Hierarchical nanospheres based on Pd nanoparticles dispersed on carbon coated magnetite cores with a mesoporous ceria shell: A highly integrated multifunctional catalyst

Article abstract of DOI:10.1039/c5dt01805f

The design and fabrication of core-shell nanostructures with steerable morphologies and tailored performances have aroused abundant scientific studies for organic transformations. We here report the preparation of multifunctional and highly efficient core-shell microspheres, which bear a carbon-protected magnetic Fe₃O₄ core, a transition layer of active Pd nanoparticles (NPs) and an outer shell of mesoporous CeO₂ (mCeO₂). The composition and structure of the as-prepared Fe₃O₄@C-Pd@mCeO₂ were thoroughly characterized by X-ray photoelectron spectroscopy, transmission electron microscopy, scanning electron microscopy, X-ray diffraction, Fourier-transform infrared spectroscopy and Brunauer-Emmett-Teller measurements. The well-designed microspheres have high dispersibility, convenient magnetic separability and good reusability as heterogeneous nanoreactors due to their unique structure. We illustrate the high efficiency of these nanocomposites in mediating the Suzuki-Miyaura cross-coupling reaction and the reduction reaction of 4-nitrophenol (4-NP). The enhanced catalytic activity can be attributed to the synergistic effect between the CeO₂ nanoparticles and noble metal NPs. A mechanism was further proposed to explain the improved catalytic activity. This peculiar core-shell nanostructure renders the nanospheres to be an approachable and attractive catalyst system for various catalytic organic industrial processes.

Full text of DOI:10.1039/c5dt01805f

Dalton
Transactions
View Article Online
View Journal
PAPER
Hierarchical nanospheres based on Pd
nanoparticles dispersed on carbon coated
magnetite cores with a mesoporous ceria shell:
a highly integrated multifunctional catalyst†
Cite this: DOI: 10.1039/c5dt01805f
Yinle Li, Zhuqing Zhang, Jianfeng Shen and Mingxin Ye*
The design and fabrication of core–shell nanostructures with steerable morphologies and tailored per-
formances have aroused abundant scientific studies for organic transformations. We here report the
preparation of multifunctional and highly efficient core–shell microspheres, which bear a carbon-protected
magnetic Fe3O4 core, a transition layer of active Pd nanoparticles (NPs) and an outer shell of mesoporous
CeO2 (mCeO2). The composition and structure of the as-prepared Fe3O4@C–Pd@mCeO2 were thoroughly
characterized by X-ray photoelectron spectroscopy, transmission electron microscopy, scanning electron
microscopy, X-ray diffraction, Fourier-transform infrared spectroscopy and Brunauer–Emmett–Teller
measurements. The well-designed microspheres have high dispersibility, convenient magnetic separability
and good reusability as heterogeneous nanoreactors due to their unique structure. We illustrate the high
efficiency of these nanocomposites in mediating the Suzuki–Miyaura cross-coupling reaction and the
reduction reaction of 4-nitrophenol (4-NP). The enhanced catalytic activity can be attributed to the syner-
gistic effect between the CeO2 nanoparticles and noble metal NPs. A mechanism was further proposed to
explain the improved catalytic activity. This peculiar core–shell nanostructure renders the nanospheres to
be an approachable and attractive catalyst system for various catalytic organic industrial processes.
Received 14th May 2015,
Accepted 18th August 2015
DOI: 10.1039/c5dt01805f
www.rsc.org/dalton
Currently, the introduction of magnetic NPs in various
solid matrices permits the assembly of well-known programs
Introduction
Owing to the extraordinary advantages of their unique optical, for catalyst heterogenization with magnetic separation.17,18
electrochemical and catalytic properties, noble metal NPs have Magnetite is an ideal support that is easy to prepare and it has
become promising materials for underlying applications in a proper active surface for the immobilization of metals and
various fields.1–5 In particular, with the property of the high ligands, which can be isolated by magnetic decantation after
surface-to-volume ratio and the superiority of incorporating the completion of a reaction, thus making it a better sustain-
various homogeneous and heterogeneous catalysis,6 the able catalyst.19–21 In the last few years, the magnetic core–shell
employment of noble metal nanoparticles (such as Pd,7 Au,8 nanostructure,22,23 a special type of universal functional com-
Pt,9 and Ag10) in catalytic reactions has aroused enormous scien- posite with a distinct microstructure, has shown a great appli-
tific research. However, naked noble metal NPs tend to aggregate cation for loading noble metal NPs.24–29 Generally, silica or
and sinter, leading to the loss of catalytic activity under reaction carbon as a protecting shell is employed to coat the magnetic
conditions. Therefore, solving this predicament and promoting particles to obtain the shape of core–shell nanostructured
environmental consciousness are imperative parts of these (Fe3O4@SiO2 or Fe3O4@C).30–33 Moreover, a silica or carbon
studies,11–14 and also enhancing efficiency of the catalysts in shell can prevent the magnetic particles from losing their mag-
chemical reactions under a friendly condition with recyclable netic properties in acidic environments and provide numerous
reuse is equally important. Loading noble metal NPs onto the functionalized groups for further modification. Several noble
special surface of supports is one of the best approaches to metal NPs dispersed on the surface of Fe3O4@SiO2 or
boost the efficiency and recyclability of the catalysts.15,16
Fe3O4@C core–shell nanostructures have been reported in an
attempt to achieve easy separation and recyclable catalytic pro-
cesses including methanol oxidation,34 Fenton-like catalysis,35
hydrogenation36 and cross-coupling reactions.31 However, the
active metal NPs on the surface of core–shell nanostructure
directly exposed in reaction condition can be easily leached.
