Polymer-bound pyridine-bis(oxazoline). Preparation through click chemistry and evaluation in asymmetric catalysis

Article abstract of DOI:10.1002/adsc.200700112

A pyridine-bis(oxazoline) ligand was efficiently immobilized by copper(I)-catalyzed azidealkyne cycloaddition onto a polystyrene resin. The so obtained click-pybox resin la was associated with various metal salts (YbCl ₃, LuCl₃, CuOTf) and the resulting resin-bound catalysts were explored in ring-opening of cyclohexene oxide, silylcyanation of benzaldehyde and alkynylation of imines. These new polymer-supported catalysts exhibit good to excellent performances in terms of catalytic activity, enantioselectivity and recyclability.

Full text of DOI:10.1002/adsc.200700112

COMMUNICATIONS
DOI: 10.1002/adsc.200700112
Polymer-Bound Pyridine-Bis(oxazoline). Preparation through
Click Chemistry and Evaluation in Asymmetric Catalysis
MØlanie Tilliet,a,b Stina Lundgren,b Christina Moberg,b,* and Vincent Levachera,*
a
Laboratoire de Chimie Organique Fine et HØtØrocyclique, UMR 6014, IRCOF, CNRS, UniversitØ et INSA de Rouen.
B.P. 08, 76131 Mont-Saint-Aignan CØdex, France
Fax : (+33)-(0)-2-3552-2962; e-mail: vincent.levacher@insa-rouen.fr
KTH School of Chemical Science and Engineering, Organic Chemistry, 100 44 Stockholm, Sweden; e-mail: kimo@kth.se
b
Received: February 28, 2007; Revised: July 6, 2007; Published online: September 11, 2007
Supporting information for this article is available on the WWW under http://asc.wiley-vch.de/home/.
Abstract: A pyridine-bis(oxazoline) ligand was effi-
ciently immobilized by copper(I)-catalyzed azide-
alkyne cycloaddition onto a polystyrene resin. The
so obtained click-pybox resin 1a was associated
with various metal salts (YbCl3, LuCl3, CuOTf) and
the resulting resin-bound catalysts were explored in
ring-opening of cyclohexene oxide, silylcyanation of
benzaldehyde and alkynylation of imines. These
new polymer-supported catalysts exhibit good to ex-
cellent performances in terms of catalytic activity,
enantioselectivity and recyclability.
tion of a new polystyrene-supported pybox by means
of “click chemistry”.[3] The performance of the result-
ing click-pybox resin 1a was investigated in asymmet-
ric ring opening of cyclohexene oxide,[4] silylcyanation
of benzaldehyde,[5] and alkynylation of imines.[6]
We were especially interested in investigating the
Cu(I)-catalyzed “click” azide-alkyne cycloaddition,[7]
since the resulting triazole ring connecting the ligand
to the polymer is formed in high yield and under mild
conditions,[8] providing a stable linker compatible with
various reaction conditions. The robustness of this tri-
azole linker and the simplicity of the method make
this approach very attractive for accessing new poly-
mer-supported catalysts. This strategy has recently
been exploited with success for the immobilization of
hydroxyproline,[9a] TEMPO,[9b] P,N-ligands,[9c] aza(bi-
soxazoline)[9d] and Cinchona alkaloid derivatives.[9e]
The click-pybox resin 1a was uneventfully prepared
Keywords: asymmetric catalysis; click chemistry;
copper(I); immobilization; lanthanides; pybox
In recent years, pyridine-2,6-bisoxazoline (pybox) de- in a three-step sequence from the readily available 4-
rivatives have emerged as a popular class of tridentate bromo-substituted phenyl-pybox 2[2c] and polystyrene-
ligands, highly efficient in a large range of asymmetric supported azide 5,[10] using the Gmeiner CuI proto-
catalytic processes.[1] The demands for economical, col.[8] The required “alkyne anchor” was installed
eco-friendly and recyclable polymer-supported cata- through a Sonogashira cross-coupling between 2 and
