CuI-Fesulphos complexes: efficient chiral catalysts for asymmetric 1,3-dipolar cycloaddition of azomethine ylides

Article abstract of DOI:10.1016/j.tet.2007.03.130

The Cu^I-Fesulphos catalyst system (≤3 mol %) shows an excellent performance in the asymmetric 1,3-dipolar cycloaddition of azomethine ylides. High to very high levels of reactivity, endo/exo selectivity, and enantioselectivity (69->99% ee) are generally achieved with a very wide range of azomethine ylides and dipolarophiles. Based on experimental and computational studies data, a model that accounts for this high enantiocontrol is proposed.

Full text of DOI:10.1016/j.tet.2007.03.130

Tetrahedron 63 (2007) 6587–6602
CuI–Fesulphos complexes: efficient chiral catalysts for asymmetric
1,3-dipolar cycloaddition of azomethine ylides
a
Silvia Cabrera, Ramon Gomez Arrayas, Belen Martın-Matute,
a
a
ꢀ
Fernando P. Cossıo and Juan C. Carretero
ꢀ
ꢀ
ꢀ
ꢀ
b
a,
*
ꢀ
aDepartamento de Quꢀımica Orgaꢀnica, Universidad Autoꢀnoma de Madrid, Cantoblanco, 28049 Madrid, Spain
bKimika Fakultatea, Euskal Herriko Unibertsitatea, P.K. 1072, San Sebastiaꢀn-Donostia, Spain
Received 15 February 2007; revised 14 March 2007; accepted 21 March 2007
Available online 27 March 2007
Abstract—The CuI–Fesulphos catalyst system (ꢀ3 mol %) shows an excellent performance in the asymmetric 1,3-dipolar cycloaddition of
azomethine ylides. High to very high levels of reactivity, endo/exo selectivity, and enantioselectivity (69–>99% ee) are generally achieved
with a very wide range of azomethine ylides and dipolarophiles. Based on experimental and computational studies data, a model that accounts
for this high enantiocontrol is proposed.
Ó 2007 Elsevier Ltd. All rights reserved.
1. Introduction
described that subtle variations in the nature of the aryl
groups on the P atom of the chiral CuI–P,N-ferrocene catalyst
lead to a switch of the diastereoselectivity, a reverse exo/
endo-selectivity being achieved in the reaction with nitroal-
kenes by using either electron-rich or electron-withdrawing
substituent. Despite the number of highly efficient examples,
a general drawback of the catalytic systems developed is the
limited scope with regard to the dipolarophile, each proce-
dure being specific for a narrow variety of activated alkenes.
Therefore, the development of highly reactive and enantio-
selective catalysts showing compatibility with a broad vari-
ety of dipolarophiles and azomethine ylides remains a great
challenge.
The catalytic asymmetric 1,3-dipolar cycloaddition reaction
is likely the most straightforward and atom-economical
method for the enantioselective synthesis of five-membered
ring heterocycles.1 In recent years a wide variety of dipoles
have been successfully applied in asymmetric catalytic
cycloadditions, including nitrones,2 carbonyl ylides,3 nitrile
oxides,2f,4 diazoalkanes,5 azomethine imines,6 nitrile im-
ines,7 and azomethine ylides.8 In particular, the synthesis
of optically active pyrrolidine derivatives by 1,3-dipolar cy-
cloaddition reaction of azomethine ylides was first reported
by Grigg and Allway in 19919 using stoichiometric amounts
of CoCl2 or MnBr2 and an ephedrine derivative as the chiral
catalyst. However, despite its synthetic relevance, the cata-
lytic enantioselective version of this cycloaddition has been
developed only recently. Highly enantio- and endo-selective
reactions of stabilized glycine-derived N-metalated azo-
methine ylides with electron-deficient alkenes have been
reported by Jørgensen [ZnII–bisoxazoline (N,N ligand),10a
or AgI–cinchona alkaloids10b], Zhang [AgI–bisferrocenyl
phosphine amide ligand (N,P ligand)],11 Schreiber [AgI–
QUINAP (N,P ligand)],12 Zhou [AgI–ferrocenyloxazoline,
(N,P ligand)],13 and Li [AgI–Taniaphos type ligand (N,P
ligand)].14 There are also some reports on enantio- and
exo-selective reactions, although these are scarce.15 As a dra-
matic example of the influence of the Lewis acid catalyst on
the stereochemical outcome of the reaction, Hou et al.15c
We have previously described a sterically and electronically
tuneable family of bidentate 1,2-disubstituted ferrocenyl
ligands16 bearing phosphorus and sulfur heteroatoms as co-
ordinating groups, and the presence of planar chirality as the
only source of asymmetry (Fig. 1). These sulfenylphosphino-
ferrocenes, named Fesulphos ligands (1),17 have proven to
be excellent chiral ligands in a variety of highly enantio-
selective Pd- and Cu-catalyzed reactions, such as allylic sub-
stitution reactions,17a,b formal aza-Diels–Alder reactions of
N-sulfonyl imines,17c Diels–Alder reaction,17d ring-opening
of meso heterobicyclic alkenes,17e,f and Mannich-type reac-
tions of N-sulfonyl imines.17g Extending the applicability of
Fesulphos ligands in asymmetric catalysis, herein we de-
scribe in detail that their silver(I) and copper(I) complexes,
especially the latter, function as highly efficient chiral Lewis
acids in the 1,3-dipolar cycloaddition of azomethine ylides
with electrophilic alkenes, providing very good reactivity,
endo/exo selectivity, and exceptional levels of enantioselec-
tivity (up to >99% ee).18 Furthermore, in contrast to most of
Keywords: 1,3-Dipolar cycloaddition; Azomethine ylide; Asymmetric cata-
lysis; Chiral ligand; Copper; Enantioselectivity; Dipolarophile.
* Corresponding author. Tel.: +34 91 497 3925; fax: +34 91 497 3966;
e-mail: juancarlos.carretero@uam.es
0040–4020/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2007.03.130

6588
S. Cabrera et al. / Tetrahedron 63 (2007) 6587–6602
t
O
R2
O
Ph
S Bu
OR1
PR2
R
H
OEt
R
Me
Fe
H2N
Ph
N
O
3a-e
2a-h
O
Fesulphos ligands
2a R = Ph
3a R = Ph
8e
4a R1
=
Me, R2 = H
=
Me, R2 = Me
4b R1 = t-Bu, R2 = H
2b R = 1-Naph
2c R = (p-F)C6H4
3b R = (p-Cl)C6H4
3c R = (p-OMe)C6H4
1a R = Ph
4c R1
1b R = (p-F)C6H4
1c R = o-Tol
1d R = 1-Naph
1e R = Cy
2d R = (p-OMe)C6H4 3d R = i-Pr
2e R = (o-Me)C6H4
2f R = 2-Py
2g R = Cy
2h R = t-Bu
Figure 1.
the chiral catalysts reported to date, the CuI–Fesulphos sys-
tem shows a broad scope with regard to the dipolarophile. A
mechanistic study based on NMR experiments and theoret-
ical calculations is also provided to rationalize the observed
enantioselectivity.
Table 1. Synthesis of a-imino esters 5, 6, 7, and 8
Entry Aldehyde/ a-Amino a-Imino ester
Conditionsa Yieldb
(%)
ketone
ester
OMe
R
N
1
2a
4a
A
89
O
5a R = Ph
2
3
4
5
6
7
8
2b
2c
2d
2e
2f
4a
4a
4a
4a
4a
4a
4a
5b, R¼2-Naph
5c, R¼(p-F)C6H4
5d, R¼(p-OMe)C6H4
5e, R¼(o-Me)C6H4
5f, R¼2-Py
A
A
A
A
A
B
B
84
56
72
80
60
37
24
2. Results and discussion
2.1. Preparation of a-imino esters
2g
2h
5g, R¼Cy
Awide variety of a-imino esters were readily prepared from
reaction of aromatic and aliphatic aldehydes and ketones
with different amino esters (Fig. 1).
5h, R¼t-Bu
t
O Bu
Ph
N
9
2a
2a
4b
A
80
O
6
a-Imino esters derived from aldehydes were prepared fol-
lowing the procedure described in the literature.19 Thus,
reaction of aromatic aldehydes 2a–f with methyl glycinate
(4a) hydrochloride or tert-butylglycinate (4b) hydrochloride
in CH2Cl2 in the presence of Et3N and MgSO4 at room tem-
perature gave a-imino esters 5a–f, and 6 in good yields
(Table 1, entries 1–6 and 9). Similarly, the bulky aliphatic
imino esters 5g and 5h were prepared by condensation of
the corresponding aldehydes (2g and 2h) with methyl glyci-
nate, which was obtained by treatment of methyl glycinate
hydrochloride (4a) with NH4OH (Table 1, entries 7 and 8).
The lower yields obtained in the case of the aliphatic a-imino
esters (24–37%) are likely due to the lower reactivity of
their starting aldehydes and, especially, their lower stability.
Substituted a-imino esters with an extra methyl group at the
a carbon (7a–d) were prepared from methyl alaninate hy-
drochloride (4c) in good yields (Table 1, entries 10–13). It
was also of great interest to study the reactivity of imino es-
ters derived from ketones, a type of azomethine ylides never
tested in asymmetric catalytic 1,3-dipolar cycloadditions.
The Schiff base derived from benzophenone and ethyl glyci-
nate is commercially available (8e). Other a-imino esters de-
rived from ketones were prepared in moderate yield by direct
condensation in benzene in the presence of molecular sieves
OMe
R
N
10
4c
A
87
O
7a, R = Ph
11
12
13
2b
2c
2d
4c
4c
4c
7b, R¼2-Naph
7c, R¼(p-OMe)C6H4
7d, R¼(o-Me)C6H4
A
A
A
72
70
73
OMe
R
14
3a
4a
C
16 (30)c
N
O
8a, R = Ph
d
15
16
17
3b
3c
3d
4a
4a
4a
8b, R¼(p-Cl)C6H4
8c, R¼(p-OMe)C6H4
8d, R¼i-Pr
C
C
C
—
—
e
38
a
Conditions A: Et3N and MgSO4, in CH2Cl2 at room temperature; B:
˚
MgSO4, in CH2Cl2 at room temperature; C: Molecular sieves 4 A,
benzene at room temperature.
b
c
d
e
Isolated yield.
In parenthesis, conversion measured by 1H NMR.
A 2.2:1 mixture of ketone/ketimine was obtained.
A mixture of ketimine/enamine was obtained.
ketone 3d was isolated in 38% yield after evaporation of
the remaining starting ketone under low pressure (Table 1,
entry 17).
˚
4 A. The ketimine 8a, derived from acetophenone, was ob-
tained in low yield, although it could be easily purified by
crystallization in hexane (Table 1, entry 14). The Schiff
base derived from p-chloroacetophenone (3b) was obtained
as an inseparable 2.2:1 mixture of ketone (3b)/ketimine (8b)
(Table 1, entry 15), and it was directly used in the cycload-
dition reactions. The ketimine derived from p-methoxyace-
tophenone (imine 8c) could not be purified due to its
prompt isomerization to the corresponding enamine (Table 1,
entry 16). Finally, the Schiff base derived from the aliphatic
2.2. Asymmetric Fesulphos-catalyzed 1,3-dipolar
cycloaddition of azomethine ylides with maleimides
As model reaction we selected the asymmetric 1,3-dipolar
cycloaddition of N-phenyl maleimide (9a) with the a-imino
ester 5a in the presence of catalytic amounts of a base
(Et3N), Fesulphos ligand 1a, and a metal salt (Table 2).
This reaction had been previously studied by Komatsu