Center of Special Materials and Technology, Fudan University, Shanghai 200433,
China. E-mail: meye@fudan.edu.cn
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
c5dt01805f
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
The novel mesoporous “shell-in-shell” hierarchical structures to a Teflon-lined stainless-steel autoclave (90 mL volume), and
are composed of a mesoporous outermost shell and protected then sealed to heat at 180 °C for 4 h. The products were col-
magnetic inside core–shell structure with the ultrafine active lected from the solution through an external magnet and
metal NPs distributed on the internal and external surface that washed three times with deionized water and ethanol, and
endows the nanocatalyst with high stability, reducing the then vacuum dried at 60 °C overnight.
leaching of active metal NPs, and an independent cavity is
Synthesis of core–shell Fe3O4@C–Pd nanoparticles
formed by two shells.36,37 Zhao et al.38 synthesized multi-
component and multifunctional Fe3O4@C–Pd@mSiO2 hier- Fe3O4@C–Pd nanoparticles were synthesized by the depo-
archical “shell-in-shell” structures for the Suzuki–Miyaura sition–precipitation method. Typically, 100 mg Fe3O4@C nano-
coupling reaction. According to our investigation, there are particles were dispersed in 120 mL 1 : 1 water/ethanol with
only a few reports about the assembling of magnetic core–shell ultrasonication for 30 min. Then, 5 mL 0.01 M PdCl2 ethanol
nanocatalyst with active hierarchical structures, because the dispersion was added into the abovementioned solution. After
precise control of the morphology, structure and assembly ultrasound treatment for another 30 min, the obtained
process of each section is always troublesome.
mixture was then mechanically stirred at 100 °C for 50 min,
Herein, we report novel, well-defined hierarchical “shell-in- and then allowed to cool to room temperature. The products
shell” structures of Fe3O4@C–Pd@mCeO2 consisting of a core were separated by applying an external magnet and washed
of carbon-protected magnetite particles, a transition layer of three times with deionized water and ethanol, and then
confined catalytic Pd nanoparticles, and an outer shell of vacuum dried at 60 °C overnight.
mesoporous CeO2. The outermost CeO2 shell can not only
prevent active Pd NPs from aggregating, sintering and leach-
Synthesis of core–shell Fe3O4@C–Pd@mCeO2 nanoparticles
ing, but also improve the catalytic activity due to a strong 100 mg Fe3O4@C–Pd nanoparticles were dispersed and soni-
synergistic effect between the CeO2 nanoparticles and noble cated for 30 min in 60 mL ethanol. 200 mg Ce(NO3)3·6H2O was
metal NPs.39 The well-designed Fe3O4@C–Pd@mCeO2 hier- dissolved in 60 mL ethanol, and mixed with the above men-
archical core–shell structures possess the advantages of large tioned solution under an ultrasound treatment process for
magnetization, being highly open mesoporous and they dis- 10 min. Subsequently, 0.8 g HMT dissolved in 80 mL pure
perse well in water. The multifunctional catalytic activities water was added to the mixture solution with ultrasonication
were systematically tested for catalyzing the Suzuki–Miyaura for another 15 min. The mixture was then mechanically stirred
cross-coupling reaction and the reduction reaction of 4-NP. for 2 h at 70 °C. The resultant products were separated with a
The studies reveal that the CeO2 nanoparticles and noble magnet, and washed with water five times to remove any
metal NPs exhibit excellent synergetic catalytic performance potential ionic remnants. Finally, the products were vacuum
and the good reusability of the system by the magnetic separ- dried at 60 °C overnight. The powder was calcined at 400 °C
ation for both the reactions. Our results, therefore afford a for
2 h to obtain Fe3O4@C–Pd@mCeO2 nanoparticles.
general way based on a hierarchical core–shell structure for The product was decomposed in concentrated nitric acid, and
the preparation of high performance magnetic catalysts loaded then analysed via an inductively coupled plasma atomic
with noble metal NPs, which will be a very useful tool for emission spectrometer (ICP-AES) to determine the content of
various catalytic organic industry processes.
Pd (3.05 wt%).
The synthesis process of Fe3O4@C@mCeO2 nanoparticles
was the same as that of Fe3O4@C–Pd@mCeO2 nanoparticles
without the Pd nanoparticles.
Experimental
Materials
Suzuki–Miyaura cross-coupling reactions
Ferric chloride hexahydrate (FeCl3·6H2O), trisodium citrate,
palladium(II) chloride (PdCl2), sodium acetate (NaAc), ethylene
glycol (EG), glucose, cerium(III) nitrate hexahydrate (Ce(NO3)3·
6H2O), hexamethylene tetramine (HMT), 4-nitrophenol (4-NP),
arylboronic acid, and aryl halide were bought from Sinopharm
Chemical Reagent (Shanghai, China). All the reagents were of
analytical grade and directly used without further purification.
The Fe3O4@C–Pd@mCeO2 catalyst (10 mg), aryl halide
(1.0 mmol), arylboronic acid (1.2 mmol) and potassium carb-
onate (2 mmol) were placed in a Schlenk tube containing a
magnetic stirrer (in the case of stirring, the catalyst has good
solubility in the solvent of our choice and solvent (5 mL) was
then added. The mixture was stirred at 80 °C in air. The reac-
tion process was monitored by GC at fixed time intervals. The
catalyst was separated from the mixture with a magnet,
washed several times with water and ethanol, and then dried
Synthesis of core–shell Fe3O4@C nanoparticles
The magnetic Fe3O4 nanoparticles modified by trisodium in vacuum at 60 °C overnight. The recycling experiment was
citrate were prepared according to the study reported by carried out following the same procedures.
Zhao.38 Fe3O4@C nanoparticles were synthesized by a versatile
Reduction of 4-NP
hydrothermal method. Briefly, 1 g of as-prepared Fe3O4 nano-
particles and 3.24 g of glucose were dispersed in 60 mL water. The reduction of 4-NP was tested in a quartz cuvette and moni-
After ultrasonication for 30 min, the mixture was transferred tored by UV-vis spectroscopy (Shimadzu UV-3600) at room
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
temperature. In a typical procedure, 4.0 mg of Fe3O4@C–
Pd@mCeO2 composites were homogeneously dispersed into
the 4.0 mL 4-NP solution (10 mg L−1), followed by a quick
injection of 1 mL of fresh NaBH4 solution (10 mg mL−1) under
stirring. The color of the mixture solution gradually changed
from yellow to colorless, indicating that the Fe3O4@C–
Pd@mCeO2 composites catalyzed the reduction of 4-NP. After
stirring every ten seconds, the mixture was rapidly moved to a
quartz cell to monitor the reduction progress by recording the
UV-vis absorption spectra of the solution. The 4-NP has a
strong absorption peak at 400 nm in the presence of NaBH4,
while the product of 4-aminophenol has a moderate absorp-
tion peak at 295 nm.
