Siliceous mesocellular foam-supported aza(bisoxazoline)-copper catalysts

Article abstract of DOI:10.1002/adsc.200700388

Aza(bisoxazoline) was easily immobilized on siliceous mesocellular foam (MCF) in a systematic approach. It was found that silanol precapping, linker group flexibility, ligand loading, and silanol postcapping were important factors to consider in optimizing silica-supported aza(bisoxazoline) catalysts. The optimized MCF-supported aza(bisoxazoline)-copper catalyst offered the same enantioselectivity as its homogeneous counterpart, and excellent recyclability. The heterogenized catalyst showed a much higher chemoselectivity in a short reaction time. The heterogenized aza(bisoxazoline)-copper(I) catalyst was successfully applied to a circulating flow-type packed bed reactor with excellent productivity and enantioselectivity. The circulating flow-type reactor was found to be very useful for gas-generating catalytic reactions such as cyclopropanation reaction.

Full text of DOI:10.1002/adsc.200700388

FULL PAPERS
DOI: 10.1002/adsc.200700388
Siliceous Mesocellular Foam-Supported Aza(bisoxazoline)-
Copper Catalysts
Jaehong Lim,a Siti Nurhanna Riduan,a Su Seong Lee,a,* and Jackie Y. Yinga,*
a
Institute of Bioengineering and Nanotechnology, 31 Biopolis Way, The Nanos, Singapore 138669
Fax : (+65)-6478-9020; e-mail: sslee@ibn.a-star.edu.sg or jyying@ibn.a-star.edu.sg
Received: August 5, 2007; Revised: December 30, 2007; Published online: May 13, 2008
Supporting information for this article is available on the WWW under http://asc.wiley-vch.de/home/.
Abstract: Aza(bisoxazoline) was easily immobilized time. The heterogenized aza(bisoxazoline)-copper(I)
on siliceous mesocellular foam (MCF) in a systemat- catalyst was successfully applied to a circulating
ic approach. It was found that silanol precapping, flow-type packed bed reactor with excellent produc-
linker groupflexibility, ligand loading, and silanol
tivity and enantioselectivity. The circulating flow-
postcapping were important factors to consider in type reactor was found to be very useful for gas-gen-
optimizing silica-supported aza(bisoxazoline) cata- erating catalytic reactions such as cyclopropanation
lysts. The optimized MCF-supported aza(bisoxazo- reaction.
line)-copper catalyst offered the same enantioselec-
tivity as its homogeneous counterpart, and excellent Keywords: asymmetric catalysis; cyclopropanation;
recyclability. The heterogenized catalyst showed a immobilization; packed bed reactor; siliceous meso-
much higher chemoselectivity in a short reaction cellular foam
Introduction
more effective diffusion of substrates. Siliceous meso-
cellular foam (MCF), with its ultralarge pores of
Chiral bisoxazoline-metal complexes are of great >10 nm, is particularly useful for reactions involving
value in various asymmetric catalytic reactions.[1] large substrates.[10]
However, these catalysts require high catalyst-to-sub-
Since chiral bisoxazoline-metal catalysts have
strate ratios to achieve high enantioselectivity and strong interactions with silica surface, most silica-sup-
high turnover frequency (TOF). Thus, there is a great ported chiral bisoxazoline-metal catalysts showed
deal of interest to immobilize these catalysts on vari- lower enantioselectivities in asymmetric catalytic re-
ous types of supports.[2–4]
actions, compared to their homogeneous counterparts.
Although inorganic supports[5] have advantages due There have only been a few reports on silica-support-
to physical strength and chemical inertness, polymer ed chiral bisoxazoline-metal catalysts, which showed
supports[6] are preferred to covalently fixate homoge- poor enantioselectivities, particularly in asymmetric
neous catalysts due to their facile immobilization. cyclopropanation.[3b,f] It was reported that postcapping
However, the polymer support tends to swell in the of the residual silanol groups on the silica support
organic reaction medium. To avoid this issue, high po- with trimethylsilyl (TMS) groups after catalyst immo-
lymer cross-linking can be used, which gives rise to bilization improved the enantioselectivity of the
problem such as brittleness and increased portion of silica-supported catalysts. However, the enantioselec-
embedded catalysts.[7] The embedded catalysts are in- tivity was still lower than that exhibited by the homo-
accessible to substrates, leading to an inefficient system. geneous catalysts.[11] Recently, our groupreported a
Mesoporous silica has attracted substantial interest novel precapping method, which might reduce the in-
as a catalytic support due to its ultrahigh surface area teraction of immobilized bisoxazoline (box) ligands
and well-defined pore structure.[8] The large internal with silica surface during the immobilization step.[12]
surface area associated with the pores could provide a It might lead to more efficient complexation of the
high dispersion of catalytic sites. Some of the resulting box ligand with copper, resulting in a higher enantio-
site-isolated catalysts offered enhanced reactivity.[9] selectivity.
Mesoporous silica with a 3-dimensional pore structure
Aza(bisoxazoline) (azabox) has a nitrogen atom in-
can be an attractive catalytic support by providing stead of a carbon atom at the center of the box. This
Adv. Synth. Catal. 2008, 350, 1295 – 1308
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1295

Jaehong Lim et al.
FULL PAPERS
for fixating bulky complexes, and for catalyzing reac-
tions with large substrates. Conventional MCF has ir-
regular particle morphology and a particle size of tens
of microns. For more effective packed bed applica-
tions, we have recently synthesized spherical MCF mi-
croparticles of ~5 mm.[10f,g,12] Their ultralarge pore size
and 3-dimensional pore structure could minimize the
diffusion limitations, and reduce the back pressure in
a packed bed reactor even at high flow rates.
The spherical MCF microparticles were precapped
nitrogen atom leads to a particularly strong interac- with TMS groups by the reaction with hexamethyldi-
tion between azabox and silica support, and more silazane (HMDS) in toluene. To achieve uniform dis-
flexibility after immobilization. This may make it dif- tribution of TMS groups on the support, MCF parti-
ficult to achieve high enantioselectivity and recyclabil- cles were well dispersed in toluene, and then HMDS
ity with silica-supported azabox. Fraile et al. reported was slowly added to the mixture at room temperature.
that azabox has a stronger binding affinity to Heating the mixture at 608C overnight allowed both
copper.[13] The strong coordination can lead to less TMS groups of HMDS to react with the silanol
leaching of copper from the heterogenized azabox- groups of the support. This partially TMS-precapped
copper catalysts, resulting in better recyclability. Re- MCF was used for immobilization of azabox.
cently, several heterogenized azabox catalysts have
To obtain fully TMS-capped MCF, a vapor-phase
been reported.[4] However, most of them were sup- grafting method was used. An excess amount of
ported on polymers. Covalent immobilization of HMDS was injected into a flask with MCF under a
azabox on silica has not been reported to date.
closed, high vacuum system, and then the flask was
Conventionally, most asymmetric reactions involve heated to vaporize HMDS for reaction with the sila-
batch processing. In contrast, the flow-type reactor nol groups of MCF.
has numerous advantages including more on-stream
time, less maintenance, more consistent product
stream, less solvent wastage, and less catalyst attri- Effect of Silanol Groups
tion/loss.[14] Despite these benefits, there have been
very few accounts of continuous systems for asymmet- Azabox 1 was prepared by the reported method in
ric catalysis.[15] Recently, continuous flow reaction sys- moderate yield.[17] Compound 2 was readily synthe-
tems for asymmetric cyclopropanation have been re- sized by reaction of 1 with methyl iodide (Scheme 1).
ported.[16] Some reactions can be quite complicated To investigate the effects of free silanol groups, homo-
for continuous flow reactor applications due to the geneous azabox-Cu(II) catalyst [2:Cu(II)] was physi-
generation of side-products that are rate-inhibiting cally mixed with bare MCF (2a) and fully TMS-
and/or gaseous in nature. Such side-products must be capped MCF (2b), respectively, in dichloromethane
removed quickly from the reactor to achieve an effi- (Figure 1). When the two mixtures were centrifuged,
cient reaction process.
the precipitates showed that all the copper complexes
Herein, we report the immobilization of azabox on in 2a were adsorbed on bare MCF, whereas the ma-
MCF for asymmetric cyclopropanation. By examining jority of the copper moiety in 2b remained in the so-
the effect of support precapping, catalyst linker lution.
group, ligand loading, and support postcapping, we
Cu(II) was reduced to Cu(I) by phenylhydrazine
were able to significantly improve the enantioselectiv- prior to asymmetric cyclopropanation of styrene. 2a
ity and regioselectivity of silica-supported azabox cat- showed 13% and 21% lower enantioselectivities for
alysts. We also successfully applied the heterogenized trans (12) and cis (13) products, respectively, com-
catalyst to a circulating flow-type packed bed reactor pared to 2b (Table 1). Compound 2a showed similar
to rapidly remove the nitrogen gas that was generated results in the second run with slightly increased enan-
inside the reactor during the catalytic reaction.
tiomeric excess (ee) values. However, 2b gave very
low yield (2%) in the second run, and turned pale
yellow in color. Photoacoustic Fourier-transform in-
frared (PA-FTIR) spectra confirmed that 2a retained
most of the organometallic catalysts after 2 runs, and
2b contained very little organometallic catalysts after
2 runs. Each run was followed by thorough washing
Results and Discussion
Preparation of Catalyst Support
MCF is a stable mesoporous silica with ultralarge cell- with dichloromethane. These results indicated that
like pores (24–42 nm) interconnected by windows of TMS capping successfully reduced the interaction be-
9–22 nm.[10] This novel support material is well-suited tween the catalyst and support surface, leading to
1296
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1295 – 1308

