Magnetic core-shell-structured nanoporous organosilica microspheres for the Suzuki-Miyaura coupling of aryl chlorides: Improved catalytic activity and facile catalyst recovery

Article abstract of DOI:10.1039/c2jm15693h

New magnetic core-shell-structured nanoporous organosilica microspheres consisting of a Fe₃O₄ core and an organosilica shell with a built-in N-heterocyclic carbene (NHC) moiety were synthesized through surfactant-directed self-assembly on surface. The structure and composition of the material thus-designed were confirmed by N₂ sorption, XRD, TEM, FT-IR, etc. Such composite microspheres possess highly open mesopores (1.6 nm) in the shell, moderate surface area, tunable shell thickness, uniform morphology, high magnetization as well as good coordination ability toward Pd. The catalytic activity test with the Suzuki-Miyaura couplings of challenging aryl chlorides reveals that this Pd-coordinated material is highly active for a wide range of substrates, suggesting highly promising application potentials of the magnetic core-shell-structured nanoporous organosilica. In particular, the catalyst derived from this material is superior to NHC-functionalized MCM-41 microspheres and a commercial Pd/C catalyst in terms of activity and recovery. The catalytic activity enhancement effects may be attributable to the purposely designed core-shell structure with significantly decreased diffusion depth, which is crucial for designing high-performance solid catalysts. The Royal Society of Chemistry 2012.

Full text of DOI:10.1039/c2jm15693h

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
Journal of
Materials Chemistry
Cite this: J. Mater. Chem., 2012, 22, 6639
www.rsc.org/materials
PAPER
Magnetic core–shell-structured nanoporous organosilica microspheres for the
Suzuki–Miyaura coupling of aryl chlorides: improved catalytic activity and
facile catalyst recovery†
*
Hengquan Yang, Guang Li and Zhancheng Ma
Received 6th November 2011, Accepted 30th January 2012
DOI: 10.1039/c2jm15693h
New magnetic core–shell-structured nanoporous organosilica microspheres consisting of a Fe3O4 core
and an organosilica shell with a built-in N-heterocyclic carbene (NHC) moiety were synthesized
through surfactant-directed self-assembly on surface. The structure and composition of the material
thus-designed were confirmed by N2 sorption, XRD, TEM, FT-IR, etc. Such composite microspheres
possess highly open mesopores (1.6 nm) in the shell, moderate surface area, tunable shell thickness,
uniform morphology, high magnetization as well as good coordination ability toward Pd. The catalytic
activity test with the Suzuki–Miyaura couplings of challenging aryl chlorides reveals that this
Pd-coordinated material is highly active for a wide range of substrates, suggesting highly promising
application potentials of the magnetic core–shell-structured nanoporous organosilica. In particular, the
catalyst derived from this material is superior to NHC-functionalized MCM-41 microspheres and
a commercial Pd/C catalyst in terms of activity and recovery. The catalytic activity enhancement effects
may be attributable to the purposely designed core–shell structure with significantly decreased diffusion
depth, which is crucial for designing high-performance solid catalysts.
fabricated a magnetic Au catalyst by encapsulation of a core–
Introduction
satellite composite within a porous silica shell.5 This material
exhibited an excellent catalytic stability. Yang’s group trapped
Au nanoparticles in yolk–shell-structured organosilica micro-
spheres, which exhibited high acidity in the reduction of 4-
nitrophenol.2h Obviously, the development of multifunctional
magnetic porous microspheres has made exciting progress and
they exhibit highly promising potential applications. However, in
contrast to the ever-expanding intended applications that often
require porous silica with specific organic or inorganic func-
tionalization, thus providing properties not obtainable from pure
silica, the applications of catalytically active magnetic core–shell
nanoporous silica microspheres are rather limited.
Currently, core–shell microspheres are attracting extensive
attention because the ready adjustability and functionality, both
of the cores and the shells, can confer them with a wide number
of potential applications in catalysis, drug delivery, biosensor,
and so forth.1 Of various core–shell structured microspheres,
magnetic porous silica microspheres consisting of a magnet core
and a nanoporous silica shell are of particular interest owing to
the unusual properties that, in addition to a magnet-responsive
function, the nanoporous silica shell can provide a versatile
platform for designing more sophisticated materials.2 For
example, Shi and co-workers synthesized magnetic core–shell-
structured nanoporous silica nanospheres, which exhibited
excellent performances in drug delivery.3 Zhao’ group prepared
multifunctional microspheres comprised of a core of Au-sup-
ported Fe3O4–silica composite and an outer shell of ordered
nanoporous silica with perpendicularly aligned pore channels.4
This multifunctional microsphere were highly active in the
reduction of 4-nitrophenol and styrene epoxidation. Yin et al.
Over the past few years, mesoporous organosilicas, synthe-
sized through co-condensation of organic moiety-bridged tri-
alkoxysilane and tetraalkoxysilane with the aid of templates in
one pot, have also aroused considerable interest in the fields of
catalysis, sorption and drug delivery due to their high surface,
large pore volume and tunable pore sizes.6 Compared with the
organo-modified mesoporous silica prepared by post grafting,
mesoporous organosilica usually has a larger surface area, higher
pore volume and less diffusion resistance because of the incor-
poration of organic moieties in the framework instead of their
being grafted on the pore surface. Moreover, organic moieties
are more homogeneously distributed in the framework of the
materials thanks to the self-assembly at the molecular level. Salen
complex-, tartardiamide-, binol-, imidazolium-, alkyl- and
School of Chemistry and Chemical Engineering, Shanxi University,
Taiyuan 030006, P. R. China. E-mail: hqyang@sxu.edu.cn; Fax: +86-
351-7011688; Tel: +86-351-7010588
† Electronic supplementary information (ESI) available: The TEM
images, XRD pattern of NHC-functionalized MCM-41, photograph of
magnetic isolation and 1H NMR data of the coupling products. See
DOI: 10.1039/c2jm15693h
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 6639–6648 | 6639

View Article Online
aryl-functionalized mesoporous organosilicas, to name a few,
were successfully prepared and showed promising performances
in catalysis.6e,7 However, the conventional synthesis protocols
usually lead to mesoporous organosilicas with non-uniform
morphology and/or particle size as large as several hundred
nanometres to micrometres. The diffusion limitation phenomena
caused by the long nano-channels in the large particles are often
observed. The activity of a complex catalyst immobilized on
a mesoporous material is generally much lower than the free
complex catalyst in solution.8
outlining the synthesis of new magnetic core–shell-structured
organosilica microspheres with molecular functionalization in
the shell and disclose their unusual properties.
