Synthesis of vicinally functionalized 1,4-dihydropyridines and diazabicycles via a pseudo-intramolecular process

Article abstract of DOI:10.1016/j.tet.2013.11.044

An α-nitro-δ-keto nitrile readily forms the corresponding ammonium salt immediately upon treatment with an amine. When the amine liberated under equilibrium, the nucleophilic amine and the electrophilic keto nitrile come close to each other to afford so-c

Full text of DOI:10.1016/j.tet.2013.11.044

Tetrahedron 70 (2014) 402e408
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Synthesis of vicinally functionalized 1,4-dihydropyridines and
diazabicycles via a pseudo-intramolecular process
y
z
*
Nagatoshi Nishiwaki , Shotaro Hirao , Jun Sawayama , Haruyasu Asahara,
Ryuichi Sugimoto, Kazuya Kobiro, Kazuhiko Saigo
School of Environmental Science and Engineering, Kochi University of Technology, Tosayamada, Kami, Kochi 782-8502, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 5 October 2013
Received in revised form 11 November 2013
Accepted 14 November 2013
Available online 21 November 2013
An a-nitro-d-keto nitrile readily forms the corresponding ammonium salt immediately upon treatment
with an amine. When the amine liberated under equilibrium, the nucleophilic amine and the electro-
philic keto nitrile come close to each other to afford so-called an intimate pair. The spatial proximity
realized an efficient reaction to give a 2-amino-3-nitro-1,4-dihydropyridine; the reaction proceeded like
an intramolecular reaction although it is actually an intermolecular reaction, namely the pseudo-
intramolecular reaction. The bifunctionality of the keto nitrile also enabled the pseudo-intramolecular
imination followed by tandem cyclization leading to diazabicyclic frameworks.
Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction
To the contrary, we proposed a new protocol, a pseudo-
intramolecular reaction, which efficiently proceeds to completion
The development of highly efficient synthetic methods for var-
ious organic compounds is crucial for the efficient utilization of
carbon resources and for low environmental impact. Reactions not
requiring a protecting group or a special reagent facilitate the
simplification of manipulations from the viewpoint of atom econ-
omy.1 The improvement of the efficiency of reactions is one of the
main approaches to environmentally benign methods, in which the
increase of the collision frequency of reactants should be consid-
ered as a significant factor. Intramolecular reactions proceed faster
than do intermolecular reactions because of the high collision
frequency of reaction sites, which can be attributed to the spatial
proximity. With regard to an intermolecular process, employment
of reaction fields, such as capsules, cages, bowls, and micelles has
been recognized as a useful method in organic synthesis to make
reaction sites close to one another, so that the reaction proceeds
rapidly to afford the product.2 However, it is still difficult to achieve
the high efficiency of the reaction under mild conditions without
using any contrivance, such as reaction field, catalyst, and activating
agent. From this viewpoint, development of the efficient protocol is
one of the challenging subjects.
under mild conditions even when no catalyst, additive, or special
manipulation is employed; in particular, this reaction can be carried
out even in the absence of the abovementioned reaction field.
Compounds having an acidic hydrogen as well as an appropriate
functionality can be used as
a substrate for the pseudo-
intramolecular reaction, and the substrate readily form an ammo-
nium salt upon treatment with an amine. When the amine is lib-
erated from the salt under equilibrium, the nucleophilic moiety
(amine) and the electrophilic functionality come close to each
other. Because of this spatial proximity, the reaction, which is ac-
tually intermolecular, behaves like an intramolecular manner, and
side reactions are suppressed. In the present process, the acidic
hydrogen atom is used as a lure for attracting the amine, and the
functional group bites it; this process is similar to the manner in
which a football fish attracts its prey (Scheme 1).
FG
FG
FG
RNH2
_
+
H
RNH2
H
RNH2
H
Intimate Pair
FG
HN
R
* Corresponding author. Tel./fax: þ81 887 57 2520; e-mail address: nishiwaki.
nagatoshi@kochi-tech.ac.jp (N. Nishiwaki).
H
y
Present address: Department of Applied Chemistry, Faculty of Engineering, Oita
University, Dannoharu, Oita, Oita 870-1192, Japan.
z
Present address: JST-ERATO Iyoda Supra-integrated Material Project, Tokyo
Scheme 1. The concept of the pseudo-intramolecular process.
Institute of Technology, Nagatsuta, Midori, Yokohama, Kanagawa 226-8503, Japan.
0040-4020/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tet.2013.11.044

N. Nishiwaki et al. / Tetrahedron 70 (2014) 402e408
403
Indeed, the a-arylated and a-nitrated b-keto esters 1a and 1b un-
O
O
N+
derwent the transfer of an acyl group to the amines very smoothly
yielding the amide 5 and deacylated ester 6, which is named as the
transacylation (Scheme 2).3e5 In these reactions, the highly acidic
keto ester reacted with the amine to form the corresponding am-
monium salt 3 immediately. Then, under equilibrium, the electro-
philic keto ester and the nucleophilic amine were regenerated
simultaneously in close proximity. The spatial proximity of the
reactants (the intimate pair 4) enhanced the reactivity, and con-
sequently, the reaction proceeded like an intramolecular process;
this has been confirmed by NMR study and by carrying out the
reaction using an amino alcohol in diluted conditions.4
O
C
N
H
+
H2N
Pr
9a
Fig. 2. The ammonium salt 9a.
When an acetonitrile solution of 9a was heated under reflux,
the 3-nitro-2-propylamino-1,4-dihydropyridine 10a was isolated
in 71% yield (Table 1, run 1).9 Although common transacylations
RNH2
2
X
X
X
H
OEt
OEt
OEt
EtO
EtO
EtO
_
rt, 1 d
CHCl3
O
O
O
O
O
O
O
O
+
O
RNH2
H
RNH2
H
3
X: 2,4-dinitrophenyl 1a
Intimate Pair
4
NO2
1b
X
OEt
RHN
EtO
+
O
O
O
6
5
Scheme 2. Transacylation of the keto esters 1 with the amine 2.
