Nitrosation and denitrosation of substituted N-methylbenzenesulfonamides. Evidence of an imbalanced concerted mechanism

Article abstract of DOI:10.1039/a801275j

The kinetics of the nitrosation reaction of several substituted sulfonamides and of the denitrosation of the resulting products have been studied. The denitrosation rate is first-order with respect to both the nitroso compound and acid concentration and no effect of added nucleophiles was observed. The denitrosation reaction is general-acid catalysed, with a Bronsted parameter α_d, of 0.7, which is independent of the substituents on the aromatic ring. Kinetic solvent isotope effects range from k^H3O+_d/ k^D3O+_d = 1.20 ± 0.05 to 2.04 ± 0.06 for denitrosation by L₃O⁺ and from k^AH_d/k^AD_d = 1.5 ± 0.2 to 2.3 ± 0.3 for denitrosation by dichloroacetic acid, which suggest that a rate-determining proton transfer is involved in this reaction. For nitrosation reaction, the absence of catalysis by nucleophilic anions, the observed general-base catalysis (β_NO = 0.3) and the substituent effects suggest a concerted nitrosation-denitrosation process. The Leffler parameters obtained for N ... H bond formation (α_nuc = 0.7) as well as for N ... N=O bond breaking (α_lg = 0.17) are in favour of an imbalance in the transition state (α_imbalance = 0.53) with the development of a positive charge on the nitrogen adjacent to the nitroso group.

Full text of DOI:10.1039/a801275j

View Article Online / Journal Homepage / Table of Contents for this issue
Nitrosation and denitrosation of substituted N-methylbenzene-
sulfonamides. Evidence of an imbalanced concerted mechanism
Luis García-Río,a J. Ramon Leis,*,a José A. Moreiraa and Fatima Norbertob,c
a Departamento de Química Física, Facultade de Química, Universidade de Santiago,
15706 Santiago, Spain
b Departamento de Química e Bioquímica, Faculdade de Ciências, Universidade de Lisboa,
1700 Lisboa, Portugal
c CECF, Faculdade de Farmácia, Universidade de Lisboa, Av. Forças Armadas, 1699, Lisboa,
Portugal
The kinetics of the nitrosation reaction of several substituted sulfonamides and of the denitrosation
of the resulting products have been studied. The denitrosation rate is first-order with respect to both
the nitroso compound and acid concentration and no effect of added nucleophiles was observed.
The denitrosation reaction is general-acid catalysed, with a Brønsted parameter, ád, of 0.7, which is
ϩ
independent of the substituents on the aromatic ring. Kinetic solvent isotope effects range from kdH O
/
3
ϩ
kdD O ؍ 1.20 0.05 to 2.04 0.06 for denitrosation by L3Oϩ and from kdAH/kdAD ؍ 1.5 0.2 to 2.3 0.3
for denitrosation by dichloroacetic acid, which suggest that a rate-determining proton transfer is
involved in this reaction. For nitrosation reaction, the absence of catalysis by nucleophilic anions, the
observed general-base catalysis (âNO ؍ 0.3) and the substituent effects suggest a concerted nitrosation–
denitrosation process. The Leffler parameters obtained for N ؒ ؒ ؒ H bond formation (ánuc ؍ 0.7) as
3
well as for N ؒ ؒ ؒ N᎐O bond breaking (á ؍ 0.17) are in favour of an imbalance in the transition state
᎐
lg
(áimbalance ؍ 0.53) with the development of a positive charge on the nitrogen adjacent to the nitroso group.
their nitroso group at moderate acidities7,8 where traditional
Introduction
nitrosating agents (NOϩ or alkyl nitrites) are either inefficient
or unstable. In spite of the large interest in nitrososulfonamides
themselves and their being included in one of the most import-
ant classes of mutagenic and carcinogenic compounds,9,10 there
is a clear lack of mechanistic studies of the nitrosation of sulf-
onamides in the literature.
By far the best-known and most widely studied nitrosation
reactions are N-nitrosations of amines and related compounds.
N-Nitrosations occur widely in organic chemistry in many
standard synthetic pathways, and a number (including diazo-
tisation and azo dye formation) have large-scale industrial
applications. Because of their importance, such reactions have
been also much studied mechanistically, and in general are now
well understood.1
The mechanisms of nitrosation of amides and ureas in an
acidic medium have been investigated in recent years2–5 and a
large number of differences have been found between nitros-
ation of these compounds and amines. In the case of amines the
attack of the nitrosating agent on the free amine is rate deter-
mining, while for amides this first step is fast, the slow step
being a proton transfer from an intermediate to the reaction
medium. In the latter case the reaction seems to occur initially
on the oxygen atom, and a subsequent rearrangement leads to
the thermodynamically more stable N-nitrosoamide (Scheme 1).
Similar behaviour would certainly be expected for denitros-
ation of both sulfonamides and carboxamides, because of their
quite similar structure and because the only study known until
now demonstrates the same kinetic characteristics for both.11
Nevertheless, N-nitrososulfonamides are the only N-nitroso
compounds able to transfer their nitroso group to other species
acting as nucleophiles, which makes them an important family
of transnitrosating agents7,8 with efficiency similar to that of
alkyl nitrites. This peculiar characteristic of nitrososulfon-
amides may be related with their nitrosation–denitrosation
mechanism.
In this paper we try to establish a direct relationship between
the denitrosation mechanism and transnitrosation ability of a
given compound. The experimental results presented here on
nitrosation–denitrosation of a series of 4-substituted N-methyl-
N-nitrosobenzenesulfonamides suggest the existence of a third
route for decomposition of some N-nitroso compounds. This
pathway involves a concerted protonation and fission of the
R
R
NO
H
R
–H+
N+
+
N
H
NO+
N
NO
R′
R′
R′
O
N
N
N ؒ ؒ ؒ N᎐O bond. This mechanism is quite similar to that
O
O
O
᎐
O
H
O
–H+
H
NO+
reported for the main family of transnitrosating agents, alkyl
nitrites, for which denitrosation involves a rate-determining
protonation with a concerted mechanism in which proton
+
R
NO
R
N+
R
N+
R
N
R′
R′
R′
R′
transfer and O ؒ ؒ ؒ N᎐O bond breaking occur simultaneously, in
᎐
Scheme 1
a slightly imbalanced transition state (Scheme 2).12
The reason for this seems to be related to the much lower bas-
icity of amides, compared to amines.
