Copper(II)-bisoxazoline catalyzed asymmetric 1,3-dipolar cycloaddition reactions of nitrones with electron-rich alkenes

Article abstract of DOI:10.1021/jo982081b

A transition-metal-catalyzed diastereo- and enantioselective 1,3- dipolar cycloaddition reaction between electrophilic nitrones and electron- rich alkenes has been developed employing chiral copper(II)- and zinc(II)- bisoxazolines as catalysts. In the presence of Cu(OTf)₂-bisoxazoline as the catalyst, the nitrones, which coordinate to the chiral catalyst in a bidentate fashion, reacted smoothly with alkenes at room temperature to give isoxazolidines in good yields and diastereoselectivity and with high enantioselectivities of up to 94% ee. The influence of the metal salts, chiral ligands, and solvents on the reaction course has been investigated, and a general reaction protocol is developed. On the basis of the absolute stereochemistry of the 1,3-dipolar cycloaddition product, the coordination of the nitrone to the catalyst is discussed and a mechanistic approach of the reaction is presented. It is proposed that the intermediate is one in which both the nitrone and alkene are coordinated to the chiral copper(II)- bisoxazoline catalyst.

Full text of DOI:10.1021/jo982081b

J . Org. Chem. 1999, 64, 2353-2360
2353
Cop p er (II)-Bisoxa zolin e Ca ta lyzed Asym m etr ic 1,3-Dip ola r
Cycloa d d ition Rea ction s of Nitr on es w ith Electr on -Rich Alk en es
Kim B. J ensen, Rita G. Hazell, and Karl Anker J ørgensen*
Center for Metal Catalyzed Reactions, Department of Chemistry,
Aarhus University, DK-8000 Aarhus C, Denmark
Received October 15, 1998
A transition-metal-catalyzed diastereo- and enantioselective 1,3-dipolar cycloaddition reaction
between electrophilic nitrones and electron-rich alkenes has been developed employing chiral copper-
(II)- and zinc(II)-bisoxazolines as catalysts. In the presence of Cu(OTf)2-bisoxazoline as the catalyst,
the nitrones, which coordinate to the chiral catalyst in a bidentate fashion, reacted smoothly with
alkenes at room temperature to give isoxazolidines in good yields and diastereoselectivity and with
high enantioselectivities of up to 94% ee. The influence of the metal salts, chiral ligands, and solvents
on the reaction course has been investigated, and a general reaction protocol is developed. On the
basis of the absolute stereochemistry of the 1,3-dipolar cycloaddition product, the coordination of
the nitrone to the catalyst is discussed and a mechanistic approach of the reaction is presented. It
is proposed that the intermediate is one in which both the nitrone and alkene are coordinated to
the chiral copper(II)-bisoxazoline catalyst.
In tr od u ction
advances in chiral Lewis acid catalyzed 1,3-dipolar
cycloaddition reactions have been achieved.31-47
The 1,3-dipolar cycloaddition reaction provides a pow-
erful method for the synthesis of five-membered hetero-
cyclic rings. The isoxazolidine adducts obtained from the
1,3-dipolar cycloaddition reaction of nitrones with alkenes
are very important compounds/intermediates in organic
chemistry; these cycloadducts have found wide applica-
tions in synthesis and as synthons in total synthesis by
their conversion into, for example, 3-amino alcohols and
alkaloids.1-6 During the past two decades numerous
enantioselective total syntheses have been described
using chiral isoxazolidines from asymmetric 1,3-dipolar
cycloaddition reactions.7,8 However, the chirality in the
cycloadduct has mostly been achieved through incorpora-
tion of chiral center(s) in both the nitrone and/or the
alkene in the 1,3-dipolar cycloaddition reaction.8,9
Contrary to the Lewis acid catalyzed asymmetric carbo-
and hetero-Diels-Alder reactions,10 the use of chiral
Lewis acids as catalyst in asymmetric 1,3-dipolar cy-
cloaddition reactions is relatively unexplored. Recently,
focus has been put on achiral Lewis acid catalyzed 1,3-
dipolar cycloaddition reactions to influence the rate and
regio- and stereoselectivity of the reaction,11-30 and
(12) Kanemasa, S.; Tsuruoka, T.; Wada, E. Tetrahedron Lett. 1993,
34, 87-90.
(13) Kanemasa, S.; Nishiuchi, M.; Kamimura, A.; Hori, K. J . Am.
Chem. Soc. 1994, 116, 2324-2339.
(14) Kanemasa, S.; Tsuruoka, T.; Yamamoto, H. Tetrahedron Lett.
1995, 36, 5019-5022.
(15) Kanemasa, S.; Tsuruoka, T. Chem. Lett. 1995, 49-50.
(16) Murahashi, S.; Imada, Y.; Kohno, M.; Kawakami, T. Synlett
1993, 395-396.
(17) Tamura, O.; Yamaguchi, T.; Noe, K.; Sakamoto, M. Tetrahedron
Lett. 1993, 34, 4009-4010.
(18) Tamura, O.; Yamaguchi, T.; Okabe, T.; Sakamoto, M. Synlett
1994, 620-622.
(19) Tamura, O.; Gotanda, K.; Terashima, R.; Kikuchi, M.; Miyawa-
ki, T.; Sakamoto, M. J . Chem. Soc., Chem. Commun. 1996, 1861-1862.
(20) Tamura, O.; Mita, N.; Gotanda, K.; Yamada, K.-I.; Nakano, T.;
Sakamoto, M. Heterocycles 1997, 46, 95-99.
(21) Katagiri, N.; Okada, M.; Morishita, Y.; Kaneko, C. J . Chem.
Soc., Chem. Commun. 1996, 2137-2138.
(22) Katagiri, N.; Okada, M.; Kaneko, C.; Furuya, T. Tetrahedron
Lett. 1996, 37, 1801-1804.
(23) Camiletti, C.; Poletti, L.; Trombini, C. J . Org. Chem. 1994, 59,
6843-6846.
(24) Castellari, C.; Lombardo, M.; Pietropaolo, G.; Trombini, C.
Tetrahedron: Asymmetry 1996, 7, 1059-1068.
(25) Degiorgis, F.; Lombardo, M.; Trombini, C. Tetrahedron 1997,
53, 11721-11730.
(26) Degioris, F.; Lombardo, M.; Trombini, C. Synthesis 1997, 1243-
1245.
(27) Tokunaga, Y.; Ihara, M.; Fukumoto, K. Tetrahedron Lett. 1996,
37, 6157-6160.
(1) Morrison, J . D. Asymmetric Synthesis; Academic Press: New
York, 1983; Vol. 1, pp 1-11.
(28) Naito, T.; Shirikawa, M.; Ikai, M.; Ninomiya, I.; Kiguchi, T. Recl.
Trav. Chim. Pays-Bas 1996, 115, 13-19.
(2) Mulzer, J .; Altenbach, H.-J .; Brauun, M.; Krohn, K.; Reissing,
H.-U. Organic Synthesis Highlights; VCH: New York, 1991; pp 77-
85.
(29) Kashima, C.; Takahashi, K.; Fukuchi, I.; Fukusaka, K. Hetero-
cycles 1997, 44, 289-304.
(3) Torssell, K. B. G. Nitrile Oxides, Nitrones and Nitronates in
Organic Synthesis; VCH: New York, 1988.
(30) Minakata, S.; Ezoe, T.; Nakamura, K.; Ryu, I.; Komatsu, M.
Tetrahedron Lett. 1998, 39, 5205-5208.