The pseudo-intramolecular process is considered to be a novel
method for synthesizing polyfunctionalized compounds. In this
process, it is crucial that suitable substratesdthose containing both
an acidic hydrogen and functionalities, such as carbonyl and cyano
groupsdare available; molecular design for these substrates is
via a pseudo-intramolecular process gave no by-product (Scheme
1), several kinds of unidentified products were formed in this
reaction presumably due to side reactions caused by reactive
multiple functionalities of the keto nitrile 7. The structure of the
dihydropyridine 10a was determined on the basis of spectral and
analytical data. The singlet signal at 7.51 ppm and the triplet signal
at 12.12 ppm were exchangeable with D2O; the former dis-
appeared immediately and the latter disappeared gradually. The
different exchange rates indicate the presence of an intra-
molecular hydrogen bond between an amino and a nitro group.
Furthermore, two gem-methyl groups were equivalently observed
despite a cyclic structure, suggesting that the dihydropyridine ring
is closely flat10 or undergoes a facile ring flipping in an NMR time
scale.
considerably easy. From this viewpoint, we focused on the
a-
nitrated
d
-keto nitriles 76 and 87 as substrates for the present
pseudo-intramolecular reaction and utilized them in the synthesis
of functionalized azaheterocyclic compounds.
2. Results and discussion
2.1. Synthesis of vicinally functionalized 1,4-dihydropyridines
Butylamine (2b) reacted in a similar manner with the keto ni-
The keto nitrile 7 was easily transformed to the ammonium salt
9a upon treatment with propylamine (2a); the formation was
confirmed by 1H NMR and IR spectra. Just after 2a was added to
a solution of the keto nitrile 7 in acetonitrile-d3, in the 1H NMR
trile
7 (Table 1, run 2), however, bulkier amines, such as
Table 1
Synthesis of the dihydropyridines 10
spectrum, the singlet signal at 6.29 ppm assigned to the acidic a-
O
R1R2NH
2
proton immediately disappeared, and the signals for the propyl
group shifted to the downfield, indicating the formation of the
ammonium salt 9a (Fig. 1). The absorption of the cyano group in the
IR spectrum of 9a was observed to be as strong as that of the car-
bonyl group, while, only very weak cyano group absorption could
be detected in the spectrum of the starting keto nitrile 7. The strong
O
_
N+
O
NO2
H
R2
MeCN
reflux 40 h
N
N
H
C
N
R1
7
10
Run
R1
R2
Yield/%
absorption of a cyano group is a typical feature of an
a-cyanonitr-
onate framework8 (Fig. 2).
1
2
3
4
5
6
7
8
EteCH2
PreCH2
Me2CH
Me3C
H
H
H
H
H
a
71
65
0
0
0
b
c
d
e
f
g
h
i
Ph
O
O
EteCH2
EteCH2
0
N
N
i-PreCH2
t-BueCH2
c-HexeCH2
PheCH2
(MeO)2CHeCH2
H
H
H
H
H
63
48
59
80
74
C
C
H
H
9
10
11
NO2
NO2
7
8
j
k
Fig. 1.
a-Nitroketo nitriles 7 and 8.

404
N. Nishiwaki et al. / Tetrahedron 70 (2014) 402e408
isopropylamine (2c), tert-butylamine (2d), aniline (2e), and dipro-
pylamine (2f) did not cause the cyclization (runs 3e6). The high
sensitivity to steric hindrance is likely due to the congestion around
the reaction site of the intimate pair 11. Hence, the insertion of
a methylene group as a spacer between the bulky group and the
amino function facilitated the cyclization to afford the dihy-
dropyridines 10gek, respectively (runs 7e11).9
prototype d
-keto nitrile 12,14 which has no nitro group, unreacted
with the amine 2a under the similar conditions used for the for-
mation of 10a (Scheme 4). Mulliken populations of atoms de-
termined by DFT calculations (B3LYP 6-31þG**) reveal the electron
density of the cyano group is not so decreased even though a nitro
group is introduced at the
the above consideration.
a
-position (Fig. 3), which well supports
1,4-Dihydropyridines containing an amino group and a nitro
group can also be used in the synthesis of insecticides11 and central
nervous system potassium channel modulators;12 the dihydropyr-
idine framework is constructed through a single procedure by using
nitroketene aminals (1,1-diamino-2-nitroethenes) or nitroketene
dithioacetals as building blocks.13 However, a serious drawback of
this method is difficulty in the modification of the amino moiety. In
sharp contrast, in our method, the amino group can be easily
modified by changing the amine 2. Hence the present reaction will
supplement with the conventional reactions.
_
+0.397
+0.626
+0.388
+0.218
O
O
N
N
C
C
-0.454
+0.023
+0.452
NO2
NO2
9
8
O
N
C
+0.047
2.2. A plausible mechanism
-0.046
12
A plausible mechanism for the present reaction is illustrated in
Scheme 3. When the salt 9 is heated, the amine parts from 9 under
equilibrium giving the intimate pair 11. The liberated amine im-
mediately attacks the very close cyano group, and then the cyano
group attacks the acyl group to form a six-membered ring. Sub-
sequent dehydration and proton transfer lead to the formation of
the dihydropyridine 10.
Fig. 3. Mulliken population (au) determined by DFT calculations.
In the case of the anionic keto nitrile 9, the cyano group looses
electrophilicity, which does not react with amine anymore. When
the amine is liberated under equilibrium, the cyano group retrieves
its electrophilicity, and the liberated amine attacks the cyano group
O
_
_
O
N+
H
O
O
O
O
_
N+
_
N+
H
O
O
C
N
O
C
N
C
N
NH2R1
H
7
R1NH2
2
+
R1
11
9
H2N
O
_
N+
O
H
HO
10
N
N
R1
Scheme 3. A plausible mechanism for the formation of dihydropyridine 10.
Since there are three carbons between the two functional
groups (the cyano and the keto groups) in 7, the nitro group should
be unable to activate the carbonyl group by an electron-
withdrawing inductive effect. The main role of the nitro group is
efficiently because of spatial proximity, although the electrophi-
licity is lower than that of carbonyl group.
These calculated results strongly indicate that the carbonyl
functionality of the keto nitrile is available for the pseudo-
intramolecular process when the dinucleophilic diamine 14 is
employed instead of the monoamine 2. Namely, when one of the
amino groups of 14 forms the salt 16 at the nitroacetonitrile moiety
of 7, the other functionalities, a nucleophilic amino group and an
electrophilic keto group, are remained with spatial proximity,
which will react easily with each other to cause the pseudo-
intramolecular imination leading to a new framework via the
iminoammonium salt 17 (Scheme 5). On the basis of this supposi-
tion, the reactions of the keto nitrile 7 with the dinucleophilic di-
amines 14 were studied.
to make the a-hydrogen acidic so that the abstraction of the hy-
drogen by the amine 2 takes place to form the ammonium salt 9.