Nitrosation of secondary sulfonamides gives rise to stable N-
nitroso derivatives.1 Some nitrosulfonamides are used to gener-
ate diazomethane6 and as nitrosating agents able to transfer
δ–
O
‡
H
NO
ROH + NO+
RONO + H+
R
Scheme 2
J. Chem. Soc., Perkin Trans. 2, 1998
1613

View Article Online
Table 1 Effect of added nucleophiles (BrϪ and SCNϪ) on kobs of acid
denitrosation of N-methyl-N-nitrosobenzenesulfonamides II and IIIa
Experimental
N-Methylbenzenesulfonamides have been synthesized by reac-
tion of the corresponding benzenesulfonyl chlorides with an
excess of methylamine in water.13 The product is extracted
with dichloromethane and washed with a solution of sodium
hydrogen carbonate and with water. N-Methyl-4-methyl-
benzenesulfonamide and its nitroso derivative II were supplied
by Ega-chemie and Merck, respectively.
Nucleophile
kobs/10Ϫ4 sϪ1
4-CH3 (II)
[BrϪ]/
[SCNϪ]/
4-Cl (III)
—
—
0.16
0.32
0.50
(5.1 0.3)
(4.7 0.3)
(5.1 0.3)
(5.4 0.3)
(5.3 0.2)
(5.4 0.3)
(4.3 0.2)
(4.3 0.2)
(4.4 0.2)
(4.9 0.3)
(5.7 0.3)
(6.4 0.3)
0.057
0.114
0.142
O
S
I: X = 4-CH3O
II: X = 4-CH3
III: X = 4-Cl
N
X
N
O
IV: X = 4-NO2
O
a [HClO4] = 1.39 × 10Ϫ2 ; T = 25 ЊC.
N-Methyl-N-nitrososulfonamides I, III and IV were prepared
using a biphasic water–dichloromethane mixture. The aqueous
phase containing sodium nitrite and the organic phase contain-
ing the parent sulfonamide were mixed together and then con-
centrated perchloric acid (5 ) was slowly added. The mixture
was stirred for 1 h. The organic phase was separated and washed
with water and the N-methyl-N-nitrososulfonamides were
finally recrystallised from dichloromethane–light petroleum
(bp 25–40 ЊC) with a final yield of 80%. This method has the
advantage of preventing the hydrolysis of the nitroso deriv-
atives by sequestering them in the organic phase as soon as they
are formed.
ϩ
Table 2 kdH O values and kinetic solvent isotope effects in the acid-
catalysed denitrosation of N-methyl-N-nitrosobenzenesulfonamides
3
ϩ
ϩ
ϩ
kdH O /10Ϫ2 Ϫ1 sϪ1
kdH O /kdD O
kdAH/kdAD a
3
3
3
Substrate
4-CH3O
4-CH3
4-Cl
4.1 0.1
3.5 0.1
2.80 0.02
2.49 0.06
1.20 0.05
1.2 0.1
1.28 0.04
2.04 0.06
1.5 0.2
1.9 0.2
1.8 0.1
2.3 0.3
4-NO2
a Obtained with dichloroacetic acid.
of buffer solutions were added to mixtures that already con-
tained the amount of acid required to achieve the desired pH.
In all cases pH values were measured at the end of the reaction
using a Radiometer 82 pH meter with a GK-2401C combined
electrode. In experiments with D2O, pD was calculated by add-
ing 0.4 units to the measured value.15
N-Methyl-4-methoxybenzenesulfonamide:
δH(300
MHz,
CDCl3) 2.64 (3H, d, CH3-N), 3.87 (3H, s, CH3O), 6.99 (2H, m,
Ph), 7.80 (2H, m, Ph).
N-Methyl-N-nitroso-4-methoxybenzenesulfonamide: δH(300
MHz, CDCl3) 3.12 (3H, s, CH3-N), 3.89 (3H, s, CH3O), 7.04
(2H, m, Ph), 7.91 (2H, m, Ph).
N-Methyl-4-chlorobenzenesulfonamide: δH(300 MHz, CDCl3)
2.68 (3H, d, CH3-N), 7.51 (2H, m, Ph), 7.81 (2H, m, Ph).
Results
The reaction of nitrosation of the N-methylbenzenesulfon-
amides is in equilibrium with the denitrosation process and this
must be taken into account when studying their kinetics.
Experiments have been carried out using N-methyl-N-nitroso-
benzenesulfonamides as substrates to study the denitrosation
reaction while the corresponding N-methylsulfonamides were
used as substrates when studying the reverse process.
N-Methyl-N-nitroso-4-chlorobenzenesulfonamide:
δH(300
MHz, CDCl3) 3.14 (3H, s, CH3-N), 7.57 (2H, m, Ph), 7.94 (2H,
m, Ph).
N-Methyl-4-nitrobenzenesulfonamide: δH(300 MHz, CDCl3)
2.74 (3H, d, CH3-N), 8.06 (2H, m, Ph), 8.38 (2H, m, Ph).
N-Methyl-N-nitroso-4-nitrobenzenesulfonamide:
δH(300
MHz, CDCl3) 3.17 (3H, s, CH3-N), 8.20 (2H, m, Ph), 8.43 (2H,
m, Ph).
Denitrosation
Nitrite produced by denitrosation of N-methyl-N-nitroso-
sulfonamides was determined using a modification of Shinn’s
method.14 The reaction mixture was brought to pH ca. 2 and
mixed with sulfanilamide (Merck) and naphthylethylene-
diamine (Carlo Erba). Absorbance at 550 nm due to the dye
formed was measured and nitrite was quantified using the
value of 4.6 × 104 dm3 molϪ1 cmϪ1 for the molar absorptivity of
the dye.
Because of their poor solubility in water, sulfonamides and
their nitroso derivatives were dissolved in a small amount of
organic solvent (usually dioxane) prior to preparation of aque-
ous solutions. The final concentration of organic solvent in the
medium was usually 3.3% (v/v).
