4
702 J . Org. Chem., Vol. 62, No. 14, 1997
Garc ´ı a-R ´ı o et al.
molecular sieves prior to use.21 Deuterated water (99.77% D)
was supplied by CIEMAT (Spain). Pyrrolidine (PYR), piperi-
dine (PIPER), diethylamine (DEA), N,N′-dimethylethylenedi-
amine (DED), N-methylpiperazine (MePIP), and morpholine
The solvent in which a reaction takes place plays a
1
5
crucial role in determining its mechanism, for example,
by controlling aggregation of the reactants or by altering
the lifetimes of any intermediates. To date, the influence
of the solvent on nitrosation reactions has not been
studied very intensively. Williams et al.16 examined the
S-nitrosation of thiourea and thioglycollic acid by alkyl
nitrites in acidified isopropyl and tert-butyl alcohols and
concluded that under these conditions the reactive species
is probably the protonated alkyl nitrite. Later these
authors also studied the nitrosation of alcohols, thiogly-
collic acid, and water by alkyl nitrites in acidified
acetonitrile, concluding that the rate-controlling step of
(MOR), all from Aldrich, were of the highest commercially
available purity and were distilled under argon shortly before
use. All other commercially supplied materials were Merck
or Aldrich products of the highest available purity and were
used as supplied.
1H NMR spectroscopy was performed on a Bruker AMX300
instrument operating at 300.1 MHz. The external reference
6
and lock signal were provided by DMSO-d contained in an
internal capillary coaxial with the NMR tube.
Alkyl nitrites (RONO) were prepared by a standard proce-
22
dure involving reaction of the corresponding alcohol with
+
these reactions was the formation of NO from the
sodium nitrite in an acid medium and were stored over 3 Å
molecular sieves pending use. N-Deuterated pyrrolidine and
N-deuterated diethylamine were prepared by repeated frac-
tional distillation of a mixture of the amine with a 10-fold
protonated nitrite or nitrous acid.1 More recently, ex-
periments in our laboratory have shown that the mech-
anism of the nitrosation of ureas in dioxane/water or
acetonitrile/water mixtures depends on the composition
of the solvent.1
1
9
2
molar excess of D O through a 20 cm Vigreux column. After
at least three runs, the product was dried over calcium
7,18
1
hydride, from which it was later distilled. H NMR spectros-
In spite of similarities between the electronic struc-
copy of the final product confirmed N-deuteration.
9
tures of alkyl nitrites and those of carboxylic esters,
In all the kinetic experiments the nitrosating agent (RONO
or MNTS) was in deficit, its concentration generally ranging
whose CdO group is isoelectronic with NdO, there are
marked differences between the chemical behaviors of
these two families of compounds. Particularly striking
is the fact that in aqueous solution alkyl nitrites react
-
4
-4
from 1 × 10 to 2 × 10 M. Reaction kinetics were studied
by monitoring absorbance (generally at a wavelength in the
range 250-270 nm) in Uvikon 930 or Milton Roy 3000 array
spectrophotometers. In all cases, absorbance-time data were
-
with OH much more slowly than the corresponding
0
fitted by integrated first order rate equations, and k , the
carboxylic esters, in spite of their reactions with other
nucleophiles, such as amines, being much faster than
those of carboxylic esters. More generally, whereas NO
donation by alkyl nitrites is thought to occur via a
concerted mechanism (as we have seen above for the
nitrosation of amines), the chemistry of carboxylic esters
is characterized by the formation of tetrahedral inter-
mediates.
corresponding first-order pseudoconstant, was reproducible to
within 3%. Regardless of the experimental conditions, the
N-nitrosamine was the product detected, and it showed no
signs of decomposition. Reaction yields calculated from the
absorbance data together with published values of the molar
2
3,24
absorption coefficients of the reagents and products
always very close to 100% ((5%). In some experiments
dioxane and THF) HPLC separation were carried out and
were
(
retention times and peak areas were compared with that of
pure N-nitrosamines. In every case we found quantitative
N-nitrosamine formation compatible with spectral changes
observed in kinetic experiments.
