Multipurpose box- and azabox-based immobilized chiral catalysts

Article abstract of DOI:10.1002/adsc.200606121

Azabis(oxazolines) can be used as chiral ligands in the copper-catalyzed enantioselective Mukaiyama aldol reaction. When supported on solids, azabis(oxazoline)-copper complexes are more easily deactivated than their analogous bis(oxazoline)-copper complex

Full text of DOI:10.1002/adsc.200606121

FULL PAPERS
DOI: 10.1002/adsc.200606121
Multipurpose box- and azabox-Based Immobilized Chiral
Catalysts
a
a
a,
b
JosØ M. Fraile, Ignacio PØrez, JosØ A. Mayoral, * and Oliver Reiser
a
Departamento de Química Orgµnica, Instituto de Ciencia de Materiales de Aragón and Instituto Universitario de
Catµlisis HomogØnea, Facultad de Ciencias, Universidad de Zaragoza – C.S.I.C., 50009 Zaragoza, Spain
Fax : (+34)-976-762-077; e-mail: mayoral@unizar.es
b
Institut für Organische Chemie, Universität Regensburg, Universitätsstr. 31, 93053 Regensburg, Germany
Received: March 21, 2006; Accepted: June 13, 2006
Abstract: Azabis(oxazolines) can be used as chiral li- lectivity due to surface effects. The deactivation is
gands in the copper-catalyzed enantioselective Mu- not irreversible and the deactivated catalysts show
kaiyama aldol reaction. When supported on solids, excellent performance in the cyclopropanation reac-
azabis(oxazoline)-copper complexes are more easily tion, providing added value to the supported multi-
deactivated than their analogous bis(oxazoline)- purpose catalysts.
copper complexes, and are not compatible with the
use of coordinating solvents in the method of prepa-
ration. The performance of the immobilized catalysts Keywords: asymmetric catalysis; azabis(oxazolines);
(
up to 86% ee) depends on the support and the reac- cyclopropanation; Mukaiyama aldol; supported cata-
tion solvent, with some positive effect on enantiose- lysts
Introduction
The use of heterogeneous catalysts in enantioselective
[1]
reactions is an area of increasing interest, which is
mainly due to the practical advantage of easy separa-
tion of catalyst and products by simple filtration. This
significant advantage makes slight reductions in activ-
ity or enantioselectivity acceptable. However, this ad-
vantage is frequently not sufficient to justify the use
of immobilized catalysts when their preparation re-
Figure 1. Chiral box (1) and azabox (2) ligands.
quires considerable synthetic effort. Although recov- several authors have attempted the immobilization of
ery and reuse have been shown to increase the effec- chiral box-metal complexes by different strategies.[3]
tive TON of the immobilized catalysts, the search for
The formation of covalent bonds between the sup-
other “added value” for these systems seems to be port and the chiral ligand is the most common immo-
[
4]
necessary to increase the acceptance and applicability bilization strategy, and this has been applied to bis-
of immobilized chiral catalysts. Among these ꢀadded (oxazolines) using different methods. The polyme[7r,i8-]
valueꢁ aspects we can consider three as particularly zation of box substituted with two[5,6] or only one
important aims; (i) a synthetic effort similar to that vinylbenzyl group has been described. This general
required to prepare homogeneous catalysts, (ii) im- strategy has very recently been applied to inorganic
[
9]
[10]
[11]
proved results in comparison with the homogeneous supports. Both organic polymers
and silicas
phase and (iii) the preparation of multipurpose cata- have been used to anchor bis(oxazolines), although
lysts. The latter two factors are ambitious goals and the functionalization of the box is not straightforward.
will require a deep understanding of the immobilized
systems.
A simpler strategy involves immobilization through
electrostatic interactions. Although less widely used
Regarding the preparation of multipurpose cata- than the covalent strategy,[12] this methodology was
lysts, bis(oxazoline) ligands (box, 1; Figure 1) appear the first system employed to immobilize box-copper
[
13]
to be suitable candidates given that their metal com- complexes,
Indeed, this approach proved to be use-
plexes are efficient catalysts for a wide variety of less for Lewis acid-promoted enantioselective reac-
enantioselective reactions.[ In view of this interest, tions.
2]
[14]
1680
ꢂ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 1680 – 1688

Multipurpose box- and azabox-Based Immobilized Chiral Catalysts
FULL PAPERS
Azabis(oxazolines) (azabox, 2) are structural ana-
logues of bis(oxazolines) and have shown excellent
behavior as chiral ligands in different asymmetric re-
actions in the homogeneous phase.[ The presence of
a nitrogen atom instead of a carbon in the central
bridge leads to higher stability, improving the efficien-
cy in different immobilized systems,[ and is advanta-
geous as far as grafting on Merrifield resins is con-
cerned. This method was used to prepare the best het-
erogeneous catalyst for enantioselective cyclopropa-
15]
16]
Scheme 1. Mukaiyama aldol reaction between methyl pyru-
vate (3) and 1-phenyl-1-(trimethylsilyloxy)ethene (4).
[
17]
nation reactions.
This system was the best not only
due to the highest selectivity but also to the almost
complete chemoselectivity.
tions were conducted at room temperature to facili-
The next objective is to extend these easier meth- tate comparison with the heterogeneous cases, thus
odologies to other enantioselective reactions. The preventing possible diffusion limitations in the case of
Mukaiyama aldol reaction is probably one of the solid catalysts.
most useful reactions for CÀC bond formation in an
The results obtained in these reactions are gathered
Bis(oxazoline)-copper in Table 1. The catalytic activity is not noticeably
enantioselective manner.[
18]
complexes lead to excellent results in the homogene-
[
19,20]
ous phase.