Kinetic runs were monitored following the change in absorb-
ance (λ = 250–275 nm) due to the formation or decomposition
of the N-nitroso derivative using a Spectronic 3000 Diode-
Array UV–VIS spectrophotometer equipped with a multiple
cell carrier thermostatted by circulating water. All experiments
were carried out at 25.0 0.1 ЊC. In all kinetic experiments
NaClO4 was used to fix the ionic strength of the medium at
0.5 . All kinetic experiments were performed under pseudo-
first-order conditions keeping in deficit either the nitroso deriv-
ative concentration (0.5–1.0) × 10Ϫ4  for denitrosation or the
nitrite concentration (6–8) × 10Ϫ4  for the nitrosation reaction.
In all cases the absorbance–time data fitted accurately the cor-
responding first-order integrated rate equations. The observed
first-order rate constants, kobs were reproducible within 3%. In
the experiments to study the effect of buffers, different amounts
Under the experimental conditions used ([N-methyl-N-
nitrososulfonamide] = 1.0 × 10Ϫ4  and [Hϩ] = 1.0 × 10Ϫ2 to
0.10 ) denitrosation occurred quantitatively. Formation of
Ϫ
NO2 as final product together with the total disappearance
of the absorption band corresponding to the N-nitroso group
confirms the irreversibility of the acid denitrosation under these
experimental working conditions for all nitrosulfonamides
studied as is reported for compound II.11
The effect of addition of BrϪ and SCNϪ on the acid denitros-
ation of N-methyl-N-nitrosobenzenesulfonamides II and III
was studied (Table 1). In neither case was catalysis observed
although these nucleophiles are highly effective (even at lower
concentrations than those used in this work) in the acid cat-
alysed decomposition of other N-nitroso compounds (i.e. the
catalytic efficiencies of ClϪ, BrϪ and SCNϪ in the denitrosation
of N-methyl-N-nitrosoaniline are in the ratio 1:50:5500).16 The
absence of nucleophilic catalysis was also observed by Williams
for the acid denitrosation of N-methyl-N-nitrosotoluene-p-
sulfonamide.11
Influence of [HClO4] on the observed rate constants of the
acid catalysed denitrosation of N-methyl-N-nitrososulfon-
amides (see Fig. 1) leads to a linear relationship implied by
eqn. (1). In Table 2 are summarised the second-order rate
ϩ
kobs = kdH O [Hϩ]
(1)
3
ϩ
constants, kdH O , obtained using eqn. (1) for all the nitroso-
3
sulfonamides studied.
1614
J. Chem. Soc., Perkin Trans. 2, 1998

View Article Online
In order to investigate whether the reaction studied is suscep-
tible to general-acid catalysis, we carried out experiments at
constant pH in monochloroacetic acid–monochloroacetate
buffers of various concentrations for denitrosation of N-
nitrososulfonamide II. In all cases a moderate catalysis was
observed (Fig. 2) in accordance with eqn. (2) in terms of the
obtained in good agreement with literature values at this ionic
strength.
Catalytic constants were also obtained from the slope of the
kobs vs. [buffer] plots, for a 1:1 ratio of the acid and the basic
forms of the buffers, using eqn. (2). Table 3 summarises the
catalytic constants for denitrosation of N-methyl-N-nitroso-
sulfonamides I–IV. In Fig. 3 a representative Brønsted plot
for N-methyl-N-nitrososulfonamide III is displayed including all
carboxylic acids studied. It is clear that this plot is essentially
straight although if the point for H3Oϩ is included a downward
curvature is observed. However, the switch from a neutral to a
cationic acid (for which temperature and medium effects would
be very different) as well as the uncertain statistical correction
for H3Oϩ could well account for these differences. From the
slope of Brønsted plots for N-methyl-N-nitrososulfonamides
I–IV αd values, always у0, were calculated and they are sum-
marised in Table 3.
[Hϩ][Buffer]
KaAH ϩ [Hϩ]
ϩ
kobs = kdH O [Hϩ] ϩ kdAH
3
(2)
total concentration of buffer, where KaAH is the acidity constant
of monochloroacetic acid.
From the slopes of kobs vs. [buffer] plots and using eqn. (2) it
is easy to obtain the catalytic constant, kdAH = (8 4) × 10Ϫ4 Ϫ1
s
Ϫ1, for the acid denitrosation of II by monochloroacetic acid.
From eqn. (2) a pKa for monochloroacetic acid of 2.42 was also
To further confirm the hypothesis that a proton transfer is
involved in the rate-determining step, the reaction was studied
in deuterium oxide both in the presence of either DClO4 (under
conditions analogous to those of Fig. 1) or dichloroacetic
ϩ
ϩ
buffers. The kinetic solvent isotope effect kdH O /kdD O varies
between 1.20 0.05 and 2.04 0.06 (for substrates I and IV
respectively) for hydrolysis by the hydronium ion in the absence
of buffer (Table 2). Experimental results are consistent with
those reported for compound II.11 The measured kinetic solvent
isotope effect on the catalytic constant of dichloroacetic acid
kdAH/kdAD, varies between 1.5 0.2 and 2.3 0.3 for substrates I
and IV respectively.
3
3
Nitrosation
The pKa for the protonation of these sulfonamides varies
between Ϫ3.4 and Ϫ6 (Table 4),17 which means that these
compounds are in their neutral form under the experimental
conditions used to study their nitrosation (pH = 1–4). The micro-
scopic reversibility principle suggests that general-base catalysis
Fig. 1 Influence of [Hϩ] upon kobs in the acid-catalysed denitrosation
of N-methyl-N-nitrosobenzenesulfonamides (᭹) I; (᭺) II; (᭡) III; (᭝)
IV, T = 25 ЊC. Ionic strength 0.50  (NaClO4).
Fig. 2 Influence of the concentration of buffers monochloroacetic
acid–monochloroacetate on kobs for the acid-catalysed denitrosation of
II at T = 25 ЊC. Ionic strength 0.50  (NaClO4). pH = (᭹) 3.30, (᭺)
3.02, (᭡) 2.64 and (᭝) 2.25.
