Copper-catalyzed skeletal rearrangement of o -propargylic aryloximes into four-membered cyclic nitrones - Chirality transfer and mechanistic insight

Article abstract of DOI:10.1055/s-0031-1290819

Copper-catalyzed skeletal rearrangement of O-propargylic aryloximes (E)-1 were carried out to afford the corresponding four-membered cyclic nitrones 2 in good to excellent yields. The optimal reactions conditions of the highly regioselective reactions inv

Full text of DOI:10.1055/s-0031-1290819

1542▌
SPECIAL TOPIC
sCpeociapl tpopeicr-Catalyzed Skeletal Rearrangement of O-Propargylic Aryloximes into
Four-Membered Cyclic Nitrones – Chirality Transfer and Mechanistic Insight
Preparation of Four-Membered Cyclic Nitrones
Itaru Nakamura,*a,b Yu Kudo,b Toshiharu Araki,b Dong Zhang,b Eunsang Kwon,a Masahiro Teradaa,b
a
Research and Analytical Center for Giant Molecules, Graduate School of Science, Tohoku University, Sendai 980-8578 Japan
Fax +81(22)7956602; E-mail: itaru-n@m.tohoku.ac.jp
b
Department of Chemistry, Graduate School of Science, Tohoku University, Sendai 980-8578, Japan
Received: 05.03.2012; Accepted after revision: 09.03.2012
R4
O
R3
Abstract: Copper-catalyzed skeletal rearrangement of O-propar-
gylic aryloximes (E)-1 were carried out to afford the corresponding
four-membered cyclic nitrones 2 in good to excellent yields. The
optimal reactions conditions of the highly regioselective reactions
involved the use of [CuCl(cod)]2 in acetonitrile at 70 °C. In the case
of (Z)-1, however, the reaction proceeded in the absence of the cop-
per catalysts to afford the identical compound 2 in good yields. Fur-
thermore, the reactions were also carried out using chiral substrates
(R)-1 in the presence of Cu catalysts to afford (R)-2 with good levels
of chirality transfer.
R3
O
cat. Cu
N
N
R4
R2
R2
R1
R1
2
1
Scheme 1 Formation of cyclic nitrones 2
As illustrated in Scheme 2, the O-propargylic aryloxime
substrate 1a was readily prepared via three steps: 1) al-
kynylation of benzaldehyde, 2) Mitsunobu reaction with
N-hydroxyphthalimide (NHPI), along with in situ remov-
al of the phthaloyl group using hydrazine, and 3) conden-
sation of corresponding propargyloxyamine 3 with
benzaldehyde to afford a 92:8 mixture of isomers (E)-1a
and (Z)-1a (readily separated by silica gel column chro-
matography).
Key words: copper, skeletal rearrangement, oxime, nitrone, chiral-
ity transfer
Skeletal rearrangements that are catalyzed by π-acidic
metal complexes have served as one of the most attractive
transformations due to the effective cleavage of σ-bonds
such as carbon–carbon, carbon–heteroatom, or heteroat-
om–heteroatom bonds.1 For the catalytic skeletal rear-
rangements, dramatic substitution effects leading to
diverse molecular skeletons are often observed. More-
over, the reactions can be carried out under mild condi-
tions using readily accessible starting materials to afford
highly elaborate molecules that remain elusive using con-
ventional synthetic methods. Investigations of such cata-
lytic skeletal rearrangements have mainly focused on 1,n-
enynes2 and propargylic esters3 as substrates. Recently,
we have found that O-propargylic oximes can act as novel
substrates for the π-acidic metal-catalyzed skeletal rear-
rangements in providing various heterocyclic com-
pounds.4 Specifically, we reported that the copper-
catalyzed skeletal rearrangement of (E)-O-propargylic ar-
yloximes 1 afforded the corresponding four-membered
cyclic nitrones 2 in good to excellent yields with high re-
gioselectivities (Scheme 1).4a,5,6 Herein, we describe the
full scope of the copper-catalyzed skeletal rearrangement
of O-propargylic aryloximes 1, including the thermal re-
arrangement of (Z)-1 in the absence of the copper cata-
lysts. Moreover, the copper-catalyzed reaction of chiral
substrates (R)-1 afforded (R)-2 with good levels of chiral-
ity transfer.
OH
1) n-BuLi
2) PhCHO
1) NHPI, DIAD
Ph3P
Ph
H
Ph
2) NH2NH2•H2O
79% (2 steps)
quant.
