Core-shell silica magnetic microspheres supported proline as a recyclable organocatalyst for the asymmetric aldol reaction

Article abstract of DOI:10.1016/j.molcata.2012.07.017

l-4-Hydroxyproline has been successfully grafted onto the core-shell structural silica magnetic microspheres and characterized by elemental analysis, thermo gravimetric analysis (TGA), vibrating sample magnetometry (VSM), high resolution transmission electron microscopy (HRTEM) and Fourier transform infrared (FT-IR). The chiral immobilized catalyst demonstrated high catalytic activity (up to 92%), diastereoselectivity (up to 85:15) and enantioselectivity (up to 80%) in the asymmetric aldol reaction between aldehyde acceptors and ketone donors. On the other hand, the synthesized catalyst could be rapidly separated from the reaction mixture through an external magnetic field and reused up to five runs without any obvious loss of activity, indicating its easy-separated property and excellent recyclability.

Full text of DOI:10.1016/j.molcata.2012.07.017

Journal of Molecular Catalysis A: Chemical 363–364 (2012) 404–410
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Core–shell silica magnetic microspheres supported proline as a recyclable
organocatalyst for the asymmetric aldol reaction
Honglei Yang, Shuwen Li, Xiaoyu Wang, Fengwei Zhang, Xing Zhong, Zhengping Dong, Jiantai Ma∗
State Key Laboratory of Applied Organic Chemistry, College of Chemistry and Chemical Engineering, Lanzhou University, Lanzhou, Gansu 730000, People’s Republic of China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 25 May 2012
Received in revised form 10 July 2012
Accepted 16 July 2012
Available online 31 July 2012
l-4-Hydroxyproline has been successfully grafted onto the core–shell structural silica magnetic micro-
spheres and characterized by elemental analysis, thermo gravimetric analysis (TGA), vibrating sample
magnetometry (VSM), high resolution transmission electron microscopy (HRTEM) and Fourier trans-
form infrared (FT-IR). The chiral immobilized catalyst demonstrated high catalytic activity (up to 92%),
diastereoselectivity (up to 85:15) and enantioselectivity (up to 80%) in the asymmetric aldol reaction
between aldehyde acceptors and ketone donors. On the other hand, the synthesized catalyst could
be rapidly separated from the reaction mixture through an external magnetic field and reused up to
five runs without any obvious loss of activity, indicating its easy-separated property and excellent
recyclability.
Keywords:
Recyclable organocatalyst
Supported proline
Silica magnetic nanoparticles
Asymmetric aldol reaction
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
the possibility to explore modifications of the properties of the
supported catalysts by employing specific characteristics of the
Since reported as a catalyst for the asymmetric aldol reaction by
List et al. [1] in 2000, l-proline has attracted a wild spread attention
in the field of asymmetric catalysis. Compared with metal catalysts,
l-proline, as a non-metallic organocatalyst, has a lot of advantages,
such as milder reaction conditions, lower price, easier immobi-
lization and recovery, better stability and less pollution. Besides
the aldol reactions [2–5], l-proline has also been applied to many
other asymmetric reactions. Quite a few researchers successively
used proline and its derivatives as catalysts for Michael addition
reactions [6–9]. Mannich reactions were also usually catalyzed by
proline [10–15]. Bui and Barbas III [16] reported that l-proline could
act as catalysts in both steps of the Robinson annulation reaction.
The research work of Thayumanavan et al. [17], Ramachary et al.
[18], and Sabitha et al. [19] revealed the proline’s catalytic activities
for Diels–Alder reactions, respectively.
At the same time, immobilization and recycling of l-proline
have received considerable concerns in recent years. Although l-
proline is not very expensive, the studies of supported proline and
its derivatives still have important significance. Supported pro-
line catalysts can be easily recovered from the reaction mixture
and keep stable catalytic activity and selectivity after being reused
for many times, which is meaningful for the environmental pro-
tection and energy conservation. Moreover, immobilization gives
supports. Several types of supports such as polymer, silica, ionic
liquid, cyclodextrin and magnetite are usually considered for the
immobilizations of proline and its derivatives. Gruttadauria and
his co-workers [20] used polystyrene-supported, proline-based
organic catalysts for the direct asymmetric aldol reaction. Zou et al.