(4) Tufariello, J . J . In 1,3-Dipolar Cycloaddition Chemistry; Padwa,
A, Ed.; Wiley: New York, 1984; Vol. 2, pp 83-168.
(5) Padwa, A. Comprehensive Organic Synthesis; Trost, B. M.,
Flemming, I., Semmelhack, M. F., Eds.; Pergamon: Oxford, 1991; Vol.
4, pp 1069-1109.
(6) Tufariello, J . J . Acc. Chem. Res. 1979, 12, 396-403.
(7) Frederickson, M. Tetrahedron 1997, 53, 403-425.
(8) Gothelf, K. V.; J ørgensen, K. A. Chem. Rev. 1998, 98, 863-909.
(9) Aggarwal, V. K.; Grainger, R. S.; Adams, H.; Spargo, P. L. J .
Org. Chem. 1998, 63, 3481-3485.
(10) Kagan, H. B.; Riant, O. Chem. Rev. 1992, 92, 1007-1019.
(11) Kanemasa, S.; Uemura, T.; Wada, E. Tetrahedron Lett. 1992,
33, 7889-7892.
(31) Gilbertson, S. R.; Dawson, D. P.; Lopez, O. D.; Marshall, K. L.
J . Am. Chem. Soc. 1995, 117, 4431-4432.
(32) Seerden, J . P. G.; Scholte op Reimer, A. W. A.; Scheeren, H. W.
Tetrahedron Lett. 1994, 35, 4419-4422.
(33) Seerden, J .-P. G.; Kuypers, M. M. M.; Scheeren, H. W. Tetra-
hedron: Asymmetry 1995, 6, 1441-1450.
(34) Seerden, J .-P. G.; Boeren, M. M. M.; Scheeren, H. W. Tetrahe-
dron 1997, 53, 11843-11852.
(35) Seebach, D.; Marti, R. E.; Hintermann, T. Helv. Chim. Acta
1996, 79, 1710-1740.
(36) (a) Hori, K.; Kodama, H.; Ohta, T.; Furukawa, I. Tetrahedron
Lett. 1996, 37, 5947-5950. (b) Hori, K.; Ito, J .; Ohta, T.; Furukawa, I.
Tetrahedron 1998, 54, 12737-12744.
10.1021/jo982081b CCC: $18.00 © 1999 American Chemical Society
Published on Web 03/11/1999

2354 J . Org. Chem., Vol. 64, No. 7, 1999
J ensen et al.
The focus in the chiral Lewis acid catalyzed 1,3-dipolar
cycloaddition reaction has mainly been devoted to the
activation of electron-deficient alkenes by the cata-
lyst.31,35-47 In previous papers, we have shown that chiral
titanium(IV), magnesium(II), and ytterbium(III) com-
plexes can act as chiral catalysts for reactions between
different nitrones and various alkenes.40-46 The electron-
deficient alkenes are activated for the 1,3-dipolar cy-
cloaddition reaction by a bidentate coordination to the
metal catalyst. By this coordination, the LUMOalkene
energy is significantly lowered compared to that of the
uncoordinated alkene, giving rise to a more favorable
HOMOnitrone-LUMOalkene interaction leading to an in-
crease of the reaction rate with the nitrone. The intro-
duction of a suitable chiral ligand on the metal can thus
lead to induced chirality in 1,3-dipolar cycloaddition
reactions.
cloaddition reactions of nitrones with electron-rich alk-
enes leading to synthetically useful 1,3-dipolar cycload-
dition adducts.
The nitrone substrates for the inverse electron-demand
1,3-dipolar cycloaddition reaction presented here contain
an ester substituent that allows for a bidentate coordina-
tion to the chiral Lewis acid. These nitrones, 1, can react
with electron-rich alkenes, 2, in a regioselective reaction
giving the isoxazolines exo-3 or endo-3 (eq 1).
In this paper, we turn our interest to the Lewis acid
catalyzed asymmetric 1,3-dipolar cycloaddition reaction
between electron-rich alkenes and electron-deficient ni-
trones, thereby changing the interaction between the
frontier molecular orbitals (FMOs) of the substrates to a
HOMOalkene-LUMOnitrone controlled reaction path. This
type of reaction can in principle be considered as an
inverse electron-demand 1,3-dipolar cycloaddition reac-
tion. To our knowledge, only Scheeren et al. have
reported a FMO interaction of this type in an asymmetric
1,3-dipolar cycloaddition reaction between an electron-
neutral nitrone and ketene acetals or vinyl ethers,
catalyzed by a chiral oxazaborolidinone.32-34 The catalysts
used in these studies were of the oxazaborolidine type,
producing up to 62% ee at -78 °C and 31% ee at room
temperature. In the case of cyclic vinyl ethers such as
2,3-dihydrofuran, a highly diastereoselective reaction
takes place, but the ee of the adduct was only 38%.
This paper presents a new metal-catalyzed 1,3-dipolar
cycloaddition reaction of electrophilic nitrones with elec-
tron-rich alkenes in which the nitrone is activated in a
bidentate fashion by chiral Lewis acids.
By coordination to the Lewis acid, the HOMO and
LUMO of 1 will be lowered compared with those of the
free nitrone, and the HOMOalkene-LUMOnitrone interaction
energy between the nitrone-Lewis acid complex and an
electron-rich alkene will be increased compared with that
of the reaction in the absence of a Lewis acid.
Kanamasa et al. were probably the first who observed
an acceleration in reaction rate of the 1,3-dipolar cy-
cloaddition reaction between nitrones and allyl alcohol
dipolarophiles by coordinating the nitrone to a Lewis acid,
such as MgX2, ZnBr2, TiCl2(Oi-Pr)2, or BF3‚Et2O.12,14,15 In
the absence of 1 equiv of the Lewis acid, a slight decrease
in reaction rate and a lack of diastereoselectivity were
observed. The lack of selectivity is presumably caused
by the Z/ E isomerization of the nitrone. In the presence
of MgBr2, the selectivity of the reaction is explained by
the favored chelation of the (Z)-nitrone and concomitant
coordination of the allyl alcohol to MgX2, as outlined in
Figure 1.
Resu lts a n d Discu ssion
To accelerate the inverse electron-demand 1,3-dipolar
cycloaddition reaction between a nitrone and an electron-
rich alkene, one can coordinate the nitrone in a mono-
dentate or bidentate fashion to a Lewis acid. In this
paper, we will use the latter approach and present the
development of catalytic enantioselective 1,3-dipolar cy-
F igu r e 1. The proposed coordination of the (Z)-nitrone and
allylic alcohol to MgX2, which accounts for the selectivity in
the 1,3-dipolar cycloaddition reaction.15
(37) Kobayashi, S.; Akiyama, R.; Kawamura, M.; Ishitani, H. Chem.
Lett. 1997, 1039-1040.
Because the nitrones 1 exhibit Z/ E isomerization in
solution at room temperature when the carbon atom
attached to the nitrogen atom has at least one hydrogen
atom,48-51 an investigation of the influence on the Z/ E
equilibrium by a bidentate coordination to a Lewis acid
was undertaken in an attempt to obtain a more con-
strained intermediate. The Z/ E equilibrium is also
important for the diastereoselectivity of the reaction
(38) Ukaji, Y.; Taniguchi, K.; Sada, K.; Inomata, K. Chem. Lett.
1997, 547-548.
(39) Gillbertson, S. R.; Lopez, O. D. J . Am. Chem. Soc. 1997, 119,
3399-3400.
(40) Gothelf, K. V.; J ørgensen, K. A. J . Org. Chem. 1994, 59, 5687-
5691.