Scheme 1 Schematic of the preparation of Fe3O4@C–Pd@mCeO2
composites.
In the recycling study, the Fe3O4@C–Pd@mCeO2 compo-
sites were separated from the mixture using a magnet after the
reduction reaction was complete. After washing with ethanol
and water three times, they were utilized in the next reaction run
in a similar manner to the abovementioned reduction process.
grounded on a high temperature reduction of FeCl3·6H2O in
the presence of EG and NaAc.38 Then, a uniform carbon
coating was fabricated on the surface of Fe3O4 nanoparticles
through hydrothermal treatment with glucose as the carbon
source, resulting in core–shell Fe3O4@C nanospheres. Sub-
sequently, Pd NPs were deposited on the surface of the carbon-
protected Fe3O4 nanospheres by refluxing in an ethanol/water
solution using PdCl2 as the precursor without any additional
reducing agents.22 The prepared Pd-immobilized Fe3O4@C
nanospheres were denoted as Fe3O4@C–Pd. Finally,
Ce(NO3)3·6H2O was used as a cerium source and HMT was
applied to slowly produce OH− for obtaining a homogenous
CeO2 coating surrounding the Fe3O4@C–Pd nanospheres
through an assembly sol–gel process. Subsequently, HMT was
removed by a calcination process and a porous CeO2 shell
could be obtained,40 resulting in Fe3O4@C–Pd@mCeO2 nano-
spheres with a sandwich-like structure.
The morphology and structure of the as-prepared compo-
sites were investigated by SEM and TEM. Fig. 1a and b show
the SEM images of the Fe3O4@C–Pd@mCeO2 nanospheres. It
can be seen from the images that the Fe3O4@C–Pd@mCeO2
nanospheres have sizes in the range of 250–300 nm with a
spherical shape. Fig. S1† shows the EDX spectrum of
Fe3O4@C–Pd@mCeO2 nanospheres. The loading amount of Pd
in the Fe3O4@C–Pd@mCeO2 catalyst is estimated to be
0.35 atom% from the EDX analysis. Fig. 2a and b show the
TEM images of the Fe3O4@C after the hydrothermal reaction
of Fe3O4 NPs; a thin carbon layer of ∼8 nm thickness was
formed on the Fe3O4 nanospheres, resulting in core–shell
Fe3O4@C nanospheres. After the deposition of Pd NPs, numer-
Characterization
FT-IR was recorded over the wavenumber range from 4000 to
400 cm−1 on a Nicolet IS10 spectrometer, and solid samples
were measured with KBr disks. XRD analyses were recorded on
a D/max-γB diffractometer using Cu Kα radiation. Trans-
mission electron microscopy (TEM) measurements were
obtained on a JEOL 2010F microscope operating at 200 kV.
The samples were dispersed in ethanol and transferred onto
holey carbon films supported with a Cu grid for TEM measure-
ments. High resolution transmission electron microscopy
(HR-TEM) and scanning transmission electron microscopy
(STEM) images were obtained on a JEM-2100F scanning trans-
mission electron microscope to characterize the morphologies
of the particles. Scanning electron microscopy (SEM) images
were obtained from a Philips XL30FEG. The samples were dis-
persed in ethanol and transferred onto a silicon chip for SEM
measurements. N2 sorption isotherms were obtained with a
Micromerits Tristar 3000 analyzer at 77 K. The powder was
degassed at 200 °C in a vacuum for 10 h. Using the Barrett–
Joyner–Halenda (BJH) model, the pore volumes and the pore
size distributions were acquired from the adsorption branches
of the isotherms. The Brunauer–Emmett–Teller (BET) method
was used to count the specific surface areas using adsorption
date. The magnetization curve of the sample was recorded on
a
superconducting quantum interference device (SQUID)
magnetometer (Quantum Design MPMS XL-7) at 300 K. X-ray
photoelectron spectroscopy (XPS) was collected on XR 5 VG
(UK) using a monochromatic Mg X-ray source. Binding energy
standardization was based on C 1s at 284.6 eV.
Results and discussion
The procedure for the preparation of hierarchical core–shell
nanospheres, designated as Fe3O4@C–Pd@mCeO2, is shown in
Scheme 1. First, the Fe3O4 nanoparticles modified by tri-
sodium citrate were prepared via a rough solvothermal reaction Fig. 1 SEM images of Fe3O4@C–Pd@mCeO2 (a) 1 μm; (b) 200 nm.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
cannot be visualized evidently from the TEM image of the
Fe3O4@C–Pd@mCeO2 nanospheres.41 In addition, the selected
areas of the red circular region of panels, shown in Fig. 2g,
provides EDX results showing that the main element present
in the outermost shell of the nanospheres is cerium. There-
fore, the HR-TEM images shown in Fig. 3a and b further
demonstrate that the relatively crude surface is composed of
ultrafine Pd and CeO2 nanocrystallites with main sizes of
approximately 4.5 nm and 5 nm. The lattice fringes are
obviously visible with d-spacings of about 0.32 nm and
0.21 nm, which readily index to the lattice spacings of (111)
planes of the cubic uorite phase of CeO2 and (111) lattice
planes of Pd NPs.42 in addition, scanning transmission elec-
tron microscopy (STEM) and EDS elemental mapping were
also implemented to uncover the elemental distribution in
these structured microspheres. The STEM images in Fig. 3c
reveal that the structures are core–shell structured micro-
spheres. The corresponding EDS mapping images results indi-
cate that the Fe, Ce, O and Pd elements are dispersed evenly
throughout the whole spheres. The elements Pd and Ce are
thoroughly distributed in the outermost surface of the micro-
spheres, which confirms the hierarchical core–shell structure
of Fe3O4@C–Pd@mCeO2 obtained by the elaborate assembly
procedures and is illustrated in Scheme 1.
