Gissot et al.
SCHEME 3. Nitrosated Species Are Needed To
Convert Primary Nitro Derivatives into Acids
sated by these nitrosating species (path C) as already
evidenced by Kornblum, thus overtaking path B.10 The
final ketone/oxime ratio obtained at the end of the
reaction largely depends on the propensity of the oxime
to react with such species (path D). While electronically
rich oximes are easily nitrosated to the corresponding
ketones, electronically poor ones are not and are obtained
as the sole product.
FIGURE 1. Nitrosation of 1a in the presence of N,N-
dimethylaniline. Conditions: NaNO2 (4 equiv), N,N-dimethy-
laniline (2 equiv), DMSO/H2O 1.4/0.2, 70 °C.
Cyclohexanone is indeed formed in that case. Conse-
quently, the nitro derivative is not only transformed into
ketone but is also responsible for the generation of
nitrosating species from NaNO2 under neutral, aqueous
conditions. Hence, the nitro derivative can be considered
as a kind of a catalyst for its own transformation as the
nitrosating species formed catalyze the transformation
(path C, Scheme 2). Yet, one could argue the nitro
derivative might simply be acidic enough to promote the
generation of nitrosating species from NaNO2 through
the simple acid-base reaction between those two species
(see the aci-nitro-aci-nitronate equilibrium I in Scheme
2). In that case, path C alone (the original Kornblum’s
mechanism) could account for the transformation of
secondary nitro derivatives into ketones. Based on this
hypothesis alone, no reaction should take place in the
absence of nitrosating species. Yet, the reaction does take
place in the presence of N,N-dimethylanilinesa potent
NO+ species scavengers21,22 even though excess NaNO2
is needed (Figure 1). Under these conditions, both
transformations involving nitrosated species (see red
arrows in Scheme 2) are shut down. The only way for
the substrate to undergo the transformation is through
paths A and B. Since the oxime and ketone are formed
under these conditions, this strongly supports the hy-
pothesis that NaNO2 can act as a nucleophile toward
the aci-nitro(nate). To the best of our knowledge, this is
the first time the nitrite ion has been shown to be able
to react as a nucleophile toward a nitroalkane. The
aptitude of nitrite ion to act as a nucleophile has been
precluded so far by the acidic conditions usually employed
to perform such transformations (the nitrite ion is
protonated at pH < 3.3). The 7:3 oxime/ketone ratio
obtained may in turn reflect the N/O-nitration addition
ratio of NaNO2 to the aci-nitro (paths B and A, respec-
tively).23 This experiment also suggests that nitrosat-
ing species are indeed good catalysts for this transforma-
tion, the reaction time being much longer in their
absence.24
Discussion
While under acidic conditions the protonation of NaNO2
is responsible for the formation of NO+-like species11,12
and in turn the reactivity observed with primary nitroal-
kanes,13 the anionic character of the nitrite remains
unaltered under our neutral conditions. Thus, it is likely
to act as a powerful nucleophile toward the aci-nitro, and
both the O- and N-nitrations of the latter can be
anticipated by virtue of the ambident character of the
nitrite ion. Several experimental observations support
this hypothesis and prove that nitrosating species are
actually formed in the course of the reaction under
neutral, aqueous conditions.
First, primary derivatives such as 2-phenylnitroethane
are not reactive and do not give the corresponding acid
unless acidic conditions are employed (Scheme 3).
Thus, if primary nitroalkanes do require the formation
of nitrosating species from NaNO2 and acetic acid to
afford the corresponding acid, secondary derivatives do
not. This result can be explained on the basis of the
nitro-aci-nitro equilibrium unfavorable to the latter with
primary derivatives;19 the amount of the aci-nitro form
in solution is then too low to observe the nucleophilic
attack of the nitrite (paths A and B). The same observa-
tion goes for the intermediately formed cyclohexanone
oxime that is not transformed into cyclohexanone under
our standard conditions unless acetic acid is added to the
solution (Scheme 4).20
This result is in agreement with our mechanism, as
the oxime-ketone transformation requires nitrosating
species previously formed from the intermediates 4 and
5 (see path D, Scheme 2). Therefore, those species must
somehow be formed from the nitro derivative itself. To
further evidence the ability of the nitro derivative to act
as the actual NO+ supply, we carried out the same test
reaction in the presence of an external secondary nitro
derivative, namely 2-nitropropane (Scheme 5).
(21) Liddell, H. F.; Saville, B. Chem. Ind. 1957, 493-494.
(22) Loeppky, R. N.; Singh, S. K.; Elomari, S.; Hastings, R.; Theiss,
T. E. J. Am. Chem. Soc. 1998, 120, 5193-5202.
(23) While the cyclohexanone 2a is formed quantitatively in the
absence of N,N-dimethylaniline with a strong nitrosating reagent such
as the nitrosyl tetrafluoroborate, it is not detected in the presence of
2 equiv of N,N-dimethylaniline. As a consequence, the cyclohexanone
formed in this test is likely to result from the O-nitration of the aci-
nitro (path A, Scheme 2).
(19) Bordwell, F. G.; Boyler, W. J., Jr.; Yee, K. C. J. Am. Chem. Soc.
1970, 92, 5926-5932.
(20) Since sodium nitrate was formed during the course of the
reaction, the reactions were also conducted in the presence of this salt.
9000 J. Org. Chem., Vol. 69, No. 26, 2004