Experimental
Reagents and materials
Pd(acac)2 (acac ¼ acetylacetonate) were obtained from Kaida
Metal Catalyst & Compounds Co. Ltd (China). Various aryl-
boronic acids were bought from Beijing Pure Chemical Co. Ltd.
Pd/C catalyst (1 wt% Pd) was purchased from the Aladdin
Company. Most of the aryl chlorides, ICl, FeCl3, tetraethyl
orthosilicate (TEOS) and cetyltriethylammonium bromide
(CTAB) were purchased from the Aladdin Company. 2,6-Dii-
sopropylaniline, concentrated ammonia solution (28 wt%) and
other reagents were obtained from Shanghai Chemical Reagent
Company of Chinese Medicine Group. 1,3-Bis(4-triethoxysilyl-
2,6-diisopropylphenyl)-imidazol-3-iumtrifluorometh-ane-sulfo-
nate (NHC-bridged organosilane) was synthesized according to
our previous publication.7c
Decreasing particle sizes down to the nanoscale may be an
effective way to mitigate the diffusion limitations because the
diffusion lengths of reactant molecules are dramatically short-
ened. However, when the particle sizes are less than 100 nm, the
isolation of these nanoparticles from a reaction system is rela-
tively difficult and time-consuming because of the unavoidable
need for high speed centrifugation. The diffusion limitations and
catalyst recovery might be simultaneously addressed if the
catalytically active part was positioned surrounding magnetic
nanospheres (resulting in a multifunctional core–shell structure)
and the catalytic reaction thus occurs only in the outer nano-
porous shell instead of throughout the solid material. However,
to the best of our knowledge, there has been no report on the
preparation and catalytic applications of magnetic core–shell-
structured nanoporous organosilica microspheres with a built-in
active phase only in the shell.
Synthesis of magnetic NHC-functionalized nanoporous
organosilica
Fe3O4 particles were synthesized according to the reported
method.11 0.4 g of Fe3O4 particles was dispersed in 200 mL of
0.1 mol HCl and was subject to ultrasonic treatment. After
10 min, this material was washed with water and isolated with the
aid of a magnet. The resultant material was re-dispersed into
a mixture solution (320 mL of ethanol, 80 mL of water and
4.2 mL of concentrated ammonia solution). Under stirring,
0.12 g of tetraethyl orthosilicate (TEOS) was added into the
mixture solution. After this mixture was further stirred for 6 h at
room temperature, the resultant solid was separated, repeatedly
washed with ethanol and dried under air, eventually leading to
SiO2-coated Fe3O4 particles.
Our recent efforts have focused on development of recoverable
solid catalysts for the Suzuki–Miyaura coupling reaction as it is
a fundamental reaction for synthesizing fine chemicals.9 Parti-
cular attention is paid to the use of aryl chlorides as substrates
because they are much cheaper and readily available than the
usually used aryl bromides and iodides. However, the catalytic
activation of aryl chlorides is considerably challenging especially
for heterogeneous catalysts. So far, heterogeneous examples for
the Suzuki–Miyaura coupling of less reactive aryl chlorides are
quite rare.10 In our recent work, we have successfully developed
active solid catalysts for this reaction by incorporation of NHC
(N-heterocyclic carbene, the most successful class of ligands in
recent organometallic chemistry) into the framework of the
mesoporous material.7c,7d However, the coupling reactions over
the resulting solid catalysts proceeded relatively sluggishly. It
required 24 h to achieve satisfactory yields and its activity was
much lower than that of the homogeneous counterpart.7c There is
a significant need to improve the catalytic activity.
0.5 g of SiO2-coated Fe3O4 particles were dispersed in 250 mL
of 0.1 mol Lꢀ1 HCl solution which was then treated under
ultrasonic conditions.4 After 10 min, the SiO2-coated Fe3O4 was
separated with a magnetic field, and then repeatedly washed with
distilled water. Next, ethanol (380 mL), water (400 mL),
concentrated ammonia solution (5.0 g, 28 wt%) and CTAB
(1.5 g) were admixed with the resultant SiO2-coated Fe3O4. After
ꢁ
the solution was stirred at 35 C for 5 h, total 4.75 mmol of the
In this context, combining the merits of magnetic particles, the
core–shell structure and nanoporous organosilica in a single
system will probably create a high-performance multifunctional
solid catalyst for the Suzuki–Miyaura coupling of challenging
substrates. In light of this idea, we herein demonstrate a novel
appraoch for the synthesis of new magnetic composite micro-
spheres consisting of a Fe3O4 core and a nanoporous organo-
silica shell with the built-in bulky NHC ligand, and we
investigate their ability in coordinating metals, and further
examine their catalytic performances. Interestingly, the Pd-
coordinated material exhibits activity enhancement effects in the
Suzuki–Miyaura coupling of aryl chlorides as compared to
NHC-functionalized MCM-41 microspheres possibly due to the
significantly shortened diffusion depth. Although magnetic core–
shell silica microspheres have been reported,2 we proceed be
total siliceous precursors dissolved in 3 mL of ethanol was added
dropwise to the above system. The molar fraction of NHC-
bridged organosilane in the total siliceous precursors was 5.26%.