Fig. 3 Brønsted plot for acid-catalysed denitrosation of N-methyl-N-
nitrosobenzenesulfonamide III using carboxylic acids and including
data for H3Oϩ (dotted line)
Table 3 Catalytic constants for the carboxylic acid-catalysed denitrosation of N-methyl-N-nitrosobenzenesulfonamides
kdAH/Ϫ1 sϪ1
Acid
4-CH3O
4-CH3
4-Cl
4-NO2
CH3COOH
ClCH2COOH
Cl2CHCOOH
Cl3CCOOH
αd
(3.03 0.01) × 10Ϫ5
(5.7 0.2) × 10Ϫ4
(5.9 0.5) × 10Ϫ3
(3.7 0.3) × 10Ϫ2
0.71 0.04
(2.93 0.09) × 10Ϫ5
(5.36 0.04) × 10Ϫ4
(6.7 0.5) × 10Ϫ3
(3.5 0.2) × 10Ϫ2
0.72 0.04
(4.24 0.02) × 10Ϫ5
(7.4 0.3) × 10Ϫ4
(8.2 0.1) × 10Ϫ3
(3.05 0.09) × 10Ϫ2
0.68 0.02
—
(6.27 0.08) × 10Ϫ4
(6.8 0.2) × 10Ϫ3
(2.77 0.04) × 10Ϫ2
0.71 0.04
J. Chem. Soc., Perkin Trans. 2, 1998
1615

View Article Online
ϩ
H2O
and KNO for nitrosation of N-methylbenzenesulfonamides and pKaBH of the N-methylbenzenesulfonamides
2
Table 4
k
NO
ϩ
H2O
NO
a
X
k
/Ϫ1 sϪ1
KNO
K1KNO
pKaBH log k0
2
4-CH3O
4-CH3
4-Cl
(1.1 0.1) × 107
(6 1) × 106
(2.64 0.26) × 108
(1.7 0.3) × 108
(5.4 1.8) × 107
7.1 × 106 b
79
51
16
7
9
5
Ϫ3.4
4.93
4.85
4.11
3.81
Ϫ4.2
Ϫ5.1
Ϫ6.0
(1.5 0.5) × 106
1.76 × 105 a
4-NO2
2.13b
a The equilibrium constant for the overall reaction HNO2 ϩ sulfonamide. b Estimated from a Hammett correlation (see text).
O
S
O
S
+
0
kHd [H+] + kAdH [AH]
3
+
H+
+ NO+
N
N
H
A
kNO0 + kNO [A–]
2
X
X
N
H
O
O
O
Scheme 3
should be involved in the nitrosation reaction (Scheme 3). For
the same reason nucleophilic catalysis should not be expected.
The absence of catalysis by nucleophiles was confirmed by
studying the influence of ClϪ on the nitrosation of the sulf-
onamide II (data not shown).
From Scheme 3, and considering the formation of NOϩ,
K1 = [NOϩ]/[HNO2][Hϩ], it is easy to calculate the following
rate equation [eqn. (3)]. In the absence of acids other than
H3Oϩ, eqn. (3) can be simplified to eqn. (4).
Ϫ
Fig. 4 Influence of N-methyl-4-methylbenzenesulfonamide concen-
H2O
kobs = K1k [Hϩ][S] ϩ K1kANO[Hϩ][S][AϪ] ϩ
tration upon kobs in nitrosation by NOϩ. [Hϩ] = 7.5 × 10Ϫ2 , T = 25 ЊC.
Ionic strength 0.50  (NaClO4).
NO
ϩ
kdH O [Hϩ] ϩ kdAH[AH] (3)
3
ϩ
H2O
kobs = kdH O [Hϩ] ϩ K1kNO [Hϩ][sulfonamide]
(4)
3
Fig. 4 shows the linear dependence obtained when studying
the influence of the concentration of sulfonamide II on kobs
.
The non-zero intercept is clear evidence of reversibility and can
be interpreted on the basis of eqn. (4) as the rate of denitros-
ϩ
ation under the experimental conditions. A value of kdH O
=
3
(3.6 0.1) × 10Ϫ2 Ϫ1 sϪ1 was obtained which agrees with
ϩ
the value of kdH O = (3.5 0.1) × 10Ϫ2 Ϫ1 sϪ1 reported above
3
when studying denitrosation (Table 2). The slope of the solid
H2O
line in Fig. 4 gives K1k
= (1.86 0.04) Ϫ2 sϪ1, which allows
NO
H2O
us to calculate k
= (6.20 0.15) × 106 Ϫ1 sϪ1 using the value
NO
of K1 = 3 × 10Ϫ7 Ϫ1 18
.
Fig. 5 shows the linear plots (kobs vs. [Hϩ]) obtained when
studying nitrosation of sulfonamides I–III. Their slopes are
related both to the nitrosation and the denitrosation process
[eqn. (5)]. From the slope in Fig. 5 and eqn. (5) the rate constant
Fig. 5 Influence of [Hϩ] upon kobs in the nitrosation of (᭹) N-methyl-
4-methoxybenzenesulfonamide, (᭺) N-methyl-4-methylbenzenesulfon-
amide and (᭡) N-methyl-4-chlorobenzenesulfonamide. [Sulfonamide] ϩ
6.67 × 10Ϫ3 ; T = 25 ЊC, ionic strength 0.50  (NaClO4).
ϩ
H2O
kobs = (kdH O ϩ K1kNO [sulfonamide])[Hϩ]
(5)
H2O
3
second-order rate constants for both nitrosation, k , and de-
NO
ϩ
ϩ
nitrosation, kdH O , seem to correlate also with the pKaBH [eqns.
(7) and (8)].
3
2
for nitrosation of sulfonamide II, k , is (6 1) × 106 Ϫ1 sϪ1
H2O
NO
which agrees with the value obtained from the variation of kobs
with sulfonamide concentration.