Ph
Ph
H
O
NH2
O
PhCHO
87%
N
Ph
Ph
Ph
3
Ph
E/Z = ca. 92:8
1a
Scheme 2 Preparation of substrate 1a
In the case of substrate (E)-1a, which possesses phenyl
groups at the alkyne terminus, the propargylic position,
and the oxime moiety, the reaction was carried out in the
presence of CuBr (10 mol%) in toluene at 100 °C for 42
hours to afford the corresponding four-membered cyclic
nitrone (E)-2a in 96% yield (Table 1, entry 1). Although
CuCl also exhibited comparable catalytic activity, the use
of CuI was not very effective (entries 2 and 3, respective-
ly). The use of PtCl2 and InCl gave (E)-2a in lower yields
(entries 4 and 5, respectively), whereas the use of AuCl,
AuCl3, and AgOTf resulted in the decomposition of (E)-
1a (entries 6–8, respectively). Brønsted acids, such as
TfOH, did not help in promoting the reaction (entry 9). In
the absence of the Cu catalysts, the reaction of (E)-1a at
100 °C did not proceed; in fact, (E)-1a was recovered
SYNTHESIS 2012, 44, 1542–1550
Advanced online publication: 24.04.2012
0
0
3
9
-
7
8
8
1
1
4
3
7
-
2
1
0
X
DOI: 10.1055/s-0031-1290819; Art ID: SS-2012-C0227-ST
© Georg Thieme Verlag Stuttgart · New York

SPECIAL TOPIC
Preparation of Four-Membered Cyclic Nitrones
1543
quantitatively (entry 10). Among the various solvents 1j quickly decomposed under the reaction conditions (en-
used, toluene, 1,4-dioxane, and THF gave (E)-2a in good try 9). A 4-(trifluoromethyl)phenyl group at the propar-
yields (entries 1, 11, and 12, respectively). Interestingly, gylic position decreased the reaction rate, whereas a p-
although faster reactions times were observed using polar anisyl group at the propargylic position and the oxime
solvents such as acetonitrile and DMF, the yields were moiety completely inhibited the reaction (entries 10, 11,
lower than that using toluene (entries 14 and 15, respec- respectively).
tively). The use of a protic solvent such as ethanol resulted
in rapid decomposition of the starting material (E)-1a (en-
Table 2 Copper-Catalyzed Reactions of (E)-1b–ma
try 16).
H
Ar
Ar
O
N
CuBr (10 mol%)
toluene, 100 °C
Table 1 Catalytic Activities for the Reactions of (E)-1a
N
O
H
Ph
Ar
R1
Ar
Ph
O
(E)-2
R1
N
catalyst (10 mol%)
100 °C
N
(E)-1
O
Ph
R1
Ph
Entry 1
Ar
Time
(h)
2
Yield
(%)b
Ph
(E)-2a
Ph
(E)-1a
1
2
1b
4-F3CC6H4
4-ClC6H4
4-MeC6H4
4-MeOC6H4
n-Pr
Ph
Ph
Ph
Ph
Ph
16
24
29
39
10
12
18
72
14
67
14
20
2b
2c
2d
2e
2f
2g
2h
–
92
85
94
83
Entry Catalyst
Solvent
Time Yield Recovery of 1a
(h)
(%)a (%)a
1c
1d
1e
1f
96b
91
41
34
42
<1
<1
<1
<1
<1
93
93
75
1
2
CuBr
CuCl
CuI
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
1,4-dioxane
THF
42
23
43
12
33
15
15
2
<1
<1
54
3
4
3
80c
63d
80e
n.r.
5
4
PtCl2
InCl
<1
<15
<1
<1
<1
36
6
1g
1h
1i
H2C=CHCH2 Ph
5
7
c-C6H11
t-Bu
Ph
Ph
Ph
6
AuCl
AuCl3
AgOTf
TfOH
none
8
7
f
9
1j
H
–
–
8
10
11
12
1k
1l
Ph
4-F3CC6H4
4-MeOC6H4
4-MeC6H4
2k
–
61
9
24
24
11
11
26
f
Ph
–
10
11
12
13
14
15
16
>99
<1
<1
16
1m
4-MeC6H4
2m
81
CuBr
CuBr
CuBr
CuBr
CuBr
CuBr
a The reaction of (E)-1 (0.4 mmol) was carried out in the presence of
CuBr (10 mol%) in toluene (0.8 mL) at 100 °C.
b Isolated yield; n.r. = no reaction.
c An 81:19 mixture of the E- and Z-isomers was obtained.
d A 78:22 mixture of the E- and Z-isomers was obtained.
e A 93:7 mixture of the E- and Z-isomers was obtained.
f Decomposition of the starting material.
hexane
MeCN
DMF
3.5 82
2.5 62
<1
<1
<1
EtOH
2
<1
In the case of substrate (E)-1n, the reaction formed a
68:32 mixture of regioisomers 2n and 2n′ (combined
yield: 96%; Table 3, entry 2), which corresponded to the
switched positions of the substituents between the olefinic
moiety and the sp3 carbon of four-membered ring, and the
substituent at the olefinic moiety of the major regioisomer
2n was derived from that at propargylic position of the
substrate. In order to improve the regioselectivity, the re-
action conditions were re-examined, as summarized in
Table 3. Although copper(I) salts, such as CuCl, CuBr,
CuI, and CuOAc were less reactive and required longer
reaction times to completely consume the starting materi-
al (entries 1–4, respectively), copper(II) salts such as
CuCl2 and Cu(OAc)2 resulted in significant decomposi-
tion of the starting substrate (entries 5 and 6, respective-
a Determined by 1H NMR spectroscopy using 1,3-benzodioxole as an
internal standard.
b Isolated yield.
