Enhanced aerobic epoxidation of styrene with copper(II), cobalt(II), iron(III), or oxovanadium(IV) salen complexes immobilized onto carbon-coated Fe3O4 nanoparticles hybridized with graphene sheets

Article abstract of DOI:10.1002/cplu.201300424

Core-shell structural magnetic nanocatalysts, involving carbon-coated magnetic Fe₃O₄ nanoparticles supported on graphene oxide sheets (mildly reduced graphene oxide (MRGO)/Fe₃O₄@C) and carbonaceous coatings on magnetic Fe₃O₄ nanoparticles (Fe₃O₄@C), for immobilizing metal (Cu^II, Co^II, Fe^III or VO^IV)-salen complexes were prepared by a facile, green, and efficient chemical approach and first applied in the aerobic epoxidation of styrene with air as the oxidizing agent. The prepared catalysts were characterized by different techniques, such as XRD, TEM, FTIR spectroscopy, inductively coupled plasma atomic emission spectroscopy, TEM, thermogravimetric analysis, Raman spectroscopy, and X-ray photoelectron spectroscopy. The magnetic nanocatalysts MRGO/Fe₃O ₄@C-salen-M (M=Cu, Co, Fe, or V) exhibited higher catalytic reactivity than that of the corresponding Fe₃O₄@C-salen-M (M=Cu, Co, Fe, or V) species possibly owing to site isolation, good dispersion of magnetic Fe₃O₄ nanoparticles, and an efficiently synergistic effect between the metal complexes and the supports. Furthermore, MRGO/Fe₃O₄@C-salen-Co exhibited up to 92.3% conversion of styrene and 59.2% selectivity to the styrene oxide upon using air as the oxidizing agent and showed good recoverability with negligible loss in activity and selectivity within successive runs owing to superparamagnetism. Most importantly, these materials can be easily recovered by simple magnetic separation. Extra support: Core-shell-structured magnetic nanocatalysts have been prepared and used as efficient heterogeneous catalysts for the aerobic oxidation of styrene. Mildly reduced graphene oxide (MRGO; see picture)/Fe ₃O₄@C-salen-Co showed over 91% conversion of styrene and over 52% selectivity to the styrene oxide. The materials can be easily recovered through simple magnetic separation. Copyright

Full text of DOI:10.1002/cplu.201300424

CHEMPLUSCHEM
FULL PAPERS
DOI: 10.1002/cplu.201300424
Enhanced Aerobic Epoxidation of Styrene with Copper(II),
Cobalt(II), Iron(III), or Oxovanadium(IV) Salen Complexes
Immobilized onto Carbon-Coated Fe3O4 Nanoparticles
Hybridized with Graphene Sheets
Zhifang Li,[a] Shujie Wu,[a] Dafang Zheng,[b] Hong Ding,[b] Xiufang Wang,[a] Xiaoyuan Yang,[a]
Qisheng Huo,[b] Jingqi Guan,*[a] and Qiubin Kan*[a]
Core–shell structural magnetic nanocatalysts, involving carbon-
coated magnetic Fe3O4 nanoparticles supported on graphene
oxide sheets (mildly reduced graphene oxide (MRGO)/
Fe3O4@C) and carbonaceous coatings on magnetic Fe3O4 nano-
particles (Fe3O4@C), for immobilizing metal (CuII, CoII, FeIII or
VOIV)–salen complexes were prepared by a facile, green, and
efficient chemical approach and first applied in the aerobic ep-
oxidation of styrene with air as the oxidizing agent. The pre-
pared catalysts were characterized by different techniques,
such as XRD, TEM, FTIR spectroscopy, inductively coupled
plasma atomic emission spectroscopy, TEM, thermogravimetric
analysis, Raman spectroscopy, and X-ray photoelectron spec-
troscopy. The magnetic nanocatalysts MRGO/Fe3O4@C–salen–M
(M=Cu, Co, Fe, or V) exhibited higher catalytic reactivity than
that of the corresponding Fe3O4@C–salen–M (M=Cu, Co, Fe,
or V) species possibly owing to site isolation, good dispersion
of magnetic Fe3O4 nanoparticles, and an efficiently synergistic
effect between the metal complexes and the supports. Fur-
thermore, MRGO/Fe3O4@C–salen–Co exhibited up to 92.3%
conversion of styrene and 59.2% selectivity to the styrene
oxide upon using air as the oxidizing agent and showed good
recoverability with negligible loss in activity and selectivity
within successive runs owing to superparamagnetism. Most
importantly, these materials can be easily recovered by simple
magnetic separation.
Introduction
Current interest in the selective epoxidation of alkenes into the
corresponding epoxides, to a great extent, has been fueled by
its utility in flexible intermediates[1] and precursors to many
useful chemical products,[2] such as drugs, agrochemicals, food
additives, perfume materials, and sweeteners.[3] Generally, H2O2,
tert-butyl hydroperoxide (TBHP), or peracids were used as oxi-
dizing agents in the oxidation of alkenes.[4] These oxidizing
agents are corrosive and hazardous in nature and might result
in serious pollution and be wasteful. To meet the increasing
demand for epoxides and to satisfy environmental require-
ments, considerable efforts have been made to produce epox-
ides by using air as the oxidizing agent. The use of air for the
oxidation of olefins is still attractive and challenging.