The mixture was further vigorously stirred for 10 h. The resultant
microspheres were separated with a magnet. CTAB infiltrated in
the shell nanopores was removed by repeated acetone extractions
under reflux conditions. After being dried under vacuum, the
magnetic composite microspheres were eventually obtained,
denoted as Fe3O4@mSiO2-NHC(x) where x depended on the
following synthesis conditions. To obtain the composite with
different shell thicknesses, the total amount of added siliceous
precursors including TEOS and NHC-bridged organosilane (the
molar ratios of NHC-bridged organosilane and TEOS were
always kept constant at 5.26%) was varied from 4.75 mmol to
3.562 and 2.375 mmol. According to the added amount of
6640 | J. Mater. Chem., 2012, 22, 6639–6648
This journal is ª The Royal Society of Chemistry 2012

View Article Online
siliceous precursor, we obtained three materials corresponding to
Fe3O4@mSiO2-NHC(1), Fe3O4@mSiO2-NHC(2) and
Fe3O4@mSiO2-NHC(3), respectively.
pressure range of 0.05 to 0.15. The total pore volume of each
sample was estimated from the amount adsorbed at the highest
P/P0 (ca. 0.99). Magnetization measurements M(T,H) were
performed by using a superconducting quantum interference
device magnetometer (LakeShore 7410). FT-IR spectra were
collected on a PE-1730 infrared spectrometer infrared spec-
trometer. Pd content was analyzed with an inductively coupled
plasma-atomic emission spectrometry (ICP-AES, AtomScan16,
TJA Co.). X-Ray photoelectron spectra (XPS) were recorded on
a Kratos Axis Ultra DLD, and the C1S line at 284.8 eV was used
as a reference. TEM micrographs were taken with a JEM-
2000EX transmission electron microscope. GC analysis was
conducted on a SP-GC6800A.
For comparison, we synthesized NHC-functionalized MCM-
41 microspheres without Fe3O4 core. In a typical synthesis, 3.52 g
of CTAB and 2.5 mL of 1 mol Lꢀ1 sodium hydroxide solution
were dissolved in 800 g of a mixture of methanol and water (50/
50, w/w). Next, 3 mL of methanol containing 8.684 mmol of
TMOS and 0.0482 mmol of NHC-bridged organosilane was
loaded to the solution with vigorous stirring at 298 K. After
further stirring for 8 h hours at this temperature, the resulting
white precipitate was isolated by filtration and thoroughly
washed with distilled water. The template was removed through
the same procedure as with Fe3O4@mSiO2-NHC(1).
Coordination of Fe3O4@mSiO2-NHC(x) with Pd(acac)2
Results and discussion
In aꢁtypical experiment, 1 g of Fe3O4@mSiO2-NHC(1) (dried at
120 C for 4 h) was dispersed in 12 mL of dry 1,4-dioxane con-
taining 0.0063 g of Pd(acac)2 (acac ¼ acetylacetonate). After the
mixed system was stirred at 100 ꢁC for 30 h under N2 atmo-
sphere, the magnetic solid was collected by employing a magnetic
field. The solid was repeatedly washed with dry 1,4-dioxane and
dried under vacuum for the catalysis tests. As for Fe3O4@mSiO2-
NHC(2) and Fe3O4@mSiO2-NHC(3), the amounts of Pd(acac)2
were decreased to 0.0053 and 0.0041 g without changing any
other conditions, respectively. The molar ratio of NHC/Pd was
kept the same.
Material synthesis
In order to integrate the NHC moiety with the framework of
shell, we introduced two hydrolysable triethoxysilyl groups onto
the terminals of the NHC ligand to obtain an NHC-bridged
organosilane
(1,3-bis(4-triethoxysilyl-2,6-diisopropylphenyl)-
imidazol-3-ium-trifluoro-methane sulfonate). The synthesis of
NHC-bridged triethoxysilane was reported in our recent pub-
lications.7c The fabrication of magnetic core–shell-structured
nanoporous organosilica microspheres consists of four steps
according to the modified method,4 as described in Scheme 1.
Firstly, the Fe3O4 microspheres were prepared according to the
reported method.4 In order to a grow uniform nanoporous shell
around the Fe3O4 core and to protect Fe3O4 core under harsh
conditions, the Fe3O4 microspheres with a rough surface were
first coated with a thin layer of SiO2. Next, a NHC-functionalized
nanoporous organosilica shell was created surrounding silica-
coated Fe3O4 cores through a sol–gel process in the presence of
CTAB as template. After removing CTAB infiltrated in the shell
nanopores, the magnetic core–shell nanoporous organosilica
microspheres with built-in NHC moieties were eventually
obtained. By gradually reducing the amount of the added sili-
ceous precursors in the third step, we expected to achieve three
samples with gradually decreasing shell thicknesses:
General procedure for the Suzuki–Miyaura coupling
A mixture of aryl chlorides (2.0 mmol), phenylboronic acid
(2.2 mmol), potassium tert-butoxide (3 mmol) and iso-propyl
alcohol (6 mL), and the solid catalyst (0.5 mol% with respect to
aryl chlorides) were stirred at 80 ꢁC under an N2 atmosphere for
a given time. At the end of reaction, the mixture was cooled to
room temperature and was repeatedly extracted with diethyl
ether. The combined organic layers were concentrated and the
resulting product was purified by a column chromatography on
silica gel. The product was confirmed by 1H NMR (see the ESI†).
Its purity was confirmed by GC.
The recycling test for the Suzuki–Miyaura coupling was con-
ducted as follows. At the end of the first run, the mixture was
cooled to room temperature and the liquid was isolated with aid
of an external field. The recovered catalyst was washed with
diethyl ether, water, methanol and acetone in sequence and then
dried under vacuum. The recovered catalyst was weighed again.
Fresh solvent and substrates were added, but the molar ratio of
substrate to Pd remained the same as that in the first run.
The procedures for the coupling reactions of benzyl chloride
and arylboronic acids were the same as those of aryl chlorides.
Characterization and analysis
Wide-angle powder X-ray powder diffraction was performed on
a Rigaku D/Max 3400 powder diffraction system. N2 physical
adsorption was carried out on a micrometritics. ASAP2020
volumetric adsorption analyzer (before measurements, samples
were out gassed at 125 ꢁC for 6 h). The Brunauer–Emmett–Teller
(BET) surface area was evaluated from data in the relative
Scheme 1 Schematic description for the synthesis of Fe3O4@mSiO2-
NHC(x), HNC–S
¼
1,3-bis(4-triethoxysilyl-2,6-diisopropylphenyl)-
imiadzol-3-ium-trifluoro-methane
organosilane).
sulfonate
(NHC-bridged
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 6639–6648 | 6641

View Article Online
Table 1 Physical parameters of the synthesized magnetic core–shell
organosilica microspheres
Fe3O4@mSiO2-NHC(1),
Fe3O4@mSiO2-NHC(3).
