe
N
bonds of each complex were exactly equal to each other in length,
indicating that the Q3FeIII complexes possess a symmetrically hex-
adentate structure. The HOMO energy for these Q3FeIII complexes
followed an increasing sequence of 1a > 1b > 1e > 1c ꢀ 1d, implying
that the Q3FeIII complexes 1b and especially 1a are more suscep-
tible to electrophilic attack of H2O2 than the three dihalogenated
Q3FeIII complexes (1c–1e) under heating. The difference (band gap)
of the LUMO and HOMO energies followed a decreasing sequence
of 1e < 1c < 1d < 1b < 1a, expecting that the dihalogenated Q3FeIII
complexes should be excited by a visible light more easily than
the Q3FeIII complexes 1a and 1b, which may lead to a good light-
promoted effect on the oxygenation of organic compounds by H2O2.
1c
1d
1e
100 200 300 400 500 600 700 800 900 1000
Temperature / oC
Fig. 3. Differential thermogravimetry (DTG) curves of Q3FeIII complexes 1a–1e.
band near 328 nm, which should be due to the ␲→␲* transitions
of ligands [45]. Both the complexes also exhibited two broad and
weak bands in a lower energy region of 450–640 nm. The band in
the 450–490 nm region should be assigned to the ligand to metal
charge-transfer transition (LMCT), as reported for other FeIII com-
plexes [46]. Another band in the 570–640 nm region is responsible
for the intense colors of the complex [47]. Notably, the two low
energy bands of 1c were shifted to a higher wavelength region
compared to both those of the 1b. This illustrates that a charge-
transfer transition of the ligand to FeIII ion occurs easily upon the
dihalogened Q3FeIII complex 1c.
3.2. Cyclohexane oxygenation
Table 2 lists the data for the Q3FeIII complexes-catalyzed cyclo-
hexane oxygenation by aqueous hydrogen peroxide (H2O2) in
acetonitrile (MeCN) at 35 ◦C under heating and visible light irradia-
tion conditions. It is seen from Table 2 that all the Q3FeIII complexes
were active for this oxygenation whether heating or light irra-
diation was used, providing cyclohexanol and cyclohexanone as
the oxygenation products. Notably, the catalytic performance of
these Q3FeIII complexes was significantly influenced by their sub-
stituents. Among the Q3FeIII complexes examined, the 1a showed
the fastest catalysis rate for this oxygenation under heating, which
could achieve its highest conversion at 2 h (8.3%, entry 1). Entry
2 shows that the 1b accelerated this oxygenation under heating,
providing ca. 16.0% of cyclohexane conversion at 6 h. In compari-
son with the 1a and 1b, the three dihalogenated Q3FeIII complexes
(1c–1e) showed an obviously slow catalysis rate for this oxygen-
ation under heating and a long reaction time (12–16 h) was usually
needed to achieve the highest conversion (ca. 12.6% at 16 h for 1c,
5.2% at 16 h for 1d and 11.3% at 12 h for 1e, see entries 3–5).
In comparison with the heating-triggered catalysis oxygenation
described above, the Q3FeIII complex-catalyzed cyclohexane oxy-
genation by H2O2 was acceleratedto some extent under visible light
irradiation, which could lead to the significantly shortened reaction
Thermal stability of these Q3FeIII complexes was checked by the
thermo-gravimetric analysis (TGA) technique and their differential
thermogravimetry (DTG) curves are shown in Fig. 3. The DTG curves
of these Q3FeIII complexes exhibited two to three main peaks for
loss of weight in the range 100–550 ◦C, which would originate from
the gradual decomposition of their ligands. Among the Q3FeIII com-
plexes examined, the decomposition temperatures for 1c, 1e and
especially 1d were higher than those for 1b and 1a, illustrating that
the three dihalogenated Q3FeIII complexes possess a better coordi-
nation stability than the 5-chlorinated 1b and the unsubstituted 1a
ones.