S. Cabrera et al. / Tetrahedron 63 (2007) 6587–6602
Table 2. 1,3-Dipolar cycloaddition of 5a with maleimides 9a and 9b
6589
R
N
O
1a (x mol%),
O
O
O
[M] (x mol%)
Ph
N
R
N
+
OMe
Et3N (18 mol%), CH2Cl2
25 °C
CO2Me
Ph
N
H
O
5a
9a R = Ph
9b R = Me
endo-10a, R = Ph
endo-11, R = Me
Entry
Maleimide
[M]
CuCl
Cu(CH3CN)4PF6
Cu(CH3CN)4ClO4
Cu(CH3CN)4ClO4
Cu(CH3CN)4ClO4
Cu(CH3CN)4ClO4
AgClO4
AgOAc
AgOAc
Pd(CH3CN)2Cl2
Pd(CH3CN)2Cl2
X (mol %)
t (h)
endo/exoa
Yieldb (%)
eec (%)
1
2
3
4
5
6
7
8
9a
9a
9a
9a
9a
9b
9a
9a
9a
9a
9a
10
10
10
1
0.5
3
10
10
1
10
10
24
94:6
96:4
>98:<2
95:5
97:3
>98:<2
96:4
>98:<2
>98:<2
96:4
51
74
78
80
86
97
86
64
90d
60
65
30
>99
>99
>99
>99
>99
89
86
86
0
0.25
0.25
0.25
1
0.25
2
0.25
2
48
48
9
10
11e
92:8
0
Determined by 1H NMR of the crude reaction mixture.
Yield of endo product after column chromatographic purification.
Determined by HPLC (Daicel Chiralpak AS-H).
a
b
c
d
e
Conversion measured by 1H NMR spectroscopy of the crude reaction mixture.
AgBF4 (20 mol %) was added.
et al. by using Cu(OTf)2/Binap or Cu(OTf)2/Segphos as
catalyst system,15a describing that under these conditions
the reaction occurred at room temperature in 24–48 h with
very high exo-selectivity, albeit moderate enantioselectivity
(up to 72% ee). In contrast to these results, the cycloaddition
between 5a and 9a catalyzed by CuCl–Fesulphos 1a
(10 mol %) proved to be highly endo-selective (endo/exo¼
94:6), yielding endo-10a with moderate chemical yield and
30% ee (Table 2, entry 1). To our delight, more electrophilic
CuI complexes, such as Cu(CH3CN)4PF6 or Cu(CH3CN)4-
ClO4, improved dramatically the yields (74–78%) and espe-
cially the enantioselectivity, affording endo-10a practically
as a single enantiomer (endo/exo¼>94:<6, >99% ee, en-
tries 2 and 3). Furthermore, the reactions took place in very
short reaction times (15 min). Due to this veryhigh reactivity,
the catalyst loading could be reduced from 10 to 1 mol %
without compromising the yield, selectivity or reaction times
(entry 4). Similar results were also obtained reducing the cat-
alyst loading to 0.5 mol % (entry 5), which resulted only in
slightly longer reaction times (1 h). Nearly complete endo-
selectivity (endo/exo¼>98:<2) and enantioselectivity
(>99% ee) were also obtained in the CuI–Fesulphos cycload-
dition to N-methylmaleimide (entry 6).
slower (24 h) and led to racemic endo-10a (entry 10). An
identical result was obtained by using Pd(CH3CN)2Cl2 in
combination with AgBF4 to facilitate the formation of
a more electrophilic palladium complex (entry 11).
Encouraged by the excellent results obtained with the
Cu(CH3CN)4ClO4/Fesulphos catalyst system, we extended
the study of the 1,3-dipolar cycloaddition reaction of N-
phenyl maleimide to a wide variety of sterically and elec-
tronically different a-imino esters derived from aldehydes
or ketones (Table 3). Aromatic a-imino esters derived
from aldehydes (R1¼H; imines 5a–e) gave excellent results.
As in the case of the phenyl a-imino ester 5a (Table 3, entry
1), the p-methoxy and o-methyl substituted substrates (5d
and 5e) gave cycloadducts endo-10d and 10e in good yields
and excellent enantioselectivities at room temperature with
1 mol % of catalyst loading (Table 3, entries 7 and 9). Under
identical reaction conditions the 2-naphthyl and p-fluoro-
phenyl substituted imino esters (5b and 5c) afforded the
corresponding pyrrolidines endo-10b and 10c with high
endo-selectivity but moderate enantioselectivity (72 and
84% ee, entries 3 and 5). Fortunately, this loss of asymmetric
induction could be overcome by lowering the reaction tem-
perature to ꢁ10 ꢂC and increasing the catalyst loading to
3 mol %, resulting in the formation of a single endo enantio-
mer in both cases (>99% ee, entries 4 and 6). In contrast, the
heteroaromatic imino ester 5f (R¼2-Py) gave a complex
reaction mixture (entry 11), likely due to the basic and/or
metal coordinating nature of the pyridyl unit. On the other
hand, although the reaction of the aliphatic imino esters 5g
and 5h was also completely endo-selective (entries 12 and
13), the reactivity and the enantioselectivity were much
lower than in the case of the aromatic azomethine ylides.
Since silver salts (e.g., AgClO4 and AgOAc) have been
widely used as Lewis acids in catalytic asymmetric 1,3-di-
polar cycloadditions of azomethine ylides,10b–14 we also
studied the Fesulphos-mediated cycloaddition of the model
imine 5a with phenyl maleimide in the presence of a catalytic
amount (1–10 mol %) of AgClO4 or AgOAc (entries 7–9).
These reactions took place with excellent endo-selectivity
and good enantioselectivity (86–89% ee), albeit not as
high as the CuI/Fesulphos-catalyzed process (>99% ee),
showing the superiority of the Cu-catalyzed process. Unlike
the Cu- and Ag-promoted reactions, the process catalyzed by
a palladium Lewis acid, such as Pd(CH3CN)2Cl2, was much
Next we studied the reactivity of C-2 substituted imino
esters, a kind of ylide precursor that had been scarcely tested

6590
S. Cabrera et al. / Tetrahedron 63 (2007) 6587–6602
Table 3. 1,3-Dipolar cycloaddition of a variety of imino esters with N-phenyl maleimide (9a)
Ph
N
O
O
O
R3
1a (x mol%),
O
Cu(CH3CN)4ClO4 (x mol%)
R
N
R1
Ph
N
OR2
Et3N (18 mol%), CH2Cl2
R3
R1
R
CO2R2
N
H
O
endo-10a-h, R1 = H, R2 = Me, R3 = H
endo-12a-d, R1 = H, R2 = Me, R3 = Me
endo-13a-b,d, R1 = Me, R2 = Me, R3 = H
5a-h, R1 = H, R2 = Me, R3 = H
7a-d, R1 = H, R2 = Me, R3 = Me
8a-b,d, R1 = Me, R2 = Me, R3 = H
9a
R1 = Ph, R2 = Et, R3 = H
endo-13e,
R1 = Ph, R2 = Et, R3 = H
8e,
Entry
R/Imine
Cu/1a (mol %)
T (ꢂC)
t (h)
0.25
0.50
5
1
5
1
8
1
2
0.50
Product
endo/exoa
Yieldb (%)
eec (%)
1
2
3
4
5
6
7
8
Ph/5a
Ph/5a
1
3
1
3
1
3
1
3
1
3
3
3
3
3
3
3
3
3
3
3
3
25
ꢁ10
25
10a
10a
10b
10b
10c
10c
10d
10d
10e
10e
10f
10g
10h
12a
12b
12c
12d
13a
13b
13d
13e
95:5
80d
81
63
81
>99
>99
72
>99
84
>99
>99
>99
>99
>99
—
>98:<2
96:4
97:3
2-Naph/5b
2-Naph/5b
(p-F)C6H4/5c
(p-F)C6H4/5c
(p-OMe)C6H4/5d
(p-OMe)C6H4/5d
(o-Me)C6H4/5e
(o-Me)C6H4/5e
2-Py/5f
ꢁ10
25
>98:<2
>98:<2
>98:<2
>98:<2
>98:<2
>98:<2
71
ꢁ10
25
82
73d
81
ꢁ10
9
25
62
85
—
20
15
50
78
10
11
12
13
14
15
16
17
19
20
21
22
ꢁ10
ꢁ10
25
e
12
—
Cy/5g
t-Bu/5h
Ph/7a
24
24
48
1
24
24
1
1
24
1
>98:<2
>98:<2
>98:<2
>98:<2
>98:<2
>98:<2
>98:<2
>98:<2
NDf
40
51
80
25
ꢁ10
ꢁ10
ꢁ10
ꢁ10
ꢁ10
ꢁ10
ꢁ10
ꢁ10
2-Naph/7b
(p-OMe)C6H4/7c
(o-Me)C6H4/7d
Ph/8a
92
<10f
<20f
78
NDg
NDg
94
(p-Cl)C6H4/8bh
i-Pr/8d
8e
80
>99
NDg
93
<15f
92
>98:<2
Determined by 1H NMR of the crude reaction mixture.
a
b
c
d
e
f
Yield of endo product after column chromatographic purification.
Determined by HPLC (Daicel Chiralpak AS-H or Chiralcel OD).
The yield increased to >96% in the presence of an excess of 5.
Complex reaction mixture.
Conversion measured by 1H NMR from the crude reaction mixture.
Not determined.
A 2.2:1 mixture of ketone (2b)/ketimine (8b) was used.
g
h
in asymmetric catalytic 1,3-dipolar cycloadditions12 despite
the fact that the addition products are pyrrolidines with a qua-
ternary stereogenic carbon at C-2. The results obtained from
a set of four imino esters derived from alanine (R3¼Me, 7a–
d) proved to be very dependent on the aryl substitution at the
nitrogen atom. Thus, while the cycloaddition of the imino
esters 7a and 7b occurred with satisfactory yield, complete
endo-selectivity and high enantioselectivity (80–92% ee, en-
tries 14 and 15), almost no reaction was observed in the case
of the imino esters 7c and 7d (entries 16 and 17). Finally, we
tested the glycinates derived from ketones (imines 8a–e),
a kind of azomethines never studied in asymmetric catalytic
1,3-dipolar cycloadditions. The commercially available im-
ine 8e derived from benzophenone (R¼R1¼Ph) provided the
pyrrolidine cycloadduct endo-13e with high enantioselectiv-
ity (entry 22, 93% ee). The reactions of the more challenging
azomethine ylides derived from acetophenones 8a and 8b
were also completely endo-selective and highly enantio-
selective (94–>99% ee, entries 19 and 20), giving rise to
a single pyrrolidine with a quaternary stereogenic center at
C-5 with full control of the four stereogenic centers. Unlike
these results, no reaction was observed from the bulky di-
alkyl imine 8d, derived from isopropyl methyl ketone
(entry 21).
2.3. 1,3-Dipolar cycloaddition of azomethine ylides with
other symmetrical 1,2-disubstituted alkenes
Taking into account the stereocontrol achieved by the CuI–
Fesulphos catalyst system in the reaction with maleimides,
we extended the study to other dipolarophiles, starting
with other electronically poor symmetrical disubstituted
alkenes.
2.3.1. Reaction with dimethyl maleate. While the 1,3-di-
polar cycloaddition of azomethine ylides with maleimides
catalyzed by CuI–Fesulphos or Ag–Fesulphos was highly
endo-selective, the results obtained with dimethyl maleate
were less homogeneous (Table 4). In our first attempt, we
used the same reaction conditions to those employed for the
cycloaddition of N-phenyl maleimide (9a). Unlike the behav-
ior of maleimide, the reaction of 5a and dimethyl maleate
(14) in CH2Cl2 catalyzed by CuI/Fesulphos (1a)/Et3N was
highly exo-selective (endo-15/exo-15¼10:90, entry 1). The
use of diisopropyl ethyl amine (DIPEA) instead of Et3N
gave very similar results (entry 2). The stereoselectivity of
this reaction proved to be highly solvent dependent. Thus,
in THF or toluene the amount of endo-15 increased dramat-
ically, being the major adduct in THF (entries 3 and 4),

S. Cabrera et al. / Tetrahedron 63 (2007) 6587–6602
Table 4. 1,3-Dipolar cycloaddition of 5a with dimethyl maleate (14)
6591
(R)-1 (3 mol%) /
Cu(CH3CN)4ClO4 or AgOAc
(3 mol%)
CO2Me
CO2Me
CO2Me
MeO2C
Ph
MeO2C
Ph
CO2Me
CO2Me
14
Ph
N
CO2Me
+
+
CO2Me
N
H
N
H
Base (18 mol%), 25 °C
endo-15
exo-15
5a
Entry
[M]
Solvent
Base
Et3N
T (ꢂC)
1
t (h)
endo/exoa
endo-15
exo-15
Yieldb (%)
eec (%)
Yieldb (%)
eec (%)
1
2
3
4
5
6
7
8
CuI
CuI
CuI
CuI
CuI
CuI
AgI
AgI
AgI
AgI
AgI
AgI
AgI
AgI
CH2Cl2
CH2Cl2
THF
25
25
25
25
25
25
25
25
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1b
1c
1d
1e
48
24
24
24
24
48
24
1
1
6
6
3
10:90
7:93
—
—
47
41
33
—
36
67
75
71
82
87
70
80
—
—
94
94
93
—
66
68
67
75
69
79
89
8
82
78
32
30
59
—
23
21
12
10
13
—
DIPEA
Et3N
DIPEA
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
Et3N
67:33
67:33
41:59
THF
Toluene
CH3CN
CH2Cl2
THF
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
d
—
54:46
85:15
90:10
>98:<2
>98:<2
66:34
NDe
NDe
NDe
—
NDe
NDe
NDe
—
9
25
10
11
12
13
14
ꢁ10
ꢁ10
ꢁ10
ꢁ10
ꢁ10
—
—
NDe
NDe
NDe
NDe
NDe
NDe
3
3
71:29
90:10
Determined by 1H NMR of the crude reaction mixture.
Yield of the pure product after column chromatographic purification.
Determined by HPLC (Daicel Chiralpak AS-H).
<20% Conversion by 1H NMR.
a
b
c
d
e
Not determined.
whereas no reaction was observed in acetonitrile (entry 6). It
is important to note that both in toluene and THF endo-15
was obtained in very high ee (93–94% ee, entries 3–5), while
the exo adduct was obtained with very low enantioselectivity
(10–13%).
toluene at ꢁ10 ꢂC good endo-selectivities were obtained
with a variety of Fesulphos ligands (entries 10–14), espe-
cially with ligands 1a and 1b, whose cycloadditions led
exclusively to the endo adduct (entries 10 and 11). However,
the enantioselectivity of these silver-promoted reactions
was significantly lower (66–75% ee with ligand 1a) than
that obtained in the copper-mediated process (93–94% ee).
As indicated before, a similar trend was observed in the
reaction with N-phenyl maleimide (Table 2).
As mentioned in the introduction, silver Lewis acids have
been frequently employed in endo-selective catalytic asym-
metric 1,3-dipolar cycloadditions of azomethine ylides.10–14
Thus, in an attempt to improve the endo-selectivity of the re-
action with dimethyl maleate, we turned to the combination
AgOAc–Fesulphos (entries 7–14). Unlike the Cu–Fesulphos
mediated reaction, in all cases the endo adduct was obtained
as the major one. A brief study on the effect of the solvent
showed that toluene provided the best enantioselectivity
(entries 7–9). In particular, by performing the reaction in
2.3.2. Reaction with dimethyl fumarate and fumaroni-
trile. Dimethyl fumarate (16) showed higher reactivity and
endo-selectivity20 than dimethyl maleate (Table 5, entries
1–3). As in the case of dimethyl maleate, the best endo-
selectivity was achieved in THF, reaching a 90:10 endo/
exo ratio at ꢁ10 ꢂC (entry 3). Under these conditions the
Table 5. 1,3-Dipolar cycloaddition of 5a with dimethyl fumarate (16) and fumaronitrile (17)
R
R
R
R
(R)-1a (3 mol%) /
R
Ph
N
CO2Me
Cu(CH3CN)4ClO4 (3 mol%)
+
+
CO2Me
Ph
Ph
CO2Me
N
H
N
H
Et3N (18 mol%)
R
16, R = CO2Me
17, R = CN
endo-18, R = CO2Me
endo-19, R = CN
exo-18, R = CO2Me
exo-19, R = CN
5a
Entry
Alkene
Solvent
T (ꢂC)
t (h)
Product
endo/exoa
Yieldb (%)
eec (%)
1
2
3
4
5
6
16
16
16
17
17
17
CH2Cl2
THF
THF
CH2Cl2
THF
THF
25
25
24
1
5
0.25
0.25
1
18
18
18
19
19
19
73:27
88:12
90:10
20:80
21:79
20:80
59
82
89
73
71
78
97
>99
>99
55
71
76
ꢁ10
25
25
ꢁ30
Determined by 1H NMR of the crude reaction mixtures.
Yield of the major adduct after column chromatographic separation.
Determined by HPLC (Daicel Chiracel OD).
a
b
c