lysts led some groups to investigate various strategies (trimethylsilyl)acetylene in refluxing CH2Cl2 in the
to attach pybox ligands on solid supports.[2] The first presence of 2 mol% Pd
(PPh3)2Cl2, 4 mol% CuI, and
A
immobilization of pybox was reported by Mayoral a large excess of NEt3. The resulting 4-[(trimethylsilyl)-
et al. by means of copolymerization of 4-vinylpybox ethynyl]-pybox 3 was subsequently, without further
derivatives with styrene and divinylbenzene.[2a,b] purification, desilylated by TBAF affording 4-ethynyl-
Moberg et al. subsequently reported the grafting of pybox 4 in 50% yield over the two steps. The “click-
pybox ligands onto a TentaGel resin via the formation immobilization” of pybox 4 with the polystyrene-sup-
of ester bonds.[2c] Finally, Portnoy at al. reported a ported resin 5 was then accomplished in THF at 358C
five-step solid-phase synthesis of various pybox li- for 3 days in the presence of excess DIEA and a cata-
gands from a Wang trichloroacetimidate resin.[2d] Al- lytic amount of CuI.[8] Analysis of resin 1a by IR spec-
though these polymer-supported pybox ligands dem- troscopy showed complete disappearance of the typi-
onstrated promising results in a number of transfor- cal azide stretch vibration (2090 cmÀ1), consistent with
mations, all these approaches suffer from serious the smooth formation of the triazole ring. The esti-
drawbacks and limitations, regarding either the mated pybox loading based on elemental analysis
number of steps or the stability of the linker connect- (6.79% N; 0.8 mmolg)À1 demonstrated that “click-im-
ing pybox to the polymeric matrix. We report herein mobilization” occurred in essentially quantitative
a rapid and straightforward method for the prepara- yield (theoretical loading: 0.8 mmolg)À1. For compari-
Adv. Synth. Catal. 2007, 349, 2079 – 2084
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2079

MØlanie Tilliet et al.
COMMUNICATIONS
Table 1. Click-pybox resin 1a-mediated asymmetric ring
son, monomeric ligand 1b was also prepared from 4-
ethynyl-pybox 4 and benzyl azide under the same
conditions in 82% yield (Scheme 1).
opening of cyclohexene oxide.[a]
The click-pybox resin 1a was first evaluated in the
ring opening of cyclohexene oxide (Table 1). The
asymmetric ring opening of meso epoxides with
TMSCN has previously been found to be catalyzed by
pybox-lanthanide complexes affording b-trimethylsilyl-
oxy nitrile ring-opened products in high yields and
fair to good enantioselectivities.[4] Given that a sys-
tematic evaluation of LnCl3 revealed that YbCl3 and
LuCl3 furnished the highest ees, both lanthanide chlo-
ride salts were selected in this study. The resin-bound
pybox 1a was first incubated with a THF solution of
YbCl3, filtered and washed. Ring opening of cyclohex-
ene oxide in CH2Cl2 was accomplished in the pres-
ence of 10 mol% of the resulting resin-bound catalyst
and 1.2 equivs. of TMSCN at room temperature. The
b-trimethylsilyloxy nitrile 6a was obtained in 98%
conversion and 64% ee within 1 hour (entry 1). When
the click-pybox resin 1a was treated with LuCl3 and
involved in the ring opening of cyclohexene oxide
under the same conditions, b-trimethylsilyloxy nitrile
6a was obtained in 97% conversion and 57% ee
(entry 2). The same experiments conducted with
Entry
Catalyst
Run
Conversion [%][b]
ee [%][c]
1
2
3
4
5
6
7
8
1a-YbCl3
1a-LuCl3
1b-YbCl3
1b-LuCl3
1a-YbCl3
1a-YbCl3
1a-YbCl3
1a-YbCl3
1
1
1
1
2
3
4
5
98
97
98
98
97
95
98
97
64
57
65
55
21
7
4
4
[a]
All reactions were carried out in CH2Cl2 at room temper-
ature for 1 hour using 10 mol% of catalyst, 1.2 equivs. of
TMSCN.
Conversion determined by GC-MS analysis.