ous catalysts for the selective oxidation of alkenes.[5] However,
homogenous catalysts have some shortcomings, such as diffi-
culty in separation of the product, easy deactivation, and im-
practicable regeneration and recovery of the catalysts. An at-
tempt to heterogenize the homogenous complexes onto solid
supports (e.g., MCM-41, SBA-15, and zeolite-X) could overcome
these problems. Therefore, the epoxidation of alkenes by using
air as the oxidizing agent over heterogeneous transition-metal-
complex catalysts is desirable. However, silicate-based materi-
als have some shortcomings, such as poor hydrothermal stabil-
ity and they are not alkali resistant. These shortcomings can be
overcome by using carbon materials. Therefore, research into
the carbon materials has received much attention.[6]
Various Schiff complexes of various transition metals (e.g.,
Cu, Co, Fe, and V) are known to be conventional homogene-
Graphene, which is a novel one-atom-thick, two-dimension-
al, graphitic carbon nanomaterial, has been received much at-
tention recently owing to its unique electrical, mechanical,
thermal, and optical properties, and high surface area.[7] In the
family of graphene derivatives, graphene oxide (GO) is an ideal
substrate for the immobilization of inorganic nanoparticles for
a wide range of applications because ample oxygen-carrying
functionalities exist on its basal planes and at its edges (i.e.,
hydroxyl, epoxide, and carboxylic groups)[8] and the aromatic
sp2 domains permit it to participate in a wide range of bond-
ing interactions.[9] Furthermore, inorganic nanoparticles can
arrest the aggregation of graphene sheets during oxidation
[a] Dr. Z. Li, S. Wu, X. Wang, X. Yang, Prof. J. Guan, Prof. Q. Kan
College of Chemistry, Jilin University
Changchun 130023 (P. R. China)
E-mail: guanjq@jlu.edu.cn
qkan@mail.jlu.edu.cn
[b] D. Zheng, H. Ding, Prof. Q. Huo
State Key Laboratory of Inorganic Synthesis
and Preparative Chemistry, College of Chemistry
Jilin University, Changchun 130012 (P. R. China)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cplu.201300424.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2014, 79, 716 – 724 716

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
and maintain the large specific surface area of the resultant
hybrid materials.[10]
plexes with chloromethyl groups immobilized onto amino-
modified graphene sheets and carbon-coated Fe3O4 nanoparti-
cle catalysts (mildly reduced graphene oxide (MRGO)/
Fe3O4@C–salen–M, M=Cu, Co, Fe, or VO). Moreover, their cata-
lytic properties in the epoxidation of styrene, with acetonitrile
as the solvent and air as the oxidizing agent, in combination
with a sacrificial coreductant, isobutyraldehyde, were investi-
gated.
Fe3O4 nanoparticles have attracted considerable attention
owing to their magnetic properties, low toxicity, and low
cost.[11] Compared with the separation of product and catalyst
from the reaction system through other chemical or physical
procedures, such as liquid–liquid extraction, chromatography,
distillation, filtration, or centrifugation, magnetic separation
can be regarded as a robust, highly efficient, and fast separa-
tion tool. Silica or carbon coatings on magnetic materials are
promising approaches[12] because they give excellent chemical
and mechanical stability against variation in pH or temperature
of the environment. Moreover, these approaches can graft
functional groups on their surfaces and enhance the specific
surface area. In this study, carbon coatings on magnetic materi-
als is an efficient approach to produce carbon shells, which are
considered to be promising new materials owing to their
higher stability in various chemical and physical environments,
such as acid or base media, as well as at high temperatures
and pressures, compared with silica shells.[13] Kong et al. report-
ed a carbon layer coated onto Fe3O4 microspheres by hydro-
thermal treatment of low-cost glucose.[12b] Furthermore, the
prepared core–shell catalyst not only had unique antacid and
magnetic properties, but also manifested higher catalytic activ-
ity in the reduction of methyl orange than that of the commer-
cially used Pd/C catalyst. Recently, Fan et al. reported that
magnetic graphene–Fe3O4@carbon (GFC) exhibited excellent
reproducibility and recyclability during dye adsorption.[10] For
instance, about 86 and 77% of the dye removal efficiency can
be maintained after five adsorption–desorption cycles in water
and 1m HCl, respectively. The immobilization of porcine pan-
creatic lipase onto graphene decorated with magnetic Fe3O4
nanoparticles was successfully achieved by an easy, one-pot
polyol approach.[7a] Additionally, the immobilized catalysts ex-
hibited excellent efficiency in immobilizing porcine pancreatic
lipase (although dependent on pH, temperature, and denatur-
ant) and had remarkable thermal stability and recyclability in
comparison with free porcine pancreatic lipase. To date, to the
best of our knowledge, there have not been any related re-
ports on the covalent attachment of transition-metal (Cu, Co,
Fe, and V)–salen complexes onto carbon-coated magnetic
Fe3O4 supported on GO sheets and carbon-coated Fe3O4 nano-
particles and no similar materials were applied in the epoxida-
tion of styrene with air as the oxidant.
Results and Discussion
Structural characteristics of catalysts
The powder XRD patterns of the prepared catalysts are shown
in Figure 1. For GO (Figure S1 in the Supporting Information).
Figure 1. XRD patterns of a) MRGO/Fe3O4, b) MRGO/Fe3O4@C, c) MRGO/
Fe3O4@C–salen–Cu, d) Fe3O4, e) Fe3O4@C, and f) Fe3O4@C–salen–Cu.