Fe3O4@mSiO2-NHC(2)
and
Samples
Sa (m2 gꢀ1
)
Pb (cm3 gꢀ1
)
Dc (nm)
Structural characterization
Fe3O4@mSiO2-NHC(1)
Fe3O4@mSiO2-NHC(2)
Fe3O4@mSiO2-NHC(3)
216
163
120
0.13
0.11
0.08
1.6
1.6
1.7
We employed N2 physical sorption, XRD and TEM to check the
structures of the synthesized samples. N2 adsorption/desorption
isotherms of these materials are shown in Fig. 1, and the physical
parameters measured with N2 sorption are summarized in Table
a
b
BET surface area. Single point pore volume calculated at a relative
c
pressure P/P0 of 0.99. Calculation from adsorption branch.
1. The sample Fe3O4@mSiO2-NHC(1) shows
a type-IV
isotherms, suggesting the presence of cylindrical nanopores. The
surface area, pore volume and pore size were determined to be
216 m2 gꢀ1, 0.13 cm3 gꢀ1 and 1.6 nm, respectively. The values
obtained for surface area and pore volume per unit mass of
Fe3O4@mSiO2-NHC(1) are large, especially considering that the
total mass of the sample is dominated by the contribution from
the nonporous Fe3O4 core. The surface areas of a typical meso-
porous material which has a fully porous structure are as large as
600–1000 m2 gꢀ1 and pore volume as high as 0.5–1.0 cm3 gꢀ1
depending on the synthetic conditions. The striking contrast is
suggestive of a fact that the obtained Fe3O4@mSiO2-NHC(1) has
a fractionally porous architecture. The determined pore size of
this sample is much smaller than that of typical MCM-41, which
is caused by the presence of a larger amount of ethanol in the
synthesis system and bulky NHC moieties present on the nano-
pore surface.12 Fe3O4@mSiO2-NHC(2) also shows type-IV
isotherms, but their surface area and pore volume have appar-
ently decreased. The same trend was observed for the
Fe3O4@mSiO2-NHC(3) sample. The decreases in these para-
meters reflect the decrease in the fraction of the porous shell in
the whole material, which is due to the reduction in the amount
of siliceous precursor added. The dependence of the physical
properties on the amount of added siliceous precursor confirms
that the structural aspects of these materials can be tunabed.
The XRD patterns of these three materials are shown in Fig. 2.
One diffraction peak at 2q ¼ 2.4ꢁ was clearly observed for
Fe3O4@mSiO2-NHC(1), confirming a somewhat ordered orga-
nization of the architecture. This diffraction peak can be
attributed to the (100) plane if the pore structure was assigned to
the MCM-41 family. The (110) and (200) diffraction peaks that
were usually observed in a standard MCM-41 sample, were not
found. This can be explained by the fact that the presence of the
bulky organosilane affects the interactions between silicon
oligomers and template molecules and the assembly ability is
thus discounted. The intensity of (100) diffraction peaks gradu-
ally decrease in the order of Fe3O4@mSiO2-NHC(1),
Fe3O4@mSiO2-NHC(2), and Fe3O4@mSiO2-NHC(3), probably
reflecting a gradual decrease in the shell thickness.
The TEM images of the samples with low and high magnifi-
cations are displayed in Fig. 3. TEM images clearly show that the
prepared Fe3O4 sample consists of monodispersed microspheres
with diameters of about 200 nm (Fig. 3a). The microsphere
surface is relatively rough, seen from more magnified TEM
images (Fig. 3b), which is in agreement with the results reported
in ref. 4. Interestingly, as for Fe3O4@mSiO2-NHC(1), uniform
microspheres with a good dispersion are clearly observed, and
the particle diameter undergoes an apparent increase (ca. 80–
100 nm) in comparison with the Fe3O4 microspheres. A gray
layer around the black Fe3O4 core appears (Fig. 3c) and its
surface become smoother than that of the Fe3O4 microspheres.
Most of the Fe3O4 cores are effectively encapsulated by an
organosilica layer. In the more magnified TEM images (Fig. 3d),
the pores of ca. 2 nm in size around the core can be observed. It is
worthwhile to note that most of channels in the shell are orien-
tated to core boundary and shell outer-surface in spite of some
distortions of the pore channels in comparison with a typical
MCM-41 structure. Such pores perpendicular to outer surface
Fig. 1 N2 adsorption/desorption isotherms of the synthesized samples:
(a) Fe3O4@mSiO2-NHC(1); (b) Fe3O4@mSiO2-NHC(2); and (c)
Fe3O4@mSiO2-NHC(3).
Fig. 2 XRD patterns of the synthesized samples: (a) Fe3O4@mSiO2-
NHC(1); (b) Fe3O4@mSiO2-NHC(2); and (c) Fe3O4@mSiO2-NHC(3).
6642 | J. Mater. Chem., 2012, 22, 6639–6648
This journal is ª The Royal Society of Chemistry 2012

View Article Online
decreases down to ca. 30–40 and 20–30 nm, respectively (Fig. 4c
and d). This trend coincides with the N2 sorption and XRD
results.
Compositional characterization
To clarify the compositions of the synthesized materials, we
employed FT-IR to characterize the synthesized samples (Fig. 5).
The prepared magnetic particles show a typical FT-IR spectrum
of Fe3O4. After coating with a thin layer of SiO2, two new peaks
at 1080 and 960 cmꢀ1 appear which are assigned to Si–O–Si and
Si–OH stretching vibrations, respectively. Notably, Fe3O4@
mSiO2-NHC(1) exhibits new peaks at 2976, 1542 and 1468 cmꢀ1
when compared with SiO2-caoted Fe3O4. These three peaks are
related to C–H stretching vibrations and vibrations of the NHC
precursor, suggesting that the NHC precursor has been
successfully integrated into the solid materials. Moreover, for
Fe3O4@mSiO2-NHC(2) the intensities of these three peaks
decrease. The further decrease in the peak intensity is observed
for the Fe3O4@mSiO2-NHC(3) sample. This trend reflects the
fact that the amount of siliceous material present (including
NHC) decreases, which is in agreement with the amount of
precursor used in the third step of the reaction.
Fig. 3 TEM images of the synthesized samples: (a) Fe3O4, the bar is
300 nm; (b) Fe3O4, magnified, the bar is 100 nm; (c) Fe3O4@mSiO2-
NHC(1), the bar 200 nm; (d) Fe3O4@mSiO2-NHC(1), magnified, the bar
is 20 nm.
are favorable to a fast mass transport of reactant and product
molecules.