The conditions as described above (Table 1, entry 1) were
employed for the reactions of substrates (E)-1b–m, which
possess identical substituents at the propargyl (R2) and the
oxime (R3) positions (cf. Scheme 1, Table 2). The reaction
of aryl substrates 1b, 1c, 1d, and 1e afforded the E-isomer
as the sole product (entries 1–4, respectively), whereas the
reaction of substrates 1f, 1g, and 1h, which contain an al-
kyl group at R1, formed the corresponding Z-isomer as a
by-product (entries 5–7). A bulky tert-butyl group dis-
rupted the reaction (entry 8). The terminal alkyne group of
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2012, 44, 1542–1550

1544
I. Nakamura et al.
SPECIAL TOPIC
ly). Among the copper complexes, [CuCl(cod)]2 exhibited formed the E-isomer (entries 2 and 3). In contrast, sub-
favorable catalytic activity for the present reaction (en- strate (E)-1r, which possesses an electron-deficient 4-(tri-
tries 9 and 11–14). Furthermore, the regioselectivity was fluoromethyl)phenyl group, resulted in products with
significantly affected by the reaction solvent – in particu- lower regioselectivity (entry 5). Substrates (E)-1s and (E)-
lar, the use of acetonitrile gave 2n with high regioselectiv- 1t, which possess an alkyl group at the alkyne terminus,
ity (entry 13). Moreover, excellent regioselectivity was afforded nitrones 2s and 2t (entries 6 and 7, respectively)
attained by conducting the reaction at 70 °C (entry 14). in excellent yields with high regioselectivities.
The use of other transition metals, such as PtCl2 and InCl,
Substrates with an alkyl group at the propargylic position
was not effective, even in acetonitrile (entries 15 and 16,
(R2) required elevated reaction temperatures (100 °C) to
respectively).
form the desired products in good yields (Table 5, entries
Subsequently, the optimal conditions (Table 3, entry 14) 1 and 2). As a note, the reaction of substrate (E)-1x, which
were employed for the reactions of the remaining (E)-1 possesses a p-anisyl group as its oxime moiety, did not
substrates, as summarized in Table 4. Substrates (E)- form the desired product due to the decomposition of the
1n,o,p, which possess an electron-rich aromatic moiety at starting compound (entry 4). The reaction of ketoxime 1y
the alkyne terminus, afforded products with high regiose- afforded the corresponding product 2y in a moderate yield
lectivities (Table 4, entries 1–3, respectively) – in partic- (Scheme 3) – interestingly, (Z)-2y was obtained as the ma-
ular, bulky substituents at the R1 position selectively jor isomer.
Table 3 Optimization of Reaction Conditions
CF3
F3C
H
O
Ph
O
N
O
N
conditions
F3C
N
+
Ph
Ph
OMe
2n'
OMe
2n
OMe
(E)-1n
Yield (%)a
Entry
Catalyst (mol%)
Solvent
Temp (°C)
Time (h)
2n/2n′
E/Z of 2n
1
CuCl (10)
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
CH2Cl2
1,4-dioxane
MeCN
100
100
100
100
100
100
100
100
100
100
100
100
100
70
24
42
48
27
4
94
96
18
91
32
45
70:30
68:32
72:28
85:15
93:7
>99:1
>99 :1
>99:1
84:16
>95:5
>98:2
–
2
CuBr (10)
3
CuI (10)
4
CuOAc (10)
5
CuCl2 (10)
6
Cu(OAc)2 (10)
[Cu(OTf)]2·toluene (5)
CuBr(PPh3)3 (10)
[CuCl(cod)]2 (5)
[CuBr(cod)]2 (5)
[CuCl(cod)]2 (5)
[CuCl(cod)]2 (5)
[CuCl(cod)]2 (5)
[CuCl(cod)]2 (5)
PtCl2 (10)
45
22
48
7
67:33
–
<1b
7
8
23
69:31
73:27
94:6
>99:1
90:10
76:24
78:22
84:16
80:20
74:26
>98:2
>98:2
9
quant
85
10
11
12
13
14
15
16
6
4.5
7
quant
46
81:19
82:18
87:13
98:2
3
(90)
99 (92)
24
MeCN
14
9.5
41
MeCN
100
100
>98:2
82:18
InCl (10)
MeCN
28
a Combined yield of 2n and 2n′. The yield and ratio were determined using 1H NMR spectroscopy with CH2Br2 as an internal standard. Isolated
yield in parentheses.
b Decomposition of (E)-1n.
Synthesis 2012, 44, 1542–1550
© Georg Thieme Verlag Stuttgart · New York

SPECIAL TOPIC
Preparation of Four-Membered Cyclic Nitrones
1545
Table 4 Copper-Catalyzed Reactions of (E)-1n–ta
CF3
F3C
H
Ph
O
[CuCl(cod)]2 (5 mol%)
N
N
O
+
F3C
O
N
MeCN, 70 °C
R1
Ph
Ph
R1
2'
2
R1
(E)-1
R1
Yield (%)b
Entry
1
Time
2
2/2′
E/Z of 2
1
2
3
4
5
6
7
1n
4-MeOC6H4
2-MeOC10H6
2,6-(MeO)2C6H3
Ph
16 h
4 d
2n
2o
2p
2q
2r
2s
2t
92
82
91
86
89
91
95
98:2
95:5
99:1
94:6
91:9
97:3
93:7
74:26
97:3
1o
1p
1q
1r
1s
1t
4 d
95:5
24 h
48 h
36 h
7 d
73:27
74:26
73:27
83:17
4-F3CC6H4
n-Pr
c-C6H11
a The reaction of (E)-1 (0.4 mmol) was conducted in the presence of [CuCl(cod)]2 (5 mol%) in MeCN (0.8 mL) at 70 °C.
b Combined yield of 2 and 2′. The ratio was determined using 1H NMR spectroscopy.