Enantiomeric excess determined by chiral GC after deri-
vatization of 6a into 6b according to ref.[12]
[b]
[c]
ligand 1b under homogenous conditions afforded a triazole linker nor the heterogeneous conditions
comparable conversions and enantioselectivities (en- perturb the performance of the catalyst. However, at-
tries 3 and 4). These results are very similar to those tempts to reuse the resin-bound catalyst after simple
reported in the literature with the parent phenyl- filtration and washings with CH2Cl2 resulted in a dras-
pybox ligand,[4] indicating that neither the presence of tic drop of the enantioselectivity, while the conversion
Scheme 1. Preparation of click-pybox resin 1a and monomeric ligand 1b
2080
asc.wiley-vch.de
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 2079 – 2084

Polymer-Bound Pyridine-Bis(oxazoline)
COMMUNICATIONS
remained unaffected after five consecutive runs (en- and LuCl3 salts were tested with click-pybox resin 1a.
tries 5–8).[11]
Reactions were conducted at room temperature in
We then turned our interest to the enantioselective the presence of 10 mol% polystyrene-bound pybox
silylcyanation of benzaldehyde (Table 2). Enantiomer- 1a and 5 mol% LnCl3. Both YbCl3 and LuCl3 com-
ically pure cyanohydrins represent an important class plexes provided comparable results. As judged by
GC-MS analysis, the reactions were completed within
1.5 h, affording cyanohydrin 7 with 87% and 78%
Table 2. Click-pybox resin 1a-mediated asymmetric silylcya-
nation of benzaldehyde.[a]
conversions, respectively (entries 2 and 6). These con-
versions are similar to those observed with the Tenta-
gel-supported pybox ligand previously reported by
us,[2c] whereas the enantioselectivity of the click-pybox
resin revealed to be somewhat lower, affording cyano-
hydrin 7 with 65% and 69% ee (entries 2 and 6). In
four consecutive runs, the polymer-supported catalysts
could be easily recovered by simple filtration and re-
cycled, although the catalytic activity was somewhat
reduced presumably due to some leaching of the
metal (entries 2–9). We were pleased to note that ees
obtained from the recycled catalysts were appreciably
improved, reaching the same level of induction as
that recorded with the Tentagel-supported pybox li-
gands previously reported (entries 5 and 9).[14]
Lastly, we examined the performance of click-
pybox resin 1a in enantioselective alkynylation of
imines (Table 3 and Table 4). Chiral propargylic
amines are useful building blocks and important pre-
cursors for the preparation of various nitrogen-con-
taining products.[15] Catalytic alkynylation of imines
by pybox-Cu(I) complexes has been reported earlier
Entry
Catalyst
Run
Conversion [%][b]
ee [%][c]
1[d]
2
3
4
5
6
7
8
9
1a-YbCl3
1a-YbCl3
1a-YbCl3
1a-YbCl3
1a-YbCl3
1a-LuCl3
1a-LuCl3
1a-LuCl3
1a-LuCl3
1
1
2
3
4
1
2
3
4
55
87
76
70
70
78
73
70
68
32
67
73
75
78
69
75
77
78
[a]
All reactions were carried out in acetonitrile: CH2Cl2
(3:2) at room temperature for 1.5 h using 1.2 equivs. of
TMSCN, 10 mol% polymer-bound ligand 1a and 5
mol% LnCl3.
Conversion determined by GC-MS analysis.
Enantiomeric excess determined by chiral GC.
Only acetonitrile used as solvent.
[b]
[c]
[d]
Table 3. Click-pybox resin 1a-mediated enantioselective ad-
dition of phenylacetylene to benzylidene aniline 8a.[a]
of chiral building blocks in organic synthesis.[13] The
use of pybox-lanthanide complexes in enantioselec-
tive silylcyanation of aldehydes has been reported
earlier, providing cyanohydrins with fair to good
enantioselectivities (up to 89% ee).[5] We have previ-
ously demonstrated that cyanohydrins could be ob-
tained with similar enantioselectivities (81% ee) by
conducting the reaction under heterogeneous condi-
tions with Tentagel-supported pybox ligands.[2c] In the
present context, we were interested in testing the
newly prepared click-pybox resin 1a in order to com-
pare its performance with the earlier reported Tenta-
gel-supported pybox. In contrast to what was ob-
served with pybox-ytterbium complexes under homo-
geneous conditions[5] and with Tentagel-supported
pybox ligands,[2c] the use of acetonitrile as solvent led
to a modest ee and rather low conversion (entry 1).
