Our laboratory recently became interested in the
N-oxide activation approach to selectively prepare 2-sub-
stituted pyridines. In a previous communication,7 we
demonstrated that the phosphonium salt PyBroP8 (bromo-
tris-pyrrolidino-phosphonium hexafluorophosphate) func-
tioned as a general and mild N-oxide activator for the
regioselective addition of amine nucleophiles. In that re-
port, we described significant reaction optimization which
resulted in an operationally simple amination procedure
devoid of common side reactions mentioned above. In
general, unhindered-aliphatic amines participated most
effectively in the transformation. There were, however, a
few exceptions to this paradigm (Table 1). Aminations
using heterocycles, such as imidazoles and pyrazoles, un-
expectedly proceeded in excellent yields (entries 1-3).
We selected a number of nucleophiles to test under our
standard protocol (Table 2). Presumably, these new sub-
strates would function in place of the amines, for which the
procedure was originally optimized. As a general guideline
for selection, we chose nucleophiles within a pKa range10 of
∼10-20, as these would approximate the acidity of the
heterocyclic amines in Table 1. When pyridine-N-oxide 6
was combined with each nucleophile (1) in dichloro-
methane and treated with iPr2EtN and PyBroP, we were
pleased to obtain a variety of 2-substituted pyridines in
modesttoexcellent yields. Innoneofthese instances did we
observe addition at the 4-position, a result consistent with
our aforementioned amination procedure. As shown in
Table 2, phenols (entries 1-5) were some of the most
effective nucleophiles. Both aromatic and aliphatic sulfo-
namides (entries 6-8) underwent smooth N-addition in
modest to excellent yields. A series of enolizable substrates
(entries 9-12) also proceeded in moderate yields. In these
cases, it was necessary to use a 3-fold excess of nucleophile
with respect to N-oxide 6 in order to mitigate overaddition
of the reaction product onto 6. We were pleased with the
reactivity of pyridones and pyrimidone (entries 13-16). In
these examples, we observed chemoselective reaction at the
nitrogen. For entry 15, we saw both a regioselective and
chemoselective reaction at the 3-nitrogen. The relatively
poor reactivity of the 1-nitrogen, as seen in entry 14, may
explain this selectivity in part. Aliphatic thiols (entries
17-18) were also capable nucleophiles and afforded the
desired 2-thiopyridines in good yields.
Table 1. Amination of Pyridine-N-Oxide with Heterocyclic
Aminesa
We further examined the reaction of select nucleophiles
with various pyridine- and quinoline-N-oxides under our
standard conditions (Table 3). With the exception of
strongly electron deficient N-oxides (entries 7-9), we were
pleased to observe good reactivity in all cases, regardless of
the nucleophile. For entries 7-9, a change in reaction
solvent to THF (vide infra), helped improve yields signifi-
cantly. Although the examples in Table 3 are derived from
commercially available N-oxides, the relative ease11 of
converting pyridines and quinolines to the corresponding
pyridine- and quinoline-N-oxides facilitates considerable
diversity in this methodology.
Analogous to amine nucleophiles, we propose that the
reaction proceeds via the activated phosphonium complex
(8) shown in Table 1. The strong regiochemical preference
for 2-position addition in all cases is most likely attributed
to a charge association12 of 8 with the incoming nucleo-
phile. Additionally, in certain examples (Table 3, entries
a Reaction Conditions: Combine N-oxide 6 (1.00 equiv), Amine 7
(1.25 equiv), iPr2EtN (3.75 equiv) in CH2Cl2 (0.25 M) and add PyBroP
(1.30 equiv) and stir for 15 h at room temperature.
These results were contrary to our assumption that only
nucleophilic amines would participate in the reaction.
Under the mild reaction conditions, these heterocycles,
although electron rich, are generally not considered
strong nucleophiles. We therefore rationalized that other
weakly nucleophilic substrates that share similar elec-
tronic and ionization properties to the heterocycles
screened in Table 1 would function as effective nucleo-
philes in our reaction. We were encouraged by reports9
in the literature on the direct addition of weak nucleophiles
to tautomerizable carbon-oxygen bonds, which had been
first activated by phosphonium salts.
(10) Equilibrium acidities discussed in this article are measured in
DMSO, as reported by Bordwell. For a review, see: Bordwell, F. G. Acc.
Chem. Res. 1988, 21, 456–463.
(11) (a) Jain, S. L.; Sain, B. Chem. Commun. 2002, 1040–1041. (b)
Fields, J. D.; Kropp, P. J. J. Org. Chem. 2000, 65, 5937–5941. (c) Caron,
S.; Do, M. N.; Sieser, J. E. Tetrahedron Lett. 2000, 41, 2299–2302. (d)
Ferrer, M.; Sanchez-Baeza, F.; Messeguer, A. Tetrahedron 1997, 53,
15877–15888.
(12) For a review on the regioselectivity of nucleophilic addition onto
activated pyridines, see: Poddubnyi, I. S. Chem. Heterocycl. Compd.
1995, 31, 682–714. Hard-hard/soft-soft interactions during the nucleo-
philic addition transition state are proposed to influence regioselectivity
in additions onto activated pyridines, and may work in concert with the
proposed charge interaction.
(7) Londregan, A. T.; Jennings, S.; Wei, L. Org. Lett. 2010, 12, 5254–
5257.
(8) Castro, B.; Coste, J. PCT Int. Appl. WO90/10009, 1990.
(9) (a) Kokatla, H. P.; Lakshman, M. K. Org. Lett. 2010, 12, 4478–
4481. (b) Kang, F. A.; Sui, Z.; Murray, W. V. Eur. J. Org. Chem. 2009, 4,
461–479. (c) Wan, Z.; Wacharasindhu, S.; Levins, C. G.; Lin, M.; Keiko
Tabei, K.; Mansour, T. S. J. Org. Chem. 2007, 72, 10194–10210. (d)
Kang, F. A.; Kodah, J.; Guan, Q.; Li, X.; Murray, W. V. J. Org. Chem.
2005, 70, 1957–1960.
Org. Lett., Vol. 13, No. 7, 2011
1841