Siliceous Mesocellular Foam-Supported Aza(bisoxazoline)-Copper Catalysts
FULL PAPERS
Scheme 1. Immobilization of azabox on MCF.
Figure 1. Mixing of homogeneous azabox-Cu(II) with calcined bare MCF (2a) and fully TMS-capped MCF (2b). The white
bar is a magnetic stir bar.
Adv. Synth. Catal. 2008, 350, 1295 – 1308
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1297

Jaehong Lim et al.
FULL PAPERS
Table 1. Asymmetric cyclopropanation of styrene over 2a
and 2b after reduction of copper.[a]
on the MCF surface. Starting with 1, the overall yield
of the final postcapped heterogenized azabox 8 or 9
was over 80%.
Complexation of Cu(II) with the immobilized li-
gands was achieved by adding a Cu(II) triflate solu-
tion in THF to a THF suspension of 8 or 9. After stir-
ring for 1 h, the suspension was filtered, washed with
THF thoroughly, and dried under vacuum. Elemental
analysis showed that most of the copper (> 99%) re-
mained on the box-functionalized MCF. Cyclopropa-
nation was performed over Cu(I) complexes, which
were generated in situ by reducing Cu(II) complexes
with phenylhydrazine.
Ligand Run
no.
Yield
[%][b]
12/13
ee [%] of ee [%] of
ratio[c]
12[d]
13[d]
Asymmetric Cyclopropanation of Styrene over MCF-
Supported Azabox-Cu(I) Catalysts with a Propyl
Linker
2a
2b
1
2
1
2
73
67
78
2
64/36
66/34
74/26
71/29
78
82
91
86
66
72
87
81
The immobilized azabox with a flexible propyl linker
groupshowed a great improvement in enantioselec-
tivity and regioselectivity (trans/cis ratio) by using the
partially precapped MCF. While 8a with a catalyst
loading of 0.240 mmolgÀ1 offered 32% ee for the trans
isomer 12 and a trans/cis ratio of 63/37, 9a with a cata-
lyst loading of 0.262 mmolgÀ1 gave 88% ee for the
trans isomer 12 and a trans/cis ratio of 71/29 in the
asymmetric cyclopropanation of styrene (Table 2).
[a]
All reactions were carried out under argon at 258C for
7 h (dropwise addition of EDA for 5 h, followed by stir-
ring for an additional 2 h), styrene/EDA ratio=1.5.
Calculated from a gas chromatography (GC) calibration
curve between n-dodecane and the products.
Determined by GC.
[b]
[c]
[d]
Determined by GC with a Chiraldex-B column.
high enantioselectivity. However, if the support was The enantioselectivity and recyclability of 9a also de-
fully precapped by TMS, then it would only allow for pended largely on the ligand loading density. When
the catalyst to be weakly adsorbed onto its surface; the ligand loading was increased from 0.120 mmolgÀ1
such a weak interaction would result in major catalyst to 0.262 mmolgÀ1, the enantioselectivity increased
leaching. This study suggested that it would be useful considerably from 68% ee to 88% ee for the trans
to have partial precapping of MCF with TMS, fol- isomer 12, and the recyclability improved significantly
lowed by covalent reaction of the catalyst with the re- as well.
sidual silanol groups of the surface-treated MCF.
The low enantioselectivity of 8a with the propyl
linker groupmight be caused by incomlpete TMS
postcapping, followed by incomplete complexation of
the immobilized ligands with the copper sources. This
incomplete TMS postcapping might have stemmed
from the strong interactions between the immobilized
Preparation of Heterogenized Azabox-Cu(I)
Catalysts
Azabox 1 was easily deprotonated with sodium hy- azabox and the support surface, or the large cone
dride in tetrahydrofuran (THF), and the resulting angle of the immobilized azabox ligands with the flex-
anion was reacted with electrophilic trialkoxysilanes ible propyl linker group.
(3-iodopropyltrimethoxysilane and p-(chloromethyl)-
To examine the above postulate, T-silyl-functional-
phenyltrimethoxysilane) to give T-silyl-functionalized ized azabox was first complexed with copper triflate,
1
azabox 5 (Scheme 1). H and 13C nuclear magnetic and then immobilized on bare MCF by stirring in di-
resonance (NMR) spectra confirmed the complete T- chloromethane for 4 days (Scheme 2). Compound 14,
silyl functionalization of 1. Different amounts of prepared by phenylhydrazine reduction, offered
ligand 5 were immobilized onto uncapped bare MCF higher enantioselectivity for the trans-isomer 12 (86%
and partially TMS-precapped MCF (0.8 mmol TMS ee) (Table 3) than the mixture of homogeneous
gÀ1) by heating in toluene to give 6 and 7, respective- 2a:Cu(I) and bare MCF (78% ee) (Table 1). It is note-
ly. The remaining silanol groups were postcapped worthy that the enantioselectivity was much higher
with TMS groups by vapor-phase treatment with than that of 8a:Cu(I) (~30% ee) (Table 2). This im-
HMDS to yield 8 and 9. 13C and 29Si cross-polarization plied that if the immobilized ligand 8a was fully coor-
magic angle spinning (CPMAS) NMR spectra re- dinated with the copper, the heterogenized catalyst
vealed that the azabox ligand 5 was well-immobilized should give a high enantioselectivity of >80% ee for
1298
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1295 – 1308