The sluggish reaction rate can be explained by the
poor swelling properties of the polystyrene matrix in
acetonitrile. To overcome this problem, the reaction
was then carried out in a mixture of acetonitrile and
CH2Cl2. The acetonitrile-CH2Cl2 ratio was optimized
to ensure a good compromise between resin swelling
and compatibility of the solvent with the reaction of
interest. Under optimized solvent conditions, YbCl3
Entry
Catalyst
Run
Conversion [%][b]
ee [%][c]
1
2
3
4
1b-Cu(I)
1a-Cu(I)
1a-Cu(I)
1a-Cu(I)
1a-Cu(I)
1a-Cu(I)
1a-Cu(I)
1
1
2
3
4
5
1
100
100
98
99
95
96
88
83
81
80
9
5
6
49
100
7[d]
89
[a]
All reactions were carried out in dichloroethane at room
temperature for 48 h using 1.5 equivs. of phenylacetylene,
10 mol% polymer-bound ligand 1a and 10 mol%
CuOTf.
[b]
1
Conversion determined by H NMR of the crude prod-
uct.
[c]
[d]
Enantiomeric excess determined by chiral HPLC
Polymer-supported catalyst from entry 6 after washings
with pyridine/methanol/CH2Cl2 and reloading with
copper triflate
Adv. Synth. Catal. 2007, 349, 2079 – 2084
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2081

MØlanie Tilliet et al.
COMMUNICATIONS
Table 4. Click-pybox resin 1a-mediated enantioselective al-
kynylation of imines 8a–h.[a]
next examined the possibility to recycle the metal-
free click-pybox resin 1a by successive washings (pyri-
dine/methanol/dichloromethane) of the deactivated
supported catalyst from run 6. Interestingly, when re-
charged with copper triflate, both the catalytic activity
and enantioselectivity of the resulting polymer-sup-
ported catalyst were fully re-established (Table 3,
entry 7).
To illustrate the performance and the effectiveness
of the click-pybox resin 1a, various N-arylimines 8a–h
were tested in this reaction (Table 4). In this set of ex-
periments, resin 1a was washed and recharged with
copper triflate between each experiment as reported
above. The corresponding propargylic amines 9a–h
were obtained in high conversions and good enantio-
selectivities ranging from 80% to 90% ee (Table 4,
entries 1–11). Although the enantioselectivity re-
mained to some extent inferior to that generally ob-
served under homogeneous conditions,[6] the click-
pybox resin 1a offered markedly higher enantioselec-
tivities to those usually obtained by Portnoyꢁs sup-
ported catalyst and showed superior recyclability of
Entry Substrate R1
R2
Conversion ee
[%][b]
[%][c]
1
2
4
5
7
8a
8b
8c
8d
8e
H
H
H
H
H
H
9a: 100
9b: 85
9c: 84
9d:78
88
90
90
84
81
4-Cl
4-Et
4-F
Ar=a-
naphthyl
H
9e:100
9
10
11
8f
8g
8h
4-t-Bu
4-Et
4-OMe
9f: 73
9g: 93
9h: 73
88
88
80
H
H
[a]
All reactions were carried out in dichloroethane at room the catalyst.
temperature for 48 h using 1.5 equivs. of phenylacetylene,
10 mol% polymer-bound ligand 1a and 10 mol%
CuOTf.
Conversion determined by H NMR of the crude prod-
uct.
In summary, we have shown for the first time that
click chemistry provides a useful strategy for immobi-
lizing pybox ligands on a polystyrene resin for asym-
metric catalytic applications. The grafting occurred
cleanly, under mild conditions, affording the click-
pybox resin 1a with an estimated pybox loading of
0.8 mmolgÀ1. Various polymeric metal complexes
could be pre-formed and exploited in a number of
catalytic benchmark reactions. Ring-opening of cyclo-
[b]
[c]
1
Enantiomeric excess determined by chiral HPLC.
and represents one of the rare enantioselective pro- hexene oxide and silylcyanation of benzaldehyde with
cesses available today for the preparation of this class TMSCN were first surveyed in the presence of poly-
of compounds with high enantioselectivity.[6] Portnoy mer-supported pybox-lanthanide complexes. In both
et al. have recently reported the use of polystyrene- cases, comparable reactivity and level of asymmetric
supported pybox-Cu(I) complexes in enantioselective induction were obtained when compared with parent
additions of phenylacetylene to N-arylimines.[2d] The pybox ligands under homogeneous conditions. The
authors obtained 83% ee for the reaction of phenyl- polymeric complex of click-pybox resin 1a with
acetylene and N-benzylideneaniline with their best copper triflate was then successfully used in enantio-
catalyst, but had problems with recycling.