ϩ
H2O
log k
= (9.5 0.5) ϩ (0.7 0.1)pKaBH
(r = 0.97) (7)
2
NO
Due to the low solubility of the N-methylbenzenesulfon-
ϩ
ϩ
log kdH O = Ϫ(1.10 0.03) ϩ (0.08 0.007)pKaBH
H2O
NO
3
2
amides, the rate constants of nitrosation, k , were determined
(r = 0.97) (8)
by varying the acid concentration while keeping the sulfon-
amide concentration constant (Table 4). Taking into account
H2O
The possibility of the existence of general-base catalysis simi-
lar to that found in the nitrosation of amides was also investi-
gated. Fig. 6 shows the influence of the total concentration of
monochloroacetic acid on the ratio kobs/[Hϩ] for the nitrosation
of II. For this set of experiments, eqn. (9) can be derived from
the second-order rate constants for nitrosation, k
(Table 4),
NO
ϩ
and denitrosation, kdH O (Table 2), one can obtain the values
3
ϩ
for the equilibrium constants for the process KNO = kNO /kdH O
H2O
3
ϩ
(Table 4). KNO correlates quite well with the pKaBH of sulfon-
2
amides I, II and III [eqn. (6)]. From this correlation it is possible
kobs
[Hϩ]
ϩ
ϩ
log KNO = Ϫ(10.6 0.5) Ϫ (0.6 0.1)pKaBH (r = 0.96) (6)
H2O
= kdH O ϩ K1kNO [S] ϩ
2
3
kdAH
KaAH
KaAH
KaAH ϩ [Hϩ]
Ϫ
to estimate the value of KNO for sulfonamide IV allowing the
calculation of its nitrosation rate constant (Table 4). The
ϩ K1kANO[S]
ͪ
[buffer]
(9)
ͩ
1616
J. Chem. Soc., Perkin Trans. 2, 1998

View Article Online
AϪ
NO
Table 5 Catalytic constants, k , for nitrosation of N-methylbenzenesulfonamides
AϪ
NO
k
/Ϫ1 sϪ1
a
4-CH3O
4-CH3
4-CH3
—
4-Cl
4-NO2
CH3COOϪ
ClCH2COOϪ
Cl2CHCOOϪ
Cl3CCOOϪ
βNO
(4.6 0.15) × 108
(1.1 0.15) × 108
(3.0 0.5) × 107
(4.2 0.7) × 107
0.28 0.05
(2.9 0.6) × 108
(6.6 1.2) × 107
(2.2 0.5) × 107
(2.5 0.5) × 107
0.28 0.04
(1.3 0.4) × 108
(2.9 1.0) × 107
(8.6 3.0) × 106
(7.0 2.4) × 106
0.32 0.02
—
(9.4 2) × 107
2.2 × 107
2.9 × 107
—
3.2 × 106
9.4 × 105
8.4 × 105
0.27 0.06
a Experimentally determined (see text).
suggest a rate-determining protonation of the substrate. This
mechanism is similar to that of amides. However, a more
detailed analysis of the results leads to a quite different
proposal.
Evidence of a concerted mechanism
Results in Tables 2 and 4 clearly show that both processes,
denitrosation and nitrosation, show similar substituent effects.
Electron withdrawing groups retard both denitrosation and
nitrosation. If a single proton transfer were involved in the rate
determining step, electron withdrawing groups would certainly
increase the rate of the nitrosation process because of the
enhanced acidity of the protonated N-nitrososulfonamide.
Then the sequence of nitrosation rates of sulfonamides would
be the opposite of what was observed (Table 4).
The substituent effects in the nitrosation process could be
achieved by a stepwise mechanism involving a fast nitrosation
pre-equilibrium followed by a slow proton transfer to the
medium [eqn. (12)]. In that case electron withdrawing groups
Fig. 6 Influence of the concentration of buffers monochloroacetic
acid–monochloroacetate on kobs for the nitrosation of the N-methyl-4-
methylbenzenesulfonamide at T = 25 ЊC. [Sulfonamide] = 6.67 × 10Ϫ3 .
Ionic strength 0.50  (NaClO4). (᭹) pH = 2.10, (᭺) pH = 2.25, (᭡)
pH = 2.64, (᭝) pH = 2.78 and (᭿) pH = 3.30.
O
S
O
S
K
N
+ NO+
N+
H
X
eqn. (3), where KaAH is the dissociation constant of the acid used,
and [S] is the concentration of the sulfonamide that is being
nitrosated.
X
H
O
O
NO
k
O
S
In accordance with this equation, reciprocals of the slopes
+ H+
N
from Fig. 6 show a linear dependence on [Hϩ] [eqn. (10)] (not
X
(12)
NO
O
1/slope =
will disfavour the formation of the protonated nitroso inter-
mediate, and substituent effects will arise from the balance of
the pre-equilibrium constant and the deprotonation rate con-
stant. We can calculate the pre-equilibrium constant K from
Ϫ
Ϫ
KaAH/(kdAH ϩ KaAHkANO[S]) ϩ [Hϩ]/(kdAH ϩ KaAHkANO[S]) (10)
shown). From this plot a value for the dissociation constant
of monochloroacetic acid was obtained (pKa = 2.60) in good
agreement with literature values. Moreover, the second-order
ϩ
KNO and (pKaBH
)
of the nitroso derivatives [eqns. (13)–(15)
2
NO
rate constant for the reaction catalysed by the monochloro-
O
O
Ϫ
acetic anion was also determined as kA = (9 2) × 107 Ϫ1 sϪ1
.
KNO
+ H+ (13)
S
N
+ NO+
S
N
Catalytic constants of dichloroacetic and trichloroacetic buf-
fers for nitrosation of sulfonamide II were obtained from the
linear relationship between kobs and buffer concentration using
eqn. (9) (Table 5). These constants can also be obtained from
the values of the denitrosation catalytic constants, the equi-
librium constant of the nitrosation reaction and the acidity
constant of the buffer used [eqn. (11)]. Both sets of values for
X
X
H
X
NO
O
O
O
O
+
2
KBaH
+ H+ (14)
S
N+
NO
H
S
N
X
NO
O
O
ϩ
K = KNO(KaBH
)
2
(15)
NO
KNOkdAH
KaAH
AϪ
NO
(pKaBH
NO = pKaBH ϩ Ϫ 10] where the obtained values for K vary
ϩ
(11)
k
=
2
2
)
from 1.05 × 10Ϫ5 to 7.10 × 10Ϫ10 for the 4-MeO and 4-NO2
derivatives, respectively. Since the measured rate constants for
the overall reaction have a 70-fold variation, the rate constants
for the slow process must vary by three orders of magnitude,
this is not consistent with a process that should be diffusion
controlled because of the difference of pKa of the involved
species (∆pKa > 12). A protonated nitrososulfonamide with
pKa < Ϫ13 according to Eigen theory19 would deprotonate in
water at a diffusion-controlled rate and no intermediate would
be formed.
sulfonamide II agree well (Table 5). In the Discussion section
AϪ
NO
we will use constants k obtained from eqn. (11).