Siliceous Mesocellular Foam-Supported Aza(bisoxazoline)-Copper Catalysts
Table 2. Asymmetric cyclopropanation of styrene over the MCF-supported azabox-Cu(I) catalysts with a propyl linker.[a]
FULL PAPERS
Ligand
Loading [mmolgÀ1]
Run #
Yield [%][c]
12/13 ratio[d]
ee [%] of 12[e]
ee [%] of 13[e]
2[f]
3
1
1
1
1
2
1
2
1
2
1
2
3
4
5
6
69
82
69
68
40
68
65
69
67
69
81
85
80
83
83
71/29
74/26
71/29
70/30
67/33
62/38
62/38
63/37
63/37
71/29
71/29
71/29
72/28
71/29
71/29
91
93
91
68
40
24
37
32
57
88
91
92
92
92
92
87
90
87
64
38
19
27
27
47
83
87
88
89
89
89
4[f]
9a
0.120
0.176
0.240
0.262
8a[b]
8a[b]
9a
[a]
All reactions were carried out under argon at 258C for 7 h (dropwise addition of EDA for 5 h, followed by stirring for an
additional 2 h), styrene/EDA ratio=1.5.
Styrene/EDA=3.
Calculated from a GC calibration curve between n-dodecane and the products.
Determined by GC.
[b]
[c]
[d]
[e]
[f]
Determined by GC with a Chiraldex-B column.
Taken from ref.[3a]
the trans-isomer, even with the strong interaction be-
Table 3. Asymmetric cyclopropanation of styrene over 14.[a]
tween the catalysts and the support surface. There-
fore, the low enantioselectivity of 8a:Cu(I) might be
caused mostly by the incomplete complexation of the
immobilized ligands with the copper sources.
Run Yield
12/13
ee [%] of
ee [%] of
#
[%][b]
ratio[c]
12[d]
13[d]
To investigate the relationships between the strong
interaction of the ligand with the support surface and
the effective complexation with copper, azabox 2 was
first mixed with bare MCF and fully TMS-capped
MCF, respectively, in dichloromethane, and then the
copper triflate was added to each mixture to give 2c
and 2d (Scheme 3). Compound 2c gave similar enan-
tioselectivity (77% ee for the trans-isomer 12)
(Table 4) as 2a (the mixture of homogeneous catalyst
2:Cu(II) and bare MCF) (78% ee) (Table 1). This im-
plied that the strong interaction between the free
1
2
3
75
79
76
62/38
64/36
65/35
86
87
87
78
78
79
[a]
All reactions were carried out under argon at 258C for
7 h (dropwise addition of EDA for 5 h, followed by stir-
ring for an additional 2 h), styrene/EDA ratio=2.
Calculated from a GC calibration curve between n-do-
decane and the products.
[b]
[c]
[d]
Determined by GC.
Determined by GC with a Chiraldex-B column.
Scheme 2. Immobilization of T-silyl-functionalized azabox-Cu(II) catalyst.
Adv. Synth. Catal. 2008, 350, 1295 – 1308
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1299

Jaehong Lim et al.
FULL PAPERS
the case of 2c. Therefore, it would be critical to
remove the sites on the support surface with which
copper could coordinate in order to enhance the
enantioselectivity.
The precapping process for 9a might facilitate the
postcapping of the silanol groups by reducing interac-
tions between the azabox ligands and the support sur-
face at the anchoring step. Therefore, 9a:Cu(I) gave
higher enantioselectivity than 8a:Cu(I). MCF of 9a
was precapped with 0.8 mmol of TMS gÀ1 prior to the
ligand immobilization. Thus, 9a with a lower ligand
loading would have more possibility for interactions
between the immobilized ligands and the residual sila-
nol groups. High ligand loading could reduce such un-
desired interactions since there would be fewer resid-
Scheme 3. Mixing of the azabox ligand 2 with bare MCF and
fully TMS-capped MCF prior to complexation with the
copper sources, and mixing of the copper sources with bare
MCF and partially precapped MCF without the azabox
ligand 2.
Table 4. Asymmetric cyclopropanation of styrene over 2c, ual silanol groups after the ligand immobilization.
2d, 15a, and 15b after reduction of Cu(II) to Cu(I).[a]
This would also improve the chance for copper to
complex with the immobilized ligands, resulting in en-
Catalyst Run Yield
12/13
ee [%] of ee [%] of
hanced enantioselectivity. Thus, for azabox immobi-
lized by a flexible linker group, high enantioselectivity
and good recyclability could be achieved via partial
precapping of MCF and high ligand loading. 8a:Cu(I)
showed slightly increased enantioselectivity in the
second run probably due to more complexation of the
immobilized ligands with free copper ions staying on
the support surface during the catalytic reaction.
#
[%][b]
ratio[c]
12[d]
13[d]
2c
2d
15a
15b
1
1
1
1
2
3
65
81
58
45
53
49
64/36
74/26
57/43
63/37
63/37
67/33
77
89
-
63
82
-
-
-
-
-
-
-
[a]
All reactions were carried out under argon at 258C for
7 h (dropwise addition of EDA for 5 h, followed by stir-
ring for an additional 2 h), styrene/EDA ratio=1.5.
Calculated from a gas chromatography (GC) calibration
curve between n-decane and the products.
Determined by GC.
Determined by GC with a Chiraldex-B column.
Asymmetric Cyclopropanation of Olefins by MCF-
supported Azabox-Cu(I) Catalysts with a
Methylphenyl Linker
[b]
[c]
[d]
When the catalyst was immobilized through a rigid
methylphenyl linker group, the ligand loading had a
greater effect on enantioselectivity than support pre-
azabox ligand 2 and the support surface did not affect capping. The rigidity of the methylphenyl linker
the complexation of the ligand with copper. In con- groupmight considerably reduce the interaction be-
trast, 2d led to superior yield (81%) and enantioselec- tween the immobilized ligand and MCF surface even
tivity (89% ee for the trans isomer 12), illustrating the at a low ligand loading and in the absence of support
importance of TMS capping of MCF.
precapping. 8b:Cu(I) without precapping offered high
It is also noteworthy that (i) the mixture of Cu(II) enantioselectivity (90% ee for trans-isomer 12) and
and bare MCF (15a) catalyzed the cyclopropanation high trans/cis ratio (69/31) (Table 5). Precapping in-
of styrene with a moderate yield (58%) after reduc- creased the trans/cis ratio from 69/31 to 72/28. And
tion to Cu(I), and (ii) the mixture of Cu(II) and parti- 9b:Cu(I) with precapping and a high ligand loading
ally TMS-capped MCF (0.8 mmol TMS gÀ1) (15b) (0.245 mmolgÀ1) achieved excellent recyclability and
also catalyzed the cyclopropanation of styrene with the same enantioselectivity (93% ee for trans-isomer
slightly higher trans/cis ratio and lower chemoselectiv- 12) (Table 5) as its homogeneous counterpart 3:Cu(I)
ity probably due to the presence of the TMS groups (Table 2). Elemental analysis showed that leaching of
(Table 4). This study showed that free copper ions copper after 8 runs was negligible (~0.30 ppm in run
could be coordinated with the support surface to #1, and ~0.23 ppm on the average for runs #2 to #8),
result in low selectivities. When the azabox ligands which could be attributed to the strong binding of
were anchored on the support surface, their mobility copper to azabox.[13] The small cone angle of the im-
would be restricted to movements within the cone mobilized ligand with a methylphenyl linker or the re-
angle of the immobilized ligand. If copper ions coor- duced interaction of this rigid ligand with the support
dinated very strongly to the support surface through surface might lead to the more effective postcapping
silanol groups, they would not be able to complex of silanol groups, followed by effective complexation
with the immobilized ligands. This was different from of the anchored ligands with the copper sources.
1300
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1295 – 1308

