that the acidic nature of the gel promoted loss of nitrogen.
The isolation of 1312 was supported by mass spectrometry.13
The fraction containing the aziridine 13 was treated with
hydrogen in the presence of 10% Pd-C in ethyl acetate as
solvent and this gave an inseparable mixture of DNJ
derivatives 14 and 15 (78%). The formation of 14 was
expected from this reduction reaction whereas the formation
of 15 is explained by hydrolysis of 13 in the presence of
adventitious water. These results support the proposal that
13 is formed from 12.14
methylene carbon atom of 20 and acid-catalyzed hydrolysis
of the acetal. It would appear that the acetal group is more
stable in the azepane 16 when compared with the piperidine
15. When this reaction was carried out at higher temperature
(70 °C) a complex mixture was obtained.
The one-pot conversion of 11 with a variety of nucleo-
philes was next investigated, and the results are summarized
in Table 1.
The possibility to carry out a one-pot conversion of 11 to
DNJ derivatives was next investigated.The decomposition
of the triazoline 12 and subsequent reactions of the resulting
aziridine 13 were expected to be promoted in acidic
conditions.The reaction of 11 in aqueous acetic acid over-
night at room temperature gave a mixture of 13 (15%),
azepane 16 (33%), and DNJ derivative 17 (14%) (Scheme
4). That azepane15 16 is obtained as the major product from
Table 1. One-Pot Conversion of 11 to DNJ Derivatives
entry
1
reagents and conditions
product (yield, %)
NaN3 (5.0 equiv), AcOH
(1.5 equiv), toluene, 110 °C
MeOH, 60 °C, 1 ha
21 (R ) N3, 35%)
Scheme 4
2
3
4
22 (R ) OMe, 20%),
13 (31%)
23 (R ) OAc, 45%)
AcOH (5.0 equiv), toluene,
110 °C, 1 h
PhSH, 15 ha
24 (R ) SPh, 57%)
a Toluene removed before addition of reagents.
Accordingly, 11 was first heated to reflux in toluene for
1 h, and then reagents (entries 1 and 3) were added and
heating continued. Reaction of 11 with sodium azide and
acetic acid in toluene by this protocol gave 21 (35%),
whereas reaction with acetic acid in toluene gave 23 (45%).
Interestingly, only the piperidines, and not azepane products,
were isolated under these conditions, contrasting with the
behavior shown in Scheme 4. This may be due to the use of
the nonpolar solvent which would favor an SN2 pathway and
reaction of the nucleophile at the least hindered carbon,
disfavoring the formation of 19. An alternative explanation
is that the reaction is under steric control and that C-N bond
breaking in 20 has not occurred to a great extent when the
transition state is reached under these conditions, favoring
nucleophilic attack at the least hindered carbon of the
aziridine 20.
The second protocol involved the initial formation of the
triazoline 12 in toluene, evaporation of the toluene and
subsequent addition of the nucleophile. Thus, conversion of
11 to 22 (20%) and 24 (57%) was achieved by this procedure
using methanol and thiophenol as nucleophiles, respectively.
For the reaction with methanol the aziridine 13 was also
isolated (31%).
this reaction where water is the nucleophile suggests that
carbocation 19 is generated under these conditions.16 An
alternative explanation is that the reaction is under electronic
control and C-N bond breaking in 20 is more complete when
the transition state (of its reaction with water) is reached
under these conditions, favoring azepane formation. The
formation of 17 is explained by the reaction of water at the
(11) For a review on applications of organic azides including alkene-azide
cycloadditions, see: Bra¨se, S.; Gil, C.; Knepper, K.; Zimmermann, V. Angew.
Chem., Int. Ed. 2005, 44, 5188–240.
(12) Kim, S.; Lee, Y. M.; Lee, J.; Lee, T.; Fu, Y.; Song, Y.; Cho, J.;
Kim, D. J. Org. Chem. 2007, 72, 4886–91.
(13) We could not obtain satisfactory NMR data as the aziridine 13
decomposed in standard deuterated solvents.
(14) Decomposition of 1,2,3-triazolines to aziridines and imines is
known. For a recent example, see: Kim, S.; Lee, Y. M.; Lee, J.; Lee, T.;
Fu, Y.; Song, Y.; Cho, J.; Kim, D. J. Org. Chem. 2007, 72, 4886–91.
(15) Le Merrer, Y.; Poitout, L.; Depezay, J.-C.; Dosbaa, I.; Geoffroy,
S.; Foglietti, M.-J. Bioorg, Med. Chem. 1997, 5, 519–33.
(16) For an example of Markovnikov addition of water to an aziridine
and proposal of a carbocation intermediate, see: Hanaoka, M.; Kim, S.;
Inoue, M.; Nagami, K.; Shimada, Y.; Yasuda, S. Chem. Pharm. Bull. 1985,
33, 1434–43.
In summary, the synthesis of a range of novel DNJ
derivatives has been achieved from D-glucono-δ-lactone. The
sequence has included a Huisgen cycloaddition reaction of
alkene and azide in a key step. The resulting triazoline can
be promoted to give an aziridine, which can be converted to
DNJ derivatives in the presence of nucleophiles and DNJ
Org. Lett., Vol. 10, No. 17, 2008
3779