(41) Gothelf, K. V.; Hazell, R. G.; J ørgensen, K. A. J . Org. Chem.
1996, 61, 346-355.
(42) Gothelf, K. V.; Thomsen, I.; J ørgensen, K. A. J . Am. Chem. Soc.
1996, 118, 59-64.
(43) Gothelf, K. V.; J ørgensen, K. A. Acta Chem. Scand. 1996, 50,
652-660.
(44) J ensen, K. B.; Gothelf, K. V.; Hazell, R. G.; J ørgensen, K. A. J .
Org. Chem. 1997, 62, 2471-2477.
(48) Inouye, Y.; Takaya, K.; Kakisawa, H. Bull. Chem. Soc. J pn.
1983, 56, 3541-3542.
(49) Aurich, H. G.; Franzke, M.; Kesselheim, H. P. Tetrahedron Lett.
1992, 48, 663-668.
(50) Inuoye, Y. Bull. Chem. Soc. J pn. 1983, 56, 244-247.
(51) (a) DeShong, P.; Dicken, M.; Staib, R. R.; Freyer, A. J .; Weinreb,
S. M. J . Org. Chem. 1982, 47, 4397-4403. (b) Burdisso, M.; Gandolfi,
R.; Gru¨nanger, P.; Rastelli, A. J . Org. Chem. 1990, 55, 3427-3429.
(45) J ensen, K. B.; Gothelf, K. V.; J ørgensen, K. A. Helv. Chem. Acta
1997, 80, 2039-2046.
(46) Sanchez-Blanco, A. I.; Gothelf, K. V.; J ørgensen, K. A. Tetra-
hedron Lett. 1997, 38, 7923-7926.
(47) Kobayashi, S.; Kawamura, M. J . Am. Chem. Soc. 1998, 120,
5840-5841.

Asymmetric 1,3-Dipolar Cycloaddition Reactions
J . Org. Chem., Vol. 64, No. 7, 1999 2355
Ta ble 1. 1,3-Dip ola r Cycloa d d ition Rea ction of N-Ben zyl-r-eth oxyca r bon ylm eth a n im in e N-Oxid e 1a w ith Eth yl Vin yl
Eth er 2a u n d er Va r iou s Rea ction Con d ition s
entrya
solvent
catalyst
reactn time (h)
temp (°C)
conversionb (%)
exo-3a :endo-3a
1
2
3
4
5
CH2Cl2
CH2Cl2
CH3CN
toluene
CH2Cl2
66
66
23
23
8
rt
40
58
45
76
69
24:76
22:78
35:65
12:88
30:70
50
50
50
rt
c
Cu(OTf)2
a
b
1
The reactions were performed on a 0.1 mmol scale. Conversion and exo-3a :endo-3a ratio were determined by H NMR spectroscopy.
c Catalyst loading, 25 mol %.
because, depending on the equilibrium, different ratios
of the diastereomers are obtained.
The Z/ E-equilibrium and the influence of Lewis acids
have been studied for N-benzyl-R-ethoxycarbonylmeth-
animine N-oxide 1a (eq 2) which is the class of nitrones
used in the present investigations.
elevated temperatures in various solvents accelerate the
reaction to different extents (entries 2-4). The exo-3a /
endo-3a ratio is dependent on the solvent, and a prefer-
ence for endo-3a is obtained especially in toluene (entry
4). It is considered that exo-3a is predominantly formed
from (Z)-1a , through an exo-transition state, and endo-
3a is produced mainly from (E)-1a , by an exo-transition
state.51 Because the (Z)-1a /(E)-1a ratio is solvent-depend-
ent with a decreased ratio when less polar solvents are
used,48-50,52 it can be seen that the reactivity of (E)-1a is
higher than that of (Z)-1a (entries 2-4).27,53 In the
presence of 25 mol % of Cu(OTf)2 as the catalyst, the
reaction of 1a proceeds at room temperature to give 3a
with 69% conversion after only 8 h (entry 5). No ap-
preciable difference is seen in the (Z)-1a /(E)-1a ratio of
the Cu(OTf)2-catalyzed reaction compared to that of the
uncatalyzed reaction. Tamura et al. have investigated the
same reaction in the presence of equimolar amounts of
Eu(fod)3 as the catalyst and found that endo-3a is the
major diastereomer formed; the stereoselectivity was
explained by an endo attack of the alkene to (Z)-1a .20
The C2-symmetric cationic MX2-bisoxazoline com-
plexes54 such as (S)-4a -c, (R)-4d ,e, (S)-5a , and (R)-5b
can act as Lewis acid catalysts for a variety of different
addition reactions such as carbo-55-58 and hetero-Diels-
Alder59-62 reactions. These catalysts can also be applied
for the 1,3-dipolar cycloaddition reaction of electron-
deficient alkene with nitrones.41
In the solid phase, nitrone 1a exists in the Z form, and
an 1H NMR spectrum in CD3CN recorded immediately
after dissolving 1a shows a (Z)-1a /(E)-1a ratio of 16:1.
After 1 h at room temperature, the (Z)-1a /(E)-1a ratio
has changed to 4.3:1. Addition of a Lewis acid changes
the (Z)-1a /(E)-1a equilibrium toward an increase (>90%)
in the Z form, and both Lewis acids, such as titanium-
(IV) and zinc(II), the latter also in the presence of chiral
ligands, have the properties to coordinate to 1a , bringing
it into the Z form.
The preliminary experiments focused on the 1,3-dipolar
cycloaddition reaction of N-benzyl-R-ethoxycarbonyl-
methanimine N-oxide 1a to ethyl vinyl ether 2a (eq 3).
The results for the 1,3-dipolar cycloaddition reaction
of N-benzyl-R-ethoxycarbonylmethanimine N-oxide 1a
(52) Inouye, Y.; Hara, J .; Kakisawa, H. Chem. Lett. 1980, 1407-
1410.
(53) Hara, J .; Inouye, Y.; Kakisawa, H. Bull. Chem. Soc. J pn. 1981,
54, 3871-3872.
(54) (a) Pfaltz, A. Acc. Chem. Res. 1993, 26, 339-345. (b) Ghosh, A.
K.; Mathivanen, P.; Cappiello, J . Tetrahedron: Asymmetry 1998, 9,
1-45.
(55) Evans, D. A.; Miller, S. J .; Lectka, T. J . Am. Chem. Soc. 1993,
115, 6460-6461.
(56) (a) Evans, D. A.; Murry, J . A.; Matt, P.; Norcross, R. D.; Miller,
S. J . Angew. Chem. Int. Ed. Engl. 1995, 34, 798-800. (b) Evans, D.
A.; Kozlowski, M. C.; Tedrow, J . S. Tetrahedron Lett. 1996, 37, 7481-
7484.
The reaction proceeds in a regioselective manner with
or without catalyst to give the isoxazolidines exo-3a and
endo-3a , which are useful substrates for further trans-
formation, i.e., to amino acids. The results for the reaction
of 1a with 2a under various reaction conditions are
presented in Table 1.
In the absence of a catalyst, the reaction between 1a
and 2a proceeds very slowly at room temperature, giving
only 40% conversion after 66 h (Table 1, entry 1), whereas
(57) Evans, D. A.; Lectka, T.; Miller, S. J . Tetrahedron Lett. 1993,
34, 7027-7030.
(58) Evans, D. A.; J ohnson, J . S. J . Org. Chem. 1997, 62, 786-787.
(59) Yao, S.; J ohannsen, M.; J ørgensen, K. A. J . Chem. Soc., Perkin
Trans. 1 1997, 2345-2349.