6592
S. Cabrera et al. / Tetrahedron 63 (2007) 6587–6602
cycloadduct endo-18 was isolated in 89% yield in practically
enantiopure form (>99% ee). A very different stereochem-
ical outcome was found in the reaction with fumaronitrile
(17) (Table 5, entries 4–6). This dipolarophile was quite re-
active, allowing to perform the reaction at ꢁ30 ꢂC (entry 6),
the reaction being moderately exo-selective (endo/exo¼
20:80) and enantioselective (up to 76% ee).
isolated in 62% yield and 95% ee (Eq. 1). The use of CH2Cl2
as solvent or bulkier a,b-unsaturated esters, such as tert-
butyl acrylate or methyl crotonate, resulted in a sluggish
reaction. For instance, the reaction with tert-butyl acrylate
resulted in 20% conversion after 24 h in THF.
In spite of the chemical interest of the resulting highly
functionalized bicyclic pyrrolidines, to the best of our
knowledge, a,b-unsaturated lactones had never been tested
in asymmetric 1,3-dipolar cycloaddition with azomethine
ylides. The Cu–Fesulphos mediated reaction of 5a with
2-butenolide (22) under the usual reaction conditions led
to a 60:40 mixture of endo/exo isomers 23. After standard
silica gel chromatographic separation, endo-23 was isolated
in 31% yield in 76% ee. This enantioselectivity could be
slightly improved to 80% ee by performing the reaction in
CH2Cl2 instead of THF (Eq. 2).
2.4. 1,3-Dipolar cycloaddition of azomethine ylides with
non-symmetrical alkenes
Extending the scope of the Cu–Fesulphos catalyst system in
asymmetric 1,3-dipolar cycloadditions with azomethine
ylides, we next undertook the study of a wide variety of
monoactivated dipolarophiles, such as methyl acrylate
(20), 2-butenolide (22), methacrolein (24), and b-nitrostyr-
ene (26). In all cases the cycloadditions occurred with
complete regiocontrol (exclusive formation of the 2,4,5-
trisubstituted pyrrolidine) and good enantioselectivity
(69–95% ee), albeit the endo/exo-selectivity showed high
dependence on the dipolarophile substitution. The most
relevant results are shown in Scheme 1.
Although some examples of catalytic asymmetric 1,3-di-
polar cycloadditions of a,b-unsaturated aldehydes with
nitrones have been reported,2h,21 there are no examples of
cycloadditions with azomethine ylides. Moreover, the use
of a-substituted acroleins would provide a synthetic route
to pyrrolidines with a quaternary carbon at C-4. As shown
in Eq. 3, the cycloaddition of 5a with methacrolein (24)
The reaction with methyl acrylate in THF was moderately
endo-selective (endo/exo¼75:25), the adduct endo-21 being
MeO2C
Ph
MeO2C
Ph
(R)-1 (3 mol%) /
CO2Me
+
Cu(CH3CN)4ClO4 (3 mol%)
Ph
N
CO2Me
(1)
+
CO2Me
CO2Me
N
H
N
H
Et3N (18 mol%),
THF, 25 °C, 24 h
endo-21
exo-21
5a
20
endo/exo = 75:25
endo: 62% yield, 95% ee
O
O
O
O
O
(R)-1 (3 mol%) /
(2)
+
Cu(CH3CN)4ClO4 (3 mol%)
Ph
N
CO2Me
O
+
Ph
CO2Me
Ph
CO2Me
N
N
H
Et3N (18 mol%),
THF, 25 °C, 24 h
H
5a
22
endo-23
exo-23
endo/exo = 60:40
endo: 31% yield, 76% ee
(80% ee in CH2Cl2)
OHC
OHC
Ph
(R)-1 (3 mol%) /
CHO
+
Cu(CH3CN)4ClO4 (3 mol%)
(3)
Ph
N
CO2Me
+
CO2Me
CO2Me
Ph
N
H
N
H
Et3N (18 mol%),
THF, 25 °C, 24 h
5a
24
endo-25
exo-25
endo/exo = >98:<2
endo: 55% yield, 41% ee;
(69% ee in CH2Cl2)
O2N
Ph
Ph
CO2Me
O2N
Ph
Ph
CO2Me
(R)-1 (3 mol%) /
NO2
+
Cu(CH3CN)4ClO4 (3 mol%)
Ph
N
CO2Me
(4)
+
N
H
N
H
Et3N (18 mol%),
THF, T, 0.5-1 h
Ph
5a
26
endo-27
exo-27
T = 25 °C, endo/exo = 14:86, exo: 65% yield, 92% ee
T = -10 °C, endo/exo = 5:95, exo: 61% yield, 94% ee
Scheme 1. Catalytic and asymmetric 1,3-dipolar cycloaddition with monoactivated alkenes.

S. Cabrera et al. / Tetrahedron 63 (2007) 6587–6602
6593
under the standard reaction conditions took place with com-
plete regio- and endo-selectivities, albeit the cycloadduct
endo-25 was obtained with moderate enantioselectivity
(41% ee; Eq. 3). After a brief study of solvents, this asym-
metric induction could be improved to 69% ee by perform-
ing the reaction in CH2Cl2.
configuration of new cycloadducts was assigned assuming
a similar reaction pathway, and by 1D and 2D NMR exper-
iments. The study of the H NMR data of the maleimide
1
adducts revealed that the endo adducts have a big coupling
constant for protons H-4 and H-5 (J4,5¼8–9 Hz), while
this value is much lower for the exo adducts (J4,5
¼
4–5.5 Hz). A couple of examples is shown in Figure 2.25 Fur-
thermore, the protons of the methoxy carbonyl group at C-2
are more deshielded for endo-10a than for exo-10a. This
difference in the chemical shift is likely due to the ring-
current effect of the carbonyl group at C-3, which is in a
cis relationship with the methoxy carbonyl group in endo-
10a. A similar effect is observed in pyrrolidines with a
quaternary carbon, such as endo-12 and endo-13.
The first systematic study on nitroalkenes in catalytic asym-
metric 1,3-dipolar cycloadditions with azomethine ylides
has been recently reported by Hou et al.15c In addition, the
ꢀ
group of Cossıo has described that pyrrolidines bearing a di-
peptide at C-2 and a nitro group at C-4 are potent inhibitors
of the enzyme a4b1-integrine, which is involved in the
hepatic melanoma metastasis.22 We were pleased to see that
the 1,3-dipolar cycloaddition of (E)-b-nitrostyrene (26) with
the model azomethine ylide precursor 5a occurred rapidly,
providing the C4-nitro-pyrrolidine adduct 27 with good
stereochemical control (Eq. 4). In this case, similar results
were obtained in CH2Cl2 and THF, although a slightly higher
enantioselectivity was achieved in THF. This dipolarophile
proved to be very reactive, allowing to perform the reaction
at ꢁ10 ꢂC (60 min reaction time), leading to the adduct 27
with excellent exo-selectivity (endo/exo¼5:95) and enantio-
selectivity (94% ee).
In the case of adductsderived fromacyclic dipolarophiles, the
chemical shift of the protons of the substituent at C-4 has
a great diagnostic value. Thus, this chemical shift is substan-
tially lower for the endo adduct (Ph group at C-5 and substit-
uent at C-4 in cis arrangement) than for the exo diastereomer,
likely due to the shielding ring-current effect of the phenyl
group at C-5 on the cis substituent at C-4. For example, the
methoxy carbonyl group at C-4 in adduct endo-15, endo-
18, and endo-21 appears around 0.6 ppm more shielded
(3.1–3.2 ppm) than the methoxy carbonyl groups on C-2
and C-3 (at 3.6–3.7 ppm; Fig. 4). Similarly, the signal corre-
sponding to the aldehyde proton of cycloadduct endo-25 also
appears at higher field (9.1 ppm) than expected (w9.7 ppm).
Additionally, the stereochemical assignment of the new
pyrrolidines endo-12a, endo-13a, endo-14e, endo-23, and
endo-25 was confirmed by NOE and NOESY experiments.
Figure 3 shows some relevant NOE correlations.
2.5. Stereochemical assignment of the cycloadducts
The assignment of the endo/exo configuration of cyclo-
adducts endo-10a,11,15a endo-10d,15a endo-11,11,15a endo-
12a,23 endo-15,11 endo-18,10a exo-19,15a exo-21,10a and
exo-2724 was made by comparison of their NMR spectro-
scopic data with those reported in the literature. The
Ph
N
Ph
N
O
O
O
O
H
4
H
Ph
H
4
5 N
Ph
5 N
J4,5= 5.1 Hz
CO2Me
CO2Me
J4,5= 8.6 Hz
H
H
Me= 3.88 ppm
Me= 3.65 ppm
H
endo-10a
exo-10a
Figure 2.
Me= 3.64 ppm
CO2Me
CO2Me
Me= 3.72 ppm
Me= 3.16 ppm
MeO2C
Ph
Me= 3.62 ppm
CO2Me
Me= 3.58 ppm
MeO2C
Ph
CO2Me
N
H
N
H
Me= 3.74 ppm
endo-15
exo-15
Ph
Ph
N
Ph
N
O
H3b
N
O
O
O
O
O
H3
OHC
O
O
H3
H3a
CO2Me
H2
H4
Ph
H5
H3
H4
CO2Me Ph
Me
H4
Ph
H5
H3
H4
Ph
Ph
Me
Ph
H5
CO2Me
H2
CO2Me
CO2Et
H2
N
H
N
N
H
N
H
N
H
Me
H2
H
endo-12a
endo-13a
endo-13e
endo-23
endo-25
H3b-CHO = 1%
H3a-Mea = 2%
H5-Mea = 2%
H2- H3a = 2%
H3-H4 = 3%
H2-Mea = 5%
H2-H3 = 5%
H3-H4 = 3%
H4-H5 = 5%
H2-H3 = 4%
H2-H5 = 4%
H3-H4 = 5%
H4-H5 = 8%
H5-Mea = 6%
H3-H4 = 5%
H3-H2 = 8%
Figure 3.

6594
S. Cabrera et al. / Tetrahedron 63 (2007) 6587–6602
Me
N
Ph
N
O
O
O
O
MeO2C
Ph
CO2Me
CO2Me
CO2Me
Ph
CO2Me
Ph
N
H
N
H
N
H
endo-10a
endo-11
endo-15
[α]D20 +114 (c 1.22, CH2Cl2); >99% ee
Lit. [α]D20 +91.1 (c 1.22, CH2Cl2); 71% ee
[α]D20 +51 (c 0.14, CH2Cl2); >99% ee
Lit. [α]D20 +61 (c 1.18, CH2Cl2); 79% ee
[α]D20 +100 (c 0.05, CH2Cl2); 94% ee
Lit. [α]D20 +64.7 (c 1.07, CH2Cl2); 87% ee
O2N
Ph
Ph
CO2Me
MeO2C
MeO2C
Ph
CO2Me
CO2Me
CO2Me
Ph
N
H
N
H
N
H
exo-27
endo-21
endo-18
[α]D20 +36 (c 0.08, CH2Cl2); 95% ee
[α]D20 +43 (c 0.10, CH2Cl2); >99% ee
Lit. [α]D20 +20 (c 0.35, CH2Cl2); 76% ee
[α]D20 +105 (c 0.11, CH2Cl2); 94% ee
Lit. [α]D20 +38 (c 0.10, CH2Cl2); 88% ee
Lit. [α]D20 +118.2 (c 1.10, CHCl3); 97% ee
Figure 4.
Finally, the absolute configuration of the cycloadducts was
established by comparison of the optical rotation values of
(+)-endo-10a,11 (+)-endo-11,11 (+)-endo-15,11 (+)-endo-
18,10a (+)-endo-21,10a and (+)-exo-2715c with those reported
in the literature (Fig. 4).
ClO4
t
Bu
t
S Bu
PPh2
S
1. Cu(CH3CN)4ClO4
CH2Cl2
Cu(5a)
P
Fe
Fe
Ph2
2. Ph
N
CO2Me
(5a)
28
(R)-1a
100% conversion
2.6. Mechanistic considerations
Scheme 2.
2.6.1. Isolation of a Fesulphos/Cu/ylide precursor com-
plex. To shed some light on the origin of the excellent enan-
tioselectivity achieved by the Cu–Fesulphos catalyst system
in the reaction with a very wide variety of azomethine ylides
and dipolarophiles, we first undertook the isolation and char-
acterization of the presumed chiral copper-ligand-imine
complex intermediate. Thus, treatment of (R)-1a with
Cu(CH3CN)4ClO4 in CH2Cl2, followed by addition of 5a
at room temperature, yielded after evaporation of the solvent
a relatively stable yellow complex (presumed compound 28)
(Scheme 2).
The formation of the presumed copper complex 28, in which
the metal atom is simultaneously coordinated with the ligand
(R)-1a and the imine 5a, is in full agreement with the high
deshielding effect observed for the imine and methylene
protons of the unit 5a. Also, the methylene protons of 5a
become an AB system upon coordination with the chiral
ligand. Additionally, a significant deshielding effect is also
observed for the three protons of the substituted Cp ring of
the ligand (Table 6).
Table 6. 1H NMR selected data of 1a, 5a, and complex 28 (in CDCl3)
H observed
ClO4
t
t
Ph
N
CO2Me
S Bu
PPh2
Bu
S
Fe
Cu(5a)
P
Fe
Ph2
(R)-1a
5a
28
Shift (ppm)/multiplicitya
Shift (ppm)/multiplicitya
Shift (ppm)/multiplicitya
8.97/s
5.11 and 5.08/AB system
N]CH
CH2
8.26/s
4.39/s
3.75/s
CO2CH3
Cp-Hb
Cp-Hb
Cp-Hb
3.95/s
4.71–4.67/m
4.50–4.46/m
4.15–4.12/m
3.98/s
4.93/br s
4.81/br s
4.59/br s
3.99/s
c
Cp0-H5
C(CH3)3
1.00/s
1.02/s
a
s¼singlet, br s¼broad singlet, and m¼multiplet.
Disubstituted cyclopentadienyl ring.
Unsubstituted cyclopentadienyl ring.
b
c

