Highly active, water-compatible and easily separable magnetic mesoporous Lewis acid catalyst for the Mukaiyama-Aldol reaction in water

Article abstract of DOI:10.1039/c4gc00178h

A novel magnetic mesoporous Lewis acid catalyst was prepared through immobilizing Yb(OTf)₃ on a sodium propylsulphonate and phenyl group co-functionalized magnetic core-mesoporous silica shell composite. The obtained Yb(OTf)₂-SO₃Na&Ph-MCMSS catalyst had a typical core-shell structure with a Fe₃O₄ magnetic core, a middle amorphous silica layer and a multifunctional mesoporous silica shell with radial pore channels. In water medium Mukaiyama-Aldol reactions, it exhibited a higher catalytic reactivity than that of the homogeneous catalyst Yb(OTf)₃, and control catalysts Yb(OTf)₂-SO₃Na-MCMSS without phenyl groups inside the mesoporous channels, Yb(OTf)₂-SO ₃Na&Ph-MCSS without a mesoporous structure, mesoporous Yb(OTf)₂-SO₃Na&Ph-MCM-41 with an irregular morphology and nonporous Yb(OTf)₂-SO₃Na-Amberlyst-15 ion-exchange resin. The systematic analysis demonstrated that this excellent catalytic performance could be attributed to the synergetic effect resulting from its radial mesoporous channels and the pore surface hydrophobicity, leading to the increased accessibility of actives sites and the decreased diffusion limitation of reactants. More importantly, this catalyst was stable in water and could be easily separated with a simple magnet and reused at least six times without loss of catalytic activity. This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c4gc00178h

Green Chemistry
View Article Online
View Journal
PAPER
Highly active, water-compatible and easily
separable magnetic mesoporous Lewis acid
catalyst for the Mukaiyama–Aldol reaction
in water†
Cite this: DOI: 10.1039/c4gc00178h
Fang Zhang,* Xiaotao Wu, Chao Liang, Xiaoyan Li, Zhen Wang and Hexing Li*
A novel magnetic mesoporous Lewis acid catalyst was prepared through immobilizing Yb(OTf)3 on a
sodium propylsulphonate and phenyl group co-functionalized magnetic core–mesoporous silica shell
composite. The obtained Yb(OTf)2-SO3Na&Ph-MCMSS catalyst had a typical core–shell structure with a
Fe3O4 magnetic core, a middle amorphous silica layer and a multifunctional mesoporous silica shell with
radial pore channels. In water medium Mukaiyama–Aldol reactions, it exhibited a higher catalytic reactivity
than that of the homogeneous catalyst Yb(OTf)3, and control catalysts Yb(OTf)2-SO3Na-MCMSS without
phenyl groups inside the mesoporous channels, Yb(OTf)2-SO3Na&Ph-MCSS without a mesoporous struc-
ture, mesoporous Yb(OTf)2-SO3Na&Ph-MCM-41 with an irregular morphology and nonporous Yb(OTf)2-
SO3Na-Amberlyst-15 ion-exchange resin. The systematic analysis demonstrated that this excellent cataly-
tic performance could be attributed to the synergetic effect resulting from its radial mesoporous channels
and the pore surface hydrophobicity, leading to the increased accessibility of actives sites and the
decreased diffusion limitation of reactants. More importantly, this catalyst was stable in water and could
be easily separated with a simple magnet and reused at least six times without loss of catalytic activity.
Received 3rd February 2014,
Accepted 21st March 2014
DOI: 10.1039/c4gc00178h
www.rsc.org/greenchem
geneous acid catalysts. Although extensive efforts have been
devoted to the fabrication of solid acid catalysts, they usually
1. Introduction
Homogeneous acid catalysts are mostly common used in the display inferior catalytic efficiencies in water-medium organic
large-scale production of industrial chemicals and fine chemi- syntheses compared to their homogeneous compartments.
cals.1 However, their inherent drawbacks such as the corro- The intrinsic problems in the heterogeneous catalytic
sivity, the environmental hazards as well as the high cost of systems are the decreased accessibility of actives sites and the
separating the acids from the reaction system, make it very enhanced diffusion limitation of reactants, leading to the
difficult to meet the requirements of environmentally friendly declined catalytic reactivity and selectivity.5 Recently, homo-
chemical processes.2 Moreover, some of them generally must geneous acids immobilized on a solid matrix with a high
be used under anhydrous conditions, which necessitates surface area or nanoscale size represented a viable solution to
massive toxic, flammable organic solvents and specialized this problem because these supports could guarantee the high
reaction vessels.3 On the contrary, enzymes have an unparal- dispersion of active sites and reduce the mass transfer resis-
leled ability to accomplish many biological processes in tance.6 However, the consequent challenge is that conventional
aqueous medium, which has inspired significant research separation methods such as filtration or centrifugation may
activities in the development of solid acid catalysts for chemi- become inefficient for these nanocatalysts.7 To overcome this
cal synthesis in water.4 The remarkable advantages of this drawback, magnetically supported acid catalysts have received
catalytic system are that it could simultaneously reduce the increasing attention since they can be easily removed from the
pollution and cost resulting from liquid acids and organic sol- reaction mixture by external magnetic separation.8 Neverthe-
vents, coupled with the ease of recovery and recycling of homo- less, the low surface area and rapid aggregation of the typical
iron oxide support make it difficult to maximize catalyst dis-
persion and activity.9 Meanwhile, the rare functionality of
the iron oxide surface limits the optimization of the surface
The Education Ministry Key Lab of Resource Chemistry and Shanghai Key Laboratory
of Rare Earth Functional Materials, Shanghai Normal University, Shanghai 200234,
chemistry, such as the hydrophobicity property, resulting in
P. R. China. E-mail: zhangfang@shnu.edu.cn, HeXing-Li@shnu.edu.cn
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
unfavorable chemical microenvironments for water-medium
organic reactions.10 Therefore, one strategy for the fabrication
c4gc00178h
This journal is © The Royal Society of Chemistry 2014
Green Chem.