(60) J ohannsen, M.; J ørgensen, K. A. J . Org. Chem. 1995, 60, 5757-
5762.
(61) J ohannsen, M.; J ørgensen, K. A. Tetrahedron 1996, 52, 7321-
7328.
(62) Ghosh, A. K.; Mathivanen, P.; Cappiello, J .; K., K. Tetrahe-
dron: Asymmetry 1996, 8, 2165-2168.

2356 J . Org. Chem., Vol. 64, No. 7, 1999
J ensen et al.
Ta ble 2. 1,3-Dip ola r Cycloa d d ition betw een N-Ben zyl-r-eth oxyca r bon ylm eth a n im in e N-Oxid e 1a w ith Eth yl Vin yl Eth er
2a in th e P r esen ce of Va r iou s Cop p er (II) a n d Zin c(II) Sa lts Coor d in a ted to th e Ch ir a l Bisoxa zolin e Liga n d s (S)-4a -c,
(R)-4d ,e, (S)-5a , a n d (R)-5b a t Room Tem p er a tu r e
entrya
catalystb
reactn time (h)
conversionc (%)
exo-3a :endo-3a c
eed (exo-3a /endo-3a , %)
1e
2e,g
3e
4e
5e
6f
4a
4b
4c
4d
4e
5a
5b
8
24
24
7
24
2.5
2.5
93
81
73
86
79
75
69
84:16
44:56
66:34
43:57
46:54
58:42
47:53
89/35
0/0
62/0
44/0
62h/0
0/0
7f
0/0
a
b
The reactions were performed on a 0.1 mmol scale. Catalyst load, 25 mol %. c Conversion and exo-3a :endo-3b ratio were determined
by 1H NMR spectroscopy. Determined by HPLC (Daicel Chiralcel OD using hexane/i-PrOH 99.5:0.5, flow 0.7 mL/min). e CH2Cl2 as the
d
solvent. f Toluene as the solvent. Catalyst load, 1 mol %. Opposite enantiomer compared with entry 3.
g
h
with ethyl vinyl ether 2a (eq 3) in the presence of various
copper(II) and zinc(II) salts coordinated to the chiral
5) compared with 93% and 86% conversions after 7-8 h
(entries 1 and 4) for the latter. The ee of exo-3a was 62%
in both cases using (S)-4c and (R)-4e as the catalyst in
the zinc(II)-bisoxazoline catalyzed reaction. The absolute
stereochemistry of the major enantiomer changed when
moving from (S)-4c to (R)-4e. Hence, there is reason to
believe that the two zinc(II)-bisoxazoline catalyst-
substrate intermediates have the same geometry. Inves-
tigations of the copper(II)(pybox) complexes (S)-5a and
(R)-5b, which were anticipated to possess a preferred
square-planar coordination or square-pyrimidial geom-
etry with one or two accessible coordination sites56
indicated that the bidentate coordination of the nitrone
1a to the copper(II)(pybox) complex is not feasible
because of missing selectivities in the reaction between
nitrone 1a and alkene 2a (entries 6 and 7).
The diastereo- and enantioselectivity in the 1,3-dipolar
cycloaddition reaction of nitrone 1a with ethyl vinyl ether
2a in the presence of the copper(II)-bisoxazoline catalysts
is very solvent dependent. The results for the reaction of
1a with 2a in the presence of (S)-4a as the catalyst in
different solvents are presented in Table 3. In relation
to the results, it should also be mentioned that in the
copper(II)-bisoxazoline catalyzed carbo- and hetero-Di-
els-Alder reaction solvent effects have shown consider-
able changes in reaction rate, yield and enantioselectiv-
ity.61,63
bisoxazoline ligands (S)-4a -c, (R)-4d ,e, (S)-5a , and (R)-
5b are presented in Table 2.
The use of (S)-4a as the catalyst for the 1,3-dipolar
cycloaddition reaction of nitrone 1a with ethyl vinyl ether
2a (eq 3) leads to a conversion of 93% (Table 2, entry 1).
The diastereoselectivity has now changed to exo-selectiv-
ity in the reaction, as exo-3a and endo-3a were obtained
in a ratio of 84:16 with 89% and 35% ee, respectively.
Application of catalyst (R)-4d leads to a lack of diaste-
reoselectivity, and a moderate ee of only 44% for exo-3a
was obtained (entry 4). The major enantiomer of exo-3a
produced using (R)-4d as the catalyst has the same
absolute stereochemistry as that produced by the use of
(S)-4a , even though the absolute stereochemistry of the
ligand has changed. This effect has also been observed
in copper(II)-bisoxazoline catalyzed hetero-Diels-Alder
reactions.60-62 The change of the counterion of the
catalyst from triflate (in (S)-4a ) to antimonate (in (S)-
4b) leads, to our surprise, to a nonselective reaction
(entry 2). Compared to the Diels-Alder reaction,56,63 the
Lewis acidity of the two zinc(II)-bisoxazoline catalysts (S)-
4c and (R)-4e was weaker than that of the corresponding
copper(II) catalyst, as seen from the 73% and 79%
conversions, respectively, for the former (entries 3 and
The catalytic reaction in toluene between N-benzyl-R-
ethoxycarbonylmethanimine N-oxide 1a and ethyl vinyl
ether 2a gives a higher ee but a lower diastereoselectivity
compared to results of the reaction performed in CH2Cl2
(Table 3, entries 1 and 2). In toluene and CH2Cl2, 93%
and 89% ee, respectively, of exo-3a were obtained.
However, a moderate ee of endo-3a is found in CH2Cl2,
whereas a racemic mixture is formed in toluene. Applica-
tion of a mixture of petroleum ether and CH2Cl2 as the
solvent (entry 3) leads to an ee of 92% of exo-3a and an
improvement of the exo-3a /endo-3a ratio compared with
that for the reaction in toluene. More polar solvents
proved to be less effective for the reaction, and consider-
able amounts of byproduct and regioisomeres were
obtained (entries 4-6).
The reaction course is dependent on the temperature
and amount of both the alkene and the catalyst in the
standard reaction of nitrone 1a with ethyl vinyl ether
(63) J ohannsen, M.; J ørgensen, K. A. J . Chem. Soc., Perkin Trans.
2 1997, 1183-1185.

Asymmetric 1,3-Dipolar Cycloaddition Reactions
J . Org. Chem., Vol. 64, No. 7, 1999 2357
Ta ble 3. Solven t Effects in th e 1,3-Dip ola r Cycloa d d ition betw een N-Ben zyl-r-eth oxyca r bon ylm eth a n im in e N-Oxid e 1a
w ith Eth yl Vin yl Eth er 2a in th e P r esen ce of th e Cop p er (II)-bisoxa zolin e (S)-4a a t Room Tem p er a tu r e
entrya
solvent
reactn time (h)
conversionb (%)
exo-3a :endo-3a b
eec (exo-3a /endo-3a , %)
1
2
3
4
5
6
toluene
CH2Cl2
petroleum ether/CH2Cl2
CH3CN
THF
CH3NO2
24
8.5
24
23
23
96
98
93
70:30
84:16
77:23
68:32
57:43
37:63f
93/0
89/35
92/37
9e/2
51/0
14/2
84
65d
80d
99d
a
b
1
The reactions were performed on a 0.1 mmol scale. Conversion and exo-3a :endo-3a ratio were determined by H NMR spectroscopy.
c HPLC (Daicel Chiralcel OD using hexane/i-PrOH 99.5:0.5, flow 0.7 mL/min). Byproducts formed. e Opposite enantiomer. f The
d
regioselectivity was 65:35 in favor of the 5-isomer relative to the 4-isomer.
