also reported that the addition of a strong base, such as Et
or DBU, resulted in a change in the distribution of products
to favor rearrangement to the methyl group.
3
N
was found that pyridine could be used as a suitable base in
this reaction at low temperatures (between -45 and -78
°C); however, hindered pyridine derivatives, such as 2,6-
lutidine, seem to work best. An unoptimized experiment
involved treatment of a solution of N-acetylbenzylamine (1.1
equiv) in CH
2
Cl
2
(0.2 M) with 2.2 equiv of 2,6-lutidine and
1
.1 equiv of Tf
2
O dropwise at 0 °C. After the mixture was
stirred for 30 min, 1.0 equiv of 4-phenyl-pyridine-1-oxide
was added all at once and the mixture was warmed to room
temperature. After aqueous workup and purification, N-ben-
zyl-N-(4-phenylpyridin-2-yl)acetamide was isolated in 46%
yield (Equation 2).
The use of 3-methylpyridine-1-oxide (Table 2, entry 4)
resulted in a 3:2 mixture of N-methyl-N-(3-methylpyridin-
2-yl)acetamide and N-methyl-N-(5-methylpyridin-2-yl)ac-
etamide in 64% combined yield. This result is in agreement
with the regioselectivity observed by Abramovitch. Quino-
lines and isoquinolines have also been shown to participate
in this transformation, giving high yields of the single isomers
shown (Table 2, entries 6 and 7).5
While the use of other carbonyl-containing functionalities
that would provide a more readily removable protecting
group would be desirable, to date, this has not been
accomplished. Substitution of the alkyl or aryl amide with
trifluoroacetyl, trichloroacetyl, or carbamates failed to pro-
duce significant quantities of iminochlorides or 2-aminopy-
ridine amides (data not shown).
In summary, we have described a significant operational
improvement in the generation of imidoyl chlorides from
secondary amides and their in situ reaction with pyridine-
1
-oxides to give 2-aminopyridine amides. The reaction
conditions have not been completely optimized in terms of
reagent equivalents partially due to the hydroscopic nature
of pyridine-1-oxides.
Acknowledgment. The authors thank Professor David A.
Evans (Harvard University) for helpful advice.
During our search for suitable replacements for phosgene,
we also discovered that Tf
2
O could be used as an activating
agent. In the absence of base or in the presence of Et N or
Hunig’s base, imidotriflate formation was not observed. It
Supporting Information Available: Experimental de-
tails. This material is available free of charge via the Internet
at http://pubs.acs.org.
6
3
OL0264556
(
4) Abramovitch, R. A.; Bailey, T. D. J. Heterocycl. Chem. 1975, 12,
079. Abramovitch, R. A.; Abramovitch, D. A. J. Chem. Soc., Chem.
Commun. 1979, 21, 956.
5) Abramovitch, R. A.; Rogers, R. B. Tetrahedron Lett. 1971, 22, 1951.
Abramovitch, R. A.; Rogers, R. B.; Singer, G. M. J. Org. Chem. 1975, 40,
1.
1
(6) For the formation of imino triflates and addition of nucleophiles,
see: Thomas, E. W. Synthesis 1993, 767. Charette, A. B.; Chua, P. J. Org.
Chem. 1998, 63, 908. Charette, A. B.; Grenon, M. Tetrahedron Lett. 2000,
41, 1677. Charette, A. B.; Grenon, M.; Lemire, A.; Pourashraf, M.; Martel,
J. J. Am. Chem. Soc. 2001, 123, 11829.
(
4
Org. Lett., Vol. 4, No. 18, 2002
3129