View Article Online
Paper
Green Chemistry
of highly active, water-compatible and easily separable solid The solution was adjusted to pH 10 by 28 wt% aqueous
acid catalysts for chemical synthesis in water, is to form multi- ammonia solution. After that, 0.25 ml tetraethyl orthosilicate
functional nanocomposites that combine the advantages of (TEOS) was introduced into the solution at 25 °C under mech-
both a nanoporous support and magnetic material, which anical stirring for 6 h. The products were collected and washed
favor a high surface area, tunable chemical composition and with ethanol and water, followed by drying at 60 °C for 12 h.
convenient separation.
2.1.2. Preparation of the multifunctional magnetic meso-
Magnetic mesoporous materials have attracted enormous porous silica support (SO3Na&Ph-MCMSS). 0.20 g SiO2@Fe3O4
interest for the development of novel heterogeneous catalysts particles were dispersed in an aqueous solution containing
owing to their outstanding features, such as unique magnetic 0.64 g CTAB, 100 ml deionized water, 2.0 ml ammonia solution
response, large surface area, tunable pore structure and (28 wt%), and 140 ml ethanol. This mixture was ultrasonicated
surface functionality.11 In this context, magnetic core–meso- for 30 min and then mechanically stirred for another 30 min.
porous silica shell (MCMSS) composites that are comprised of Then, 0.625 ml TEOS was added dropwise to the above solu-
a superparamagnetic Fe3O4 core and an outer shell of ordered tion. After pre-hydrolysis for 30 min, 0.116 ml mercaptopropyl-
mesoporous silica with radial pore channels have great poten- trimethoxysilane (MPTS) and 0.117 ml phenyltrimethoxysilane
tial. Owing to their distinctive advantages of both short pore (PhTS) were introduced. After stirring for 12 h at 25 °C, the
channels and highly open pore structure, several different products were collected with a magnet and washed repeatedly
MCMSS supported catalysts containing metal nanoparticles or with ethanol and deionized water. Finally, the CTAB template
metal oxides have shown remarkable enhancements of cataly- was removed by repeated ethanol extraction under reflux con-
tic activity.12 However, to the best of our knowledge, Lewis ditions. Next, 0.20 g of the solid sample was suspended in
acids bonded to the mesoporous channel of a MCMSS support 30 ml H2O2 solution and stirred for 24 h. The obtained solid
for water-medium organic synthesis has not been reported so was washed and then added to 30 ml saturated NaCl solution
far. Additionally, insight into the adsorption and diffusion for ion-exchange treatment, resulting in a sodium propylsul-
behaviors in the confined microenvironment of the MCMSS phonate and phenyl group co-functionalized magnetic core–
supported catalysts has been rarely discussed. Herein, we mesoporous silica shell composite.
report for the first synthesis of a novel magnetic mesoporous
2.1.3. Fabrication of the magnetic mesoporous Lewis acid
Lewis acid catalyst (Yb(OTf)2-SO3Na&Ph-MCMSS) through catalyst (Yb(OTf)2-SO3Na&Ph-MCMSS). The SO3Na&Ph-MCMSS
immobilizing Yb(OTf)3 on a sodium propylsulphonate and support was introduced into a 35 ml ethanol solution contain-
phenyl group co-functionalized magnetic core–mesoporous ing 0.50 g Yb(OTf)3 under mechanical stirring for 3 h at 80 °C.
silica shell composite. It should be noted that the catalyst The black powder product was filtered and washed thoroughly
exhibited a higher activity compared to the homogeneous cata- with absolute ethanol to eliminate any un-coordinated
lyst Yb(OTf)3 and control catalysts (including Yb(OTf)2-SO3Na- Yb(OTf)3 compound, followed by vacuum drying at 60 °C for
MCMSS without phenyl groups inside the mesoporous 24 h.
channels, Yb(OTf)2-SO3Na&Ph-MCSS without a mesoporous
structure, Yb(OTf)2-SO3Na&Ph-MCSS mesoporous catalyst with
2.2. Characterization
an irregular morphology and a nonporous Yb(OTf)2-SO3Na- The ytterbium loading was measured by an inductively
Amberlyst-15 ion-exchange resin), in the water-medium coupled plasma optical emission spectrometer (Varian VIS-
Mukaiyama–Aldol reaction. This excellent catalytic perform- TA-MPX). The sulfur content was determined by elemental
ance could be attributed to the synergic effect resulting from analysis on an Element Vario EL III analyzer. Fourier trans-
the radially short mesoporous channels and the surface hydro- form infrared (FTIR) spectra were obtained using a Thermo
phobicity, which efficiently enriched the reactants and mean- Nicolet Magna 550 spectrometer. The X-ray powder diffraction
while decreased the mass transfer limitation. Moreover, it (XRD) data was acquired on a Rigaku D/maxr B diffractometer
could be easily separated using an external magnet and reused using Cu Kα radiation. N2 adsorption–desorption isotherms
up to 6 times without loss of catalytic activity.
were analyzed by a Quantachrome NOVA 4000e analyzer.
Specific surface areas (SBET) and average pore diameters (DP)
are calculated using BET and BJH models, respectively. The
morphologies were observed by transmission electron
microscopy (TEM, JEM-2011). The surface electronic states
were analyzed by X-ray photoelectron spectroscopy (XPS,
2. Experimental section
2.1. Sample preparation
2.1.1. Synthesis of amorphous silica coated Fe3O4 micro- Perkin-Elmer PHI 5000C ESCA). All the binding energy values
spheres (SiO2@Fe3O4). Fe3O4 microspheres were prepared were calibrated by using C1S = 284.6 eV as a reference. The
according to the method previously described by Zhao et al.12a magnetic property was measured by
a superconducting
In a typical run, 0.12 g Fe3O4 was added to 100 ml HCl quantum interference device magnetometer (SQUID). The
aqueous solution (0.10 M) and then the mixture was treated by hysteresis loops of these samples were recorded at temperatures
ultrasonication for 30 min and washed by ethanol and water. of 300 K. Water and toluene vapour absorption measurements
The solid was filtrated and subsequently dispersed in a solu- were carried out on an intelligent gravimetric analyse (Hiden
tion containing 160 ml ethanol and 30 ml deionized water. Isochema IGA-002/3) by introducing a dosed amount of high-
Green Chem.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
purity vapor directly into the sample chamber and recording Scheme 1. We started by depositing a thin silica layer on the
the weight change after a stable equilibrium pressure was prefabricated Fe3O4 microspheres through a sol–gel coating
reached.