Ta ble 4. Asym m etr ic 1,3-Dip ola r Cycloa d d ition Rea ction s betw een th e Nitr on es 1a ,b a n d Alk en es 2a ,c Ca ta lyzed by 25
m ol % of Cop p er (II)-bisoxa zolin e Ca ta lyst (S)-4a
entrya
nitrone
alkene
product
reactn time (h)
yieldb (%)
exo-3:endo-3c
ee (exo-3/endo-3 %)d
1
2
3
4
1a
1a
1a
1b
2a
2b
2c
2a
3a
3b
3c
3d
20
20
20
38
83
83
43
52
77:23
31:69
50:50
50:50
89/16
90/94
12/0
0/0
a
b
The reactions were performed on a 0.25 mmol scale. For details, see Experimental Section. Isolated yields. c The endo:exo ratio was
determined by 1H NMR spectroscopy. The ee was determined by HPLC (Daicel Chiralcel OD or OJ using hexane/i-PrOH).
d
2a in CH2Cl2 as the solvent (eq 3). In the presence of 50
mol % of catalyst (S)-4a , a quantitative conversion was
obtained and the selectivities were similar to the reaction
using 25 mol % of the catalyst. Reduction of the catalyst
loading to 10 mol % reduces the diastereo- and enanti-
oselectivity, and an exo-3a /endo-3a ratio of approximately
50:50, with only moderate ee of the exo-3a (<20% ee),
was obtained. The addition of twice as much alkene
increased the reaction rate but decreased the selectivity
of the reaction slightly. The ee of the 1,3-dipolar cycload-
dition reaction in the presence of 25 mol % loading of
the catalyst changes slightly as the reaction proceeds.
In the beginning of the reaction course, a very high
enantioselectivity of 97% ee of exo-3a was obtained;
however, the ee of exo-3a went to a level of 89% when
the reaction had gone to nearly completion.
The more general application of this new exo-selective
catalytic approach to the 1,3-dipolar cycloaddition reac-
tion is demonstrated by performing the reaction of the
different nitrones 1a ,b with the alkenes 2a -d in the
presence of (S)-4a as the catalyst (eq 4). The reactions
1
were allowed to continue until H NMR spectroscopy of
the reaction mixtures reveal that >90% conversion has
taken place at room temperature, and the results are
presented in Table 4.
3b are formed, respectively. Cyclic alkenes, such as 2,3-
dihydrofurane 2c, react with nitrone 1a in a nondiaste-
reoselective reaction, giving 50% of each of the two
diastereomers with only 12% ee of exo-3c (entry 3). It is
also notable that an exchange of the ethyl ester in 1a
with a tert-butyl ester, N-benzyl-R-tert-butoxycarbonyl-
methanimine N-oxide 1b , leads to a nonselective dia-
stereo- and enantioselective 1,3-dipolar cycloaddition
reaction. The catalytic 1,3-dipolar cycloaddition reaction
of nitrone 1a proceeds also with tert-butyl vinyl ether 2d .
However, adducts exo-3e and endo-3e are formed in only
35% yield in an exo-3e/endo-3e ratio of 26:74, and both
diastereomers are formed in 50% ee.
The nitrone 1a reacts with ethyl vinyl ether 2a in the
presence of (S)-4a as the catalyst to give exo-3a as the
major diastereomer formed in 89% ee under these reac-
tion conditions (Table 4, entry 1). The 1,3-dipolar cy-
cloaddition of nitrone 1a with 2-methoxy propene 2b
leads to a change in diastereoselectivity, as endo-3b now
is the major diastereomer formed in a exo-3b/endo-3b
ratio of 31:69 (entry 2). Both diastereomers are formed
in very high ee as 90% and 94% ee of exo-3b and endo-

2358 J . Org. Chem., Vol. 64, No. 7, 1999
J ensen et al.
To determine the absolute stereochemistry of the major
enantiomer of exo-3a formed by reaction of nitrone 1a
with 2a in the presence of catalyst (S)-4a , the chiral 1,3-
dipolar cycloaddition adduct exo-3f was synthesized (eq
5). The exo-3a adduct was separated from the endo-3a
adduct and hydrolyzed with LiOH in a THF/H2O solution
giving the corresponding carboxylic acid which was
converted to the crystalline potassium salt exo-3f by
addition of 0.5 equiv K2CO3.
The square-planer copper(II)-bisoxazoline-nitrone in-
termediate 6 has the si-face of the nitrone available for
approach of the alkene. However, this approach is not in
accordance with the absolute stereochemistry of the
isoxazolidine adduct, determined by X-ray analysis in
Figure 2, as approach to the si-face will lead to the
(3S,5R)-isoxazolidine of exo-3a . This absolute stereo-
chemistry of the 1,3-dipolar cycloaddition adduct is not
in accordance with what should be “expected” on the basis
of previous assignments of the absolute stereochemistry
of products obtained using (S)-4a as the catalyst for
addition reactions.54b,60,64-66
To account for the absolute stereochemistry of the 1,3-
dipolar cycloaddition adduct obtained by reaction of
nitrone 1a with the electron-rich alkenes catalyzed by
(S)-4a , different models for the intermediate have been
investigated. On the basis of these studies and the
experimental observation that addition of ethyl vinyl
ether 2a to the copper(II)-bisoxazoline-nitrone interme-
diate gives a significant color change from yellow-green
to green, it is postulated that both 1a and 2a can be
coordinated to the chiral catalyst during the reaction,
leading to a pentacoordinated intermediate.
For a pentacoordinated intermediate containing the
chiral bisoxazoline ligand, the nitrone, and ethyl vinyl
ether, the requirement is that both the bisoxazoline
ligand and the nitrone have to be in a cis fashion at the
metal. This leads to an intermediate, 7 (Figure 3, bottom),
in which the nitrone is coordinated in the equatorial
plane and the bisoxazoline in the equatorial and axial
positions, leaving the remaining axial position available
for the coordination of ethyl vinyl ether. The coordination
of ethyl vinyl ether in this axial position allows the alkene
fragment to approach the re-face of the nitrone in an exo-
selective fashion, leading to the formation of the isox-
azolidine with the same absolute (3R,5S) stereochemistry
as observed experimentally.
The absolute stereochemistry of the isoxazolidine exo-
3f as determined by X-ray analysis is presented in Figure
2. The chiral centers formed in the 1,3-dipolar cycload-
dition reaction are assigned as (3R,5S). The absolute
stereochemistry of exo-3f indicates that the alkene ap-
proaches the re-face (relative to the nitrone carbon atom)
of nitrone 1a when (S)-4a is applied as the catalyst.
The absolute stereochemistry of the 1,3-dipolar cy-
cloaddition adduct can also be accounted for by a tetra-
hedral copper(II)-bisoxazoline-nitrone intermediate, 8
(Figure 3, top). This intermediate has the re-face of
nitrone 1a available for approach of ethyl vinyl ether and
is also in accordance with the experimental results.
However, such a tetrahedral intermediate (8) is not in
accordance with what should be “expected” on the basis
of previous assignments of the absolute stereochemistry
of products obtained using the same catalyst addition
reactions.54b,60,64-66 Nevertheless, we leave both interme-
diates 7 and 8 in Figure 3 as possible candidates for the
intermediate in these reactions.
F igu r e 2. The X-ray structure of exo-3f. The hydrogen atoms
are shown only for the isoxazolidine ring.