[21] investigated the catalytic behaviors of PVC–TEPA-supported
l-proline in the aldol reaction. Lu et al. [22] synthesized l-proline
functionalized polymers as supported organocatalysts. Zamboulis
et al. [23] grafted l-proline onto the heterogenized silica for cat-
alyzing the asymmetric aldolization. Doyagüez et al. [24] catalyzed
the asymmetric aldol reaction by proline on mesoporous materi-
als. Miao and Chan [25] prepared an ionic-liquid-anchored proline
catalyst which was efficient and recyclable for asymmetric aldol
reaction. Luo et al. [26] used proline functionalizing chiral ionic
liquids as highly efficient asymmetric organocatalysts in Michael
addition reaction. Immobilizations of proline and its derivatives
into the ␤-cyclodextrin cavity as catalysts for direct asymmetric
aldol reactions were also reported [27,28].
Magnetic materials, especially Fe3O4 and ␥-Fe2O3, have many
outstanding properties such as superparamagnetism and low toxic-
ity. Fe3O4 coated with silica was commonly applied as the support
of metal and non-metal catalysts [29–33]. These silica magnetic
microspheres have large surface area and can be functionalized
easily. In catalytic applications, the magnetic particles have high
stability, which can be used in kinds of organic and inorganic
solutions and easy to be separated from the reaction mixture by
magnetic decantation.
∗
Corresponding author. Tel.: +86 931 8912577; fax: +86 931 8912582.
E-mail address: majiantai@lzu.edu.cn (J. Ma).
1381-1169/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molcata.2012.07.017

H. Yang et al. / Journal of Molecular Catalysis A: Chemical 363–364 (2012) 404–410
405
In our present work, l-4-hydroxyproline has been successfully
2.1.5. Preparation of Fe3O4@SiO2-Pro
anchored onto the Fe3O4@SiO2 nanoparticles, consequently named
Fe3O4@SiO2-Pro, and confirmed by corresponding characterization
means. Moreover, the catalytic performances and recycling utiliza-
tion rate of the catalyst Fe3O4@SiO2-Pro in the asymmetric aldol
reaction were investigated. Although the catalyst based on pro-
line and magnetite has been studied [31,34], our catalyst had much
more excellent mechanical strength and recyclability compared
with those related reports. Furthermore, the effects of the catalyst
support Fe3O4@SiO2 for the reaction got a discussion.
The solution of the proline derivative (compound 5, 1 g) in
toluene (20 mL) was added into the suspension of the inor-
ganic support of Fe3O4@SiO2 (3 g) in a mixture of toluene/water
(40 mL/40 ␮L). The mixture was refluxed for 24 h. The solid was
then filtered and washed with several solvents of different polarity
(MeOH, AcOEt, CH2Cl2, hexane and ether) abundantly, to remove
the remaining non-supported proline derivative. The brown yellow
solid was dried at 40 ◦C under vacuum.
Several catalysts with different amount of proline were also pre-
pared through adjusting the proportion of the proline derivative
and Fe3O4@SiO2.
2. Experimental
2.2. Characterizations of the catalyst Fe3O4@SiO2-Pro
2.1. Preparation of the catalyst Fe3O4@SiO2-Pro
The synthesized catalyst Fe3O4@SiO2-Pro was confirmed by cor-
responding characterization means. The C, H and N contents in
Fe3O4@SiO2-Pro were determined by Elementar Analysensysteme
GmbH varioEL cube. Thermal gravimetric analysis (TGA) was mea-
sured under nitrogen atmosphere to 800 ◦C with a Perkin Elmer
Thermal Analyzer at a heating rate of 10 ◦C min−1. Magnetic mea-
surements of Fe3O4@SiO2 and Fe3O4@SiO2-Pro were investigated
by a quantum design vibrating sample magnetometry (VSM) at
room temperature in an applied magnetic field sweeping from
−15 to 15 kOe. The morphology and microstructure of Fe3O4@SiO2-
Pro were characterized by high-resolution transmission electron
microscopy (HRTEM). The HRTEM images were obtained through
Tecnai G2 F30 electron microscope operating at 300 kV. Fourier
transform infrared (FT-IR) spectra were recorded on a Nicolet
NEXUS 670 FTIR spectrometer with a DTGS detector, and samples
were measured with KBr pellets.
The core–shell structural silica magnetic nanoparticles as sup-
ports of the catalyst were synthesized according to the previously
published methods [35]. Then, the catalyst Fe3O4@SiO2-Pro was
prepared by steps showed in Fig. 1.