The optimized structure of the Q3FeIII complexes was investi-
gated by means of the B3LYP method of density functional theory
with 6-311G(d) basis set (DFT B3LYP/6-311G (d)), in which all
Table 1
The iron content of Q3FeIII complexes measured by chemical titration and their Fe O (N) bond distances, HOMO and LUMO energies calculated by DFT B3LYP/6-311+G (d)
basis set.a
Q3FeIII
Fe contenta (wt%)
Fe O (N) bond distancesb (Å)
Orbital energyb (kcal mol−1
)
REBd (kcal mol−1
)
Fe O1
Fe N1
Fe O2
Fe N2
Fe O3
Fe N3
HOMO
LUMO
Band gapc
11.46
1.944
2.254
1.944
2.254
1.944
2.254
−124.81
−38.23
1a
1b
1c
1d
86.58
84.01
82.68
83.34
81.80
3.05
2.82
6.12
5.90
6.12
11.42e
9.46
1.942
2.255
1.942
2.255
1.942
2.255
−133.46
−49.45
9.49e
8.01
1.944
2.249
1.944
2.249
1.944
2.249
−138.94
−56.26
8.06e
5.77
1.943
2.248
1.943
2.248
1.943
2.248
−139.13
−55.79
5.82e
5.73
1.944
2.248
1.944
2.248
1.944
2.248
−136.78
−54.98
1e
5.77e
a
The iron content of Q3FeIII complexes was measured by EDTA complexometric titration.
b
c
The Fe O or Fe N bond distances, HOMO and LUMO energies for the Q3FeIII complexes were calculated by the DFT B3LYP/6-311+G (d) basis set.
Band gap = LUMO energy − HOMO energy.
d
e
REB (reaction energy barrier) = The steady-state energy of the species 2 in Scheme 1 − (the steady-state energy of H2O2 + the steady-state energy of Q3FeIII).
The calculated values from the corresponding formulae of the hexadentate Q3FeIII complexes.

Y. Wang et al. / Journal of Molecular Catalysis A: Chemical 383–384 (2014) 46–52
49
Table 2
Various Q3FeIII complexes-catalyzed cyclohexane oxygenation by H2O2 in MeCN at 35 ◦C under heating or visible light irradiation.a
Entry
Q3FeIII
Under heating condition
Entry
Under visible light condition
Time (h)
Conv. (%)
Yield (mol%)
Kb
Time (h)
Conv. (%)
Yield (mol%)
Kb
Ac
Ac
1
2
3
4
5
1a
1b
1c
1d
1e
2.0
6.0
16.0
16.0
12.0
8.3
16.0
12.6
5.2
4.1
8.5
5.1
2.2
6.0
4.2
7.5
7.8
3.0
5.3
1ꢀ
2ꢀ
3ꢀ
4ꢀ
5ꢀ
2.0
4.0
8.0
8.0
8.0
9.5
15.1
17.1
5.3
4.8
8.3
8.4
2.2
5.3
4.7
6.8
8.7
3.1
4.9
11.3
10.2
a
Reaction conditions: cyclohexane (2 mmol), catalyst (0.01 mmol), 30% H2O2 (4 mmol), MeCN (3 ml) at 35 ◦C with stirring. Conversion and selectivity were determined by
GC.
b
K indicated cyclohexanone.
A indicated cyclohexanol.
c
time to achieve the highest conversion (see entries 1ꢀ–5ꢀ in Table 2).
hardly occurred in 8 h, after that, it proceeded at a higher rate and
eventually achieved ca. 11.3% cyclohexane conversion at 12 h (see
curve 1 in Fig. 5). This indicates that the 1e-catalyzed reaction has
an induction period due to a good coordination stability of 1e. If the
1e-catalyzed reaction described above was carried out under visible
light, its induction period was shortened to about 4 h and its final
conversion was ca. 10.2% at 8 h (see curve 2 in Fig. 5). Additionally,
the oxygenation of cyclohexane by H2O2 over the other two dihalo-
genated Q3FeIII complexes 1c and 1d was also found to have a long
induction period under heating and could be significantly accel-
erated under visible light irradiation (the results were not given
here). It is concluded from these findings that (i) the accelerated
effect of Q3FeIII complexes on this oxygenation is basically propor-
tional to their HOMO energy and inversely proportional to their
coordination stability, (ii) the dihalogenated Q3FeIII complexes,
with a relatively low band gap energy, are easily excited by visi-
ble light, which can lead to an outstanding visible light-accelerated
effect on the oxygenation of cyclohexane with H2O2 catalyzed by
them.
Finally, the oxygenation of other substrates like benzene,
toluene, ethylbenzene and thioanisole by H2O2 in MeCN at 35 ◦C
under heating or visible light was examined using 1b and 1e as a
catalyst, respectively, and the results are summarized in Table 3.