To further confirm the presence of Pd NPs and CeO2 in the
shell, surface analysis of the prepared nanospheres was carried
out using XPS (Fig. 4). In Fig. 4a, it is difficult to observe the
peak of iron element from the XPS spectrum of Fe3O4@C–Pd
nanospheres, suggesting that the Fe3O4 nanospheres were
thoroughly coated by a carbon layer to form a core–shell struc-
ture.42 The regional XPS spectrum of Fig. 4b shows two peaks
at 341.1 and 335.7 eV, which correspond to 3d3/2 and 3d5/2 of
Pd(0), respectively. This indicates that the Pd(0) NPs were
Fig. 2 TEM images of (a and b) Fe3O4@C nanospheres; (c and d)
Fe3O4@C–Pd nanospheres; (e and f) Fe3O4@C–Pd@mCeO2 nano-
spheres; (g) EDX corresponding to the red circular region of panel f.
ous monodispersed Pd NPs were uniformly deposited on the
carbon shell with an average size of 4–5 nm, producing
Fe3O4@C–Pd nanospheres (Fig. 2c and d). After further depo-
sition of a homogenous CeO2 shell, Fe3O4@C–Pd@mCeO2
nanospheres with
a regular spherical morphology were
obtained (Fig. 2e and f).
However, because of the high electron density of polycrys-
talline CeO2,41 the small size of Pd NPs and the specific core–
shell architecture, it is difficult to recognize the interface
between Pd NPs and CeO2. Therefore, the encapsulated Pd NPs
Fig. 3 (a and b) HR-TEM images of Fe3O4@C–Pd@mCeO2 nano-
spheres, (c) HADDF-STEM images of Fe3O4@C–Pd@mCeO2 nano-
spheres and corresponding EDS mapping images.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
Fig. 4 XPS spectra of (a) Fe3O4@C–Pd nanospheres and (b) Pd 3d; (c) Fe3O4@C–Pd@mCeO2 nanospheres and (d) Ce 3d.
formed on the surface of the Fe3O4@C nanospheres.42 The tion peaks of Fe3O4@C–Pd nanospheres and Fe3O4@C–
XPS spectrum of Fe3O4@C–Pd@mCeO2 nanospheres (Fig. 4c) Pd@mCeO2 nanospheres are similar to those of the Fe3O4@C
reveals that carbon, oxygen and cerium are present on the nanospheres. However, it is noteworthy that the intensity of
surface of the nanospheres. Concerning the characteristic absorption peaks was significantly weaker than that of
signals of the Ce3d spectrum, the four primary 3d3/2 peaks Fe3O4@C nanoparticles after being anchored with Pd nano-
were located around 917.8 eV, 907.8 eV, 904.5 eV and 901.6 eV, particles and coated by CeO2.
matching the α1, α2, α3 and α4 components, respectively,
The phase and composition of the resulting products are
whereas the Ce3d5/2 peaks were located around 898.7 eV, 888.0 profoundly characterized by XRD in Fig. 5. In Fig. 5a, all the
eV, 885.3 eV, and 883.0 eV in accordance with β1, β2, β3 and β4, peaks at about 29.9°, 35.3°, 43.0°, 56.9° and 62.5° are associ-
respectively (Fig. 4d).43 Both the signals α3 and β3 are the ated with Fe3O4 (220), (311), (440), (511) and (220) peaks of a
characteristic peaks of Ce3+, while the other peaks correspond face centered cubic. According to Fig. 5b, there is no obvious
to the characteristic peaks of Ce4+. The presence of Ce3+ can be change in the XRD of the Fe3O4@C nanospheres compared
conducive to the synergistic effect between the CeO2 nano- with that of the Fe3O4 nanospheres. The XRD pattern of
particles and noble metal NPs.44 The Ce3+ : Ce4+ ratio in the Fe3O4@C–Pd is shown in Fig. 5c; addition the peaks of Fe3O4
Fe3O4@C–Pd@mCeO2 catalyst was ca. 18%.
core, additional peaks are located near 40.0°, which are well-
Fig. S2† shows the FT-IR spectra of Fe3O4 nanospheres, indexed to face centered Pd (111) peaks (JCPDS no. 05-0681).42
core–shell Fe3O4@C nanospheres, Fe3O4@C–Pd nanospheres Fig. 5d and e show the XRD patterns of Fe3O4@C@mCeO2
and Fe3O4@C–Pd@mCeO2 nanospheres synthesized in this nanoparticles and core–shell Fe3O4@C–Pd@mCeO2 nano-
study. The FT-IR spectrum of magnetite nanoparticles particles; both the figures exhibit peaks at 28.7°, 33.2°, 47.8°,
(Fig. S2a†) shows absorption bands at 583 and 1650 cm−1
,
and 56.3°, corresponding to the (111), (200), (220), and (311)
which are attributed to the vibration mode of Fe–O and the Bragg diffraction of CeO2 (JCPDS no. 34-0394).39 There are no
carbonyl group of trisodium citrate.45 The bands around diffraction peaks matching to those of Pd nanoparticles in the
3400 cm−1 are ascribed to the OH vibration. From Fig. S1b,† Fig. 5e owing to their high dispersity and the low content
for core–shell Fe3O4@C nanoparticles, the peaks around loading.42
1620 cm−1 are related to the aromatization of glucose during
hydrothermal treatment.46 From Fig. S2c and S2d,† the absorp- gated using a SQUID magnetometer at 300 K from −20 000 to
The magnetic performances of the samples were investi-
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
3.5 nm, indicating a mesoporous CeO2 shell. This result is in
good agreement with the literature.40
The successful synthesis of Fe3O4@C–Pd@mCeO2 nano-
spheres can be attributed to the following outlines. First, the
Fe3O4 nanoparticles synthesized by the solvothermal reaction
are modified by citrate groups. Therefore, they can be highly
dispersible in water and ethanol, favouring the further coating
with other oxides or polymers.45 Thus, a thin carbon layer can
be readily coated onto the Fe3O4 nanoparticles by the partial
carbonization and polymerization of glucose under hydro-
thermal treatment.46 Second, the existence of reductive groups
(such as –OH and –CHO) in the carbon shell, appeared due to
the dehydration of the glucose, can facilitate in situ reduction
of PdCl2. Therefore, no additional reducing or linker agents
are needed.22 Third, because the obtained Fe3O4@C–Pd nano-
spheres are hydrophilic and have good dispersibility in
ethanol and water, the subsequent surface coating of the
mesoporous CeO2 can be achieved.