Thus, the acidic hydrogen is essential for triggering the pseudo-
intramolecular process, as shown in Scheme 4. Indeed, the simple
O
O
PrNH2
N
2a
Pr
N
H
N
H
MeCN
reflux 40 h
12
13a
NH2
2.3. Tandem cyclization including a pseudo-intramolecular
imination
H2N
N
NH2
N
N
14a
MeCN
rt, 1 h
When the 1,2-diaminoethane (14a) was added to a solution of
the keto nitrile 7 in acetonitrile, a white precipitate was immedi-
ately formed. The results of an elemental analysis and HRMS
12
15a
Scheme 4. Reactions of the keto nitrile 12 with propylamine (2a) and the diamine 14a.

N. Nishiwaki et al. / Tetrahedron 70 (2014) 402e408
405
_
_
O
O
O
O
_
N+
_
N+
H
_
O
N+
O
O
H2N H
C
N
C
N
H
NH2
C
N
O
7
+
N
NH2
14a
17a
H2N
H2N
16a
Scheme 5. The pseudo-intramolecular imination of the keto nitrile 7 with the diamine 14a.
revealed that the molecular formula of the precipitated compound
was C10H18N4O2, which corresponds to a dehydrated form of an
adduct obtained from 7 and 14a. The IR spectrum of the precipitate
showed a strong absorption at 2190 cmꢀ1 and a relatively weak
absorption at 1672 cmꢀ1; these two absorptions could be assigned
N3
C1
O1
C7
N4
to a cyano group in an a
-cyanonitronate structure8 and an imino
N1
O2
C2
group, respectively. On the basis of these spectral and analytical
data, the product was confirmed to be the zwitterionic imine 17a,
rather than the ammonium salt 16a (Scheme 5). It is noteworthy
that the imination proceeded quantitatively at room temperature
to afford 17a even in the absence of a catalyst. In the case of the keto
nitrile 12, the corresponding imine 15a was not formed upon
treatment with the diamine 14a under similar conditions (Scheme
4), which revealed the imination of the keto nitrile 7 proceeded in
a pseudo-intramolecular manner.
C6
C8
C5
C4
C3
N2
C10
Upon refluxing a suspension in acetonitrile, the imine 17a could
be easily transformed to the diazabicyclononane derivative 18a
(yield: 76%; Table 2, run 1).7 The structure of 18a was determined by
X-ray crystallography (Fig. 4) as well as by spectral analyses. In-
terestingly, the bridgehead nitrogen and the three adjacent carbons
C9
ꢁ
ꢀ
Fig. 4. An ORTEP drawing of 18a bond lengths [A] and angles [ ]: N1eC1 1.320(3),
N1eC5 1.482(3), N1eC7 1.468(3), C1eC2 1.457(3), C1eN3 1.331(3), N4eO1 2.567(3),
C2eN4 1.351(3); C1eN1eC5 123.3(2), C1eN1eC7 125.0(2), C5eN1eC7 111.5(2),
N1eC1eN3 119.2(2), N3eC1eC2 122.7(2), N1eC1eC2 118.2(2).
ꢀ
in 18a were located in a plane with a derivation of only 0.025 A; this
was thought to be owing to the conjugation with the
tronic acid moiety.
a-iminoni-
The formation of a bicyclic framework is considered to proceed
as illustrated in Scheme 6. When the imine 18a is heated, a small
amount of the imino amine 19a is formed under equilibrium, and
the nucleophilicity of the amino group and the electrophilicity of
the cyano group are reverted. As a result, two rings are simulta-
neously constructed by the nucleophilic attack of the amino group
to the imino carbon and the subsequent attack of the imino ni-
trogen to the cyano group to give the resulting product 18a with
a bicyclic structure.
O
H
N
_
N+
H
_
O
NH2
17a
N
O
C
N
N+
O
N
N
H
18a
H
19a
Scheme 6. Tandem cyclization leading to the bicyclic product 18a.
20 having the same bicyclic structure is rather low (run 2), when
a substrate with no substituent at the -position of the -keto ni-
trile skeleton (i.e., the compound 8) was used. This result would
We next tried to use other diamines 14. Except for run 2, the
reactions proceeded smoothly to give the corresponding bicyclic
compounds 18, as was expected (Table 2). The yield of the product
b
d
Table 2
Synthesis of bicyclic compounds 18 using the diamines 14
R2
N
R1
R1
R1
O
R1
1) rt, 1 h
MeCN
R3
_
NO2
H
R2
N
O
R3
NH2
+
N
H
N+
O
n
n
2) 80 °C, 2 h
C
N
N
H
H
14
7 (R1 = Me)
8 (R1 = H)
18 (R1 = Me)
20 (R1 = H)
Run
R1
R2
R3
n
Diamine
Product
Yield/%
1
2
3
4
5
6
Me
H
Me
Me
Me
Me
H
H
H
Et
H
H
H
H
Me
H
H
H
1
1
1
1
2
3
14a
14a
14b
14c
14d
14e
18a
20
76
26
a
18b and 18b0
65a
79
18c
18d
18e
85
23
a
A mixture of two regioisomers 18b (8-methyl isomer) and 18b0 (9-methyl isomer) was formed in a 1:1 ratio.

406
N. Nishiwaki et al. / Tetrahedron 70 (2014) 402e408
arise from the instability of 8 at the reaction temperature; 8
decomposed within a few days even when stored in a refrigerator.
When 1,2-diaminopropane (14b) was employed, an equimolar
mixture of the regioisomers 18b (8-methyl isomer) and 18b0 (9-
methyl isomer) was formed, indicating that the initial imination
unfortunately could not be controlled by the presence of the methyl
group on the ethylene chain (run 3).
The reactivity of the diamine 14c, which has an N-ethyl group,
was similar to that of the unsubstituted diamine 14a; the N-ethyl
group in 14c did not interfere the reaction, and the bicyclic product
18c was obtained (run 4). Worth to note is that the present reaction
completely recognized the difference in reactivity between the
primary amino group and the secondary amino group even though
they have similar basicity. It was also possible to construct rela-
tively larger condensed rings by employing the diamines 14d and
14e. When using 1,3-diaminopropane (14d), the bicyclization pro-
ceeded in the same way to afford the diazabicyclo[4.4.0]decane 18d
in 85% yield (run 5). The use of 1,4-diaminobutane (18e) resulted in
the formation of a seven-membered ring (run 6).