The results presented in Table 2 for the use of copper-
(II)-bisoxazoline catalyst (S)-4a , which has the possibility
for a bidentate coordination, compared with the results
for the application of the copper(II)(pybox) (S)-5a and (R)-
5b catalysts, indicate that nitrone 1a coordinates to the
catalyst in a bidentate fashion. In relation to this
observation, it should be mentioned that the copper(II)-
bisoxazolines generally give the best enantiomeric results
for substrates which can coordinate in a bidentate fashion
and that coordination of substrates, such as R-dicarbonyl
compounds, leads to a square-planar complex.54b,60,64-66
Coordination of nitrone 1a to (S)-4a in this fashion leads
to intermediate 6 (Figure 3, top).
Con clu sion
A new copper(II)-bisoxazoline catalyzed diastereo- and
enantioselective 1,3-dipolar cycloaddition reaction be-
tween electron-poor nitrones that can coordinate to the
chiral Lewis acid in a bidentate fashion and electron-rich
alkenes has been developed. In the presence of Cu(OTf)2-
bisoxazoline as the catalyst, the nitrones react smoothly
with vinyl ethers at room temperature, giving isoxazo-
lidines in good yields and diastereoselectivity and with
enantioselectivities up to 94% ee, especially for ethyl
vinyl ether and 2-methoxy propene. The reaction leads
to an approach of the alkene fragment of the vinyl ether
to the re-face of the nitrone. On the basis of the absolute
stereochemistry of the isoxazolidine formed, a penta-
(64) Evans, D. A.; Burgey, C. S.; Paras, N. A.; Vojlovsky, T.; Tregay,
S. W. J . Am. Chem. Soc. 1998, 120, 5824-5825.
(65) Yao, S.; J ohannsen, M.; Audrain, H.; Hazell, R. G.; J ørgensen,
K. A. J . Am. Chem. Soc. 1998, 120, 8599-8605.
(66) Thorhauge, J .; J ohannsen, M.; J ørgensen, K. A. Angew. Chem.,
Int. Ed. Engl. 1998, 39, 2404-2406.

Asymmetric 1,3-Dipolar Cycloaddition Reactions
J . Org. Chem., Vol. 64, No. 7, 1999 2359
F igu r e 3. Top: a square-planar copper(II)-bisoxazoline-nitrone intermediate (6) in which the re-face of the nitrone is shielded
by the tert-butyl substituent of the bisoxazoline ligand, and a tetrahedral copper(II)-bisoxazoline-nitrone intermediate (8) showing
the possible approach to the re-face of the nitrone. Bottom: a pentacoordinated intermediate (7). To the left is a representation
of the coordination of both the nitrone and ethyl vinyl ether to the copper(II)-bisoxazoline catalyst. To the right is a model for the
intermediate in which only the hydrogen atoms of the vinyl ether are shown. The latter model is viewed from the same site as
intermediates 6 and 8 above.
hydrochloride to the corresponding glyoxylates according to
the literature.69 Cu(OTf)2, Zn(OTf)2, CuBr2, AgSbF6, AgPF6,
2,2′-isopropylidenebis[(4S)-4-tert-butyl-4,5-dihydrooxazole], 2,2′-
isopropylidenebis[(4R)-4-phenyl-4,5-dihydrooxazole], 2,6-bis-
[(4S)-tert-butyl-4,5-dihydrooxazole-2-yl]pyridine, 2,6-bis[(4R)-
phenyl-4,5-dihydrooxazole-2-yl]pyridine, benzylhydroxylamine
hydrochloride, ethyl vinyl ether, 2,3-dihydrofurane, and 2-meth-
oxy propene were purchased from Aldrich and distilled over
solid sodium prior to use. Cu(OTf)2, Zn(OTf)2, CuBr2, and
AgSbF6 were stored under N2.
Asym m etr ic 1,3-Dipolar Cycloaddition Reaction s; Gen -
er a l P r oced u r e for th e Rea ction Usin g 25 m ol % of
Ca ta lyst (S)-4a . A flame-dried Schlenk tube was charged with
Cu(OTf)2 (22.6 mg, 0.0625 mmol) and the ligand 2,2′-
isopropylidenebis[(4S)-4-tert-butyl-4,5-dihydrooxazole] (21.5
mg, 0.073 mmol) under a stream of N2. Dry CH2Cl2 (5 mL)
was added, and the mixture was then stirred at room tem-
perature for 1-2 h. To the green solution nitrone 1 (0.25 mmol)
and alkene 2 (0.50 mmol) were added. The solution was then
stirred at room temperature for the time given in the tables.
The reaction mixture was then filtered through a 30 mm layer
of silica gel. The silica gel layer was washed with 25-50 mL
of 0.3-0.6% Et3N in CH2Cl2, and the solvent was evaporated.
The crude product was purified by FC (silica gel, hexane/Et2O
95:5 containing 0.3% of Et3N). The less polar isomer was the
exo-isomer, and the more polar compound was the endo-isomer.
In most of the cases, the diastereomers could not be fully
separated. The ratio and assignment of the diastereomers were
possible by both 1H and 13C NMR spectroscopic data, except
in one case. Exo-3b and endo-3b are given as combinations of
the 13C NMR spectra below.
coordinated or a tetrahedral intermediate are proposed
to account for the absolute stereochemistry. In the former
intermediate, the chiral bisoxazoline ligand and the
nitrone occupy four of the five available coordination sites
at the copper(II) center, and the vinyl ether is coordinated
to the last coordination site.
Exp er im en ta l Section
Gen er a l Meth od s. The 1H and 13C NMR spectra were
recorded at 300 and 75 MHz, respectively. Chemical shifts for
1H and 13C NMR are reported in CDCl3 as the solvent and in
ppm downfield from tetramethylsilane (TMS). HPLC was
performed using a 4.6 mm × 25 cm Daicel Chiracel OD or OJ
columns. Mass spectra were recorded at 70 eV with a direct
inlet. Optical rotation was measured on a Perkin-Elmer 241
polarimeter. TLC was performed on Merck analytical silica
gel 60 F254 plates and visualized with blue stain. Solvents were
dried using standard procedures. All glass equipment was
dried in an oven at 150 °C before use.
Ma ter ia ls. The starting materials ethyl glyoxalate and tert-
butyl glyoxalate were prepared as described in the litera-
ture,67-69 stored at -18 °C, and distilled under water vacuum
prior to use. N-Benzyl-R-ethoxycarbonylmethanimine N-oxide
1a and N-benzyl-R-tert-butoxycarbonylmethanimine N-oxide
1b were synthesized from addition of benzylhydroxylamine
(67) J ung, M. E.; Shishido, K.; Davis, L. H. J . Org. Chem. 1982, 47,
891-892.
(68) Subasinghe, N.; Schulte, M.; Chan, M. Y.-M.; Roon, R. J .;
Koerner, J . F.; J ohnson, R. L. J . Med. Chem. 1990, 33, 2734-2744.
(69) Inouye, Y.; Watanabe, Y.; Takahashi, S.; Kakisawa, H. Bull.
Chem. Soc. J pn. 1979, 52, 3763-3764.
(+)-(3R,5S)-2-N-Ben zyl-5-et h oxy-isoxa zolid in e-3-ca r -
boxylic Acid Eth yl Ester . Total yield: 83%. Exo-3a : [R]D

2360 J . Org. Chem., Vol. 64, No. 7, 1999
J ensen et al.
