Enantioselective carbonyl-ene reaction on BINOLate/titanium catalyst encapsulated in magnetic nanoreactors

Article abstract of DOI:10.1039/c2cc17950d

An asymmetric multicomponent catalyst, BINOLate/titanium, successfully encapsulated in the nanocages of mesoporous silicas exhibits much higher activity than the homogeneous counterpart in quasi solvent-free enantioselective carbonyl-ene reaction, owing t

Full text of DOI:10.1039/c2cc17950d

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 3191–3193
www.rsc.org/chemcomm
COMMUNICATION
Enantioselective carbonyl–ene reaction on BINOLate/titanium catalyst
encapsulated in magnetic nanoreactorsw
Xiao Liu,ab Shiyang Bai,ab Yan Yang,ab Bo Li,a Bing Xiao,a Can Lia and Qihua Yang*a
Received 20th December 2011, Accepted 30th January 2012
DOI: 10.1039/c2cc17950d
An asymmetric multicomponent catalyst, BINOLate/titanium,
successfully encapsulated in the nanocages of mesoporous silicas
exhibits much higher activity than the homogeneous counterpart
in quasi solvent-free enantioselective carbonyl–ene reaction,
owing to the confinement effect of the nanoreactor.
proximity of the ligands in the monomer was controlled.13 Ding
and co-workers demonstrated a successful immobilization of
multicomponent metal complexes via a self-supporting strategy
for the heterogeneous asymmetric catalysis.14 Although the
methods mentioned above are efficient, the immobilization of
asymmetric multicomponent catalysts that consist of two or more
chiral ligands and/or metals still remains a difficult task due to the
difficulty in controlling the proper proximity of two or more
catalytic centers needed for the generation of active species on a
solid catalyst.
Asymmetric multicomponent catalysts that consist of two or more
chiral ligands and/or metals have attracted much research attention
in the asymmetric catalytic transformations due to the synergistic
cooperation among different components. For example, the
titanium complex of BINOL (1,1-bis-2-naphthol) and its deriva-
tives are among the most widely used asymmetric multicomponent
catalysts for a wide variety of enantioselective reactions.1–8 For
BINOLate/titanium catalyst, the catalytic species is usually gener-
ated in situ and subtle changes in reaction conditions (ligand-to-Ti
ratio, solvent, addition sequence, and so on) may have a big
influence on the catalytic performance of BINOLate/titanium
catalyst.1,9 This is due to the co-existence of different kinds of
species during the catalytic process as a result of a dynamic mixture
of BINOLate/titanium species equilibrating with each other in
solution.10,11 This makes the usage of asymmetric catalysts ineffi-
cient and the mechanistic study of the asymmetric catalysis difficult.
Immobilization of asymmetric multicomponent catalysts on
solid supports may help to keep most multicomponent complexes
in their active forms by carefully fixing two or more chiral ligands
and/or metals in proper position, which will favor the usage
efficiency of asymmetric catalysts and the research on the cata-
lytic mechanism. Moreover, the transformation of homogeneous
asymmetric catalysts into heterogeneous ones would benefit
the economic and green process for the production of chiral
chemicals.12 Several strategies have been developed for immobili-
zation of the asymmetric multicomponent catalysts. Sasai and
co-workers immobilized bisBINOL ligands on the polymer-
support via a ‘‘catalyst analogue’’ concept, in which the proper
Herein, we report an efficient and facile strategy for the hetero-
genization of the multicomponent metal complex, BINOLate/
titanium, by its encapsulation in the magnetic nanocages of
mesoporous silicas. The confinement effect of the nanoreactor
prohibits the formation of a dynamic mixture of BINOLate/
titanium species. Consequently, BINOLate/titanium encapsulated
in the nanoreactor exhibits higher activity than the homogeneous
counterpart for the quasi solvent-free enantioselective carbonyl–ene
reaction. The comparison results of varying the BINOL-to-Ti
molar ratio in the nanoreactor from 1 to 3 suggest that
Ti(BINOLate)2 species are the active catalysts for the carbonyl–ene
reaction. This work demonstrates a new strategy to realize the
heterogenization of asymmetric multicomponent catalysts and also
a new method for elucidation of active species among a dynamic
mixture evolved in the catalytic process.
