Evidence for radical cations in linked mechanisms of N,N-Dialkyl aromatic amine nitration and nitrosative dealkylation

Article abstract of DOI:10.1021/ja9732744

N,N,-Dialkyl aromatic amines react rapidly with nitrous acid to competitively produced a nitrosamine and a nitro compound. The mechanism of nitro compound formation involves a reaction of an amine radical cation with NO₂, while the nitrosamine

Full text of DOI:10.1021/ja9732744

J. Am. Chem. Soc. 1998, 120, 5193-5202
5193
Evidence for Radical Cations in Linked Mechanisms of N,N-Dialkyl
Aromatic Amine Nitration and Nitrosative Dealkylation
Richard N. Loeppky,* Sukhjeet P. Singh, Saleh Elomari, Riley Hastings, and
Thomas E. Theiss
Contribution from the Department of Chemistry, UniVersity of Missouri-Columbia,
Columbia, Missouri 65211
ReceiVed September 17, 1997. ReVised Manuscript ReceiVed March 23, 1998
Abstract: N,N-Dialkyl aromatic amines react rapidly with nitrous acid to competitively produce a nitrosamine
and a nitro compound. The mechanism of nitro compound formation involves a reaction of an amine radical
cation with NO2, while the nitrosamine is produced by two competing pathways, one of which involves N-R-
CH deprotonation of a radical cation with subsequent oxidative generation of an imminium ion, and the other
of which occurs through NOH elimination of a nitrosammonium ion (R3N-NdO+). All three pathways are
linked through the reversible homolysis of R3N-NdO+ to NO and a radical cation.
Introduction
Scheme 1
Much of the recent organic chemistry of nitrous acid has been
focused on the possible inadvertent formation of potentially
carcinogenic N-nitroso compounds such as nitrosamines. In-
deed, we were recently led to consider the possible formation
of nitrosamines from esters of p-N,N-dialkylbenzoic acid, which
are used commercially as sun-blocking agents, among other
things.1,2 Prior work in our laboratory and those of others raised
significant questions, however, regarding the ability of N,N-
dialkyl aromatic amines to nitrosate sufficiently fast to produce
significant quantities of nitrosamines at ambient temperatures.3,4
Most, but certainly not all, tertiary aliphatic amines react too
slowly with nitrous acid at 25 °C to produce significant
quantities (0.1%) of nitrosamines.3,4 The nitrosative dealkylation
of triethylamine, typical of many tertiary amines, can be
estimated to be 6.7 × 10-7 M-1 s-1 at 25 °C,5 but is too slow
for measurement at this temperature.3,4 For simple amines, the
rate of nitrosamine formation only competes with the thermal
decomposition of HNO2 at temperatures above 50-60 °C. The
low nitrosation rates of simple aliphatic tertiary amines are due,
in part, to their high basicity and protonation under acidic
conditions required for the generation of nitrosating agents.
While N,N-dialkylanilines are much less basic than triethy-
lamine, their rapid nitrosative dealkylation was not anticipated.
A review of the early literature, however, reveals that N,N-
dialkyl aromatic amines are both nitrosatively dealkylated and
nitrated by nitrous acid,6,7 but the kinetics and mechanisms of
these transformations have not been reported. We found that
ethyl 4-(dimethylamino)benzoate (1, X ) CO2Et) reacts with
HNO2 in HOAc to form large quantities of ethyl 4-methylnit-
rosaminobenzoate (2, X ) CO2Et) and smaller amounts of ethyl
4-(dimethylamino)-2-nitrobenzoate (3, X ) CO2Et) in less than
an hour at 25 °C (Scheme 1).1,2 This qualitative finding
naturally led us to ask if nitrosamine formation in this system
was occurring by a mechanism different than elucidated for the
nitrosation of trialkyl tertiary amines (Scheme 2, path A).3,4
The mechanism and kinetics of nitrosative dealkylation of
trialkyl tertiary amines have been investigated.3,4 A key marker
of this mechanistic pathway, shown as path A (1 f 4 f 8 f
2) in Scheme 2, is the formation of 0.5 mol of N2O for every
mole of nitrosamine formed. We have identified other structural
features of tertiary amines which render them highly reactive
toward nitrosative dealkylation and demonstrated that they occur
by different mechanisms than that shown in path A of Scheme
2, but none of these would appear to be applicable to the
reactions of N,N-dialkyl anilines.8-14 Our preliminary inves-
tigations of the nitrosation of 1 (X ) CO2Et) showed that the
(1) Loeppky, R. N.; Hastings, R.; Sandbothe, J.; Heller, D.; Bao, Y.;
Nagel, D. In ReleVance of Human Cancer of N-Nitroso Compounds,
Tobacco Smoke, and Mycotoxins, IARC Scientific Publication ed.; O’Neill,
I. K., Chen, J., Bartsch, H., Eds.; International Agency for Research on
Cancer: Lyon, France, 1991; Vol. 105, pp 244-252.
(2) Singh, S.; Hastings, R.; Loeppky, R. N. In Nitrosamines and related
N-nitroso compounds: chemistry and biochemistry; Loeppky, R. N.;
Michejda, C. J., Eds.; American Chemical Society: Washington, DC, 1994;
pp 309-311.
(7) Hodgson, H. H.; Nicholson, D. E. J. Chem. Soc. 1941, 470-475.
(8) Loeppky, R. N.; Yu, L. Tetrahedron Lett. 1990, 31, 3263-3266.
(9) Loeppky, R. N.; Outram, J. R.; Tomasik, W.; Faulconer, J. M.
Tetrahedron Lett. 1983, 24, 4271-4274.
(10) Loeppky, R. N.; Tomasik, T. J. Org. Chem. 1983, 48, 2751-2756.
(11) Loeppky, R. N.; Shevlin, G.; Yu, L. In Significance of N-Nitrosation
of Drugs (Drug DeV. EVal); Eisenbrand, G., Bolzer, G., Nicolai, H. v., Eds.;
Gustav Fischer Verlag: New York, 1990; Vol. 16, pp 253-266.
(12) Loeppky, R. N.; Bae, J. Y. Chem. Res. Toxicol. 1994, 7, 861-867.
(13) Bae, J.-Y.; Mende, P.; Shevlin, G.; Spiegelhalder, B.; Preussmann,
R.; Loeppky, R. N. Chem. Res. Toxicol. 1994, 7, 868-876.
(14) Loeppky, R. N.; Bao, Y. T.; Bae, J. Y.; Yu, L.; Shevlin, G. In
Nitrosamines and Related N-Nitroso Compounds: Chemistry and Biochem-
istry; Loeppky, R. N.; Michejda, C. J., Eds.; American Chemical Society:
Washington, DC, 1994; pp 52-65.
(3) Smith, P. A. S.; Loeppky, R. N. J. Am. Chem. Soc. 1967, 89, 1147-
1157.
(4) Gowenlock, B.; Hutcheson, R. J.; Little, J.; Pfab, J. J. Chem. Soc.,
Perkin Trans. 2 1979, 1110-1114.
(5) Calculated from rate constants and activation parameters in ref 4.
(6) Pinow, J. Chem. Ber. 1898, 31, 2982-2987.
S0002-7863(97)03274-5 CCC: $15.00 © 1998 American Chemical Society
Published on Web 05/14/1998

5194 J. Am. Chem. Soc., Vol. 120, No. 21, 1998
Loeppky et al.
Scheme 2
thorough investigation of the possible role of radical cations in
the nitrous acid chemistry of N,N-dialkyl tertiary amines.
Overview
To improve the readability of this paper, we present here an
overview of the mechanistic hypothesis that has arisen from
this work. This is shown in Scheme 2. We propose that all
reactions proceed through the reversible formation of a nitro-
sammonium ion 4, which either suffers reversible homolysis to
generate NO and radical cation 5 or loses NOH ultimately to
give a nitrosamine (path A) by the established mechanism.3,4
The radical cation 5 either reacts with NO2 to give the nitro
compound 3 (path C), loses a proton from the N-bound alkyl
group to eventually produce a nitrosamine (path B), a new
mechanism of nitrosative dealkylation, or reacts with NO to
regenerate 4.
In the presentation of the data and the attending discussion
we will first show that nitration by nitrous acid in this system
involves a radical cation. We will then show that nitrosamine
formation occurs by more than one mechanism, and will present
data consistent with the existence of a radical cation pathway,
which competes with path A (NOH elimination). We present
data supporting the hypothesis that the homolysis of the
nitrosammonium ion is reversible, and that this equilibrium links
nitrosamine and nitro compound formation in such a way that
minor changes in reaction conditions and substrate structure
change the pathway taken.
relative yields of nitrosamine and nitro compound vary signifi-
cantly with relatively minor changes in reaction conditions.
These data suggested the operation of facile mechanistic
switching and led us to consider a possible role for radical cation
intermediates in the nitrosation chemistry of these aromatic
amines.
General Characteristics of the Transformation
The reactions of p-substituted-N,N-dialkylanilines were stud-
ied under several different nitrosating conditions resulting from
the additon of concentrated aqueous sodium nitrite to either
glacial acetic acid (HOAc) or 60% HOAc/H2O buffered to pH
3.8 with NaOAc. (Even though some transformations produce
the nitro compound 3 as the major product, we will use the
term nitrosating agent in connection with the reagent nitrous
acid, regardless of the nature of the products generated.) The
literature suggests that the principal nitrosating agent under these
conditions is N2O3.24 Its initial steady state concentration is
significantly higher in HOAc than in buffer. Nitrosations in
HOAc occur more rapidly, but significant amounts of N2O3 are
lost due to reversible homolysis to NO2 and NO, followed by
escape of the volatile gases. Accordingly, some transformations
were examined in closed systems with no headspace above the
liquid.