selective addition of imines, affording the correspond-
Alkynylation of N-benzylideneaniline 8a was per- ing propargylic amines with high enantioselectivities
formed at room temperature in dichloroethane in the (up to 90% ee). In all cases, the metal-free resin 1a
presence of 10 mol% of the resin-supported catalyst, could be easily recycled by simple washings of the
generated in situ from the click-pybox resin 1a and supported catalyst. With the exception of ring open-
copper triflate. The aminoalkyne 9a was obtained in ing of cyclohexene oxide catalyzed by lanthanide
quantitative conversion and 88% ee (Table 3, complexes, in all other cases, the polymer-supported
entry 2), whereas under homogeneous conditions catalyst could be recovered by simple filtration and
pybox 1b afforded ees up to 96% (Table 3, entry 1). reused at least in four consecutive runs with no signif-
After filtration, the recycled catalyst could be reused icant erosion of the conversion and enantioselectivity,
in three consecutive runs with no apparent loss of the without adding further metal salts.
catalytic activity, while the enantioselectivity de-
creased somewhat from 88% to 80% ee (Table 3, en-
tries 2–5). A marked drop of both the conversion and
enantioselectivity was observed in run 5 (Table 3, en-
tries 6), probably due to partial oxidation of the
active catalytic copper(I) species to copper(II). We
2082
asc.wiley-vch.de
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 2079 – 2084

Polymer-Bound Pyridine-Bis(oxazoline)
COMMUNICATIONS
filtered off and the resin was rinsed twice with CH2Cl2
(2 mL). Enantiomeric excesses were determined after deri-
Experimental Section
vatization of 6a, by adding Sc(OTf)3 (4.9 mg, 0.01 mmol)
A
General Remarks
and acetyl chloride (0.1 mL, 2 mmol) to the filtrate and the
resulting mixture was stirred for 2.5 h.[10] The resulting b-
acetyloyl nitrile 6b was obtained with quantitative conver-
sion. Enantiomeric excesses of 6b were determined by GC
using a chiral column [Chiraldex, G-TA (gamma cyclodex-
trin trifluoroacetyl), 30 m”0.25 mm, 1208C, tR (minor)=
19 min, tR (major)=27 min].
The Merrifield resin (1.2 mmolgÀ1, cross-linked with 1%
DVB) was purchased from Fluka. The following compounds
were prepared according to literature methods: 4-bromopy-
box 2,[2c] azidomethylpolystyrene 5.[10] All other chemicals
were purchased and used without further purification. 1H
and 13C NMR were recorded on a Bruker Avance 300 with
chloroform-d1 (d=7.26, 1H; d=77.0, 13C) as internal stan-
dard unless otherwise indicated. The IR spectra were re-
corded on a Perkin–Elmer Paragon 500. Elemental analyses
were performed by the University of Rouen Microanalytical
Service Laboratory on a Carlo Erba 1160. Mass spectrome-
try was performed by the University of Rouen Spectroscopy
Center. Electron impact (EI) and chemical ionization (IC)
spectra were performed on a JEOL JMS AX-500 spectrom-
eter. Routine monitoring of reaction was performed by
TLC, using 0.2 Kieselgel 60 F254 precoated aluminium
sheets, commercially available from Merck. Flash chroma-
tographies were performed on Gerudan SI-60 (70–230 mesh
ASTM) from Merck. Tetrahydrofuran (THF) was distilled
on sodium-benzophenone ketyl under nitrogen, CH2Cl2
from NaH, and MeCN from CaH2. Resin 1a was prepared
on the First Mate apparatus from Argonaut. All heterogene-
ous catalytic reactions with resin 1a were performed on a
Quest 210 Parallel Synthesizer from Argonaut.
General Procedure for Lanthanide-Catalyzed
Asymmetric Addition of TMSCN to Benzaldehyde
Using Click-Pybox Resin 1a
Polymer-supported pybox 1a (30 mg, 0.8 mmolgÀ1,
0.024 mmol) and LnCl3 (0.01 mmol) were transferred to a
reaction vessel and agitated for 1 h in dry MeCN (0.4 mL)
and dry CH2Cl2 (0.6 mL) under nitrogen. After adding ben-
zaldehyde (106 mg, 0.1 mmol) and TMSCN (12 mg,
0.12 mmol), the reaction mixture was stirred at room tem-
perature and conversions were determined by GC-MS.