S. Cabrera et al. / Tetrahedron 63 (2007) 6587–6602
6595
2% NOE
Ph
hindered t-Bu group, could explain the high enantioselectiv-
ity observed in the reaction with most of the dipolarophiles
(Scheme 3).
ClO4
ClO4
t
t
OMe
Bu
Bu
O
Cu
N
S
S
Cu
P
P
N
Ph
Fe
Fe
O
Ph2
Ph2
ClO4
OMe
t
Bu
Ph
X
X
N
X
X
S
Cu
P
Fe
O
Ph2
OMe
(RP,R)-28b
(RP,R)-28a
2 CO2Me
Ph
5
N
H
Figure 5.
(2S,5R)
28a
Based on the X-ray structure of other Fesulphos–CuI com-
plexes,17c we proposed for complex 28 a distorted tetrahedral
arrangement of the ligands in the coordination sphere of the
Cu atom, and the formation of a single epimer at the sulfur
atom, the one with the bulky t-Bu group in anti relationship
with regard to the ferrocene moiety. Two different structures
can then be proposed for complex 28, depending on the ori-
entation of the a-imino ester (5a), as shown in Figure 5. Al-
though, unfortunately, suitable crystals for X-ray diffraction
analysis could not be obtained, we tentatively proposed the
structure 28a based on the NOE observed between the t-Bu
group and the protons of the phenyl ring at the imine moiety.
Scheme 3.
2.6.2. Computational studies. In order to gain a better
understanding of the structure and reactivity of these com-
plexes, we carried out several computational studies on the
Cu and Ag catalysts complexed with the azomethine ylide
derived form imino ester 5a. The fully optimized structures
of these complexes, denoted as 29 and 30, are reported in
Figure 6. According to our results, imino ester 5a can form
two diastereomeric silver and copper azomethine ylides in
which the metal atoms are coordinated with at least the O
and N atoms of the azomethine ylide moiety and the P
atom of the catalyst. In the case of Cu complex 29a, the
P–Cu and S–Cu bond distance structures are significantly
shorter than those computed for 29b (Fig. 6). The P–Cu–N
and S–Cu–N bond angles are 154 and 113ꢂ, respectively.
Assuming that 28a is the chiral metal-complexed ylide pre-
cursor involved in the cycloaddition process, the approach of
the dipolarophile to its least hindered face (re face of the di-
pole), that avoiding the interaction with the sterically
Figure 6. Fully optimized (B3LYP/6-31G* and LANL2DZ level of theory) of complexes 29a and 29b, and 30a and 30b. Bond distances are given in angstrom.
Hydrogen atoms have been omitted for clarity. Numbers in parentheses are the relative energies, calculated at the B3LYP/6-31G* and LANL2DZ-DZPVE
levels.

6596
S. Cabrera et al. / Tetrahedron 63 (2007) 6587–6602
In the case of complexes 30a and 30b, our calculations sug-
gest that there areno S–Ag covalent bonds, the corresponding
Wiberg bond indexes being of only 0.02. Instead, the Natural
Bonding Orbital (NBO) analysis indicates that in both cases
the interaction between these atoms consists of a two-
electron donation between the two lone pairs of the sp3-
hybridized sulfur atom and the empty 5s AO of AgI. The
second-order perturbational energy associated with these
stabilizing donations is of ca. ꢁ12.0 kcal/mol. For both
complexes, the coordination pattern around the silver atom is
T-shaped and the oxygen and nitrogen atoms of the azome-
thine ylide are in a cis-relationship, the phosphine moiety oc-
cupying the distal position with respect to the nitrogen atom.
these latter complexes there is a strong steric interaction be-
tween the phenyl group of the imine moiety and one of the
phenyl groups of the PPh2 subunit (Fig. 6). In addition,
both 29a and 30a benefit from a stabilizing electrostatic inter-
action between the oxygen atom of the azomethine carbonyl
group and the iron atom. For instance, in the case of 30a the
˚
O–Fe interatomic distance is 4.772 A, the corresponding nat-
ural charges of the O and Fe atoms being ꢁ0.79 and +0.24
a.u., respectively. The computed energy differences indicate
that 29a and 30a are virtually the exclusive complexes pres-
ent in the reaction mixtures previous to the cycloaddition
steps. These results are in line with those obtained in the
previously described NOE experiments (Fig. 5).
Complexes 29a and 30a are calculated to be 3.6 and 2.5 kcal/
mol more stable than 29b and 30b, respectively, because in
Ourcalculationsalsoindicatethatinbothcomplexesoneofthe
faces of the complexed azomethine ylide is closed to further
coordination with the dipolarophile because of the high steric
congestion imposed by the tert-butylsulfenyl subunit linked to
the ferrocenyl group. This means that in the major complexes
29a and 30a the cycloaddition will take place through the
(2Si,4Re) face and therefore the preferred products will be
the cycloadducts indicated in Scheme 3. On the other hand,
theshorterdistance betweenthebulkytert-butylsulfenyl group
H
H
O
NH
H
H
H
S
N
O
M
+
P
Fe
H2
O
9c
31a [M=Cu(I)]
32a [M=Ag(I)]
˚
andthecopperatomin29a(distanceS–Cu¼2.71 A)compared
˚
to that in the silver complex 30a (distance S–Ag¼3.13 A)
wouldnicelyexplainthesignificantlyhigherenantioselectivity
always observed in our copper catalyzed cycloadditions.
O
H
O
H
NH
NH
The next step in our study was to understand the origins of the
high endo-selectivity observed in the [3+2] cycloaddtion be-
tween imino esters 5 and symmetric cis-dipolarophiles such
as maleimides 9a and 9b. Given the size of the interacting sys-
tems, simplified models such as those gathered in Scheme 4
were studied. The four possible transition structures for the
CuI and AgI complexes 31a and 32a to yield the endo- and
exo-cycloadducts33and34(Scheme4)arereportedinFigure7.
H
O
H
S
O
N
O
N
H
H
S
M
M
P
P
O
+
Fe
Fe
H
H
H2
H2
endo-33 [M=Cu(I)]
endo-34 [M=Ag(I)]
exo-33 [M=Cu(I)]
exo-34 [M=Ag(I)]
Scheme 4.
Figure 7. Fully optimized (B3LYP/6-31G* and LANL2DZ level of theory) of endo and exo transition structures associated with [3+2] cycloadditions depicted
in Scheme 3. Bond distances are given in angstrom. Hydrogen atoms have been omitted for clarity. Numbers in parentheses are the relative energies, calculated
at the B3LYP/6-31G* and LANL2DZ-DZPVE level. The asterisks indicate the position of the susbstituents in azomethine ylides derived form imino esters 5.

S. Cabrera et al. / Tetrahedron 63 (2007) 6587–6602
6597
According to our calculations, endo transition structures
TS1 and TS2 lie 4.6 kcal/mol and 4.1 kcal/mol lower in
energy than their exo analogs, respectively (Fig. 7). These
results are in agreement with the nearly complete endo-
selectivity found for this kind of dipolarophiles (vide supra,
Table 3). The reasons for this result cannot be completely at-
tributed to orbital interactions between the dipolarophile and
the metallic center. Thus, in endo-TS1 the combined orbital
interactions between the maleimide moiety and the Cu atom
result in total second-order perturbation energy of ca.
ꢁ1.5 kcal/mol. In contrast, the N–H moiety of imide 9c in
endo-TS1 exhibits a Coulombic interaction with the Cu
atom with an electrostatic energy of ꢁ21.5 kcal/mol. The
carbonyl groups of the imide moiety also interact with the
metallic center, the corresponding electrostatic potential en-
ergy being +17.0 kcal/mol. Therefore, the total electrostatic
energy between the endo imide moiety and the Cu atom is
ꢁ4.5 kcal/mol, a value in line with the computed relative
energies for endo- and exo-TS1. The analysis of endo- and
exo-TS2 resulted in similar results, thus validating our con-
clusion that the endo-selectivity in this kind of [3+2] cyclo-
additions is mainly electrostatic in nature. Since Coulombic
interactions are very sensitive to substituent and solvent ef-
fects, the analysis of the stereochemical outcome with other
dipolarophiles will require specific studies in each case.
for 1H and 20.4 for 13C). Mass spectra (MS) were determined
at an ionizing voltage of 70 eV. HPLC experiments were
conducted using Daicel Chiralcel OD and Chiralpak AS-H
columns. Flash column chromatography was performed us-
ing silica gel Merk-60 (230–400 mesh). Ethyl N-(diphenyl-
methylene)glycinate (8e) was purchased from Aldrich.
4.1.1. General procedure for the synthesis of a-imino
esters (5a–h, 7a–d, and 8a, 8b, and 8d). MethodA:Toasus-
pension of the corresponding amino acid ester hydrochloride
(23.9 mmol) and MgSO4 (25.0 mmol) in CH2Cl2 (25 mL)
was added Et3N (3.4 mL, 23.9 mmol). The mixture was
stirred at room temperature for 1 h, and then the correspond-
ing aldehyde (20.0 mmol) was added. The reaction was
stirred at room temperature overnight, and the resulting pre-
cipitate was removed by filtration. The filtrate was washed
with water (15 mL), the aqueous phase was extracted with
CH2Cl2 (3ꢃ10 mL), and the combined organic phases were
washed with brine, dried over MgSO4, and concentrated.
The resulting imino esters were obtained pure and used in
1,3-dipolar cycloadditions without further purification.
Method B: A suspension of methyl glycinate hydrochloride
(935 mg, 7.50 mmol) in CH2Cl2 (8 mL) was washed with
NH4OH (30%, 5 mL). The organic phase was dried over
MgSO4, and filtered. To this solution, MgSO4 (9.94 mmol)
and the corresponding aldehyde (5 mmol) were added. The
reaction mixture was stirred for 12 h at ambient temperature.
Afterfiltration,theorganicphasewaswashedwithasaturated
solution of NaCl. The aqueous phase was extracted with
CH2Cl2 (3ꢃ10 mL) and the combined organic phases were
dried over MgSO4 and concentrated. Method C: A suspen-
sion of methyl glycinate hydrochloride (879 mg, 7.00 mmol)
in CH2Cl2 (8 mL) was washed with NH4OH (30%, 5 mL).
The organic phase was dried over MgSO4, filtered, and con-
centrated to a final volume of 1 mL. This solution was added
to a suspension of the corresponding ketone (6.6 mmol) and
3. Conclusions
Among different Lewis acids tested, we have found that the
combination Cu(CH3CN)4ClO4/Fesulphos (0.5–3 mol %) is
a very efficient catalyst system for the 1,3-dipolar cycloaddi-
tion of azomethine ylides. This catalyst system displays
a very high reactivity and provides good yields and stereo-
selectivities with an exceptionally wide range of dipolaro-
philes, including diactivated alkenes, such as maleimides,
dimethyl maleate, dimethyl fumarate, and fumaronitrile,
and monoactivated alkenes, such as a,b-unsaturated esters,
methacrolein, and b-nitrostyrene. With the exception of
fumaronitrile and b-nitrostyrene, the cycloaddition was
endo-selective and in all cases good to excellent enantio-
selectivities were achieved (69–>99% ee). A variety of azo-
methine ylide precursors derived from methyl glycinate and
methyl alaninate were also tested (17 cases), including the
first reported examples of catalytic asymmetric 1,3-dipolar
cycloadditions of azomethine ylides of ketimines. We have
also isolated a presumed Cu–Fesulphos–azomethine com-
plex intermediate and performed some theoretical calcula-
tions that shed some light on the origin of the high
efficiency of the CuI–Fesulphos catalyst system.
˚
MS 4 A (15 g) in benzene (50 mL). The reaction mixture was
stirred for 12 h, filtered, and the solvent evaporated.
4.1.1.1. Methyl (E)-N-benzylideneglycinate (5a). Yield
89%; H NMR (200 MHz): d 8.26 (s, 1H, N]CH), 7.77–
1
7.74 (m, 2H, Ar), 7.42–7.39 (m, 3H, Ar), 4.39 (s, 2H,
CH2), 3.75 (s, 3H, CO2Me).
4.1.1.2. Methyl (E)-N-(2-naphthylmethylidene)gly-
cinate (5b). Yield 84%; mp: 95–96 ꢂC; 1H NMR
(200 MHz): d 8.45 (s, 1H, N]CH), 8.10–8.01 (m, 2H,
Ar), 7.92–7.85 (m, 3H, Ar), 7.55–7.51 (m, 2H, Ar), 4.48
(s, 2H, CH2), 3.80 (s, 3H, CO2Me).
4.1.1.3. Methyl (E)-N-(p-fluorobenzylidene)glycinate
(5c). Yield 56%; 1H NMR (200 MHz): d 8.26 (s, 1H,
N]CH), 7.82–7.74 (m, 2H, Ar), 7.15–7.06 (m, 2H, Ar),
4.40 (s, 2H, CH2), 3.78 (s, 3H, CO2Me).
4. Experimental
4.1. General methods
All the reactions were carried out in anhydrous solvents and
under an argon atmosphere. Melting points were taken in
open-end capillary tubes. NMR spectra were recorded at
4.1.1.4. Methyl (E)-N-(p-methoxybenzylidene)gly-
cinate (5d). Yield 72%; mp: 67–68 ꢂC; 1H NMR
(200 MHz): d 8.21 (s, 1H, N]CH), 7.72 (d, J¼8.8 Hz,
2H, Ar), 6.93 (d, J¼8.8 Hz, 2H, Ar), 4.38 (s, 2H, CH2),
3.84 (s, 3H, OMe), 3.77 (s, 3H, CO2Me).
1
200 or 300 MHz for H, and at 50 or 75 MHz for 13C, at
room temperature in CDCl3 unless stated, using the chloro-
form residual peaks for calibration (7.26 ppm for H and
1
77.0 ppm for 13C). The NMR spectra recorded in toluene-
d8 were calibrated using the toluene residual peak (2.09
4.1.1.5. Methyl (E)-N-(o-methylbenzylidene)glycinate
(5e). Yield 80%; mp: 50–51 ꢂC; 1H NMR (200 MHz):