This method has even been adapted
Table 1. Results obtained in the homogeneous Mukaiyama
[
7]
to biphasic conditions, either on a solid support or
aldol reactions between methyl pyruvate (3) and 1-phenyl-1-
with water-soluble polymeric ligands.[
21]
However, (trimethylsilyloxy)ethene (4).
[a]
both of these strategies require considerable synthetic
effort to modify the chiral ligands. Furthermore, these
procedures were not problem-free. The polymerized
catalyst needed the presence of additional molecular
[b]
Entry
Ligand
Solvent
t [h]
Yield [%]
% ee
1
2
3
4
5
6
7
8
9
1a
THF
24
24
24
18
21
24
24
24
24
24
24
48
76
45
64
41
59
46
49
91
55
86
66
60
69
CH Cl
100
100
100
80
2
2
[
7]
sieves
and, moreover, additional fresh MS and
Hexane
THF
copper salt were necessary after each cycle. The
2a
2b
2c
[
11]
CH Cl
water-soluble catalysts showed poor recyclability
2
2
Hexane
THF
72
and the copper salt was leached, leading to recovery
of the polymeric ligand alone. In this paper we pres-
ent our efforts to use easily immobilized chiral bis(ox-
azoline)- and azabis(oxazoline)-copper complexes in
enantioselective Mukaiyama aldol reactions.
100
100
100
100
99
CH Cl
2
2
Hexane
THF
10
11
12
CH Cl
2
2
Hexane
100
[a]
Reaction conditions: 1 mmol 3, 1.2 mmol 4, 10% catalyst,
Results and Discussion
room temperature. The catalysts were prepared by treat-
ment of the ligand with Cu(OTf) in CH Cl .
Determined by HPLC with a Chiralcel OD column; 5 is
the major product.
A
H
R
U
G
2 2 2
Homogeneous Mukaiyama Aldol Reactions with
Azabis(oxazoline)-Cu
[b]
A
H
R
U
G
Previous to the immobilization work, it seemed advis-
able to study the behavior of azabis(oxazoline) li- modified by the chiral ligand and high conversions
gands in homogeneous Mukaiyama aldol reactions. In are obtained after a 24 h period. As far as enantiose-
addition to the three azabis(oxazoline) ligands lectivities are concerned, bis(oxazoline) 1a led to
(
Figure 1), only the bis(oxazoline) bearing phenyl moderate results (41–64% ee), in agreement with the
substituents (1a) was tested for the sake of compari- results reported for the trimethylsilylketene acetal of
[
20]
son, given the problems encountered in the immobili- tert-butyl thioacetate,
although in our case the re-
zation of 1b-Cu complexes.[
13]
action was carried out at room temperature. It is
The reaction between methyl pyruvate (3) and 1- noteworthy that in this case the use of dichlorome-
phenyl-1-(trimethylsilyloxy)ethene (4) (Scheme 1) thane is more efficient than THF or hexane, a situa-
was chosen as the test reaction. Three different sol- tion in contrast with Evansꢁ results on using ligand 1b.
vents (tetrahydrofuran, dichloromethane and hexane)
The analogous azabis(oxazoline) 2a gave a slightly
were used in order to assess the effect reported by less active catalyst, with enantioselectivities in the
[
20]
Evans et al.
and the surface effect observed in the same range (46–59% ee). In this case, however, THF
case of cyclopropanation with catalysts immobilized is the most efficient solvent. The results are slightly
by electrostatic interactions on Laponite.[ The reac- better when an isopropyl group is present in the aza-
22]
Adv. Synth. Catal. 2006, 348, 1680 – 1688
ꢂ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
1681

JosØ M. Fraile et al.
FULL PAPERS
bis(oxazoline) ligand (2c) and 69% ee is obtained in
hexane, clearly better than the 23% ee reported for
the silylketene acetal.[ As expected, the best azabis-
20]
(
(
oxazoline) ligand is that bearing tert-butyl groups
2b). THF is again the best solvent (91% ee), with the
use of hexane leading to only slightly lower enantiose-
lectivity. In this case dichloromethane is clearly less
efficient and only 55% ee was obtained. If we com-
pare these results with those obtained with silylketene
[
20]
acetal,
it can seen that the solvent effect is highly
dependent on the silylenol ether used. It is important
to note the excellent results obtained with this azabox
ligand, which was used in this reaction for the first
time.
Scheme 2. Immobilization by electrostatic interactions.
Heterogeneous Catalysts: Preparation,
Characterization, and Catalytic Tests
sistent with our previous results. The reduced intensi-
ty of the (001) diffraction line in the X-ray diffraction
patterns indicates partial delamination of the layered
Given the excellent results obtained in the cyclopro- structure of the clay. This line is slightly shifted to
panation reactions, we initially prepared the heteroge- smaller angles, representing partial intercalation of
neous catalysts following the methods reported in pre- the complexes between the clay layers. This intercala-
vious papers. These methods for electrostatic and co- tion is more marked in L1a, with basal spacing of
valent immobilization are denoted by the letter A.