Discussion
According to the principle of microscopic reversibility, the
transition state (TS) of both the nitrosation and denitrosation
reactions must be the same. At a first glance, the results
obtained for denitrosation: absence of catalysis by nucleophiles,
general-acid catalysis, and the kinetic solvent isotope effect,
With regard to the denitrosation process the observed
J. Chem. Soc., Perkin Trans. 2, 1998
1617

View Article Online
sequence of reactivity is in keeping with a rate-determining
protonation although the difference in reactivity between the
4-CH3O and 4-NO2 is only 1.65, which is rather small when
compared with the corresponding difference in acidity, 1000
between the 4-CH3O and 4-NO2.
The concomitant substituent effects on both rates of denitros-
ation and nitrosation lead us to think that the rate-determining
step is not a simple proton transfer. Moreover, for denitrosation
the rate-limiting step should be a concerted process where
substrate protonation must be simultaneous with N ؒ ؒ ؒ NO
bond breaking. This mechanism accounts for the effect of the
substituents assuming an asymmetric TS with protonation
advanced to N ؒ ؒ ؒ NO bond breaking. Obviously, for nitros-
ation the TS will also be asymmetric, with N ؒ ؒ ؒ NO bond
formation advanced to the deprotonation.
ϩ
The values observed for the solvent isotope effect are kdH O
/
/
3
ϩ
kdD O = 1.20–2.04 for the denitrosation by H3Oϩ and kdAH
3
kdAD = 1.5–2.3 for the reaction catalysed by dichloroacetic acid.
Any pre-equilibrium protonation would be characterised by
a solvent isotope effect of less than 0.5.20 The value observed
Fig. 7 Jencks and More O’Ferrall diagram for the TS
ϩ
ϩ
for the isotope effect, kdH O /kdD O = 1.20–2.04, is small for a
rate-limiting proton transfer, but is in fact quite typical of
protonation by L3Oϩ.20,21
3
3
Considering both values, the position of the TS in a typical
Jencks and More O’Ferrall diagram is shown in Fig. 7.
Leffler methodology26 has been frequently used for the
discussion of asymmetric reactions, such as phenolysis of
sulfurylpyridines27 or aminolysis of triazenes.28 If the extent of
bond formation (αnuc) and bond fission (αlg) are not the same in
the TS, the Leffler parameters αnuc and αlg will not be the same
and an imbalance in the effective charge will result.29 The differ-
ence αnuc Ϫ αlg provides the apparent Leffler parameter αimbalance
which is a measure of the sign and amount of charge accum-
ulated in the TS relative to the total amount possible if the
transition structure resembled either the dissociative intermedi-
ate (resulting from N ؒ ؒ ؒ NO fission when αimbalance would be
Ϫ1) or the associative intermediate (resulting from the N-
nitrosamine protonation when αimbalance would be ϩ1). The ratio
O
S
‡
N
N
O
O
X
O
φL
1
Lφ
3
L
φ
2
ϩ
ϩ
For the transition state shown above the ratio kdH O /kdD O is
3
3
given by eqn. (16) where l is the fractionation factor of L3Oϩ
ϩ
kdH O
l3
3
(16)
=
αnuc/αlg for the denitrosation reaction is greater than unity (αnuc
/
D3Oϩ
2
k
φ1φ2
αlg = 4.12). This indicates that there is an imbalance29 with bond
formation being advanced over bond breaking. The resultant
build-up of a positive charge on the nitrogen atom adjacent to
the nitroso group corresponds to a hypothetical Leffler value
‘αimbalance’ of 0.53 as is illustrated below.
d
(l = 0.69), φ1 is that of the proton being transferred (the
‘in-flight’ proton) and φ2 is that of the other L atoms. Since the
‘in-flight’ proton is loosely bound, the value of φ1 is assumed to
be small and that of φ2 is intermediate between l and 1.22 The
fractionation factor of the N-nitrososulfonamide is assumed to
be unity, as is usual for uncharged substrates. Only the small
numerator on the right-hand side of eqn. (16) can bring about a
αlg = 0.17
NO+
+H
‡
αnuc = 0.7
ϩ
ϩ
low kdH O /kdD O ratio by balancing the small value of φ1 in the
3
3
N
O
O
S
denominator.20 Proton transfer to N-nitrososulfonamide from
some other general acid LA with a fractionation factor, φLA
close to unity would give rise to a larger primary isotope
,
X
effect.23
αimbalance = 0.53
About the transition state
The situation of the TS along the reaction coordinate can be
schematically represented by a typical Jencks and More
O’Ferrall 24,25 (Fig. 7) diagram. In our case protonation and
bond-breaking processes are represented by each axis. The loc-
ation of the TS can be approximately plotted using Brønsted αd
values as a measure of proton transfer and αlg of Leffler for the
N ؒ ؒ ؒ NO bond breaking. The Brønsted αd values for denitros-
ation of N-methyl-N-nitrososulfonamides are close to 0.7
Intrinsic rate constants for nitrosation of N-methylsulfonamides
AϪ
The catalytic constants for nitrosation, k , and for denitros-
NO
ation, kdAH, of sulfonamides by carboxylic acids fit linear
Brønsted relationships. The values of αd and βNO (Tables 3 and
5, respectively) allow us to calculate a β for the equilibrium
equal to one. The intersection of the two Brønsted plots in
Fig. 8 allows a direct determination of the intrinsic rate constant
for nitrosation of N-methylsulfonamides according to Marcus
Ϫ
(Table 3). These values show that protonation of sulfonamide
theory, defined as k0 = kA = kdAH when pKaAH ϩ log KNO = 0.29
Additional insights may be gained from consideration of the
intrinsic rate constants for nitroso group transfer. The intrinsic
rate constant, k0, provides a measure of the purely kinetic
barrier (intrinsic reactivity) of the reaction. The log k0 values
are included in Table 4. It is difficult to give precise error limits
for these parameters, because they depend not only on
uncertainties in the rate constants and pKaAH values but also on
the linear extrapolation and on the assumptions made to cal-
culate the intrinsic rate constants. Log k0 decreases following
ϩ
is nearly complete in the TS. From the values of kdH O for de-
3
nitrosation by H3Oϩ we obtain a value for ανlg ≅ 0.1 [eqn. (8)].