The magnified TEM images of silica-coated Fe3O4 before and
after encapsulation by the nanoporous organosilica shell are
presented in Fig. 4. The layer thickness of silica-coated Fe3O4 is
not identified because the layer is very thin (Fig. 4a). The shell
thickness of most of the Fe3O4@mSiO2-NHC(1) particles is
estimated to be 40–50 nm from the TEM images (Fig. 4b).
Consistent with N2 sorption and XRD results, the TEM obser-
vations further confirm that Fe3O4@mSiO2-NHC(1) indeed
possesses a core–nanoporous shell structure. Moreover, as the
amount of the added siliceous precursor decreases, the shell
thickness decreases. The shell thickness for most of the
Fe3O4@mSiO2-NHC(2) and Fe3O4@mSiO2-NHC(3) particles
Magnetic properties
Fig. 6 shows the wide angle XRD patterns of Fe3O4 and
Fe3O4@mSiO2-NHC(1). These two samples show the same
diffraction peaks at 2q ¼ 30.1, 35.6, 43.1, 53.5, 57.1 and 62.7,
which correspond to the (220), (311), (400), (422), (511) and (440)
lattice planes of a typical Fe3O4 structure. Fe3O4@mSiO2-
NHC(1) is easily isolated magnetically by placing a magnet
beside the reaction vessel wall. At the same time, this material can
re-disperse into the reaction medium when the external magnet is
removed (a negligible magnetic hysteresis loop at room temper-
ature). As expected, Fe3O4@mSiO2-NHC(1) has
a good
magnetization that can be inferred from the magnetic field
dependence of the magnetization displayed in Fig. 7. The
magnetization saturation values of Fe3O4, Fe3O4@mSiO2-
NHC(1), Fe3O4@mSiO2-NHC(2) and Fe3O4@mSiO2-NHC(3)
were estimated to be 52.2, 21.9, 28.1, and 34.1 emu gꢀ1 from the
Fig. 4 TEM images of the synthesized samples: (a) SiO2-coated Fe3O4;
(b) Fe3O4@mSiO2-NHC(1); (c) Fe3O4@mSiO2-NHC(2); (d) Fe3O4@
mSiO2-NHC(3). The bar is 100 nm.
Fig. 5 FT-IR spectra of the synthesized samples (A) and regional
spectra (B).
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 6639–6648 | 6643

View Article Online
amount of Pd present on Fe3O4@mSiO2-NHC(1) was deter-
mined with ICP-AES to be 0.22 wt%, which was approximately
equal to the theoretic value. This reveals that the all the Pd(acac)2
in solution was loaded onto Fe3O4@mSiO2-NHC(1) under the
investigated conditions. The complete coordination confirms the
good ability of Fe3O4@mSiO2-NHC(1) in coordination with
metals. XPS spectroscopy was also employed to check the
surface elements and the valence states of Pd species on the
organosilica surface (Fig. 8). The XPS elemental survey scan of
the surface of the Pd-coordinated Fe3O4@mSiO2-NHC(1)
reveals that silicon, oxygen, carbon, nitrogen and palladium are
present in the material, as initially expected (Fig. 8a). However,
iron, solely belonging to the core, was also detected although its
intensity was very low. The XPS technique is used for surface
inspection and its detection depth is usually ca. 10 nm, which is
much less than the shell thickness. The presence of a weak signal
from iron probably indicates that some of the cores were not fully
Fig. 6 Wide angle XRD patterns of Fe3O4 and Fe3O4@mSiO2-NHC(1).
Fig. 7 Magnetization curves of the samples at 298 K: (a) Fe3O4; (b)
Fe3O4@mSiO2-NHC(3); (c) Fe3O4@mSiO2-NHC(2); and (d)
Fe3O4@mSiO2-NHC(1).
magnetization curves, respectively. Such magnetizations are
sufficient for performing magnetic separation. From the change
in magnetization saturation value, it can be deduced that the
weight fraction of the organosilica layer (including the NHC
units and nonporous silica layer) in Fe3O4@mSiO2-NHC(1),
Fe3O4@mSiO2-NHC(2) and Fe3O4@mSiO2-NHC(3) corre-
spond to 58%, 46% and 35%. The decrease in the weight fraction
is the consequence of reducing the total amount of siliceous
precursor.
Coordination capacity
To probe the coordination capacity of N-heterocyclic carbene
(NHC) in the solid framework of the shell, we chose Pd(acac)2 to
coordinate with the hybrid material because the formed
Pd(NHC)(acac)Cl is stable in air and catalytically active for the
Suzuki–Miyaura reaction.13 0.63 wt% Pd(acac)2 with respect to
Fe3O4@mSiO2-NHC(1) was used. The coordination was carried
out in a 1,4-dioxane solution under reflux conditions. The
Fig. 8 XPS spectra: (a) the elemental survey scan of Fe3O4@mSiO2-
NHC(1); (b) Pd XPS regional spectra of Fe3O4@mSiO2-NHC(1).
6644 | J. Mater. Chem., 2012, 22, 6639–6648
This journal is ª The Royal Society of Chemistry 2012

View Article Online
encapsulated by an organosilica shell. Fluorine and sulfur per-
taining to trifluoromethane sulfonate (CF3SO3ꢀ) groups of the
NHC-bridged organosilane were not found. It is likely that the
trifluoromethane sulfonate (CF3SO3ꢀ) groups of the NHC
precursor were exchanged by OHꢀ during the alkali-assisted sol–
gel process. In Fig. 8b, the regional XPS spectrum of
Fe3O4@mSiO2-NHC(1) exhibits two peaks centered at 335.2 and
treated with Pd(acac)2 using the same procedure as Fe3O4@
mSiO2-NHC(1). The Pd sites-related pure SiO2 and Fe3O4 are
both almost inactive in the Suzuki–Miyarura coupling of chlo-
robenzene. As the shell thickness decreases, the fraction of active
Pd site decreases and the fraction of inactive Pd sites increases,
thus leading to the decrease in the activity.