2.1.1. Synthesis of compound 2
In ice bath, benzyl chloroformate (24 mmol, 4.09 g) was added
dropwise to a mixture of l-4-hydroxyproline (20 mmol, 2.62 g) and
NaHCO3 (60 mmol, 5.04 g) in acetone (20 mL) and distilled water
(40 mL) under continuous stirring. The resulting mixture was kept
stirring in ice environment for 30 min and followed by reacting at
room temperature for 2 h. Then the pH value of the reaction sys-
tem was adjusted to 3–4 with 1 mol/L HCl. Acetone was removed
by vacuum distillation and the aqueous phase was extracted by
CH2Cl2 (4 × 30 mL). The organic layer was dried on anhydrous
Na2SO4, filtered and evaporated. (2S,4R)-1-benzyloxycarbonyl-4-
hydroxyproline (compound 2) was obtained as colorless oil (4.51 g,
yield 85%).
2.3. Reaction procedures for activity evaluation of catalyst
Fe3O4@SiO2-Pro in the direct aldol reaction
2.3.1. Optimization experiments of reaction conditions
2.1.2. Synthesis of compound 3
4-Nitrobenzaldehyde (0.5 mmol, 76 mg), cyclohexanone and
corresponding kinds and amounts of catalysts were added in the
solvent. The reaction mixture was stirred for 24–72 h at room tem-
perature and treated with 10 mL of saturated ammonium chloride
solution and extracted with ethyl acetate (3 × 10 mL). The organic
layer was dried over anhydrous Na2SO4, and concentrated in vacuo.
The products were isolated by column chromatography on sil-
ica gel (hexane/AcOEt, 4:1). The enantiomeric excess of products
was determined by chiral high performance liquid chromatogra-
phy (HPLC) using CHIRALPAK AD-H column with n-hexane and
isopropyl alcohol (95:5) as eluants.
Triethylamine was added to the solution of compound 2
(15 mmol, 3.98 g) and benzyl bromide (16.5 mmol, 2.82 g) in THF
(25 mL) at 0 ◦C. After the mixture was stirred for 18 h at room
temperature, the solvent was evaporated in vacuo. The residue
was dissolved in 50 mL of CH2Cl2, washed with HCl (1 N), H2O,
Na2CO3 (5%), and H2O, and then dried over Na2SO4. The sol-
vent was evaporated, and the residue was purified by column
chromatography on silica gel (hexane/AcOEt, 2:1) to afford (2S,4R)-
1,2-dibenzyloxycarbonyl-4-hydroxypyrrolidine (compound 3) as
pale yellow oil (2.66 g, yield 50%).
2.1.3. Synthesis of compound 4
2.3.2. The direct aldol reactions between different aldehydes and
ketones
The solution of compound 3 (10 mmol, 3.55 g), triethoxysilyl-
propyl isocyanate (15 mmol, 3.71 g), and triethylamine (20 mmol,
2.02 g) in THF (25 mL) was refluxed for 24 h and then cooled to room
temperature. After removal of solvent under vaccum, the crude
product was purified by column chromatography on silica gel (hex-
ane/AcOEt, 3:1) to give 4.99 g of (2S,4R)-1,2-dibenzyloxycarbonyl-
4-(3-triethoxysilylpropylaminocarboxy) pyrrolidine (compound 4)
as a brown oil (yield 83%).
The catalyst Fe3O4@SiO2-Pro (20 mol%) and corresponding alde-
hyde (0.5 mmol) and ketone (2 mmol) were stirred in 2 mL ethanol
for 24–72 h at room temperature. The mixture was treated with
10 mL of saturated ammonium chloride solution and extracted
with ethyl acetate (3 × 10 mL). The organic layer was dried over
anhydrous Na2SO4, and concentrated in vacuo. The residue was
purified by column chromatography on silica gel, eluting with
hexanes/ethyl acetate. The enantiomeric excess of products was
determined by chiral HPLC using CHIRALPAK AD-H and OB-H
columns with n-hexane and isopropyl alcohol as eluants.
2.1.4. Synthesis of compound 5
Compound 4 (2 mmol, 1.20 g) was hydrogenated over palla-
dium on carbon (0.16 g, 10%) in methanol (40 mL) for 7 h at room
temperature under H2 atmosphere (3 atm). The catalyst Pd/C was
filtered off, and the filtrate was concentrated in vacuo yielding
(2S,4R)-2-carboxy-4-(3-triethoxysilylpropylaminocarboxy) pyrro-
lidine (compound 5) quantitatively as pale yellow oil. No
purification was possible due to the instability of the product.