Mesoporous silicas, FDU-12 with cage-like pore structure, were
used as nanoreactors for encapsulation of the BINOLate/titanium
complex. With an aim of obtaining solid asymmetric catalysts
which could be easily separated from the reaction system, the
magnetic nanoreactor was prepared by putting Fe3O4 nano-
particles on the surface of the as-synthesized FDU-12. The
as-synthesized FDU-12 containing magnetic Fe3O4 nanoparticles
was carbonized under inert gas conditions and F127 surfactant in
the nanocages could generate a thin layer of carbon species. This
synthetic method can selectively modify the inner surface of the
mesoporous FDU-12, which is important for microenvironment
modification of the mesoporous materials and difficult to be
achieved using other synthetic strategies. The synthetic process
for this bifunctionalized mesoporous materials, FDU-CM (C and
M refer to the thin layer of carbon species and magnetic property,
respectively), is outlined in Scheme 1A (for details of synthesis and
characterization see ESIw, Fig. S1–S4).
a State Key Laboratory of Catalysis, Dalian Institute of Chemical
Physics, Chinese Academy of Sciences, 457 Zhongshan Road,
Dalian 116023, China. E-mail: yangqh@dicp.ac.cn;
Fax: +86-411-84694447
b Graduate School of the Chinese Academy of Sciences,
Beijing 100049, China
w Electronic supplementary information (ESI) available: Experimental
details, nitrogen adsorption isotherms, TEM images, Raman spectrum,
solid state 13C and 29Si NMR, UV-vis spectra, and recyclability of solid
catalyst. See DOI: 10.1039/c2cc17950d
The (R)-6,60-Br2-BINOL–Ti complex (simplified as BnTi,
where n refers to the molar ratio of (R)-6,60-Br2-BINOL-to-Ti)
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 3191–3193 3191

View Article Online
Table 1 The catalytic performance of the solid catalyst with different
ratios of (R)-6,60-Br2-BINOL to Ti(OiPr)4 for the quasi solvent-free
enantioselective carbonyl–ene reactiona
Entry
n
S/C
Yieldb (%)
eec (%)
1
2
1 : 1
2 : 1
3 : 1
2 : 1
2 : 1
1000
1000
1000
5000
10 000
6 (14)
95 (58)
4 (31)
71 (53)
55 (22)
27.3 (43.1)
92.1 (94.7)
34.6 (84.3)
91.7 (94.4)
92.3 (94.6)
3
4d
5d
a
Unless otherwise stated, ethyl glyoxylate was 2.0 molar equivalents to
a-methylstyrene; room temperature; 48 h. Ti contents for solid catalysts
BnTi@FDU-CM with n of 1, 2 and 3 are 0.154, 0.142, 0.130 wt%,
respectively. The values in the parentheses are for homogeneous catalysts.
Scheme 1 Schematic description for synthesis process of FDU-CM
with Fe3O4 nanoparticles on the outer surface and carbon species on
the inner surface of the nanoreactor (A) and encapsulating the chiral
(R)-6,60-Br2-BINOL–Ti complex with different ratios of ligand to
metal in the nanoreactor through the silylation method (B).
b
c
Isolated yields. ee value of the product was determined by HPLC
d
analysis using a Chiralcel OJ column. The reaction time was 96 h.
may contribute to the catalytic activity and enantioselectivity of
homogeneous B1Ti and B3Ti. The fact that B1Ti@FDU-CM
and B3Ti@FDU-CM are almost inactive for the carbonyl–ene
reaction and B2Ti@FDU-CM shows very high activity and ee
under quasi solvent-free conditions suggests that little equili-
brating aggregates or oligomeric species exist in the nanoreactor
and most of the BnTi complexes remain in the form of BnTi.