strategy and then employed a surfactant-directed assembly
approach to further generate the thiol- and phenyl-functiona-
lized mesostructured silica/CTAB composite on the
SiO2@Fe3O4 surface by using cetyltrimethyl ammonium
bromide (CTAB) as a pore template, tetraethyl orthosilicate
[TEOS, Si(OEt)4], mercaptopropyltrimethoxysilane [MPTS, SH-
(CH2)3Si(OMe)3] and phenyltrimethoxysilane [PTS, PhSi-
(OMe)3] as functional precursors. After removing CTAB by
ethanol extraction, this bifunctionalized magnetic core–meso-
porous silica shell composite was treated by H2O2 oxidation
and the subsequent NaCl ion-exchange processes, resulting in
the SO3Na and Ph groups co-functionalized support
(SO3Na&Ph-MCMSS). Then, the Yb(III) Lewis acids can be intro-
duced into the mesoporous channels of the SO3Na&Ph-
MCMSS support by exposure to the Yb(OTf)3 solution to form
the final catalyst, denoted as Yb(OTf)2-SO3Na&Ph-MCMSS. For
comparison, Yb(III) Lewis acids immobilized on the SO3Na-
functionalized magnetic core–mesoporous silica shell support
without hydrophobic phenyl groups inside the mesoporous
channels, named as Yb(OTf)2-SO3Na-MCMSS, was also syn-
thesized using a similar protocol. Also, Yb(III) Lewis acids
grafted on the SO3Na- and Ph-functionalized magnetic core–
silica shell support without a mesoporous structure, denoted
as Yb(OTf)2-SO3Na&Ph-MCSS, were prepared through directly
coordinating Yb(OTf)3 with SO3Na and Ph groups terminally
bonded to the SiO2@Fe3O4 microspheres (Scheme S1†). Mean-
while, two non-magnetic control catalysts were fabricated
including Yb(OTf)2-SO3Na&Ph-MCM-41, which was prepared
by immobilizing Yb(OTf)3 in the SO3Na- and Ph-functionalized
traditional mesoporous support MCM-41, and the Yb(OTf)2-
SO3Na-Amberlyst-15 catalyst was synthesized by immobilizing
Yb(OTf)3 on the SO3Na-functionalized commercial ion-
exchange resin Amberlyst-15.
2.3. Activity test
In a typical run of the water-medium Mukaiyama–Aldol reac-
tion, a catalyst containing 0.050 mmol Yb(III), 0.5 mmol benz-
aldehyde, 1.0 mmol trimethyl-(1-phenylprop-1-enyloxy)silane
and 3.0 ml distilled water were mixed and allowed to react at
20 °C for 16 h under mild stirring. The products were extracted
with ethyl acetate, followed by analysis on a high performance
liquid chromatography analyzer (HPLC, Agilent 6410 series
Triple Quad) equipped with an Agilent C18 column. The reac-
tion conversion was calculated based on benzaldehyde since
trimethyl(1-phenylprop-1-enyloxy)silane was in excess.
In order to determine the catalyst recyclability, the catalyst
was allowed to settle down by a magnet after each run of reac-
tion and the clear supernatant liquid was decanted slowly. The
residual solid catalyst was reused with a fresh charge of water
and reactant for subsequent cycles under the same reaction
conditions. In all the tests, the reproducibility was checked by
repeating each result at least three times and was found to be
within 5%. Meanwhile, the liquid phase of the reaction
mixture was collected for ICP analysis after each reaction to
test for Yb leaching.
2.4. Adsorption test
To determine the adsorption performances of the magnetic
catalysts, 50 mg catalyst was firstly soaked in 100 ml water and
oscillated at 25 °C for 12 h. Then, 100 ml aqueous solution
including 10 ppm benzaldehyde was added into the above
solution. The solution was sampled at given time intervals and
the concentration of the remaining benzaldehyde in the
solution was determined on a HPLC analyzer. The adsorption
capacity was measured after reaching the saturation
adsorption.
We firstly optimized the sulfur content of the functiona-
lized support through variation of the molar ratio between
TEOS and MPTS silanes and obtained a 1.39 mmol g−1 sulfur
content in the SO3Na&Ph-MCMSS support (Table S1†). Next,
we investigated the immobilization conditions using different
solvents and the molar ratio between the Yb(OTf)3 catalyst
and the SO3Na-ligands, resulting in an optimal catalyst with
3. Results and discussion
The synthetic protocol used to achieve the magnetic meso-
porous Yb(OTf)2-SO3Na&Ph-MCMSS catalyst is presented in
Scheme 1 Schematic illustration of the synthesis of the magnetic mesoporous Yb(OTf)2-SO3Na&Ph-MCMSS catalyst.
This journal is © The Royal Society of Chemistry 2014
Green Chem.

View Article Online
Paper
Green Chemistry
ascribed to the C–F stretching absorption of the CF3 groups. In
addition, the sulfate stretching vibration absorption at
1250 cm−1 could also be found.16 These results clearly demon-
strated the successful incorporation of the Yb(OTf)3 complex
in the multifunctional magnetic mesoporous silica support.