Radical cations have been shown to be intermediates in the
nitrous acid-catalyzed HNO3 nitration of N,N-dialkyl aromatic
amines and other aromatic compounds.15-21 The relatively
facile one-electron oxidation of aromatic amines also has led
Colona et al. to propose the involvement of radical cations in
nitrosative N-dealkylations of aromatic amines.22 This group
reported the ipso (p) and o-nitration of a group of p-substituted-
N,N-dimethylanilines by nitrite in HOAc.22 In two cases
nitrosative demethylation was also observed. On the basis of
their earlier work, they proposed that the nitration occurred
through radical recombination of NO with radical cation
followed by oxidation to the nitro compound, and that dem-
ethylation also occurred through a radical cation by an unspeci-
fied process. Verardo et al. observed the formation of nitro
compounds, nitrosamines, and other substances when aromatic
tertiary amines were heated with nitrite esters at reflux,23 and
proposed that nitro compound formation arose from the reaction
of a radical cation with NO2, but gave no direct role to radical
cations in nitrosamine formation. These observations, and our
own findings regarding the unexpectedly high reactivity of 1
(X ) CO2Et) toward nitrosative dealkylation, called for a more
In a typical nitrosation, N,N-dimethyl-4-chloroaniline 1 (X
) Cl) (0.23 mM) was reacted with NaNO2 (0.69mM) in 60%
HOAc buffer at pH 3.8 at 21 °C. The reaction was complete
in 90 min with the production of the nitrosamine 2 (X ) Cl)
(83%) and the nitro compound 3 (X ) Cl) (10%). Under similar
conditions the ratio of the nitro compound to nitrosamine (3/2
X ) Cl) increased linearly as the initial [NO2-] to [amine] ratio
increased. For example, the ratio 3/2 (X ) Cl) was 0.6 with a
(15) Ridd, J. H. Chem. Soc. ReV. 1991, 20, 149-165.
(16) Eberson, L.; Radner, F. Acc. Chem. Res. 1987, 20, 53-59.
(17) Eberson, L.; Hartshorn, M. P.; Radner, F. Acta Chem. Scand. 1994,
48, 937-950.
-
3× excess of NO2 but increased to 2.8 with a 15× excess.
While this system appeared to be well behaved, other reaction
variables can change the relative product composition. We have
made numerous observations and a sample of these data are
provided in Table 1. The following conclusions can be made:
(1) The ratio of 3/2 was much greater when the reaction was
conducted in glacial HOAc. (2) The Cl-substituted compound
not only reacted more rapidly than the CO2Et-substituted
derivative but the ratio of 3/2 was also much greater. (3)
(18) Kochi, J. K. Acc. Chem. Res. 1992, 25, 39-47.
(19) Al-Obaidi, U.; Moodie, R. B. J. Chem. Soc., Perkin Trans. 2 1985,
467-472.
(20) Dix, L. R.; Moodie, R. B. J. Chem. Soc., Perkin Trans. 2 1986,
1097-1101.
(21) Beake, B. D.; Constantine, J.; Moodie, R. B. J. Chem. Soc., Perkin
Trans. 2 1994, 335-340.
(22) Colona, M.; Greci, L.; Poloni, M. J. Chem. Soc., Perkin Trans. 2
1984, 165-169.
(23) Verardo, G.; Giumanini, A. G.; Strazzolini, P. Tetrahedron 1990,
46, 4303-4332.
(24) Williams, D. H. L. Nitrosation; Cambridge University Press:
Cambridge, 1988; pp 1-10.

Linked Mechanisms of Nitration and NitrosatiVe Dealkylation
J. Am. Chem. Soc., Vol. 120, No. 21, 1998 5195
Table 1. Variation of the 3/2 Ratio with Reaction Conditions for
the Nitrosation of 1a
reaction complexity and the mechanistic switching observed for
some substrates bearing the CO2Et substitutent, we do not know
the nitrite order for this substrate.
run
X
[NO2]o [Am]o
solv
time % rxn 3/2 % 2
1
2
3
Cl
Cl
Cl
0.69
3.3
1
2.6
2.6
.25 buffer 78
96
98
65
38
30
81
48
52
53
The Mechanism of Nitro Compound Formation
1.1
1.1
HOAc
HOAc
3
3
6
6.2
4b Cl
0.23 buffer
0.23 buffer
To our knowledge, the mechanism of nitrous acid nitration
has not been the object of a modern study, but a closely related
transformation, the nitrous acid-catalyzed nitration by HNO3
of relatively electron rich aromatic substrates, has been carefully
examined. While the work in this area has been reviewed by
several authors who have made seminal contributions,15-21 the
reader’s attention is directed to a short review by Ridd,15 who
presented strong evidence for the mechanism shown in Scheme
3.26-30
5
6
7
Cl
CO2Et 67
CO2Et 18
6.7
6.9
2.7
2.7
HOAc 47
buffer 50
buffer 61
buffer 61
8b CO2Et 27
9
CO2Et 27
a General notes: concentrations are in mM; buffer is 60% HOAc/
H2O with NaOAc at pH 3.8; HOAc is glacial acetic acid; times are in
min; unless noted specifically reactions were open to the atmosphere.
b Without headspace.
In Ridd’s work, and also in this paper, NO+ is used to
represent the various species which are effective NO+ donors.
The key transformation is the oxidation of the aniline to a radical
cation by NO+ (eq 1). The nitronium ion or a related species
is then seen to oxidize the NO produced back to NO+ with the
formation of NO2. Reaction of the radical cation with NO2 then
leads to the formation of the nitro compound. This mechanism
is based on kinetics, 15N CIDNP NMR data, product studies,
electrochemical data, and ESR experiments.15,26-30 Related
work on other aromatic substrates by Kochi,18 Eberson and
Radner,16,17 and Moodie et al.20,21 present similar mechanisms.
We show here that a closely related mechanism is responsible
for nitration by HNO2.
No NO3- and hence NO2+ are present in our reaction media,
ruling out conventional electrophilic aromatic nitration. More-
over, it has been demonstrated with several substrates that
nitration by nitrosating agents occurs much more rapidly than
conventional HNO3 acid-catalyzed nitration of aromatic amines
and leads to a different regiochemistry.31 It has been assumed
that the nitro compounds arise in nitrosation reactions by
C-nitrosation followed by oxidation.15,22,31,32 To test this
hypothesis, we sought the synthesis of 4-chloro-2-nitroso-N,N-
dimethylaniline. Despite numerous attempts we were unable
to prepare this compound and, surprisingly, 2-nitroso-N,N-
dialkylanilines appear to be unknown. We then examined the
possible oxidation of 4-nitroso-N,N-dimethylyaniline under the
conditions in which 3 was produced in high yield. The C-nitroso
compound was recovered unchanged under conditions where
the consumption of 1 (X ) Cl) by nitrosating agents was
complete. These data provide strong evidence that 3 (X ) Cl)
was not being produced in our system by the oxidation of a
C-nitroso compound.
Table 2. The Effect of Added Gases on Product Ratios from the
Nitrosation of 1
gas
% rxn
% 2
% 3
3/2
N2
NO
O2
52
76
97
26
49
15
24
24
75
0.9
0.5
5
Regardless of the ring substituent or the solvent, the 3/2 ratio
increased with [NO2-]. (4) When reactions were conducted
without headspace in closed vessels, the 3/2 ratio was greater
regardless of the ring substituent. In the case of the nitrosation
of 1 (X ) CO2Et) the product ratio was seen to change over
the course of runs in HOAc. For example, the ratio of 3/2
(X ) CO2Et) changed smoothly, from 0.23 at 11% reaction to
0.12 at 63% reaction.
As may be expected from the headspace studies, reaction of
1 (X ) Cl) in HOAc with a 10× [NO2-] excess in the presence
of added gases change the 3/2 ratio dramatically. In these
experiments, a HOAc solution of 1 was run through a set of
freeze-thaw cycles under vacuum to remove residual gases.
The headspace above the liquid was then saturated with either
N2, NO, or O2,, and a 10× excess of concentrated aqueous
sodium nitrite was introduced by syringe. The product yields
are given in Table 2. Addition of NO increased the yield of
nitrosamine 2 (X ) Cl) while addition of O2 markedly increased
the relative and absolute quantity of nitro compound 3
(X ) Cl).
The nature of N-alkyl substituents can also alter the 3/2 ratio.
Deuteration of the methyl groups bound to N increases the
relative amount of nitro compound (vide infra). N-Cyclopropyl-
N-methyl-4-chloro- or 4-carboethoxyaniline react with HNO2
in HOAc to give only the nitrosamine 2 and no detectable
amount of nitro compound. This phenomenon was seen for all
compounds bearing a cyclpropyl group on N.25
Kinetic data for nitrosation of 1 (X ) Cl or CO2Et) are
reported in Table 3. Reactions were followed by determining
the concentrations of the amine, nitrosamine, and nitro com-
pound by HPLC as a function of time. The rate data given in
Table 3 are for the disappearance of the amine. Rate constants
are the average of two to three determinations derived from
linear regression of ln([amine]) vs time at nitrite concentrations
3- to 10-fold greater than that of the amine (pseudo first order).