Siliceous Mesocellular Foam-Supported Aza(bisoxazoline)-Copper Catalysts
Table 5. Asymmetric cyclopropanation of styrene over MCF-supported azabox-Cu(I) catalysts with a methylphenyl linker.[a]
FULL PAPERS
Ligand
Loading [mmolgÀ1]
0.148
Run #
Yield [%][c]
12/13 ratio[d]
ee [%] of 12[e]
ee [%] of 13[e]
9b
1
2
1
2
3
4
1
2
3
4
5
6
7
8
55
60
79
80
79
81
83
84
86
86
87
80
80
80
72/28
72/28
69/31
69/31
68/32
67/33
72/28
72/28
72/28
72/28
72/28
72/28
72/28
72/28
84
88
90
90
89
87
93
93
93
93
93
93
93
93
80
84
85
85
83
80
91
91
91
91
91
91
91
91
8b[b]
0.157
9b
0.245
[a]
All reactions were carried out under argon at 258C for 7 h (dropwise addition of EDA for 5 h, followed by stirring for an
additional 2 h), styrene/EDA ratio=1.5.
Styrene/EDA=3.
Calculated from a GC calibration curve between n-dodecane and the products.
Determined by GC.
[b]
[c]
[d]
[e]
Determined by GC with a Chiraldex-B column.
Azabox catalyst was also immobilized on bare the immobilized ligands with copper sources through
MCF by
a
methylphenylethyl linker group the effective postcapping of surface silanol groups.
A
Catalyst 9b:Cu(I) offered high enantioselectivity
species 16:Cu(I) (Scheme 4) gave a higher enantiose- (93% ee) also in the asymmetric cyclopropanation of
lectivity (55% ee for trans-isomer 12) than 8a:Cu(I) 1,1-diphenylethylene (Table 6). Catalyst 9b:Cu(I)
(24% ee) and a lower enantioselectivity than 8b:Cu(I) showed a lower yield than the polymer-supported cat-
(90% ee). The flexibility of the methylphenylethyl alyst 4:Cu(I)[4a] probably due to the lower substrate/
linker lies between the propyl and the methylphenyl ethyl diazoacetate (EDA) ratio (2 instead of 3) and
linkers. This showed that the rigidity of the linker shorter reaction time. However, 9b:Cu(I) offered
groupplayed an important role in reducing the inter-
action between the immobilized ligands and the sup-
port surface, and/or facilitating the complexation of
slightly higher enantioselectivity than 4:Cu(I).
Table 6. Asymmetric cyclopropanation of 1,1-diphenylethy-
lene over 9b:Cu(I).[a]
Ligand Run # 17/11 ratio Yield [%][c] ee [%] of 18[d]
4[b]
9b
1
1
2
3
2
2
78
62
60
90
93
93
[a]
All reactions were carried out under argon at 258C for
7 h (dropwise addition of EDA for 5 h, followed by stir-
ring for an additional 2 h).
[b]
[c]
[d]
Taken from ref.[4a] Total reaction time was 11 h.
Isolated yield.
Scheme 4. MCF-supported azabox with methylphenylethyl
linker group.
Determined by HPLC (Chiralcel OD-H column,
hexane:2-propanol=99.4:0.6).
Adv. Synth. Catal. 2008, 350, 1295 – 1308
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1301

Jaehong Lim et al.
FULL PAPERS
Maximization of Azabox Loading Density on MCF
It is noteworthy that the catalyst 9b:Cu(I) with a
high ligand loading (0.245 mmolgÀ1) achieved high
To maximize the loading of azabox, an excess of T- chemoselectivity even when the reaction time was re-
silyl-functionalized azabox was reacted with dried duced to 75 min (Table 7). This could lead to superior
MCF without precapping. 0.507 mmolgÀ1 of ligand productivity. In contrast, the homogeneous counter-
loading was achieved (Scheme 5), which translated to part 3:Cu(I) showed significantly lower chemoselec-
0.641 mmol of azabox being anchored onto 1.0 g of tivity over 75 min of reaction time (Table 7). This il-
bare MCF. Therefore, azabox ligands were located lustrated a further benefit of the covalent immobiliza-
very close to each other, and might be suffering from tion of azabox on MCF.
steric hindrance. The steric effect led to lower chemo-
selectivity and yield as the reaction time was reduced.
For example, when the reaction time was decreased Circulating Flow-Type Reactor
from 420 min to 180 min, 19:Cu(I) showed a large re-
duction in yield from 81% to 61%, although EDA The optimal catalyst 9b (ligand loading=
was almost consumed (Table 7). More by-products 0.245 mmolgÀ1) was initially applied to a continuous
(such as diethyl maleate and fumarate, and their olig- flow reactor [Figure 2, a)]. 9b:Cu(OTf)2 microparti-
G
omers derived from carbene dimerization) were gen- cles were packed into a small empty HPLC column
erated when the chemoselectivity was decreased. The (50 mm”4.6 mm I.D.) by a commercial slurry packer,
lower chemoselectivity at a shorter reaction time and Cu(II) was readily reduced to Cu(I) by flowing
could be attributed to the high packing density of phenylhydrazine solution in CH2Cl2 through the
bulky azabox catalysts on the support surface.
HPLC system. After washing away the excess phenyl-
hydrazine by flowing CH2Cl2, EDA was diluted in
CH2Cl2 (0.33M), and continuously mixed with an
equal volume of styrene solution (0.67M) in CH2Cl2
by the HPLC mixer. The resulting solution was slowly
supplied into the packed bed reactor at a flow rate of
0.2 mLminÀ1 by using the HPLC system to obtain full
conversion at the initial reaction time. This generated
a large amount of nitrogen gas as a by-product, which
was trapped within the packed bed reactor system for
a relatively long time at the low flow rate before elu-
tion. Both conversion and enantioselectivity de-
creased steadily with reaction time. Although the con-
Scheme 5. Maximization of azabox loading density on MCF. version and enantioselectivity were slightly improved
Table 7. Asymmetric cyclopropanation of styrene over MCF-supported azabox-Cu(I).[a]
Ligand
Run #
Reaction time [min]
Yield [%][b]
12/13 ratio[c]
ee [%] of 12[d]
ee [%] of 13[d]
19
1
2
3
1
2
1
2
3
4
5
6
7
1
1
420
420
420
180
180
75
75
75
35
35
81
77
82
61
63
82
83
84
79
77
61
54
52
35
69/31
68/32
68/32
70/30
70/30
72/28
72/28
72/28
72/28
72/28
72/28
72/28
74/26
74/26
94
94
94
93
94
94
94
94
94
94
94
94
94
94
90
90
90
91
91
90
90
90
90
90
90
90
91
91
19
9b
15
15
75
35
3
3
[a]
All reactions were carried out under argon at 258C for 7 h (dropwise addition of EDA for 5 h, followed by stirring for an
additional 2 h), styrene/EDA ratio=2.
Calculated from a gas chromatography (GC) calibration curve between n-dodecane and the products.
Determined by GC.
[b]
[c]
[d]
Determined by GC with a Chiraldex-B column.
1302
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1295 – 1308

Siliceous Mesocellular Foam-Supported Aza(bisoxazoline)-Copper Catalysts
FULL PAPERS
Figure 3. Asymmetric cyclopropanation of styrene by the
circulating flow-type reactor (50 mm”4.6 mm I.D.) with
~
9b:Cu(I) (total amount of catalyst: 230 mg). ( ) % ee for
!
*
the trans isomer, ( ) % ee for the cis isomer, ( ) % trans
&
isomers, and ( ) yield. Total amount of reaction medium: 18
ml for runs #1 to #10, 14 mL for runs #11 to #20.
When the total volume of solvent decreased from
18 mL to 14 mL at the start of run #11, the yield was
increased by ~10%. The reaction time of one cycle
was reduced to 75 min for higher productivity; it was
5.6 times shorter than the typical reaction time in a
batch reactor (420 min). TOF of the reaction (mmol of
cyclopropanation products/mmol copper/h) was almost
68, which was 15 times higher than that by pybox-Ru
monolithic mini-flow reactors, and similar to that by
PhGli-box-copper monolithic mini-flow reactors using
supercritical solvents.[16] However, the enantioselectivi-
ty achieved with our circulating flow-type reactor was
much higher than that by the box-copper monolithic
mini-flow reactors (59% ee for the trans isomer). In
Figure 2. Experimental set-up for (a) the continuous flow re-
actor, and (b) the circulating flow-type reactor using a
HPLC pump system (inset: high magnification SEM micro-
graph of spherical MCF).
with an increased amount of styrene (upto 10 equiva-
lents of EDA), those values also decreased with reac- fact, similar enantioselectivity was attained by
tion time. The nitrogen gas generated inside the 9b:Cu(I) in the circulating flow-type reactor as in the
packed bed reactor might have greatly damaged the batch reactor (Table 7). The circulating flow-type reac-
efficiency of the catalytic system. These results did tor could be regarded as a hybrid between a batch re-
not agree with that for a monolithic mini-flow reactor actor and a continuous flow reactor. It retained the
with immobilized PhGli-box-Cu
(OTf)2,[16a] which did benefits of a continuous flow reactor (such as more on-
G
not show any noticeable changes after 5 h of continu- stream time and less catalyst attrition/loss), as well as
ous run. This might be due to the differences in sup- the benefits of a batch reactor (such as quick removal
port materials and box ligands employed.
of gaseous byproducts).
In the modified reactor system, a styrene solution in
The ultralarge p ore size of MCF microp articles p ro-
CH2Cl2 was circulated through the packed bed reactor vided the packed bed system with a relatively low back
from a reservoir at a high flow rate (5 mLminÀ1), and pressure. Scale-up of the reactor system could be opti-
EDA was added slowly into the circulating solution mized by increasing the particle size and/or pore size
[Figure 2, b)]. The high flow rate was applied to of the support to further reduce the back pressure.
remove the nitrogen byproduct quickly from the
packed bed reactor, giving a slightly high back pressure
(upto 4.48 MPa). This circulating flow-tyep reactor
showed excellent performance, attaining the same
Asymmetric Cyclopropanation of 2-Methylpropene
enantioselectivity (93% ee for the trans-isomer) and The circulating packed bed reactor did not work in
high yield (upto 80%) during 20 cycles (Figure 3). the asymmetric cyclopropanation of 2-methylpropene
Adv. Synth. Catal. 2008, 350, 1295 – 1308
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1303