As shown in Table 3, the catalyst 1b still showed a good acceler-
ating effect on these oxygenations under heating condition and
achieved a relatively high conversion in a short reaction time
(2 h), respectively providing phenol for benzene, benzaldehyde and
o/m-cresols for toluene, ␣-phenethylalcohol, acetophenone and
phenylacetaldehyde for ethylbenzene, methyl phenyl sulfoxide for
thioanisole as their main oxygenation products. The reactivity of
various substrates over 1b followed an increasing sequence of
Moreover, the accelerating effect of visible light on this oxygen-
ation was more strikingly apparent upon the three dihalogenated
Q3FeIII complexes. For example, the 1b-catalyzed oxygenation
could achieve its highest cyclohexane conversion under sustained
visible light irradiation of 4 h (ca. 15.1%, entry 2ꢀ) and this time was
shortened by 2 h compared to that under heating. The 1c-catalyzed
oxygenation under visible light achieved ca. 17.1% of cyclohexane
conversion at 8 h (entry 3ꢀ), its reaction time was shortened by half
and its conversion increased by 4.5% compared to both those under
heating.
In the following experiments, the effect of reaction time on
cyclohexane oxygenation was checked using the 1b and 1e as two
representatives under heating and visible light, respectively. Fig. 4
illustrates that the curve of cyclohexane conversion over 1b under
heating climbed slowly with time and achieved ca. 16.0% at 6 h.
After that, the conversion hardly increased with the time. This
is likely due to the deactivation of 1b, which can be confirmed
from a fact that the dissolution and decolourization of 1b was
observed clearly after 6 h. Cyclohexanol selectivity continuously
and slowly decreased with time, with a concomitant increase in
cyclohexanone selectivity (see curves 3 and 4 in Fig. 4). The effect
of reaction time on the 1b-catalyzed cyclohexane oxygenation by
H2O2 under visible light was very similar to that under heating, but
its conversion curve in 1–2 h climbed faster than that under heat-
ing and achieved ca. 15.1% at 4 h. After that, the conversion slightly
decreased with time (see curve 2 in Fig. 4), illustrating that when a
long reaction time was used, the deep oxygenation of its oxidative
products cyclohexanol and cyclohexanone occurred more easily
under light irradiation than under heating. Fig. 5 illustrates that
the 1e-catalyzed cyclohexane oxygenation by H2O2 under heating
60
60
1b
1e
3
5
4
50
3
50
5
4
6
6
40
1: cyclohexane conversion under heating
2: cyclohexane conversion under visible light
3: cyclohexanol selectivity under heating
4: cyclohexanol selectivity under visible light
5: cyclohexanone selectivity under visible light
6: cyclohexanone selectivity under heating
40
1: Conversion under heating
2: Conversion under visible light
3: Cyclohexanol selectivity under heating
4:Cyclohexanone selectivity under heating
5:Cyclohexanol selectivity under visible light
6: Cyclohexanone selectivity under visible light
30
30
20
10
20
10
0
2
2
1
1
0
1
2
3
4
5
6
7
8
9
2
4
6
8
10
12
14
Time / h
Time / h
Fig. 4. Reaction time-dependence of conversion and selectivity for the 1b
(0.01 mmol)-catalyzed oxygenation of cyclohexane (2 mmol) by H2O2 (4 mmol) at
35 ◦C in MeCN under heating or light.
Fig. 5. Reaction time-dependence of conversion and selectivity for the 1e
(0.01 mmol)-catalyzed oxygenation of cyclohexane (2 mmol) by H2O2 (4 mmol) at
35 ◦C in MeCN under heating or light.