Therefore, most of the B2Ti complexes in the nanoreactor keep in
their active forms, which results in much higher activity than the
homogenous catalyst. At an S/C ratio as high as 5000, B2Ti@
FDU-CM could smoothly catalyze the reaction with product
yield of 71% with 91.7% ee, while homogeneous B2Ti showed
only 52% product yield with 94.4% ee. Even at an S/C of 10 000,
55% product yield with 92.3% ee could still be obtained with
solid catalysts and the homogeneous catalyst only showed 22%
product yield. The above results show that B2Ti encapsulated in
the nanoreactor is more active than the homogeneous counter-
part. As far as we know, B2Ti@FDU-CM is among the few solid
catalysts that can catalyze the carbonyl–ene reaction at such low
amounts of catalyst usage. This result suggests that the free
transformation of different species of BINOLate/Ti is difficult
in the confined space of the nanoreactor due to the spatial
limitation. The encapsulation of metal complexes in the nano-
reactor may be an efficient method for immobilization of metal
complexes with ligand/metal ratio more than 1.
was encapsulated in the nanocage of FDU-CM and FDU-12
according to a modified method based on our previous reports
using propyltrimethoxysilane (C3) as silylation reagent to
reduce the pore entrance size (Scheme 1B).15,16 Solid catalysts
with the confined molecular complexes are denoted as
BnTi@FDU-CM and BnTi@FDU-12, respectively, using
FDU-CM and FDU-12 as starting materials (for the preparation
details and characterizations of the solid catalysts see the ESIw,
Fig. S5–S7). The characterizations of UV-vis diffuse reflectance
spectroscopy and 13C CP/MAS NMR spectrum confirm that the
homogeneous catalyst is successfully encapsulated in the nano-
reactor (Fig. S8 and S9, ESIw).
The asymmetric carbonyl–ene reaction catalyzed by titanium
complexes of BINOL derivatives is an important approach for
C–C bond formation in organic synthesis.3,17 The catalytic
performance of BnTi@FDU-CM and BnTi with different
(R)-6,60-Br2-BINOL-to-Ti ratios for quasi solvent-free enantio-
selective carbonyl–ene reaction is summarized in Table 1.
B2Ti@FDU-CM could smoothly catalyze the asymmetric
carbonyl–ene reaction with up to 95% yield and 92.1% ee at
an S/C of 1000 (Table 1, entry 2). However, low product yield
and ee were obtained for B1Ti@FDU-CM (6% yield with 27.3%
ee, Table 1, entry 1) and B3Ti@FDU-CM (4% yield with 34.6%
ee, Table 1, entry 3) with (R)-6,60-Br2-BINOL-to-Ti molar ratio
other than 2. This result suggests that the B2Ti complex is the
active catalyst for the carbonyl–ene reaction.3,8 Similar to the
results of solid catalysts, the homogeneous catalyst, B2Ti, showed
the highest product yield and ee among BnTi (Table 1, entry 2). It
is interesting to find that B2Ti@FDU-CM showed much higher
yield (95% versus 58%) than and comparable ee to B2Ti under
similar conditions. However, B1Ti@FDU-CM and B3Ti@
FDU-CM showed lower product yield and ee than the homo-
geneous counterparts (Table 1, entry 1 and 3).
The proper proximity of BINOL ligands can be adjusted by
the loadings of (R)-6,60-Br2-BINOL–Ti complexes in the nano-
reactor. The catalytic performance of B2Ti@FDU-CM with
different metal complex loadings encapsulated in the nanoreactor
is summarized in Table 2 for the carbonyl–ene reaction. The yield
and ee of products increased as the Ti content increased and
reached the maximum at a Ti content of 0.142 wt% (Table 2).