The chemical state of the Yb(OTf)2-SO3Na&Ph-MCMSS cata-
lyst was investigated by its XPS spectrum. As shown in Fig. 2,
the results revealed that the 4d5/2 and 4d3/2 electronic states of
all the Yb species in the Yb(OTf)2-SO3Na&Ph-MCMSS catalyst
were observed at 187.6 eV and 202.6 eV, suggesting that the Yb
element was present in a trivalent oxidation state.17 In com-
parison with the Yb(OTf)3 sample, the binding energy of the
ytterbium species in the Yb(OTf)2-SO3Na&Ph-MCMSS catalyst
was shifted negatively by 0.70 eV. This shift could be ascribed
to the stronger electron-donating property of the sulfur
element in the propylsulphonate groups than that in the tri-
flate ligands. This explanation could be further confirmed by
XPS analysis of the sulfur element. The binding energy of the
sulfur species in the Yb(OTf)2-SO3Na&Ph-MCMSS catalyst
shifted positively by 0.40 eV compared to the Yb(OTf)3 sample.
Fig. 1 FTIR spectra of the Fe3O4, SiO2@ Fe3O4 and Yb(OTf)2-
SO3Na&Ph-MCMSS samples.
0.265 mmol g−1 Yb(III) loading (Table S2†). The Yb(OTf)2- Thus, the presence of the electron transfer between Yb(OTf)3
SO3Na&Ph-MCMSS catalyst was characterized before and after and functionalized support correlated with the observation of
each functionalization step. The FT-IR spectra (Fig. 1) revealed the FT-IR spectrum, confirming that the Yb(III) Lewis acid was
that the as-made Fe3O4 microspheres displayed a vibration immobilized by replacing one triflate ligand in Yb(OTf)3 with a
band at 580 cm−1, which is indicative of the typical Fe–O SO3Na ligand in the mesoporous channels of the SO3Na&Ph-
bond.13 The SiO2@Fe3O4 sample showed new broad peaks at MCMSS support.
1084 and 800 cm−1, which were attributed to the asymmetric
We then investigated the morphological properties of the
stretching bonds of Si–O–Si,14 suggesting the successful Yb(OTf)2-SO3Na&Ph-MCMSS catalyst. The TEM image (Fig. 3a)
coating of a thin silica layer on the Fe3O4 surface. In compari- indicated that it had a regular spherical shape with around
son with the SiO2@Fe3O4 sample, the Yb(OTf)2-SO3Na&Ph- 500 nm size. It further revealed that a typical core–shell struc-
MCMSS sample exhibited several addition absorption peaks. ture of the Yb(OTf)2-SO3Na&Ph-MCMSS catalyst was formed
Two additional absorbance peaks around 650 and 1461 cm−1 with the magnetic core (Fe3O4, dark) surrounded by a silica
were indicative of the δC–H and νC–C vibrations of the phenyl shell (SiO2, bright). Moreover, the silica shell consisted of a
groups while two additional absorbance bands around 2925 middle thin silica shell with a thickness of ca. 15 nm and a top
and 2855 cm−1 were attributed to the asymmetric and sym- mesoporous silica shell with a thickness of about 70 nm.
metric stretching modes of the C–H bond in the propyl Notably, the HRTEM image of the Yb(OTf)2-SO3Na&Ph-MCMSS
groups, revealing the co-existence of phenyl and SO3Na moi- catalyst (Fig. 3b) demonstrated that wormhole-like mesopores
eties.15 Meanwhile, the new peak around 1172 cm−1 was with diameters of ca. 2–3 nm were observed and the pore
Fig. 2 XPS spectra of the Yb(OTf)2-SO3Na&Ph-MCMSS catalyst ((a) Yb 4d; (b) S 2p).
Green Chem.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
The Mukaiyama–Aldol reaction is an important carbon–
carbon bond formation process that has the ability to obtain
β-hydroxycarbonyl compounds in a regio-, stereo-, and enantio-
selective manner.20 To investigate the utility of our novel
Yb(OTf)2-SO3Na&Ph-MCMSS catalyst, the water-medium
Mukaiyama–Aldol reaction of benzaldehyde and trimethyl-
(1-phenylprop-1-enyloxy)silane was firstly carried out under the
procedure described in the Experimental section. Firstly, the
control experiments indicated that the blank experiments with
SiO2@Fe3O4 and SH&Ph-MCMSS samples could not give any
products in the water-medium Mukaiyama–Aldol reaction,
indicating that the Yb3+ Lewis acid was the active species
(Table S3†). Furthermore, the preliminary experiments demon-
strated that the optimum reaction temperature was 20 °C
Fig. 3 TEM pictures of the Yb(OTf)2-SO3Na&Ph-MCMSS catalyst.
channels were radially aligned in the top mesoporous silica (Fig. S1†). By adjusting the catalyst amount in the reaction
shell, which result from the unique co-assembly of CTAB/silica system, the total Yb(III) amount in each reaction was fixed with
on the surface of the SiO2@Fe3O4 microsphere.12 The meso- a 0.05 mmol loading (Fig. S2†). Under this optimal reaction
structure of the Yb(OTf)2-SO3Na&Ph-MCMSS catalyst was condition, the Yb(OTf)2-SO3Na&Ph-MCMSS catalyzed reaction
further analyzed using low-angle XRD analysis. It exhibited a proceeded rapidly and cleanly to the coupling product with
well-resolved diffraction peak at around 2.2°, indicative of the 97.2% yield after 16 h (Fig. 5). Interestingly, the homogeneous
(100) reflection and two other peaks at 3.8° and 4.4°, corres- Yb(OTf)3 catalyst displayed a significantly decreased catalytic
ponding to (110) and (200) reflections, suggesting that the hexa- efficiency, with only a 72.4% yield under the same conditions.