Table 5 Copper-Catalyzed Reactions of (E)-1u–xa
H
R3
R3
O
R2
O
N
N
N
O
[CuCl(cod)]2 (5 mol%)
MeCN, 70 °C
R2
R3
+
R2
OMe
OMe
2
2'
OMe
(E)-1
R2
R3
Yield (%)b
Entry
1
Time (h)
2
2/2′
E/Z of 2
1c
2c
3
1u
1v
1w
1x
n-Pr
i-Pr
Ph
4-F3CC6H4
4-F3CC6H4
2-naphthyl
p-anisyl
24 h
25 h
18 h
16 h
2u
2v
2w
–
77
71
88
nd
98:2
95:5
97:3
–
73:27
74:26
74:26
–
4
Ph
a The reaction of (E)-1 (0.4 mmol) was conducted in the presence of 5 mol% of [CuCl(cod)]2 in MeCN (0.8 mL) at 70 °C.
b Combined yield of 2 and 2′. The ratio was determined using 1H NMR spectroscopy. nd = not detected.
c At 100 °C.
Ph
O
Ph
Ph
Ph
Ph
O
O
N
Ph
N
N
[CuCl(cod)]2 (5 mol%)
Ph
MeCN, 70 °C
12 h
Ph
Ph
OMe
OMe
2y
2y'
OMe
1y
45% (2y/2y' = 95:5)
E/Z of 2y = 12:88
Scheme 3 Cyclization of ketoxime 1y
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2012, 44, 1542–1550

1546
I. Nakamura et al.
SPECIAL TOPIC
F3C
F3C
H
Ph
O
O
N
conditions
N
O
N
+
F3C
MeCN, 70 °C
Ph
(Z)-1q
Ph
Ph
Ph
2q
20 h
110 h
2q'
Ph
with [CuCl(cod)]2 (5 mol%)
without Cu catalyst
45% (2q/2q' = 96:4)
E/Z of 2q = 72:28
91% (2q/2q' = 87:13)
E/Z of 2q = 83:17
Scheme 4 Cyclization reaction of (Z)-1q
Ar
O
Ar
O
Ph
O
without Cu
N
N
N
+
MeCN
Ph
Ar
p-anisyl
p-anisyl
p-anisyl
2n
(E)-2n'
Ph
(Z)-2n
at 100 °C, 12 h 83%, 2n/2n' = 41:59, E/Z of 2n = 93:7
at 70 °C, 14 h 92%, 2n/2n' = 92:8, E/Z of 2n = 20:80
Ar = 4-F3CC6H4
Scheme 5 Isomerization of (Z)-2n
In the presence of [CuCl(cod)]2, both the reactions of (Z)-
1q and (E)-1q (Scheme 4) afforded 2q. In the absence of
copper catalysts, however, only the reaction of (Z)-1q
formed 2q, but required a longer reaction time.
R3
O
N
R3
R1
R2
O
N
O
R2
(Z)-2
N
The isomerization experiments were further examined to
gain insight into the mechanism of the reactions. Product
(Z)-2n, albeit a minor product, was successfully converted
to (E)-2n and (E)-2n′ in the absence of Cu catalysts at
100 °C. However, the conversion did not occur at a lower
reaction temperature of 70 °C (Scheme 5). In contrast,
(E)-2n remained unchanged, even at 100 °C (Scheme 6).7
These results strongly suggest that only the Z-isomer can
undergo isomerization, and help explain the higher regio-
selectivity and lower E/Z selectivity of the reaction of (E)-
1n under lower reaction temperatures (70 °C rather than
100 °C) (Table 3, entries 13 and 14). We believe that the
isomerization of (Z)-2 to (E)-2 and (E)-2′ proceeds
through cleavage and reformation of the sp3 carbon–nitro-
gen bond via zwitterionic intermediate 4, which possesses
an allylic cation moiety (Scheme 7). Presumably, the re-
laxation of the 1,3-allylic strain between R1 and R2 within
(Z)-2 helps drive the ring-opening step. Accordingly, the
low regioselectivity of substrate (E)-1r can be explained
as the stabilization of the anionic moiety of intermediate 4
due to the electron-withdrawing 4-(trifluoromethyl)phe-
nyl group at R1 that facilitates the isomerization to the mi-
nor regioisomer 2r′, even at 70 °C (Table 4, entry 5).
R1
R3
R1
R2
(E)-2'
R3
O
4
N
R2
R1
(E)-2
Scheme 7 Isomerization of (Z)-2 through zwitterionic intermediate 4
Moreover, the results of our 13C-labeling experiments
(Scheme 8) are in agreement with the isomerization mech-
anism involving zwitterionic intermediate 4 (Scheme 7).5a
Specifically, the reaction of (E)-1a-c (R2 = R3 = Ph), in
which the alkyne sp-carbons were enriched with 13C (15%
and 85%, Scheme 8), was carried out in the presence of
CuBr in toluene at 100 °C to afford (E)-2a-c, which pos-
sessed 15% 13C at the nitrone carbon and 85% 13C at the
sp2-carbon of the four-membered ring bound to the ben-
zylidene group. This result indicates that our reaction does
not proceed via cleavage of the carbon–carbon bond of the
alkyne triple bond, but rather via cleavage of the carbon–
oxygen bond.