The copper complexes of ligands 1a, 2a, 2b and 2c ligand), than in the other solids (basal spacing around
were prepared with Cu(OTf) in dichloromethane, 15.4 Š). This result is also in agreement with the
and this solution was filtered to remove any non-com- higher copper content in L1a. Nafion-based catalysts
plexed Cu(OTf) particles. The complexes were dis- show a much higher copper loading, ranging between
solved in methanol to carry out the cationic exchange 0.66 mmol/g in N1a and 0.91 mmol/g in N2c.
Scheme 2) onto Laponite (L) and Nafion N50 (N) The IR spectra of the supported complexes are
according to the standard method used previously for shown in Figure 2. The most prominent band is that
17.2 Š (cf. 13.8 Š in Cu-Laponite without a chiral
AHCTREUNG
2
ACHTREUNG
2
(
cyclopropanation catalysts.[
13,15]
In each case the mix- of the C=N group, which appears in the copper com-
À1
ture was stirred for 24 h at room temperature and the plex at about 1650–1690 cm . The Laponite-immobi-
exchanged solids were filtered off, washed with di- lized complexes generally present a double band in
À1
chloromethane and dried under vacuum.
that zone. The band at 1657 cm in Cu-1a appears in
À1
These solids were characterized by copper analysis, L1a as two bands at 1653 and 1631 cm . The same
X-ray diffraction (Laponite-based catalysts) and IR band in Cu-2a appears at 1682 cm , whereas two
spectroscopy. Laponite-based catalysts contain bands at 1707 and 1668 cm are present in the spec-
À1
À1
amounts of copper in the range 0.13–0.23 mmol/g trum of L2a. The presence of different sites on the
(
L1a 0.23, L2a 0.15, L2b 0.18, L2c 0.17), which is con- solids may account for these differences. Characteris-
Figure 2. IR spectra of laponite supported copper complexes: A) box complexes: a) 1a-Cu
plexes: a) 2a-Cu(OTf) , b) L2a, c) L2b d) L2c.
A
H
R
U
G
2
ACHTREUNG
2
1682
www.asc.wiley-vch.de
ꢂ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 1680 – 1688

Multipurpose box- and azabox-Based Immobilized Chiral Catalysts
FULL PAPERS
À1
tic bands in the 1600–1350 cm region are present in Table 2. Results obtained in the Mukaiyama aldol reactions
between methyl pyruvate (3) and 1-phenyl-1-(trimethylsilyl-
oxy)ethene (4) promoted by solids prepared by method A.
the solids (Figure 2), demonstrating the integrity of
the copper complexes.
[a]
Covalent immobilization onto a commercially avail-
able Merrifield resin was carried out as described
[b]
Entry Catalyst Solvent Run t [h]
Yield [%] % ee
[
17]
elsewhere.
Conversion of the chloromethyl to the
1
L1a
THF
1
2
1
2
1
1
1
1
1
2
1
2
1
1
1
72
73
8
86
65
2
72
more reactive bromomethyl group was readily ach-
3
CH Cl
48
100
72
27
<5
<5
<5
31
9
67
58
79
n.d.
n.d.
n.d.
35
2
2
ieved by treatment with NaBr/NBu Br. Subsequent
4
4
48
coupling of the brominated resin with deprotonated
azabis(oxazoline) in THF under reflux (Scheme 3)
proceeded cleanly to give 7b. Complexation with
5
6
7
8
9
10
hexane
any
any
any
THF
168
24
24
L2a
L2b
L2c
N1a
Cu(OTf) was carried out in methanol, and the solid
A
C
H
T
R
E
U
N
G
2
24
catalyst was thoroughly washed with the same solvent
and dried under vacuum.
312
600
96
120
144
18
11
12
13
14
15
CH Cl
100
74
5
44
43
30
2
2
Hexane
any
7b-Cu
>144 <5
24 96
n.d.
[c]
CH Cl
97, 92
2
2
[
[
[
a]
Reaction conditions: 1 mmol 3, 1.2 mmol 4, 10% catalyst,
room temperature.
Determined by HPLC with a Chiralcel OD column; 5 is
the major product.
Results of the cyclopropanation reaction between styrene
and ethyl diazoacetate (Scheme 4). Enantioselectivities
are for trans and cis isomers respectively (trans/cis=
b]
c]
7
0:30).
cult to explain this behavior in the Laponite-immobi-
lized catalysts but it could be due to a surface effect,
[
22]
as shown in the case of cyclopropanation.
Howev-
Scheme 3. Covalent immobilization on polystyrene-divinyl- er, this effect was observed in solvents with low di-
benzene resin (Merrifieldꢁs resin).
electric constants, because of the increase in the ion-
pair interaction between support and complex. The
effect observed in THF seems to indicate the super-
The copper content of this resin is high and is in imposition of the unexplained solvent effect in solu-
the range 0.75–0.90 mmol/g depending on the batch tion with the surface effect of the solid catalyst. In
of resin. The presence of the azabox ligand in 7b was this case immobilization gave rise to a better enantio-
also confirmed by IR.
selectivity, which constitutes added value over simple
Solid catalysts prepared by method A were tested separation, especially given the ease of preparation of
in the Mukaiyama aldol reaction between methyl pyr- the solid catalyst.
uvate (3) and 1-phenyl-1-(trimethylsilyloxy)ethene (4)
With regard to recoverability, the catalyst used in
THF was almost completely deactivated after the first
(
Scheme 1). The results are gathered in Table 2.