Further increasing the Ti content above 0.142 wt% led to the
decreasing of yield and ee. The ‘‘volume active site density’’ of
the nanoreactor can be used to explain this tendency.16 With the
increase of catalyst loadings in the nanoreactor from 0.064 to
0.211 wt%, the molecular number of the metal complex in each
cage increases from 5 to 17 (Table 2). Consequently, the contact
frequency between substrates and catalysts becomes increased
and the activity of the solid catalyst increases. At high catalyst
loading, the nanoreactor becomes crowded, which makes the
For enantioselective carbonyl–ene reaction, the active species is
the B2Ti complex. The 1H NMR spectrum of BnTi (n = 1, 2, 3) in
C6D6 showed broad bulge-like resonances at the aromatic region
(Fig. S10, ESIw), indicating the existence of a mixture of slowly
equilibrating aggregates or oligomeric species.1 Consequently, the
presence of slowly equilibrating aggregates or oligomeric species
c
3192 Chem. Commun., 2012, 48, 3191–3193
This journal is The Royal Society of Chemistry 2012

View Article Online
Table 2 The catalytic performance of solid catalysts with different
contents of B2Ti for the quasi solvent-free enantioselective carbonyl–ene
reactiona
the activity of filtrate was investigated. The solid catalyst, B2Ti@
FDU-CM, was filtrated out after the product yield reached 23%.
The filtrate was continuously stirred. After 48 h, the product
yield was about 26%. This indicated that the enantioselective
carbonyl–ene reaction was mainly catalyzed by the solid catalyst.
In summary, we have demonstrated the successful encapsu-
lation of asymmetric multicomponent catalysts in the nano-
cage of mesoporous silicas using BINOLate/Ti complexes as a
model catalyst. The solid catalyst exhibits significantly higher
activity than the homogenous catalyst under quasi solvent-free
conditions for enantioselective carbonyl–ene reaction. The
confinement effect of the nanoreactor could effectively suppress
the formation of a dynamic mixture of BINOLate/Ti species and
increase the usage efficiency of the asymmetric catalysts, which
mainly contribute to the high activity of the solid catalyst. It is
found that the solid catalyst can catalyze the enantioselective
carbonyl–ene reaction even at an S/C as high as 10 000. This
work provides a new strategy to realize the heterogenization of
the asymmetric multicomponent catalysts.
Ti content/ Number of catalyst molecules in
each nanocageb
Yieldc eed
Entry wt%
(%)
(%)
1
2
3
4
0.064
0.093
0.142
0.186
0.211
0.137
5.2
7.6
11.6
15.3
17.2
11.2
82
89
95
67
51
79
80.5
87.2
92.1
60.6
46.3
61.3
5
6e
a
S/C = 1000; ethyl glyoxylate is 2.0 molar equivalents to a-methyl-
styrene; no solvent was added; room temperature; 48 h. For calculation
b
c
of number of catalyst molecules in each nanocage see ESI. Isolated
d
e
yields. HPLC analysis using a Chiralcel OJ column. The catalyst is
B2Ti@FDU-12.
diffusion of reactants and products in the nanoreactor difficult
and reduces the activity of the solid catalyst. The ‘‘volume density
of the active site’’ has great influence on the proximity of BINOL
ligands and can change the coordination status between ligand
and metal, which may cause the changes in ee of the solid
catalysts with encapsulation of different B2Ti contents. The
activity and ee of encapsulated molecular catalysts could be
improved by adjusting the ‘‘volume density of the active site’’
in the nanoreactors.
We gratefully acknowledge the financial support of the
National Basic Research Program of China (2009CB623503,
2010CB833300, 2008DFB50130) and National Natural Science
Foundation of China (20920192).
Notes and references
1 H. L. Bao, J. Zhou, Z. Wang, Y. L. Guo, T. P. You and
K. L. Ding, J. Am. Chem. Soc., 2008, 130, 10116.
2 X. Du, Q. Wang, X. He, R. G. Peng, X. Zhang and X. Q Yu,
Tetrahedron: Asymmetry, 2011, 22, 1142.
3 K. Mikami and S. Matsukawa, Nature, 1997, 385, 613.
4 K. Mikami and M. Shimizu, Chem. Rev., 1992, 92, 1021.
5 K. Mikami, M. Terada, T. Korenaga, Y. Matsumoto and
S. Matsukawa, Acc. Chem. Res., 2000, 33, 391.
6 S. Pandiaraju, G. Chen, A. Lough and A. K. Yudin, J. Am. Chem.
Soc., 2001, 123, 3850.