In order to further evaluate the performance of this catalyst,
we compared the activity of Fe3O4@mSiO2-NHC(1) with that of
NHC-functionalized MCM-41 microspheres that were synthe-
sized under the same conditions except without the addition of
the Fe3O4 cores (its TEM images and XRD pattern were
included in Fig. S1 and Fig. S2 in the ESI†). Under the same
conditions, NHC-functionalized MCM-41 microspheres gave
a conversion of 51%. It is much lower than Fe3O4@mSiO2-
NHC(1) although Fe3O4@mSiO2-NHC(1) included a fraction of
inactive Pd sites. The comparison suggests activity enhancement
effects caused by a core–shell structure. The enhanced activity
may be attributed to the significantly reduced diffusion pathway
because the catalytic reaction solely occurs in outer nanoporous
shell of Fe3O4@mSiO2-NHC(1), not throughout the whole
microspheres of NHC-functionalized MCM-41. Impressively,
under identical reactions conditions, commercial Pd/C (with a Pd
loading of 1 wt%) gave a conversion of only 16%, which was also
much lower than Fe3O4@mSiO2-NHC(1). These striking
contrasts further highlight the superiority of Fe3O4@mSiO2-
NHC(1). Several types of heterogeneous catalysts for Suzuki–
Miyaura couplings were reported such as Pd/C, mesoporous
silica-supported palladium, polymer-incarcerated palladium,
polyurea-encapsulated palladium and dendrimer-supported
palladium catalysts.14 These heterogeneous systems successfully
catalyzed the Suzuki–Miyaura couplings of aryl bromides but
failed in the couplings of aryl chlorides except in only limited
examples.10 In this context, our developed catalyst Fe3O4@
mSiO2-NHC(1) outperforms most of the reported heterogeneous
catalysts in the terms of activity and catalyst recovery. The high
activity is attributed to a favorable combination of the presence
of a molecularly functional moiety only in the shell, decreased
diffusion length as well as the radially aligned pore architecture.
To further evaluate the catalytic performances of the Pd-
341.1 eV, which are assigned to Pd(0) 3d5/2, and Pd(0) 3d3/2
,
respectively. Based on these results, we speculate that Pd(0)
species are formed on the surface of Fe3O4@mSiO2-NHC(1)
during the coordination process even though a Pd(II) precursor
was employed. These results are different from our previous
observation that Pd(II) complexes were yielded on a NHC-
functionalized material synthesized under the acidic conditions.
Catalytic performances
To investigate the catalytic activities of the synthesized materials,
we coordinated Fe3O4@mSiO2-NHC(2) and Fe3O4@mSiO2-
NHC(3) with
a given amount of Pd(acac)2 along with
Fe3O4@mSiO2-NHC(1) (the molar ratio of NHC to Pd was kept
at the same level). Their catalytic activities were evaluated with
the Suzuki–Miyarura coupling of chlorobenzene and phenyl-
boronic acid in iso-propyl alcohol at 80 ꢁC under a N2 atmo-
sphere. Potassium tert-butoxide was used as a base to promote
this reaction. The ratio of Pd to chlorobenzene was kept at
0.5 mol% for all the reactions. The reaction results are reflected in
Fig. 9. Fe3O4@mSiO2-NHC(1) gave a 63% conversion within 8 h,
whereas Fe3O4@mSiO2-NHC(2) and Fe3O4@mSiO2-NHC(3)
afforded 59% and 56% conversions, respectively. It seems
contrary to our initial expectation that catalytic activity should
increase with decreasing the shell thickness because of the
decreased diffusion depth. Considering the unique structure and
composition of Fe3O4@mSiO2-NHC(x), it is not surprising
because Pd(acac)2 can coordinate with the nonporous SiO2
surface and naked Fe3O4 surface as well as the NHC units. ICP-
AES analyses confirmed our assumption that SiO2-encapsulated
Fe3O4 could indeed also capture Pd(acac)2 form the solution.
0.083 wt% Pd was detected on SiO2-encapsulated Fe3O4 that was
coordinated
Fe3O4@mSiO2-NHC(1),
Suzuki–Miyaura
couplings of various arylboronic acids and chlorobenzene were
conducted using iso-propyl alcohol as the solvent at 80 ꢁC under
a N2 atmosphere (Table 2). At a Pd loading of 0.5 mol%, for
phenylboronic acids, the Pd-coordinated Fe3O4@mSiO2-
NHC(1) gave a 73% yield within 12 h (Table 2, entry 1). 73% and
72% yields were achieved in the cases of arylboronic acids
bearing electron-donating methyl and tert-butyl groups (Table 2,
entries 2 and 3). The moderately sterically demanding arylbor-
onic acids were also smoothly converted to the corresponding
products in 70–86% yields (Table 2, entries 4 and 5). For fluoro-
substituted arylboronic acid, 69% yield was achieved (Table 2,
entry 6). Aryl boronic acid bearing strong electron-withdrawing
groups needs a relatively longer time to obtain a moderate yield
(Table 2, entry 7).
Encouraged by the impressive results for the Suzuki–Miyaura
couplings of chlorobenzene and boronic acids, we wanted to test
the catalytic activity towards the couplings of benzylic chloride
and arylboronic acids because this coupling reaction provids an
important alternative to the Friedel–Crafts reaction for synthesis
Fig. 9 The chlorobenzene conversions over different solid catalysts.
Reaction conditions: aryl chlorides (1 mmol), phenylboronic acid
(1.1 mmol), potassium tert-butoxide (1.5 mmol) and iso-propyl alcohol
(3 mL), and the solid catalyst (0.5 mol% with respect to aryl chlorides),
80 ꢁC, 8 h, under a N2 atmosphere.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 6639–6648 | 6645

View Article Online
Table 2 The Suzuki–Miyaura couplings of various phenylboronic acids and chlorobenzene over the Pd-coordinated Fe3O4@mSiO2-NHC(1)
Entry
Substrates
Products
Time (h)
Yielda (%)
1
2
12
12
73
73
3
4
12
12
12
72
63
79
5
6
12
24
69
59
7
a
Isolated yields.
of diarylmethane-type aromatics.15 Using phenylboronic acid as
a substrate, the Pd-coordinated Fe3O4@mSiO2-NHC(1) gave
diphenylmethane in a yield up to 79% (Table 3, entry 1). In
followed reactions, in spite of electron-rich arylboronic acids
(Table 3, entries 2–7) and electron-poor arylboronic acid (Table
3, entry 8), various diarylmethane compounds in 71–89% yields
could be obtained.