The first conclusion that can be drawn is that the run (entry 2), whereas the catalyst used in dichloro-
heterogeneous catalysts are noticeably less active methane maintained a significant part of its activity
than the analogous homogeneous ones. In spite of (entry 4) although it gave slightly lower enantioselec-
this, L1a (entries 1–5) gives rise to high yields in THF tivity. The analysis of the used catalysts (0.18 mmol/g)
(
73%) and dichloromethane (quantitative), although showed a partial loss of copper. From the mechanism
it has a poor activity in hexane. The use of Laponite of the copper catalyzed Mukaiyama aldol reaction[
as the solid support improves significantly the enan- an intermediate copper aldolate is formed, and in the
tioselectivity of the process. Remarkably, the reaction case of catalysts prepared by cationic exchange, a sily-
in THF displays 86% ee, which is much higher than lated Laponite must be generated instead of the tri-
the value (45% ee) obtained in solution, and a similar methylsilyl triflate in solution. One possible source of
result is observed on using hexane, 79% ee vs. 41% leaching might be the formation of a copper bis-
18]
ee in solution. In contrast, the enantioselectivity in di-
A
H
R
U
G
chloromethane does not change perceptibly. It is diffi- were less efficient than triflate in the silylation step.
Adv. Synth. Catal. 2006, 348, 1680 – 1688
ꢂ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
1683

JosØ M. Fraile et al.
FULL PAPERS
However, this loss of around 20% of the initial types of immobilized catalyst in the absence of coor-
copper does not account for a complete deactivation dinating solvents.
(
entry 2), that may be due to strong complexation of
The 2b-Cu
A
H
R
U
G
copper with products or by-products.
chloromethane and the Laponite support was sus-
When the azabis(oxazoline)-based catalysts were pended in the same solution. After evaporation of the
tested (entries 6–8), the results were highly disap- solvent, the complex was immobilized on the Lapon-
pointing. Reactions did not take place irrespective of ite surface. The nature of the complex-support inter-
the solvent used and it appeared as if the freshly pre- action was confirmed by the lack of solubility of the
pared catalysts were deactivated in some way.
immobilized complex in dichloromethane, and the
In view of these results, Nafion N-50 was only used non-detectable leaching after one cyclopropanation
with bis(oxazoline) 1a (entries 9–13). N1a is even less reaction. This solid was used in the Mukaiyama aldol
active than L1a, in spite of the similar natures of reaction under the same conditions as the solids pre-
Nafion and triflate. The best solvent in this case was pared by method A.
dichloromethane, with total conversion and 44% ee.
As can be seen, the use of a non-layered support pre- Cu
In the case of the Merrifield-based catalyst,
(OTf) was dissolved in dichloromethane or THF
ACHTREUNG
2
cludes any surface effect, and the enantioselectivities in the presence of the required amount of methyl pyr-
are slightly lower than those obtained in solution: uvate for the catalytic reaction. The polymeric ligand
35% vs. 45% ee in THF, 44% vs. 64% ee in dichloro- 7b was suspended in that solution and, after a certain
methane. This improved behavior of N1a in dichloro- equilibration time, the enol silane (4) was added to
methane is complemented by better recoverability the mixture.
(
entry 12), a situation in agreement with the behavior
The catalysis results are gathered in Table 3. As can
of L1a.
be seen, both catalysts L2b and 7b-Cu prepared by
As can be seen, the results with catalysts immobi- method B are active in the Mukaiyama aldol reaction
lized by electrostatic interactions are highly depen- – in contrast with the same catalysts prepared by
dent on several interrelated factors: support, solvent method A. This result demonstrates the role of meth-
and chiral ligand.
anol as a poison for copper-based Lewis acid catalysts.
The polymeric ligand 7b was used in the same reac- However, it should be noted that the activity of the
tion (entry 14) but, as with the exchanged catalysts, it new catalysts is also limited. The exchanged complex
did not show any significant catalytic activity.
L2b is only active in dichloromethane (entry 2), with
These results in Mukaiyama aldol reaction contrast- a moderate yield and a good enantioselectivity (76%
ed with the high performance of the same batches of ee) obtained. This latter value is better than that ob-
solids in the cyclopropanation reaction between sty- tained with the homogeneous catalyst in the same sol-
rene and ethyl diazoacetate (Scheme 4),[16] for exam- vent (55% ee, see Table 1). The polymeric catalyst
ple, in the case of 7b (entry 15). This difference in be- 7b-Cu was only tested in THF (entry 5) and dichloro-
havior may arise from the different mechanisms and methane (entry 8) due to the low solubility of the pyr-
active speciesꢁ in the two processes. The presence of uvate-Cu(OTf) complex in hexane. In both solvents
A
H
R
U
G
2
methanol as a solvent in the immobilization of the results were similar, with slightly better enantiose-
azabox-copper complexes may poison the catalysts if lectivity in THF (84% ee), as occurs in solution. The
methyl pyruvate were not able to displace methanol recycled catalysts are less active but the original enan-
molecules from the coordination sphere of copper.
tioselectivity is maintained in the subsequent reaction.