+96° (c ) 1.0, CHCl3). Rf ) 0.35 (Et2O/petroleum ether 40:
60). HPLC (Daicel Chiralcel OD, hexane/i-PrOH ) 99.5:0.5,
flow rate ) 0.7 mL/min) tR ) 34.0 min (major), tR ) 38.8 min
(minor). Ee ) 89%. 1H NMR δ 1.15 (t, J ) 7.1 Hz, 3H), 1.24 (t,
J ) 7.1 Hz, 3H), 2.54 (ddd, J ) 13.2, 7.1, 2.2 Hz, 1H), 2.64
(ddd, J ) 13.2, 8.8, 5.5 Hz, 1H), 3.39 (dq, J ) 9.4, 7.1 Hz, 1H),
3.48 (dd, J ) 8.8, 7.7 Hz, 1H), 3.73 (dq, J ) 9.9, 7.1 Hz, 1H),
4.02 (d, J ) 14.3 Hz, 1H), 4.14 (q, J ) 7.1 Hz, 1H), 4.15 (q, J
) 7.1 Hz, 1H), 4.22 (d, J ) 14.3 Hz, 1H), 5.14 (dd, J ) 5.5, 2.2
14.3 Hz, 1H), 5.11 (dd, J ) 6.0, 2.2 Hz, 1H), 7.27-7.42
(m, 5H). 13C NMR δ 15.0, 28.0, 39.9, 61.3, 63.5, 66.1, 82.1,
100.7, 127.4, 128.2, 129.4, 135.0, 211.3. MS m/z ) 307 (M+).
En d o-3d : (mixture of endo-3d /exo-3d ) Rf ) 0.45 (Et2O/
petroleum ether 40:60). HPLC (Daicel Chiralcel OD, hexane/
i-PrOH ) 99.5:0.5, flow rate ) 0.5 mL/min) tR ) 37.5 min, tR
) 53.3 min. Ee ) 0%. 1H NMR δ 1.21 (t, J ) 7.1 Hz, 3H), 1.39
(s, 9H), 2.49 (ddd, J ) 12.6, 7.1, 1.0 Hz, 1H), 2.69 (ddd, J )
13.1, 8.2, 4.9 Hz, 1H), 3.46 (dq, J ) 9.3, 7.1 Hz, 1H), 3.76-
3.84 (m, 2H), 4.20 (d, J ) 13.2 Hz, 1H), 4.30 (d, J ) 12.6 Hz,
1H), 5.21 (d, J ) 4.9 Hz, 1H), 7.26-7.42 (m, 5H). 13C NMR δ
15.1, 27.9, 39.5, 63.2, 65.1, 66.0, 81.8, 103.3, 127.5, 128.3, 129.3,
137.4, 169.9. MS m/z ) 307 (M+).
P r oced u r e for th e Con ver sion of exo-3a in to (+)-
(3R,5S)-2-N-Ben zyl-5-et h oxy-isoxa zolid in e-3-ca r b oxyl-
ic Acid . To the ester exo-3a (45.7 mg, 0.164 mmol) dissolved
in THF/H2O (1:1, 4.0 mL) at room temperature was added
LiOH (14.0 mg, 0.58 mmol). The solution was stirred at room
temperature for 1 h. The solution was acidifed to pH 1 with 4
M HCl and extracted twice with EtOAc. The combined organic
phase was washed with brine, and the organic phase was dried
over Na2SO4 and evaporated to dryness. The crude compound
was used without further purification in the next step (40.0
mg, 97%). [R]D ) +78° (c ) 1.0, CH3OH). Rf ) 0.10-0.26
(MeOH/CH2Cl2 5:95). 1H NMR δ 1.15 (t, J ) 7.1 Hz, 3H), 2.32
(ddd, J ) 13.2, 10.4, 4.9 Hz, 1H), 2.60 (dd, J ) 13.1, 1.6 Hz,
1H), 3.42 (dq, J ) 9.4, 7.1 Hz, 1H), 3.72 (dq, J ) 9.3, 7.1 Hz,
1H), 3.73 (dd, J ) 9.9, 1.1 Hz, 1H), 4.00 (d, J ) 14.2 Hz, 1H),
4.22 (d, J ) 13.3 Hz, 1H), 5.25 (d, J ) 4.4 Hz, 1H), 7.35 (s,
5H). 13C NMR δ 14.9, 38.4, 62.0, 62.8, 63.2, 100.3, 128.4, 128.9,
129.4, 134.5, 171.7. MS m/z ) 251.
Hz, 1H), 7.27-7.33 (m, 3H), 7.40 (dd, J ) 8.2, 1.6 Hz, 2H). 13
C
NMR δ 14.0, 14.9, 39.8, 61.3, 61.5, 63.3, 65.5, 100.5, 127.3,
128.1, 129.2, 136.0, 169.7. MS m/z ) 279 (M+).
En d o-3a : (mixture of endo-3a /exo-3a ) Rf ) 0.32 (Et2O/
petroleum ether 40:60). HPLC (Daicel Chiralcel OD, hexane/
i-PrOH ) 99.5:0.5, flow rate ) 0.7 mL/min) tR ) 55.9 min
1
(major), tR ) 60.8 min (minor). Ee ) 16%. H NMR δ 1.20 (t,
J ) 7.1 Hz, 3H), 1.22 (t, J ) 7.1 Hz, 3H), 2.56 (ddd, J ) 12.6,
7.2, 1.1 Hz, 1H), 2.74 (ddd, J ) 13.2, 8.2, 5.5 Hz, 1H), 3.47
(dq, J ) 9.9, 7.1 Hz, 1H), 3.80 (dq, J ) 9.4, 7.2 Hz, 1H), 3.91
(t, J ) 7.7 Hz, 1H), 4.11 (q, J ) 7.1 Hz, 2H), 4.18 (d, J ) 12.6
Hz, 1H), 4.34 (d, J ) 13.2 Hz, 1H), 5.24 (d, J ) 5.0 Hz, 1H),
7.26-7.35 (m, 3H), 7.40 (dd, J ) 8.2, 1.7 Hz, 2H). 13C NMR δ
14.0, 15.0, 39.7, 61.3, 63.2, 65.0, 65.3, 103.4, 127.5, 128.3, 129.1,
136.8, 170.7. MS m/z ) 279 (M+).
2-N-Ben zyl-5-m et h oxy-5-m et h yl-isoxa zolid in e-3-ca r -
boxylic Acid Eth yl Ester . Total yield: 83%. Exo-3b: (mix-
ture of endo-3b/exo-3b) Rf ) 0.22 (Et2O/petroleum ether 40:
60). HPLC (Daicel Chiralcel OJ , hexane/i-PrOH ) 99.5:0.5,
flow rate ) 0.5 mL/min) tR ) 32.3 min (major), tR ) 48.3 min
1
(minor). Ee ) 90%. H NMR δ 1.25 (t, J ) 7.1 Hz, 1H), 1.42
(s, 3H), 2.34 (dd, J ) 13.2, 9.3 Hz, 1H), 2.70 (dd, J ) 13.2, 7.1
Hz, 1H), 3.23 (s, 3H), 3.58 (dd, J ) 9.3, 7.1 Hz, 1H), 4.02 (d, J
) 13.8 Hz, 1H), 4.14 (q, J ) 7.1 Hz, 1H), 4.15 (q, J ) 7.1 Hz,
1H), 4.25 (d, J ) 13.7 Hz, 1H), 7.28-7.42 (m, 5H). 13C NMR
δ 14.0, 14.1, 19.3, 21.0, 45.0, 45.5, 49.2, 49.3, 61.3, 61.7,
65.0, 66.3, 66.9, 104.9, 107.7, 127.5, 127.5, 128.2, 128.3,
129.2, 129.3, 136.1, 136.7, 169.8, 170.8. MS m/z ) 279 (M+).