However, whereas Brønsted αd values are by definition normal-
ised, αν values must be normalised with respect to the αeq of
lg
a suitable calibration equilibrium. We have used the value of
ϩ
αeq = 0.6 corresponding to the variation of KNO with the pKaBH
2
of sulfonamides [eqn. (6)]. In this way, a normalised value for
αlg = 0.17 was obtained. This can be considered an index of the
degree of N ؒ ؒ ؒ NO bond breaking in the TS for denitrosation.
1618
J. Chem. Soc., Perkin Trans. 2, 1998

View Article Online
Transnitrosation and denitrosation
N-Nitrososulfonamides are well known for their ability to
transfer their nitroso group to nucleophiles in slightly acidic
or basic media, which makes them as good nitrosating agents as
alkyl nitrites. Acidic hydrolysis of alkyl nitrites occurs by a con-
certed mechanism with an asymmetric transition state where
the protonation on the oxygen adjacent to the nitroso group
and the O ؒ ؒ ؒ NO bond breaking occur simultaneously with a
slight development of charge on the oxygen, in a quite similar
way to the mechanism proposed here for N-methyl-N-nitroso-
sulfonamides.12 For a given nitroso compound the ability to
transfer the nitroso group seems to be related to its capacity
for stabilising negative charge on the atom adjacent to the
nitroso group during transnitrosation. The alcohols used to
prepare alkyl nitrites have pKa values between 12 and 16.
Those with the larger pKa values generate the less reactive alkyl
nitrites. Sulfonamides7,8 have deprotonation pKa values ranging
between 10 and 12. Both alkyl nitrites and nitrososulfonamides
easily stabilise negative charge on the oxygen of the alkoxide
group and on the nitrogen of the sulfonamide, respectively. We
therefore conclude that both the ability of transnitrosation and
the concerted mechanism for denitrosation involving proton-
ation and N ؒ ؒ ؒ NO or O ؒ ؒ ؒ NO bond breaking are related to
the ability to stabilise negative charge.
Fig. 8 Brønsted plots for nitrosation–denitrosation of III
the sequence: 4-MeO > 4-Me > 4-Cl > 4-NO2. The main con-
tributing factor to this decrease is destabilisation of the positive
charge generated on the nitrogen atom adjacent to the nitroso
group by electron withdrawing substituents.
Formation of nitroso compounds
It is interesting to compare the mechanism of nitrosation and
denitrosation of N-methyl-N-nitrososulfonamides with that of
other families of N-nitroso compounds. The absence of reac-
tion intermediates during the denitrosation of nitrososulfon-
amides is in contrast to what is observed for N-nitrosamines
and N-nitrosamides. For N-nitrosamines the mechanism of
denitrosation involves a rapid protonation equilibrium followed
by nucleophilic attack on the nitroso group.2–5 The presence of
a nitroso group diminishes the protonation pKa of amines by
about 10 units19 and one can conclude that the pKa of a pro-
tonated nitrosamine must be around 0, making it sufficiently
stable in aqueous medium. Using a similar approach, a pro-
tonated N-methyl-N-nitrosobenzenesulfonamide would have
pKa = Ϫ14, and thus, its water deprotonation should be
diffusion-controlled.30 Such an intermediate would be too
unstable and the reaction presumably occurs by an ‘enforced
concerted mechanism’.24,31 It is not expected that the N ؒ ؒ ؒ NO
bond breaking occurs during denitrosation of N-nitrosamines
with partial generation of the negative charge on the nitrogen
atom simply because the second pKa of deprotonation of
amines (pKa = 26 and approx. 16 for a nitrosamine) is high in
comparison with that of sulfonamides (pKa = 11 and approx.
1 for N-methyl-N-nitrosobenzenesulfonamide).13 Amides and
ureas possess a protonation pKa around 3 that could account
for the occurrence of concerted mechanisms of nitrosation
and denitrosation. A protonated nitrosamide would have a
pKa = Ϫ12 and consequently following Eigen theory30 would
deprotonate in water at a diffusion-controlled rate and no
intermediate is formed. This apparent disagreement was inter-
preted in the past29 by considering that nitrosation takes place
on oxygen and then is followed by a slow proton transfer. This
mechanism implies that it is the iminium (pKa close to 0) and
not the nitrosamide which is losing the proton.
Conclusions
The kinetic studies and in particular the substituent effects on
the nitrosation–denitrosation of sulfonamides led us to propose
a concerted mechanism for this process. The proton transfer
and the N ؒ ؒ ؒ NO bond breaking occur simultaneously in the
transition state, though not synchronously. The results also
show the existence of a significant imbalance between form-
ation and fission of bonds, generating an important positive
charge on the nitrogen adjacent to the nitroso group. Gener-
ation of this positive charge is due to the extent of protonation
of the nitrososulfonamide (ca. 70%) in the denitrosation pro-
cess and the extent of the nucleophilic attack of the sulfon-
amide on the NOϩ (ca. 83%) in the transition state for nitros-
ation. Thus we propose a third denitrosation pathway between
those of nitrosamines and nitrosoureas. This concerted mech-
anism comes from the instability of a possible intermediate of
nitrososulfonamide protonation and the capacity of the sulf-
onyl group to stabilise the negative charge on the nitrogen atom
adjacent to the nitroso group. The mechanistic similarities
found between acid denitrosation of nitrososulfonamides and
alkyl nitrites suggest a relationship between nitrosation–
denitrosation mechanisms and the ability to transfer their
nitroso group.
Acknowledgements
Financial support from the Dirección General de Enseñansa
Superior of Spain (project PB96-0954) is gratefully acknow-
ledged. J. A. M. thanks Junta Nacional de Investigação
Cientifíca e Tecnológica of Portugal (PRAXIS XXI/BD/5219/
95).
All these arguments are in keeping with a concerted mechan-
ism. Furthermore, the hydrolysis of N-methyl-N-nitrosobenz-
enesulfonamides fulfils Jencks criteria24 for the possibility of
proton transfer occurring concertedly with the formation or
breaking of covalent bonds: the site of protonation undergoes a
large change in pKa during the course of the reaction and the
pKa of the catalyst lies between the initial and final pKa values
of the protonation site. The hydrolysis of N-methyl-N-nitroso-
benzenesulfonamides under the conditions employed certainly
satisfies these criteria since the pKa of the N atom changes from
ca. Ϫ15 to ca. 11 and the catalyst is H3Oϩ (pKa = Ϫ1.74) or a
carboxylic acid.