Key evidence for the intermediacy of an amine radical cation
in the nitrous acid-catalyzed nitration of N,N-dimethyl-p-
toluidine by H15NO3 involved the Ridd group’s observation of
enhanced emission in the 15N CIDNP NMR spectra of 4-methyl-
2-nitro-N,N-dimethylaniline.27,30 To determine whether a radical
was involved in the production of 3 we nitrosated 1 with Na-
15NO2 and the 15N-NMR spectra were recorded as a function
(26) Giffney, J. C.; Mills, D. J.; Ridd, J. H. J. Chem. Soc., Chem.
Commun. 1976, 19-20.
(27) Giffney, J. C.; Ridd, J. H. J. Chem. Soc., Perkin Trans. 2 1979,
618-623.
(28) Al-Omran, F.; Fujiwara, K.; Giffney, J. C.; Ridd, J. H.; Robinson,
S. R. J. Chem. Soc., Perkin Trans. 2 1981, 518-525.
(29) Ridd, J. H.; Sandall, J. P. B. J. Chem. Soc., Chem. Commun. 1981,
402-403.
(30) Clemens, A. H.; Helsby, P.; Ridd, J. H.; Al-Omran, F.; Sandall, J.
P. J. Chem. Soc., Perkin Trans. 2 1985, 1217-1225.
(31) Aitken, M. F.; Reade, T. H. J. Chem. Soc. 1926, 1896-1897.
(32) Crowley, G. P.; Milton, G. J. G.; Reade, T. H.; Todd, W. M. J.
Chem. Soc. 1940, 1286-1289.
-
Good linearity was observed in all cases. The order in NO2
was determined only for 1 (X ) Cl) in buffer by plotting log-
(kobs) vs log([NO2-]) for runs where the [NO2-]0 was varied
giving a nitrite order of 0.97 ( 0.08 for this substrate. Second-
order rate constants were calculated for all runs but, given the
(25) Loeppky, R. N.; Elomari, S. Manuscript in preparation.

5196 J. Am. Chem. Soc., Vol. 120, No. 21, 1998
Loeppky et al.
Table 3. Rates of Amine Disappearance upon Nitrosation
compda
104kobs
[NO2
]
[NO2-]/[Am]
100k2nd
solvent
b
-
c
d
run
T, °C
1
2
3
4
5
6
21
21
30
30
30
30
Cl
Cl (D6)
CO2Et
CO2Et (D6)
CO2Et
6.6 ( 0.2
0.69
0.69
20.8
23.5
67
67
2.8
3.2
3
3
10
10
96 ( 3
buffere
buffer
buffer
buffer
HOAcf
HOAc
1.51 ( 0.08
4.4 ( 0.2
22 ( 1
2.1 ( 0.1
1.18 ( 0.04
6.0 ( 0.2
2.50 ( 0.04
0.50 ( 0.02
0.89 ( 0.03
0.373 ( 0.006
CO2Et (D6)
a p-Substituted N,N-dimethylanilines 1; (D6) ) N(CD3)2. b Pseudo first order rate constant (s-1). c mM. d 2nd order rate constant calculated by
dividing kobs by [NO2-] (s-1 M-1). e 60% aqueous HOAc buffered to pH 3.8 with NaOAc (1.66 M). f Glacial acetic acid.
equation,33 modified for 15N,34 led to a similar conclusion.35
Reaction of an amine radical cation with NO2 to generate 6
(Scheme 2) is completely consistent with the observed CIDNP
emission. We propose that the mechanism of nitration by
nitrous acid is adequately represented by the transformations
depicted in path C of Scheme 2 (1 f 4 f 5 f 6 f 3). The
radical cation 5 is proposed to arise through the homolysis of
the nitrosammonium ion 4 (vide infra), but could be produced
by an equivalent set of transformations involving oxidation by
N2O3, another NO+ donor, or NO2. Under our conditions, NO2
is generated through the well-known facile homolysis of N2O3.
The reaction of 5 with NO2 generates the familiar nitration
intermediate 6, which rapidly deprotonates to generate 3.
This mechanistic hypothesis is supported by several other
pieces of data. The fact that the 15N CIDNP evolved over the
Figure 1. 15N-NMR CIDNP spectra for the reaction of Na15NO2 with
same time frame as the production of the nitro compound, as
1 (X ) Cl) in D4-HOAc. Spectra were accumulated over the time
1
intervals shown following the addition of Na15NO2 in D2O. Line
assignments: a, Ph15NO2 standard; b, 3 (X ) Cl) ; c, 2 (X ) Cl); d,
H15NO2. For each trace the line intensities should be compared to that
of the nitrobenzene standard because of computer manipulation of
graphical data for presentation purposes.
independently measured by H NMR, provides evidence that
the amine radical cation is on the reaction path to the nitro
compound and that the CIDNP does not result from some minor
side reaction. While the rate of disappearance of 1 (X ) Cl) is
first order in [NO2-], the ratio of nitro compound to nitrosamine
3/2 (X ) Cl) increased linearly with increasing [NO2-],
suggesting that nitro compound formation is second order in
[NO2-]. Rate determining formation of the radical cation 5
followed by a more rapid reaction of it with NO2, as shown in
Scheme 2¸ would meet this requirement. Other observations
which support the proposed mechanism include the fact that
the 3/2 (X ) Cl) ratio increased when the reaction was
conducted in a closed vial without headspace (compare entries
4 with 5 and 8 with 9 in Table 1). These conditions retained
the volatile NO2 and increased its reaction with the radical
cation. Similarly, the saturation of the system with O2 increased
the concentration of NO2 through the oxidation of NO and
increased the formation of the nitro compound (Table 2).
Scheme 3
When one of the N-bound substituents is capable of a reaction
that rapidly transforms the radical cation, no nitro compound
is produced. Neither 4-chloro-N-cyclopropyl-N-methylaniline
nor any other N-cyclopropyl-N-alkylaniline examined by us
produced any nitro compound upon reaction with HNO2.25 The
sole amine derived product in these reactions is the nitrosamine
2 (Scheme 4). It is well-known that the generation of radicals
at atoms attached to the cyclopropyl group results in ring
opening. This process is so fast as to preclude the possibility
of NO2/amine radical cation combination at the aromatic ring.
of time. To enhance sensitivity, higher concentrations ([1(X
) Cl)] ) 0.04 M and [NO2-] ) 0.30 M in D4-HOAc) had to
be employed than those used in the kinetics experiments. 1H
NMR spectra were recorded under the same conditions to verify
1
the reaction rate, course, and product identity. The H NMR
spectra showed no evidence for the formation of an ipso (para)
intermediate, as was observed by Giffney and Ridd27 in the
nitration of N,N-dimethyl-p-toluidine. 15N-NMR spectra were
accumulated over the intervals shown in Figure 1 after Na-
15NO2 addition. The spectra at the first two time intervals show
a strong but diminishing emission signal for the nitro group of
3 (X ) Cl). In the latter time runs the emission signal has
disappeared and only the normal absorption peak of the product
nitro signal was observed.
(33) Kaptein, R. J. J. Chem. Soc. D 1971, 732-733.
(34) Porter, N. A.; Dubay, G. R.; Green, J. G. J. Am. Chem. Soc. 1978,
100, 920-925.
(35) The Kaptein equation modified for 15N is Γ ) -µꢀ∆gaN, where µ
is positive for F-radical pairs formed by diffusion, ꢀ is positive for product
formation from geminate collapse of these radicals (NO2 and 5 (X ) Cl)),
∆g ) gNO - grad.cat is negative (gNO ) 2.000 and grad.cat can be estimated
2
to be close to 2.0032, the value for 2the radical cation derived from N,N-
dimethyl-p-toluidine30), and aN for NO2 is negative,30 giving Γ ) -(emis-
sion).
Our CIDNP data were analogous to those obtained by the
Ridd group29,30 and an analogous application of the Kaptein

Linked Mechanisms of Nitration and NitrosatiVe Dealkylation
J. Am. Chem. Soc., Vol. 120, No. 21, 1998 5197
Table 4. Products and Yields from Reactions of Radical Cations and Controls
run
substrate
reactant
% rxna
% 2
other products
1
2
3
1 (X ) N(CH3)2)
5 (X ) N(CH3)2)
5 (X ) N(CH3)2)
NO2- /HOAc
NO2- /H2O
NO/H2O
97
99
89
31
74
58
12 (59%), 13 (7%)
12 (4%), 13 (2%)
12 (12%), 13 (3%)
a Percent reaction determined by measuring the amount of unreacted amine recovered.
Scheme 4
These model transformations demonstrate the following: (1)
-
Neither NO nor NO2- is reactive toward the amines. (2) NO2
does not react with these radical cations to produce significant
quantities of nitro compounds, especially when compared to
the nitrous acid nitrosations (Table 4, run 1). (3) The major
product from reactions of the radical cations with either NO or
-
NO2 is the nitrosamine. (4) NO2 reacts with both radical
cations and their parent amines, blurring somewhat the distinc-
tions between the transformations, but supporting the hypothesis
that NO2 can both oxidize the amine to a radical cation and
react with the latter to produce a nitro compound. We discuss
the mechanisms of these transformations further below. Al-
though every attempt was made to keep these systems free of
O2, it is difficult to avoid some contamination and we believe
that the small relative yields of the nitro compound in the
Wurster’s blue transformation with NO (Table 4, run 3) likely
resulted from production of NO2 by the oxidation of NO.
These data and those discussed above are completely consistent
with the mechanism of nitro compound formation (path C) in
Scheme 2.