The reaction of an equimolar mixture of (E)-1a and (E)-
1m afforded only the normal products (E)-2a and (E)-2m;
the formation of crossover products was not observed us-
ing LC-MS (Scheme 9). This result clearly indicates that
the present reaction proceeds in an intramolecular man-
ner.
Ar
O
without Cu
MeCN
N
no reaction
Ph
p-anisyl
(E)-2n
Ar = 4-F3CC6H4
Scheme 6 Attempted isomerization of (E)-2n
Synthesis 2012, 44, 1542–1550
© Georg Thieme Verlag Stuttgart · New York

SPECIAL TOPIC
Preparation of Four-Membered Cyclic Nitrones
1547
The reaction of the corresponding Z-isomer (R,Z)-1q
(Scheme 11) afforded the same enantiomer (–)-2q as that
of the reaction of (R,E)-1q (Scheme 10).
H
O
Ph
Ph
O
N
C2 C1
N
CuBr (10 mol%)
toluene, 100 °C, 24 h
Ph
Ph
Ph
C2
C1
H
Ph
F3C
(E)-2a–c
(E)-1a–c
F3C
92%
C1: 13C (15%)
C2: 13C (85%)
H
O
conditions
N
Scheme 8 13C-Labeling experiments
N
O
*
MeCN, 70 °C
Ph
Ph
Ph
2q
Ph
H
O
Ph
H
O
Ar
(R,Z)-1q
99% ee
N
N
CuBr (10 mol%)
+
with [CuCl(cod)]2 (5 mol%) 20 h
70%, 2q/2q' = 95:5, E/Z of 2q: 73:27
ee of (–,E)-2q = 87%
toluene, 100 °C, 24 h
Ph
Ar
ee of (+,Z)-2q = 89%
Ph
Ar
O
without Cu catalyst
110 h
90%, 2q/2q' = 92:8, E/Z of 2q: 76:24
ee of (–,E)-2q = 95%
(E)-1a
(E)-1m
Ar = 4-tolyl
ee of (+,Z)-2q = 99%
Ph
Ar
O
Scheme 11 Cyclization of chiral Z-isomer (R,Z)-1q
N
N
+
Ph
Ar
Ph
Ar
H
(E)-2a
H
Similarly, the chirality of substrates 1s, 1aa, and 1ab were
effectively maintained during the formation of products
2s, 2aa, and 2ab as summarized in Table 6.
(E)-2m
77%
80%
Scheme 9 Crossover experiments with (E)-1a and (E)-1m
To determine the absolute configuration of the major
product, the reaction of substrate (R,E)-1z, which has a 2-
bromo-5-nitrophenyl group at the oxime moiety, was car-
ried out under standard reaction conditions (Scheme 12).
Although the R-configuration at the sp3-carbon of major
product (E)-2z was confirmed using X-ray crystallo-
graphic analysis (see Supporting Information); the yield,
the ratio between 2z and 2z′, the E/Z ratio of 2z, and the
ee of (E)-2z were not determined because the crystallinity
of 2z was too high to conduct these analyses. And, the ab-
solute configuration of the Z-product (Z)-2z cannot be de-
termined at the present stage due to its poor crystallinity.
To gain further insight into the reaction mechanism, we
investigated the transfer of chirality during the Cu-cata-
lyzed skeletal rearrangements (Scheme 10) using chiral
substrates. Optically active (R,E)-1q was readily prepared
via In-catalyzed asymmetric alkynylation,8 then subjected
to the rearrangement reaction under optimal reaction con-
ditions (Table 3, entry 14). The resulting products (–,E)-
and (+,Z)-2q were obtained in 77% ee and 80% ee, respec-
tively, suggesting the effective transfer of chirality of the
starting material. Furthermore, the choice of solvent sig-
nificantly affected the chirality transfer – acetonitrile was
the most effective – whereas less polar solvents such as
CH2Cl2 and toluene decreased the enantioselectivity of
the reaction (see Supporting Information). As a note, it
was confirmed that racemization of (R,E)-1q and the iso-
lated chiral products (–,E)-2q did not occur under the re-
action conditions.
A plausible mechanism for the reaction of the O-propar-
gylic aldoxime (E)-1 is illustrated in Scheme 13. First, the
π-acidic copper catalyst coordinates with the alkyne moi-
ety of (E)-1 to allow the nucleophilic attack by the oxime
nitrogen atom onto the electrophilically-activated triple
bond. The resulting five-membered cyclic intermediate 6
undergoes cleavage of the carbon–oxygen bond and elim-
ination of the copper catalyst9 to afford N-allenylnitrone
CF3
F3C
F3C
H
[CuCl(cod)]2 (5 mol%)
N
O
O
+
O
MeCN, 70 °C
24 h
N
N
*
*
92%
Ph
Ph
Ph
Ph
Ph
(R,E)-1q
99% ee
Ph
(–,E)-2q
(+,Z)-2q
2q/2q' = 96:4 E/Z of 2q: 72:28
ee of (–,E)-2q = 77%
ee of (+,Z)-2q = 80%
Scheme 10 Cyclization of chiral substrate (R,E)-1q
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2012, 44, 1542–1550