50
Y. Wang et al. / Journal of Molecular Catalysis A: Chemical 383–384 (2014) 46–52
Table 3
Q3FeIII-catalyzed oxygenation of other organic compounds by H2O2 in MeCN at 35 ◦C.a
Entry
Q3FeIII
Substrate
Exited condition
Time (h)
Conv. (%)
Main products (yield, mol%)
1
2
3
4
5
6
7
8
Heating
Light
Heating
Light
Heating
Light
Heating
Light
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.5
8.9
8.6
9.1
12.5
30.3
65.0
82.8
Phenol (2.5)
Phenol (8.9)
Benzene
Benzaldehyde (3.4), o/m-cresol (3.7)
Benzaldehyde (4.1), o/m-cresol (3.9)
PEA (2.7), ACP (6.4), PEAD (2.5)
PEA (1.4), ACP (21.1), PEAD (6.1)
Methylphenyl sulfoxide (62.9)
Methylphenyl sulfoxide (81.1)
Tolueneb
1b
Ethylbenzenec
Thioanisoled
9
10
11
12
13
14
15
16
Heating
Light
Heating
Light
Heating
Light
Heating
Light
12.0
12.0
12.0
12.0
12.0
12.0
12.0
12.0
0.3
7.1
0.4
13.5
0.8
35.5
59.2
98.2
Phenol (0.3)
Phenol (7.1)
Benzaldehyde (0.4)
Benzaldehyde (7.6), o/m-cresol (3.2)
PEA (0.1), ACP (0.6), PEAD (0.1)
PEA (5.3), ACP (22.7), PEAD (6.1)
Methylphenyl sulfoxide (58.8)
Methylphenyl sulfoxide (56.7)
Benzene
Tolueneb
1e
Ethylbenzenec
Thioanisoled
a
Reaction conditions: substrate (1 mmol), catalyst (0.02 mmol), 30% H2O2 (4 mmol), MeCN (3 ml), reaction temperature, 35 ◦C. Main oxygenation products were identified
by GC–MS, conversion and selectivity were determined by GC.
b
A small amount of benzoquinone was detected.
PEA, ACP and PEAD presented ␣-phenethylalcohol, acetophenone and phenylacetaldehyde, respectively, in addition, a small amount of benzaldehyde was detected.
A small amount of methylphenyl sulfone was detected.
c
d
thioanisole > ethylbenzene > toluene > benzene (see entries 1, 3, 5
and 7). In comparison with the above-described reactions under
heating, these reactions catalyzed by 1b were improved to some
extent under visible light irradiation, which could lead to an
increase in the conversion of substrates (entries 2, 4, 6 and 8). For
example, toluene and ethylbenzene conversions over 1b under vis-
ible light were ca. 9.1 and 30.3%, being 0.06- and 1.42-fold higher
than both those obtained under heating, respectively (entries 4 vs.
3 and 6 vs. 5). Notably, the heating-driven oxygenation of benzene,
toluene or ethylbenzene over 1e hardly occurred in the reaction
time examined (12 h, see entries 9, 11 and 13), further illustrat-
ing that the stable 1e has a very small accelerating effect on these
oxygenations. If the 1e-catalyzed reactions were carried out under
visible light irradiation, the conversion of these substrates was
improved significantly (see entries 10, 12 and 14). For example,
benzene and ethylbenzene conversions over 1e were ca. 7.1 and
35.5% under visible light, being 22.67- and 43.38-fold higher than
both those obtained under the heating, respectively (entries 10
vs. 9 and 14 vs. 13). This further indicates that the dihalogenated
1e complex is easily excited by visible light, which can lead to its
good activity for the oxygenation of these substrates by H2O2 under
visible light.
3.3. Catalytic mechanism
In order to investigate the reaction pathway, a stepwise overlay
of UV–vis spectra for the 1b catalysis system in a MeCN medium
at room temperature (20 ◦C) is illustrated in Fig. 6. As previously
described in Fig. 3, the 1b showed two weak and broad coordination
bands at ꢁmax ∼465 and ∼596 nm in the presence of cyclohexane
(curve A in Fig. 6). Followed by the addition of H2O2 to the A solu-
tion, an obvious decrease in the two bands was observed after 5 min
(curve B1), illustrating that the 1b can react rapidly with H2O2.
After that, both the bands continuously decayed with prolonging
the interaction time of H2O2 (see curves B2–B5). On the other hand,
the absorbance near 350–380 nm grew gradually with time (see
inset in Fig. 6), which is indicative of the phenoxyl radical forma-
tion [46]. This spectral change likely implies that the 1b induces
homolytic cleavage of the O O bond of H2O2 to generate free radi-
cals in solution, which leads to spurious reactions that can degrade
the catalyst [48], as observed by the above catalytic experiments.