the spatial restrictions imposed by nanopores and electrostatic
interactions with the NHC unit on the nanopore surface.7d
To check the reaction nature, namely heterogeneous catalysis
and homogeneous catalysis, we tested the activity of the filtrate
of the catalytic reaction. After the coupling reaction of chloro-
benzene with phenylboronic acid had proceeded for 6 h (with
conversion at 54%) the reaction was stopped and the filtrate was
immediately collected under the hot conditions. A further 5%
increase in conversion was observed after heating the filtrate at
The recyclability of the Pd-coordinated Fe3O4@mSiO2-
NHC(1) was investigated with the consecutive Suzuki–Miyaura
reactions of chlorobenzene with phenylboronic acid (Table
4).The fresh solid catalyst gave the product in a 73% yield within
12 h. After the first reaction cycle, the catalyst could be separated
with the reaction liquid by using an external magnetic field
(Fig. S3 in the ESI†). The magnetic separation was easy, fast and
clean, avoiding the need for centrifugation and filtration that
require extra energy. After the reaction liquid was removed,
diethyl ether and acetone in sequence were added to wash the
recovered solid catalyst. After washing and drying, the recovered
catalyst was directly used for the next reaction cycle. For the
second and third reaction cycles, yields greater than 69% were
achieved within the same number of hours as the first run. From
the fourth reaction cycle, the activity of the reused catalyst was
observed to slightly decrease. A 57% yield was obtained within 15
h. For the fifth reaction cycle, a reaction time of 20 h led to a 51%
yield and the prepared catalyst could still separated with
a magnet. The good recyclability may be attributable to the
NHC-functionalized nanopores that can efficiently prevent the
aggregation or agglomeration of Pd nanoparticles through
ꢁ
80 C for another 4 h (50% of the initial amount of potassium
tert-butoxide was added to the filtrate). These results may indi-
cate that the observed conversion was at least partially contrib-
uted to by leached Pd species in the filtrate. The Pd loss may be
attributed to two factors. One is related to the Pd leaching from
the solid catalyst during catalytic reaction. The other results from
the brushing off of the nanoporous organosilica layer from the
magnetic core (Si was detected in the filtrate by ICP-AES)
because of the persistent stirring during reaction, yielding fine
particles
Although the activity of the solid catalyst decreases during the
consecutive recycling reactions, it represents a highly recoverable
heterogeneous catalyst for the Suzuki–Miyaura coupling of less
reactive aryl chlorides, considering a relatively low Pd loading
(0.5 mol%). Additionally, this solid catalyst exhibited a high
stability in air. It was found that there was no decrease in activity
and magnetic isolation ability even after the catalyst was exposed
to air for several days. In this aspect, the developed catalyst is
also superior to immobilized phosphine catalysts.
6646 | J. Mater. Chem., 2012, 22, 6639–6648
This journal is ª The Royal Society of Chemistry 2012

View Article Online
Table 3 The coupling reactions of benzylic chloride and various arylboronic acids over the Pd-coordinated Fe3O4@mSiO2-NHC(1)
Entry
Arylboronic acids
Products
Time (h)
Yielda (%)
1
2
8
8
79
80
3
4
5
10
8
71
81
89
8
6
7
8
73
8
85
71
8
10
a
Isolated yield.
Table 4 Recyclability test of the Pd-coordinated Fe3O4@mSiO2-
NHC(1) for the Suzuki–Miyaura coupling reaction
Suzuki–Miyaura coupling of less active aryl chlorides, and its
activity was found to be dependent on the organosilica shell
thickness. Impressively, the resultant catalyst showed activity
enhancement effects in comparison with the analogous MCM-41
microspheres due to the significantly decreased diffusion depth.
Furthermore, the developed catalysts can be easily isolated by
using a magnetic field and directly used in the next reaction cycles
without a significant loss of its activity, suggesting promising
application potentials. This study, other than supplying a novel
synthesis of magnetic core–shell structured nanoporous orga-
nosilica microspheres with molecular functionalization, demon-
strates the catalytic activity enhancement effects, which is crucial
for designing high-performance solid catalysts.
Cycles
Time (h)
Yielda (%)
1
12
73
2
12
75
3
12
69
4
15
57
5
20
51
a
Isolated yield.
Conclusions
We report a successful synthesis of magnetic core–shell-struc-
tured nanoporous organosilica microspheres with built-in NHC
ligands by growing a layer of NHC-functionalized nanoporous
organosilica around a Fe3O4 core with the aid of a template.
These microspheres thus-obtained have nanopores (1.6 nm) in
the shell, moderate surface area, uniform morphology and high
magnetization. The growth of a nanoporous organosilica shell
with controlled thickness around Fe3O4 can be achieved by
varying the amount of siliceous precursor added. Such multi-
functional materials exhibit a good coordination ability toward
Pd. The Pd-coordinated material was highly active toward the
Acknowledgements
We acknowledge Specialized Research Fund for the Doctoral
Program of Higher Education (200801081035), Shanxi Natural
Science Foundation for Youths (2009021009) and the Natural
Science Foundation of China (20903064, 21173137).
References
€
1 (a) F. Caruso, R. A. Caruso and H. Mohwald, Science, 1998, 282,
1111; (b) L. M. Liz-Marzan, M. Giersig and P. Mulvaney,
Langmuir, 1996, 12, 4329; (c) Z. H. Xu, Y. L. Hou and S. H. Sun,
ꢀ
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 6639–6648 | 6647