The peak at approximately 11.28 corresponded to the (001) re-
flection of graphite oxide, which indicated that oxygen species
were successfully inserted into the graphitic layers. On the con-
trary, for a series of MRGO/Fe3O4 composites (Figure 1a–c), the
diffraction peak at 2qꢀ11.28 disappears and a new peak at 2q
ꢀ25.38 is observed owing to the removal of the oxygen-con-
taining groups and recovery of the ordered structure of gra-
phene.[11a] For Fe3O4, the peaks at 2q =18.4, 30.2, 35.6, 43.3,
53.6, 57.1, and 62.88 are ascribed to (111), (220), (311), (400),
(422), (511), and (440) reflections of Fe3O4 (Figure 1d), respec-
tively; these values are consistent with standard XRD data for
cubic-phase Fe3O4 (JCPDS no. 89-4319) with a face-centered
cubic (fcc) structure.[15] For MRGO/Fe3O4 (Figure 1a), the char-
acteristic diffraction peaks of Fe3O4 and graphene are still ob-
served; this indicates that Fe3O4 and graphene coexist in the
hybrid catalyst. Meanwhile, for MRGO/Fe3O4@C (Figure 1b),
a wide peak at 24.28 appears; this suggests that the carbon
shell is amorphous.[10,12b] However, compared with MRGO/
Fe3O4@C, this diffraction peak of the carbon shell of Fe3O4@C is
not clear. A similar pattern is found for MRGO/Fe3O4@C–salen–
Cu and Fe3O4@C–salen–Cu (Figure 1c and f), which suggests
Previously, we found that tethering of copper(II), cobalt(II),
iron(III), and oxovanadium(IV) transition-metal complexes onto
SBA-15 or GO material resulted in catalysts that were highly
active in the epoxidation of styrene.[14] However, these catalysts
sustain a certain level of difficulties in separation and inevitably
result in time and energy consumption. To resolve these prob-
lems, one possible solution is to prepare magnetically separa-
ble nanocatalysts that allow easy separation of the catalyst by
simply applying an external magnetic field. Therefore, hybridi-
zation of graphene sheets and carbon-coated Fe3O4 nanoparti-
cles is an ideal support for immobilizing transition-metal com-
plexes. Herein, we report the design and characterization of
magnetic functionalized CuII, CoII, FeIII, or VOIV Schiff base com-
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2014, 79, 716 – 724 717

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
that the structures of MRGO/Fe3O4@C and Fe3O4@C are not de-
stroyed during immobilization of the complexes. Similar results
are obtained for MRGO/Fe3O4@C–salen–M (M=Co, Fe, V) and
Fe3O4@C–salen–M (M=Co, Fe, V; Figures S2 and S3 in the Sup-
porting Information).
Figure 2 provides typical TEM and high-resolution (HR)-TEM
images of the MRGO/Fe3O4@C–salen–Cu catalyst. The well-sep-
Figure 3. FTIR spectra of a) GO, b) Fe3O4, c) MRGO/Fe3O4, d) MRGO/Fe3O4@C,
and e) MRGO-Fe3O4@C–salen–Cu.
Figure 2. TEM (a) and HR-TEM (b) images of MRGO/Fe3O4@C–salen–Cu.
immobilized onto MRGO/Fe3O4@C. Similar results were ob-
tained for MRGO/Fe3O4@C–salen–M (M=Co, Fe, V) and
Fe3O4@C–salen–M (M=Cu, Co, Fe, V; Figures S4 and S5 in the
Supporting Information).
arated, dark Fe3O4 nanoparticles are evenly covered by a light
layer of amorphous carbonaceous shell (Figure 2a) and the
hybrid particle has a carbonaceous shell with an average thick-
ness of 6 nm (inset in Figure 2a). The HR-TEM image (Fig-
ure 2b) reveals an interplanar spacing of 0.25 nm, which corre-
sponds to the (311) plane of Fe3O4.[16] Meanwhile, the d-spacing
value of 0.45 nm measured corresponds to the spacing of the
(002) planes of graphene (Figure 2b). Furthermore, the BET sur-
face areas of MRGO/Fe3O4@C–salen–Cu and Fe3O4@C–salen–Cu
are 15.6 and 8.0 m2 gÀ1, respectively.
Raman spectroscopy was employed to provide information
about structural changes to the carbon materials during the
chemical reactions. The G band is generally attributed to the
in-plane vibration of sp2 carbon atoms in a 2D hexagonal lat-
tice and the D band is the vibrations of sp3-carbon atoms of
disordered graphite.[16,19] The Raman spectra for all of the cata-
lysts are depicted in Figure 4. The spectra of all of the catalysts
exhibit two strong bands at n˜ ꢀ1351 and 1595 cmÀ1, owing to
the well-defined D and G bands, respectively. The ratio of ID/IG
(the Raman intensity ratio of the D and G bands) is regarded
as a unique characteristic tool to evaluate the ordered and dis-
ordered crystal structures of carbon. For all of the catalysts, the
Spectroscopic characterization
The FTIR spectra of GO, as-prepared Fe3O4, and the magnetic
hybrids are provided in Figure 3. The spectrum of GO exhibits
characteristic bands at n˜ =3387, 1728, 1623, 1224, and
1056 cmÀ1 (Figure 3a), corresponding to the presence of OÀH,
C=O, C=C, CÀOÀC, and CÀO, respectively.[17] Furthermore, for
MRGO/Fe3O4, the characteristic bands at n˜ ꢀ1623 and
1395 cmÀ1 are absent and a new absorption band at n˜
ꢀ1569 cmÀ1 is observed owing to skeletal vibrations of the
graphene sheets (Figure 3c).[18] The spectrum of MRGO/
Fe3O4@C shows characteristic peaks at n˜ =1707 and 1576 cmÀ1,
which are due to C=O and C=C vibrations (Figure 3d), respec-
tively; this indicates that glucose was successfully coated onto
MRGO/Fe3O4 during hydrothermal treatment.[12b] Moreover, the
bands in the range n˜ =1000–1300 cmÀ1 are observed owing to
CÀO stretching and OÀH bending vibrations, which suggests
the existence of a large number of residual hydroxyl groups on
the surface.[12b] However, compared with Fe3O4, the intensity of
the FeÀO bond at n˜ ꢀ585 cmÀ1 decreases in MRGO/Fe3O4
owing to a reduction in the loading of Fe3O4. The spectrum of
MRGO/Fe3O4@C–salen–Cu shows new weak peaks in the range
of n˜ =1600–1100 cmÀ1 owing to CÀO, CÀN and aromatic ring
vibrations; this indicates that the salen ligand is successfully
Figure 4. Raman spectra of a) GO, b) MRGO/Fe3O4@C–salen–Cu, c) MRGO/
Fe3O4@C–salen–Co, d) MRGO/Fe3O4@C–salen–Fe, and e) MRGO/Fe3O4@C–
salen–V.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2014, 79, 716 – 724 718

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
values of ID/IG are shown in Figure S6 in the Supporting Infor-
mation. The values of ID/IG in the catalysts increase in the se-
quence GO <MRGO/Fe3O4@C–salen–Cu<MRGO/Fe3O4@C–
salen–Fe<MRGO/Fe3O4@C–salen–V<MRGO/Fe3O4@C–salen–
Co. The ID/IG value of MRGO/Fe3O4@C–salen–Co is larger than
that in GO (Figure S6 in the Supporting Information), owing to
more numerous but smaller sp2 graphitic domains created
upon reduction of GO and the presence of some unrepaired
defects.[16]
ions and metal–salen complexes were successfully immobilized
onto MRGO/Fe3O4@C.