Based on the above findings, we propose that the Q3FeIII (1 in
Scheme 1) directly reacts with H2O2 to form a pentadente Q3FeIII
OOH with a pentadentate hydroxyl group (2) through opening one
of the Fe O bonds under heating or visible light irradiation. Then,
-
OH
O
O
H2O2
OH
N
O
N
OH
N
N
III
III
Homolysis
III
O
Fe
N
N
O
Fe
N
N
OH
OH
Degradation
O
Cy=O
OH
Fe
O
O
O
N
or
1
2
3
3 or
CyH, Cyclohexane
CyOH, Cyclohexanol
Cy=O, Cyclohexanone
OH
OH
N
CyOH
N
Cy
OH
CyH
-H2O
III
O
Fe
O
N
1a-1e: Q3FeIII
R1
O
N
O
a: R1=H, R2=H
b: R1=Cl,R2= H
c: R1= Cl, R2= Cl
d: R1= Br, R2= Br
e: R1= Cl, R2= I
N
III
Fe
N
O
R2
N
O
N
O
Scheme 1. Proposed catalysis cycle for the Q3FeIII-catalyzed oxygenation of cyclohexane by H2O2.

Y. Wang et al. / Journal of Molecular Catalysis A: Chemical 383–384 (2014) 46–52
51
exploiting other activation methods for the Q3FeIII complexes-
catalyzed selective oxygenation of various organic compounds by
H2O2 and especially dioxygen or air.
0.6
0.5
0.4
0.3
0.2
0.1
A
1.5
1.2
0.9
0.6
B1
B2
B3
B4
B5
Acknowledgements
320
340
360
380
400
We acknowledge the financial support for this work by the
National Natural Science Fund of China (20873040), the Special-
ized Research Fund for the Doctoral Program of Higher Education
(20124306110005), the Natural Science Fund of Hunan Province
(10JJ2007), the Innovation Platform Open Fund of Hunan College
(11K044), Program for Science and Technology Innovative Research
Team in Higher Educational Institutions of Hunan Province, the
100 Talents Program of Hunan Province and the Hunan Provincial
Innovation Foundation for Postgraduate of China (CX2013B207 and
CX2012B208).
400
500
600
Wavelength / nm
700
800
Fig. 6. Liquid UV–vis spectra of (A) a saturated solution of Q3FeIII 1b in MeCN
(6.0 × 10−6 M, 10 ml) and cyclohexane (27 ␮mol); B1: 5 min later of adding 13 ␮mol
30% H2O2 to the solution A; B2: 15 min later; B3: 35 min later; B4: 55 min later; B5:
110 min later. Inset is the amplified traces in 320–400 nm.
References
the 2 may undergo a homolytic cleavage of its O OH bond to
generate a Q3FeIIIO• (3) and •OH radicals in solution, as reported
in the 8-quinolinolato MnIII-catalyzed thioanisole oxygenation by
H2O2 in the absence of additives [40]. Finally, the 3 or •OH radical
may directly oxidize cyclohexane or other organic substrates to the
corresponding oxygenation products and simultaneously give the
regenerated catalyst. In addition, the 3 or •OH radical, as observed
by the above spectral change, may oxidize the opened ligand’s phe-
nolic OH of 2 to the corresponding phenoxyl radical, which can lead
to the catalyst’s degradation. Evidently, the catalytic rate of Q3FeIII
is inversely proportional to its reaction energy barrier (REB) with
H2O2. Table 1 lists the REB data of the Q3FeIII complexes obtained
by a DFT B3LYP/6-311G (d) calculation method (for the computa-
tional formula see the annotation d in Table 1). In that, the REB
value for the Q3FeIII complexes followed a decreasing sequence
of 1b < 1a < 1d < 1c = 1e, further supporting the idea that the com-
plexes 1a and 1b more easily suffer from an electrophilic attack of
H2O2 and open one of the Fe O bonds to form the intermediate 2
compared to the three dihalogenated Q3FeIII complexes, which can
lead to a good accelerating effect on the present oxygenations. In
addition, the Q3FeIII complexes can absorb a visible light to form
the corresponding excitation states, and the excited Q3FeIII com-
plexes may react with H2O2 to form the intermediate 2 more easily
than their ground-state counterparts, which can lead to the visible
light significantly accelerating these oxygenations catalyzed by the
Q3FeIII complexes. Moreover, the dihalogenated Q3FeIII complexes
with a lower band gap energy (see Table 1) are excited by visible
light to form their corresponding excitation states more easily than
the 1a and 1b complexes with a higher band gap energy, so that the
visible light-accelerated effect on the present oxygenations is more
strikingly apparent upon them, as supported by the above reaction
results.