Model Radical Cation Reactions
While the mechanism for nitro compound formation appears
secure within the framework of data and literature precedent
provided, several questions remained. Why do we apparently
see no products arising from the combination or recombination
-
of NO with the amine radical cation? Is it possible that NO2
Evidence for More than One Mechanism of Nitrosamine
Formation
reacts with the radical cation to produce a nitro compound? To
answer these questions we prepared amine radical cations and
examined their transformations with NO, NO2, and NO2-. In
each case control reactions involving the parent amine also were
carried out. The reactions of radical cations derived from either
1 (X ) Cl or X ) N(CH3)2, N,N,N′,N′-tetramethyl-1,4-
benzendiamine, the precursor of Wurster’s blue) have been
examined. Because of its greater stability, more reliable
quantitative data were obtained from the transformations of
Wurster’s blue, although those from the more reactive radical
cation derived from 1 (X ) Cl) were similar, but have been
omitted from Table 4.
Aqueous solutions of purified Wurster’s blue 5 (X )
N(CH3)2) reacted with both NaNO2 and NO resulted in
predominant formation of the nitrosamine 2 (X ) N(CH3)2)
(Table 4, runs 2 and 3). The major product from the reaction
of NO2 with 5 (X ) N(CH3)2) (CH3CN, 10 min) was the nitro-
nitrosamine 12 with much smaller amounts of 13. However,
The data presented so far provide strong evidence that nitro
compound formation in these systems involves a radical cation
intermediate. Several pieces of data also implicate a role for
radical cations in the mechanisms of nitrosamine formation.
These include the finding that N-cyclpropyl-N-alkylanilines only
give nitrosamines upon reaction with HNO2, and the observation
that the model radical cations react with NO2- or NO to produce
nitrosamines as the major products. In seeking to determine
whether more than one mechanistic pathway was giving rise to
nitrosamine formation in these reactions, we examined the ratio
of 2 (X ) Cl) to N2O produced in the nitrous acid transforma-
tions. The mechanism given in path A of Scheme 2 predicts
that 2 mol of nitrosamine should be produced for every mole
of N2O. The production of N2O in the nitrosation of triben-
zylamine had been confirmed,3 but no studies have been carried
out to verify the predicted product ratio. We reasoned that
competition from another mechanism of nitrosamine formation
would either eliminate N2O formation or significantly increase
the 2 (X ) Cl)/N2O ratio.
Quantitation of N2O Production. To test this hypothesis
we developed a method for quantitating the amount of N2O
produced in the nitrosation of 1 (X ) Cl, CO2Et) as a function
of reaction conditions. N2O was determined by sampling the
headspace above stirred reaction mixtures and quantifying by
GC. Immediately after the gas sampling, the nitrosamine yield
was determined by HPLC. The data are given in Table 5. To
ensure sensitivity in the N2O determinations most of the
reactions were run to near completion, but this condition also
produces limitations in the interpretation of the data because of
the possibility of mechanistic switching during the course of
the run. Nevertheless, it is evident that nitrosamine/N2O ratios
as high as 9.5 were observed in some cases, clearly indicating
the operation of more than one mechanism of nitrosamine
formation. From these data it can be seen that the fraction of
nitrosamine (frad) produced by a radical cation pathway was a
function of the initial reactant concentrations, the solvent, the
substituent, and, in glacial acetic acid, added base in the form
of sodium acetate (run 3).
reaction of 1 (X ) N(CH3)2) with NO2 under similar conditions
(30 min) resulted in the formation of 2 (X ) N(CH3)2) as the
major product with minor amounts of 13. The predominant
formation of 12 from the reaction of Wurster’s blue with NO2
supports the mechanism proposed for nitro compound formation.
Subsequent oxidation of the resulting amine and reaction with
-
the accumulating reduction product (NO2 or HNO2) leads to
nitrosamine formation. In the case of the reaction of the parent
amine (1, X ) N(CH3)2) with NO2, the primary reaction was
an oxidation producing 5 and NO2-. The relative quantity of
NO2 was significantly reduced at the expense of NO2- formation
and the predominant reaction was nitrosamine formation as was
-
observed in the reaction of 5 with NO2
.

5198 J. Am. Chem. Soc., Vol. 120, No. 21, 1998
Loeppky et al.
2),3,4 and another, described here for the first time, occurs
through the deprotonation of a radical cation (1 f 4 f 5 f 7
f 8 f 2, path B of Scheme 2). A better understanding of the
evidence and arguments which support our hypotheses is
facilitated by consideration of the rate equation (eq 4) derived
Table 5. Nitrosamine/N2O Ratios as a Function of Conditions and
Substrate for the Nitrosation of 1
-
b
runa
X
[Am]b [NO2
]
solv ratioc fradd % 2 3/2 % rxn
1
2
3
4
5
6
7
8
9
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
0.052 0.157 HOAc 2.00 0.00 11.8 7.39
0.052 0.052 HOAc 2.00 0.00 24.1 1.70
0.052 0.052 HOAce 4.44 0.55 28.8 1.56
99
65
74
0.012 0.037 buffer 9.52 0.79 43.5 1.29 100
K4[Am][N2O3]
[Am]
dt
0.024 0.037 buffer 4.65 0.57 54.9 0.79
0.034 0.037 buffer 3.70 0.46 32.0 0.43
0.052 0.037 buffer 3.33 0.40 26.9 0.30
0.052 0.157 buffer 2.17 0.08 42.9 1.22
0.012 0.095 buffer 2.00 0.00 34.6 1.86
98
46
35
95
99
- d
)
k8 +
-
{
[NO2 ]
k5k7[Y-] + k5k6[NO2]
(4)
-
}
k-5[NO] + k6[NO2] + k7[Y ]
10 CO2Et 0.050 0.152 HOAc 2.47 0.19 40.5 1.48 100
11 CO2Et 0.050 0.152 buffer 2.00 0.00 29.9 0.11
12 CO2Et 0.012 0.037 buffer 4.76 0.58 32.3 0.08
33
35
from the mechanistic scaffold presented in Scheme 2. The
derivation required several important assumptions: (1) Steady-
state conditions were assumed for both the nitrosammonium
ion 4 and the radical cation 5. (2) The radical 7 and the
imminium ion 8 were presumed to be rapidly oxidized and
converted to the nitrosamine, respectively. (3) Hydrolysis of
the nitrosammonium ion 4 to 1 was presumed to be more rapid
than its other transformations.4 And (4) the major base (Y-)
involved in the deprotonation of the radical cation 5 was AcO-.
The latter assumption will hold under the buffer conditions used
here, and when [NO2-] . [Am], but will break down if that
condition is not met and would thus invalidate eq 4. The kinetic
measurements of the nitrosation of 1 (X ) Cl) were carried out
under high acetate concentrations ([AcO-] ) [Y-]). As a result,
terms containing k7 in eq 4 may be large in comparison to
summed components (k7[Y-] . k6[NO2] and k7[Y-] . k-5[NO]).
Under these constraints it can be shown that eq 4 reduces to eq
5, where Ka is the dissociation constant for AmH+, K2 is the
a Each entry is the average of four to five separate determinations.
b Concentrations are in M. c Molar ratio of nitrosamine 2 to N2O.
d Fraction of the nitrosamine being produced by a pathway (presumed
to involve a radical cation) other than that involving the coproduction
of N2O. e NaOAc added (1.52 M).
Table 6. Kinetic Deuterium Isotope Effects on Amine Nitrosation
Rates and Rates of Product Formation
X
amine
nitrosamine
nitro
1a
2a
3b
pClc
4.38 ( 0.08
3.69 ( 0.09
2.39 ( 0.04
6.98 ( 0.25
3.84 ( 0.13
7.47 ( 0.12
0.43 ( 0.03
0.28 ( 0.01
0.42 ( 0.05
pCO2Etc
pCO2Etd
a In buffer. b In glacial acetic acid. c T ) 21 °C. d T ) 30 °C.
Deuterium Kinetic Isotope Effects. To gain more informa-
tion regarding the nature of the dealkylation step which gives
rise to the nitrosamine, we prepared the perdeuteriomethyl
isotopomers of 1 (X ) Cl, CO2Et) and measured their nitrosation
rates. Rate constants are reported in Table 3 and kH/kD values
are presented in Table 6, including kH/kD values for the amine
nitrosation, nitrosamine formation, and nitro compound forma-
tion. Rate constants for the disappearance of the respective
amines were determined as discussed above. The kH/kD values
obtained for the nitrosamine and the nitro compound were
estimated from plots of ln([amine]0 - [Z]) vs t, where [Z] )
[nitrosamine] or [nitro compound], respectively. Errors were
estimated from the standard deviation of the slope, obtained by
linear regression, for each kinetic run and were propagated in
kH/kD. All plots were well behaved except for several of the
nitro compounds. Isotope effects of less than one exhibited for
the formation of 3 (X ) Cl, CO2Et) clearly reflect the
mechanistic switching that occurs upon deuterium substitution.
The deuterated compounds show an enhanced rate of nitro
compound formation.
K4(k8 + k5)[Am][N2O3]
[Am]
dt
- d
)
)
-
[NO2 ]
-
KaK3K4(k8 + k5)[Am]T[NO2 ][H+]2
(5)
K2 ([H+] + Ka)
2
dissociation constant of HNO2, K3 is the equilibrium constant
for the formation of N2O3 from HNO2, and [Am]T is the total
amine concentration. With these simplifications in eq 4, eq 5,
which shows a first-order dependence on both [NO2-] and
[Am]T, is in agreement with our experimental observations.
Nevertheless, eq 4 is a useful tool in understanding the
mechanistic switching observed under some of the experimental
conditions used in this work.