X-ray photoelectron spectroscopy studies
The X-ray photoelectron spectroscopy (XPS) results for MRGO/
Fe3O4@C–salen–M (M=Cu, Co, Fe, V) are listed in Figure 6 and
give signals mainly associated with Si 2p, C 1s, N 1s, O 1s,
and Fe 2p. For all supported catalysts, the binding energy
peaks at 711.8 and 725.7 eV, which corresponds to Fe 2p3/2
and Fe 2p1/2, respectively. Compared with our early characteri-
zation of GO, the intensities of the characteristic peaks of
oxygen-containing functional groups decrease significantly for
the MRGO/Fe3O4@C–salen–Cu catalyst (Figure 6 f); this indi-
cates that most of the oxygen-containing functional groups
are removed and the conjugated graphene networks are re-
stored after the solvothermal process.[10] Meanwhile, the bind-
ing energies of Cu 2p for the MRGO/Fe3O4@C–salen–Cu cata-
lyst (Figure 6g) are 934.4 and 954.3 eV, which can be assigned
to copper(II) atoms in the supported copper catalyst. These re-
sults give another piece of evidence for the existence of metal
complexes in the hybrid materials. Similar results were ob-
tained for the other tethered catalysts (Figure S7 in the Sup-
porting Information).
Thermogravimetric studies
The thermogravimetric (TG) curves of GO and MRGO/Fe3O4@C–
salen–M (M=Cu, Co, Fe, V) are presented in Figure 5. The TG
Magnetization studies
Magnetic measurements were performed on Fe3O4 and a series
of the tethered catalysts, as shown in Figure 7. The magnetiza-
tion hysteresis loops of all samples are S-like curves, which im-
plies that there is almost no remaining magnetization when
the external magnetic field is removed and indicate that the
prepared catalysts show superparamagnetic behavior.[10,16,21]
The saturation magnetization values for Fe3O4, MRGO/
Figure 5. TG curves of a) GO, b) MRGO/Fe3O4@C–salen–Cu, c) MRGO/
Fe3O4@C–salen–Co, d) MRGO/Fe3O4@C–salen–Fe, and e) MRGO/Fe3O4@C–
salen–V.
Fe3O4@C–salen–Cu,
MRGO/Fe3O4@C–salen–Co,
MRGO/
Fe3O4@C–salen–Fe, and MRGO/Fe3O4@C–salen–V (not shown)
nanocomposites are 89.6, 4.1, 4.0, 3.8, and 4.1 emugÀ1, respec-
tively. The tethered catalysts show a decrease in saturation
magnetization, owing to a decrease in the amount of magnetic
component in the composites. In addition, the tethered cata-
lyst (MRGO/Fe3O4@C–salen–Cu) can be easily separated by
means of an external magnetic force (inset in Figure 7), which
suggests that the nanocomposite catalyst is a promising candi-
date for practical applications in the field of catalytic reactions
with great separability and recyclability.
curve of graphite oxide exhibits two clear weight losses at
temperatures ranging from 25 to 6008C under a nitrogen at-
mosphere (Figure 5a); these are attributed to the loss of physi-
cally adsorbed water and the decomposition of oxygen-carry-
ing functionalities.[20] However, the TG curves of the tethered
catalysts show a three-step weight loss in the range from 25–
6008C (Figure 5b–e). The weight loss below 2508C is attribut-
ed to physically adsorbed water and the decomposition of un-
digested oxygen-carrying functionalities, which do not interact
with (3-aminopropyl)trimethoxysilane (APTMS). Weight loss in
the temperature range from 250 to 4508C is allocated to the
combustion of amine. The weight loss of 9.1–13.1% in the
temperature range from 450 to 6008C is probably attributed
to the decomposition of Schiff ligands. The loadings of Schiff
base ligands are 0.334–0.480 mmolgÀ1 based on the weight
losses for the tethered catalysts and the metal contents are
0.116–0.206 mmolgÀ1, as estimated by inductively coupled
plasma atomic emission spectroscopy (ICP-AES; Table S1 in the
Supporting Information). Consequently, the molar ratios of
ligand/metal are greater than 1.9 for all of the supported cata-
lysts; this suggests that all ligands have coordinated to metal
Catalytic tests
Table 1 presents catalytic data for the epoxidation of styrene
by molecular oxygen under atmospheric pressure with CH3CN
as the solvent and isobutyraldehyde coreductant over MRGO/
Fe3O4@C–salen–M (M=Cu, Co, Fe, or V) catalysts. Notably,
these catalysts are active. The MRGO/Fe3O4@C–salen–Co cata-
lyst shows 91.4% conversion of styrene with a TOF of
197.0 hÀ1 and 52.3% selectivity to the styrene oxide after 8 h
in the first run. In addition, the MRGO/Fe3O4@C–salen–Cu cata-
lyst gives lower activity and epoxide selectivity, corresponding
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2014, 79, 716 – 724 719

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
Figure 6. XPS spectra of a) MRGO/Fe3O4@C–salen–Cu, b) MRGO/Fe3O4@C–salen–Co, c) MRGO/Fe3O4@C–salen–Fe, and d) MRGO/Fe3O4@C–salen–V; e) Fe 2p of
MRGO/Fe3O4@C–salen–Cu; f) C 1s of MRGO/Fe3O4@C–salen–Cu; and g) Cu 2p of MRGO/Fe3O4@C–salen–Cu.