En d o-3b: Rf ) 0.22 (Et2O/petroleum ether 40:60). HPLC
(Daicel Chiralcel OJ , hexane/i-PrOH ) 99.5:0.5, flow rate )
0.5 mL/min) tR ) 35.4 min (major), tR ) 42.1 min (minor). Ee
P r oced u r e for th e Con ver sion of Ca r boxylic Acid of
exo-3a in to (+)-(3R,5S)-2-N-Ben zyl-5-eth oxy-isoxa zoli-
d in e-3-ca r boxylic Acid P ota ssiu m Sa lt (exo-3f). To the
carboxylic acid (40.0 mg, 0.16 mmol) dissolved in MeOH (1.0
mL) was added K2CO3 (11.0 mg, 0.08 mmol). After the mixture
stirred at 40 °C overnight, half of the mixture was evaporated
to dryness. Recrystallation of the white powder in CH3CN
provided exo-3f as colorless crystals (5.0 mg). [R]D ) +166° (c
1
) 1.0, CH3OH). H NMR (CD3OD) δ 1.06 (t, J ) 7.2 Hz, 3H),
1
) 94%. H NMR δ 1.18 (t, J ) 7.1 Hz, 3H), 1.46 (s, 3H), 2.56
2.28 (ddd, J ) 13.2, 9.9, 2.7 Hz, 1H), 2.71 (ddd, J ) 13.2, 8.3,
6.6 Hz, 1H), 3.18 (m, 1H), 3.31 (dq, J ) 9.3, 7.1 Hz, 1H), 3.57
(dq, J ) 9.8, 7.2 Hz, 1H), 3.70 (d, J ) 14.2 Hz, 1H), 4.27 (d, J
) 14.3 Hz, 1H), 5.02 (dd, J ) 6.6, 3.3 Hz, 1H), 7.15-7.25 (m,
3H), 7.35 (d, J ) 7.1 Hz, 2H). 13C NMR (CD3OD) δ 15.4, 42.5,
62.3, 64.5, 71.3, 102.3, 128.0, 129.0, 130.6, 138.8, 177.0.
(dd, J ) 12.1, 9.3 Hz, 1H), 2.65 (dd, J ) 12.0, 7.1 Hz, 1H),
3.32 (s, 3H), 3.96 (dd, J ) 9.4, 7.2 Hz, 1H), 4.09 (q, J ) 7.1
Hz, 2H), 4.15 (d, J ) 13.2 Hz, 1H), 4.33 (d, J ) 13.2 Hz, 1H),
7.28-7.42 (m, 5H). 13C NMR δ 14.0, 14.1, 19.3, 21.0, 45.0, 45.5,
49.2, 49.3, 61.3, 61.7, 65.0, 66.3, 66.9, 104.9, 107.7, 127.5, 127.5,
128.2, 128.3, 129.2, 129.3, 136.1, 136.7, 169.8, 170.8. MS m/z
) 279 (M+).
X-r a y Deter m in a tion s. Data were collected from a needle-
shaped crystal of exo-3f on a SMART diffractometer using Mo
KR radiation (λ ) 0.71073 Å). Exo-3f is tetragonal, space group
I41, a ) b ) 17.8839(6) Å, c ) 17.7409(8) Å. A total of 41591
reflections were measured, 6376 unique, internal agreement
0.041. The structure was solved by direct methods (SIR97)70
and refined by least squares to R ) 0.024, Rw ) 0.025 for 472
parameters. The absolute configuration was determined from
the anomalous scattering contribution of potassium by least-
squares refinement of the Rogers parameter,71 giving a value
of 0.96(3). In all, 2916 Friedel pairs were included in the
refinement.
2-N-Ben zyl-h exa h yd r o-fu r o[3,2d ]isoxa zole-3-ca r boxy-
lic Acid Eth yl Ester . Total yield: 43%. Exo-3c: (mixture of
endo-3c/exo-3c) Rf ) 0.31 (Et2O/petroleum ether 40:60). HPLC
(Daicel Chiralcel OD, hexane/i-PrOH ) 98:2, flow rate ) 1.0
mL/min) tR ) 17.6 min (major), tR ) 20.4 min (minor). Ee )
1
12%. H NMR δ 1.28 (t, J ) 7.1 Hz, 3H), 1.83-2.10 (m, 3H),
3.51 (d, J ) 7.1 Hz, 1H), 3.81 (d, J ) 13.7 Hz, 1H), 3.93-4.04
(m, 2H), 4.17 (d, J ) 13.7 Hz, 1H), 4.19 (q, J ) 7.1 Hz,
2H), 5.69 (d, J ) 5.5 Hz, 1H), 7.25-7.39 (m, 5H). 13C
NMR
δ 14.2, 28.6, 49.9, 60.8, 61.2, 69.6, 70.9, 104.8,
127.5, 128.3 129.1, 136.0, 168.7. MS m/z ) 277 (M+).
En d o-3c: Rf ) 0.26 (Et2O/petroleum ether 40:60). HPLC
(Daicel Chiralcel OD, hexane/i-PrOH ) 98:2, flow rate ) 1.0
mL/min) tR ) 23.3 min, tR ) 30.1 min. Ee ) 0%. 1H NMR δ
1.29 (t, J ) 7.1 Hz, 3H), 1.90-2.10 (m, 3H), 3.34-3.40 (m, 2H),
4.03 (d, J ) 13.7 Hz, 1H), 4.03 (m, 1H), 4.12 (d, J ) 13.7 Hz,
1H), 4.17 (q, J ) 7.2 Hz, 2H), 5.78 (d, J ) 4.9 Hz, 1H), 7.26-
7.40 (m, 5H). 13C NMR δ 14.1, 30.7, 51.1, 59.1, 61.1, 67.9, 71.2,
106.0, 127.3, 128.1, 128.8, 136.1, 169.4. MS m/z ) 277 (M+).
2-N-Ben zyl-5-eth oxy-isoxa zolid in e-3-ca r boxylic Acid
ter t-Bu tyl Ester . Total yield: 52%. Exo-3d : (mixture of endo-
3d /exo-3d ) Rf ) 0.50 (Et2O/petroleum ether 40:60). HPLC
(Daicel Chiralcel OD, hexane/i-PrOH ) 99.5:0.5, flow rate )
Ack n ow led gm en t. We are indebted to The Danish
National Research Foundation and Statens Teknisk
Videnskabelige Forskningsråd for financial support.
Su p p or t in g In for m a t ion Ava ila b le: 1H and 13C NMR
and MS data for products and X-ray data for exo-3f. This
material is available free of charge via the Internet at
http://pubs.acs.org.
J O982081B
1
0.5 mL/min) tR ) 26.0 min, tR ) 28 min. Ee ) 0%. H NMR δ
1.14 (t, J ) 7.1 Hz, 3H), 1.44 (s, 9H), 2.51 (ddd, J ) 13.1, 7.7,
2.2 Hz, 1H), 2.57-2.68 (m, 1H), 3.34-3.44 (m, 2H), 3.73 (dq,
J ) 9.9, 7.1 Hz, 1H), 3.98 (d, J ) 14.4 Hz, 1H), 4.30 (d, J )
(70) Altomare, A.; Cascarano, G.; Giacivazzo, C.; Guagliardi, A.;
Burla, M. C.; Polidori, G.; Camilli, M. J . Appl. Chem. 1994, 27, 435.
(71) Rogers, D. Acta Crystallogr. 1981, A37, 734-741.