Jaehong Lim et al.
FULL PAPERS
because the HPLC pump was not appropriate for
A fresh 9b:Cu(I) catalyst achieved 92% ee and
eluting such a volatile substrate. Instead, the catalyst 90% yield (Table 9) under the reaction conditions re-
collected from the previous packed bed reactor study ported in the literature (0.1 mol%, 19 h, 08C). With a
was used to catalyze the cyclopropanation of 2-meth- shortened reaction time and an increased reaction
ylpropene. The reused 9b:Cu(I) catalyst (0.2 mol%) temperature, comparably high yield and high enantio-
showed excellent enantioselectivity (91% ee) and high selectivity were still obtained.
yield (90%) in a shorter reaction time (3 h) and at a
higher reaction temperature (108C) (Table 8), com-
pared to the reaction conditions from the literature Conclusions
(0.1 mol% homogeneous bisoxazoline catalysts, 19 h,
08C; 0.2 mol% heterogeneous bisoxazoline catalyst, Azabox was successfully immobilized on MCF. It was
22 h, 08C). Elemental analysis showed that the leach- found that support precapping, linker group flexibili-
ing of copper after the catalytic reaction (TON= ty, ligand loading, and support postcapping were im-
3,500) was ~8% of the initial copper content.
portant factors to consider in achieving an excellent
heterogenized catalyst. The optimal MCF-supported
azabox-Cu(I) catalyst offered as high enantioselectivi-
ty and yield as the homogeneous counterpart, and ex-
cellent recyclability. This study demonstrated that
silica-supported catalysts could achieve comparable
enantioselectivity and yield as their homogeneous
counterparts through proper design. In addition, the
heterogenized catalyst with the optimal ligand loading
density could attain superior chemoselectivity in a
short reaction time with increased productivity. The
optimal heterogenized catalyst was successfully ap-
plied to a circulating flow-type packed bed reactor,
while retaining the attractive enantioselectivity, yield
and recyclability. The circulating flow-type reactor
was more suitable than the conventional continuous
flow reactor for gas-generating processes, such as the
cyclopropanation reaction.
Table 8. Asymmetric cyclopropanation of 2-methylpropene[a]
over 9b:Cu(I) recovered from the packed bed reactor after
20 cycles.
Run 20/11
Chemoselectivity Yield
[%]
ee [%] of
#
ratio
[%][b]
21[c]
1
2
3
7
7
7
98
98
97
90
89
89
91
90
90
[a]
All reactions were carried out under argon for 3 h (drop-
wise addition of EDA at 108C for 2 h, followed by warm-
ing upto room temperature over 1 h).
Isolated yield.
Determined by GC (Chiraldex G-TA, 50 m”0.25 mm
I.D.).
Experimental Section
[b]
[c]
Spherical MCF was synthesized according to the literature
procedure.[12] Other chemicals were purchased from com-
mercial suppliers, and were used without further purifica-
tion. l-tert-Leucine, anhydrous THF, toluene, anhydrous
CH2Cl2, styrene, phenylhydrazine and EDA were purchased
from Aldrich. Azabox was prepared according to the litera-
ture procedure.[16] 3-Bromopropyltrimethoxysilane, p-(chloro-
methyl)phenyltrimethoxysilane and [(chloromethyl)phenyl-
ethyl]trimethoxysilane (para- and meta-isomers) were pur-
chased from Gelest Inc. PA-FTIR spectra were recorded on
a Bio-rad FTS-60 A spectrometer with an MTEC Model 300
Table 9. Asymmetric cyclopropanation of 2-methylpropene
by fresh 9b:Cu(I) (0.1 mol%).
Run #
20/11 ratio
Yield [%][c]
ee [%] of 21[e]
1[a]
2[a]
3[a]
4[b]
5[b]
7
7
7
7
7
90
92
92
91
91
91
86[d]
86[d]
86[d]
86[d]
1
photoacoustic cell. H and 13C NMR spectra were recorded
on a 400 MHz Bruker spectrometer at ambient temperature.
13C and 29Si CPMAS NMR spectra were recorded on a
400 MHz Bruker spectrometer. Elemental analyses were
performed with a CE440 CHN Analyzer (Exeter Analyti-
cal).
[a]
Dropwise addition of EDA at 08C for 5 h, followed by
warming upto room temperature over 14 h.
Dropwise addition of EDA at 58C for 2 h, followed by
warming upand stirring at 10 8C for 1.5 h, and warming
upto room temperature over 0.5 h.
Isolated yield.
Overall yield from runs #2 to #5.
Determined by GC (Chiraldex G-TA, 50 m”0.25 mm
I.D.).
[b]
Preparation of Precapped MCF
[c]
[d]
[e]
MCF (5.0 g) was dried under vacuum at 1208C overnight
before use, and then well dispersed in toluene (60 mL).
HMDS (422 mL, 2.0 mmol) was slowly added into the sus-
pension. The mixture was stirred for 1 h at room tempera-
1304
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1295 – 1308