2.3.3. Recycling tests of catalyst Fe3O4@SiO2-Pro
4-Nitrobenzaldehyde (0.5 mmol, 76 mg), catalyst Fe3O4@SiO2-
Pro (20 mol%), cyclohexanone (2 mmol, 196 mg) and ethanol (2 mL)
were mixed and stirred at room temperature for 48 h. The cata-
lyst Fe3O4@SiO2-Pro was removed by an external magnetic field,

406
H. Yang et al. / Journal of Molecular Catalysis A: Chemical 363–364 (2012) 404–410
Fig. 1. Synthesis of Fe3O4@SiO2-Pro.
washed with MeOH and AcOEt for several times and dried under
vacuum. Then another portion of reactants was added. The products
in every recycling test were isolated by column chromatography
and analyzed by chiral HPLC.
3. Results and discussion
3.1. Characterizations of Fe3O4@SiO2-Pro
Through elemental analysis of C, H and N, the proline content of
Fe3O4@SiO2-Pro was evaluated to be 0.693 mmol/g.
The results of TGA analysis for Fe3O4@SiO2 and Fe3O4@SiO2-
Pro were depicted in Fig. 2. The weight losses of both Fe3O4@SiO2
and Fe3O4@SiO2-Pro below 200 ◦C can be assigned to the release
of physisorbed and chemisorbed water on the surface of the silica
shell. The weight loss at the temperature range of 200–600 ◦C in
the TGA curve of Fe3O4@SiO2-Pro (Fig. 2b) was mainly attributed
to the decomposition of organic groups from the grafted pro-
line. Meanwhile, the weight loss above 200 ◦C was also possibly
caused by the loss of structural water within amorphous SiO2.
In this way, the slow slope appeared at the temperature range
of 200–600 ◦C in the TGA curve of Fe3O4@SiO2 (Fig. 2a) can be
explained. The total weight loss was evaluated to be 10% and 23% for
Fe3O4@SiO2 and Fe3O4@SiO2-Pro, respectively. Through the TGA
analysis, the proline content of Fe3O4@SiO2-Pro was evaluated to
be 0.605 mmol/g, which was close to the CHN elemental analysis
result (0.693 mmol/g). The little difference between the results of
the proline content in Fe3O4@SiO2-Pro was mainly attributed to
the difference of detection methods and the instrument error of
the two detection means.
Fig. 2. TGA curves of (a) Fe3O4@SiO2 and (b) Fe3O4@SiO2-Pro.
at room temperature were given in Fig. 3. The saturation magne-
tization value of Fe3O4@SiO2-Pro was measured to be 2.04 emu/g.
The zero coercivity and resonance of each magnetization loop evi-
denced the superparamagnetism of the magnetic materials which
was very useful for the catalyst’s rapid dispersion and separation.
As shown in the lower right corner of Fig. 3, the nanoparticles of
Fe3O4@SiO2-Pro had a good dispersion in the solvent of ethanol,
and an excellent magnetic separation capability appeared when
there was a magnet near the vessel.
The magnetic property of the synthesized catalyst was exam-
ined through VSM analysis. The hysteresis loops of the silica
magnetic nanoparticles before and after grafting proline measured

H. Yang et al. / Journal of Molecular Catalysis A: Chemical 363–364 (2012) 404–410
407
the separation of the catalyst from the reaction mixture. Gener-
ally, compared with the large-size support solid, nanoparticles with
small particle sizes have larger specific surface area and better dis-
persion in the solvent, which can have beneficial effects to the
catalysis. However, it easily leads to an agglomeration in the pro-
cess of reaction, and then seriously affects the activity of the catalyst
when the particle size is too small. In this paper, the distribution
of the particle sizes was controlled in a range of 60–100 nm. The
particles of the catalyst did not agglomerate and kept high activity
even after five runs, evidenced by the HRTEM images of the reused
catalyst (Fig. 4e and f) and the recycling experiments (Table 3).
The FT-IR spectra of Fe3O4@SiO2, Fe3O4@SiO2-Pro, and
Fe3O4@SiO2-Pro after five runs were illustrated in Fig. 5. The strong
absorption band at 1095 cm−1 was attributed to Si
O Si vibra-
tions, indicating the presence of SiO2. Compared to the FT-IR
spectrum of Fe3O4@SiO2, there were two new peaks near 1560 and
1635 cm−1 obviously appearing in that of Fe3O4@SiO2-Pro, corre-
sponding to bending vibration of the N H groups in proline. These
results demonstrated that l-4-hydroxyproline has been success-
fully grafted on the surface of Fe3O4@SiO2 and the original structure
of silica magnetic particles was retained.
Fig. 3. Magnetization curves of (a) Fe3O4@SiO2 and (b) Fe3O4@SiO2-Pro.
3.2. Activity evaluations of Fe3O4@SiO2-Pro
The HRTEM images shown in Fig.