Fig. 5 XRD pattern of (a) Fe3O4 nanospheres, (b) Fe3O4@C nano-
spheres, (c) Fe3O4@C–Pd nanospheres, (d) Fe3O4@C@mCeO2 nano-
spheres and (e) Fe3O4@C–Pd@mCeO2 nanospheres.
Based on the high dispersibility, we evaluated the catalytic
efficiency for the Suzuki–Miyaura reaction (cross-coupling of
4-iodoanisole and phenylboronic acid) in a green medium of
water or ethanol without any external promoters, phosphine
ligands or biphasic media. From the results in Table S1,† we
can see that a low-grade yield of the desired resultant was
obtained by the Fe3O4@C@mCeO2 without Pd NPs in an
ethanol solvent (Table S1,† entry 1). No obvious increase in
yield was observed by increasing the reaction time (Table S1,†
entry 2). This suggests that ceria has a catalytic effect for the
Suzuki–Miyaura reaction.47 With water as the solvent, the reac-
tion did not occur at all (Table S1,† entry 3). Significantly, the
model reaction using Fe3O4@C–Pd or Fe3O4@C–Pd@mCeO2 as
the catalyst in water or ethanol or their mixture at 80 °C
afforded the desired product within 3 h in excellent isolated
yield (Table S1,† entries 4–9). Interestingly, other solvents,
such as isopropanol, tetrahydrofuran, N,N-dimethyl-
formamide, 1,4-dioxane, dimethyl sulfoxide, dimethyl-
acetamide, toluene and dimethylbenzene, supported the same
catalytic reaction of Fe3O4@C–Pd@mCeO2 fairly well
(Table S2,† entries 1–8).
20 000 Oe. Fig. S3† shows the magnetization curves of Fe3O4
nanospheres (black), Fe3O4@C nanospheres (red), Fe3O4@C–
Pd nanospheres (blue) and Fe3O4@C–Pd@mCeO2 nano-
particles (pink). At 300 K, the saturation magnetization value
is 74.9, 25.5, 24.6 and 21.3 emu per g, respectively. The inset
in Fig. S3† shows the photograph of the dispersion of 5 mg
mL−1 Fe3O4@C–Pd@mCeO2 nanospheres in water before and
after magnetic separation. To further analyse the pore size dis-
tribution, BET measurements were carried out. As shown in
Fig. 6, the N2 adsorption–desorption isotherms of the
Fe3O4@C–Pd@mCeO2 composites exhibit representative type-
IV curves, and the BET surface area is about 85 m2 g−1. The
pore size distribution shows that the pores size mostly range
from 2.0 to 4.5 nm with an intense peak appearing at around
To explore the expansion of this reaction, a series of aryl
iodides were selected to react with phenylboronic acid with a
mixed solvent of ethanol/water (1 : 1) and K2CO3. It is worth
noting that Fe3O4@C–Pd@mCeO2 as the catalyst can tolerate a
wider range of functional groups (Table 1, entries 2–7) such as
–Me, –OCH2CH3, –OCF3, –OH, –NH2 and –COCH3. Interest-
ingly, aryl bromides delivered the products with high yields
under identical conditions (Table 1, entries 8–9). However, aryl
chloride is relatively troublesome for the same process provid-
ing a medium yield (Table 1, entries 10–11). An array of substi-
tuted arylboronic acid substrates also underwent the cross-
coupling reaction with aryl halide with our catalytic system
(Table 1, entries 13–30). Ortho-methyl phenylboronic acid gave
good yields of the 2-arylation product (Table 1, entries 13, 18,
23, and 27), which was not impaired by steric hindrance. A sat-
isfactory yield was also obtained by 1-naphthylboronic acid for
the formation of 1-arylation naphthalane (Table 1, entries 16,
Fig. 6 Nitrogen adsorption–desorption isotherm plots of Fe3O4@C–
Pd@mCeO2 and pore size distribution curves of Fe3O4@C–Pd@mCeO2
(inset).
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
The catalytic property of the Fe3O4@C–Pd@mCeO2 compo-
sites was also evaluated by the reduction of 4-NP to 4-amino-
phenol (4-AP) in the presence of NaBH4, which is one of the
representative model reactions for estimating the catalytic
activity of the noble metal NPs.42 As shown in Fig. 7a, when a
trace amount of the Fe3O4@C–Pd@mCeO2 nanocatalyst (2 mg)
is introduced into the solution, the absorption peak at 400 nm
decreases in a certain period of time and, simultaneously, a
new absorption peak at 295 nm arises, which can be attributed
to the reduction of 4-NP and the emergence of 4-AP, respecti-
vely. For comparison, the catalytic activities of Fe3O4@C@
mCeO2 and Fe3O4@C–Pd as reference catalysts have also been
investigated following the identical experimental conditions.
The conversion can be directly determined from the curves
made of the percentage of the absorbance intensity of 4-NP at
an interval time to its initial absorbance intensity value, as dis-
played in Fig. 7b. It is obvious that Fe3O4@C@mCeO2 exhibits
negligible catalytic reduction activity, suggesting that the
reduction reaction is mainly catalyzed by the Pd NPs, whereas
Table 1 The Suzuki reactions catalyzed by Fe3O4@C–Pd@mCeO2
catalysta
Entry
R1
X
R2
Yieldb
TOFd
1
2
3
4
5
6
7
8
H
OCF3
OH
OCH2CH3
NH2
CH3
COCH3
H
I
I
I
I
I
I
I
Br
Br
Cl
Cl
I
I
I
I
I
I
I
H
H
H
H
H
H
H
H
H
H
H
H
>99c
93
95
92
95
95
90
95
90
50
58
99
90
95
98
85
82
75
88
93
73
88
68
78
85
72
75
85
86
77
345.4
108.2
116.3
107.0
116.3
116.3
104.6
116.3
104.6
58.2
9
CH3
H
CH3
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
67.5
OCH3
OCH3
OCH3
OCH3
OCH3
COCH3
COCH3
COCH3
COCH3
COCH3
H
H
H
H
H
CH3
CH3
CH3
CH3
115.1
104.6
116.3
113.9
98.9
95.4
88.4
102.3
108.2
84.9
102.3
79.1
90.7
98.9
2-CH3
4-CH3
4-F
1-Naphthyl
4-OCH3
2-CH3
4-CH3
4-F
1-Naphthyl
4-OCH3
2-CH3
4-CH3
4-F
1-Naphthyl
2-CH3
4-OCH3
4-F
I
I
I
Br
Br
Br
Br
Br
Br
Br
Br
Br
83.7
88.4
98.9
100.0
90.7
1-Naphthyl
a Reaction conditions: aryl halide (1.0 mmol), arylboronic acid (1.2 mmol),
potassium carbonate (2 mmol), 5 mL ethanol–water (1 : 1), 10 mg
Fe3O4@C–Pd@mCeO2 (Pd 3.05 wt%) at 80 °C for 3 h. b Isolated yields.