1548
I. Nakamura et al.
SPECIAL TOPIC
Table 6 Chirality Transfer for Copper-Catalyzed Reactions of 1s,
1aa, and 1ab
intermediate 7, which then rotates to rotational conformer
7′ that undergoes cyclization to afford product 2.10 In the
case of (Z)-1, the [2,3] rearrangement proceeds in a con-
certed manner to form N-allenylnitrone intermediate (Z)-
7, without the aid of the copper catalyst, due to the lower
steric repulsion between the alkyne substituent R1 and the
oxime moiety in the five-membered cyclic transition state
8 (Scheme 14).4e Should the cyclization of the chiral al-
lene intermediate (E)-7′ proceed via a conrotatory 4π-
electrocyclization, the sp3-carbon would adopt an S-con-
figuration. However, the resulting R-configuration sug-
gests that the aldonitrone moiety undergoes an E/Z
isomerization to favor the more stable (Z)-7′ prior to the
thermal cyclization.11 It should be noted that the reaction
of ketoxime 1y (Scheme 3) suggests that cyclization of
(E)-7′ can also take place. Accordingly, the lower level of
the chirality transfer during the formation of (E)-2q can be
attributed to the competitive 4π-cyclizations of the E- and
Z-forms of N-allenylnitrone intermediate 7′. This is in
sharp contrast to the high level of chirality transfer for (Z)-
1q, which proceeds directly via (Z)-N-allenylnitrone in-
termediate (Z)-7′. Moreover, isomerization from (S,Z)-2
to (S,E)-2 could result in low ee of the E-isomer (E)-2
(Scheme 5). At the present stage, it is unclear whether the
loss of ee takes place during the copper-catalyzed [2,3] re-
arrangement step from (E)-1 to 7.
H
R3
R3
O
[CuCl(cod)]2 (5 mol%)
MeCN, 70 °C
N
N
O
*
Ph
R1
Ph
(E)-2
R1
(R,E)-1
99% ee
R1
R3
E/Zb
1
Time 2
(h)
Yield
(%)a
ee
(E/Z)c
92d
90e
89f
1
1s
n-Pr 4-F3CC6H4 36
2s
72:28 78/78
72:28 81/85
2
3
1aa Ph
1ab Ph
4-BrC6H4
16
2aa
2ab
4-MeC6H4 20
–
76/75
a Isolated yield.
b The ratio was determined by 1H NMR analysis.
c The value for ee was determined by HPLC analysis using a chiral
stationary phase.
d The ratio of 2s/2s′ was 96:4.
e The ratio of 2aa/2aa′ was >99:1.
f The ratio of 2ab/2ab′ was 98:2
Br
NO2
R3
O
H
H
O
R3
R1
NO2
O
N
O
[CuCl(cod)]2 (5 mol%)
MeCN, 70 °C
N
N
N
R2
•
R3
O
N
H
Br
Ph
R2
R2
R1
(Z)-7
H
Ph
Ph
R1
Ph
(Z)-1
8
(R,E)-1z
99% ee
(R,E)-2z
Scheme 14 Thermal [2,3] rearrangement of Z-oxime (Z)-1
Scheme 12 Cyclization of (R,E)-1z
O
N
H
O
R3
Cu(I)
R2
•·
H
N
R1
R3
(E)-7
R2
R3
O
R1
R3
R3
H
conrotatory
(E)-1
O
N
H
O
R3
4π
H
N
+
H
R2
N
•·
O
R1
R2
N
R1
R1
R2
(R,Z)-2
(S,E)-2
(E)-7'
R2
R1
R3
H
Cu
O
R3
5
R3
O
O
conrotatory
H
N
N
4π
H
+
R2
N
·
•
R1
R2
R2
R1
H
O
R2
(S,Z)-2
R1
(Z)-7'
(R,E)-2
N
R3
Cu
R1
6
Scheme 13 A plausible mechanism for the formation of four-membered nitrones
Synthesis 2012, 44, 1542–1550
© Georg Thieme Verlag Stuttgart · New York

SPECIAL TOPIC
Preparation of Four-Membered Cyclic Nitrones
1549
IR (neat): 3057, 3023, 3002, 2965, 2938, 2839, 1691, 1599, 1586,
1547, 1473, 1446, 1432, 1402, 1322, 1257, 1165, 106, 1065, 1019,
859, 818 cm–1.
1H NMR (500 MHz, CDCl3): δ = 3.88 (s, 6 H), 6.16 (s, 1 H), 6.35
(s, 1 H), 6.63 (d, J = 8.5 Hz, 2 H), 7.08–7.18 (m, 5 H), 7.41 (d, J =
8.5 Hz, 1 H), 7.64 (d, J = 8.0 Hz, 2 H), 7.71 (d, J = 8.0 Hz, 2 H).
13C NMR (125 MHz, CDCl3): δ = 56.00, 84.66, 103.22, 103.87,
104.11, 116.16, 120.59, 122.77, 124.93, 125.85, 125.88, 125.91,
125.94, 127.09, 127.30, 127.42, 128.50, 129.11, 131.06, 131.32,
131.83, 131.97, 132.83, 135.05, 136.02, 148.84, 159.36.