6598
S. Cabrera et al. / Tetrahedron 63 (2007) 6587–6602
d 8.60 (s, 1H, N]CH), 7.93 (dd, J¼7.5, 1.9 Hz, 1H, Ar),
7.37–7.17 (m, 3H, Ar), 4.44 (s, 2H, CH2), 3.78 (s, 3H,
CO2Me), 2.52 (s, 3H, Me).
4.1.2. General procedure for the asymmetric 1,3-dipolar
cycloaddition of azomethine ylides. To a solution of Fesul-
phos ligand 1a (2.7 mg, 0.006 mmol) and Cu(CH3CN)4-
ClO4 (2.0 mg, 0.006 mmol) in CH2Cl2 (0.5 mL) or THF
(0.5 mL), at the optimal temperature (typically room tem-
perature), was successively added a solution of the imine
(0.20 mmol) in CH2Cl2 (1.0 mL) or THF (0.5 mL), Et3N
(5.0 mL, 0.035 mmol), and the corresponding dipolarophile
(0.22 mmol in the case of maleimides 9a and 9b, 30 mmol
for the remaining dipolarophiles). Once the starting material
was consumed as monitored by TLC, the mixture was fil-
tered through Celite, and the filtrate was concentrated to dry-
ness. The residue was analyzed by 1H NMR to determine the
endo/exo ratio and purified by flash chromatography (the
eluent is indicated in each case). The ee values were deter-
mined by HPLC analysis. Racemic endo/exo mixtures
were obtained as above by using Cu(CH3CN)4ClO4/PPh3
(5 mol %) and Cu(CH3CN)4ClO4/rac-1a (5 mol %) or
AgOAc/(ꢄ)-Binap (5 mol %) as catalyst system.
4.1.1.6. Methyl (E)-N-(2-pyridylmethylidene)glyci-
nate (5f). Yield 72%; H NMR (200 MHz): d 8.66–8.64
1
(m, 1H, Py), 8.39 (s, 1H, N]CH), 8.08 (d, J¼8.0 Hz, 1H,
Py), 7.79–7.72 (m, 1H, Py), 7.37–7.31 (m, 1H, Py), 4.47
(s, 2H, CH2), 3.78 (s, 3H, CO2Me).
4.1.1.7. Methyl (E)-N-(cyclohexylmethylidene)glyci-
nate (5g).26 Yield 37%; 1H NMR (200 MHz): d 7.49
(d, J¼5.1 Hz, 1H, N]CH), 4.15 (s, 2H, CH2), 3.74 (s,
3H, CO2Me), 1.86–1.67 (m, 5H, Cy), 1.56–1.21 (m, 6H,
Cy).
4.1.1.8. Methyl (E)-N-(2,2-dimethylpropylidene)glyci-
nate (5h).26 Yield 24%; 1H NMR (200 MHz): d 7.55 (s, 1H,
N]CH), 4.16 (s, 2H, CH2), 3.73 (s, 3H, CO2Me), 1.01 (s,
9H, t-Bu).
4.1.2.1. (1S,3R,3aS,6aR)-Methyl-4,6-dioxo-3,5-diphe-
nyl-octahydropyrrole [3,4-c]pyrrole-1-carboxylate (endo-
10a).11 SiO2 chromatography: n-hexane–EtOAc 1:1;
mp¼180–181 ꢂC (lit.11 mp: 160–162 ꢂC). [a]2D0 +114 (c
1.22, CH2Cl2), >99% ee; lit.11 [a]D20 +91.1 (c 1.22,
CH2Cl2) for a 71% ee sample. HPLC: Daicel Chiralpak
AS-H, i-PrOH/hexane 50:50, flow rate 0.8 mL/min, tR:
13.6 min (1S,3R,3aS,6aR)-10a and 29.2 min (1R,3S,3aR,
4.1.1.9. Methyl (E)-N-benzylidenealaninate (7a). Yield
87%; H NMR (200 MHz): d 8.31 (s, 1H, N]CH), 7.80–
7.75 (m, 2H, Ar), 7.44–7.40 (m, 3H, Ar), 4.16 (q,
J¼6.8 Hz, 1H, CH), 3.74 (s, 3H, CO2Me), 1.53 (d,
J¼6.8 Hz, 3H, Me).
1
4.1.1.10. Methyl (E)-N-(2-naphthylmethylidene)alani-
nate (7b). Yield 72%; mp: 73–74 ꢂC; 1H NMR (200 MHz):
d 8.48 (s, 1H, N]CH), 8.10–8.01 (m, 2H, Ar), 7.93–7.84
(m, 3H, Ar), 7.58–7.50 (m, 2H, Ar), 4.23 (q, J¼6.7 Hz,
1H, CH), 3.77 (s, 3H, CO2Me), 1.58 (d, J¼6.7 Hz, 3H,
Me).
1
6aS)-10a, 220 nm. H NMR (200 MHz): d 7.50–7.32 (m,
8H), 7.17–7.12 (m, 2H), 4.63 (dd, J¼8.6, 5.1 Hz, 1H),
4.16 (dd, J¼6.5, 5.1 Hz, 1H), 3.88 (s, 3H), 3.78–3.71 (m,
1H), 3.58 (t, J¼8.2 Hz, 1H), 2.55–2.50 (m, 1H).
exo-10a: Mp: 139–140 ꢂC (lit.11 mp: 138–140 ꢂC); 1H NMR
(200 MHz): d 7.55–7.31 (m, 10H), 4.65 (d, J¼5.1 Hz, 1H),
4.22 (d, J¼4.3 Hz, 1H), 4.07 (dd, J¼8.9, 4.3 Hz, 1H), 3.65
(s, 3H), 3.67–3.59 (m, 1H), 2.73 (br s, 1H).
4.1.1.11. Methyl (E)-N-(p-methoxybenzylidene)alani-
nate (7c). Yield 70%; H NMR (300 MHz): d 8.24 (s, 1H,
1
N]CH), 7.72 (d, J¼8.7 Hz, 2H, Ar), 6.92 (d, J¼8.8 Hz,
2H, Ar), 4.12 (q, J¼6.8 Hz, 1H, CH), 3.84 (s, 3H, OMe),
3.74 (s, 3H, OMe), 1.51 (d, J¼6.8 Hz, 3H, Me).
4.1.2.2.
(1S,3R,3aS,6aR)-Methyl-4,6-dioxo-3-naph-
thyl-5-phenyl-octahydropyrrole [3,4-c]pyrrole-1-carb-
oxylate (endo-10b). SiO2 chromatography:n-hexane/EtOAc
1:1; mp: 226–227 ꢂC. [a]D20 +165 (c 0.11, CH2Cl2), >99%
ee. HPLC: Daicel Chiralpak AS-H, i-PrOH/hexane 50:50,
flow rate 0.8 mL/min, tR: 16.9 min (1S,3R,3aS,6aR)-10b
and 60.1 min (1R,3S,3aR,6aS)-10b, 220 nm; 1H NMR
(300 MHz): d 7.98 (s, 1H), 7.88–7.83 (m, 3H), 7.57–7.46
(m, 3H), 7.39–7.28 (m, 3H), 7.15–7.11 (m, 2H), 4.78 (dd,
J¼8.6, 5.0 Hz, 1H), 4.21 (dd, J¼6.4, 5.0 Hz, 1H), 3.92 (s,
3H), 3.80 (t, J¼6.4 Hz, 1H), 3.68 (t, J¼8.6 Hz, 1H), 2.65–
2.62 (m, 1H); 13C NMR (75 MHz): d 175.1, 173.6, 170.1,
134.3, 133.4, 133.3, 131.6, 129.0, 128.5, 128.2, 128.0,
127.9, 126.3, 126.1, 126.0, 125.6, 125.3, 64.3, 61.9, 52.4,
49.4, 48.4; EIMS m/z (%): 400 (M+, 18), 227 (78), 196
(33), 167 (100).
4.1.1.12. Methyl (E)-N-(o-methylbenzylidene)alani-
nate (7d). Yield 73%; H NMR (200 MHz): d 8.61 (s, 1H,
N]CH), 7.93–7.90 (m, 1H, Ar), 7.35–7.15 (m, 3H, Ar),
4.06 (q, J¼6.8 Hz, 1H, CH), 3.74 (s, 3H, CO2Me), 2.50 (s,
3H, Me–C6H4), 1.53 (d, J¼6.8 Hz, 3H, Me).
1
4.1.1.13. Methyl N-(1-phenylethylidene)glycinate (8a).
Yield 16%; 30% conversion); mp: 57–58 ꢂC; 1H NMR
(200 MHz, C6D6): d 7.97–7.92 (m, 2H, Ar), 7.21–7.17 (m,
3H, Ar), 4.13 (s, 2H, CH2), 3.44 (s, 3H, CO2Me), 1.66 (s,
3H, CH3).
4.1.1.14. Methyl N-(1-p-chlorophenylethylidene)glyci-
nate (8b). It was obtained as a 2.1:1 mixture of 8b/2b, and
used without further purification; 1H NMR (200 MHz,
C6D6): d 7.65 (d, J¼8.6 Hz, 2H, Ar), 7.14 (d, J¼8.6 Hz,
2H, Ar), 4.07 (s, 2H, CH2), 3.43 (s, 3H, CO2Me), 1.53 (s,
3H, Me).
exo-10b: Mp: 169–170 ꢂC; 1H NMR (200 MHz): d 7.89–7.82
(m, 4H), 7.60–7.36 (m, 8H), 4.83 (d, J¼5.1 Hz, 1H), 4.28 (d,
J¼4.1 Hz, 1H), 4.13 (dd, J¼8.8, 4.1 Hz, 1H), 3.73 (dd,
J¼8.8, 5.1 Hz, 1H), 3.60 (s, 3H, CO2Me), 2.84 (br s, 1H);
13C NMR (75 MHz): d 176.0, 175.9, 171.9, 137.6, 133.1,
133.0, 131.7, 129.2, 128.8, 128.0, 127.7, 126.5, 126.4,
126.3, 125.2, 124.7, 65.9, 62.7, 52.8, 51.9, 49.1; EIMS m/z
(%): 400 (M+, 38), 341 (88), 227 (76), 194 (87), 167 (100).
4.1.1.15. Methyl N-(1-ethyl-2-methylpropylidene)gly-
cinate (8d). Yield 38%; 1H NMR (200 MHz, C6D6):
d 3.96 (s, 2H), 3.39 (s, 3H, CO2Me), 2.44–2.30 (m, 1H),
1.33 (s, 3H), 1.07 (d, J¼6.6 Hz, 6H).

