5722
M. A. Schmidt, X. Qian / Tetrahedron Letters 54 (2013) 5721–5726
Table 1
time to 19 h; additional amounts had no effect on reaction profile
Optimization of the reaction parameters
or rate. These conditions led to a 77% isolated yield of 1a (Table 1,
entry 8). A greater than stoichiometric amount of acetic acid was
necessary since 1 equiv was needed to neutralize the ammonia
byproduct of the reaction. For example, if ammonium acetate is
used instead of acetic acid, allowing the ammonia to build up,
the reaction proceeds slowly (67% conversion after 24 h). Since a
common method for the formation of imidates 5 is through a Pin-
ner reaction, which often yields imidate hydrochloride salts,10 we
investigated the use of ethyl benzimidate hydrochloride (5b)
(Table 1, entry 9). Since the presence of HCl is detrimental (entry
7) and 1.5 equiv of acetic acid is the optimum quantity (entry 8),
we added 1.0 equiv of sodium acetate to buffer the HCl in situ,
and added an additional 0.5 equiv of acetic acid. This resulted in
an identical reaction profile to entry 8, affording 1a in 78% yield.
The reaction of 2-hydrazinopyridine dihydrogen chloride (4b)
and 5b in the presence of 3.0 equiv of sodium acetate gave an iden-
tical reaction profile; but 1a was isolated in 90% yield (Table 1, en-
try 10). The ꢀ12% increase in yield was not attributed to the excess
3.0 equiv of acetic acid or the presence of 3.0 equiv of sodium chlo-
ride byproduct, as control experiments with 4a and 5a gave 1a in
77% and 78% isolated yield, respectively. We cautiously suggest
that the reagents are more stable when stored and used as salts,
therefore we chose to use the salt forms of the 2-pyridylhydrazine
whenever possible. We next applied these conditions to a range of
electronically and structurally diverse substrates to examine the
scope and functional group tolerance of the reaction.
H
N
NH2
NH
N
N
HN
HN
EtO
N
NH
N
N
4a
4b (2 HCl)
5a
5b (HCl)
6a
1a
Entry Solventa Additive (equiv)
Time
(h)
Yield (conversion)b
(%)
1
2
3
4
5
THF
AcOH (1.0)
AcOH (1.0)
AcOH (1.0)
AcOH (1.0)
None
BzOH (1.0)
HCl (1.0)c
AcOH (1.5)
NaOAc (1.0)/AcOH
(0.5)
24
24
24
24
24
24
24
19
19
(46)
(52)
(90)
PhMe
MeCN
EtOH
EtOH
EtOH
EtOH
EtOH
EtOH
(99)
(100 6a)
(99)
(84 6a)
77
6
7
8
9d
78
10e
EtOH
NaOAc (3.0)
19
90
a
b
c
Concentration 5 mL/g 5.
Isolated yields are an average of two runs, conversion measured by HPLC.
Reaction between 4a and 5b with no additional additives.
Reaction of 4a and 5b.
d
e
Reaction of 4a and 5b.
We found that both electron rich and electron deficient imi-
dates are acceptable substrates in the reaction. The reaction be-
tween 4b and the electron rich p-methoxybenzimidate
hydrochloride (5c) proceeded cleanly to afford 1b in 90% yield (Ta-
ble 2, entry 1). Likewise, the reaction of 4b and electron deficient
ethyl picolinimidate (5d) afforded the desired product 1c in 80%
(Table 2, entry 2). While the electronic nature of the imidate only
subtly affected the reaction rate, the electronic properties of the
hydrazinopyridine had a stronger effect on the reaction. The com-
paratively more electron rich 4-methyl substituted hydrazine 4c
reacted with 5b to afford the amidrazone rapidly, however it cy-
clized to 1d much slower than the parent reaction (Table 1, entry
8) at 50 °C presumably due to the greater basicity of the pyridine
ring, leading to more NPyridine protonation and hence deactivation.
At 70 °C however, the reaction was rapid, giving the product 1d in
87% yield in 4 h (Table 2, entry 3). The addition of an electron with-
drawing group on the 5-position of the hydrazinopyridine ring
greatly reduced the reaction rate with 5b. The reaction between
4d and 5b required heating to 70 °C for 24 h to form the product
1e in 87% yield (Table 2, entry 4).
The substitution patterns of the hydrazinopyridine ring had a
drastic effect on the reaction rate. Interestingly, if a functional
group is located at the 3-position, the reaction is significantly fas-
ter. The reaction of 3-bromo-2-hydrazinopyridine (4f) with 5b
afforded product 1f in 90% yield in 3.5 h (Table 3, entry 1, com-
pared to Table 2, entry 4). We attribute this rate enhancement to
a conformational bias that would not be available to the 5-bromo
containing amidrazone. We hypothesize that the 3-position substi-
tution in the pyridine ring will increase the population of produc-
tive conformers by restricting the rotation of the amidrazone
around the N–C2Pyridine ring bond (Fig. 2).11 This rate enhancement
could effectively compensate for substrates with otherwise dimin-
ished reactivity. Even the electron deficient hydrazinopyridine 4g
reacted at 50 °C to give the product 1g (Table 3, entry 2). An elec-
tron rich hydrazinopyridines with 3-substitution such as 4h re-
acted the fastest out of all the hydrazinopyridines studied
(compare Table 3, entry 3 with Table 2, entry 1). This enhance-
ment, coupled with the mildness of the conditions allowed for
the rapid formation of products with sensitive functional groups
H2N
N
HN
H
Cl
N
Figure 1. Amidrazone HCl salt 6aÁHCl.
limited examples suffered from low yields due to incomplete con-
versions, or utilized pyridine as a solvent or co-solvent and/or thi-
oimidates as reaction partners. In this study, we report our efforts
in expanding the scope of this reaction and developing a mild, ro-
bust, high yielding protocol for the formation of [1,2,4]triazolopyri-
dines 1.
The reaction between 2-hydrazinopyridine (4a) and ethyl ben-
zimidate (5a) with 1.0 equiv of acetic acid was examined as a mod-
el system in a variety of solvents at 50 °C (Table 1, entries 1–4).9 In
all the solvents examined, the corresponding amidrazone 6a was
formed cleanly in less than 30 min. The subsequent cyclization of
the amidrazone to product 1a, was clean and complete in ethanol
after 24 h, however it was incomplete in THF, MeCN and toluene. A
brief survey of acids was conducted (Table 1, entries 5–7). The con-
trol experiment confirmed that in the absence of acid, the amidraz-
one 6a was the sole product, and no further cyclization took place.
Benzoic acid was as effective as acetic acid, however stronger acids
such as HCl (introduced by using ethyl benzimidate hydrochloride
(5b)) immediately formed the amidrazone HCl salt (6aÁHCl), with-
out further cyclization to 1a due to substrate deactivation (Fig. 1).
Based on these results, we chose acetic acid for our study. Increas-
ing the amount of acetic acid to 1.5 equiv reduced the reaction