We show the radical cation 5 to arise through reVersible
homolysis of the nitrosammonium ion 4, a process first
considered prior to the advent of modern mechanistic tools,36
but as stated before 5 could be produced by other processes.
The key step is the recombination of 5 with NO to regenerate
the nitrosammonium ion 4, a hypothesis that successfully
accounts for the mechanistic switching we observe and ties
numerous experimental observations together in a comprehen-
sible format. Our evidence in support of the reversible
homolysis comes from three sources: (1) Independently gener-
ated radical cations reacted with NO in H2O to give nitros-
amines. These transformations are well rationalized as occurring
through the nitrosammonium ion 4 as depicted in Scheme 2.
A deuterium isotope effect of 3.5 (80 °C) has been determined
previously for the nitrosative dealkylation of tribenzylamine by
means of intramolecular competition.3 While the values of kH/
kD measured for amine disappearance shown in Table 6 are in
this range, our other data, as well as the change in isotope effects
for the formation of the nitrosamine (X ) CO2Et) with solvent
change, suggest that steps involving C-H bond breakage may
not be the sole contributors to kobs. The kH/kD for nitrosamine
formation from 1 (X ) Cl) in buffer or 1 (X ) CO2Et) in HOAc
are significant, as expected for processes in which the breakage
of a C-H bond occurs through a well-centered transition state.
Discussion
-
Competitive hydrolysis of 4 provides the HNO2 or NO2
required for the conversion of the imminium ion to the
nitrosamine. (2) As shown in Table 2, addition of NO nearly
We propose that the data presented above argue for linked
competitive mechanisms of nitration and nitrosative dealkylation
by two different pathways: one, the classical mechanism,
involves NOH elimination (1 f 4 f 8 f 2, path A of Scheme
(36) Glazer, J.; Hughes, E. D.; Ingold, C. K.; James, A. T.; Jones, G.
T.; Roberts, E. J. Chem. Soc. 1950, 2657-2678.

Linked Mechanisms of Nitration and NitrosatiVe Dealkylation
J. Am. Chem. Soc., Vol. 120, No. 21, 1998 5199
doubled the nitrosamine yield compared to the control, while
the addition of O2 decreased the nitrosamine yield, but increased
the production of 3 (X ) Cl). Here, NO reacted with 5 to
generate 4 from which the nitrosamine is produced through the
classical path (A), or was oxidized to NO2 by O2, decreasing
the extent of the back reaction but increasing the amount of
NO2 and thereby the rate of nitro compound formation through
5 as is predicted by the rate equation. (3) Careful studies of
the mechanisms of both secondary arylamine nitrosation and
diazonium ion formation from some primary arylamines present
strong evidence for the reaction of NO with the corresponding
amine radical cations to produce nitrosammonium ions from
which subsequent transformations occur.37,38
Scheme 5
A priori, there are two reasonable mechanisms for the
conversion of the radical cation 5 to the imminium ion 8: (1)
deprotonation to a radical 7, which is rapidly oxidized to the
imminium ion 8, or (2) hydrogen atom abstraction to produce
the imminium ion directly. Our data, considered in the light
of literature precedent, are in best agreement with path 1. A
significant deuterium isotope effect is observed for nitrosamine
formation in two cases (Table 6, entries 1 and 3). In the first
case, the conditions are the same for which we observed 80%
of nitrosamine formation to be occurring by a pathway other
than NOH elimination, as determined from the N2O/nitrosamine
yield data (Table 5, entry 4). Under these conditions a kH/kD
) 7 is observed for the rate of nitrosamine formation, a value
consistent with deprotonation of the radical cation.39,40 The
resulting radical 7 will be oxidized rapidly39,40 in the nitrosation
media to the imminium ion 8. Dinnocenzo et al.39 and Parker
and Tilset40 have independently shown that the R-C-H of
aromatic amine radical cations are significantly acidic. Parker
and Tilset40 determined a pKa of 9 for 5 (X ) Cl), and
Dinnocenzo et al. have measured kH/kD values of 7-8 for the
deprotonation rates of similar radical cations,39 although such
effects are certain to be a function of both the ring substituent
and the base strength. Thus, literature precedent on the
properties of aromatic amine radical cations supports our
hypothesis that the radical cation is being deprotonated. Refer-
ence to Scheme 2 and eq 4, however, shows that isotope effects
on kobs occur through changes in the magnitudes of k7 and k8,
which appear in summed terms. Moreover, k7 appears in both
the numerator and denominator of eq 4, and changes in its
magnitude resulting from deuterium substitution will be modu-
lated by concentration changes. The lower values of kH/kD for
amine disappearance (Table 6) compared to nitrosamine forma-
tion are manifestations of these effects.
greatest. It is unclear, without further work, why and how the
fraction of the radical pathway depends so on the initial reactant
concentrations. A limitation of the current work is that the
nitrosamine/N2O ratios were observed, in most cases, over the
whole reaction and could be distorted by the occurrence of
mechanistic switching resulting from reactant concentration
changes during a given run.
This analysis also helps clarify how the reaction of Wurster’s
-
blue with NO2- in H2O gives nitrosamine. In this case, NO2
acts at first principally as a base to deprotonate the radical cation
5 (X ) N(CH3)2) and generate a radical 7 (X ) N(CH3)2),
(Scheme 5). Precedent exists for the oxidation of similar
radicals to imminium ions by radical cations.39 Thus, the
oxidation of 7 (X ) N(CH3)2) to the imminium ion 8 (X )
N(CH3)2) by 5 (X ) N(CH3)2) leads ultimately to the nitros-
amine 2 (X ) N(CH3)2) by known pathways.
Our data show that mechanistic switching occurs in response
factors other than changes in the magnitude of rate constants
arising from deuterium substitution. An increase in any
concentration found in the denominator of the second term of
eq 4 could reduce the overall magnitude of that term and the
participation of the radical cation pathways compared to NOH
elimination (k8, path A). As discussed above, addition of acetate
ion to glacial acetic acid significantly increases the proportion
of the nitrosamine which is produced through the radical cation
pathway from the nitrosation of 1 (X ) Cl). Inspection of data
from runs 1-8 of Table 5 for the nitrosation of 1 (X ) Cl) in
either glacial acetic acid or buffer shows that the extent of
nitrosamine formation through the radical cation pathway is
greater in the more basic buffer media where the concentration
of Y- ) AcO- is large. In glacial acetic acid, because of the
low H2O concentration, the concentrations of N2O3 and its
homolysis products, NO and NO2, will be higher, and will
increase the magnitude of the denominator of the second term
of eq 4, also reducing the participation of the radical cation
pathway to 2.
Additional support for these postulates is found in the data
of Table 5. Although the reaction of 1 (X ) Cl) was too rapid
for convenient kinetic studies and hence kH/kD determinations
in HOAc, comparison of runs 2 and 3 (Table 5) shows that the
percent of nitrosamine formation occurring by the radical
pathway is increased from 0 to 55% merely through the
inclusion of 0.5 M NaOAc in the reaction media. The data of
Table 5 also show that the fraction of the radical pathway to
the nitrosamine is increased by changing the solvent form HOAc
to buffer, where the concentration of the base necessary for
deprotonation reactions of 5 (X ) Cl) is necessarily greater.
Under these conditions (runs 4-7) the participation of the radical
cation pathway is greatest when the [NO2-]0/[amine]0 ratio is
It is also apparent that a change in the ring substitutent
significantly influences the nature of these transformations.
Amines containing the CO2Et substituent exhibited smaller
relative yields of nitro compound and nitrosamine formation
through a radical cation path. While a more complete under-
standing of the influence of ring substituents and other reaction
variables on the mechanistic pathways must await additional
experiments, a few preliminary conclusions can be drawn and
(37) Morkovnik, A. S.; Divaeva, L. N.; Okhlobystin, O. Y. J. Gen. Chem.
USSR 1989, 59, 2459-2471.
(38) Koshechko, V. G.; Inozemtsev, A. N. J. Gen. Chem. USSR 1983,
53, 1911-1914.
(39) Dinnocenzo, J. P.; Banach, T. E. J. Am. Chem. Soc. 1989, 111,
8646-8653.
(40) Parker, V. D.; Tilset, M. J. Am. Chem. Soc. 1991, 113, 8778-8781.

5200 J. Am. Chem. Soc., Vol. 120, No. 21, 1998
Loeppky et al.
Nicolet-FTIR 20DXB. UV data were obtained with a HP84 52A diode
array spectrophotometer. Elemental analyses were done by Desert
Analytics.
are well supported by the literature of amine radical cation
chemistry. The rate of amine radical cation formation in these
reactions will parallel their oxidation potential: X ) (CH3)2N
. Cl > CO2Et.40 On the other hand, the C-H acidity (and
probable deprotonation rates) of their resulting radical cations
will exhibit the opposite substituent effect. Thus, k5 is expected
to be significantly less when X ) CO2Et and the path involving
NOH loss may predominate. When a radical cation is produced
from an amine bearing the CO2Et substituent, however, it is
likely to be rapidly deprotonated, resulting in nitrosamine
formation at the expense of the nitro compound formation (see
run 12 Table 5).
Materials. Solvents used for HPLC were of Fisher Optima grade
and were filtered before use. Solvents used for flash chromatography
and moisture sensitive reactions were purchased from Fisher Scientific
and purified. Tetrahydrofuran was dried over calcium hydride and
distilled from lithium aluminum hydride. Methylene chloride was
washed with concentrated sulfuric acid, water, and saturated potassium
carbonate, predried over calcium sulfate, and distilled over calcium
hydride or phosphorus pentoxide. Acetonitrile was distilled over
calcium hydride. Hexane was distilled over anhydrous sodium sulfate.