(10 mL), propylamine (2a) (41 mL, 0.5 mmol) was added. After stirring
at room temperature for 10 min, the solvent was evaporated to afford
the propylammonium salt 9a (122 mg, 0.5 mmol, quant.).
Pale brown oil. IR (neat/cmꢀ1) 3200e2900 (br), 2198 (strong),
1711, 1341, 1228, 733; 1H NMR (400 MHz, CDCl3)
d
0.99 (t, J¼7.4 Hz,
3H), 1.26 (s, 6H), 1.71 (dt, J¼7.5, 7.4 Hz, 2H), 2.10 (s, 3H), 2.97 (s, 2H),
3.00 (t, J¼7.5 Hz, 2H), 7.2e8.2 (br, 4H); 13C NMR (100 MHz, CDCl3)
d 9.0 (CH3), 19.3 (CH2), 24.2 (CH3), 29.0 (CH3), 32.3 (CH2), 39.6 (CH2),
48.6 (CH2), 103.6 (C), 116.1(C), 206.0 (C).
4.1.2. 1,4-Dihydro-4,4,6-trimethyl-3-nitro-2-(1-propylamino)pyri-
dine (10a). To a solution of keto nitrile 7 (92 mg, 0.5 mmol) in
acetonitrile (10 mL), propylamine (2a) (41 mL, 0.5 mmol) was added,
and the solution was heated under reflux for 40 h. After removal of
the solvent, the residue was purified by column chromatography on
silica gel to afford the dihydropyridine 10a (eluted with ethyl ace-
tate, 80 mg, 0.35 mmol, 71%). Pale yellow needles (recrystallized
from ethyl acetate). Mp 202e204 ꢁC. IR (Nujol/cmꢀ1) 1720, 1626,
1535, 1333; 1H NMR (400 MHz, CDCl3)
d
1.01 (t, J¼7.2 Hz, 3H), 1.48
Although several synthetic methods15 for diazabicyclic com-
pounds have been established with the aim of studying the bi-
ological activities16 and of producing agrochemicals,17 these
methods are not effective for synthesizing multiply functionalized
systems because of the following drawbacks: (1) these methods
involve multistep reactions and (2) the starting materials are not
readily available. To the best of our knowledge, there is only one
report on the synthesis of diazabicyclononanes possessing two
functional groups at the 2- and 3-positions.17 In contrast, the
present method is a valuable synthetic method that affords the
vicinally functionalized diazabicyclic compounds 18 and 20 via
simple manipulations without no special reagents nor reaction
conditions.
(s, 6H), 1.72 (tq, J¼7.2, 7.2 Hz, 2H), 1.83 (s, 3H), 3.36 (dt, J¼7.2, 5.2 Hz,
2H), 4.47 (s, 1H), 7.51 (s, 1H), 12.14 (t, J¼5.2 Hz, 1H); 13C NMR
(100 MHz, CDCl3)
d 11.4 (CH3), 18.4 (CH3), 22.0 (CH2), 27.0 (CH3),
36.0 (C), 43.5 (CH2), 113.7 (C), 114.4 (CH), 124.7 (C), 152.5 (C); MS
(FAB) 226 (Mþþ1, 100). Anal. Calcd for C11H19N3O2: C, 58.64; H,
8.50; N, 18.65. Found: C, 58.49; H, 8.51; N, 18.59.
The other dihydropyridines were synthesized in a similar
manner.
4.1.3. 2-(1-Butylamino)-1,4-dihydro-4,4,6-trimethyl-3-nitro-pyri-
dine (10b). Yield (78 mg, 0.33 mmol, 65%). Yellow plates (recrys-
tallized from acetonitrile). Mp 192e193 ꢁC. IR (Nujol/cmꢀ1) 1720,
1626, 1531, 1331; 1H NMR (400 MHz, CDCl3)
d
0.94 (t, J¼7.3 Hz, 3H),
1.43 (tq, J¼7.3, 7.3 Hz, 2H), 1.48 (s, 6H), 1.67 (tt, J¼7.3, 7.3 Hz, 2H),
3. Conclusions
1.83 (s, 3H), 3.42 (dt, J¼7.3, 5.1 Hz, 2H), 4.45 (s, 1H), 7.85 (s, 1H),
12.12 (t, J¼5.1 Hz, 1H); 13C NMR (100 MHz, CDCl3)
d 13.6 (CH3), 18.4
We successfully synthesized vicinally functionalized azaheter-
ocyclic compounds, such as the 1,4-dihydropyridines 10 and the
diazabicyclic compounds 18 and 20 upon treatment of the keto
nitriles 7 with the amines 2 and the diamines 14 under mild con-
ditions, in which the pseudo-intramolecular process is a key step.
Molecular design of the substrates with an acidic hydrogen and
a functionality, is not difficult. Hence, the pseudo-intramolecular
reaction is expected to emerge as a powerful tool for the synthe-
sis of polyfunctionalized compounds that cannot be prepared by
conventional methods.
(CH3), 20.1 (CH2), 27.0 (CH3), 30.6 (CH2), 36.0 (C), 41.5 (CH2), 113.6
(C), 114.4 (CH), 124.7 (C), 152.4 (C). Anal. Calcd for C12H21N3O2: C,
60.23; H, 8.84; N, 17.56. Found: C, 60.27; H, 8.91; N, 17.43.
4.1.4. 1,4-Dihydro-4,4,6-trimethyl-2-(2-methyl-1-propyl)amino-3-
nitropyridine (10g). Yield (75 mg, 0.32 mmol, 63%). Yellow needꢀle1s
(recrystallized from acetonitrile). Mp 201e205 ꢁC. IR (Nujol/cm
)
1724, 1620, 1537, 1313; 1H NMR (400 MHz, CDCl3)
d 1.04 (d,
J¼6.6 Hz, 6H), 1.49 (s, 6H), 1.83 (s, 3H), 2.0e2.1 (m, 1H), 3.13 (dd,
J¼6.6, 5.1 Hz, 2H), 4.50 (s,1H), 7.27 (s,1H),12.19 (t, J¼5.1 Hz,1H); 13
C
NMR (100 MHz, CDCl3)
d 18.4 (CH3), 20.2 (CH3), 27.0 (CH3), 27.8
4. Experimental
4.1. General
(CH), 36.0 (C), 49.2 (CH2), 113.6 (C), 114.5 (CH), 124.5 (C), 152.5 (C).