4 revealed that the
silica–magnetite composites prepared as supports of the catalyst
had good spherical morphologies and regular core–shell structures.
Grafting proline onto the Fe3O4@SiO2 did not lead to a significant
increase in the size of the individual particles. The average particle
size of as-synthesized catalyst was about 80 nm and the diameter
of the magnetic core was about 10 nm.
Through the seeded sol–gel approach, the thickness of silica
shell of the Fe3O4@SiO2 particles can be conveniently adjusted
by controlling the addition amount of silica source (TEOS) [35].
With the increase of the particle sizes, the magnetic property of
the particles certainly would decrease, which was unfavorable for
With the synthesized catalyst Fe3O4@SiO2-Pro in hand, its cat-
alytic activity was evaluated in the asymmetric aldol reaction at
room temperature.
The
well-documented
aldol
reaction
between
4-
nitrobenzaldehyde and cyclohexanone was chosen as a model to
optimize the reaction conditions (Table 1). In all cases, the desired
aldol product was obtained as a mixture of diastereomers.
In order to find out the optimum molar ratio of substrates,
different stoichiometric ratios of 4-nitrobenzaldehyde and cyclo-
hexanone (1:1, 1:2, 1:4, 1:10 and 1:20) were used in the
experiments respectively (entries 1–5). When the stoichiometric
Fig. 4. HRTEM images of (a), (b) Fe3O4@SiO2, (c), (d) Fe3O4@SiO2-Pro and (e), (f) Fe3O4@SiO2-Pro after five runs.

408
H. Yang et al. / Journal of Molecular Catalysis A: Chemical 363–364 (2012) 404–410
Table 1
Optimization conditions for the direct aldol reaction catalyzed by Fe3O4@SiO2-Pro.
Entry
Catalyst
Loading (mol%)
Time (h)
Solvent
Yielda (%)
drb (anti:syn)
eeb (%) anti (syn)
1c
2c
3
Fe3O4@SiO2-Pro
Fe3O4@SiO2-Pro
Fe3O4@SiO2-Pro
Fe3O4@SiO2-Pro
Fe3O4@SiO2-Pro
Fe3O4@SiO2-Prod
Fe3O4@SiO2-Prod
Fe3O4@SiO2-Prod
Fe3O4@SiO2-Pro
Fe3O4@SiO2-Pro
–
20
20
20
20
20
20
20
20
20
20
–
48
48
48
48
48
48
48
48
48
48
48
48
48
48
24
72
EtOH
EtOH
EtOH
EtOH
EtOH
EtOH
EtOH
EtOH
CHCl3
H2O
EtOH
EtOH
EtOH
EtOH
EtOH
EtOH
32
62
92
94
91
89
92
90
57
<10
n.d.
37
56
94
63
91
84:16
90:10
85:15
80:20
75:25
88:12
81:19
90:10
94:6
82 (23)
78 (6)
80 (12)
76 (10)
81 (31)
76 (5)
81 (23)
77 (19)
58 (10)
n.d.
n.d.
–
77 (15)
84 (26)
85 (7)
67 (10)
4c
5c
6
7
8
9
10
11
12
13
14
15
16
n.d.e
n.d.
–
95:5
84:16
76:24
97:3
f
Fe3O4@SiO2
–
Fe3O4@SiO2-Pro
Fe3O4@SiO2-Pro
Fe3O4@SiO2-Pro
Fe3O4@SiO2-Pro
10
30
20
20
a
Isolated yield of mixture of all four diastereomers.
Determined by chiral HPLC (Chiralpak AD-H column, 95:5 n-hexane/isopropanol, 1 mL/min, detected at 254 nm).
The molar ratios of 4-nitrobenzaldehyde and cyclohexanone were 1:1, 1:2, 1:10, 1:20 respectively, and 1:4 was always used in the other groups of experiments.
Catalysts with different amount of proline (0.412 mmol/g, 0.791 mmol/g, 1.17 mmol/g) were used respectively, and catalyst with 0.693 mmol/g of proline was always
b
c
d
used in the other groups of experiments.
e
n.d. = not determined.
The mass equal to 20 mol% Fe3O4@SiO2-Pro was added.
f
content of proline per unit mass of the catalyst was too high, the
mechanical loss of the catalyst solid in the process of recycling
would cause more loss of proline in the catalyst, which possi-
bly decreased the catalytic activity seriously. Therefore, from an
economic point of view, 0.693 mmol/g was the reasonable proline
content in Fe3O4@SiO2-Pro.