to 84.3% conversion of styrene with a TOF of 102.3 hÀ1 and
38.3% selectivity to the corresponding epoxide. Meanwhile,
the MRGO/Fe3O4@C–salen–Fe and MRGO/Fe3O4@C–salen–V cat-
alysts exhibit 86.2 versus 72.3% conversion of styrene with
a TOF of 158.5 versus 149.4 hÀ1 and 34.5 versus 44.4% selectiv-
ity to the corresponding epoxide, respectively. In addition, it is
also observed that the fresh supported catalysts exhibit some-
what higher activity than the corresponding homogeneous
complex under the same reaction conditions, owing to isola-
tion of the active species, which prevents their dimerization,
and the synergistic beneficial catalytic interaction between the
metal complexes and the supports. A similar result for asym-
metric epoxidation of unfunctionalized olefins was also report-
ed for a new type of immobilized chiral salen MnIII complex
and chiral salen MnIII complex.[22]
lized metal-complex catalysts MRGO/Fe3O4@C–salen–M (M=
Cu, Co, Fe, or V). For example, the Fe3O4@C–salen–Co catalyst
shows a conversion of 75.6% and the Fe3O4@C–salen–Cu cata-
lyst displays a conversion of 58.8%. In this case, graphene
sheets, acting as 2D substrates, can not only effectively pre-
vent the aggregation of Fe3O4 nanoparticles, but also afford
good dispersion of these magnetic nanoparticles.[10] Mean-
while, the Fe3O4 nanoparticles anchored on the graphene
sheets can stop restacking of the graphene sheets to avoid or
reduce the loss of specific surface area; thus enhancing the
catalytic activity.[10]
In addition to the strong influence of support on the catalyt-
ic performance of the metal complexes, the catalytic activity is
highly dependent on the type and loadings of the central
metals. In our experimental results, the conversion of styrene
over supported catalysts decreases in the order Co>Fe>Cu>
V (Table 1), which is opposite to the results reported in our
early work.[14c] However, the metal loading amount decreases
in the order Cu>Fe>V>Co. These results suggest that the re-
activity is not mainly determined by the metal loading
Furthermore, for comparison, the catalytic properties of the
Fe3O4@C–salen–M (M=Cu, Co, or Fe) catalyst was also exam-
ined under identical reaction conditions. The results confirm
that fresh Fe3O4@C–salen–M (M=Cu, Co, Fe, or V) catalysts ex-
hibit worse catalytic activity than the corresponding immobi-
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2014, 79, 716 – 724 720

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
performed by truly heterogeneous copper(II), cobalt(II), iron(III),
and oxovanadium(IV) catalysts.
Conclusion
Magnetic nanocomposites of MRGO/Fe3O4@C–salen–M (M=
Cu, Co, Fe, or V) were synthesized and their applications as cat-
alysts for the aerobic oxidation of styrene were investigated.
Characterization results revealed that metal (Cu, Co, Fe, or V)
complexes were successfully grafted onto amino-functionalized
MRGO/Fe3O4@C and the structure of MRGO/Fe3O4@C was pre-
served. Meanwhile, the magnetic nanocatalysts MRGO/
Fe3O4@C–salen–M (M=Cu, Co, Fe, or V) exhibited higher cata-
lytic activity for the aerobic oxidation of styrene with air as the
oxidant than the analogous homogeneous catalysts or
Fe3O4@C–salen–M (M=Cu, Co, Fe, or V) catalysts under the
same conditions owing to site isolation, good dispersion of the
magnetic Fe3O4 nanoparticles, and efficient synergistic effects
between the metal complexes and the supports. Moreover,
these catalysts were stable and could be recycled many times.
Experimental Section
Materials
Flaky graphite, H2SO4, KMnO4, 30% H2O2, 5% HCl, salicylaldehyde
(99%), ZnCl2, diethyl ether, petroleum ether, MgSO4, NaHCO3, ethyl-
enediamine, styrene (98%), isobutylaldehyde, FeCl3·6H2O,
CH3COONa·3H2O, glucose, APTMS (Aldrich), Fe(NO3)3·9H2O, V2O5,
acetylacetone (acac), Co(NO3)2·6H2O, and Cu(CH3COO)2 were com-
mercially available. All organic solvents were of A.R. grade. Na/di-
phenylketone ketyl was used to dry toluene, then toluene was dis-
tilled under an N2 atmosphere and stored over 4 ꢁ molecular
sieves.
Figure 7. Magnetization curves of a) Fe3O4, b) MRGO/Fe3O4@C–salen–Cu,
c) MRGO/Fe3O4@C–salen–Co, and d) MRGO/Fe3O4@C–salen–Fe.
Synthetic procedures (Scheme 1)
amount, but by the different redox nature of transition metals
and efficient synergistic effects between the metal complexes
and supports.
A schematic illustration of the covalent attachment of copper(II),
cobalt(II), iron(III), or oxovanadium(IV) salen complexes to MRGO/
Fe3O4@C is depicted in Scheme 1 and described below.