Anal. Calcd for C12H21N3O2: C, 60.23; H, 8.84; N, 17.56. Found: C,
59.86; H, 8.95; N, 17.29.
The melting points were determined on a Yanaco micro-melt-
ing-points apparatus, and were uncorrected. All the reagents and
solvents were commercially available and used as received. The 1H
NMR spectra were measured on a Bruker DPX-400 at 400 MHz with
TMS as an internal standard. The 13C NMR spectra were measured
on a Bruker DPX-400 at 100 MHz, and the assignments of 13C NMR
spectra were performed by DEPT experiments. The IR spectra were
recorded on a Horiba FT-200 IR spectrometer. The mass spectra
were recorded on a JEOL JMS-AX505HA mass spectrometer. The
high resolution mass spectra were measured on a JEOL JMS-
DX303HF. The elemental microanalyses were performed using
a Yanaco MT-3 CHN corder.
4.1.5. 1,4-Dihydro-2-(2,2-dimethyl-1-propyl)amino-3-nitro-4,4,6-
trimethylpyridine (10h). Yield (61 mg, 0.24 mmol, 48%). Pale yel-
low needles (recrystallized from acetonitrile). Mp 219e222 ꢁC. IR
(Nujol/cmꢀ1) 1722, 1620, 1543, 1313; 1H NMR (400 MHz, CDCl3)
d
1.07 (s, 9H), 1.50 (s, 6H), 1.83 (s, 3H), 3.08 (d, J¼7.0 Hz, 2H), 4.52
(s, 1H), 5.73 (s, 1H), 12.37 (t, J¼7.0 Hz, 1H); 13C NMR (100 MHz,
DMSO-d6)
d 16.4 (CH3), 25.0 (CH3), 25.3 (CH3), 29.4 (C), 33.5 (C),
51.0 (CH2), 110.4 (C), 111.5 (CH), 123.0 (C), 150.3 (C). Anal. Calcd for
13H23N3O2: C, 61.63; H, 9.15; N, 16.59. Found: C, 61.40; H, 9.29; N,
C
16.51.
4.1.6. 2-(Cyclohexylmethyl)amino-1,4-dihydro-4,4,6-trimethyl-3-
nitropyridine (10i). Yield (109 mg, 0.40 mmol, 80%). Yellow nee-
dles (recrystallized from acetonitrile). Mp 202e203 ꢁC. IR (Nujol/
4.1.1. Conversion of 3,3-dimethyl-2-nitro-5-oxohexanenitrile (7) to
propylammonium 3,3-dimethyl-5-oxohexanenitrile-2-nitronate
9a. To a solution of keto nitrile 7 (92 mg, 0.5 mmol) in acetonitrile
cmꢀ1) 1720, 1626, 1537, 1313; 1H NMR (400 MHz, CDCl3)
d
1.40 (s,
6H), 1.76 (s, 3H), 4.53 (s, 1H), 4.61 (d, J¼5.4 Hz, 2H), 7.32e7.44 (m,

N. Nishiwaki et al. / Tetrahedron 70 (2014) 402e408
407
5H), 8.33 (s, 1H), 12.15 (t, J¼5.4 Hz, 1H); 13C NMR (100 MHz, CDCl3)
1.38 (s, 3H), 1.48 (s, 3H), 1.58 (s, 3H), 1.86 (d, J¼13.5 Hz, 1H), 1.96 (d,
J¼13.5 Hz, 1H), 3.10 (10.8, 9.4 Hz, 1H), 3.3e3.4 (m, 1H), 3.83 (dd,
d
18.5 (CH3), 27.2 (CH3), 36.1 (C), 45.9 (CH2), 113.4 (C), 115.1(CH),
123.4 (C), 127.1 (CH), 127.5 (CH), 128.5 (CH), 128.8 (CH), 129.4 (CH),
J¼10.8, 6.0 Hz, 1H); 13C NMR (CD3OD)
d 16.4 (CH3), 23.7 (CH3), 25.7
135.1 (C), 152.4 (C); MS (FAB) 274 (Mþþ1, 100). Anal. Calcd for
(CH3), 28.1 (CH3), 33.9 (C), 50.4 (CH), 50.9 (CH2), 53.3 (CH2), 77.2 (C),
154.8 (C), 188.6 (C); MS (EI) 240 (Mþ, 21), 225 (100), 208 (40), 179
(47), 164 (95), 99 (86). HRMS (EI, magnetic field) Calcd for
C
15H19N3O2: C, 65.91; H, 7.01; N, 15.37. Found: C, 66.13; H, 7.07; N,
15.43.
C11H20N4O2: 240.1586. Found: 240.1582. A correlation between
4.1.7. 2-Benzylamino-1,4-dihydro-4,4,6-trimethyl-3-nitropyridine
(10j). Yield (109 mg, 0.40 mmol, 80%). Yellow needles (recrystal-
lized from acetonitrile). Mp 202e203 ꢁC. IR (Nujol/cmꢀ1) 1720,
a proton of the methyl group at the 5-position and a methyne
proton at the 7-position was observed in the 1He1H NOESY 2D
spectrum.
1626, 1537, 1313; 1H NMR (400 MHz, CDCl3)
d
1.40 (s, 6H), 1.76 (s,
18b0 (measured using a mixture with 18b) 1H NMR (CD3OD)
3H), 4.53 (s, 1H), 4.61 (d, J¼5.4 Hz, 2H), 7.32e7.44 (m, 5H), 8.33 (s,
d
1.34 (d, J¼6.1 Hz, 3H), 1.41 (s, 3H), 1.44 (s, 3H), 1.59 (s, 3H), 1.85 (d,
1H), 12.15 (t, J¼5.4 Hz, 1H); 13C NMR (100 MHz, CDCl3)
d
18.5
J¼13.4 Hz, 1H), 2.12 (d, J¼13.4 Hz, 1H), 2.99 (dd, J¼10.4, 8.5 Hz, 1H),
3.73 (dd, J¼10.4, 7.1 Hz, 1H), 3.7e3.8 (m, 1H).