S. Cabrera et al. / Tetrahedron 63 (2007) 6587–6602
6599
1
4.1.2.3. (1S,3R,3aS,6aR)-Methyl 4,6-dioxo-3-(p-fluo-
rophenyl)-5-phenyl-octahydropyrrole [3,4-c]pyrrole-1-
carboxylate (endo-10c). SiO2 chromatography: n-hexane/
EtOAc 1:1; mp: 179–180 ꢂC. [a]D20 +118 (c 0.11, CH2Cl2),
>99% ee. HPLC: Daicel Chiralpak AS-H, i-PrOH/hexane
50:50, flow rate 0.8 mL/min, tR: 13.6 min (1S,3R,3aS,6aR)-
exo-10e: Mp: 138–139 ꢂC; H NMR (300 MHz): d 7.45–
7.04 (m, 9H), 4.86 (br s, 1H), 4.10–4.07 (m, 2H), 3.64–
3.61 (m, 1H), 3.47 (s, 3H), 2.60–2.50 (m, 1H), 2.38 (s,
3H); 13C NMR (75 MHz): d 176.4, 176.2, 172.0, 138.2,
136.2, 131.7, 131.1, 129.1, 128.7, 128.0, 126.3, 126.1,
125.4, 62.6, 62.1, 52.7, 50.9, 48.7, 19.6; EIMS m/z (%):
364 (M+, 8), 305 (100), 191 (52), 158 (71), 131 (47).
1
10c and 28.5 min (1R,3S,3aR,6aS)-10c, 220 nm; H NMR
(300 MHz): d 7.38–7.23 (m, 5H), 7.09–7.05 (m, 2H), 7.00–
6.95 (m, 2H), 4.53 (d, J¼8.7 Hz, 1H), 4.07 (d, J¼6.6 Hz,
1H), 3.80 (s, 3H), 3.65 (dd, J¼7.8, 6.6 Hz, 1H), 3.47 (dd,
J¼8.7, 7.9 Hz, 1H), 2.42 (br s, 1H); 13C NMR (75 MHz):
d 174.9, 173.5, 169.9, 162.6 (d, JF–C¼245.5 Hz), 132.4,
131.6, 129.1, 128.8 (d, JF–C¼8.2 Hz), 128.6, 128.0, 115.5
(d, JF–C¼21.3 Hz), 63.5, 61.8, 52.3, 49.2, 48.1; EIMS m/z
(%): 368 (M+, 7), 309 (53), 195 (100), 162 (68), 135 (96),
108 (22); Anal. Calcd for C20H17FN2O4: C, 62.21; H, 4.65;
N, 7.60. Found: C, 64.82; H, 4.86; N, 7.23. HRMS Calcd
for: C20H18FN2O4 (M+H)+: 369.1251; found: 369.1263.
4.1.2.6. (1S,3S,3aS,6aR)-Methyl 3-cyclohexyl-(4,6-
dioxo)-5-phenyl-octahydropyrrole [3,4-c]pyrrole-1-carb-
oxylate (endo-10g).27 SiO2 chromatography: n-hexane/
EtOAc 1:1. HPLC: Daicel Chiralpak AS-H, hexane/i-PrOH
60:40, flow 0.7 mL/min, tR: 11.0 min (1S,3S,3aS,6aR)-10g
and 12.9 min (1R,3R,3aR,6aS)-10g, 210 nm; 1H NMR
(300 MHz): d 7.41–7.30 (m, 3H, Ar), 7.19–7.17 (m, 2H,
Ar), 3.93 (d, J¼8.1 Hz, 1H, H-1), 3.75 (s, 3H, CO2Me),
3.55 (t, J¼8.0 Hz, 1H, H-6a), 3.41 (t, J¼7.4 Hz, 1H, H-3a),
2.90 (dd, J¼10.1, 7.2 Hz, 1H, H-3), 2.31 (d, J¼11.2 Hz,
1H, NH), 1.87 (d, J¼12.3 Hz, 1H, Cy), 1.70–1.59 (m, 3H,
Cy), 1.31–0.76 (m, 7H, Cy); 13C NMR (75 MHz): d 175.1,
174.5, 170.6, 131.7, 129.1, 128.7, 126.5, 68.4, 63.1, 52.5,
49.1, 47.7, 38.4, 31.3, 31.2, 26.3, 25.7, 25.5.
1
exo-10c: Mp: 63–64 ꢂC; H NMR (300 MHz): d 7.44–7.18
(m, 7H), 7.02–6.95 (m, 2H), 4.51 (d, J¼5.4 Hz, 1H), 4.14
(d, J¼4.1 Hz, 1H), 3.95 (dd, J¼9.0, 4.5 Hz, 1H), 3.63 (s,
3H), 3.45 (dd, J¼9.0, 5.7 Hz, 1H), 2.40 (br s, 1H); 13C
NMR (75 MHz): d 175.7, 171.7, 164.1, 160.8, 136.1,
131.6, 129.2, 128.9, 128.4, 126.3, 115.6 (d, JF–C¼21.0 Hz),
65.1, 62.4, 52.8, 52.2, 49.1; EIMS m/z (%): 368 (M+, 8), 309
(100), 195 (30), 162 (84), 135 (41).
4.1.2.7. (1S,3S,3aS,6aR)-Methyl 3-tert-butyl-(4,6-
dioxo)-5-phenyl-octahydropyrrole [3,4-c]pyrrole-1-carb-
oxylate (endo-10h).27 SiO2 chromatography: n-hexane/
EtOAc 1:1. HPLC: Daicel Chiralpak AS-H, hexane/
i-PrOH 80:20, flow 0.8 mL/min, tR: 14.5 min (1S,3S,3aS,
1
4.1.2.4.
(1S,3R,3aS,6aR)-Methyl
4,6-dioxo-3-(p-
6aR)-10h and 16.1 min (1R,3R,3aR,6aS)-10h, 210 nm; H
methoxyphenyl)-5-phenyl-octahydropyrrole [3,4-c]pyr-
role-1-carboxylate (endo-10d).15a SiO2 chromatography:
n-hexane/EtOAc 1:1; mp: >200 ꢂC (decomp.) (lit.15a (ꢄ)-
endo-10d mp: 194–196 ꢂC). [a]D20 +106 (c 0.16, CH2Cl2),
>99% ee. HPLC: Daicel Chiralpak AS-H, i-PrOH/hexane
50:50, flow rate 0.8 mL/min, tR: 18.3 min (1S,3R,3a-
S,6aR)-10d and 36.9 min (1R,3S,3aR,6aS)-10d, 220 nm;
1H NMR (200 MHz): d 7.45–7.32 (m, 5H), 7.19–7.14 (m,
2H), 6.91–6.87 (m, 2H), 4.58 (dd, J¼8.6, 5.8 Hz, 1H), 4.13
(dd, J¼6.4, 4.8 Hz, 1H), 3.87 (s, 3H), 3.79 (s, 3H), 3.73 (t,
J¼6.6 Hz, 1H), 3.53 (t, J¼8.4 Hz, 1H), 2.49 (br s, 1H).
exo-10d: Mp: 117–118 ꢂC (lit.15a mp: 133–134 ꢂC); 1H
NMR (200 MHz): d 7.54–7.31 (m, 7H), 6.93–6.87 (m,
2H), 4.57 (d, J¼5.4 Hz, 1H), 4.19 (d, J¼4.6 Hz, 1H), 4.04
(dd, J¼8.9, 4.6 Hz, 1H), 3.80 (s, 3H), 3.67 (s, 3H), 3.55
(dd, J¼9.0, 5.4 Hz, 1H), 2.67 (br s, 1H).
NMR (300 MHz): d 7.41–7.28 (m, 3H), 7.17–7.14 (m,
2H), 3.88 (d, J¼7.4 Hz, 1H), 3.76 (s, 3H), 3.61 (t,
J¼7.7 Hz, 1H), 3.32 (t, J¼7.9 Hz, 1H), 3.02 (d, J¼7.9 Hz,
1H), 2.12 (br s, 1H), 1.09 (s, 9H); 13C NMR (75 MHz):
d 176.0, 174.8, 170.5, 131.8, 129.2, 128.8, 126.5, 73.2,
62.3, 52.4, 50.1, 47.4, 33.0, 27.2; EIMS m/z (%): 330 (M+,
1), 273 (100), 94 (71).
4.1.2.8. (1S,3R,3aS,6aR)-Methyl 5-methyl-4,6-dioxo-
3-phenyl-octahydropyrrole [3,4-c]pyrrole-1-carboxylate
(endo-11).11,28 SiO2 chromatography: n-hexane/EtOAc
1:1; mp: 166–167 ꢂC (lit.11 mp: 176–177 ꢂC). [a]2D0 +51 (c
0.14, CH2Cl2), >99% ee; lit.11 [a]D20 +61.0 (c 1.18,
CH2Cl2) for a 79% ee sample. HPLC: Daicel Chiralcel
OD, i-PrOH/hexane 50:50, flow rate 0.6 mL/min, tR:
31.3 min (1S,3R,3aS,6aR)-11 and 36.8 min (1R,3S,3aR,
1
6aS)-11, 220 nm; H NMR (200 MHz): d 7.37–7.33 (m,
5H), 4.51 (dd, J¼8.3, 5.4 Hz, 1H), 4.07 (dd, J¼6.5,
4.4 Hz, 1H), 3.89 (s, 3H), 3.64–3.54 (m, 1H), 3.48–3.39
(m, 1H), 2.89 (s, 3H), 2.44 (br s, 1H).
4.1.2.5. (1S,3R,3aS,6aR)-Methyl 4,6-dioxo-3-(o-meth-
ylphenyl)-5-phenyl-octahydropyrrole [3,4-c]pyrrole-1-
carboxylate (endo-10e). SiO2 chromatography: n-hexane/
EtOAc 1:1; mp: 154–155 ꢂC. [a]D20 +136 (c 0.11, CH2Cl2),
>99% ee. HPLC: Daicel Chiralpak AS-H, i-PrOH/hexane
50:50, flow rate 0.8 mL/min, tR: 10.9 min (1S,3R,3aS,
exo-11: Mp: 151–152 ꢂC (lit.11 mp: 137–139 ꢂC); 1H NMR
(300 MHz): d 7.43–7.28 (m, 5H), 4.52 (d, J¼5.0 Hz, 1H),
4.09 (d, J¼4.2 Hz, 1H), 3.92 (dd, J¼8.8, 4.4 Hz, 1H), 3.66
(s, 3H), 3.49 (dd, J¼8.8, 5.0 Hz, 1H), 3.06 (s, 3H), 2.67
(br s, 1H).
1
6aR)-10e and 50.5 min (1R,3S,3aR,6aS)-10e, 220 nm. H
NMR (300 MHz): d 7.69–7.63 (m, 1H), 7.38–7.19 (m,
6H), 7.09–7.04 (m, 2H), 4.75–4.73 (m, 1H), 4.16–4.14 (m,
1H), 3.88 (s, 3H), 3.74 (t, J¼6.2 Hz, 1H), 3.65 (t,
J¼8.6 Hz, 1H), 2.45 (s, 3H); 13C NMR (75 MHz): d 175.1,
173.3, 170.0, 135.8, 135.3, 131.6, 130.0, 128.9, 128.4,
128.0, 126.1 (2C), 125.3, 61.6, 60.6, 52.3, 48.3, 47.0, 19.4;
EIMS m/z (%): 368 (M+, 7), 364 (7), 305 (41), 191 (100),
158 (50), 131 (86); Anal. Calcd for C21H20N2O4: C, 69.22;
H, 5.53; N, 7.69. Found: C, 68.82; H, 5.57; N, 7.34.
4.1.2.9. (1S,3R,3aS,6aR)-Methyl 1-methyl-4,6-dioxo-
3,5-diphenyl-octahydropyrrole
[3,4-c]pyrrole-1-carb-
oxylate (endo-12a).23,27 SiO2 chromatography: n-hexane/
EtOAc 1:1; mp: 199–200 ꢂC. [a]D20 +73 (c 0.12, CH2Cl2),
80% ee. HPLC: Daicel Chiralcel OD, i-PrOH/hexane
50:50, flow rate 0.6 mL/min, tR: 17.3 min (1R,3S,3aR,
6aS)-12a and 28.4 min (1S,3R,3aS,6aR)-12a, 210 nm.