gonal arrayed pore structure (p6mm) could be well preserved The different catalytic reactivities of the Yb(OTf)2-SO3Na&Ph-
after anchoring the Yb Lewis acid catalyst (Fig. 4a).18 The N2 MCMSS and Yb(OTf)3 catalysts could be interpreted based on
sorption curve of the Yb(OTf)2-SO3Na&Ph-MCMSS sample the reaction mechanism. The carbonyl groups in benzaldehyde
(Fig. 4b) exhibited a type IV isotherm with the capillary con- firstly interact with the Yb(III) Lewis acid through σ bonding
densation step in the low pressure region and an almost con- coordination and trimethyl(1-phenylprop-1-enyloxy)silane
stant adsorption in the high pressure region, confirming the attacks the activated benzaldehyde to generate the final aldol
existence of a uniform mesoporous structure.19 The narrow product. Owing to the low solubility of the two reactants in
and sharp pore size distribution profile derived from the water, the diffusion and adsorption of the reactants plays a
adsorption branch (inset) further suggested that the meso- critical role in determining the yield. Therefore, the excellent
pores were uniform. In addition, the calculated BET surface catalytic reactivity of the Yb(OTf)2-SO3Na&Ph-MCMSS catalyst
area and total pore volume were 248 m2 g−1 and 0.14 cm3 g−1
,
could be attributed to the strong surface hydrophobicity
indicating a high degree of porosity. Compared to the derived from the location of the phenyl groups in the meso-
SO3Na&Ph-MCMSS support (Table 1), the Yb(OTf)2-SO3Na&Ph- porous channels. Accordingly, we used the control Yb(OTf)2-
MCMSS sample displayed slightly decreased values of SBET, VP SO3Na-MCMSS catalyst without phenyl groups inside the
and DP due to the presence of the bulky Lewis acid moieties mesoporous channels to confirm the merit of the hydrophobic
inside the mesoporous channels.
surface.21 As expected, it exhibited a remarkably declined cata-
Fig. 4 Low-angle XRD pattern (a), N2 sorption curve (b) and pore size distribution (inset) of the Yb(OTf)2-SO3Na&Ph-MCMSS catalyst.
This journal is © The Royal Society of Chemistry 2014
Green Chem.

View Article Online
Paper
Green Chemistry
Table 1 Chemical composition and physical parameters of the different Yb catalysts
Sample
S content (mmol g−1
)
Yb loading (mmol g−1
)
SBET (m2 g−1
)
Vp (cm3 g−1
)
Dp (nm)
SO3Na&Ph-MCMSS
1.39
1.39
1.41
0.20
1.80
0.17
—
289
248
265
36
515
40
0.21
0.14
0.17
0.02
0.55
—
2.5
2.2
2.3
1.2
2.5
—
Yb(OTf)2-SO3Na&Ph-MCMSS
Yb(OTf)2-SO3Na-MCMSS
Yb(OTf)2-SO3Na&Ph-MCSS
Yb(OTf)2-SO3Na&Ph-MCM-41
Yb(OTf)2-SO3Na-Amberlyst-15
0.265
0.270
0.0370
0.385
0.0320
Fig. 7 TEM images of the Yb(OTf)2-SO3Na&Ph-MCSS catalyst.
toluene than that of the Yb(OTf)2-SO3Na-MCMSS sample, indi-
cating that it was more hydrophobic. Meanwhile, the adsorp-
tion capacity of Yb(OTf)2-SO3Na&Ph-MCMSS for toluene
(24.2 wt%) was much higher than that of water (2.41 wt%). On
the contrary, Yb(OTf)2-SO3Na-MCMSS showed the higher
adsorption capacity (16.4 wt%) of water than that of toluene
(4.59 wt%) and water. These results demonstrated that
Yb(OTf)2-SO3Na-MCMSS has a selective adsorption for hydro-
phobic organic substances.
Fig. 5 Reaction profiles over time in the water-medium Mukaiyama–
Aldol reaction between benzaldehyde and trimethyl(1-phenylprop-1-
enyloxy)silane over different catalysts. Reaction conditions are given in
Table 1.
Moreover, the Yb(OTf)2-SO3Na&Ph-MCMSS catalyst was
more active than the other control Yb(OTf)2-SO3Na&Ph-MCSS
catalyst with 69.5% yield under the same reaction conditions.
Firstly, it had a similar chemical microenvironment with that
of the Yb(OTf)2-SO3Na&Ph-MCMSS catalyst due to the same
Yb(OTf)3 immobilization approach (Experimental section in
ESI†). Moreover, we also calculated the density of Yb(III) active
sites in the supports for these two catalysts; the CYb/SBET value
calculated from the Yb content and the specific surface area of
the Yb(OTf)2-SO3Na&Ph-MCSS and Yb(OTf)2-SO3Na&Ph-MCSS
catalysts was 1.07 × 10−3 and 1.02 × 10−3 mmol m−2, respect-
ively. The approximate data confirmed an almost identical
active site density. Furthermore, we considered the diffusion
distance of the reactants since longer reaction distances
decrease the molecular interactions and thus leads to lower
catalytic activities.23 However, the TEM image (Fig. 7) revealed
that the Yb(OTf)2-SO3Na&Ph-MCSS catalyst was present in
uniform spheres with diameters of about 440 nm. Meanwhile,
Fig. 6 Hydrophobicity tests of the Yb(OTf)2-SO3Na&Ph-MCMSS and
Yb(OTf)2-SO3Na-MCMSS catalysts.
lytic activity with only 56.2% yield. To further validate the it displayed the core–shell structure with a magnetic core and
a thin silica shell of around 30 nm in thickness, demonstrating
that the reaction should occur in the catalyst surface without
enhanced hydrophobicity of the Yb(OTf)2-SO3Na&Ph-MCMSS
catalyst, both water and toluene absorption experiments were
performed (Fig. 6). All the isotherms were of type V, indicative significant diffusion resistance. Therefore, the higher catalytic
of the weak absorbent–absorbate interactions.22 However, the
efficiency of the Yb(OTf)2-SO3Na&Ph-MCSS catalyst could be
mainly attributed to the existence of a mesoporous structure
Yb(OTf)2-SO3Na&Ph-MCMSS sample had a higher affinity for
Green Chem.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
catalysts, including a SO3Na- and Ph-functionalized MCM-41
supported Yb(OTf)3 catalyst (Yb(OTf)2-SO3Na&Ph-MCM-41)
and a conventional ion-exchange resin supported Yb(OTf)3
catalyst (Yb(OTf)2-SO3Na-Amberlyst-15). The XRD pattern, N2
sorption isotherm and TEM image of Yb(OTf)2-SO3Na&Ph-
MCM-41 demonstrated its two-dimensional hexagonal ordered
mesoporous structure. However, the SEM images revealed that
it displayed an irregular morphology with a very broad particle
size distribution (Fig. S3†). The SEM and TEM images of the
Yb(OTf)2-SO3Na-Amberlyst-15 catalyst showed that it was a
nonporous solid (Fig. S4†). Under the same reaction
conditions (Table S3†), Yb(OTf)2-SO3Na&Ph-MCM-41 obtained
a relatively high yield of 87.3%, while Yb(OTf)2-SO3Na-Amber-
lyst-15 achieved a yield of 75.4%. The inferior catalytic
efficiency of Yb(OTf)2-SO3Na&Ph-MCM-41 could be attributed
to its increased diffusion resistance, while the low surface area
of Yb(OTf)2-SO3Na-Amberlyst-15 caused the decreased catalytic
activity (Table 1), furthermore demonstrating the advantage of
the Yb(OTf)2-SO3Na-MCMSS catalyst.