Moreover, the recoverability and stability of tethered cop-
per(II), cobalt(II), iron(III), and oxovanadium(IV) catalysts were
also investigated in the epoxidation of styrene under similar re-
action conditions, and the results are listed in Table 1. It is
clear that the supported catalysts could be successfully recy-
cled three times without a significant decrease in activity or se-
lectivity under the test conditions. In addition, it is interesting
to note that the conversion of styrene and selectivity to epox-
ide increase for MRGO/Fe3O4@C–salen–Co in the second run;
this might be owing to the better synergistic effect between
the metal complexes and the supports after the first run. The
catalysts were recovered by magnetic separation after the re-
cycled reaction. The solution and recovered catalysts were ana-
lyzed by inductively coupled plasma analysis techniques. The
results suggest that there are almost no metal ions in the filter
liquor. Nevertheless, the metal contents for the used catalysts
just decreased slightly (after 3 cycles) compared with the fresh
ones (Table 1). Consequently, the majority of the catalysis is
Synthesis of GO
GO was prepared by using the harsh oxidation of graphite accord-
ing to a modified Hummers method.[23] In brief, graphite powder
(1.0 g) and concentrated H2SO4 (23 mL) were added to a 250 mL
three-necked flask. The mixture was stirred in an ice batch for
120 min, and then KMnO4 (3.0 g) was added to the mixture gradu-
ally under vigorous stirring, and the reaction mixture was stirred at
room temperature for 30 min. The mixture was then placed in
a water bath at 358C and stirred for 2 h. Deionized water (46 mL)
was then added under vigorous stirring. The mixture was succes-
sively heated to 958C for 15 min. Finally, the mixture was trans-
ferred to a 1000 mL beaker and deionized water (140 mL) was
added, followed by the addition of 30% H2O2 solution (3 mL). The
resulting product was obtained by filtration, washing with a large
amount of 5% HCl solution and deionized water, and finally
drying.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2014, 79, 716 – 724 721

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
Table 1. Catalytic data of various catalysts in the aerobic epoxidation of styrene.
[b]
Catalysts
Cycle
Metal content [mmolgÀ1
]
Styrene conversion [%][a]
TOF [hÀ1
]
Product selectivity [mol%][c]
Reference
So
Bza
Others
MRGO/Fe3O4@C–salen–Co
MRGO/Fe3O4@C–salen–Co
MRGO/Fe3O4@C–salen–Co
MRGO/Fe3O4@C–salen–Cu
MRGO/Fe3O4@C–salen–Cu
MRGO/Fe3O4@C–salen–Cu
MRGO/Fe3O4@C–salen–Fe
MRGO/Fe3O4@C–salen–Fe
MRGO/Fe3O4@C–salen–Fe
MRGO/Fe3O4@C–salen–V
MRGO/Fe3O4@C–salen–V
MRGO/Fe3O4@C–salen–V
Fe3O4@C–salen–Co
Fe3O4@C–salen–Cu
Fe3O4@C–salen–Fe
Co–salen
Cu–salen
Fe–salen
VO–CM–salen
1
2
3
1
2
3
1
2
3
1
2
3
1
1
1
1
1
1
1
0.116
–
0.115
0.206
–
0.205
0.136
–
0.119
0.121
–
0.112
0.106
0.195
0.114
3.075
3.032
3.104
–
91.4
92.3
91.3
84.3
83.0
83.1
86.2
75.0
73.1
72.3
71.3
60.2
75.6
58.8
75.4
85.9
84.2
83.1
69.0
197.0
–
198.5
102.3
–
101.3
158.5
–
153.6
149.4
–
134.4
178.3
75.4
165.4
69
52.3
59.2
58.0
38.3
40.1
43.8
34.5
41.9
35.0
44.4
28.4
30.5
42.9
12.5
40.1
54.4
62.4
51.2
58.2
47.7
40.8
42.0
61.7
59.9
56.2
65.5
58.1
65.0
55.6
71.6
69.5
57.1
87.5
59.9
36.2
33.1
44.1
41.8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
this study
this study
this study
this study
this study
this study
this study
this study
this study
this study
this study
this study
this study
this study
this study
[14c]
9.4
4.5
4.7
–
70
67
63.4
[14c]
[14c]
[14a]
[a] Reaction conditions: catalyst (50 mg (5 mg for the untethered catalyst)), styrene (1.14 mL, 10 mmol), CH3CN (10 mL), flow of air=80 mLminÀ1, isobutyr-
aldehyde (2.28 mL, 25 mmol), temperature=808C, time=8 h. [b] TOF=turnover frequency [hÀ1]: moles of substrate converted per mole of metal ion per
hour. [c] So=styrene oxide, Bza=benzaldehyde; Others: including benzoic acid, phenylacetaldehyde, and 1-phenylethane-1,2-diol.
solid was isolated by filtration. 5-Chloromethyl-2-hydroxybenzalde-
hyde was recovered by crystallization.
Synthesis of MRGO/Fe3O4 composite
Graphene–Fe3O4 was synthesized based on the following method:
GO was dispersed in a mixture of diethylene glycol (DEG; 20 mL)
and ethylene glycol (EG; 20 mL) in an ultrasonic bath for 2 h to
produce solution A for later use. At the same time, FeCl3·6H2O and
CH3COONa·3H2O were dissolved in a mixture of DEG (48 mL) and
EG (48 mL) to produce solution B. Solution B was added to solutio-
n A, and then the mixture was sonicated for 30 min and transferred
into an autoclave, sealed, and heated at 2208C for 10 h. The reac-
tion solution was cooled to room temperature and the solid prod-
uct was separated by magnetic separation. The solid product was
washed with deionized water and ethanol, and dried in a vacuum
oven at 408C overnight. As a control experiment, Fe3O4 was pre-
pared under the same conditions without the addition of GO.
N,N’-Bis(5-chloromethylsalicylidene)ethylenediamine was also pre-
pared and characterized as described previously.[14a] Briefly, a solu-
tion of 5-chloromethyl-2-hydroxybenzaldehyde (10 mmol, 10 mL)
was added to a solution of ethylenediamine (5 mmol, 10 mL) in
CH2Cl2 and then the resulting mixture was stirred at room temper-
ature under a N2 atmosphere for 24 h. The bright-yellow solid was
obtained by rotary evaporation. The obtained sample was denoted
as CM–salenH2.
Synthesis of MRGO/Fe3O4@C–salen–M (M=CuII, CoII, FeIII, or
VOIV)
Amino-functionalized MRGO/Fe3O4@C was synthesized according
to the following method. The general procedure is as follows:
MRGO/Fe3O4@C (1.0 g), APTMS (2.3 mmol), and toluene (20 mL)
were added to a 50 mL two-necked flask and then the mixture was
heated at reflux at 1108C under an N2 atmosphere for 24 h. The re-
sulting solid was isolated by filtration, washed with toluene, and
dried. The obtained solid was denoted as MRGO/Fe3O4@C–NH2.