Siliceous Mesocellular Foam-Supported Aza(bisoxazoline)-Copper Catalysts
FULL PAPERS
ture, and heated at 608C overnight. The suspension was fil-
tered, washed with toluene, and dried under vacuum. Ele-
mental analysis (%): C 2.86, H 0.94; loading of TMS:
0.794 mmolgÀ1.
With 2.5 g of MCF: elemental analysis (%) of 6a: C 5.57,
H 0.99, N 1.08; loading of 1: 0.257 mmolgÀ1. With 3.0 g of
MCF: elemental analysis (%) of 6a: C 3.71, H 0.82, N 0.82;
loading of 1: 0.195 mmolgÀ1.
Preparation of Fully TMS-Capped MCF
Preparation of 6b
MCF (2.0 g) was placed in a flask, and dried under vacuum
at 1208C overnight. Excess HMDS (1.0 mL) was added to
the flask under the closed vacuum conditions. The flask was
cooled with liquid N2 under high vacuum. It was sealed and
then warmed to room temperature. The flask was then
placed in the oven at 758C for 5 h. After reaction, excess
HMDS was removed under vacuum to yield fully TMS-
capped MCF. Elemental analysis (%): C 5.35, H 1.35; load-
ing of TMS: 1.486 mmolgÀ1.
The same procedure described for 6a was followed, except
that
p-(chloromethyl)phenyltrimethoxysilane
(162 mL,
0.75 mmol) was used instead of 3-bromopropyltrimethoxysi-
lane. 1H NMR (CDCl3) of 5b: d=7.58 (d, J=8.0 Hz, 2H,
Ph), 7.44 (d, J=8.0 Hz, 2H, Ph), 5.18 (d, J=14.9 Hz, 1H,
CH2PhSi), 4.96 (d, J=14.9 Hz, 1H, CH2PhSi), 4.31 (dd, J=
9.5, 8.4 Hz, 2H, CH2CH), 4.19 (dd, J=8.4, 6.8 Hz, 2H,
CH2CH), 3.76 (dd, J=9.5, 6.8 Hz, CH2CH), 3.59 (s, 9H, Si-
OCH3), 0.75 [s, 18H, C
(CH3)3]; 13C NMR (CDCl3) of 5b:
G
d=157.3, 140.7, 134.9, 134.7, 127.8, 73.6, 70.4, 53.5, 50.9,
34.1, 25.8; PA-FT-IR of 6b: n=2960, 1666, 1090, 845, 806,
758 cmÀ1; elemental analysis (%) of 6b: C 3.71, H 0.82, N
0.82; loading of 1: 0.195 mmolgÀ1.
Preparation of 3
THF (5 mL) was added to NaH (22 mg, 0.92 mmol) and
azabox 1 (200 mg, 0.75 mmol). The resulting solution was
stirred for 6 h at 608C to give a clear solution. Benzyl chlo-
ride (103 mL, 0.9 mmol) was added and stirred at 608C over-
night, giving a white precipitate (NaCl). After stirring over-
night, aqueous NH4Cl was then added to the reaction mix-
ture, and the aqueous layer was extracted with ethyl acetate.
The combined organic extracts were washed with brine, and
then dried over MgSO4, filtered, and rotary-evaporated to
give the crude product. The crude product was purified by
flash column chromatography to give product 3; yield:
247 mg (92%); [a]23: À8.23 (c 1.05, CHCl3); 1H NMR
(CDCl3): d=7.41 (dD, J=7.2 Hz, 2H, Ph), 7.28 (td, J=7.2,
1.2 Hz, 2H, Ph), 7.21 (td, J=7.2, 1.2 Hz, 1H, Ph), 5.14 (d,
J=15.2 Hz, 1H, CH2PhSi), 4.96 (d, J=15.2 Hz, 1H,
CH2PhSi), 4.31 (td, J=9.2, 1.2 Hz, 2H, CH2CH), 4.20 (td,
J=6.4, 1.2 Hz, 2H, CH2CH), 3.78 (ddd, J=9.2, 6.4, 1.2 Hz,
Preparation of 7a
The same procedure described for 6a was followed, except
that 2.0 g or 4.0 g of TMS-precapped MCF was used as the
support. With 2.0 g of TMS-precapped MCF: PA-FT-IR: n=
2960, 2912, 2876, 2852, 1701, 1656, 1540, 1076, 811 cmÀ1; ele-
mental analysis (%): C 7.45, H 1.50, N 1.15; loading of 1:
0.274 mmolgÀ1.
With 4.0 g of TMS-precapped MCF: PA-FT-IR: (cmÀ1)
n=2960, 2858, 1708, 1662, 1535, 1072, 811 cmÀ1; elemental
analysis (%):
C 5.85, H 1.10, N 0.61; loading of 1:
0.145 mmolgÀ1.
Preparation of 7b
2H, CH2CH), 0.78 [s, 18H, C
(CH3)3]; 13C NMR (CDCl3):
A
The same procedure described for 6b was followed, except
that 2.0 g or 4.0 g of TMS-precapped MCF (0.8 mmol TMS
gÀ1) was used as the support. With 2.0 g of TMS-precapped
MCF: PA-FT-IR: n=3080, 2960, 2912, 2875, 2851, 1701,
1653, 1478, 1080, 811 cmÀ1; elemental analysis (%): C 9.07,
H 1.58, N 1.09; loading of 1: 0.260 mmolgÀ1.
d=157.2, 137.8, 128.1, 127.1, 73.4, 70.1, 53.2, 33.9, 25.4; ele-
mental analysis: calcd. (%) for C21H31N3O2: C 70.55, H 8.74,
N 11.75; found: C 70.50, H 8.87, N 11.54.
Preparation of 6a
With 4.0 g of TMS-precapped MCF: PA-FTIR: n=3079,
THF (5 mL) was added to NaH (22 mg, 0.92 mmol) and
azabox 1 (200 mg, 0.75 mmol). The resulting solution was
stirred for 6 h at 608C to give a clear solution. 3-Bromopro-
pyltrimethoxysilane (141 mL, 0.75 mmol) was added and
stirred at 608C overnight. White precipitates (NaBr) were
formed and removed using centrifugation. The clear solution
portion was added to MCF (2.5 g or 3.0 g) in toluene
(40 mL), and heated to 1208C overnight. The suspension
was filtered through a filter funnel, and washed with toluene
(20 mL”3), CH2Cl2 (20 mL”3), water (20 mL”10), metha-
nol (20 mL”3) and CH2Cl2 (20 mL”3). After drying in
vacuum, the desired product (6a) was obtained. 1H NMR
(CD2Cl2) of 5a: d=4.25 (ddd, 2H, J=9.6, 8.8, 1.2 Hz,
CH2CH), 4.16 (ddd, 2H, J=8.8, 7.2, 2.4 Hz, CH2CH), 3.75
(m, 4H, CH2CH and CH2CH2CH2Si), 3.53 (s, 9H, Si-OCH3),
2959, 2852, 1701, 1665, 1523, 1077, 812 cmÀ1; elemental anal-
ysis (%):
C 7.26, H 1.33, N 0.68; loading of 1:
0.162 mmolgÀ1.
Preparation of 8a
Catalyst 6a (1.0 g) was dried at 808C for 1 day. Excess
HMDS (0.75 mL) was added to the solid under vacuum.
The flask was cooled under vacuum with liquid N2. It was
sealed and then warmed to room temperature. The flask
was then placed in the oven at 758C for 5 h. After reaction,
excess HMDS was removed under vacuum to yield the de-
sired product 8a.
With 2.5 g of MCF: elemental analysis (%): C 7.83, H
1.78 (m, 2H, CH2CH2CH2Si), 0.86 [s, 18H, C(CH3)3], 0.61 (t,
A
1.33, N 1.01; loading of 1: 0.240 mmolgÀ1. With 3.0 g of
2H, J=8.4 Hz, CH2CH2CH2Si); 13C NMR (CD2Cl2) of 5a:
d=157.6, 75.0, 69.8, 53.5, 34.2, 25.4, 21.4, 7.9, 6.5; PA-FT-IR
of 6a: n=2960, 1666, 1090, 845, 806, 758 cmÀ1.
MCF: PA-FT-IR: n=2961, 1664, 1087, 843, 807, 758 cmÀ1
;
elemental analysis (%): C 6.69, H 1.22, N 0.74; loading of 1:
0.176 mmolgÀ1.
Adv. Synth. Catal. 2008, 350, 1295 – 1308
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1305