Fig. 8 Benzaldehyde adsorption profiles over the Yb(OTf)2-SO3Na&Ph-
MCMSS and Yb(OTf)2-SO3Na&Ph-MCSS catalysts.
with radial pore alignment, in which the unique confined
environments concentrated the reactants and thus led to the
enhanced activity.24 In order to prove the above hypothesis,
benzaldehyde adsorption tests in water over Yb(OTf)2-
SO3Na&Ph-MCMSS and Yb(OTf)2-SO3Na&Ph-MCSS catalysts
were performed. As shown in Fig. 8a, the Yb(OTf)2-SO3Na&Ph-
MCMSS catalyst displayed a higher saturated adsorption
capacity (65.0%) than the Yb(OTf)2-SO3Na&Ph-MCSS catalyst
(42.3%). Thus, we concluded that the higher enrichment
capacity of the organic reactants promoted the catalytic
reactivity of the Yb(OTf)2-SO3Na&Ph-MCMSS catalyst in water-
medium organic reactions.
Encouraged by these results, we examined the scope of the
Yb(OTf)2-SO3Na&Ph-MCSS catalyzed Mukaiyama–Aldol coup-
ling on a series of substrates. Similarly, 4-substituted benz-
aldehyde derivates possessing either an electron-accepting
group (NO2–) or electron-donating group (CH3O–) also led to
high yields of the coupling products. Notably, 2-naphthalde-
hyde (with a large molecular size) could also be converted to
product with an excellent yield of around 90%. Next, we inves-
tigated the conversion of aliphatic aldehydes with trimethyl-
(1-phenylprop-1-enyloxy)silane, especially for water-soluble
aldehydes, such as formaldehyde. As shown in Table 2, good
yields were observed for all the aliphatic aldehydes. In the
same way, the Yb(OTf)2-SO3Na&Ph-MCSS catalyst exhibited
a higher catalytic activity and selectivity than those of the
In addition, we also compared the catalytic performance of
the Yb(OTf)2-SO3Na-MCMSS catalyst with two non-magnetic
Table 2 Catalytic performances of the different Yb catalysts in the water-medium Mukaiyama–Aldol reactiona
Catalyst
R1
Conversion (%)
Selectivity (%)
Yield (%)
Yb(OTf)2-SO3Na&Ph-MCMSS
Yb(OTf)2-SO3Na-MCMSS
Yb(OTf)2-SO3Na&Ph-MCSS
92.5
53.6
64.3
99
99
99
92.4
53.5
64.2
Yb(OTf)2-SO3Na&Ph-MCMSS
Yb(OTf)2-SO3Na-MCMSS
Yb(OTf)2-SO3Na&Ph-MCSS
95.5
62.3
68.6
99
99
99
95.4
62.2
68.5
Yb(OTf)2-SO3Na&Ph-MCMSS
Yb(OTf)2-SO3Na-MCMSS
Yb(OTf)2-SO3Na&Ph-MCSS
Yb(OTf)2-SO3Na&Ph-MCMSS
Yb(OTf)2-SO3Na-MCMSS
Yb(OTf)2-SO3Na&Ph-MCSS
Yb(OTf)2-SO3Na&Ph-MCMSS
Yb(OTf)2-SO3Na-MCMSS
Yb(OTf)2-SO3Na&Ph-MCSS
90.0
47.8
62.1
95.2
59.2
71.0
92.1
75.9
83.5
99
99
99
99
99
99
84c
56
68
89.9
47.7
62.0
95.1
59.1
70.9
77.3
42.5
56.8
b
C5H11
Hb
a Reaction conditions: 0.50 mmol benzaldehyde, 1.0 mmol silane, 0.050 mmol Yb, 3.0 ml water, 20 °C, 16 h. b 10 mmol HCHO in 37% aqueous
solution, 1.0 mmol silane. c The selectivity was determined based on its ratio to the byproduct propiophenone from the hydrolysis of silane.
This journal is © The Royal Society of Chemistry 2014
Green Chem.

View Article Online
Paper
Green Chemistry
Fig. 9 (a) Wide-angle XRD patterns of the Fe3O4 and Yb(OTf)2-SO3Na&Ph-MCMSS samples, (b) magnetization curve of the Yb(OTf)2-SO3Na&Ph-
MCMSS catalyst determined at room temperature, (c) the reaction mixture containing the Yb(OTf)2-SO3Na&Ph-MCMSS catalyst, and (d) the reaction
mixture containing the Yb(OTf)2-SO3Na&Ph-MCMSS catalyst with the application of a magnet.
Yb(OTf)2-SO3Na-MCMSS and Yb(OTf)2-SO3Na&Ph-MCSS cata- rather than the result of any dissolved acid species leached
lysts for all the reactants, demonstrating its universal advan- from the Yb(OTf)2-SO3Na&Ph-MCSS catalyst.
tage in the water-medium Mukaiyama–Aldol reaction.