c 60 °C for 45 min. d TOF was defined as mol product mol−1 Pd h−1
.
21, 26, and 30). Both electron-withdrawing (4-F) and electron-
donating (4-OCH3) groups did not show an inhibition effect
under the same conditions (Table 1, entries 15, 17, 20, 25, 28,
and 29).
We further compared the catalytic results obtained with
this hierarchical “shell-in-shell” structure with other tra-
ditional supported Pd-based catalysts such as mesoporous
carbon, polymer, silica or oxide. With the reaction of iodo-
benzene and phenylboronic acid as an example, the results are
listed in Table S4.† It can be clearly seen that the results
achieved in this study are superior to others. Even though
some of these materials also can acquire high yield, toxic sol-
vents (such as DMF, CH2Cl2, or THF) or a long time were
required. These solvent are less favorable compared to the
ethanol–water mixture used in this study. The comparison of
the catalytic performance between the catalyst of Fe3O4@C–
22
Pd@mSiO2 and our Fe3O4@C–Pd@mCeO2 system has also
Fig. 7 (a) UV-vis spectra of 4-NP and NaBH4 solution in the presence
been summarized in Table S5.† It can be clearly observed that
the catalytic performance of our Fe3O4@C–Pd@mCeO2 system
is better than that of the catalyst of Fe3O4@C–Pd@mSiO2.
of
2 mg Fe3O4@C–Pd@mCeO2; (b) Performance of Fe3O4@C–
Pd@mCeO2, Fe3O4@C–Pd, and Fe3O4@C@mCeO2 nanocomposites for
the reduction of 4-NP.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
the supported Fe3O4@C–Pd possesses inferior catalytic
performance
as
compared
to
Fe3O4@C–Pd@mCeO2
nanocomposites.
In addition to the outstanding catalytic activity, isolation
and reusability of the catalyst is a considerable requirement
for any practical application in accordance with budget and
environmental protection,48–50 are the greatest merits in our
study. In our systems, the catalysts can be easily retrieved
using an external magnet, as proved shown in Fig. S3† inset.
In order to investigate the reusability of the catalyst, both the
Suzuki–Miyaura reaction (cross-coupling of 4-iodoanisole and
phenylboronic acid) and the reduction of 4-NP were chosen. As
illustrated in Fig. S4a and S4b,† the catalyst Fe3O4@C–
Pd@mCeO2 can be reused ten times without an obvious
decrease in conversion and selectivity for both the reactions by
a simple magnetic separation. However, the results of Fig. S4c
and S4d† reveal that the catalytic activity of the Fe3O4@C–Pd
was found to distinctly decrease during the recycling test after
the two reduction process were repeated four times. Fig. S5†
shows the TEM images of Fe3O4@C–Pd@mCeO2 after 10 runs
of recycling experiments for the Suzuki reactions. The images
indicate that the catalyst still maintained a spherical shape
after the recycling experiments.
Scheme 2 The proposed mechanism for the synergistic effect between
the CeO2 NPs and Pd NPs with improved catalytic activity.
Based on the abovementioned data, it could be readily con-
cluded that the catalytic activity of Fe3O4@C–Pd@mCeO2 is
better than that of Fe3O4@C–Pd for the two reactions, and it
can be easily deduced that the Fe3O4@C–Pd@mCeO2 core–
shell nanostructure can effectively block the aggregation of the
Pd NPs due to the robust protection of CeO2 shells, thereby
holding the high catalytic property of the Pd NPs. The other
more remarkable advantage of Fe3O4@C–Pd@mCeO2 as a dis-
tinct compared to Fe3O4@C–Pd for mediating these reactions
in solution is that the CeO2 shells can effectively hinder the
leaching of Pd NPs. The leaching test of Pd was analysed by
ICP-AES after ten cycles, and the results are listed in Table S3.†
Both reactions show determinable Pd loss; the trace content of
Pd loss indicates that Fe3O4@C–Pd@mCeO2 can be a reusable
catalyst with additional anti-deactivation performance. This
prominent characteristic can be confirmed by the very steady
recycling catalytic activity test on Fe3O4@C–Pd@mCeO2
(Fig. S4a and S4b†) and its steady nanostructure.
Over the past few decades, the comprehending of the inter-
facial effect between the noble metal NPs and the metal oxide
supports is still ambiguous for heterogeneous reactions
because multiple factors influence the catalytic activity of Pd
nanocatalysts.51,52 Many metal oxides have been employed as
catalyst supports for the loading of the noble metal NPs to
boost their catalytic activity and stability. CeO2, which has a
Ce4+/Ce3+ redox cycle, is an ideal carrier due to its high oxygen
storage capability, high oxygen mobility and low cost.53,54
CeO2 belongs to the fluorite structure, in which each O2−
anion is encircled by a tetrahedron of Ce4+ cations situated at
the center of a cubic array of equivalent O2− atoms.55 However,
the non-stoichiometric oxides CeO2−x, which were obtained
in the CeO2 structure can form O vacancy. The surplus elec-
trons may either thoroughly delocalize in the conductor band,
or distribute among a few Ce3+ cations to surround the O
vacancies, or localize on the Ce4+ to form tetravalent Ce.56 Ce3+
is an electron-rich species and tends to afford electrons to metal
catalysts.57 Simultaneously, water dissociation occurring at the
O vacancy can produce the reactive OH groups.58 Both the Ce3+
species and OH groups are reactive δ− sites, which can endow
electrons to Pd NPs. Therefore, the Ce3+ fraction and O-vacancy
concentration can be used to comprehend the synergistically
catalytic effect for enhancing the catalytic activity.