[1] K.S. Suslick, C.L. Hill (Eds.), Activation and Functionalization of Alkenes, Wiley,
New York, 1989, pp. 219–236.
[2] K.A. Jorgensen, Chem. Rev. 89 (1989) 431–458.
[3] B. Meunier, Chem. Rev. 92 (1992) 1411–1456.
[4] U. Schuchardt, W.A. Carvalho, E.V. Spinace, Synlett 10 (1993) 713–718.
[5] U. Schuchardt, D. Cardoso, R. Sercheli, R. Pereira, R.S. da Cruz, M.C. Guerreiro,
D. Mandelli, E.V. Spinace, E.L. Pires, Appl. Catal. A: Gen. 211 (2001) 1–17.
[6] G.W. Parshall, S.D. Ittel, Homogenous Catalysis, second ed., Wiley, New York,
1992 (Chapter 10).
[7] H.H. Szmant, Organic Building Blocks of the Chemical Industry, Wiley, New
York, 1989.
[8] H.A. Wittcoff, B.G. Reuben, Industrial Organic Chemicals, John Wiley & Sons
Inc., New York, 1996.
[9] R.A. Sheldon, J.K. Kochi, Metal-Catalyzed Oxygenation of Organic Compounds,
Academic Press, New York, 1981 (Chapters 2 and 11).
[10] U. Schuchardt, R. Pereira, M. Rufo, J. Mol. Catal. A: Chem. 135 (1998) 257.
[11] C.L. Hill, B.C. Schardt, J. Am. Chem. Soc. 102 (1980) 6374–6375.
[12] S. Hai, B. Fritz, F. Heinz, J. Am. Chem. Soc. 116 (1994) 7951–7952.
[13] D.L. Vanoppen, D.E. De Vos, M.J. Genet, P.G. Rouxhet, P.A. Jacobs, Angew. Chem.
Int. Ed. Engl. 34 (1995) 560–563.
[14] J.M. Vincent, B.B. Stephane, C. Pierre, J.B. Verlhac, J. Chem. Soc. Dalton Trans.
(1999) 1913–1914.
[15] C.C. Guo, H.P. Li, J.B. Xu, J. Catal. 185 (1999) 345–351.
[16] X.B. Zhang, C.C. Guo, J.B. Xu, R.Q. Yu, J. Mol. Catal. A: Chem. 154 (2000)
31–38.
[17] J.D. Chen, R.A. Sheldon, J. Catal. 153 (1995) 1–8.
[18] E.V. Spinace, U. Schuchardt, D. Cardoso, Appl. Catal. A: Gen. 185 (1999)
L193–L197.
[19] A. Ebadi, N. Safari, M.H. Peyrovi, Appl. Catal. A: Gen. 321 (2007) 135–139.
[20] C.G. Homer, S. Dharmarathna, C. Eric, J. Phys. Chem. C 116 (2012) 12066–12078.
[21] Y. Li, M.Z. Wu, W. Liu, Z.Z. Yi, J.C. Zhang, Catal. Lett. 123 (2008) 123–128.
[22] J. Haber, L. Matachowski, K. Pamin, J. Połtowicz, J. Mol. Catal. A: Chem. 162
(2000) 105–109.
[23] C.C. Guo, M.F. Chu, Y. Liu, D.C. Guo, X.Q. Liu, Appl. Catal. A: Gen. 246 (2003)
303–309.
[24] B.C. Hu, W.Y. Zhou, D.S. Ma, Z.L. Liu, Catal. Commun. 10 (2008) 83–85.
[25] W.Y. Zhou, B.C. Hu, Z.L. Liu, Appl. Catal. A: Gen. 358 (2009) 136–140.
[26] D.S. Ma, B.C. Hu, C.X. Lu, Catal. Commun. 10 (2009) 781–783.
[27] Y. Ishii, J. Mol. Catal. A: Chem. 117 (1997) 123–137.
[28] G. Lu, H. Gao, J. Suo, S. Li, J. Chem. Soc. Chem. Commun. (1994) 2423–2424.
[29] P. Du, J.A. Moulijn, G. Mul, J. Catal. 238 (2006) 342–352.
[30] W.F. Wu, X.L. He, Z.F. Fu, Y.C. Liu, Y.L. Wang, X.L. Gong, X.L. Deng, H.T. Wu, Y.H.