These results represent the first use of a chiral cata-
lyst immobilized by ion-pairing in an enantioselective
reaction promoted by Lewis acids. This study increas-
es the scope of this approach and shows that the sur-
face effect improves the enantioselectivity from that
obtained in solution. The results are strongly depen-
dent on the Lewis acid character of the catalytic cen-
ters. Therefore, the stronger complexation of azabox
ligands (2), which is advantageous from the point of
[
16]
Scheme 4. Cyclopropanation reaction between styrene and
ethyl diazoacetate.
view of catalyst stability, reduces the catalytic activ-
ity as a Lewis acid.
Problems in catalyst recovery are common in Lewis
acid-promoted reactions given that catalytic centers
can be easily poisoned by product, solvent, or ambi-
ent water molecules, as clearly shown by the role of
Heterogeneous Catalysts Prepared in the Absence of
Methanol (Method B)
In view of this hypothesis, we developed alternative methanol in this case. In this respect the multipurpose
methods (denoted by the letter B) to prepare both character of the catalyst may be an additional advant-
1684
www.asc.wiley-vch.de
ꢂ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 1680 – 1688

Multipurpose box- and azabox-Based Immobilized Chiral Catalysts
FULL PAPERS
Table 3. Results obtained in Mukaiyama aldol reactions between methyl pyruvate (3) and 1-phenyl-1-(trimethylsilyloxy)-
[a]
ethene (4) promoted by solids prepared by method B.
[b]
Entry
Catalyst
Solvent
Run
Reaction
t [h]
Yield [%]
Trans/cis
% ee
1
2
3
4
5
6
7
8
9
L2b
THF
1
1
2
1
1
2
3
1
2
3
Mukaiyama
Mukaiyama
cyclopropanation
Mukaiyama
Mukaiyama
Mukaiyama
cyclopropanation
Mukaiyama
Mukaiyama
cyclopropanation
240
24
24
144
48
48
24
24
24
24
<5
53
48
<5
44
25
99
38
10
98
-
-
n.d.
76
82, 76
n.d.
84
85
97, 92
71
CH Cl
2
2
[
c]
[d]
70:30
-
-
-
71:29
-
-
hexane
THF
7b-Cu
[
[
c]
c]
[d]
[d]
CH Cl
2
2
71
95, 91
1
0
70:30
[
[
a]
Reaction conditions: 1 mmol 3, 1.2 mmol 4, 10% catalyst, room temperature.
b]
Mukaiyama: determined by HPLC with a Chiralcel OD column; 5 is the major product. Cyclopropanation: determined
by GC with a Cyclodex-B column, (1R,2R) and (1R,2S) are the major products.
Cyclopropanation reactions were always carried out in dichloromethane, irrespective of the solvent used in the previous
Mukaiyama reaction.
[
[
c]
d]
Enantioselectivities for the trans and the cis isomers, respectively.
age, as it can be reused in a different reaction, thus
providing added value to the easily obtained solid cat-
alyst. In order to investigate this hypothesis, we tested
the used catalysts from entries 2, 6 and 9 (Table 3) in
cyclopropanation reactions between styrene and ethyl
diazoacetate (Scheme 4). Surprisingly, these solids
were at least as active as the fresh catalyst reported
[
16,17]
elsewhere.
7b-Cu (entries 7 and 10) led to very
high yields of cyclopropanes without excess styrene
and more than 90% ee for both the trans and cis iso-
mers. The results with L2b were more modest but are
comparable to those obtained with the fresh catalyst,
i.e., 48% yield and 82% ee for the trans isomers.
Given the differences in behavior of 7b-Cu catalysts
prepared by methods A and B, IR spectra of the dif-
ferent resins were recorded (Figure 3).
Figure 3. IR spectra of solids based on Merrifield resins: a)
Merrifield resin, b) 7b, c) 7b-Cu prepared by method B, d)
As can be seen, the supported ligand 7b shows a
broad signal of C=N stretching (spectrum b), with a
7
b-Cu prepared by method B and used in Mukaiyama aldol
reaction, e) 7b-Cu prepared by method A.
À1
maximum at 1640 cm . Upon complexation by
À1
method B, the band is shifted to 1670 cm (spectrum
c), and the Mukaiyama reaction leads to a deactivated
solid (spectrum d) that presents a new band near
These results demonstrate that deactivation of a
À1
1
700 cm . This band may be ascribed to the carbonyl copper complex is highly dependent on the test reac-
groups of the coordinated product, although it is diffi- tion, and catalysts that are inactive for Lewis acid-
cult to explain why this molecule of product cannot promoted reactions are highly active and selective for
be replaced by a new silylenol ether molecule in the cyclopropanations. The reversible character of this de-
second run. The difference with 7b-Cu prepared by activation and methods to reactivate the catalysts are
method A is clear from the IR spectrum (spectrum currently under investigation.
e). In this solid it seems that part of the azabox moiet-
ies are non-complexed, whereas an important band at
À1
1
700 cm , analogous to that observed in the deacti- Conclusions
vated catalyst, is also present. The origin of this band
is not clear, but a similar band has been observed in Azabis(oxazolines) show similar performance to bi-
other cases in which partial hydrolysis of the ligand s(oxazolines) as chiral ligands in homogeneous
[
13d]
occurred.
copper-catalyzed enantioselective Mukaiyama aldol
Adv. Synth. Catal. 2006, 348, 1680 – 1688
ꢂ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
1685

JosØ M. Fraile et al.
FULL PAPERS
reactions. In spite of this similar behavior in solution, Synthesis of 7b
when supported on solids the azabis(oxazoline)-
copper complexes are more easily deactivated than Merrifield resin (1 g) was suspended in benzene (30 mL)
and treated with a solution of NaBr (4.12 g, 40 mmol) and
their analogous bis(oxazoline)-copper complexes,
probably as a consequence of their higher stability.