A proposed reaction mechanism for the synergistically cata-
lytic effect is illustrated in Scheme 2. During the wet-chemical
synthesis process of the CeO2, the Ce3+ cationic species (which
was also confirmed by XPS in Fig. 3d) and O vacancies are
formed. The active OH groups will be produced through the
dissociation of water on the O-vacancy sites.59 Afterwards, the
electron pair donors of the Ce3+ cationic species and OH
groups can be transferred to the Pd NPs via the electron-donat-
ing effect, leading to a higher electron density Pd NPs. This
serves to improve the first step of the oxidative addition reac-
tion, formation of the radical ligand Ar–PdII–Br. This step is
regarded as the key step in the C–C coupling reactions.52 The
high density of electron of Pd NPs also can rapidly catalyze the
reduction of 4-NP.
Conclusions
through the wet-chemical synthesis, have a high density of In summary, we demonstrate a successful preparation of
oxygen vacancies. The partial elimination of O atoms from O2− multicomponent and multifunctional Fe3O4@C–Pd@CeO2
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
nanospheres with well-defined magnetic core–shell nano- 18 S. Sá, M. B. Gawande, A. Velhinho, J. P. Veiga,
structures, confined catalytic Pd NPs and accessible meso-
porous CeO2 by combining the hydrothermal method,
interfacial in situ deposition and a sol–gel process. The well-
N. Bundaleski, J. Trigueiro, A. Tolstogouzov, O. M. N. D.
Teodoro, R. Zboril, R. S. Varma and P. S. Branco, Green
Chem., 2014, 16, 3494–3500.
designed hierarchical nanospheres have high magnetization 19 M. B. Gawande, R. Luque and R. Zboril, ChemCatChem,
(21.3 emu per g), with a highly open mesoporous character- 2014, 6, 3312–3313.
istics (∼3.5 nm in diameter), and smoothly restricted but 20 D. Wang, Astruc. Chem. Rev., 2014, 114, 6949–6985.
uncovered catalytic Pd NPs that uniformly disperse between 21 H. Hildebrand, K. Mackenzie and F. D. Kopinke, Environ.
the carbon-protected Fe3O4 core and the mesoporous CeO2
shell. The as-obtained multifunctional Fe3O4@C–Pd@CeO2 22 Z. Sun, J. Yang, J. Wang, W. Li, S. Kaliaguine, X. Hou,
nanospheres show excellent catalytic performance in the Y. Deng and D. Zhao, J. Mater. Chem. A, 2014, 2, 6071–6074.
Suzuki–Miyaura cross-coupling reaction and the reduction of 23 Q. M. Kainz and O. Reiser, Acc. Chem. Res., 2014, 47, 667–
4-nitrophenol both with convenient separability and remark- 677.
able reusability without loss activity after reusing ten times. 24 X. Le, Z. Dong, Y. Liu, Z. Jin, T. Huy, M. Le and J. Ma,
Therefore, as a novel Pd-based catalyst system, this multifunc- J. Mater. Chem. A, 2014, 2, 19696–19706.
tional core–shell nanostructure holds great promise for 25 X. X. Han, A. M. Schmidt, G. Marten, A. Fischer,
Sci. Technol., 2009, 43, 3254–3259.
various catalytic reactions. In addition, the design idea for the
hierarchical nanomaterials can be expanded to the synthesis
I. M. Weidinger and P. Hildebrandt, ACS Nano, 2013, 7,
3212–3220.
of other multicomponent nanomaterials with integrated and 26 J. Cao, J. C. Li, L. Liu, A. J. Xie, S. K. Li, L. G. Qiu, Y. P. Yuan
enhanced capabilities for various applications.
and Y. H. Shen, J. Mater. Chem. A, 2014, 2, 7593–7597.
27 Z. J. Wu, C. G. Sun, Y. Chai and M. G. Zhang, RSC Adv.,
2011, 1, 1179–1182.
28 X. B. Zhang, H. W. Tong, S. M. Liu, G. P. Yong and
Y. F. Guan, J. Mater. Chem. A, 2013, 1, 7488–7494.
29 Q. M. Kainz, R. Linhardt, R. N. Grass, G. Vilé, W. J. Stark
and O. Reiser, Adv. Funct. Mater., 2014, 24, 2020–2027.
30 M. B. Gawande, Y. Monga, R. Zboril and R. K. Sharma,
Coord. Chem. Rev., 2015, 288, 118–143.
31 R. Li, P. Zhang, Y. Huang, P. Zhang, H. Zhong and
Q. Chen, J. Mater. Chem. A, 2012, 22, 2275–22755.
32 T. Yao, T. Cui, J. Wu, Q. Chen, X. Yin, F. Cui and K. Sun,
Carbon, 2012, 50, 2287–2295.
References
1 G. V. Hartland, Chem. Rev., 2011, 111, 3858–3887.
2 S. N. Tian, Z. Zhou, S. Sun, Y. Ding and Z. L. Wang, Science,
2007, 316, 732–735.
3 T. K. Sau, A. L. Rogach, F. Jäckel, T. A. Klar and
J. Feldmann, Adv. Mater., 2010, 22, 1805–1825.
4 S. Linic, P. Christopher and D. B. Ingram, Nat. Mater.,
2011, 10, 911–921.
5 A. Scholl and A. Dionne, Nature, 2012, 483, 421–427.
6 L. Chen, G. Chen, C. Leung, S. Yiu, C. Ko, E. Anxolabéhère- 33 T. Zeng, X. Zhang, S. Wang, Y. Ma, H. Niu and Y. Cai, Chem
Mallart, M. Robert and T. Lau, ACS Catal., 2015, 5, 356–364. – Eur. J., 2014, 20, 6474–6481.