(CH3), 27.2 (CH3), 36.1 (C), 45.9 (CH2), 113.4 (C), 115.1(CH), 123.4
(C), 127.1 (CH), 127.5 (CH), 128.5 (CH), 128.8 (CH), 129.4 (CH), 135.1
(C), 152.4 (C); MS (FAB) 274 (Mþþ1, 100). Anal. Calcd for
4.1.11. 1,7-Diaza-1-ethyl-2-imino-4,4,6-trimethyl-3-aci-nitrobicyclo
[4.3.0]nonane (18c). Yield (201 mg, 0.79 mmol, 79%). Colorless
prisms (recrystallized from a mixed solvent of acetonitrile and
methanol (1:1, v/v)). Mp 201e204 ꢁC (dec.). IR (KBr) 3250, 1605,
C
15H19N3O2: C, 65.91; H, 7.01; N, 15.37. Found: C, 66.13; H, 7.07; N,
15.43.
4.1.8. 1,4-Dihydro-2-[(2,2-dimethoxy)ethyl]amino-4,4,6-trimethyl-3-
1564, 1389, 1338 cmꢀ1
;
1H NMR (DMSO-d6)
d
1.05 (t, J¼7.1 Hz,
nitropyridine (10k). Yield (100 mg, 0.37 mmol, 74%). Yellow needꢀle1s
3H), 1.09 (s, 3H), 1.31 (s, 3H), 1.44 (s, 3H), 1.51 (d, J¼13.2 Hz, 1H),
2.15 (d, J¼13.2 Hz, 1H), 2.20e2.25 (m, 1H), 2.6e2.7 (m, 2H),
3.3e3.45 (m, 2H), 3.5e3.6 (m, 1H), 7.3e7.7 (br, 1H), 10.4e10.8
(recrystallized from ethyl acetate). Mp 171e174 ꢁC. IR (Nujol/cm
)
1724, 1622, 1531, 1311; 1H NMR (400 MHz, CDCl3)
d 1.40 (s, 6H), 1.76
(s, 3H), 3.45 (dd, J¼5.8, 4.8 Hz, 2H), 3.51 (s, 6H), 4.45 (t, J¼4.8 Hz,
(br, 1H); 13C NMR (DMSO-d6)
d 13.7 (CH3), 16.8 (CH3), 26.4
1H), 4.50 (s, 1H), 7.38 (s, 1H), 11.89 (t, J¼5.8 Hz, 1H); 13C NMR
(CH3), 29.0 (CH3), 32.6 (C), 41.4 (CH2), 44.0 (CH2), 45.5 (CH2),
50.9 (CH2), 75.1 (C), 112.3 (C), 153.1 (C); MS (EI) 254 (Mþ, 20),
239 (100), 222 (36), 193 (36), 178 (91), 113 (63). HRMS (EI,
magnetic field) Calcd for
254.1744.
(100 MHz, CDCl3)
d 18.6 (CH3), 27.2 (CH3), 36.1 (C), 45.5 (CH2), 55.6
(CH3), 104.4 (CH), 113.6 (C), 115.0 (CH), 123.8 (C), 154.0 (C). Anal.
Calcd for C12H21N3O4: C, 53.12; H, 7.80; N, 15.49. Found: C, 52.94; H,
7.92; N, 15.33.
C12H22N4O2: 254.1743. Found:
4.1.9. 1,7-Diaza-2-imino-4,4,6-trimethyl-3-aci-nitrobicyclo[4.3.0]
nonane (18a). To a solution of the keto nitrile 7 (184 mg,
1.0 mmol), in acetonitrile (15 mL), 1,2-diaminoethane 14a (67 mL,
4.1.12. 1,7-Diaza-2-imino-4,4,6-trimethyl-3-aci-nitrobicyclo[4.4.0]
decane (18d). Yield (204 mg, 0.85 mmol, 85%). Pale yellow plates
(eluted with chloroform). Mp 204e205 ꢁC (dec.). IR (KBr) 3259,
1.0 mmol) was added; the imine 17a was immediately pre-
cipitated as a pale yellow solid. The resultant mixture was heated
under reflux for 2 h and concentrated under reduced pressure.
The brown residual oil was purified by column chromatography
on silica gel to afford the diazabicyclononane 18a (172 mg,
0.76 mmol, 76%, eluted with AcOEt/methanol (70:30, v/v)) as
a colorless solid. Single colorless crystals for X-ray analysis were
obtained by recrystallization from a mixed solvent of methanol
and acetonitrile (1:2, v/v). Mp 241e243 ꢁC (dec.). IR (KBr) 3271,
1606, 1537, 1338 cmꢀ1 1H NMR (DMSO-d6)
; d 1.31 (s, 3H), 1.34 (s,
3H), 1.44 (s, 3H), 1.55e1.70 (m, 2H), 1.78 (d, J¼13.8 Hz, 1H), 1.83 (d,
J¼13.8 Hz, 1H), 2.44 (br, 1H), 2.70 (br d, J¼12.8 Hz, 1H), 2.91 (br dd,
J¼11.8, 11.8 Hz, 1H), 3.09 (ddd, J¼12.8, 12.0, 3.6 Hz, 1H), 3.71 (br d,
J¼12.0 Hz, 1H), 7.9e8.3 (s, 1H), 11.4e11.9 (br, 1H); 13C NMR (DMSO-
d6)
d 23.2 (CH3), 27.5 (CH2), 28.9 (CH3), 31.3 (CH3), 34.3 (C), 38.6
(CH2), 41.8 (CH2), 54.9 (CH2), 70.5 (C), 116.1 (C), 158.3 (C); MS (EI)
240 (Mþ, 3), 225 (88), 194 (63), 179 (93), 164 (100). HRMS (EI,
magnetic field) calcd for C11H20N4O2: 240.1586. Found: 240.1585.