6600
S. Cabrera et al. / Tetrahedron 63 (2007) 6587–6602
1H NMR (300 MHz): d 7.37–7.18 (m, 8H), 7.01–6.98 (m,
2H), 4.81 (d, J¼9.1 Hz, 1H), 3.80 (s, 3H), 3.61 (dd, J¼9.1,
7.7 Hz, 1H), 3.37 (d, J¼7.6 Hz, 1H), 2.59 (br s, 1H), 1.58
(s, 3H); 13C NMR (75 MHz): d 174.8, 173.5, 172.7, 136.8,
131.5, 129.0, 128.5, 128.4 (2C), 127.1, 126.0, 67.6, 62.4,
55.7, 52.7, 50.2, 23.9; FABMS m/z (%): 365 (M+H, 100),
305 (43), 154 (64), 149 (62), 136 (67), 69 (60).
J¼7.7 Hz, 1H), 3.69 (d, J¼7.7 Hz, 1H), 3.12–3.07 (m, 2H),
1.18 (t, J¼7.2 Hz, 3H); 13C NMR (75 MHz): d 174.8,
173.9, 170.2, 144.5, 141.4, 131.7, 129.0, 128.9, 128.5, 128.0,
127.7, 127.6, 127.2, 126.4, 126.3, 73.8, 61.7, 60.3, 53.0, 49.1,
14.1; FABMS m/z (%): 441 (M+1, 100), 367 (14), 154 (18).
4.1.2.14. (2S,3R,4S,5R)-Trimethyl 5-phenylpyrrol-
idine-2,3,4-tricarboxylate (endo-15).11 SiO2 chromato-
graphy: n-hexane/EtOAc 4:1; mp: 94–95 ꢂC (lit.11 mp:
91–93 ꢂC). [a]2D0 +100 (c 0.05, CH2Cl2), 94% ee; lit.11
[a]2D0 +64.7 (c 1.07, CH2Cl2) for a 87% ee sample. HPLC:
Daicel Chiralpak AS-H, i-PrOH/hexane 50:50, flow rate
0.8 mL/min, tR: 7.1 min (2S,3R,4S,5R)-15 and 12.6 min
(2R,3S,4R,5S)-15, 220 nm; 1H NMR (300 MHz):
d 7.28–7.18 (m, 5H), 4.41 (d, J¼6.8 Hz, 1H), 4.09 (d,
J¼8.9 Hz, 1H), 3.74 (s, 3H), 3.66 (d, J¼8.2 Hz, 1H), 3.62
(s, 3H), 3.53–3.48 (m, 1H), 3.29 (br s, 1H), 3.16 (s, 3H).
4.1.2.10. (1S,3R,3aS,6aR)-Methyl 1-methyl-3-naphthyl-
4,6-dioxo-5-phenyl-octahydropyrrole [3,4-c]pyrrole-1-
carboxylate (endo-12b).27 SiO2 chromatography: n-hexane/
EtOAc 1:1; mp: >210 ꢂC (decomp.). [a]2D0 +105 (c 0.12,
CH2Cl2), 92% ee. HPLC: Daicel Chiralpak AS-H, i-PrOH/
hexane 50:50, flow rate 0.7 mL/min, tR: 13.2 min (1S,3R,3aS,
1
6aR)-12b and 15.9 min (1R,3S,3aR,6aS)-12b, 220 nm; H
NMR (300 MHz): d 7.84 (s, 1H), 7.77–7.71 (m, 3H), 7.44–
7.37 (m, 3H), 7.23–7.18 (m, 3H), 6.96–6.93 (m, 2H), 4.98–
4.93 (m, 1H), 3.83 (s, 3H), 3.68 (dd, J¼9.1, 7.6 Hz,
1H), 3.41 (d, J¼7.6 Hz, 1H), 2.65 (d, J¼6.5 Hz, 1H), 1.61
(s, 3H); 13C NMR (75 MHz): d 174.9, 173.6, 172.8, 134.5,
133.4, 133.3, 131.5, 129.0, 128.4, 128.2, 128.0, 127.8,
126.3, 126.1, 126.0, 125.5, 67.6, 62.4, 55.7, 52.7, 50.1, 23.9;
EIMS m/z (%): 414 (M+, 17), 355 (84), 241 (100), 208 (40),
181 (96).
1
exo-15: H NMR (300 MHz): d 7.39–7.36 (m, 2H), 7.28–
7.19 (m, 3H), 4.59 (d, J¼7.7 Hz, 1H), 4.30 (d, J¼6.1 Hz,
1H), 3.72 (s, 3H), 3.64 (s, 3H), 3.58 (s, 3H), 3.56–3.53 (m,
1H), 3.17 (dd, J¼16.8, 7.7 Hz, 1H), 2.55 (br s, 1H).
4.1.2.15. (2S,3S,4S,5R)-Trimethyl 5-phenylpyrrol-
idine-2,3,4-tricarboxylate (endo-18).10a SiO2 chromato-
graphy: n-hexane/EtOAc 4:1. [a]2D0 +43 (c 0.10, CH2Cl2),
>99% ee; lit.10a [a]D20 +20 (c 0.35, CH2Cl2) for a 76% ee
sample. HPLC: Daicel Chiralcel OD, i-PrOH/hexane 30:70,
flow rate 0.8 mL/min, tR: 16.1 min (2S,3S,4S,5R)-18 and
4.1.2.11. (1S,3R,3aS,6aR)-Methyl 3-methyl-4,6-dioxo-
3,5-diphenyl-octahydropyrrole [3,4-c]pyrrole-1-carboxyl-
ate (endo-13a).27 SiO2 chromatography: n-hexane/EtOAc
1:1; mp: 185–186 ꢂC. [a]D20 +87 (c 0.27, CH2Cl2), 94% ee.
HPLC: Daicel Chiralpak AS-H, i-PrOH/hexane 50:50,
flow rate 0.7 mL/min, tR: 11.7 min (1S,3R,3aS,6aR)-13a
and 24.6 min (1R,3S,3aR,6aS)-13a, 210 nm; 1H NMR
(300 MHz): d 7.70–7.66 (m, 2H), 7.43–7.28 (m, 6H), 6.98–
6.95 (m, 2H), 4.45 (t, J¼6.7 Hz, 1H), 3.90 (s, 3H), 3.82–
3.77 (m, 1H), 3.40 (d, J¼7.7 Hz, 1H), 2.41 (d, J¼6.5 Hz,
1H), 1.74 (s, 3H); 13C NMR (75 MHz): d 174.8, 173.7,
170.7, 141.8, 131.4, 129.1, 128.9, 128.4, 128.2, 127.7,
126.2, 126.0, 67.3, 60.0, 56.1, 52.4, 49.4, 29.1; EIMS m/z
(%): 364 (M+, 10), 349 (100), 305 (81), 191 (67), 158 (71).
1
26.6 min (2R,3R,4R,5S)-18, 210 nm; H NMR (200 MHz):
d 7.33–7.27 (m, 5H), 4.65 (d, J¼7.8 Hz, 1H), 4.20 (d,
J¼7.3 Hz, 1H), 3.84 (s, 3H), 3.78 (s, 3H), 3.73–3.52 (m,
2H), 3.20 (s, 3H), 2.76 (br s, 1H).
exo-18: 1H NMR (200 MHz): d 7.49–7.27 (m, 5H), 4.33 (d,
J¼8.9 Hz, 1H), 4.23 (d, J¼8.3 Hz, 1H), 3.81–3.71 (m, 1H),
3.76 (s, 3H), 3.71 (s, 3H), 3.65 (s, 3H), 3.42 (dd, J¼8.9,
7.8 Hz, 1H), 2.68 (br s, 1H).
4.1.2.16. (2S,3R,4R,5R)-Methyl 3,4-dicyano-5-phenyl-
pyrrolidine-2-carboxylate (exo-19).15a SiO2 chromato-
graphy: n-hexane/EtOAc 4:1; mp: 120–121 ꢂC (lit.15a mp:
128–130 ꢂC). [a]D20 +356 (c 1.0, CH2Cl2), 76% ee. HPLC:
Daicel Chiralcel OD, i-PrOH/hexane 40:60, flow rate
0.7 mL/min, tR: 21.9 min (2S,3R,4R,5R)-19 and 28.7 min
4.1.2.12. (1S,3R,3aS,6aR)-Methyl 3-(p-chlorophenyl)-
3-methyl-4,6-dioxo-5-phenyl-octahydropyrrole [3,4-c]-
pyrrole-1-carboxylate (endo-13b).27 SiO2 chromatography:
n-hexane/EtOAc 1:1; mp: >210 ꢂC (decomp.). [a]D20 +151 (c
0.16, CH2Cl2), >99% ee. HPLC: Daicel Chiralpak AS-H,
i-PrOH/hexane 50:50, flow rate 0.7 mL/min, tR: 12.6 min
(1S,3R,3aS,6aR)-13b and 37.8 min (1R,3S,3aR,6aS)-13b,
1
(2R,3S,4S,5S)-19, 210 nm; H NMR (300 MHz): d 7.44–
7.30 (m, 5H), 4.32 (t, J¼8.5 Hz, 1H), 4.22 (t, J¼7.3 Hz,
1H), 3.84 (s, 3H), 3.61 (t, J¼8.4 Hz, 1H), 3.15 (t,
J¼8.6 Hz, 1H), 2.73 (br s, 1H).
endo-19: 1H NMR (300 MHz): d 7.45–7.40 (m, 5H), 4.57 (t,
J¼6.1 Hz, 1H), 4.11 (t, J¼6.6 Hz, 1H), 3.81 (s, 3H), 3.65
(dd, J¼6.6, 4.3 Hz, 1H), 3.51 (dd, J¼6.4, 4.4 Hz, 1H),
2.73 (br s, 1H).
1
220 nm; H NMR (300 MHz): d 7.61–7.58 (m, 2H), 7.36–
7.29 (m, 5H), 6.97–6.94 (m, 2H), 4.41 (t, J¼6.7 Hz, 1H),
3.87 (s, 3H), 3.77 (t, J¼7.7 Hz, 1H), 3.37 (d, J¼7.7 Hz, 1H),
2.29 (d, J¼6.3 Hz, 1H), 1.68 (s, 3H); 13C NMR (75 MHz):
d 174.7, 173.6, 170.6, 140.5, 133.7, 131.4, 129.0, 128.6,
128.3, 127.8, 126.0, 66.8, 60.0, 55.8, 52.5, 49.0, 29.1.
4.1.2.13. (1S,3aS,6aR)-Ethyl 1-methyl-4,6-dioxo-3,3,5-
triphenyl-octahydropyrrole [3,4-c]pyrrole-1-carboxyl-
ate (endo-13e).27 SiO2 chromatography: n-hexane/EtOAc
1:1; mp: 185–186 ꢂC. [a]D20 +161 (c 0.14, CH2Cl2), 93% ee.
HPLC: Daicel Chiralpak AS-H, i-PrOH/hexane 30:70, flow
rate 0.8 mL/min, tR: 12.9 min (1S,3aS,6aR)-13e and
19.1 min (1R,3aR,6aS)-13e, 220 nm; H NMR (300 MHz):
d 7.57–7.54 (m, 2H), 7.46–7.43 (m, 2H), 7.30–7.01 (m,
11H), 4.40–4.29 (m, 1H), 4.23–4.12 (m, 1H), 3.85 (t,
4.1.2.17. (2S,4S,5R)-Dimethyl 5-phenylpyrrolidine-
2,4-dicarboxylate (endo-21).10a,11 SiO2 chromatography:
n-hexane/EtOAc 4:1. [a]2D0 +36 (c 0.08, CH2Cl2), 95% ee;
lit.10a [a]D20 +38 (c 0.10, CH2Cl2) for a 88% ee sample.
HPLC: Daicel Chiralpak AS-H, i-PrOH/hexane 10:90,
flow rate 1.0 mL/min, tR: 15.0 min (2S,4S,5R)-21 and
24.5 min (2R,4R,5S)-21, 210 nm; 1H NMR (200 MHz):
d 7.27–7.16 (m, 5H), 4.47 (d, J¼7.8 Hz, 1H), 3.92 (t,
1