After each run, the solvent was filtered off, the resin was
rinsed twice with CH3CN (2 mL) and dried before being
used in another run. Enantiomeric excesses were deter-
mined by GC analyses using a chiral column [Chiraldex, G-
TA (gamma cyclodextrin trifluoroacetyl), 30 m”0.25 mm,
1158C, tR (R)=12 min, tR (S)=14 min].
Preparation of the Resin-Supported Pybox (1a)
General Procedure for Cu(I)-Catalyzed Asymmetric
Alkynylation of Imines Using Click-Pybox Resin 1a
Azidomethylpolystyrene 5 (1.2 mmolgÀ1, 500 mg, 0.6 mmol),
pybox 4 (475 mg, 1.2 mmol), CuI (12 mmol, 23 mg), and
DIEA (0.986 g, 7.6 mmol) were weighed into the reaction
vessel (16”115 mm test tube). After adding THF (15 mL),
the reaction vessel was equipped with an agitation magnet,
topped with a Teflon cap and fitted with the parallel reflux
bar of the First Mateꢂ. The resulting mixture was stirred at
358C for 3 days. The beads were filtered, successively
washed with pyridine (5”6 mL), methanol (5”6 mL), and
CH2Cl2 (5”6 mL) then dried at room temperature under
vacuum overnight to afford resin 1a. Anal. found: N 6.79%
corresponding to 0.8 mmol of pybox/g of resin. IR (KBr):
n=3063, 3026, 2921, 2852, 1945, 1871, 1808, 1639, 1603,
1565, 1512, 1492, 1475, 1450, 1401, 1345, 1311, 1258, 1227,
1184, 1150, 1102, 1045, 1028, 978, 947, 923, 898, 746, 695,
606, 530 cmÀ1.
Polymer-supported pybox 1a (30 mg, 0.8 mmolgÀ1,
0.024 mmol), CuOTf (10 mg, 0.02 mmol), phenylacetylene
(33 mL, 0.3 mmol), and imine (0.2 mmol) were transferred to
a reaction vessel and agitated for 24 h in dry dichloroethane
(1 mL). The resin was filtered and washed with CH2Cl2 (2”
10 mL). The crude material was purified by flash chroma-
tography on silica gel (eluent: pure cyclohexane then 1%
EtOAc in cyclohexane). Enantiomeric excesses were deter-
mined by chiral HPLC (Daicel Chiralcel OD column). The
polymer could either be reused directly or rinsed successive-
ly with pyridine, methanol and dichloromethane and then
reloaded with the metal salt.
Supporting Information
Experimental procedures for compounds 3, 4, 1b, rac-prop-
argylic amines 9a–h. Spectroscopic data for all compounds
(including H NMR spectra of 4, 1b, IR spectra of 5, 1a, b
General Procedure for Lanthanide-Catalyzed Ring-
Opening of Cyclohexene Oxide Using Click-Pybox
Resin 1a
1
and 4).
Polymer-supported pybox 1a (150 mg, 0.8 mmolgÀ1,
0.12 mmol) and LnCl3 (0.10 mmol) were mixed in dry THF
(6 mL) and agitated for 1.5 h at room temperature. THF
was filtered off and the supported catalyst was rinsed twice
with dry THF (2”5 mL) under N2. The resin was then dried
thoroughly before adding CH2Cl2 (2 mL). Cyclohexene
oxide (102 mL, 1.00 mmol) was then added followed by
TMSCN (160 mL, 1.20 mmol). The mixture was stirred at
room temperature and the reaction rate was monitored by
GC-MS. Each sample was filtered through a plug of silica
before analysis by GC-MS. After each run, the solvent was
Acknowledgements
We thank the CNRS, the rØgion Haute-Normandie, EGIDE
and the Swedish Foundation for Strategic Research for finan-
cial and technical support. M. T. thanks the CNRS and
EGIDE for a grant.
Adv. Synth. Catal. 2007, 349, 2079 – 2084
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2083

MØlanie Tilliet et al.