Jaehong Lim et al.
FULL PAPERS
lane. 1H NMR (CD2Cl2) of T-silyl-functionalized azabox
(major para isomer): d=7.2 (m, 4H, Ph), 5.05 (d, 1H, J=
14.8 Hz, CH2PhCH2CH2Si), 4.98 (d, 1H, J=14.8 Hz,
CH2PhCH2CH2Si), 4.27 (m, 2H, CH2CH), 4.18 (m, 2H,
CH2CH), 3.76 (ddd, 2H, J=9.2, 6.8, 2.0 Hz, CH2CH), 3.54
(s, 9H, Si-OCH3), 2.67 (m, 2H, CH2PhCH2CH2Si), 0.94 (m,
Preparation of 8b
The procedure described above was followed using 6b. PA-
FT-IR: n=2960, 2929, 2859, 1709, 1664, 1089, 844, 801,
760 cmÀ1; elemental analysis (%): C 10.35, H 1.60, N 0.74;
loading of 1: 0.176 mmolgÀ1.
2H, CH2PhCH2CH2Si), 0.79 [s, 18H, C
(CH3)3]; 13C NMR
A
Preparation of 9a
(CD2Cl2): d=158, 144, 139, 128, 127, 73, 70, 34, 29, 25, 11;
PA-FT-IR of MCF-supported azabox: n=2960, 1666, 1090,
845, 806, 758 cmÀ1.
The procedure described for 8a was followed using 7a. With
2.0 g of TMS-precapped MCF: 13C CPMAS NMR
(100 MHz): d=162.5, 73.5, 67.9, 44.5, 33.0, 23.6, 8.8, À0.8;
With 3.0 g of MCF: elemental analysis (%): C 3.71, H
0.82, N 0.82; loading of 1: 0.195 mmolgÀ1.
29Si CPMAS NMR (79.5 MHz): d=13.5 [Si
A
Post-capping: The procedure described for 8a was fol-
lowed using the MCF-supported azabox. PA-FT-IR: n=
3728, 2963, 1706, 1660, 1084, 796 cmÀ1; elemental analysis
(%): C 9.58, H 1.37, N 0.76; loading of 1: 0.181 mmolgÀ1.
(T1), À57.5 (T2), À65.2 (T3), À101.5 (Q3), À108.6 (Q4); PA-
FT-IR: n=2959, 2908, 2874, 2852, 1670, 1521, 1081, 845, 811,
758 cmÀ1); elemental analysis (%): C 9.49, H 1.96, N 1.10;
loading of 1: 0.262 mmolgÀ1.
With 4.0 g of TMS-precapped MCF: PA-FT-IR: n=2959,
2907, 2852, 1672, 1523, 1076, 846, 810, 758 cmÀ1; elemental
analysis (%): C 7.79, H 1.53, N 0.53; loading of 1:
0.126 mmolgÀ1.
Maximizing Azabox Loading on MCF (19)
MCF (2.0 g) was dried under vacuum at 1208C overnight,
and then well dispersed in toluene. The excess of the T-silyl-
functionalized azabox was added to the reaction mixture,
and then heated at 1208C overnight. The suspension was fil-
tered through a filter funnel, and washed with toluene
(20 mL”3), CH2Cl2 (20 mL”3), water (20 mL”10), metha-
nol (20 mL”3) and CH2Cl2 (20 mL”3). After drying in
vacuum at room temperature, the desired product was dried
in vacuum at 808C. The remaining silanol groups were
capped by the reaction with an excess of HMDS in the
vapor phase. The excess HMDS was removed in high
vacuum to give the final product (19). 13C CPMAS NMR
(100 MHz): d=157.3 (C=N), 140.4, 133.9, 127.2, 73.6, 68.5,
49.8, 32.9, 23.7, À0.2; 29Si CPMAS NMR (79.5 MHz): d=
Preparation of 9b
The procedure described for 8a was followed using 7b. With
2.0 g of TMS-precapped MCF: 13C CPMAS NMR
(100 MHz): d=159.5 (C=N), 142.6, 136.0, 129.2, 75.6, 70.4,
51.9, 34.9, 25.7, 1.7; 29Si CPMAS NMR (79.5 MHz): d=14.6
[Si
C
1648, 1479, 1395, 1079, 846, 812, 758 cmÀ1; elemental analy-
sis (%):
C 11.21, H 1.94, N 1.03; loading of 1:
0.245 mmolgÀ1.
With 4.0 g of TMS-precapped MCF: PA-FT-IR: n=3077,
2960, 2906, 2851, 1683, 1519, 1077, 840, 811, 758 cmÀ1; ele-
mental analysis (%): C 8.93, H 1.69, N 0.62; loading of 1:
0.148 mmolgÀ1.
13.8 (Si
C
Preparation of 14
THF (5 mL) was added to NaH (15 mg, 0.62 mmol) and
azabox 1 (150 mg, 0.56 mmol). The resulting solution was
stirred for 6 h at 608C to give a clear solution. 3-Bromopro-
pyltrimethoxysilane (105 mL, 0.56 mmol) was added and
stirred overnight at 608C. White precipitates (NaBr) were
formed and removed using centrifugation. THF was evapo-
Typical Asymmetric Cyclopropanation
Cu(OTf)2 (3.6 mg, 0.01 mmol) was dissolved in THF (2 mL),
A
and then added to the MCF-immobilized bisoxazolines
(0.022 mmol) well-dispersed in THF (2 mL). The mixture
was stirred for 5 h, and filtered. The powder was thoroughly
washed with THF, and then dried under vacuum. The cata-
lyst was dispersed in CH2Cl2 (4 mL), and phenylhydrazine
(25 mL of a 5% solution) was added. After addition of sty-
rene (172 mL, 1.5 mmol), a solution of EDA (1.0 mmol, di-
luted with 2 mL of CH2Cl2) was introduced dropwise over
5 h using a syringe pump. The mixture was stirred for 2 h,
and centrifuged. The solution portion was collected, and the
trans/cis ratio and yield were determined by GC. The ee was
determined by GC using a Chiraldex-B column. The precipi-
tate was washed with CH2Cl2 (10 mL), and centrifuged 3
times. The recovered catalyst was reused directly for the
next experiment.
rated in vacuum and CH2Cl2 was added. Cu(OTf)2 was in-
A
troduced to the dichloromethane solution to give a green so-
lution. The resulting solution was stirred for 1 day at room
temperature, and was added to MCF (2.5 g) in dichlorome-
thane. The suspension was stirred for 4 days at room tem-
perature, filtered through a filter funnel, and washed with
dichloromethane (20 mL ” 5). After drying in vacuum, the
desired product (14) was obtained. PA-FT-IR: n=3741,
2966, 1676, 1627, 1088, 809 cmÀ1; elemental analysis (%): C
4.34, H 0.80, N 0.72, Cu 0.42; loading of Cu complexes:
0.066 mmolgÀ1.
Preparation of 16
Circulating Flow-Type Reactor
The same procedure described for 6a was followed to give
MCF-supported azabox, except that ((chloromethyl)phenyl-
ethyl)trimethoxysilane (para- and meta-isomers) (282 mL,
1.12 mmol) was used instead of 3-bromopropyltrimethoxysi-
Catalyst 9b:Cu(OTf)2 (230 mg, 0.03 mmol) was packed into
A
an empty HPLC column (50 mm”4.6 mm I.D.). A pump
system of HPLC was used to flow a solution into the
1306
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1295 – 1308