Azabis(oxazolines) are not compatible with the use of
coordinating solvents in the immobilization method,
and a new method has been developed that allows
high enantioselectivity to be obtained with moderate
yields. The performance of the immobilized catalysts
Bu NBr (0.97 g, 3 mmol) in water (30 mL). The mixture was
4
stirred at 608C for 5 d. The solid was filtered off and
washed with THF to yield the brominated resin.
Aza-bis(oxazoline) 2b (0.41 g, 1.7 mmol) was dissolved in
tetrahydrofuran (5 mL) and the solution was cooled to
À788C. n-BuLi (1.17 mL, 1.87 mmol) was added dropwise
by syringe and the mixture was allowed to warm up to room
temperature. The mixture was stirred for 10 min and was
(
up to 86% ee) depends on the nature of the support,
the immobilization method, and the reaction solvent. slowly transferred to a suspension of brominated resin
There is also some positive effect on enantioselectivi- (500 mg) in tetrahydrofuran (8 mL). The resulting mixture
ty due to surface effects. The deactivation observed in was heated under reflux for 40 h. The solid was filtered off,
the Mukaiyama aldol reaction is not irreversible, and washed with tetrahydrofuran, dichloromethane and metha-
nol, and dried under vacuum at 508C overnight to give PS-
bound-ligand 7b.
the deactivated catalysts show excellent performance
in a cyclopropanation reaction, a factor that provides
added value to these supported multipurpose cata-
lysts.
Preparation of 7b-Cu
Method A: The PS-bound-ligand 7b (0.1 g) was suspended
in a solution of Cu
A
H
R
U
G
(
1.7 mL) and shaken for 24 h at room temperature. After
Experimental Section
this time the solid was filtered off, washed with methanol
and dried under vacuum to yield 7b-Cu.
Method B: Methyl pyruvate (3, 0.5 mmol) and Cu
14 mg, 0.04 mmol) were dissolved in anhydrous tetrahydro-
A
C
H
T
R
E
U
N
G
(OTf)
General Remarks
2
(
furan (1 mL) and the solution was added to a suspension of
PS-bound-ligand 7b (0.05 mmol of ligand) in anhydrous tet-
rahydrofuran (2 mL). After 30 min the Mukaiyama aldol re-
action was started by addition of 1-phenyl-1-(trimethylsily-
loxy)ethane (0.6 mmol).
Box ligand 1a was prepared from phenylglycinol and dime-
[
23]
thylmalononitrile by a modified procedure.
Azabox li-
gands (2a, 2b and 2c) were prepared by methods described
elsewhere.[15]
Preparation of Immobilized Catalysts by Exchange
Procedures
Characterization of Immobilized Catalysts
Copper analyses were carried out by plasma emission spec-
troscopy on a Perkin–Elmer Plasma 40 emission spectrome-
ter. Elemental analyses were carried out on a Perkin–Elmer
2400 elemental analyzer. Step-scanned X-ray diffraction pat-
terns of oriented samples were collected at room tempera-
ture from 38 in 2q up to 608, using a D-max Rigaku system
with a rotating anode. The diffractometer was operated at
Laponite and Nafion NR-50 (sodium form) were dried
under vacuum at 1408C for 24 h prior to use.
Method A: The chiral ligand (0.11 mmol) and Cu
40 mg, 0.11 mmol) were dissolved in the minimum amount
A
H
R
U
G
(
of anhydrous dichloromethane under an argon atmosphere.
The solution was stirred for 15 min and filtered through a
PTFE microfilter. The solvent was removed under vacuum,
the complex was redissolved in methanol (4 mL), dried La-
ponite (375 mg) was added and the suspension was stirred
for 24 h at room temperature. The solid was filtered off,
washed with methanol (10 mL) and dichloromethane
4
0 kV and 80 mA, and the CuKa radiation was selected
using a graphite monochromator. Transmission FT-IR spec-
À4
tra of self-supported wafers evacuated (<10 torr) at 508C
were taken with a Nicolet Avatar 360 FTIR spectrophotom-
eter.
(
20 mL) and finally dried under vacuum for 24 h. The
Nafion-based catalyst was prepared in the same way from
box (0.22 mmol), Cu(OTf) (0.2 mmol) and Nafion NR-50
AHCTREUNG
2
Representative Procedure for Homogeneous
Mukaiyama Aldol Reactions
(
200 mg).
Method B: The chiral ligand (0.17 mmol) and Cu
55 mg, 0.151 mmol) were dissolved in anhydrous dichloro-
A
H
R
U
G
(
methane (5 mL) under an argon atmosphere. The solution
was stirred for 30 min and filtered through a PTFE microfil-
ter. This solution was added to dried Laponite (1 g) and the
solvent was slowly evaporated under reduced pressure. The
solid was dried under vacuum for 24 h.
A suspension of the corresponding ligand (0.05 mmol) and
Cu(OTf)2 (0.05 mmol) in anhydrous dichloromethane
A
H
R
U
G
(1 mL) was stirred for 15 min under an inert atmosphere.