S. Cabrera et al. / Tetrahedron 63 (2007) 6587–6602
6601
J¼8.1 Hz, 1H), 3.76 (s, 3H), 3.25 (q, J¼6.9 Hz, 1H), 3.15 (s,
3H), 2.72 (br s, 1H), 2.35 (dd, J¼8.1, 6.9 Hz, 2H).
J¼20.3 Hz, 1H, CH2), 4.93 (br s, 1H, Cp–H), 4.81 (br s, 1H,
Cp–H), 4.59 (br s, 1H, Cp–H), 3.99 (br s, 5H, Cp–H0), 3.95
(br s, 3H, OMe), 1.02 (s, 9H, t-Bu).
4.1.2.18. (1S,3R,3aS,6aR)-Methyl 3-phenyl-4-(oxo)octa-
hydrocyclopentane [c]pyrrol-1-carboxylate (endo-23).
SiO2 chromatography: n-hexane/EtOAc 5:1; mp: 150–
4.2. Computational studies
1
151 ꢂC; H NMR (300 MHz): d 7.36–7.30 (m, 5H), 4.55
All the calculations reported in this paper were performed
within Density Functional Theory,29 using the hybrid three-
parameter functional commonly denoted as B3LYP.30 The
standard 6-31G* and LANL2DZ (for Fe, Ag, and Cu atoms)
basis sets,31,32 as implemented in the Gaussian 0333 suite of
programs, were used in all cases. Donor–acceptor interac-
tions were also computed using the Natural Bond Orbital
(NBO)34 method. The energies associated with these two-
(d, J¼9.1 Hz, 1H), 4.47 (t, J¼9.3 Hz, 1H), 4.15 (dd,
J¼9.4, 6.8 Hz, 1H), 4.07 (d, J¼5.5 Hz, 1H), 3.81 (s,
3H), 3.53–3.43 (m, 1H), 3.26 (t, J¼9.1 Hz, 1H), 2.47 (br
s, 1H); 13C NMR (75 MHz): d 175.2, 170.5, 137.4,
128.4, 128.3, 127.3, 69.8, 64.5, 63.0, 52.3, 48.5, 41.1;
EIMS m/z (%): 261 (M+, 1), 202 (100), 144 (9). HPLC:
Daicel Chiralpak AS-H, hexane/i-PrOH 50:50, flow rate
0.7 mL/min, tR: 22.7 min (1S,3R,3aS,6aR)-23 and 25.1 min
(1R,3S,3aR,6aS)-23, 220 nm.
electron interactions were computed by means of the
according to the
ð2Þ
second-order perturbation energy DE
following equation:
ffꢅ
exo-23: 1H NMR (200 MHz): d 7.51–7.28 (m, 5H), 4.57 (d,
J¼4.2 Hz, 1H), 4.48 (d, J¼3.8 Hz, 2H), 3.78 (s, 4H), 3.36–
3.14 (m, 2H), 2.57 (br s, 1H).
2
ꢅ
^
Cf jFjfD
ð2Þ
DE ¼ ꢁnf
ð1Þ
ffꢅ
ꢅ
3f ꢁ 3f
where fꢅ and f are the non-Lewis and Lewis localized
orbitals, F is the Fock operator, nf is the occupation of the
4.1.2.19. (2S,4S,5S)-Methyl 4-formyl-4-methyl-5-phe-
nylpyrrolidine-2-carboxylate (endo-25). SiO2 chromato-
graphy: n-hexane/EtOAc 2:1. [a]2D0 ꢁ8 (c 0.36, CH2Cl2),
^
ꢅ
f localized orbital and 3f and 3f are the respective energies.
1
The stationary points were subjected to harmonic analysis.35
The relative energies were computed including the zero-
point vibrational energy corrections (not scaled).
69% ee; H NMR (300 MHz): d 9.16 (s, 1H), 7.40–7.28
(m, 5H), 4.13 (s, 1H), 4.05 (dd, J¼9.1, 6.5 Hz, 1H), 3.82
(s, 3H), 2.62 (dd, J¼13.4, 6.5 Hz, 1H), 2.01 (dd, J¼13.4,
9.1 Hz, 1H), 1.28 (s, 3H); 13C NMR (75 MHz): d 203.8,
174.0, 136.9, 128.6, 128.2, 127.2, 72.7, 57.9, 55.3, 52.3,
37.3, 19.1; EIMS m/z (%): 247 (M+, 8), 177 (65), 117 (100),
106 (32); HRMS Calcd for: C14H17NO3 (M+): 247.1208;
found: 247.1205. HPLC: Daicel Chiralpak AS-H, hexane/
i-PrOH 70:30, flow rate 0.5 mL/min, tR: 15.7 min
(2S,4S,5S)-25 and 29.5 min (2R,4R,5R)-25, 210 nm.
References and notes
1. For general reviews on 1,3-dipolar cycloadditions, see:
(a) Synthetic Applications of 1,3-Dipolar Cycloaddition
Chemistry Towards Heterocycles and Natural Products;
Padwa, A., Pearson, W. H., Eds.; Wiley: New York, NY,
2003; (b) Broggini, G.; Zecchi, G. Synthesis 1999, 905; (c)
Gothelf, K. V.; Jørgensen, K. A. Chem. Rev. 1998, 98, 863;
(d) 1,3-Dipolar Cycloaddition Chemistry; Padwa, A., Ed.;
Wiley: New York, NY, 1984.
4.1.2.20. (2S,3S,4R,5S)-Methyl 4-nitro-3,5-diphenyl-
pyrrolidine-2-carboxylate (exo-27).15c,24 SiO2 chro-
matography: n-hexane/EtOAc 4:1; mp: 105–106 ꢂC (lit.24
(ꢄ)-exo-27 mp: 113–115 ꢂC). [a]D20 +105 (c 0.11, CH2Cl2),
94% ee; lit.15c [a]D20 +118.2 (c 1.10, CHCl3) for a 97% ee sam-
ple. HPLC: Daicel Chiralpak AS-H, i-PrOH/hexane 20:80,
flow rate 0.4 mL/min, tR: 25.2 min (2S,3S,4R,5S)-27 and
2. Review: (a) Gothelf, K. V.; Jørgensen, K. A. Chem. Commun.
2000, 1449; For some recent examples, see: (b) Palomo, C.;
ꢀ
ꢀ
ꢀ
Oiarbide, M.; Arceo, E.; Garcıa, J. M.; Lopez, R.; Gonzalez,
A.; Linden, A. Angew. Chem., Int. Ed. 2005, 44, 6187; (c)
Kano, T.; Hashimoto, T.; Maruoka, K. J. Am. Chem. Soc.
2005, 127, 11926; (d) Sibi, M. P.; Ma, Z.; Jasperse, C. P.
J. Am. Chem. Soc. 2005, 127, 5764; (e) Suga, H.; Nakajima,
T.; Itoh, K.; Kakehi, A. Org. Lett. 2005, 7, 1431; (f) Sibi,
M. P.; Ma, Z.; Itoh, K.; Prabagaran, N.; Jasperse, C. P.
Org. Lett. 2005, 7, 2349; (g) Ashizawa, T.; Ohtsuki, N.;
Miura, T.; Ohya, M.; Shinozaki, T.; Ikeno, T.; Yamada,
T. Heterocycles 2006, 68, 1801; (h) Argyropoulos, N. G.;
Panagiotidis, T.; Coutouli-Argyropoulou, E.; Raptopoulou,
C. Tetrahedron 2007, 63, 321.
1
27.1 min (2R,3R,4S,5R)-27, 220 nm; H NMR (200 MHz):
d 7.56–7.52 (m, 2H), 7.45–7.18 (m, 8H), 5.19 (t, J¼8.0 Hz,
1H), 4.73 (d, J¼7.8 Hz, 1H), 4.48 (d, J¼9.1 Hz, 1H), 4.35
(t, J¼7.8 Hz, 1H), 3.26 (s, 3H), 2.71 (br s, 1H).
endo-27:15c,24 1H NMR (300 MHz): d 7.42–7.26 (m, 10H),
5.28 (dd, J¼6.1, 3.1 Hz, 1H), 4.92 (dd, J¼10.9, 6.7 Hz,
1H), 4.22 (d, J¼7.4, 3.5 Hz, 1H), 4.12 (d, J¼7.2 Hz, 1H),
3.81 (s, 3H), 3.36 (br s, 1H).
4.1.2.21. CuI-complex (Rp,R)-28. To a solution of 1a
(20.0 mg, 0.044 mmol) and Cu(CH3CN)4ClO4 (14.6 mg,
0.044 mmol) in CH2Cl2 (1 mL), a solution of 5a (8.4 mg,
0.044 mmol) in CH2Cl2 (0.5 mL) was added under a nitrogen
atmosphere at ambient temperature. The reaction mixture
was stirred for 30 min and the solvent was evaporated to
yield the presumed complex (RP,R)-28 in quantitative yield
3. (a) Kitagaki, S.; Anada, M.; Kataoka, O.; Matsuno, K.; Umeda,
C.; Watanabe, N.; Hashimoto, S. J. Am. Chem. Soc. 1999, 121,
1417; (b) Doyle, M. P.; Forbes, D. C. Chem. Rev. 1998, 98, 911.
4. (a) Sibi, M. P.; Itoh, K.; Jasperse, C. P. J. Am. Chem. Soc.
2004, 126, 5366; (b) Tsuji, M.; Ukaji, Y.; Inomata, K. Chem.
Lett. 2002, 1112; (c) Zou, N.; Jiang, B. J. Comb. Chem. 2000,
2, 6.
1
(35.7 mg): H NMR (300 MHz): d 8.97 (s, 1H, CH]N),
7.97 (d, J¼7.4 Hz, 2H, Ar), 7.81–7.77 (m, 2H, Ar), 7.55–
7.23 (m, 9H, Ar), 6.73 (t, J¼7.7 Hz, 2H, Ar), 5.11 (part A,
AB system, J¼20.3 Hz, 1H, CH2), 5.08 (part B, AB system,
5. (a) Kano, T.; Hashimoto, T.; Maruoka, K. J. Am. Chem. Soc.
2006, 128, 2174; (b) Kanemasa, S.; Tanai, T. J. Am. Chem.
Soc. 2000, 122, 10710.

6602
S. Cabrera et al. / Tetrahedron 63 (2007) 6587–6602
ꢀ
6. (a) Suarez, A.; Downey, W.; Fu, G. C. J. Am. Chem. Soc. 2005,
127, 11244; (b) Shintani, R.; Fu, G. C. J. Am. Chem. Soc. 2003,
125, 10778.
(b) Kano, T.; Hashimoto, T.; Maruoka, K. J. Am. Chem. Soc.
2005, 127, 11926.
22. Zubia, A.; Mendoza, L.; Vivanco, S.; Aldaba, E.; Carrascal,
T.; Lecea, B.; Arrieta, A.; Zimmerman, T.; Vidal-
7. Sibi, M. P.; Stanley, L. M.; Jasperse, C. P. J. J. Am. Chem. Soc.
2005, 127, 8276.
8. Reviews: (a) Najera, C.; Sansano, J. M. Angew. Chem., Int. Ed.
ꢀ
Vanaclocha, F.; Cossıo, F. P. Angew. Chem., Int. Ed. 2005,
44, 2903.
ꢀ
2005, 44, 6272; (b) Husinec, S.; Savic, V. Tetrahedron:
Asymmetry 2005, 16, 2047.
23. Amornraksa, K.; Grigg, R.; Guranatne, H. Q. N.; Kemp, J.;
Sridharan, V. J. Chem. Soc., Perkin Trans. 1 1987, 2285.
24. Vivanco, S.; Lecea, B.; Arrieta, A.; Prieto, P.; Morao, I.;
9. (a) Allway, P.; Grigg, R. Tetrahedron Lett. 1991, 32, 5817; (b)
Grigg, R. Tetrahedron: Asymmetry 1995, 6, 2475.
10. (a) Gothelf, A. S.; Gothelf, K. V.; Hazell, R. G.; Jørgensen,
K. A. Angew. Chem., Int. Ed. 2002, 41, 4236; (b) Alemparte,
C.; Blay, G.; Jørgensen, K. A. Org. Lett. 2005, 7, 4569.
11. Longmire, J. M.; Wang, B.; Zhang, X. J. Am. Chem. Soc. 2002,
124, 13400.
ꢀ
Linden, A.; Cossıo, F. P. J. Am. Chem. Soc. 2000, 122, 6078.
25. A similar behaviour was observed in related cycloadducts.
See for example: Woller, P. B.; Cromwell, N. H. J. Org.
Chem. 1970, 35, 888.
26. Due to the low stability of 5g and 5h, they have to be immedi-
ately used after preparation.
12. Chen, C.; Li, X.; Schreiber, S. L. J. Am. Chem. Soc. 2003, 125,
10174.
13. Zeng, W.; Zhou, Y.-G. Org. Lett. 2005, 7, 5055.
14. Zeng, W.; Chen, G.-Y.; Zhou, Y.-G.; Li, Y.-X. J. Am. Chem.
Soc. 2007, 129, 750.
27. The racemic reaction using Cu(CH3CN)4ClO4/PPh3 (5 mol %)
or Cu(CH3CN)4ClO4/rac-1a (5 mol %) only yielded endo
adduct.
28. Tsuge, O.; Kanemasa, S.; Yoshioka, M. J. Org. Chem. 1988, 53,
1384.
15. (a) Oderaotoshi, Y.; Cheng, W.; Fujitomi, S.; Kasano, Y.;
Minakata, S.; Komatsu, M. Org. Lett. 2003, 5, 5043; (b) Gao,
W.; Zhang, X.; Raghunath, M. Org. Lett. 2005, 7, 4241; (c)
Yan, X.-X.; Peng, Q.; Zhang, Y.; Zhang, K.; Hong, W.; Hou,
X.-L.; Wu, Y.-D. Angew. Chem., Int. Ed. 2006, 45, 1979.
16. For a recent review on chiral ferrocene ligands in asymmetric
29. Parr, R. G.; Yang, W. Density-Functional Theory of Atoms and
Molecules; Oxford: New York, NY, 1989.
30. (a) Kohn, W.; Becke, A. D.; Parr, R. G. J. Phys. Chem. 1996,
100, 12974; (b) Becke, A. D. J. Chem. Soc. 1993, 98, 5648;
(c) Becke, A. D. Phys. Rev. A 1988, 38, 3098.
31. Hehre, W. J.; Radom, L.; Schleyer, P. v. R.; Pople, J. A. Ab ini-
tio Molecular Orbital Theory; Wiley: New York, NY, 1986;
pp 76–87 and references therein.
ꢀ
catalysis, see: Gomez Arrayas, R.; Adrio, J.; Carretero, J. C.
Angew. Chem., Int. Ed. 2006, 45, 7674.
ꢀ
ꢀ
17. (a) Priego, J.; Garcıa Mancheno, O.; Cabrera, S.; Gomez
Arrayas, R.; Llamas, T.; Carretero, J. C. Chem. Commun.
˜
ꢀ
32. Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299.
33. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.;
Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.;
Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi,
M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.;
Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.;
Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.;
Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.;
Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.;
Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.;
Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.;
Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski,
V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.;
Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman,
J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.;
Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.;
Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith,
T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.;
Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03,
Revision C.02; Gaussian: Wallingford, CT, 2004.
ꢀ
ꢀ
2002, 2512; (b) Garcıa Manchen˜o, O.; Priego, J.; Cabrera, S.;
Gomez Arrayas, R.; Llamas, T.; Carretero, J. C. J. Org.
ꢀ
ꢀ
ꢀ
Chem. 2003, 68, 3679; (c) Garcıa Mancheno, O.; Gomez
Arrayas, R.; Carretero, J. C. J. Am. Chem. Soc. 2004, 126,
˜
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
456; (d) Garcıa Manchen˜o, O.; Gomez Arrayas, R.;
Carretero, J. C. Organometallics 2005, 24, 557; (e) Cabrera,
ꢀ
S.; Gomez Arrayas, R.; Carretero, J. C. Angew. Chem., Int.
Ed. 2004, 43, 3944; (f) Cabrera, S.; Gomez Arrayas, R.;
ꢀ
ꢀ
ꢀ
Alonso, I.; Carretero, J. C. J. Am. Chem. Soc. 2005, 127,
ꢀ
17938; (g) Salvador Gonzalez, A.; Gomez Arrayas, R.;
Carretero, J. C. Org. Lett. 2006, 8, 2977.
ꢀ
ꢀ
ꢀ
18. For a preliminary communication, see: Cabrera, S.; Gomez
Arrayas, R.; Carretero, J. C. J. Am. Chem. Soc. 2005, 127,
ꢀ
16394.
19. For the synthesis of aryl- and alkylimino esters, see: Cooper,
D. M.; Grigg, R.; Hargreaves, S.; Kennewell, P.; Redpath, J.
Tetrahedron 1995, 51, 7791.
20. For trans alkenes, endo refers to the cycloadduct with the
electron-withdrawing group at C-4 and the aromatic groups at
C-5 are cis to each other. Similarly, exo refers to the cycload-
duct where the same groups are trans to each other.
21. For 1,3-dipolar cycloadditions of nitrones with acroleins, see:
34. (a) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988,
88, 899; (b) Reed, A. E.; Weinstock, R. B.; Weinhold, F.
J. Chem. Phys. 1985, 83, 735.
35. McIver, J. W., Jr.; Komornicki, A. J. Am. Chem. Soc. 1972, 94,
2625.
ꢀ
(a) Carmona, D.; Lamata, M. P.; Viguri, F.; Rodrıguez, R.;
Oro, L. A.; Balana, A. I.; Lahoz, F. J.; Tejero, T.; Merino, P.;
Franco, S.; Montesa, I. J. Am. Chem. Soc. 2004, 126, 2716;