In conclusion, we have developed novel synthetic meth-
odology of preparing four-membered cyclic nitrones. In
particular, nitrogen-containing chiral four-membered
rings can be constructed with chirality transfer using read-
ily accessible starting materials. For example, therefore,
our methodology can be useful for the synthesis of azeti-
dine derivatives. Further investigations on manipulation
of the products are currently under way in our laboratory.
1H and 13C NMR spectra were recorded on a JEOL JNM-α 500 (500
MHz for 1H and 125 MHz for 13C) spectrometer. Chemical shifts are
reported in ppm relative to CHCl3 (for 1H, δ = 7.24), and CDCl3 (for
HRMS (ESI): m/z calcd for (C25H20F3NO3 + Na)+: 462.1287; found:
462.1284.
1
13C, δ = 77.00). H NMR data are reported as follows: chemical
Acknowledgment
shift, multiplicity (standard abbreviations were used to indicate
multiplicities), coupling constants (Hz), and integration. IR spectra
were recorded on a JASCO FT/IR-4100 spectrometer. High-resolu-
tion mass spectra analysis was performed on a Bruker Daltonics
APEX III FT-ICR-MS spectrometer at the Instrumental Analysis
Center for Chemistry, Graduate School of Science, Tohoku Univer-
sity. X-ray crystallographic data was obtained by Rigaku/MSC Sat-
urn Cu-CCD device at Graduate School of Science, Tohoku
University. Flash column chromatography was performed with
Kanto Chemical silica gel 60N (spherical, neutral, 40–50 μm). An-
alytical TLC was performed on Merck precoated TLC plates (silica
gel 60 F254). All reactions were carried out under argon atmosphere.
Anhyd DMF, MeCN, 1,4-dioxane, THF, toluene, hexane (WAKO),
and CuBr (WAKO, 99.9%) were purchased and used without fur-
ther purification. CuCl (WAKO) was purified by recrystallization
prior to use. [CuCl(cod)]2 was prepared in accordance with the lit-
erature method.12
We thank Prof. Takeaki Iwamoto and Prof. Shintaro Ishida (Depart-
ment of Chemistry, Graduate School of Science, Tohoku Universi-
ty) for X-ray crystallographic expertise [(R,E)-2z]. This work was
supported by a Grant-in-Aid for Scientific Research on Innovative
Areas ‘Molecular Activation Directed toward Straightforward Syn-
thesis’ from the Ministry of Education, Culture, Sports, Science and
Technology, Japan.
Supporting Information for this article is available online at
http://www.thieme-connect.com/ejournals/toc/synthesis.omnorinIftangonmSorniuifIatoprtngiSutpor
References
(1) (a) Rautenstrauch, V. J. Org. Chem. 1984, 49, 950. (b) Trost,
B. M.; Lautens, M. J. Am. Chem. Soc. 1985, 107, 1781.
(c) Trost, B. M.; Tanoury, G. J. J. Am. Chem. Soc. 1988, 110,
1636. (d) Chatani, N.; Morimoto, T.; Muto, T.; Murai, S.
J. Am. Chem. Soc. 1994, 116, 6049.
(2) (a) Aubert, C.; Buisine, O.; Malacria, M. Chem. Rev. 2002,
102, 813. (b) Diver, S. T.; Giessert, A. J. Chem. Rev. 2004,
104, 1317. (c) Añorbe, L.; Domínguez, G.; Pérez-Castells, J.
Chem.–Eur. J. 2004, 10, 4938. (d) Bruneau, C. Angew.
Chem. Int. Ed. 2005, 44, 2328. (e) Zhang, L.; Sun, J.;
Kozmin, S. A. Adv. Synth. Catal. 2006, 348, 2271.
(f) Michelet, V.; Toullec, P. Y.; Genêt, J.-P. Angew. Chem.
Int. Ed. 2008, 47, 4268. (g) Jiménez-Núñez, E.; Echavarren,
A. M. Chem. Rev. 2008, 108, 3326. (h) Tobisu, M.; Chatani,
N. Chem. Soc. Rev. 2008, 37, 300.
CuBr-Catalyzed Synthesis of (E)-3-Benzylidene-2,4-diphenyl-
2,3-dihydroazete 1-Oxide [(E)-2a]; Typical Procedure
To a mixture of CuBr (5.72 mg, 0.040 mmol) and oxime (E)-1a
(124.5 mg, 0.40 mmol) in a vial was added toluene (0.8 mL) under
an argon atmosphere. After stirring at 100 °C for 24 h, the crude
mixture was passed through a pad of silica gel with EtOAc (ca. 30
mL). Upon removal of the solvents in vacuo, the residue was puri-
fied using flash column chromatography with hexane–EtOAc (4:1)
as the eluent to afford (E)-2a (121 mg, 96%) as a white solid; mp
189.8–190.6 °C.
IR (neat): 3052, 3028, 2964, 1954, 1893, 1685, 1576, 1545, 1488,
1409, 1311, 1224, 1207, 1193, 1152, 859 cm–1.
1H NMR (500 MHz, CDCl3): δ = 6.17 (s, 1 H), 6.94 (d, J =1.5 Hz,
1 H), 7.13–7.20 (m, 5 H), 7.34–7.38 (m, 3 H), 7.45–7.53 (m, 5 H),
8.18–8.20 (m, 2 H).