Regrettably, H2O, as an environmentally friendly solvent, was
eliminated owing to its restraint for the activity of the catalyst
(entry 10). However, the reaction proceeded with high yield and
good enantioselectivity in ethanol (entry 3). Therefore, ethanol,
another low-polluting solvent, was chosen as the most suitable
solvent for this reaction. Several groups of controlled experiments
were arranged to investigate the influence of the catalyst’s amount
for the reaction. It could be seen that when no catalyst participat-
ing, the reaction almost hardly occurred (entry 11). Meanwhile,
Fe3O4@SiO2 nanoparticles also had a catalytic activity in the aldol
reaction (entry 12), due to the silanol groups on the silica shell,
which were considered as very weak acids. The weakly acidic sur-
face hydroxyl groups often form hydrogen bonds with oxygenates
and the H-bonds probably catalyzed the aldol reactions in this work
[36]. This result seemed to demonstrate that the solid support
Fe3O4@SiO2 also acted via its acidic sites in the Fe3O4@SiO2-Pro
catalyzing aldol reactions. However, the products of this group
of controlled experiment were racemic, which revealed that pro-
line played a very important catalytic role in the asymmetric aldol
reaction. The yield and enantioselectivity did not improve appar-
ently after increasing the catalyst loading from 20 to 30 mol%. And
although a good enantioselectivity was still obtained, the yield was
poor when the amount of catalyst was reduced to 10 mol%. From an
economical point of view, 20 mol% was the preferred catalyst load-
ing. Compared to reacting for 48 h, the yield of reacting for 24 h was
lower, although the enantiomeric excess values (ee values) were
quite similar. And little decrease even appeared both in the yield
and enantioselectivity when the reaction proceeded for 72 h. So the
optimum reaction time was decided to be 48 h.
Fig. 5. FT-IR spectra of (a) Fe3O4@SiO2, (b) Fe3O4@SiO2-Pro and (c) Fe3O4@SiO2-Pro
after five runs.
ratio was changed from 1:1 to 1:4, the yield had a significant
increase and became flat after 1:4. In all experiments with dif-
ferent stoichiometric ratios, good enantioselectivities were always
obtained. So the molar ratio of 1:4 was chosen in the subsequent
experiments.
Several catalysts with different amount of proline were designed
(0.412 mmol/g, 0.791 mmol/g, 1.17 mmol//g) for comparison. Obvi-
ous differences were not observed in the data of their catalytic
activities given in Table 1 (entries 6–8), when 20 mol% of proline
was added in the all reactions. However, when the content of pro-
line in Fe3O4@SiO2-Pro was too low, the amount of the support
would be over-used in the reaction. On the other hand, when the
Under the optimized conditions, the catalyst Fe3O4@SiO2-Pro
was tested in aldol reactions between a variety of aromatic

H. Yang et al. / Journal of Molecular Catalysis A: Chemical 363–364 (2012) 404–410
409
Table 2
Evaluation of the catalyst Fe3O4@SiO2-Pro for the direct aldol reactions between different aromatic aldehydes and ketones.
Entry
R
n
Time (h)
Yielda (%)
drb (anti:syn)
eeb (%) anti (syn)
1
2
3
4
5
6
7
4-NC
2
2
2
48
48
48
24
72
72
24
85
87
72
95
37
24
96
86:14
98:2
6:94
27:73
45:55
64:36
–
79 (7)
23 (20)
6 (40)
99 (27)
91 (96)
11 (79)
43e
3-O2N
2-O2N
4-O2N
4-O2N
4-O2N
4-O2N
1
3c
4c
d
–
a
Isolated yield of mixture of all four diastereomers.
Determined by chiral HPLC (Chiralpak AD-H column, 95:5 n-hexane/isopropanol, 1 mL/min, detected at 254 nm).
0.025 mmol of acetic acid was added.
Acetone was used as the ketone component.
Determined by chiral HPLC (Chiralpak OB-H column, 85:15 n-hexane/isopropanol, 1 mL/min, detected at 254 nm).
b
c
d
e
Table 3
Recycling experiments of the catalyst Fe3O4@SiO2-Pro for the aldol reaction between 4-nitrobenzaldehyde and cyclohexanone.
Run
Yielda (%)
drb (anti:syn)
eeb (%)
1
2
3
4
5
92
87
88
85
89
85:15
77:23
87:13
95:5
80 (12)
86 (27)
81 (17)
86 (13)
84 (20)
80:20
a
Isolated yield of mixture of all four diastereomers.