View Article Online
J. Am. Chem. Soc., 2007, 129, 8698; (d) Y. H. Deng, W. L. Yang,
C. C. Wang and S. K. Fu, Adv. Mater., 2003, 15, 1729; (e) C. Graf,
D. L. J. Vossen, A. Imhof and A. van Blaaderen, Langmuir, 2003,
19, 6693; (f) L. Zhang, S. Z. Qiao, Y. G. Jin, H. G. Yang,
S. Budihartono, F. Stahr, Z. F. Yan, X. L. Wang, Z. P. Hao and
11, 3302; (d) Y. Wan, D. Q. Zhang, Y. P. Zhai, C. M. Feng,
J. Chen and H. X. Li, Chem.–Asian J., 2007, 2, 875.
ꢀ
7 (a) A. Corma, D. Das, H. Garcıa and A. Leyva, J. Catal., 2005, 229,
322; (b) P. Y. Wang, X. Liu, J. Yang, Y. Yang, L. Zhang, Q. H. Yang
and C. Li, J. Mater. Chem., 2009, 19, 8009; (c) H. Q. Yang, G. Li,
Z. C. Ma, J. B. Chao and Z. Q. Guo, J. Catal., 2010, 276, 123; (d)
H. Q. Yang, X. J. Han, G. Li, Z. C. Ma and Y. J. Hao, J. Phys.
Chem. C, 2010, 114, 22221; (e) H. Q. Yang, Y. W. Wang, Y. Qin,
Y. Z. Chong and Q. Z. Yang, Green Chem., 2011, 13, 1352; (f)
W. J. Hunks and G. A. Ozin, J. Mater. Chem., 2005, 15, 3716; (g)
F. Hoffmann, M. Cornelius, J. Morell and M. Frc¸ba, Angew.
Chem., Int. Ed., 2006, 45, 3216; (h) S. Fujita and S. Inagaki, Chem.
Mater., 2008, 20, 891; (i) P. Nguyen, P. Hesemann, P. Gaveau and
J. J. E. Moreau, J. Mater. Chem., 2009, 19, 4164.
8 (a) C. Li, Catal. Rev. Sci. Eng., 2004, 46, 419; (b) H. Q. Yang, J. Li,
J. Yang, Z. M. Liu, Q. H. Yang and C. Li, Chem. Commun., 2007, 1086.
9 (a) N. Miyaura, T. Yanagi and A. Suzuki, Synth. Commun., 1981, 11,
513; (b) J. S. Carey, D. Laffan, C. Thomson and M. T. Williamsd,
Org. Biomol. Chem., 2006, 4, 2337; (c) M. Lamblin, L. Nassar-
Hardy, J. C. Hierso, E. Fouquet and F. X. Felpin, Adv. Synth.
Catal., 2010, 352, 33.
€
G. Q. Lu, Adv. Funct. Mater., 2008, 18, 3203; (g) W. Schaartl,
Nanoscale, 2010, 2, 829; (h) C. J. Zhong and M. M. Maye, Adv.
Mater., 2001, 13, 1507; (i) D. C. Niu, Z. Ma, Y. S. Li and J. L. Shi,
J. Am. Chem. Soc., 2010, 132, 15144.
2 (a) Z. H. Xu, C. X. Li, X. J. Kang, D. M. Yang, P. P. Yang, Z. Y. Hou
and J. Lin, J. Phys. Chem. C, 2010, 114, 16343; (b) E. Katz and
I. Willner, Angew. Chem., Int. Ed., 2004, 43, 6042; (c) Y. H. Deng,
D. W. Qi, C. H. Deng, X. M. Zhang and D. Y. Zhao, J. Am.
Chem. Soc., 2008, 130, 28; (d) L. Zhang, S. Z. Qiao, Y. G. Jin,
Z. G. Chen, H. C. Gu and G. Q. Lu, Adv. Mater., 2008, 20, 805; (e)
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., Int.
Ed., 2009, 48, 5875; (f) E. Katz and I. Willner, Angew. Chem., Int.
Ed., 2004, 43, 6042; (g) C. Q. Yang, G. Wang, Z. Y. Lu, J. Sun,
J. Q. Zhuang and W. S. Yang, J. Mater. Chem., 2005, 15, 4252; (h)
Y. Yang, J. Liu, X. B. Li, X. Liu and Q. H. Yang, Chem. Mater.,
2011, 23, 3676; (i) J. Liu, S. Z. Qiao, Q. H. Hu and G. Q. Lu,
Small, 2011, 7, 425; (j) J. Liu, B. Wang, S. B. Hartono, T. Liu,
P. Kantharidis, A. P. J. Middelberg, G. Q. Lu, L. He and
S. Z. Qiao, Biomaterials, 2012, 33, 970.
3 W. R. Zhao, J. L. Gu, L. X. Zhang, H. R. Chen and J. L. Shi, J. Am.
Chem. Soc., 2005, 127, 8916.
4 Y. H. Deng, Y. Cai, Z. K. Sun, J. Liu, C. Liu, J. Wei, W. Li, C. Liu,
Y. Wang and D. Y. Zhao, J. Am. Chem. Soc., 2010, 132, 8466.
5 J. P. Ge, Q. Zhang, T. R. Zhang and Y. D. Yin, Angew. Chem., Int.
Ed., 2008, 47, 8924.
6 (a) S. Inagaki, S. Guan, Y. Fukushima, T. Ohsuna and O. Terasaki, J.
Am. Chem. Soc., 1999, 121, 9611; (b) T. Asefa, M. J. MacLachlan,
N. Coombs and G. A. Ozin, Nature, 1999, 402, 867; (c) B. J. Melde,
B. T. Holland, C. F. Blanford and A. Stein, Chem. Mater., 1999,
10 (a) M. J. Jin and D. H. Lee, Angew. Chem., Int. Ed., 2010, 49, 1119; (b)
D. H. Lee, J. Y. Jung and M. J. Jin, Green Chem., 2010, 12, 2024.
11 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., Int.
Ed., 2009, 48, 5875.
12 D. C. Niu, Z. Ma, Y. S. Li and J. L. Shi, J. Am. Chem. Soc., 2010, 132,
15144.
ꢀ
13 N. Marion, P. D. Fremont, I. M. Puijk and E. C. Ecarnot, Adv. Synth.
Catal., 2007, 349, 2380.
14 D. H. Lee, J. H. Kim, B. H. Jun, H. Kang, J. Park and Y. S. Lee, Org.
Lett., 2008, 10, 1609 and references therein.
15 (a) M. J. Burns, I. J. S. Fairlamb, A. R. Kapdi, P. Sehnal and
R. J. K. Taylor, Org. Lett., 2007, 9, 5397; (b) B. P. Bandgar,
S. V. Bettigeri and J. Phopase, Tetrahedron Lett., 2004, 45, 6959.
6648 | J. Mater. Chem., 2012, 22, 6639–6648
This journal is ª The Royal Society of Chemistry 2012