All solvents were stored over molecular sieves. NO, N2O, and NO2
were purchased from Matheson. NO was purified by passing it through
solid KOH and anhydrous CaCl2. All other common reagents were
purified by normal procedures, when required.
Conclusions
Our investigation of an apparently simple transformation,
known since the earliest days of organic chemistry, shows that
the mechanisms of nitrosamine formation and nitro compound
generation from a N,N-dialkylaromatic amine and nitrous acid
are mechanistically linked. Much of the chemistry of nitrosating
agents with N-alkyl tertiary aromatic amines presented in the
literature, including that of Colona et al.,22 is well rationalized
by our mechanistic scaffold in which the reversible formation
aromatic amine radical cation from a nitrosammonium ion plays
a key role. Mechanistic changes, and as a result product
distributions, are influenced by subtle changes in concentrations
and reactant structure. From the perspective of human health
and safety, it is significant that nitrosamines arise rapidly, at
low concentrations from the reaction of acidic nitrite with N,N-
dialkylaromatic amines. The greater nitrosation reactivity of
these amines in comparison to trialkylamines is likely a
manifestation of both their lower basicity and lower oxidation
potentials, but the relative significance of the latter factor
remains to be determined. Few studies of the carcinogenicity
of aromatic nitrosamines have been conducted. Methylphe-
nylnitrosamine is a potent carcinogen.42 It is likely that
nitrosamine formation rates and nitrosamine carcinogenicity
levels are significantly influenced by the nature of both the ring
and N-substituents. Until more work is done to clarify these
determinants, steps should be taken to prevent the formation of
nitrosamines in commercial intermediates or formulations
containing N,N-dialkylaromatic amines.
Synthesis of Known Amines and Nitrosamines. Unless noted
specifically below, known compounds were prepared by published
procedures or slight modifications thereof. NMR and other charac-
teristic data are given in the Supporting Information. Formanilides
were reduced to the corresponding N-methylanilines by the method
similar to that of Brown and Heim.42 Compounds: 4-Chloroforma-
nilide,43 N-methyl-4-chloroaniline,44 N,N′-diformyl-p-phenylenedi-
amine,45 N-nitroso-N-methyl-4-chloroaniline46 (2 X ) Cl), N,N′-
dinitroso-N,N′-dimethyl-p-phenylenediamine47 (13), N,N-dimethyl-4-
chloroaniline48 (prepared by a modification of the method of Borsch
and Hassid49), N,N,N′,N′-tetramethyl-2-nitro-p-phenylenediamine50 (3,
X ) N(CH3)2), and 4-chloro-2-nitrodimethylaniline51 (3, X ) Cl).
Nitrosation of N,N,N′-Trimethyl-p-phenylenediamine and Sepa-
ration of the Products by Chromatography. According to Loeppky
et al.10 isopropyl nitrite was added dropwise to the stirring N,N,N′-
trimethyl-p-phenylenediamine (0.45 g, 3 mmol) at 0-5 °C. The
reaction was allowed to warm to 20 °C. After 1 h, the excess isopropyl
nitrite was removed in vacuo and the resulting solid was subjected to
flash column chromatography (7:3 hexane:ethyl acetate over silica gel).
Two major fractions were collected. Fraction 1: N′-nitroso-N,N,N′-
trimethyl-p-phenylenediamine (2, X ) N(CH3)2); mp 95-98 °C (lit.52
1
mp 98-99 °C); H NMR (500 MHz, CDCl3) δ 7.36-7.34 (d, 2H),
6.77-6.75 (d, 2H), 3.42 (s, 3H), 2.99 (s, 6H). 13C NMR (125 MHz,
CDCl3) δ 149.86, 131.93, 121.04, 112.42, 40.54, 32.44. Fraction 2:
N′-Nitroso-N,N,N′-trimethyl-3-nitro-p-phenylenediamine (12); mp 84-
1
86 °C (lit.53 mp 87 °C); H NMR (500 MHz, CDCl3) δ 7.88 (d, 1H),
7.68 (dd, 1H), 7.14 (d, 1H), 3.43 (s, 3H); 13C NMR (125 MHz, CDCl3)
δ 145.10, 138.03, 132.55, 124.39, 118.99, 117.18, 42.39, 31.43.
N,N-(D6)-Dimethyl-4-chloroaniline (1D6, X ) Cl). To a cooled
(-78 °C), stirred dry THF (50 mL) solution of 4-chloroaniline (2.0 g,
16 mmol) under N2 was added, dropwise, a 2.5 M solution of n-BuLi
(13 mL, 33 mmol) in THF. After the mixture was stirred for 30 min,
a solution of CD3I (1.5 mL, 23.6 mmol) in THF (2 mL) was added,
stirring was then continued for 3 h at -78 °C and the mixture was
allowed to warm to room temperature and then stirred for an additional
12 h. Ethanol (50 mL) was added and the resulting mixture was
extracted with ether (4 × 30 mL). The combined extracts were washed
with brine, dried over Na2SO4, filtered, and concentrated in vacuo. The
crude product (2.4 g) was subjected to column chromatography on silica
gel (16:1 hexane:ether) to give 0.51 g of the deuterium compound in
20% yield. Recrystallization from ethanol yielded 0.22 g of a white
solid; mp 32-34 °C. 1H NMR (300 MHz, CDCl3) δ 7.18-6.65 (dd,
4H). 13C NMR (125 MHz, CDCl3) δ 149.20, 128.75, 121.26, 113.50,
Experimental Section
Instrumentation. Melting points were obtained on a Thomas-
Hoover capillary tube apparatus and are not corrected. 1H, 13C NMR,
and 15N NMR were recorded on either a Joel FX-90Q (90 MHz for
1
1H), Nicolet NT (300 MHz for H and 75 MHz for 13C), or Bruker
AMX 500 (500 MHz for 1H, 125 MHz for 13C, and 51 MHz for 15N).
Gas chromatography was done on a Hewlett-Packard 5890 series gas
chromatograph equipped with a Flame ionization or thermal energy
analyzer (TEA model 502) detectors with a SPB-1 (0.25 mm × 30 m)
fused silica capillary column. GC-mass spectral analyses were
performed on a Hewlett-Packard 5970 mass selective detector with a
20 M SPB5 capillary column and controlled with a Hewlett-Packard
59970 Chemstation computer. For analyses of nitrous oxide a Gowmac
series 550 thermal conductivity detector (TCD) equipped with a 9 ft.
× 1/8 in. stainless steel column packed with Poropak Q (80-100 mesh)
was used. High-pressure liquid chromatography (HPLC) was per-
formed with a Waters Maxima 820 system controller, a Waters model
490 programmable multiwavelength detector, a Waters model 710B
Wisp autosampler, and two Waters model 510 pumps. The columns
used for HPLC analyses were Zorbax Rx-C8 (4.6 mm × 25 cm) and
Zorbax ODS-C18 (4.6 mm × 25 cm). IR spectra were recorded on
(43) Chattaway, F. D.; Orton, K. J. P.; Huntley, W. H. Chem. Ber. 1899,
32, 3636.
(44) Groggins, P. H.; Stirton, A. J. Ind. Eng. Chem. 1937, 29, 1353.
(45) Wundt, E. Chem. Ber. 1878, 11, 828.
(46) Schmidt, E.; Fischer, H. Chem. Ber. 1920, 53, 1529.
(47) Willstatter, R.; Pfannensteil, A. Chem. Ber. 1905, 38, 2249.
(48) Ley, H.; Pfeller, G. Chem. Ber. 1921, 54, 1414.
(49) Borch, R. F.; Hassid, A. I. J. Org. Chem. 1972, 37, 1637.
(50) Corbet, J. F.; Fooks, A. G. J. Chem. Soc. C 1967 1136.
(51) Clemo, G. R.; Smith, J. M. J. Chem. Soc. 1928, 2414.
(52) Wurster, C.; Schobig, E. Chem. Ber. 1879, 12, 1809.
(53) Michaelis, L.; Granick, S. J. Am. Chem. Soc. 1943, 65, 1747.
(41) Druckrey, H.; Preussmann, R.; Ivankovic, S.; Schmaehl, D. Z.
Krebsforsch. 1967, 69, 103-201.
(42) Brown, H. C.; Heim, P. J. J. Org. Chem. 1973, 38, 912.

Linked Mechanisms of Nitration and NitrosatiVe Dealkylation
J. Am. Chem. Soc., Vol. 120, No. 21, 1998 5201
39.60 (m, J ) 20.5 Hz). HRMS (EI) calcd for C8H4D6N35Cl 161.0872,
found 161.0873.
combined extracts were washed with brine, dried over Na2SO4, filtered,
and analyzed by GC-FID and HPLC utilizing nitrobenzene as an internal
standard.