Determined by chiral HPLC (Chiralpak AD-H column, 95:5 n-hexane/isopropanol, 1 mL/min, detected at 254 nm).
b
aldehyde acceptors and ketone donors (Table 2). 4-
Cyanobenzaldehyde was a good acceptors and gave the desired
anti products in good yield with high diastereoselectivity and
enantioselectivity (entry 1), which indicated that the aromatic
aldehydes carrying strong electron withdrawing groups such as
nitro or cyano group were good substrates for the aldol reaction.
Unfortunately, attempts changing the substituting group positions
onto the aromatic aldehydes resulted in lower ee values, although
the yields remained good (entries 2 and 3). These phenomena
were probably due to the electronic and steric effects of aromatic
aldehydes. The formation of aldol products related with the
electron accepting capacity of the substituent group on aldehy-
des and the steric hindrance at position 4 on the aromatic ring
played a positive role for enantioface discrimination in transition
state [37]. Good ee values were still obtained when other cyclic
ketones including cyclopentanone, cycloheptanone and cyclooc-
tanone were used as donors, whereas the yields had a significant
drop with the increasing of the carbon numbers onto the cyclic
ketones (entries 4–6). Acetone was also proved to be a suitable
donor providing excellent yield with moderate enantioselectivity
(entry 7).
without any significant loss of activity. Recently, Wang et al. [31]
and Riente et al. [34] designed organocatalysts based on proline
derivatives as active centers and magnetic nanoparticles as support
and investigated their catalytic performances in the asymmetric
Michael addition reactions, successively. Both of their catalysts
have high activities and enantioselectivity in the reactions. How-
ever, the same phenomenon occurred in the recycling tests of the
catalysts that after the third run, a significant decrease in catalytic
activity was observed, although the enantioselectivity was main-
tained. Compared with their reports, our catalyst had much more
excellent mechanical strength and recyclability.
Through elemental analysis, the catalyst Fe3O4@SiO2-Pro did
not have an obvious change in proline content (0.622 mmol/g) after
being reused for five times. As was shown in Fig. 5c, the FT-IR spec-
trum of Fe3O4@SiO2-Pro after five runs still kept the characteristic
peaks of proline compared with that of fresh catalyst (Fig. 5b). From
the HRTEM images (Fig. 4e and f), it can be seen that the reused
catalyst did not have an obvious agglomeration and still retained
the good spherical morphology and regular core–shell structure,
which revealed that the synthesized catalyst Fe3O4@SiO2-Pro had
an excellent mechanical strength and recyclability.
The silica magnetic microspheres supported catalyst can be eas-
ily separated from the reaction media and subsequently reused
through an external magnetic field, which is one of the outstand-
ing advantages associated with supporting catalysts onto magnetic
materials. The possibility of the reuse of Fe3O4@SiO2-Pro in the
reaction between 4-nitrobenzaldehyde and cyclohexanone was
studied and the results were shown in Table 3. This magnetic
nanoparticles supported proline can be reused up to five runs
4. Conclusion
A new chiral immobilized organocatalyst based on proline as
active sites and silica magnetic nanoparticles as supports has been
successfully prepared and applied to the asymmetric aldol reac-
tion between aromatic aldehydes and ketones. Moderate to good

410
H. Yang et al. / Journal of Molecular Catalysis A: Chemical 363–364 (2012) 404–410
yields, diastereoselectivities and enantioselectivities were shown
in the activity evaluation experiments. This synthesized catalyst
had excellent magnetic separation ability, mechanical strength and
recyclability, evidenced by being extensively reused without any
substantial loss of activity.
[12] W. Notz, F. Tanaka, S. Watanabe, N.S. Chowdari, J.M. Turner, R. Thayumanavan,
C.F. Barbas 3rd, J. Org. Chem. 68 (2003) 9624–9634.
[13] A.V. Malkov, L. Dufkova, L. Farrugia, P. Kocovsky, Angew. Chem. Int. Ed. 42
(2003) 3674–3677.
[14] A. Córdova, Synlett (2003) 1651–1654.
[15] N.S. Chowdari, J.T. Suri, C.F. Barbas 3rd, Org. Lett. 6 (2004) 2507–2510.
[16] T. Bui, C.F. Barbas, Tetrahedron Lett. 41 (2000) 6951–6954.
[17] R. Thayumanavan, B. Dhevalapally, K. Sakthivel, F. Tanaka, C.F. Barbas, Tetra-
hedron Lett. 43 (2002) 3817–3820.
[18] D.B. Ramachary, N.S. Chowdari, C.F. Barbas, Tetrahedron Lett. 43 (2002)
6743–6746.