The mixture was filtered through a microfilter and the sol-
vent was evaporated under vacuum. Anhydrous solvent
1686
www.asc.wiley-vch.de
ꢂ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 1680 – 1688

Multipurpose box- and azabox-Based Immobilized Chiral Catalysts
FULL PAPERS
(
2 mL), methyl pyruvate (3, 0.5 mmol) and 1-phenyl-1-(tri-
E. Malmstrçm, A. Hult, C. Moberg, Bioorg. Med.
Chem. Lett. 2002, 12, 1857.
methylsilyloxy)ethene (4, 0.6 mmol) were added by syringe
and the reaction mixture was stirred at room temperature
under an inert atmosphere. The consumption of enolsilane
was monitored by GC. The mixture was filtered through a
silica pad (2 cm), which was washed with dichloromethane.
The resulting solution was concentrated under vacuum and
the mixture analyzed by HPLC; Chiralcel OD column,
hexane/ethyl acetate (97:3), 1 mL min . The absolute con-
figurations of 5 and 6 were assigned by comparison with the
results reported by Evans et al.
[4] a) N. E. Leadbeater, M. Marco, Chem. Rev. 2002, 102,
3217; b) C. A. McNamara, M. J. Dixon, M. Bradley,
Chem. Rev. 2002, 102, 3275; c) Q.-H. Fan, Y.-M. Li,
A. S. C. Chan, Chem. Rev. 2002, 102, 3385; d) C. E.
Song, S.-G. Lee, Chem. Rev. 2002, 102, 3495.
[5] a) M. I. Burguete, J. M. Fraile, J. I. García, E. García-
Verdugo, S. V. Luis, J. A. Mayoral, Org. Lett. 2000, 2,
3905; b) 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.
À1
[20]
Representative Procedure for Heterogeneous
Mukaiyama Aldol Reactions
[6] M. I. Burguete, J. M. Fraile, J. I. García, E. García-Ver-
dugo, C. I. Herrerías, S. V. Luis, J. A. Mayoral, J. Org.
Chem. 2001, 66, 8893.
[
7] S. Orlandi, A. Mandoli, D. Pini, P. Salvadori, Angew.
To a solution of methyl pyruvate (3, 0.5 mmol) and 1-
phenyl-1-(trimethylsilyloxy)ethene (4, 0.6 mmol) in anhy-
drous solvent (2 mL) was added the solid catalyst (required
amount for 0.05 mmol Cu). The resulting suspension was
stirred at room temperature under an inert atmosphere. The
consumption of enolsilane was monitored by GC. The mix-
ture was filtered and the filtrate was concentrated under
vacuum and analyzed by HPLC. The solid catalyst was re-
peatedly washed with anhydrous dichloromethane, dried
under vacuum and reused under the same conditions or in a
cyclopropanation reaction.
Chem. Int. Ed. 2001, 40, 2519.
[
8] a) A. Mandoli, S. Orlandi, D. Pini, P. Salvadori, Chem.
Commun. 2003, 2466; b) A. Mandoli, S. Orlandi, D.
Pini, P. Salvadori, Tetrahedron: Asymmetry 2004, 15,
3
233.
[
9] J. M. Fraile, J. I. García, C. I. Herrerías, J. A. Mayoral,
Chem. Commun. 2005, 4669.
[
10] a) K. Hallman, C. Moberg, Tetrahedron: Asymmetry
2
001, 12, 1475; b) J. G. Knight, P. E. Belcher, Tetrahe-
dron: Asymmetry 2005, 16, 1415.
[
11] Some examples: a) D. Rechavi, M. Lemaire, Org. Lett.
2
001, 3, 2493; b) R. J. Clarke, I. J. Shannon, Chem.
Commun. 2001, 1936; c) J. K. Park, S.-W. Kim, T.
Hyeon, B. M. Kim, Tetrahedron: Asymmetry 2001, 12,
Representative Procedure for Heterogeneous
Cyclopropanation Reactions
2
931; d) A. Corma, H. García, A. Moussaif, M. J.
Sabater, R. Zniber, A. Redouane, Chem. Commun.
002, 1058; e) T. M. Lancaster, S. S. Lee, J. Y. Ying,
To a solution of styrene (2.5 mmol) and n-decane (internal
standard, 50 mg) in anhydrous dichloromethane (2.5 mL)
was added the solid catalyst (required amount for
2
Chem. Commun. 2005, 3577; f) D. Rechavi, B. Albela,
L. Bonneviot, M. Lemaire, Tetrahedron 2005, 61, 6976.
12] Examples of electrostatic immobilization: a) M.
Mazzei, W. Marconi, M. Riocci, J. Mol. Catal. 1980, 9,
0.05 mmol Cu). Ethyl diazoacetate (2.5 mmol) was slowly
[
added (4 h) with a syringe pump, and the resulting suspen-
sion was stirred at room temperature under an inert atmos-
phere for 24 h. The mixture was filtered and the filtrate was
analyzed by GC.[
3
81; b) R. Selke, H. Häupke, H. W. Krause, J. Mol.
Catal. 1989, 56, 315; c) H. Brunner, E. Bielmeier, J.
Wiehl, J. Organomet. Chem. 1990, 384, 223; d) I. Toth,
B. E. Hanson, M. E. Davis, J. Organomet. Chem. 1990,
16,17]
3
97, 109; e) S. Shimazu, K. Ro, P. Sento, N. Ichikuni, T.
Uematsu, J. Mol. Catal. A 1996, 107, 297; f) T. Sento, S.