Synthesis of MRGO/Fe3O4@C
Typically, MRGO/Fe3O4 (2.5 mgmLÀ1) was added to a solution of
glucose (80 mL, 0.25m), which was transferred into an autoclave,
sealed, and heated at 1808C for 6 h. The resulting product was iso-
lated by magnetic separation, washed with deionized water and
ethanol, and dried in a vacuum oven at 408C overnight. Fe3O4@C
was prepared under the same conditions.
The immobilization of CM–salenH2 on MRGO/Fe3O4@C was synthe-
sized by a nucleophilic reaction between CM–salenH2 and MRGO/
Fe3O4@C–NH2. Briefly, CM–salenH2 (1.15 mmol) was added to a sus-
pension of above-prepared MRGO/Fe3O4@C–NH2 in toluene and
then the mixture was heated at reflux under an N2 atmosphere for
24 h. The resulting solid was isolated by filtration, washed with tol-
uene, Soxhlet extraction with CH2Cl2, and dried. The obtained solid
was denoted as MRGO/Fe3O4@C–salen.
Synthesis of N,N’-bis(5-chloromethylsalicylidene)ethylenedi-
amine (CM–salenH2)
5-Chloromethyl-2-hydroxybenzaldehyde was also prepared and
characterized as described previously.[14a] In brief, a mixture of sali-
cylaldehyde (80 mmol), 38% aqueous formaldehyde (12 mL), and
ZnCl2 (0.6 g) were added to a 100 mL three-necked flask, and then
concentrated HCl (50 mL) was added and the mixture was churned
at room temperature under an N2 atmosphere for 24 h. The white
CuII, CoII, FeIII, and VOIV salen supported complexes MRGO/
Fe3O4@C–salen–M (M=Cu, Co, Fe, or VO) were synthesized as fol-
lows.
A certain amount of Cu(CH3COO)2, Co(NO3)2·6H2O, Fe-
(NO3)3·9H2O, or [VO(acac)2] was added to a solution of MRGO/
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2014, 79, 716 – 724 722

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
ter. XPS was performed on an ES-
CALAB 250 X-ray electron spec-
trometer by using AlKa radiation.
The magnetic hysteresis curves
were acquired on
a vibrating
sample magnetometer. Morpho-
logical features of the materials
were characterized by a JEOL JEM-
2010 transmission electron micro-
scope at an accelerating voltage of
200 kV. Raman spectra were re-
corded on
a Renishaw Raman
system model 1000 spectrometer
with an excitation wavelength of
514 nm. TG analyses were per-
formed on a TGA Q500 instrument
working at
a heating rate of
108CminÀ1 from 25 to 8008C
under flowing N2.
Catalytic tests
The catalytic epoxidation of sty-
rene was performed in a 20 mL
flask equipped with a reflux con-
denser.
Typically,
styrene
(10 mmol), solvent (MeCN, 10 mL),
isobutylaldehyde (25 mmol), and
catalyst (50 mg) were mixed. The
mixture was heated to 808C and
then oxidant (air) was introduced
at
a
flow rate of about
80 mLminÀ1. The reaction products
were analyzed by GC with a Shi-
madzu GC-8A (Japan) instrument
equipped by using an HJ-5 capilla-
ry column and flame ionization de-
tector. At the end of the reaction,
the catalyst was obtained by mag-
netic separation, washing with co-
pious amounts of MeCN, and
drying. The catalyst was then
reused in the next reaction with-
out further purification.
Scheme 1. Schematic outlines of the synthesis of a) CM–salen and b) the covalent attachment of CuII, CoII, FeIII, or
VOIV salen complexes onto MRGO/Fe3O4@C.
Fe3O4@C–salen in methanol and then the mixture was stirred at
room temperature for 24 h. The resulting solid was then obtained
by filtration, washing with large quantities of methanol and water,
and extraction with CH2Cl2. As a control experiment, Fe3O4@C–
salen–M (M=Cu, Co, Fe, or V) was prepared under the same condi-
tions.
Acknowledgements
This study was supported by the National Natural Science Foun-
dation of China (21303069) and Jilin province (201105006).
Keywords: epoxidation
· graphene · iron · supported
catalysts · transition metals
Characterization
The IR spectra were recorded with a NICOLET impact 410 spec-
trometer as KBr pellets. Powder XRD measurements were per-
formed on a Shimadzu XRD-6000 diffractometer by using a Ni-fil-
tered CuKa radiation scanning from 5 to 708 (2q). N2 adsorption–de-
sorption isotherms were measured with a Micromeritics ASAP 2020
system at liquid N2 temperature. The BET surface area was calculat-
ed by using the BET method. The metal content was estimated by
ICP-AES analysis conducted on a PerkinElmer emission spectrome-
[1] Y. Zhang, J. Zhao, L. He, D. Zhao, S. Zhang, Microporous Mesoporous
Mater. 2006, 94, 159–165.
[2] S. Bhattacharjee, J. A. Anderson, J. Mol. Catal. A: Chem. 2006, 249, 103–
110.
[3] B. Tyagi, B. Shaik, H. C. Bajaj, Appl. Catal. A 2010, 383, 161–168.
[4] a) J. Jiang, K. Ma, Y. Zheng, S. Cai, R. Li, J. Ma, Appl. Clay Sci. 2009, 45,
117–122; b) Y. Liu, H. Tsunoyama, T. Akita, T. Tsukuda, Chem. Commun.
2010, 46, 550–552.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2014, 79, 716 – 724 723

CHEMPLUSCHEM
FULL PAPERS
www.chempluschem.org
[5] a) Y. H. Lin, I. D. Williams, P. Li, Appl. Catal. A 1997, 150, 221–229; b) N.