References
1 D. L. H. Williams, Nitrosation, Cambridge University Press, UK,
1988.
2 (a) A. Castro, E. Iglesias, J. R. Leis, M. E. Peña and J. Vázquez Tato,
J. Chem. Soc., Perkin Trans. 2, 1986, 1725; (b) J. Casado, A. Castro,
F. Meijide, M. Mosquera and J. Vázquez Tato, J. Chem. Soc., Perkin
Trans. 2, 1987, 1759; (c) C. Bravo, P. Hervés, J. R. Leis and M. E.
Peña, J. Chem. Soc., Perkin Trans. 2, 1991, 2091.
3 J. Casado, A. Castro, M. Mosquera, M. F. Rodriguez Prieto and
J. Vázquez Tato, Ber. Bunsenges. Phys. Chem., 1983, 87, 1211.
4 G. Hallett and D. L. H. Williams, J. Chem. Soc., Perkin Trans. 2,
1980, 1372.
J. Chem. Soc., Perkin Trans. 2, 1998
1619

View Article Online
5 J. Casado, A. Castro, J. R. Leis, M. Mosquera and M. E. Peña,
Monatsh. Chem., 1984, 115, 1047.
22 A. J. Kresge, Pure Appl. Chem., 1964, 8, 243.
23 The magnitude of the isotope effects is similar for hydron transfer
from the hydronium ion and from the carboxylic acids, which
implies a more symmetrical transition state for the former than for
the latter. However experimental data show much more variation in
reactivity within the carboxylic acid set than between it and the
hydronium ion. As one of the referees suggests this anomaly could
be rationalised by considering that isotope effects between
electronegative atoms are always secondary with no primary
component [see (a)]. Nevertheless the secondary isotope effect of
1.25, which is obtained for denitrosation by carboxylic acid by using
the fractionation factors [see (b)], is not consistent with experimental
data. We think that the similarity in reactivity between carboxylic
acids and the hydronium ion is due to the switch from a neutral to a
cationic acid as is reported in the following references. (a) C. G.
Swain, D. A. Kuhn and R. L. Schowen, J. Am. Chem. Soc., 1965, 87,
1553; (b) K. B. J. Schowen, in Transition States of Biochemical
Processes, eds. R. D. Gandour and R. L. Schowen, Plenum Press,
New York, 1978, ch. 6.
6 T. H. Black, Aldrichimica Acta, 1983, 16, 3.
7 L. García-Río, J. R. Leis, J. A. Moreira and F. Norberto, J. Phys.
Org. Chem., in the press.
8 (a) L. García-Río, E. Iglesias and J. R. Leis, J. Org. Chem., 1997, 62,
4701; (b) L. García-Río, E. Iglesias and J. R. Leis, J. Org. Chem.,
1997, 62, 4712.
9 (a) R. Montesano and H. Bartsch, Mutation Res., 1976, 32, 179; (b)
A. E. Pegg, Adv. Cancer Res., 1977, 25, 195.
10 (a) Transplacental Carcinogenics, eds. L. Tomatis and U. Mohr,
IARC Scientific Publications no. 4, IARC, Lyon, 1973; (b)
D. Yarosh, Mutation Res., 1985, 145, 1.
11 D. L. H. Williams, J. Chem. Soc., Perkin Trans. 2, 1976, 1838.
12 E. Iglesias, L. García-Río, J. R. Leis, M. E. Peña and D. L. H.
Williams, J. Chem. Soc., Perkin Trans. 2, 1992, 1673.
13 G. Dauphin and A. Kergomard, Bull. Soc. Chim. Fr., 1961, 5, 486.
14 C. N. Berry and B. C. Challis, J. Chem. Soc., Perkin Trans. 2, 1974,
1638.
15 P. Salomaa, L. L. Schaleger and F. A. Long, J. Am. Chem. Soc.,
1964, 86, 1.
16 I. D. Biggs and D. L. H. Williams, J. Chem. Soc., Perkin Trans. 2,
1975, 107.
24 W. P. Jencks, Chem. Rev., 1972, 72, 705.
25 R. A. More O’Ferrall, J. Chem. Soc. B, 1970, 274.
26 (a) A. Williams, Acc. Chem. Res., 1984, 17, 425; (b) A. Williams,
Adv. Phys. Org. Chem., 1992, 27, 1.
17 L. García-Río, J. R. Leis, J. A. Moreira and F. Norberto,
unpublished results.
18 (a) N. S. Bayliss, R. Dingle, D. W. Watts and R. C. Wilkie, Aust.
J. Chem., 1963, 16, 933; (b) J. H. Ridd, Adv. Phys. Org. Chem., 1978,
15, 1.
19 B. C. Challis and J. A. Challis, in The Chemistry of Amino, Nitroso
and Nitro Compounds and Their Derivatives, ed. S. Patai, Wiley, New
York, 1983, ch. 26.
20 R. P. Bell, The Proton in Chemistry, Chapman and Hall, London,
1973.
27 A. R. Hopkins, R. A. Day and A. Williams, J. Am. Chem. Soc.,
1983, 105, 6062.
28 J. Shakes, C. Raymond, D. Rettura and A. Williams, J. Chem. Soc.,
Perkin Trans. 2, 1996, 1553.
29 (a) C. F. Bernasconi, Tetrahedron, 1985, 41, 3219;(b) C. F. Bernasconi,
Acc. Chem. Res., 1987, 20, 301; (c) C. F. Bernasconi, Adv. Phys. Org.
Chem., 1992, 27, 119.
30 M. Eigen, Angew. Chem., Int. Ed. Engl., 1964, 3, 1.
31 A. Williams, Chem. Soc. Rev., 1994, 23, 93.
21 (a) L. Melander and W. H. Saunders, Reaction Rates of Isotopic
Molecules, Wiley, New York, 1980; (b) J. Albery, in Proton Transfer
Reactions, eds. E. Caldin and V. Gold, Chapman and Hall, London,
1975, p. 263.
Paper 8/01275J
Received 13th February 1998
Accepted 21st April 1998
1620
J. Chem. Soc., Perkin Trans. 2, 1998