Ethyl 4-((D6)-Dimethylamino)benzoate (1, X ) CO2Et; D6). A
dry 100 mL, three-necked round-bottomed flask equipped with a
magnetic stirrer and N2 balloon was charged with p-carboethoxyaniline
(0.44 g, 2.6 mmol) in CH3CN (25 mL). Deuterated formaldehyde (D2O,
98%, 20% solution in D2O, 6.0 mL, 35.3 mmol) and deuterated sodium
cyanoborohydride (NaBD3CN) (0.51 g, 7.7 mmol) were added in
succession to the stirring solution of the amine. After the mixture was
stirred for 10 min, glacial HOAc (2 mL) was added dropwise, then
stirring was continued for 2 h followed by the addition of HOAc (1
mL) and continued stirring for 13 h. The solvent was removed in vacuo,
and 10% NaOH was added. The resulting aqueous mixture was
extracted with ether (3 × 25 mL), and the ether extracts were washed
with brine, dried over Na2SO4, and concentrated in vacuo to yield 418.3
mg (80.8% yield) of a white solid which was recrystallized from
ethanol. 1H NMR (500 MHz, CDCl3) δ 7.90-6.63 (dd, 4H), 4.34 (q,
2H), 1.36 (t, 3H). 13C NMR (125 MHz, CDCl3) δ 166.99, 153.24,
131.14, 117.21, 110.54, 60.04, 39.18 (m, 20.9 Hz), 14.43. HRMS
(EI): calcd for C11H9D6NO2 199.1471, found 199.1478.
N-Methyl-N-cyclopropyl-4-chloroaniline (11). According to the
two-step procedure described by Kang and Kim,54 in a dry flask under
N2, 2.0 g (14.12 mmol) of N-methyl-4-chloroaniline, 3.5 g (21.18 mmol)
of 1-ethoxy-1-bromocyclopropane, and 2.14 g (21.18 mmol) of
triethylamine were all dissolved in 7 mL of dry CH2Cl2. The mixture
was allowed to reflux while being stirred for 46 h. After dilution with
30 mL of CH2Cl2, the mixture was rinsed successively with 2 × 20
mL of H2O and 20 mL of brine. The aqueous washes were back
extracted with 20 mL of CH2Cl2. The two organic layers were
combined, dried over anhydrous MgSO4, filteredm and tripped of
solvent. The resulting residue was purified by column chromatography
on silica gel (5% ethyl acetate in hexanes) to give the N-methyl-N-(1-
ethoxycyclopropyl)-4-chloroaniline as a colorless oil in 67% yield (74%
corrected), which was used in the next step without further purification.
1H NMR (CDCl3, 500 MHz) δ 7.18 (dt, 2H, J ) 3.3, 9.2 Hz), 6.91 (dt,
2H, J ) 3.3, 9.2 Hz), 3.51 (q, 2H, J ) 7.0 Hz), 3.06 (s, 3H), 1.22 (br.
s, 2H), 1.11 (t, 3H, J ) 7.0 Hz), 0.90 (br. d, 2H, J ) 1.9 Hz); 13C
NMR (CDCl3, 125 MHz) δ 146.2, 128.48, 122.49, 114.69, 75.09, 62.24,
37.94, 16.42, 15.42.
(b) Nitrosation of 1 under No Headspace Conditions. In a typical
procedure, a series of 12 mL vials containing a mini stir bar were
charged with 10 mL of a standard stock buffer solution (pH ) 3.8,
1.66 M NaOAC in 60% HOAc) containing the amine 1 (X ) Cl or
CO2Et) and nitrobenzene as an internal standard. The vials were crimp
sealed with a Teflon-backed rubber seal and equilibrated at 21 °C for
20 min prior to the addition of 2 mL of a standard solution of aqueous
NaNO2. In the case of 1 (X ) Cl) the mixture was 0.23 mM in amine
and 2.3 mM in NO2- (see Table 1). The reaction mixture was stirred
for 6 min and the entire contents of each vial quenched by addition of
20% K2CO3 solution (7 mL). The aqueous layer was extracted with
ether (5 × 1 mL) and analyzed by HPLC.
(c) Nitrosation of 1 (X ) Cl) under Headspace Conditions. A
procedure identical to that described above was utilized except that
the headspace above the liquid was covered with N2 which permitted
ample space for gas-liquid equilibration.
(d) Nitrosation of 1 (X ) Cl) in the Presence of Added NO. A
100 mL, three-necked flask was charged with 1 (X ) Cl) (0.50 g, 0.32
mmol) in HOAc (10 mL) and attached to a vacuum manifold. To a
10 mL, two-necked flask was added 1 mL of 0.37 M NaNO2 solution
in water. The contents of both flasks were subjected to 4 freeze-
thaw degassing cycles. After bringing the solutions to room temperature
purified NO gas was introduced into the vacuum manifold until
atmospheric pressure was attained. The equilibration was repeated.
The nitrite was transferred into the amine solution with a cannula under
NO gas and NO gas was introduced until a pressure of 1 atm was
reached. The reaction mixture was stirred for 40 min and the excess
NO was removed. The reaction flask was detached from the manifold
and the quantitation standard nitrobenzene (0.018 g, 0.15 mmol) was
added in ether. The reaction was quenched by addition of 30% K2-
CO3 solution (125 mL) and extracted with ether (3 × 40 mL), dried
over Na2SO4, filtered, concentrated in vacuo, and subjected to
chromatographic analysis. The products of the reaction of 2 and 3
were identified by their mass spectral data and comparison of GC
retention times with those of authentic standards. HPLC analysis was
used for determination of product yields.
(e) Nitrosation of 1 (X ) Cl) under N2. The nitrosation reaction
was done exactly as above except it was done under a N2 atmosphere.
Nitrosation of 1 (X ) Cl) in the Presence of Added O2. The
reaction was carried out as described for that carried out in an NO
atmosphere except that NO was replaced with O2.
In a dry flask equipped with a stirring bar and a rubber septa, 671
mg (17.73 mmol) of NaBH4 were suspended in 15 mL of dry THF.
The solution was cooled to 0 °C and 2.66 g (17.73 mmol) of BF3‚Et2O
was added via a syringe under N2 atmosphere. The mixture was stirred
at 0 °C for 45 min, and 2.0 g (8.86 mmol) of N-methyl-N-(1-
ethoxycyclopropyl)-4-chloroaniline dissolved in 3 mL of dry THF was
added dropwise via a gastight syringe. The mixture was allowed to
warm slowly to room temperature and stirring was continued for an
additional 6 h. The reaction was carefully quenched with water and
extracted in 40 mL of diethyl ether followed by washing with 3 × 20
mL of H2O and 1 × 20 mL of brine. The organic layer was dried
(MgSO4), filtered, and concentrated. Column chromatographic purifica-
tion (3% ethyl acetate in hexanes) afforded the desired product as a
colorless oil in 89% yield. 1H NMR (CDCl3, 500 MHz) δ 7.17 (dt,
2H, J ) 3.3, 9.2 Hz), 6.9 (dt, 2H, J ) 3.3, 9.2 Hz), 2.94 (s, 3H), 2.53
(m, 1H), 0.81 (m, 2H), 0.60 (m, 2H); 13C NMR (CDCl3, 125 MHz) δ
149.39, 128.54, 122.19, 114.80, 39.07, 33.30, 9.07; IR (neat) 3086 w,
3045 w, 3008 w, 2939 w, 2878 w, 2817 w, 1599 m, 1497 s, 1456 w,
1360 m, 1336 m, 1308 w, 1306 w, 1267 w, 1236 w, 1184 w, 1116 m,
1097 w, 1025 w, 998 w, 963 w, 814 s, 734 m cm-1. Anal. Calcd for
C10H12ClN: C, 66.11; H, 6.66; N, 7.71. Found: C, 66.29; H, 6.52; N,
7.58
Identification and Quantification of N2O and Nitrosamines as
Products in the Nitrosation of 1 (X ) Cl or CO2Et). Calibration:
A calibration curve was constructed by determining the amount of N2O
produced in the headspace from the introduction of 0.1 mL of a 0.39
M solution of NaN3, 0.8 mL of H2O, and 0.1 mL of a 1.2 M solution
of NaNO2 into 20 mL contained in a 100 mL, three-necked flask fitted
with a stirrer, a control valve, and a leveling bulb filled with water.
After 15-25 min, and adjustment to atmospheric pressure, the
headspace above the solution was sampled by use of a 5 mL gastight
syringe and analyzed by GC-TCD. The above procedure was repeated
at a minimum of five different concentrations of NaN3. A plot of the
concentration of NaN3 (N2O) against the peak height of N2O was linear.
Calibration was performed each time before conducting the nitrosation
reaction. Analyses: The amines were subjected to nitrosation under
typical nitrosating conditions, which involves their dissolution in an
acidic solvent and addition of saturated NaNO2. The reaction was
allowed to proceed typically for 15-30 min. The gaseous species
evolved in the reaction and contained in the headspace above the
solution were sampled with the help of a 5 mL gastight syringe and
analyzed for N2O by GC-TCD in the manner described above.
Immediately after the gas sampling, the reaction mixture was worked
up by addition of a 30% K2CO3 solution (125 mL) followed by
extraction with ether (4 × 25 mL), dried over Na2SO4, and analyzed
by HPLC. Nitrosamines were quantitated by HPLC by using our
standard methodology. All products were identified by comparison
of retention times with authentic standards.
Nitrosation Reactions. (a) Nitrosation of N,N-Dimethyl-4-chlo-
roaniline (1, X ) Cl). The general procedure for the nitrosation of 1
involved its dissolution in an acidic solvent, either glacial HOAc or
60% aqueous HOAc containing sodium acetate (1.66 M), followed by
dropwise addition of a three molar excess of aqueous NaNO2 solution.
The reaction was allowed to stir for 30 min, neutralized with saturated
K2CO3 solution, and extracted with ether (4 × 25 mL), and the
(54) Kang, J.; Kim, K. S. J. Chem. Soc., Chem. Commun. 1987, 897-
898.