7 L. Zhong, A. Chokkalingam, W. S. Cha, K. S. Lakhi, X. Su, 34 O. O. Fashedemi, H. A. Miller, A. Marchionni, F. Vizza and
G. Lawrence and A. Vinu, Catal. Today, 2015, 243, 195–198. K. I. Ozoemena, J. Mater. Chem. A, 2015, 3, 7145–7156.
8 A. Sanchez, S. Abbet, U. Heiz, W. D. Schneider, 35 M. Shokouhimehr, T. Kim, S. W. Jun, K. Shin, Y. Jang,
H. Häkkinen, R. N. Barnett and U. Landman, J. Phys. Chem.
A, 1999, 103, 9573–9578.
B. H. Kim, J. Kim and T. Hyeon, Appl. Catal., A, 2014, 476,
133–139.
9 E. Yoo, T. Okata, T. Akita, M. Kohyama, J. Nakamura and 36 B. Liu, Q. Wang, S. Yu, T. Zhao, J. Han, P. Jing, W. Hu,
I. Honma, Nano Lett., 2009, 9, 2255–2259.
10 C. An, S. Peng and Y. Sun, Adv. Mater., 2010, 22, 2570–2574.
L. Liu, J. Zhang, L. Sun and C. Yan, Nanoscale, 2013, 5,
9747–9757.
11 K. Soukup, P. Topka, V. Hejtmánek, D. Petráš, V. Valeš and 37 B. Liu, Y. Niu, Y. Li, F. Yang, J. Guo, Q. Wang, P. Jing,
O. Šolcová, Catal. Today, 2014, 236, 3–11.
12 R. G. Chaudhuri and S. Paria, Chem. Rev., 2012, 112, 2373–
2433.
13 J. M. Thomas, B. F. G. Johnson, R. Raja, G. Sankar and
P. A. Midgley, Acc. Chem. Res., 2003, 36, 20–30.
J. Zhang and G. Yun, Chem. Commun., 2014, 50, 12356–
12359.
38 Y. Deng, Y. Cai, Z. Sun, J. Liu, C. Liu, J. Wei, W. Li, C. Liu,
Y. Wang and D. Zhao, J. Am. Chem. Soc., 2010, 132, 8466–
8473.
14 M. Liu, R. Z. Zhang and W. Chen, Chem. Rev., 2014, 114, 39 X. Li, X. Wang, D. Liu, S. Song and H. Zhang, Chem.
5117–5160. Commun., 2014, 50, 7198–7201.
15 Y. Zhou, H. Wang, M. Gong, Z. Sun, K. Cheng, X. Kong, 40 G. Cheng, J. Zhang, Y. Liu, D. Sun and J. Ni, Chem.
Z. Guo and Q. Chen, Dalton Trans., 2013, 42, 9906–9913. Commun., 2011, 47, 5732–5734.
16 M. B. Gawande, R. Zboril, V. Malgras and Y. Yamauchi, 41 S. Zhang, J. Li, W. Gao and Y. Qu, Nanoscale, 2015, 7, 3016–
J. Mater. Chem. A, 2015, 3, 8241–8245. 3021.
17 M. B. Gawande, P. S. Branco and R. S. Varma, Chem. Soc. 42 T. J. Yao, T. Y. Cui, X. Fang and J. Wu, Nanoscale, 2013, 5,
Rev., 2013, 42, 3371–3393.
5896–5904.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
43 F. Lin, D. T. Hoang, C. K. Tsuang, W. Y. Huang and 51 G. Chen, Y. Zhao, G. Fu, P. N. Duchesen, L. Gu, Y. Zheng,
P. D. Yang, Nano Res., 2011, 4, 61–71.
X. Weng, M. Chen, P. Zhang, C. W. Pao and N. F. Zhen,
44 Q. Wang, W. J. Jia, B. C. Liu, X. Gong, C. Y. Li, P. Jing,
Science, 2014, 344, 495–499.
Y. J. Li, G. R. Xu and J. Zhang, J. Mater. Chem. A, 2013, 1, 52 J. Shi, Chem. Rev., 2012, 113, 2139–2181.
12732–12741.
53 S. Tsunekawa, K. Ishikawa, Z.-Q. Li, Y. Kawazoe and
A. Kasuya, Phys. Rev. Lett., 2000, 85, 3440–3443.
45 J. Liu, Z. K. Sun, Y. H. Deng, Y. Zou, C. Y. Li, X. H. Guo,
L. Q. Xiong, Y. Gao, F. Y. Li and D. Y. Zhao, Angew. Chem., 54 Y. Y. Chu, Z. B. Wang, Z. Z. Jiang, D. M. Gu and G. P. Yin,
Int. Ed., 2009, 48, 5875–5879. Adv. Mater., 2011, 23, 3100–3104.
46 M. Y. Zhou and G. W. Diao, J. Phys. Chem. C, 2011, 115, 55 Q. Wang, Y. J. Li, B. C. Liu, Q. Dong, G. R. Xu and
24743–24749. L. Zhang, J. Mater. Chem. A, 2015, 3, 139–147.
47 A. S. Diez, M. G. Mayer and M. A. Volpe, Appl. Catal. A, 56 M. F. Camellone and S. Fabris, J. Am. Chem. Soc., 2009, 131,
2014, 482, 24–30. 10473–10483.
48 V. Polshettiwar, R. Luque, A. Fihri, H. B. Zhu, M. Bouhrara 57 H. Y. Kim, H. M. Lee and G. Henkelman, J. Am. Chem. Soc.,
and J. M. Basset, Chem. Rev., 2011, 111, 3036–3075. 2012, 134, 1560–1570.
49 J. Liu, S. Z. Qiao, Q. H. Hu and G. Q. Lu, Small, 2011, 4, 58 Y. G. Wang, D. Mei, J. Li and R. Rousseau, J. Phys. Chem. C,
425–443. 2013, 117, 23082–23089.
50 Y. H. Deng, Y. Cai, Z. K. Sun and D. Y. Zhao, Chem. Phys. 59 Y. Wang, F. Wang, Q. Song, Q. Xin, S. Xu and J. Xu, J. Am.
Lett., 2011, 510, 1–13. Chem. Soc., 2013, 135, 1506–1515.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015