Shimazu, N. Ichikuni, T. Uematsu, J. Mol. Catal. A
Acknowledgements
1
999, 137, 263; g) R. Margalef-Catalꢃ, C. Claver, P.
Salagre, E. Fernµndez, Tetrahedron: Asymmetry 2000,
1, 1469; h) H. H. Wagner, H. Hausmann, W. F. Hçld-
This work was supported by the C.I.C.Y.T. (project
CTQ2005–08016), the DGA, and the Fonds der Chemischen
Industrie. I. P. is indebted to the MEC for a grant.
1
erich, J. Catal. 2001, 203, 150; i) C. Simons, U. Hane-
feld, I. W. C. E. Arends, A. J. Minnaard, T. Maschmey-
er, R. A. Sheldon, Chem. Commun. 2004, 2830.
[
13] a) J. M. Fraile, J. I. García, J. A. Mayoral, T. Tarnai, Tet-
rahedron: Asymmetry 1997, 8, 2089; b) J. M. Fraile, J. I.
García, J. A. Mayoral, T. Tarnai, Tetrahedron: Asymme-
try 1998, 9, 3997; c) J. M. Fraile, J. I. García, J. A. May-
oral, T. Tarnai, M. A. Harmer, J. Catal. 1999, 186, 214;
d) J. M. Fraile, J. I. García, C. I. Herrerías, J. A. Mayor-
al, M. A. Harmer, J. Catal. 2004, 221, 532.
References
[
1] Chiral Catalysts Immobilization and Recycling; (Eds.:
D. E. De Vos, I. F. J. Vankelecom, P. A. Jacobs), Wiley-
VCH, Weinheim, 2000.
[
[
2] H. A. McManus, P. J. Guiry, Chem. Rev. 2004, 104,
4
151.
[14] J. M. Fraile, J. I. García, M. A. Harmer, C. I. Herrerías,
3] a) D. Rechavi, M. Lemaire, Chem. Rev. 2002, 102, 3467;
b) O. Reiser, Chimica Oggi 2002, 20, 73; c) C. Jçnsson,
K. Hallman, H. Andersson, G. Stemme, M. Malkoch,
J. A. Mayoral, J. Mol. Catal. A 2001, 165, 211.
[15] a) M. Glos, O. Reiser, Org. Lett. 2000, 2, 2045; b) H.
Werner, R. Vicha, A. Gissibl, O. Reiser, J. Org. Chem.
Adv. Synth. Catal. 2006, 348, 1680 – 1688
ꢂ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
1687

JosØ M. Fraile et al.
FULL PAPERS
2
003, 68, 10166; c) C. Geiger, P. Kreitmeier, O. Reiser,
zlowski, C. S. Burgey, D. W. C. MacMillan, J. Am.
Chem. Soc. 1997, 119, 7893; c) D. A. Evans, D. W. C.
MacMillan, K. R. Campos, J. Am. Chem. Soc. 1997,
119, 10859; d) D. A. Evans, M. C. Kozlowski, J. A.
Murry, C. S. Burgey, K. R. Campos, B. N. Connel, R. J.
Staples, J. Am. Chem. Soc. 1999, 121, 669.
Adv. Synth. Catal. 2005, 347, 249; d) A. Gissibl, M. G.
Finn, O. Reiser, Org. Lett. 2005, 7, 2325.
[
16] a) J. M. Fraile, J. I. García, C. I. Herrerías, J. A. Mayor-
al, O. Reiser, A. SocuØllamos, H. Werner, Chem. Eur. J.
2
004, 10, 2997; b) J. M. Fraile, J. I. García, C. I. Herre-
rías, J. A. Mayoral, O. Reiser, M. Vaultier, Tetrahedron
Lett. 2004, 45, 6765; c) J. M. Fraile, J. I. García, M. A.
Harmer, C. I. Herrerías, J. A. Mayoral, O. Reiser, H.
Werner, J. Mater. Chem. 2002, 12, 3290.
[
[
20] D. A. Evans, C. S. Burgey, M. C. Kozlowski, S. W.
Tregay, J. Am. Chem. Soc. 1999, 121, 686.
21] a) B.-Y. Yang, X.-M. Chen, G.-J. Deng, Y.-L. Zhang,
Q.-H. Fan, Tetrahedron Lett. 2003, 44, 3535; b) M. Be-
naglia, M. Cinquini, F. Cozzi, G. Celentano, Org.
Biomol. Chem. 2004, 2, 3401.
[
17] 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.
[
22] A. I. Fernµndez, J. M. Fraile, J. I. García, C. I. Herre-
[
18] E. Carreira, in: Comprehensive Asymmetric Catalysis,
rías, J. A. Mayoral, L. Salvatella, Catal. Commun. 2001,
(Eds.: E. N. Jacobsen, A. Pfaltz, H. Yamamoto),
2
, 165.
Springer-Verlag, Berlin, Heidelberg, 1999; Ch. 29.1,
pp. 998–1065.
[
23] A. Cornejo, J. M. Fraile, J. I. García, M. J. Gil, V. Martí-
nez-Merino, J. A. Mayoral, E. Pires, I. Villalba, Synlett
[
19] a) D. A. Evans, J. A. Murry, M. C. Kozlowski, J. Am.
2
005, 2321.
Chem. Soc. 1996, 118, 5814; b) D. A. Evans, M. C. Ko-
1688
www.asc.wiley-vch.de
ꢂ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 1680 – 1688