13C NMR (125 MHz, CDCl3): δ = 86.30, 117.12, 126.46, 127.08,
127.62, 127.89, 128.57, 128.61, 128.81, 129.04, 129.78, 130.15,
130.21, 131.22, 134.95, 149.68.
(3) (a) Wang, S.; Zhang, G.; Zhang, L. Synlett 2010, 692.
(b) Marion, N.; Nolan, S. P. Angew. Chem. Int. Ed. 2007, 46,
2750. (c) Correa, A.; Marion, N.; Fensterbank, L.; Malacria,
M.; Nolan, S. P.; Cavallo, L. Angew. Chem. Int. Ed. 2008,
47, 718.
(4) (a) Nakamura, I.; Araki, T.; Zhang, D.; Kudo, Y.; Kwon, E.;
Terada, M. Org. Lett. 2011, 13, 3616. (b) Nakamura, I.;
Zhang, D.; Terada, M. J. Am. Chem. Soc. 2010, 132, 7884.
(c) Nakamura, I.; Zhang, D.; Terada, M. J. Am. Chem. Soc.
2011, 133, 6862. (d) Nakamura, I.; Okamoto, M.; Terada,
M. Org. Lett. 2010, 12, 2453. (e) Nakamura, I.; Zhang, D.;
Terada, M. Tetrahedron Lett. 2011, 52, 6470. (f) Nakamura,
I.; Iwata, T.; Zhang, D.; Terada, M. Org. Lett. 2012, 14, 206.
(5) (a) We previously reported the structure of the product as 4-
arylidene-β-lactam. However, thorough analysis of X-ray
crystallographic data indicated that the structure of cyclic
nitrone was more valid than that of β-lactam. Moreover,
vibrational analysis by DFT calculations suggests the cyclic
nitrone structure. (b) Nakamura, I.; Araki, T.; Terada, M.
J. Am. Chem. Soc. 2009, 131, 2804 (withdrawn).
HRMS (ESI): m/z calcd for (C22H17NO + Na)+: 334.1202; found:
334.1201.
CuBr-Catalyzed Synthesis of (E)-3-Benzylidene-4-(2,6-dimeth-
oxyphenyl)-2-[4-(trifluoromethyl)phenyl]-2,3-dihydroazete 1-
Oxide [(E)-2p]; Typical Procedure
To a mixture of [CuCl(cod)]2 (8.3 mg, 0.020 mmol) and oxime (E)-
1p (175.8 mg, 0.40 mmol) in a vial was added MeCN (0.8 mL) un-
der an argon atmosphere. After stirring at 70 °C for 14 h, the crude
mixture was passed through a pad of silica gel with EtOAc (ca. 30
mL). Upon removal of the solvents in vacuo, the residue was puri-
fied using flash column chromatography with hexane–EtOAc (4:1)
as the eluent to afford a 95:5 mixture of (E)-2p and (Z)-2p (161 mg,
91%). The E/Z (95:5) isomers were separated using flash silica gel
column chromatography.
(c) Nakamura, I.; Araki, T.; Terada, M. J. Am. Chem. Soc.
2011, 133, 6861.
(E)-2p
White solid; mp 75.5–77.6 °C.
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2012, 44, 1542–1550

1550
I. Nakamura et al.
SPECIAL TOPIC
(6) Four-membered cyclic nitrones: (a) Pennings, M. L. M.;
Reinhoudt, D. N.; Harkema, S.; van Hummel, G. J. J. Am.
Chem. Soc. 1980, 102, 7570. (b) Pennings, M. L. M.;
Reinhoudt, D. N. J. Org. Chem. 1983, 48, 4043.
(c) Verboom, W.; van Eijk, P. J. S. S.; Conti, P. O. M.;
Reinhoudt, D. N. Tetrahedron 1989, 45, 3131.
(7) The reaction of (E)-2n at 100 °C for 12 hours afforded a
>97:3 mixture of (E)-2n and (E)-2n′.
(8) (a) Takita, R.; Yakura, K.; Ohshima, T.; Shibasaki, M.
J. Am. Chem. Soc. 2005, 127, 13760. (b) Harada, S.; Takita,
R.; Ohshima, T.; Matsunaga, S.; Shibasaki, M. Chem.
Commun. 2007, 948.
(9) Pd(II)-catalyzed [2,3] rearrangement of O-allyl oximes:
Grigg, R.; Markandu, J. Tetrahedron Lett. 1991, 32, 279.
(10) 4π-Azabutadiene electrocyclization: (a) Bongini, A.;
Panunzio, M.; Tamanini, E.; Martelli, G.; Vicennati, P.;
Monari, M. Tetrahedron: Asymmetry 2003, 14, 993.
(b) Liang, Y.; Jiao, L.; Zhang, S.; Xu, J. J. Org. Chem. 2005,
70, 334.
(11) Aurich, H. G.; Franzke, M.; Kesselheim, H. P. Tetrahedron
1992, 48, 663.
(12) Chi, K. M.; Shin, H.-K.; Hampden-Smith, M. J.; Duesler, E.
N.; Kodas, T. T. Polyhedron 1991, 10, 2293.
Synthesis 2012, 44, 1542–1550
© Georg Thieme Verlag Stuttgart · New York