3093, 1641, 1566, 1392, 1371, 1348 cmꢀ1 1H NMR (DMSO-d6)
;
d
1.22 (s, 3H), 1.32 (s, 3H), 1.47 (s, 3H), 1.69 (d, J¼13.3 Hz, 1H), 2.04
4.1.13. 1,7-Diaza-2-imino-4,4,6-trimethyl-3-aci-nitrobicyclo[5.4.0]
undecane (18e). Yield (58 mg, 0.23 mmol, 23%). Orange granules
(recrystallized from methanol). Mp 173e178 ꢁC (dec.). IR (KBr)
(d, J¼13.3 Hz, 1H), 2.9e3.0 (br, 1H), 3.2e3.3 (m, 3H), 3.4e3.5 (m,
1H), 7.3e7.6 (br, 1H), 10.4e10.7 (br, 1H); 13C NMR (DMSO-d6)
d
23.9 (CH3), 26.3 (CH3), 29.0 (CH3), 33.2 (C), 42.2 (CH2), 46.6
3275, 2930, 1611, 1560, 1544, 1335 cmꢀ1; 1H NMR (DMSO-d6)
d 1.31
(CH2), 50.5 (CH2), 75.7 (C), 112.3 (C), 152.7 (C); MS (FAB) 227
(Mþþ1, 100). Anal. Calcd for C10H18N4O2: C, 53.08; H, 8.02; N,
24.76%. Found: C, 53.18; H, 8.33; N, 24.69%. CCDC 804331 con-
tains the supplementary crystallographic data for this compound.
These data can be obtained free of charge from the Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_-
request/cif.
By using the other diamines 14bee and the keto nitrile 8, the
diazabicyclo compounds 18bee and 20 were synthesized in a sim-
ilar way.
(s, 3H), 1.33 (s, 3H), 1.35e1.40 (m, 2H), 1.39 (s, 3H), 1.50e1.55 (m,
1H), 1.56 (d, J¼14.3 Hz, 1H), 1.66e1.69 (m, 1H), 1.90 (d, J¼14.3 Hz,
1H), 2.49e2.53 (m, 1H), 2.6e2.7 (br, 1H), 2.75e2.80 (m, 1H),
3.45e3.50 (m, 2H), 7.4e7.8 (s, 1H), 11.4e11.9 (br, 1H); 13C NMR
(DMSO-d6)
d 25.7 (CH2), 25.8 (CH3), 26.2 (CH3), 28.8 (CH2), 29.5
(CH3), 40.7 (CH2), 41.5 (CH2), 49.2 (CH2), 71.9 (C),112.8 (C),154.8 (C);
MS (EI) 254 (Mþ, 7), 239 (30), 208 (81), 178 (36), 110 (57), 70 (100).
HRMS (EI, magnetic field) calcd for C12H22N4O2: 254.1743. Found:
254.1743.
4.1.14. 1,7-Diaza-2-imino-6-methyl-3-aci-nitrobicyclo[4.3.0]nonane
(20). Yield (52 mg, 0.26 mmol, 26%). Colorless plates (recrystal-
lized from acetonitrile). Mp 234e235 ꢁC (dec.). IR (KBr) 3271, 3101,
4.1.10. 1,5-Diaza-2-imino-4,4,6,8-tetramethyl-3-aci-nitrobicyclo
[4.3.0]nonane (18b) and 1,5-diaza-2-imino-4,4,6,9-tetramethyl-3-
aci-nitrobicyclo[4.3.0]nonane (18b0). Yield of 18b and 18b0 (1:1)
(eluted with a mixed solvent of dichloromethane and methanol
(9:1, v/v), 156 mg, 0.65 mmol, 65%). Colorless prisms (recrystallized
from a mixed solvent of acetonitrile and methanol (1:2, v/v)). Mp
237e238 ꢁC (dec.). IR (KBr) 3258, 2978, 2398,1618,1557,1389,1356,
1624, 1570, 1406 cmꢀ1 1H NMR (CD3OD)
; d 1.27 (s, 3H), 1.63 (ddd,
J¼13.7, 12.8, 6.1 Hz, 1H), 2.19 (ddd, J¼12.8, 6.0, 1.8 Hz, 1H), 2.68
(ddd, J¼17.4, 13.7, 6.0 Hz, 1H), 3.05 (ddd, J¼17.4, 6.1, 1.9 Hz, 1H),
3.35e3.45 (m, 3H), 3.48e3.59 (m, 1H); 13C NMR (CD3OD)
d 22.3
(CH3), 23.3 (CH2), 32.6 (CH2), 44.1 (CH2), 46.8 (CH2), 79.8 (C), 108.0
1315, 1196, 1069 cmꢀ1
;
1H NMR (CD3OD)
d
1.30 (d, J¼6.5 Hz, 3H),
(C), 189.8 (C); MS (EI) 198 (Mþ, 40), 181 (29), 166 (54), 152 (82), 85

408
N. Nishiwaki et al. / Tetrahedron 70 (2014) 402e408
4. (a) Hirai, S.; Asahara, H.; Hirao, S.; Sawayama, J.; Sugimoto, R.; Saigo, K.;
Nishiwaki, N. RSC Adv. 2013, http://dx.doi.org/10.1039/C3RA45568H; (b) Na-
kaike, Y; Taba, N.; Itoh, S.; Tobe, Y.; Nishiwaki, N.; Ariga, M. Bull. Chem. Soc. Jpn.
2007, 80, 2413e2417.
(100). HRMS (EI, magnetic field) calcd for C8H14N4O2: 198.1117.
Found: 198.1113.
5. Ballini et al. reported that an acyl group of a-nitroketones is highly activated.
This is considered to proceed in the pseudo-intramolecular process: Ballini, R.;
Bosica, G.; Fiorini. Tetrahedron 2003, 59, 1143e1145.
4.1.15. 7-Ammonio-5-aza-1-cyano-2, 2, 4-trimethyl-4-
heptenenitronate (17a). To a solution of the keto nitrile 7 (184 mg,
1.0 mmol), in acetonitrile (15 mL), 1,2-diaminoethane (14a) (67 mL,
6. Nishiwaki, N.; Nogami, T.; Ariga, M. Heterocycles 2008, 75, 675e681.
7. Nishiwaki, N.; Hirao, S.; Sawayama, J.; Saigo, K.; Kobiro, K. Chem. Commun. 2011,
4938e4940.
8. Nishiwaki, N.; Takada, Y.; Inoue, Y.; Tohda, Y.; Ariga, M. J. Heterocycl. Chem. 1995,
32, 473e475.
9. Nishiwaki, N.; Kakutani, K.; Tamura, M.; Ariga, M. Chem. Lett. 2009, 38,
680e681.
