4712
J . Org. Chem. 1997, 62, 4712-4720
Nitr osa tion of Am in es in Non a qu eou s Solven ts. 2.
Solven t-In d u ced Mech a n istic Ch a n ges
L. Garc´ıa-R´ıo,† J . R. Leis,*,† and E. Iglesias‡
Departamento de Quı´mica Fı´sica, Facultad de Qu´ımica, Universidad de Santiago,
15706 Santiago de Compostela, Spain, and Departamento de Qu´ımica Fundamental e Industrial,
Facultad de Ciencias, Universidad de La Corun˜a, La Corun˜a, Spain
Received February 3, 1997X
We studied the nitrosation of amines (pyrrolidine, piperidine, diethylamine, N-methylpiperazine,
N,N ′-dimethylethylenediamine, and morpholine) by alkyl nitrites (2-bromoethyl nitrite or 2,2-
dichloroethyl nitrite) or by N-methyl-N-nitroso-p-toluenesulfonamide (MNTS) in the solvents
chloroform, acetonitrile, and dimethyl sulfoxide (DMSO). The mechanism of nitrosation by alkyl
nitrites depends on the solvent: in chloroform, all the results were in keeping with formation of a
hydrogen-bonded complex between the amine and alkyl nitrite being followed by rate-controlling
formation of a tetrahedral intermediate T( that rapidly decomposes to afford the final products; in
acetonitrile, a situation intermediate between those obtaining in chloroform and cyclohexane results
in the [amine] dependence of the first-order pseudoconstant k0 being qualitatively influenced by
temperature and by the identities of both the amine and the alkyl nitrite; in DMSO, the results
suggest a mechanism close to the mechanism acting in water. For nitrosation by MNTS, k0 depended
linearly on [amine] in all three solvents. The Grunwald-Winstein coefficients correlating the rate
constants k for nitrosation by MNTS in the chloroform, acetonitrile, DMSO, dioxane, dichlo-
romethane, and water were l ) 0.12 and m ) 0.29. Correlation with the Kamlet-Abboud-Taft
equation confirmed that k depends largely on the dipolarity of the solvent and, to a lesser extent,
its capacity for hydrogen bonding.
In tr od u ction
for solvation of ions; and proton donation by chloroform5
might affect HBC formation. We did not use alcohols
(which would ideally have been desirable because of their
structural similarity to water) because they are them-
selves readily nitrosated,6,7 and preliminary experiments
in which nitrosation of formamides was detected simi-
larly deterred us from further investigation of the nit-
rosation of amines in solvents of this family.
As expected, we found that the more polar the solvent,
the more strongly was the mechanism deduced for the
reaction in cyclohexane distorted toward the mechanism
acting in water. Of particular interest, however, was the
finding that, for nitrosation by alkyl nitrites, acetonitrile
represents a kind of singular point in the continuum of
solvent properties, in that the mechanism acting in this
solvent depends qualitatively on both the nature of the
leaving group of the nitrosating agent and the nature of
the amine (whereas in other solvents the identity of the
amine generally has only a quantitative influence). This
finding underlines the difficulty of classifying solvents
or predicting their effects on even a single class of
reaction3 and provides an elegant example of how the
mechanism of a nucleophilic substitution reaction can
depend critically on the nature of nucleophile, leaving
group, and solvent.2
Studies of structure-reactivity relationships and sol-
vent effects have widely popularized the idea that reac-
tion mechanisms can be profoundly altered by changing
reactant substituents and/or reaction medium.2 In the
case of nucleophilic substitution and addition reactions,
extensive application of this concept has allowed several
new conclusions to be drawn with regard to the role of
nucleophile, leaving group, and solvent in reaction mech-
anism. The classification of a series of solvents can
present numerous difficulties.3 With this series of ar-
ticles on nitroso group transfer, we set out to classify
solvents phenomenologically by examining their effects
on nitrosation kinetics. In part 11 we examined the
reactions of several secondary amines with alkyl nitrites
(RONO) and N-methyl-N-nitroso-p-toluenesulfonamide
(MNTS) in nonaqueous apolar media (isooctane, cyclo-
hexane, dichloromethane, 1,4-dioxane, and tetrahydro-
furan) and found them to differ significantly from the
corresponding aqueous phase reactions.4
We now report the extension of this study to the more
polar nonaqueous solvents chloroform, acetonitrile, and
dimethyl sulfoxide (DMSO), which were chosen in view
of the results of part 1:1 acetonitrile and DMSO both have
large dielectric constants (ꢀ ) 36 for acetonitrile, 45 for
DMSO); DMSO in particular has a very high capacity
Exp er im en ta l Section
Chloroform (HPLC grade), acetonitrile (anhydrous), and
DMSO (spectophotometric grade) were supplied by Aldrich and
had nominal purities >99.9% and water contents <0.01%,
<0.005%, and <0.05%, respectively. Solubilization of atmo-
spheric moisture during experiments was to some extent
† Universidad de Santiago.
‡ Universidad de La Corun˜a.
X Abstract published in Advance ACS Abstracts, J une 1, 1997.
(1) Garcia-Rio, L.; Leis, J . R.; Iglesias, E. J . Org. Chem. 1997, 62,
4701.
(2) (a) J encks W. P. Chem. Rev. 1985, 6, 511. (b) Buncel, E.; Wilson,
H. Adv. Phys. Org. Chem. 1977, 14, 133.
(3) Reichardt, C. Solvents and Solvent Effects in Organic Chemistry;
Verlag Chemie: Weinheim, 1986.
(4) Garcia Rio, L.; Iglesias, E.; Leis, J . R.; Pen˜a, M. E.; Rios, A. J .
Chem. Soc., Perkin Trans. 2 1993, 29.
(5) Sandorfy, C.; Buchet, R.; Lussier, L. S.; Menassa, P.; Wilson, L.
Pure Appl. Chem. 1986, 58, 1115.
(6) Iglesias, E.; Garcia Rio, L.; Leis, J . R.; Pen˜a, M. E.; Williams, D.
L. H. J . Chem. Soc., Perkin Trans. 2 1992, 1673.
(7) Doyle, M. P.; Terpstra, J . A.; Pickering, R. A.; Le Poire, D. M. J .
Org. Chem. 1983, 48, 3379.
S0022-3263(97)00189-8 CCC: $14.00 © 1997 American Chemical Society