In this work we investigated the nitrosation of second-
ary amines in several different organic solvents by alkyl
nitrites or by N-methyl-N-nitroso-p-toluenesulfonamide
(
MNTS). The reactions of MNTS with amines in water
Resu lts
have rates similar to those of alkyl nitrites activated by
electron-withdrawing â-substituents (e.g., 2-ethoxyethyl
This section summarizes the results of an exhaustive
study of the reactions of 2-bromoethyl nitrite, 2,2-
dichloroethyl nitrite, 2,2,2-trichloroethyl nitrite, and
MNTS with each of several secondary aliphatic amines
in cyclohexane, and those obtained in selected experi-
ments using other aprotic solvents (isooctane, dichlo-
romethane, 1,4-dioxane, and THF).
1
2
nitrite ), have solvent isotope effects close to unity, and
are thought, unlike those of alkyl nitrites, to proceed via
a concerted mechanism which does not involve either
protonation of the leaving group or its water-assisted
expulsion.13 We found that a reaction mechanism analo-
gous to that proposed for the aminolysis of carboxylic
esters19 is able to account for the effect of the solvent on
the kinetics of these reactions.
In flu en ce of Am in e Con cen tr a tion . For nitrosation
0
by RONO, plotting k against [amine] (generally in the
-
3
range 3 × 10 to 1.00 M) gave sigmoid curves: Figures
1 and 2 show those obtained for the reactions between
pyrrolidine and 2-bromoethyl nitrite (Figure 1) and
between N-methylpiperazine and 2,2-dichloroethyl nitrite
Exp er im en ta l Section
Isooctane (nominal purity >99%, from Merck) and cyclo-
hexane and dichloromethane (nominal purities >99.9%, water
contents <0.01% and 0.02%, respectively; both from Aldrich)
were used as supplied. Published data for the solubility of
(
Figure 2) in cyclohexane. At an amine concentration
that was lower for the more basic amines (pyrrolidine
and piperidine) than for the others, k began to level off
at a limiting value that likewise depended on the amine
2
0
water in these solvents indicated that no special precautions
would be necessary in the kinetic experiments involving them.
0
1
,4-Dioxane and tetrahydrofuran (nominal purities >99%,
(Figure 3). The linear dependence of k
0
/[amine] on
water contents <0.005% and 0.02%, respectively; both from
Aldrich) are miscible with water and were dried over 3 Å
[amine] at low amine concentration (Figure 4) shows that
under these conditions the reaction has both a pathway
of first order with respect to amine and one of second
order.
No leveling off was observed in the corresponding plots
for nitrosation by MNTS. For example, for the nitrosa-
(
15) Reichardt, C. Solvents and Solvent Effects in Organic Chemistry;
Verlag Chemie: Weinheim, 1986.
16) Crookes, M. J .; Williams, D. L. H. J . Chem. Soc., Perkin Trans.
1989, 759.
17) Bravo, C.; Herves, P.; Iglesias, E.; Leis, J . R.; Pe n˜ a, M. E. J .
Chem. Soc., Perkin Trans. 2 1990, 1969.
18) Bravo, C.; Herves, P.; Leis, J . R.; Pe n˜ a, M. E. J . Chem. Soc.,
Perkin Trans. 2 1991, 2091.
(
2
(
(
(21) Coetzee, J . F.; Chang, T. H. Pure Appl. Chem. 1985, 57, 633.
(22) Noyes, W. A. Organic Syntheses; Wiley: New York, 1943;
Collect. Vol. II, p 108.
(23) Haszeldine, R. N.; Mattinson, B. J . H. J . Chem. Soc. 1955, 4172.
(24) Tarte, P. J . Chem. Phys. 1952, 20, 1570.
(
(
19) Menger, F. M.; Smith, J . H. J . Am. Chem. Soc. 1972, 94, 3824.
20) Sorensen, J . M.; Arlt, W. Liquid-Liquid Equilibrium Data
Collection; Dechema: Frankfurt, 1979.