Siliceous Mesocellular Foam-Supported Aza(bisoxazoline)-Copper Catalysts
FULL PAPERS
column. Phenylhydrazine (3.5 mL, 0.036 mmol) solution in
CH2Cl2 (5 mL) was passed through the column to reduce
Cu(II) to Cu(I), and the column was washed by flowing
CH2Cl2. Styrene (688 mL, 6.0 mmol) solution in CH2Cl2
(9 mL) was flowed from a reservoir flask to the column at
5 mLminÀ1, and then EDA (315 mL, 3.0 mmol) and dodec-
ane (342 mL, 1.5 mmol) in CH2Cl2 (9 mL) were added into
the reservoir flask over 55 min. The solution of the reservoir
was further circulated for 20 min to give a clear solution,
which was then collected at the end of the column. The
yield and trans/cis ratio were determined by GC. The ee was
determined by GC using a Chiraldex-B column. The column
was washed by flowing CH2Cl2 (15 mL), and then a further
cycle of the reaction was performed with the same proce-
dure without further reduction of the catalyst with phenyl-
hydrazine. From run #11, the amount of CH2Cl2 for dilution
of styrene was reduced from 9 mL to 5 mL.
[6] a) N. E. Leadbeater, M. Marco, Chem. Rev. 2002, 102,
3217–3274; b) C. A. McNamara, M. J. Dixon, M. Brad-
ley, Chem. Rev. 2002, 102, 3275–3300; c) D. Bergbreit-
er, Chem. Rev. 2002, 102, 3345–3384; d) M. Benaglia,
A. Puglisi, F. Cozzi, Chem. Rev. 2003, 103, 3401–3429.
[7] A. Mandoli, S. Orlandi, D. Pini, P. Salvadori, Chem.
Commun. 2003, 2466–2467.
[8] a) J. Y. Ying, C. P. Mehnert, M. S. Wong, Angew. Chem.
1999, 111, 58–82; Angew. Chem. Int. Ed. 1999, 38, 56–
77; b) E. D. V. Dirk, M. Dams, B. F. Sels, P. A. Jacobs,
Chem. Rev. 2002, 102, 3615–3640; c) C. Li, Catal. Rev.
2004, 46, 419–492; d) A. Tafuchi, F. Schüth, Micropor.
Mesopor. Mater. 2005, 77, 1–45; e) X. S. Zhao, X. Y.
Bao, W. Guo, F. Y. Lee, Materials Today 2006, 9, 32–
39.
[9] a) J. M. Thomas, R. Raja, D. W. Lewis, Angew. Chem.
2005, 117, 6614–6641; Angew. Chem. Int. Ed. 2005, 44,
6456–6482; b) J. Jarupatrakorn, T. D. Tilley, J. Am.
Chem. Soc. 2002, 124, 8380–8388; c) C. Nozaki, C. G.
Lugmair, A. T. Bell, T. D. Tilley, J. Am. Chem. Soc.
2002, 124, 13194–13203.
Acknowledgements
[10] a) D. Zhao, J. Feng, Q. Huo, N. Melosh, G. H. Fredrick-
son, B. F. Chmelka, G. D. Stucky, Science 1998, 279,
548–552; b) P. Schmidt-Winkel, W. W. Lukens, Jr., D.
Zhao, P. Yang, B. F. Chmelka, G. D. Stucky, J. Am.
Chem. Soc. 1999, 121, 254–255; c) P. Schmidt-Winkel,
W. W. Lukens, Jr., P. Yang, J. S. Lettow, J. Y. Ying,
G. D. Stucky, Chem. Mater. 2000, 12, 686–696; d) J. S.
Lettow, Y. J. Han, P. Schmidt-Winkel, P. Yang, D. Zhao,
G. D. Stucky, J. Y. Ying, Langmuir 2000, 16, 8291–
8295; e) J. S. Lettow, T. M. Lancaster, C. J. Glinka, J. Y.
Ying; Langmuir 2005, 21, 5738–5746; f) Y. Han, S. S.
Lee, J. Y. Ying, Chem. Mater. 2006 18, 643–649; g) Y.
Han, S. S. Lee, J. Y. Ying, Chem. Mater. 2007, 19, 2292–
2298.
[11] a) D. Rechavi, M. Lemaire, Org. Lett. 2001, 3, 2493–
2496; b) D. Rechavi, M. Lemaire, J. Mol. Catal. A 2002,
182–183, 239–247; c) D. Rechavi, B. Albela, B. Lau-
rent, M. Lemaire, Tetrahedron 2005, 61, 6976–6981;
d) T. M. Lancaster, S. S. Lee, J. Y. Ying, Chem.
Commun. 2005, 3577–3579; e) S. S. Lee, J. Y. Ying, J.
Mol. Catal. A: Chem. 2006, 256, 219–224.
This work was funded by the Institute of Bioengineering and
Nanotechnology (Biomedical Research Council, Agency for
Science, Technology and Research, Singapore).
References
[1] a) A. K. Gosh, P. Mathivanan, J. Capiello, Tetrahedron:
Asymmetry 1998, 9, 1–45; b) A. Pfaltz, Synlett 1999,
835–842; c) J. S. Johnson, D. A. Evans, Acc. Chem. Res.
2000, 33, 325–335; d) F. Fache, E. Schulz, M. L. Tom-
masino, M. Lemaire, Chem. Rev. 2000, 100, 2159–2232.
[2] D. Rechavi, M. Lemaire, Chem. Rev. 2002, 102, 3467–
3494.
[3] a) M. I. Burguete, J. M. Fraile, J. I. García, E. García-
Verdugo, S. V. Luis, J. A. Mayoral, Org. Lett. 2000, 2,
3905–3908; b) M. I. Burguete, J. M. Fraile, J. I. García,
E. García-Verdugo, C. I. Herrerías, S. V. Lui, J. A. May-
oral, J. Org. Chem. 2001, 66, 8893–8901; c) R. Annun-
ziata, M. Benglia, M. Cinquini, F. Cozzi, M. J. Pitillo, J.
Org. Chem. 2001, 66, 3160–3166; d) S. Orlandi, A.
Mandoli, D. Pini, P. Salvadori, Angew. Chem. 2001, 113,
2587–2589; Angew. Chem. Int. Ed. 2001, 40, 2519–
2521; e) E. Díez-Barra, J. M. Fraile, J. I. García, E.
García-Verdugo, C. I. Herrerías, S. V. Luis, J. A. Mayor-
al, P. Sµnchez-Verdffl, J. Tolosa, Tetrahedron: Asymme-
try 2003, 14, 773–778; f) R. J. Clarke, I. J. Shannon,
Chem. Commun. 2001, 1936–1937; g) A. Lee, W. Kim,
J. Lee, T. Hyeon, B. M. Kim, Tetrahedron: Asymmetry
2004, 15, 2595–2598; h) J. G. Knight, P. E. Belcher, Tet-
rahedron: Asymmetry 2005, 16, 1415–1418.
[12] S. S. Lee, S. Hadinoto, J. Y. Ying, Adv. Synth. Catal.
2006, 348, 1248–1254.
[13] J. M. Fraile, J. I. García, C. I. Herrerias, J. A. Mayoral,
O. Reiser, A. Socuellamos, H. Werner, Chem. Eur. J.
2004, 10, 2997–3005.
[14] G. Jas, A. Kirschning, Chem. Eur. J. 2003, 9, 5708–
5723.
[15] a) A. J. Sandee, D. G. I. Petra, J. N. H. Reek, P. C. J.
Kamer, P. W. N. M. Leeuwen, Chem. Eur. J. 2001, 7,
1202–1208; b) N. Künzle, J.-W. Solꢁr, A. Baiker, Catal.
Today 2003, 79, 503–509; c) A. Mandoli, S. Orlandi, D.
Pini, P. Salvadori, Tetrahedron: Asymmetry 2004, 15,
3233–3244; d) D. A. Annis, E. N. Jacobsen, J. Am.
Chem. Soc. 1999, 121, 4147–4154; e) M. I. Burguete, E.
Garcia-Verdugo, M. J. Vicent, S. V. Luis, H. N. Penne-
mann, N. G. Keyserling, J. Martens, Org. Lett. 2002, 4,
3947–3950; f) B. Altava, M. I. Burguete, E. García-Ver-
dugo, S. V. Luis, M. J. Vicent, Green Chem. 2006, 8,
717–726.
[4] a) M. Glos, O. Reiser, Org. Lett. 2000, 2, 2045–2048;
b) J. M. Fraile, J. I. García, C. I. Herrerías, J. A. Mayor-
al, O. Reiser, M. Vaultier, Tetrahedron Lett. 2004, 45,
6765–6768; c) H. Werner, C. I. Herrerías, M. Glos, A.
Gissibl, J. M. Fraile, I. PØrez, J. A. Mayoral, O. Reiser,
Adv. Synth. Catal. 2006, 348, 125–132; d) J. M. Fraile, I.
PØrez, J. A. Mayoral, O. Reiser, Adv. Synth. Catal.
2006, 348, 1680–1688.
[5] C. E. Song, S. Lee, Chem. Rev. 2002, 102, 3495–3524.
Adv. Synth. Catal. 2008, 350, 1295 – 1308
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1307

Jaehong Lim et al.
FULL PAPERS
[16] a) M. I. Burguete, A. Cornejo, E. García-Verdugo, J.
García, M. J. Gil, S. V. Luis, V. Martínez-Merino, J. A.
Mayoral, M. Sokolova, Green Chem. 2007, 9, 1091–
1096; b) M. I. Burguete, A. Cornejo, E. García-Verdu-
go, M. J. Gil, S. V. Luis, J. A. Mayoral, V. Martínez-
Merino, M. Sokolova, J. Org. Chem. 2007, 72, 4344–
4350.
[17] H. Werner, R. Vicha, A. Gissibl, O. Reiser, J. Org.
Chem. 2003, 68, 10166–10168.
1308
asc.wiley-vch.de
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2008, 350, 1295 – 1308