Anand, K. H. P. Reddy, V. Swapna, K. S. R. Rao, D. R. Burri, Microporous
Mesoporous Mater. 2011, 143, 132–140; c) S. Rayati, M. Koliaei, F.
Ashouri, S. Mohebbi, A. Wojtczak, A. Kozakiewicz, Appl. Catal. A 2008,
346, 65–71; d) S. K. Ginotra, V. K. Singh, Tetrahedron 2006, 62, 3573–
3581.
[6] S. Dorbes, C. Pereira, M. Andrade, D. Barros, A. Pereira, S. Rebelo, J.
Araffljo, J. Pires, A. Carvalho, C. Freire, Microporous Mesoporous Mater.
2012, 160, 67–74.
[7] a) X. Li, H. Zhu, J. Feng, J. Zhang, X. Deng, B. Zhou, H. Zhang, D. Xue, F.
Li, N. J. Mellors, Y. Li, Y. Peng, Carbon 2013, 60, 488–497; b) Y. Zhu, S.
Murali, W. Cai, X. Li, J. W. Suk, J. R. Potts, R. S. Ruoff, Adv. Mater. 2010,
22, 3906–3924; c) Y. Li, W. Gao, L. Ci, C. Wang, P. M. Ajayan, Carbon
2010, 48, 1124–1130.
[14] a) Y. Yang, Y. Zhang, S. Hao, J. Guan, H. Ding, F. Shang, P. Qiu, Q. Kan,
Appl. Catal. A 2010, 381, 274–281; b) Z. Li, S. Wu, H. Ding, D. Zheng, J.
Hu, X. Wang, Q. Huo, J. Guan, Q. Kan, New J. Chem. 2013, 37, 1561–
1568; c) Y. Yang, Y. Zhang, S. Hao, Q. Kan, Chem. Eng. J. 2011, 171,
1356–1366.
[15] a) L. Ai, C. Zhang, Z. Chen, J. Hazard. Mater. 2011, 192, 1515–1524; b) J.
Hu, Y. Wang, M. Han, Y. Zhou, X. Jiang, P. Sun, Catal. Sci. Technol. 2012,
2, 2332–2340; c) Y. Jang, J. Chung, S. Kim, S. W. Jun, B. H. Kim, D. W.
Lee, B. M. Kim, T. Hyeon, Phys. Chem. Chem. Phys. 2011, 13, 2512–2516.
[16] H. He, C. Gao, ACS Appl. Mater. Interfaces 2010, 2, 3201–3210.
[17] a) Y. Lin, J. Jin, M. Song, J. Mater. Chem. 2011, 21, 3455–3461; b) P.
Zhang, X. Zhang, S. Zhang, X. Lu, Q. Li, Z. Su, G. Wei, J. Mater. Chem. B
2013, 1, 6525–6531; c) S. Choudhary, H. P. Mungse, O. P. Khatri, J. Mater.
Chem. 2012, 22, 21032–21039.
[8] a) Y. Xue, H. Chen, D. Yu, S. Wang, M. Yardeni, Q. Dai, M. Guo, Y. Liu, F.
Lu, J. Qu, L. Dai, Chem. Commun. 2011, 47, 11689–11691; b) S. Lin, C.
Shen, D. Lu, C. Wang, H.-J. Gao, Carbon 2013, 53, 112–119; c) K. Haub-
ner, J. Murawski, P. Olk, L. M. Eng, C. Ziegler, B. Adolphi, E. Jaehne,
ChemPhysChem 2010, 11, 2131–2139; d) X. Yang, C. Chen, J. Li, G. Zhao,
X. Ren, X. Wang, RSC Adv. 2012, 2, 8821–8826.
[9] X.-L. Wu, L. Wang, C.-L. Chen, A.-W. Xu, X.-K. Wang, J. Mater. Chem.
2011, 21, 17353–17359.
[10] W. Fan, W. Gao, C. Zhang, W. W. Tjiu, J. Pan, T. Liu, J. Mater. Chem. 2012,
22, 25108–25115.
[18] C. Hou, Q. Zhang, M. Zhu, Y. Li, H. Wang, Carbon 2011, 49, 47–53.
[19] W. Zhang, X. Shi, Y. Zhang, W. Gu, B. Li, Y. Xian, J. Mater. Chem. A 2013,
1, 1745–1753.
[20] H. P. Mungse, S. Verma, N. Kumar, B. Sain, O. P. Khatri, J. Mater. Chem.
2012, 22, 5427–5433.
[21] F. He, J. Fan, D. Ma, L. Zhang, C. Leung, H. L. Chan, Carbon 2010, 48,
3139–3144.
[22] B. Gong, X. Fu, J. Chen, Y. Li, X. Zou, X. Tu, P. Ding, L. Ma, J. Catal. 2009,
262, 9–17.
[23] K. Krishnamoorthy, M. Veerapandian, K. Yun, S. J. Kim, Carbon 2013, 53,
38–49.
[11] a) S. Wu, Q. He, C. Zhou, X. Qi, X. Huang, Z. Yin, Y. Yang, H. Zhang, Nano-
scale 2012, 4, 2478–2483; b) J. S. Chen, Y. Zhang, X. W. Lou, ACS Appl.
Mater. Interfaces 2011, 3, 3276–3279.
[12] a) M. Masteri-Farahani, N. Tayyebi, J. Mol. Catal. A: Chem. 2011, 348, 83–
87; b) L. Kong, X. Lu, X. Bian, W. Zhang, C. Wang, ACS Appl. Mater. Inter-
faces 2011, 3, 35–42.
[13] Y. Ye, J. Chen, Q. Ding, D. Lin, R. Dong, L. Yang, J. Liu, Nanoscale 2013,
5, 5887.
Received: December 16, 2013
Revised: February 20, 2014
Published online on March 18, 2014
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemPlusChem 2014, 79, 716 – 724 724