(f) Nitrosation of N-Cyclopropyl-N-methyl-4-chloroaniline (11,

5202 J. Am. Chem. Soc., Vol. 120, No. 21, 1998
Loeppky et al.
X ) Cl). With use of the standard nitrosation procedure, 500 mg (2.75
mmol) of N-cyclopropyl-N-methyl-4-chloroaniline was nitrosated in 6
mL of glacial HOAc with 380 mg (5.51 mmol) of NaNO2 dissolved in
2 mL of H2O. The resulting reddish residue was purified by flash
column chromatography (10% ethyl acetate in hexane) to give
N-methyl-N-nitroso-4-chloroaniline as a yellow crystalline solid in 61%
yield as the sole benzenoid product (97% by HPLC analysis before
purification).
Kinetic Studies of Nitrosation. In a typical procedure, a 12 mL
reaction vial containing a mini stir bar was charged with 10 mL of
standard stock solution consisting of 0.214 mM N,N-dimethyl-4-
chloroaniline (1, X ) Cl) and 0.235 mM nitrobenzene in buffer of pH
3.8-3.9 (27.2 g of NaOAc in 200 mL of 60% aqueous HOAc) and
stabilized in a constant-temperature bath at 21 °C before the addition
of 1.8 mL of 0.0275 M aqueous NaNO2 solution. The reaction mixture
was stirred, and at regular intervals 1 mL aliquots were withdrawn and
quenched with 30% K2CO3 (5 mL) solution. The aqueous layer was
extracted with ether (5 × 1 mL) and analyzed by HPLC. The
concentrations of the various components were determined with the
internal standard method.
a 3.3 M solution of NaNO2. The reaction in one vial was allowed to
proceed for 5 min and the other 40 min. At the end of the reaction
time, phenetole (0.018 g, 0.14 mmol) was added and the reaction
mixture was quenched by addition of a 30% K2CO3 solution (10 mL)
and a few drops of 10% FeSO4. The aqueous layer was extracted with
CH2Cl2 (5 × 5 mL), dried over Na2SO4, filtered, and subjected to
chromatographic analysis. The products of the reaction were identified
by their mass spectral data and comparison of GC retention times with
those of authentic standards. HPLC analysis was used to determine
product yields.
Reaction of Wurster’s Blue Perchlorate with Nitric Oxide Gas
(NO). A 50 mL 0.1 mM solution of Wurster’s blue perchlorate was
placed in a 100 mL, three-necked flask. Two necks of the flask were
closed with a subaseal stopper. After attachment to the vacuum
manifold, the contents of the flask were degassed by several freeze-
thaw degassing cycles. NO gas, which was purified by passing through
solid KOH and anhydrous CaCl2, was introduced into the evacuated
manifold until atmospheric pressure was attained. The valve to the
reaction flask was closed and the excess gas was removed. The reaction
mixture was allowed to warm to room temperature and was stirred for
2 h. The unreacted NO was removed and N2 was allowed into the
system until atmospheric pressure was attained. Phenetole (0.04 g,
0.44 mmol), which served as an internal standard, was added and the
reaction mixture quenched by addition of 30% K2CO3 solution (10 mL)
and 10% FeSO4 solution (5 mL). The aqueous layer was extracted
with CH2Cl2 (5 × 20 mL), dried over Na2SO4, filtered, concentrated
in vacuo, and subjected to chromatographic analysis. The products of
the reaction were identified by their mass spectral data and comparison
of GC retention times with those of authentic standards. HPLC analysis
was used to determine product yields.
The Reaction of N,N,N′,N′-Tetramethylphenylenediamine with
NO in Acetonitrile. A 100 mL, three-necked flask was charged with
the amine (0.021 g, 0.13 mmol), phenetole (0.042 g, 0.35 mmol), and
anhydrous CH3CN (50 mL). The solution was subjected to reaction
with NO under conditions identical with those described above for
Wurster’s blue perchlorate. The workup procedure was the same,
except that no FeSO4 was added. Unreacted starting material was
recovered after workup.
Reaction of Wurster’s Blue Perchlorate with Nitrogen Dioxide
(NO2). To a stirred CH3CN (10 mL) solution of Wurster’s blue
perchlorate (0.027 g, 0.10 mmol) was added liquid NO2 (0.01 g, 0.22
mmol) by use of a syringe. The reaction mixture was stirred at room
temperature for 10 min. The reaction mixture was quenched by the
addition of water and a few drops of 10% FeSO4. The aqueous layer
was extracted with CH2Cl2 (4 × 5 mL), dried over Na2SO4, filtered,
and subjected to chromatographic analysis. The products of the reaction
were the nitrosamine 13 and the nitrosonitro compound 12, which were
identified by their mass spectral data and comparison of their HPLC
retention time with those of authentic standards.
Measurement of Rate Constants kH and kD To Determine Isotope
Effects in the Nitrosation of Ethyl 4-(Dimethylamino)benzoate (1,
X ) CO2Et). A small reaction vial was charged with ethyl 4-(di-
methylamino)benzoate (0.016 g, 0.083 mmol) and the internal standard
ethyl 4-(benzylnitrosamino)benzoate (0.0075 g, 0.026 mmol) in 10 mL
of buffer of pH 3.8-3.9 (1.66 M NaOAc in 60% aqueous HOAc).
The reaction vial was stabilized in a constant-temperature bath at 30
°C before the addition of 2 mL of a 0.13 M aqueous NaNO2 solution.
The reaction mixture was stirred and at regular time intervals, 1 mL
aliquots were withdrawn and quenched with 30% K2CO3 (5 mL)
solution. The aqueous layer was extracted with ether (5 × 1 mL) and
analyzed by HPLC. Similarly, the D6 isotopomer was nitrosated under
identical conditions and the rate constant determined.
1H NMR Studies in Preparation for 15N CIDNP. The nitrosation
1
of 1 (X ) Cl) was followed by H NMR spectroscopy at 25 °C with
use of a Bruker AMX 500 spectrometer. In a typical procedure, 0.5
mL of a D4-acetic acid solution of 0.07 M 1 (X ) Cl) and 0.017 M
nitromethane (internal standard) was introduced into a 5 mm spin tube
and 0.35 mL of a 0.71 M solution of NaNO2 in D2O was added. The
extent of the reaction was calculated from the percentage of unreacted
starting material. The concentration of the starting amine was
determined from the integration value of the signal for the N,N-dimethyl
protons of the amine 1 and the methyl proton of the internal standard.
15N NMR CIDNP Studies. The studies involving 15N NMR
spectroscopy were carried out at 25 °C with a Bruker AMX 500
spectrometer. The concentrations of the reagents were the same as
1
those used in the corresponding H NMR experiment except that the
internal standard used was 15N labeled nitrobenzene (0.029 g, 0.23
mmol). The spectra were measured at various intervals, using a pulse
angle of 30 and a pulse delay of 0.2 s. The identity and the 15N NMR
shifts of the various peaks obtained during the nitrosation reaction are
as follows: δ -5.42, “NO+”; -9.31, nitrobenzene; -8.03, N,N-
dimethyl-4-chloro-2-nitroaniline; 160.16, N-nitroso-N-methyl-4-chlo-
roaniline; 201.37, NO2-. Shifts are referenced to CH3NO2 as an external
standard (δ ) 0).
Reaction of N,N,N′,N′-Tetramethylphenylenediamine with NO2.
To a stirred CH3CN (10 mL) or CH2Cl2 (10 mL) solution of the amine
(0.033 g, 0.20 mmol) was added liquid NO2 (0.02 g, 0.44 mmol), and
the reaction was stirred for 30 min. At the end of 30 min, the reaction
mixture was analyzed by HPLC and GC-MS. The products for the
reaction in CH3CN were the nitrosamine 2 (X ) N(CH3)2) and the
di-N-nitroso compound 13.
Nitrosation of N,N,N′,N′-Tetramethyl-p-phenylenediamine. To
a stirred, glacial HOAc (10 mL) solution of the amine (0.057 g, 0.35
mmol) and phenetole (internal standard) (0.099 g, 0.81 mmol) was
added a 1.4 M solution of NaNO2 (1 mL, 1.4 mmol) in water. The
reaction mixture was stirred for 1 h prior to quenching the solution
with 30% K2CO3 (5 mL) solution. The aqueous solution was extracted
with CH2Cl2 (3 × 25 mL), dried over Na2SO4, filtered, and subjected
to chromatographic analysis. The initial identity of the products of
nitrosation was established from their mass spectral data and their
presence confirmed by comparison of their GC retention time with those
of authentic standards. HPLC analysis was used to determine product
yields.
Acknowledgment. The support of this research by a grant
(R37 CA 26914) from the National Cancer Institute is gratefully
acknowledged.
Supporting Information Available: Plots of (1) Concentra-
tions of 1-3 (X ) Cl) from the nitrosation of 1 (X ) Cl), (2)
first-order treatment of data for 1 (X ) Cl) from the same
experiment, (3) log (kobs) vs log ([NO2-]) for nitrosation of 1
(X ) Cl), (4) the ratio of 3/2 (X ) Cl) vs [NO2-], and (5)
NMR and physical data for all known compounds used in this
work (2 pages, print/PDF). See any current masthead page for
ordering information and Web access instructions.
Reaction of Wurster’s Blue Perchlorate with Sodium Nitrite in
Water. A stock solution (10 mL each, 2.6 mM) of Wurster’s blue
perchlorate53 was added to two separate vials containing a mini-stir
bar. To the stirring solutions of the perchlorate was added 0.04 mL of
JA9732744