COMMUNICATIONS
References
J. N. H. Reek, J. H. van Maarseveen, Org. Lett. 2006, 8,
3227; d) A. Gissibl, M. G. Finn, O. Reiser, Org. Lett.
2005, 7, 2325; e) K. M. Kacprzak, N. M. Maier, W.
Linder, Tetrahedron Lett. 2006, 47, 8721.
[1] G. Desimoni, G. Faita, P. Quadrelli, Chem. Rev. 2003,
103, 3119.
[2] a) A. Cornejo, J. M. Fraile, J. I. Garcia, E. Garcia-ver-
dugo, M. J. Gil, G. Legarreta, S. V. Luis, V. Martinez-
Merino, J. A. Mayoral, Org. Lett. 2002, 4, 3927; b) M. I.
Burguete, J. M. Fraile, E. Garcia-Verdugo, S. V. Luis, V.
Martinez-Merino, J. A. Mayoral, Ind. Eng. Chem. Res.
2005, 44, 8580; c) S. Lundgren, S. Lutsenko, C. Jçnsson,
C. Moberg, Org. Lett. 2003, 5, 3663; d) A. Weissberg,
B. Halak, M. Portnoy, J. Org. Chem. 2005, 70, 4556.
[3] a) H. C. Kolb, M. G. Finn, K. B. Sharpless, Angew.
Chem. Int. Ed. 2001, 40, 2004; b) H. C. Kolb, K. B.
Sharpless, Drug Discov. Today 2003, 8, 1128.
[4] S. E. Schaus, E. N. Jacobsen, Org. Lett. 2000, 2, 1001.
[5] a) H. C. Aspinall, N. Greeves, P. M. Smith, Tetrahedron
Lett. 1999, 40, 1763; b) H. C. Aspinall, J. F. Bickley, N.
Greeves, R. V. Kelly, P. M. Smith, Organometallics
2005, 24, 3458.
[10] S. Arseniyadis, A. Wagner, C. Mioskowski, Tetrahedron
Lett. 2002, 43, 9717.
[11] YbCl3 may be imprisoned in the polymer matrix after
partial decomplexation from the pybox ligand. The
presence of uncomplexed YbCl3 in the polymer would
promote the ring-opening of cyclohexene oxide in the
next run, however without any stereoselectivity. This
scenario would explain that while the conversion of the
reaction remains unaltered between two consecutive
runs, a significant drop of the stereoselectivity was ob-
served. A control experiment conducted under homo-
genous conditions in the presence of a suspension of
YbCl3 in CH2Cl2 showed that the reaction was complet-
ed in 16 h at room temperature.
[12] S. Norsikian, I. Holmes, F. Lagasse, H. B. Kagan, Tetra-
hedron Lett. 2002, 43, 5715.
[6] a) C. Wei, C. J. Li, J. Am. Chem. Soc. 2002, 124, 5638;
b) A. Bisai, V. K. Singh, Org. Lett. 2006, 8, 2405.
[7] V. D. Bock, H. Hiemstra, J. H. van Maarseveen, Eur. J.
Org. Chem. 2006, 51.
[13] J-M. Brunel, I. P. Holmes, Ian P. Angew. Chem. Int.
Ed. 2004, 43, 2752.
[14] As suggested by a referee, competitive complexation of
LnCl3 on the triazole linker may be responsible for the
lower enantioselectivity obtained in the first runs. For a
comparable situation with copper(II) complexes of
bis(oxazoline)ligands having a triazole side arm, see:
R. Rasappan, M. Hager, A. Gissibl, O. Reiser, Org.
Lett. 2006, 8, 6099.
[8] S, Lçber, P. Rodriguez-Loaiza, P. Gmeiner, Org. Lett.
2003, 5, 1753.
[9] a) D. Font, C. Jimeno, M. A. Pericàs, Org. Lett. 2006, 8,
4653; b) A. Gheorghe, A. Matsuno, O. Reiser, Adv.
Synth. Catal. 2006, 348, 1016; c) R. J. Detz, S. A. Heras,
R. de Gelder, P. W. N. M. van Leeuwen, H. Hiemstra,
[15] L. Zani, C. Bolm, Chem. Commun. 2006, 4263.
2084
asc.wiley-vch.de
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2007, 349, 2079 – 2084