10. Wandas, M.; Talik, Z. J. Mol. Struct. 2013, 1043, 15e27.
11. (a) Sun, C.-W.; Chen, Y.-X.; Liu, T.-Y. Acta Crystallogr. 2012, E68, o1146 and ref-
erences are cited in; (b) Shiokawa, K.; Tsuboi, S.; Sasaki, S.; Moriya, K.; Hattori,
Y.; Shibuya, K. Eur. Pat. Appl. EP 296,453, 1988.
1.0 mmol) was added; a pale yellow solid was immediately pre-
cipitated, which was collected by filtration to give the imine 17a
(226 mg, 1.0 mmol, quant.). Pale yellow prisms (recrystallized from
methanol). Mp 129e130 ꢁC (dec.). IR (KBr) 2189, 1655, 1464, 1234,
1061 cmꢀ1; MS (EI) 226 (Mþ, 28), 211 (95), 194 (94), 150 (97), 85
(100), 69 (99). HRMS calcd for C10H18N4O2: 226.1430. Found:
226.1427. Anal. Calcd for C10H18N4O2: C, 53.08; H, 8.02; N, 24.76%.
Found: C, 52.70; H, 8.18; N, 25.07%.
12. Urbahns, K.; Goldmann, S.; Heine, H.-G.; Junge, B.; Schohe-Loop, R.; Somme-
meyer, H.; Glaser, T.; Wittka, R.; de Vry, J. Ger. Offen. DE 4,430,638, 1996.
13. (a) Ram, V. J.; Agarwal, N.; Sharon, A.; Maulik, P. R. Tetrahedron Lett. 2002, 43,
Acknowledgements
€
€
307; (b) Troschutz, R.; Luckel, A.; Mertens, H. Arch. Pharmacol. 1993, 326,
335e339; (c) Troscheitz, R.; Lueckel, A. Arch. Pharmacol. 1993, 326, 199e202;
(d) Tokumitsu, T. Bull. Chem. Soc. Jpn. 1990, 63, 1921e1924.
This work was supported by Grants-in-Aid for Scientific Re-
search (No. 22550043) from Japan Society for the Promotion of
Science and by Kyoto-Advanced Nanotechnology Network from the
Ministry of Education, Culture, Sports, Science and Technology,
Japan.
14. Fischer, D. F.; Sarpong, R. J. Am. Chem. Soc. 2010, 132, 5926e5927.
15. (a) Pilling, A. W.; Boehmer, J.; Dixon, D. J. Angew. Chem., Int. Ed. 2007, 46,
5428e5430; (b) IJzendoorn, D. R.; Botman, P. N. M.; Blaauw, R. H. Org. Lett.
2006, 8, 239e242; (c) Sannigrahi, M.; Pinto, P.; Chan, T. M.; Shih, N.-Y.; Njoroge,
F. G. Tetrahedron Lett. 2006, 47 4877ꢀ4800; (d) Vink, M. K. S.; Schortinghuis, C.
€
A.; Mackova-Zabelinskaja, A.; Fechter, M.; Pochlauer, P.; Castelijns, A. M. C. F.;
van Maarseveen, J. H.; Hiemstra, H.; Griengl, H.; Schoemaker, H. E.; Rutjes, F. P.
J. T. Adv. Synth. Catal. 2003, 345, 483e487; (e) Katritzky, A. R.; He, H.-Y.; Wang, J.
J. Org. Chem. 2002, 67, 4951e4956; (f) Jones, R. C. F.; Snaith, J. S.; Anderson, M.
W.; Smallridge, M. J. Tetrahedron 1997, 53, 1111e1122; (g) Chastanet, J.; Roussi,
G. J. Org. Chem. 1988, 53, 3808e3812; (h) Takahashi, K.; Tachiki, A.; Ogura, K.;
Iida, H. Heterocycles 1986, 24, 2835e2840; (i) Alder, R. W.; Eastment, P.; Moss,
R. E.; Sessions, R. B.; Stringfellow, M. A. Tetrahedron Lett. 1982, 23, 4181e4184;
(j) Aberli, P.; Houlihan, W. J. J. Org. Chem. 1969, 34, 165e170; (k) Shapiro, S. L.;
Soloway, H.; Freedman, L. J. Org. Chem. 1961, 26, 818e820.
References and notes
1. Anastas, P. T.; Warner, J. C. Green Chemistry Theory and Practice; Oxford Uni-
versity: New York, NY, 1998.
2. For reviews: (a) Yoshizawa, M.; Klosterman, J. K.; Fujita, M. Angew. Chem., Int.
Ed. 2009, 48, 3418e3438; (b) Pluth, M. D.; Bergman, R. G.; Raymond, K. N. Acc.
Chem. Res. 2009, 42, 1650e1659; (c) Onel, L.; Buurma, N. J. Annu. Rep. Prog.
Chem., Sect. B: Org. Chem. 2009, 105, 363e379; (d) Liu, S.; Gibb, B. C. Chem.
Commun. 2008, 3709e3716; (e) Koblenz, T. S.; Wassenaar, J.; Reek, J. N. H. Chem.
Soc. Rev. 2008, 37, 247e262; (f) Dwars, T.; Paetzold, E.; Oehme, G. Angew. Chem.,
Int. Ed. 2005, 44, 7174e7199; (g) Yoshizawa, M.; Fujita, M. Pure Appl. Chem.
2005, 77, 1107e1112; (h) Hof, F.; Craig, S. L.; Nuckolls, C.; Rebek, J., Jr. Angew.
Chem., Int. Ed. 2002, 41, 1488e1508.
16. (a) Hering, K. W.; Karaveg, K.; Moremen, K. W.; Pearson, W. H. J. Org. Chem. 2005,
70, 9892e9904; (b) Ghavale, D. D.; Matin, M. M.; Sharma, T.; Sabharwal, S. G.
Bioorg. Med. Chem. 2004, 12, 4039e4044; (c) Kayakiri, H.; Takase, S.; Shibara, T.;
Okamoto, M.; Terano, H.; Hashimoto, M. J. Org. Chem. 1989, 54, 4015e4016.
17. Shao, X.; Xu, Z.; Zhao, X.; Xu, X.; Tao, L.; Li, Z.; Qian, X. J. Agric. Food Chem. 2010,
58, 2690e2695.
3. Nishiwaki, N.; Nishida, D.; Ohnishi, T.; Hidaka, F.; Shimizu, S.; Tamura, M.; Hori,
K.; Tohda, Y.; Ariga, M. J. Org. Chem. 2003, 68, 8650e8656.