K. Ghosh et al. / Tetrahedron Letters 48 (2007) 6308–6311
COCl, Et N
RBr, K CO , cat. KI
6309
COCl, Et N
3
3
2
3
O
1
dry CH Cl
dry acetone
NH2
dry CH Cl
2 2
2
2
N
N
NHR
N
N
R
2: R = n-butyl,
Scheme 1. Syntheses of compounds 1 and 2.
features. However, on storing in the absence of solvent
(26 °C). The low temperature favors the conversion pos-
sibly due to the insoluble nature of 1a that separates as a
solid from the starting material, which is gummy in nat-
ure. The reaction is reversible and the kinetically favored
product 1a is fully converted to the thermodynamically
stable amide 1 on heating in boiling xylene for 1–2 h.
Reformation of amide 1 in boiling xylene is presumably
ascribed to the azeotropic loss of water. The conversion
of 1a into 1 is also feasible in solution phase. In dry ethyl
acetate, 1a was completely converted to 1 at room tem-
perature after 2 h. In dry CHCl , CH OH, and THF the
same conversion was only 20%, 10%, and 50% complete,
respectively, after one day. Methanol, a polar H-bond
donor solvent, stabilizes 1a by solvating the carboxylate
ion as well as the amine group through H-bonding and
makes the conversion of 1 to 1a less facile. The free
amine salt 1a was very labile such that we could not
form an amide with butyryl chloride; instead we isolated
in a freezer for a couple of days, compound 1 produced
6
a white, chloroform insoluble solid 1a, which exhibited
1
a different H NMR spectrum (in DMSO-d ). Careful
6
1
analysis of the H NMR spectrum of the newly formed
solid revealed the disappearance of the vinylic protons
(
d 6.45, 6.25, and 5.77 ppm) with the appearance of
new signals centered at d 4.26 and 2.33 ppm. Also, FTIR
analysis showed significant changes in the carbonyl
À1
stretching frequencies (1676 cm
1
in 1 changed to
À1
À1
668 cm in 1a). The amide NH stretch at 3270 cm
3
3
À1
in 1, appeared as two peaks (3408 and 3290 cm ) in
the chloroform insoluble solid. This indicated the pres-
ence of a free amine functionality in the newly formed
solid. Mass analysis showed a peak at m/z 181.2
(
M+H) instead of m/z 162 for 1. A single crystal of this
solid was prepared from chloroform/methanol and anal-
8
ysis confirmed structure 1a.
1
as the sole product. The presence of an electron-donat-
ing methyl group in 1 promotes the amide cleavage pos-
sibly by increasing the donor capacity of the pyridine
ring nitrogen in the Michael type addition reaction in
Scheme 2. This was established by reacting compound
N
O
NH2
+
4
, which under similar conditions was stable, although
9
b
Szafran et al. had reported the nucleophilic reaction
of 2-aminopyridine itself with 3-bromopropionic acid
to afford 3-(2-aminopyridinium)propionate. Compound
-
O
1
a
5
was also stable under similar conditions. In addition,
this conversion was not observed with derivatives 2
and 3 suggesting the kinetic inertness of the 3° and sat-
urated 2° amides, respectively. Compound 1a crystal-
lized in the triclinic system with one water molecule in
the asymmetric unit and is interesting in the context of
the synthesis of new hydrogen bonding synthons in
supramolecular chemistry. During the rearrangement
of 1 as shown in Scheme 2, the hydrogen-bonding pat-
tern (AD; A = H-bond acceptor; D = H-bond donor)
was altered to an AADD pattern in 1a (Fig. 1).
This finding is interesting and to the best of our knowl-
edge, is unknown in the literature. The generation of 1a
can be explained by invoking the following suggested
mechanism where the pyridine amide proton on tauto-
8
meric shift presumably initiates the reaction. Michael
type addition of the pyridine nitrogen to the pendant
group at C-2 of pyridine gives the cyclic imine 6, which
on hydrolysis in the presence of atmospheric water, pro-
duces 1a. In this regard, cleavage of amide linkage fol-
lowed by intermolecular alkylation on the pyridyl ring
nitrogen to afford 1a cannot be excluded. Compound
This hydrogen bond alteration particularly, has a pro-
found role in the synthesis of supramolecules of great
1
, in absence of solvent, was found to undergo complete
conversion to 1a after 35 days in a freezer at 5 °C. The
same conversion was 10% complete on keeping for 35
days in the open atmosphere at room temperature
interest. Figure 2 shows the ORTEP plot of 1aÆH O
with atom numbering scheme. In the unit cell, two
charged pyridyl rings with carboxylate ions are aligned
2
H O
H O
2
2
1a
N
N
N
H
N
N
N
N
N
N
NH
OH
N
NH
+
+
+
OH
OH
O
OH
O
6
Scheme 2. Probable mechanism of the rearrangement.