7 M. Turlington, A. M. DeBerardinis and L. Pu, Org. Lett., 2009,
11, 2441.
8 Y. Yuan, X. Zhang and K. L. Ding, Angew. Chem., Int. Ed., 2003,
42, 5478.
9 A. Baeza, C. Najera, J. M. Sansano and J. M. Saa, Tetrahedron:
Asymmetry, 2011, 22, 1282.
In addition, if B2Ti was encapsulated in the nanocage of
FDU-12 without inner surface modification of carbon, only
79% yield with 61.3% ee was obtained in CH2Cl2 (Table 2,
entry 6), which was lower than that of B2Ti@FDU-CM with
almost the same catalyst loading (Table 2, entry 3). This is
probably due to the strong interaction of (R)-6,60-Br2-BINOL–Ti
species with the pore wall of FDU-12. Careful screening of the
modification of the microenvironment of the nanoreactors is
important to acquire high activity and enantioselectivity for the
heterogeneous asymmetric catalysis.
The substrate scope of enantioselective carbonyl–ene reaction
was investigated under optimized conditions using B2Ti@FDU-
CM as a model catalyst (Table S2, ESIw). Higher catalytic activity
and enantioselectivity (95.5%–94.9% ee) were obtained for
F- and Cl-substituted olefins than those for CH3-substituted olefin.
The result exhibits that B2Ti@FDU-CM is a highly efficient
asymmetric catalyst for electron-withdrawing substituents. The
recycling performance of the solid catalysts was conducted by
using B2Ti@FDU-CM as the model catalyst (Fig. S11, ESIw).
After the reaction was complete, the solid catalyst was recovered
easily from the reaction system through magnet and was washed
thoroughly with CH2Cl2. The separated solids were recharged with
toluene and substrates for the next run. As shown in Fig. S11,
the catalyst was used for three cycles in the enantioselective
carbonyl–ene reaction. However, longer time was needed to obtain
higher yield and the enantioselectivity dropped slightly from 92.0%
to 86.4%. In order to determine the heterogeneity of the reaction,
10 J. Balsells, T. J. Davis, P. Carroll and P. J. Walsh, J. Am. Chem.
Soc., 2002, 124, 10336.
11 T. J. Davis, J. Balsells, P. J. Carroll and P. J. Walsh, Org. Lett.,
2001, 3, 699.
12 (a) C. Li, Catal. Rev. Sci. Eng., 2004, 46, 419; (b) J. M. Fraile,
J. I. Garcı
´
(c) A. Corma and H. Garcı
a and J. A. Mayoral, Chem. Rev., 2009, 109, 360;
a, Adv. Synth. Catal., 2006, 348, 1391.
´
13 T. Sekiguti, Y. Iizuka, S. Takizawa, D. Jayaprakash, T. Arai and
H. Sasai, Org. Lett., 2003, 5, 2647.
14 (a) X. S. Wang, X. W. Wang, H. C. Guo, Z. Wang and K. L. Ding,
Chem.–Eur. J., 2005, 11, 4078; (b) H. C. Guo, X. W. Wang and
K. L. Ding, Tetrahedron Lett., 2004, 45, 2009; (c) K. L. Ding,
Z. Wang, X. W. Wang, Y. X. Liang and X. S. Wang, Chem.–Eur.
J., 2006, 12, 5188; (d) Z. Wang, G. Chen and K. L. Ding, Chem.
Rev., 2009, 109, 322.
15 H. Q. Yang, L. Zhang, P. Wang, Q. H. Yang and C. Li, Green
Chem., 2009, 11, 257.
16 H. Q. Yang, L. Zhang, L. Zhong, Q. H. Yang and C. Li, Angew.
Chem., Int. Ed., 2007, 46, 6861.
17 K. Mikami, S. Matsukawa, T. Volk and M. Terada, Angew.
Chem., Int. Ed. Engl., 1997, 36, 2768.
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 3191–3193 3193