An attractive advantage of the Yb(OTf)2-SO3Na&Ph-MCSS
To determine whether the heterogeneous or the dissolved catalyst is its easy separation using an external magnet, which
homogeneous Yb(III) compound was the real acid active drastically reduces the operating time and costs arising from
species, the procedure proposed by Sheldon group was carried the separation of products from a reaction mixture. The reusa-
out.25 After reacting for 8 h the conversion exceeded 50% in bility of the magnetic mesoporous catalysts relies on their
the Mukaiyama–Aldol reaction between benzaldehyde and magnetic-response property, since the external coating on the
trimethyl-(1-phenylprop-1-enyloxy)silane, the mixture was sep- Fe3O4 surface usually decreased their magnetic responsivity.26
arated by an external magnet to remove the solid catalyst and The wide-angle XRD pattern exhibited well resolved diffraction
then the mother liquor was allowed to react for another 10 h peaks for the Yb(OTf)2-SO3Na&Ph-MCMSS catalyst (Fig. 9a). All
under the same reaction conditions. No significant change in the peaks can be exactly indexed to the diffractions of the Fe3O4
the benzaldehyde conversion was observed, suggesting that crystal in the cubic spinel structure, suggesting that the coating
the catalytic reactivity by the leaching Yb(III) catalyst could be process and surface modification have no effect on the crystal-
approximately excluded in the present reaction. Moreover, the line properties of magnetite. We furthermore analyzed its mag-
liquid phase of the reaction mixture was collected after each netic behavior using a magnetometer at 300 K (Fig. 9b). The
reaction. ICP-AES analysis confirmed a very low amount of Yb results indicated that the Yb(OTf)2-SO3Na&Ph-MCMSS catalyst
(less than 2.0 ppm) in the reaction mixture. Even if a long time had a magnetization saturation value of 21.3 emu g−1, demon-
reaction (36 h) was carried out or a large amount of the catalyst strating its superparamagnetic behavior.27 Thus, it could
(0.50 g) was added, a low Yb leaching amount (less than simply be recovered in a short time (<10 s) by fixing a magnet
3.5 ppm) was detected after reaction in both cases. Therefore, near to the reaction vessel (Fig. 9c and d). Fig. 10 shows the
the present catalysis was indeed heterogeneous in nature recyclability of the Yb(OTf)2-SO3Na&Ph-MCMSS catalyst during
Green Chem.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
exhibited a higher catalytic reactivity and selectivity. Control
experiments and systematic characterizations confirmed that
its unique mesoporous confinement effect and surface hydro-
phobicity could enrich the reactants and simultaneously reduce
the diffusion transfer limitation in water, resulting in the
enhanced catalytic efficiency. Moreover, it could be easily recycled
using an external magnet and reused at least six times without
loss of catalytic activity. This work provides a novel synthetic
approach towards the preparation of highly efficient and easily
recyclable heterogeneous catalysts for more environmentally
friendly chemical transformations.
Acknowledgements
This work was supported by the Natural Science Foundation of
China (21107071 and 51273112), PCSIRT (IRT1269) and the
Shanghai Government (13QA1402800 and 12CG52).
Fig. 10 The recycling tests of the Yb(OTf)2-SO3Na&Ph-MCMSS catalyst
in the water-medium Mukaiyama–Aldol reaction. Reaction conditions
are given in Table 1.
the water-medium Mukaiyama–Aldol reaction between
benzaldehyde and trimethyl-(1-phenylprop-1-enyloxy)silane.
No remarkable decrease could be observed in the catalytic
activity after being used repetitively six times, revealing its
superiority over the corresponding Yb(OTf)3 homogeneous
catalyst for reducing the cost and environmental pollution
from homogeneous acid and organic solvents. The ICP analy-
sis confirmed that only about 4.8% of the Yb species leached
off after six cycles in the recycled Yb(OTf)2-SO3Na&Ph-MCMSS
catalyst, demonstrating that the existence of the mesoporous
structure could effectively restrict the leaching of active sites.28
In addition, the N2 sorption isotherm and low-angle XRD
pattern (Fig. S5†) confirmed that the ordered 2-D mesostruc-
ture and high surface area were also well preserved. Moreover,
the TEM image (Fig. S6†) of the recycled Yb(OTf)2-SO3Na&Ph-
MCMSS catalyst after six repetitions indicated that it could
retain the spherical morphology and mesoporous structure and
only partial particle aggregation existed, suggesting the relatively
high structure stability in water-medium organic synthesis.
Notes and references
1 J. A. Melero, R. van Grieken and G. Morales, Chem. Rev.,
2006, 106, 3790–3812.
2 T. M. Jyothi, M. L. Kaliya and M. V. Landau, Angew. Chem.,
Int. Ed., 2001, 40, 2881–2884.
3 A. Corma and H. García, Chem. Rev., 2003, 103, 4307–4365.
4 R. N. Butler and A. G. Coyne, Chem. Rev., 2010, 110, 6302–
6337.
5 K. Inumaru, T. Ishiharam, Y. Kamiya, T. Okuhara and
S. Yamanaka, Angew. Chem., Int. Ed., 2007, 46, 7625–7628.
6 F. D. Clippel, M. Dusselier, S. V. D. Vyver, L. Peng, P. Jacobs
and B. F. Sels, Green Chem., 2013, 15, 1398–1430.
7 J. P. Ge, T. Huynh, Y. X. Hu and Y. D. Yin, Nano Lett., 2008,
8, 931–934.
8 (a) V. Polshettiwar, R. Luque, A. Fihri, H. B. Zhu, M. Bouhrara
and J. M. Basset, Chem. Rev., 2011, 111, 3036–3075;
(b) S. Shylesh, V. Schünemann and W. R. Thiel, Angew. Chem.,
Int. Ed., 2010, 49, 3428–3459; (c) M. B. Gawande, P. S. Brance
and R. S. Varma, Chem. Soc. Rev., 2013, 42, 3371–3393.
9 M. Feyen, C. Weidenthaler, F. Schüth and A. H. Lu, Chem.
Mater., 2010, 22, 2955–2961.
4. Conclusion
10 F. C. Zheng, Q. W. Chen, L. Hu, N. Yan and X. K. Kong,
Dalton Trans., 2014, 43, 1220–1227.