Zou, N.Y. Yu, D.L. Yin, J. Catal. 286 (2012) 6–12.
[31] U.R. Pillai, E. Sahle-Demessie, Chem. Commun. (Camb.) 21 (2002)
2142–2143.
[32] M.C. Esmelindro, E.G. Oestreicher, H. Marquez-Alvarez, C. Dariva, S.M. Egues, C.
Fernandes, A.J. Bortoluzzi, V. Drago, O.A. Antunes, J. Inorg. Biochem. 99 (2005)
2054–2061.
[33] H.P. Liu, Z.H. Fu, D.L. Yin, D.H. Yin, H.Y. Liao, Catal. Commun.
638–643.
[34] M.R. Maurya, A. Arya, P. Adao, P.J. Costa, Appl. Catal. A: Gen. 351 (2008)
239–252.
[35] X.Q. Wang, J.P. Wu, M.W. Zhao, Y.F. Lv, G.Y. Li, C.W. Hu, J. Phys. Chem. C 113
(2009) 14270–14278.
[36] F.L. Liu, Z.H. Fu, Y.C. Liu, C.L. Lu, Y.Y. Wu, F. Xie, Z.P. Ye, X.P. Zhou, D.L. Yin, Ind.
Eng. Chem. Res. 49 (2010) 2533–2536.
4. Conclusion
In summary, for the first time we have designed and syn-
thesized a series of 8-quinolinolato iron(III) complexes (Q3FeIII
)
with a symmetrical octahedral geometry. The proposed Q3FeIII
complexes can catalyze the selective oxygenation of cyclohexane,
benzene, toluene, ethylbenzene and thioanisole by H2O2 in MeCN
under heating, and their catalysis rate is strongly dependent on
their ligand’s substituents. More importantly, these oxygenations
catalyzed by the Q3FeIII complexes are accelerated to some extent
under visible light irradiation, which can lead to improved catalysis
efficiency. Additionally, this accelerating effect of visible light on
these reactions is obviously influenced by the substituents of
Q3FeIII complexes. Having these findings, we are interested in
6 (2005)
[37] S. Zhong, Y.M. Tan, Z.H. Fu, Q.J. Xie, F. Xie, X.P. Zhou, Z.P. Ye, G.S. Peng, D.L. Yin,
Adv. Synth. Catal. 350 (2008) 802–806.
[38] S. Zhong, Y.M. Tan, Z.H. Fu, Q.J. Xie, F. Xie, X.P. Zhou, Z.P. Ye, G.S. Peng, D.L. Yin,
J. Catal. 256 (2008) 154–158.

52
Y. Wang et al. / Journal of Molecular Catalysis A: Chemical 383–384 (2014) 46–52
[39] Z.P. Ye, Z.H. Fu, S. Zhong, F. Xie, X.P. Zhou, F.L. Liu, D.L. Yin, J. Catal. 261 (2009)
110–115.
[44] H.F. Aly, F.M. Abdel Kerim, A.T. Kandil, J. Inorg. Nucl. Chem. 33 (1971)
4340–4344.
[40] X. Fang, Z.H. Fu, S. Zhong, Z.P. Ye, X.P. Zhou, F.L. Liu, C.Y. Rong, L.Q. Mao, D.L.
Yin, J. Mol. Catal. A: Chem. 307 (2009) 93–97.
[45] C.Y. Yue, F.L. Jiang, Y. Xu, M.C. Hong, J. Cryst. Growth
2721–2728.
8 (2008)
[41] C.L. Lu, Z.H. Fu, Y.C. Liu, F.L. Liu, Y.Y. Wu, J.W. Qin, X.L. He, D.L. Yin, J. Mol. Catal.
A: Chem. 331 (2010) 106–111.
[46] T. Kurahashi, Y. Kobayashi, S. Nagatomo, T. Tosha, T. Kitagawa, Inorg. Chem. 44
(2005) 8156–8166.
[42] E.T. James, T.S. Donald, Inorg. Chem. 3 (1964) 692–696.
[43] O. Noboru, N. Khzuo, Inorg. Chem. 10 (1971) 798–805.
[47] M.S. Shongwe, B.A. Rashdi, H. Adams, Inorg. Chem. 46 (2007) 9558–9568.
[48] B.S. Lane, K. Burgess, Chem. Rev. 103 (2003) 2457–2473.