[19] G. Sabitha, N. Fatima, E.V. Reddy, J.S. Yadav, Adv. Synth. Catal. 347 (2005)
1353–1355.
[20] M. Gruttadauria, A.M.P. Salvo, F. Giacalone, P. Agrigento, R. Noto, Eur. J. Org.
Chem. 2009 (2009) 5437–5444.
Acknowledgments
This work was supported by the Fundamental Research Funds
for the Central Universities (lzujbky-2012-68).
Appendix A. Supplementary data
[21] J. Zou, W. Zhao, R. Li, H. Zhang, Y. Cui, J. Appl. Polym. Sci. 118 (2010)
1020–1026.
[22] A. Lu, T.P. Smart, T.H. Epps, D.A. Longbottom 3rd, R.K. O’Reilly, Macromolecules
44 (2011) 7233–7241.
[23] A. Zamboulis, N.J. Rahier, M. Gehringer, X. Cattoën, G. Niel, C. Bied, J.J.E. Moreau,
M.W.C. Man, Tetrahedron: Asymmetry 20 (2009) 2880–2885.
[24] E.G. Doyaguez, F. Calderon, F. Sanchez, A. Fernandez-Mayoralas, J. Org. Chem.
72 (2007) 9353–9356.
[25] W. Miao, T.H. Chan, Adv. Synth. Catal. 348 (2006) 1711–1718.
[26] S. Luo, X. Mi, L. Zhang, S. Liu, H. Xu, J.P. Cheng, Angew. Chem. Int. Ed. 45 (2006)
3093–3097.
[27] Z. Shen, J. Ma, Y. Liu, C. Jiao, M. Li, Y. Zhang, Chirality 17 (2005) 556–558.
[28] J. Huang, X. Zhang, D.W. Armstrong, Angew. Chem. Int. Ed. 46 (2007)
9073–9077.
[29] G. Chouhan, D. Wang, H. Alper, Chem. Commun. (2007) 4809–4811.
[30] L. Zhou, C. Gao, W. Xu, Langmuir 26 (2010) 11217–11225.
[31] B.G. Wang, B.C. Ma, Q. Wang, W. Wang, Adv. Synth. Catal. 352 (2010)
2923–2928.
[32] Q. Zhang, H. Su, J. Luo, Y. Wei, Green Chem. 14 (2012) 201.
[33] S. Luo, X. Zheng, J.P. Cheng, Chem. Commun. (2008) 5719–5721.
[34] P. Riente, C. Mendoza, M.A. Pericás, J. Mater. Chem. 21 (2011) 7350.
[35] D. Yang, J. Hu, S. Fu, J. Phys. Chem. C 113 (2009) 7646–7651.
[36] F. Niu, L. Zhang, S.Z. Luo, W.G. Song, Chem. Commun. 46 (2010) 1109–1111.
[37] G.-G. Liu, H. Zhao, Y.-B. Lan, B. Wu, X.-F. Huang, J. Chen, J.-C. Tao, X.-W. Wang,
Tetrahedron 68 (2012) 3843–3850.
Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/j.
molcata.2012.07.017.
References
[1] B. List, R.A. Lerner, C.F. Barbas, J. Am. Chem. Soc. 122 (2000) 2395–2396.
[2] K. Sakthivel, W. Notz, T. Bui, C.F. Barbas, J. Am. Chem. Soc. 123 (2001)
5260–5267.
[3] Q. Pan, B. Zou, Y. Wang, D. Ma, Org. Lett. 6 (2004) 1009–1012.
[4] A.B. Northrup, I.K. Mangion, F. Hettche, D.W.C. MacMillan, Angew. Chem. Int.
Ed. 43 (2004) 2152–2154.
[5] J.T. Suri, S. Mitsumori, K. Albertshofer, F. Tanaka, C.F. Barbas, J. Org. Chem. 71
(2006) 3822–3828.
[6] B. List, P. Pojarliev, H.J. Martin, Org. Lett. 3 (2001) 2423–2425.
[7] J.M. Betancort, C.F. Barbas, Org. Lett. 3 (2001) 3737–3740.
[8] D. Enders, A. Seki, Synlett (2002) 26–28.
[9] D. Gryko, Tetrahedron: Asymmetry 16 (2005) 1377–1383.
[10] B. List, P. Pojarliev, W.T. Biller, H.J. Martin, J. Am. Chem. Soc. 124 (2002) 827–833.
[11] A. Córdova, W. Notz, G. Zhong, J.M. Betancort, C.F. Barbas, J. Am. Chem. Soc. 124
(2002) 1842–1843.