In summary, we have developed a novel approach to prepare a
highly active and water-compatible magnetic mesoporous 11 (a) J. Liu, S. Z. Qiao, Q. H. Hu and G. Q. Lu, Small, 2011, 7,
Lewis acid catalyst (Yb(OTf)2-SO3Na&Ph-MCMSS) by introdu-
cing Yb(OTf)3 into the mesoporous channels of a SO3Na- and
Ph-functionalized magnetic core–silica shell support. Notably,
it displayed a remarkable enhancement in the catalytic activity
for the water-medium Mukaiyama–Aldol reaction compared
to the homogeneous Yb(OTf)3 catalyst. In comparison to the
Yb(OTf)2-SO3Na-MCMSS catalyst without phenyl groups inside
the mesoporous channels, Yb(OTf)2-SO3Na&Ph-MCSS catalyst
without a mesoporous structure, Yb(OTf)2-SO3Na-MCM-41 meso-
porous catalyst with an irregular morphology and nonporous
Yb(OTf)2-SO3Na-Amberlyst-15 ion-exchange resin catalyst, it also
425–443; (b) J. P. Ge, Q. Zhang, T. R. Zhang and Y. D. Yin,
Angew. Chem., Int. Ed., 2008, 47, 8924–8928; (c) A. H. Lu,
W. Schmidt, N. Matoussevitch, H. Boennemann,
B. Spliethoff, B. Tesche, E. Bill, W. Kiefer and F. Schüeth,
Angew. Chem., Int. Ed., 2004, 43, 4303–4306; (d) D. Damodara,
R. Arundhathiab and P. R. Likhar, Catal. Sci. Technol.,
2013, 3, 797–802; (e) H. Q. Yang, Y. W. Wang, Y. Qin,
Y. Z. Chong, Q. Z. Yang, G. Li, L. Zhang and W. Li, Green
Chem., 2011, 13, 1352–1361; (f) R. B. Nasir Baig and
R. S. Varma, Green Chem., 2013, 15, 398–417; (g) Z. M. Cui,
Z. Chen, C. Y. Cao, L. Jiang and W. G. Song, Chem.
This journal is © The Royal Society of Chemistry 2014
Green Chem.

View Article Online
Paper
Green Chemistry
Commun., 2013, 49, 2332–2334; (h) J. H. Yang, D. F. Wang, 18 D. Y. Zhao, J. L. Feng, Q. S. Huo, N. Melosh,
W. D. Liu, X. Zhang, F. L. Bian and W. Yu, Green
Chem., 2013, 15, 3429–3437; (i) H. R. Shaterian and
M. Aghakhanizadeh, Catal. Sci. Technol., 2013, 3, 425–428.
12 (a) Y. H. Deng, Y. Cai, Z. K. Sun, J. Liu, C. Liu, J. Wei, W. Li,
G. H. Fredrickson, B. F. Chmelka and G. D. Stucky, Science,
1998, 279, 548–552.
19 K. Nakajima, I. Tomita, M. Hara, S. Hayashi, K. Domen and
J. N. Kondo, Adv. Mater., 2005, 17, 1839–1842.
C. Liu, Y. Wang and D. Y. Zhao, J. Am. Chem. Soc., 2010, 20 S. B. Jennifer Kan, K. K. H. Ng and I. Paterson, Angew.
132, 8466–8473; (b) S. Okada, S. Ikurumi, T. Kamegawa, Chem., Int. Ed., 2013, 52, 9097–9108.
K. Mori and H. Yamashita, J. Phys. Chem. C, 2012, 116, 21 R. Siegel, E. Domingues, R. De Sousa, F. Jérôme,
14360–14367; (c) Y. Chi, Q. Yuan, Y. J. Li, L. Zhao, N. Li,
X. T. Li and W. F. Yan, J. Hazard. Mater., 2013, 262, 404–
C. M. Morais, N. Bion, P. Ferreira and L. Mafra, J. Mater.
Chem., 2012, 22, 7412–7419.
411; (d) P. Li, Y. Yu, H. Liu, C. Y. Cao and W. G. Song, Nano- 22 F. Zhang, J. Yin, W. Chai and H. X. Li, ChemSusChem, 2010,
scale, 2014, 6, 442–448. 3, 724–727.
13 S. Shylesh, L. Wang, S. Demeshko and W. R. Thiel, Chem- 23 W. H. He, F. Zhang and H. X. Li, Chem. Sci., 2011, 2, 961–966.
CatChem, 2010, 2, 1543–1547.
14 C. S. Gill, B. A. Price and C. W. Jones, J. Catal., 2007, 251,
145–152.
24 M. Sanlés-Sobrido, M. Pérez-Lorenzo, B. Rodríguez-
González, V. Salgueiriño and M. A. Correa-Duarte, Angew.
Chem., Int. Ed., 2012, 51, 3877–3882.
15 Q. H. Yang, M. P. Kapoor and S. Inagaki, J. Am. Chem. Soc., 25 R. A. Sheldon, M. I. Wallau, W. C. E. Arends and
2002, 124, 9694–9695. U. Schuchardt, Acc. Chem. Res., 1998, 31, 485–493.
16 R. Kumar, J. P. Sharma and S. S. Sekhon, Eur. Polym. J., 26 M. Shokouhimehr, Y. Piao, J. Kim, Y. Jang and T. Hyeon,
2005, 41, 2718–2725. Angew. Chem., Int. Ed., 2007, 46, 7039–7043.
17 S. Suga, S. Ogawa, H. Namatame, M. Taniguchi, 27 M. J. Jin and D. H. Lee, Angew. Chem., Int. Ed., 2010, 49,
A. Kakizaki, T. Ishii, A. Fujimori, S. J. Oh, T. Miyahara, 1119–1022.
A. Ochiai, T. Suzuki and T. Kasuya, J. Phys. Soc. Jpn., 1989, 28 P. Sreekanth, S. Kim, T. Hyeon and B. Kim, Adv. Synth.
58, 4534–4543. Catal., 2003, 345, 936–938.
Green Chem.
This journal is © The Royal Society of Chemistry 2014