Kinetics and mechanism of the reaction of N-chlorosuccinimide with I - and OH-

Article abstract of DOI:10.1002/kin.20073

In this kinetic study of the reaction between the iodide ion and the N-chlorosuccinimide, we have found that the reaction is first order in N-chlorosuccinimide: namely, first order with respect to the iodide ion and first order with respect to the proton concentration. Based on the experimental results, we propose a mechanism consistent in Cl⁺ transfer from the N-chlorosuccinimide to the iodide ion. We have found that the reaction presents general acid catalysis with a Broensted exponent value of 0.10. In addition, given that the N-chlorocompounds are not stable in aqueous solution, we have carried out a study on the influence of bases in the presence of N-chlorosuccinimide. We have found that the reaction is first order in hydroxide ion, as well as in the 1,1,1,3,3,3-hexafluoroisopropanoxide ion and the 2,2,2-trifluorethoxide ion. which shows that the reaction is general-base catalyzed.

Full text of DOI:10.1002/kin.20073

Kinetics and Mechanism of
the Reaction of
N-Chlorosuccinimide with
I− and OH−
´
B. AGULLA, J. M. ANTELO, F. ARCE, M. PARAJO
´
´
´
˜
Dpto. Quımica Fısica, Facultad de Quımica, Universidad de Santiago, 15706 Santiago de Compostela, Espana
Received 12 September 2004; accepted 19 November 2004
DOI 10.1002/kin.20073
Published online in Wiley InterScience (www.interscience.wiley.com).
ABSTRACT: In this kinetic study of the reaction between the iodide ion and the N-
chlorosuccinimide, we have found that the reaction is first order in N-chlorosuccinimide:
namely, first order with respect to the iodide ion and first order with respect to the proton
concentration. Based on the experimental results, we propose a mechanism consistent in
Cl+ transfer from the N-chlorosuccinimide to the iodide ion. We have found that the reaction
¨
presents general acid catalysis with a Bronsted exponent value of 0.10. In addition, given that
the N-chlorocompounds are not stable in aqueous solution, we have carried out a study on the
influence of bases in the presence of N-chlorosuccinimide. We have found that the reaction
is first order in hydroxide ion, as well as in the 1,1,1,3,3,3-hexafluoroisopropanoxide ion and
the 2,2,2-trifluorethoxide ion, which shows that the reaction is general-base catalyzed. ꢀ 2005
C
Wiley Periodicals, Inc. Int J Chem Kinet 37: 243–252, 2005
gens in this regard. However, it has been shown that
iodoform was formed in the presence of chloramines,
in a region where the formation of the most classical
chlorinated or brominated trihalomethanes is usually
unfavoured [7].
Besides, the characteristic medicinal odors that ap-
pear quite frequently in chlorinated finished drinking
waters have been shown to be associated with the for-
mation of iodinated trihalomethanes [8,9].
Formation of various halomethanes in chlorination
of river water with added bromide and iodide was
first report by Bun et al. [10]; they found all pos-
sible single halogen and mixed halogen-containing
trihalomethanes.
INTRODUCTION
The general class of organic compounds known as
chloramines has received much attention as a possible
source of chlorating agents for water treatment [1,2].
These compounds generally have an advantage over
chlorine gas in that they do not appreciably chlorinate
organic impurities to produce toxic haloforms.
Although much information is available on the
formation of chlorine and bromine containing tri-
halomethanes [3–5], very little research has been per-
formed on the formation and occurrence of iodinated
trihalomethanes [6]. This is reasonable, given that the
iodine is the least active of the three significant halo-
If iodide is present in natural waters that are chlori-
nated, the added chlorine will oxidize the iodide very
rapidly [11]. This reaction forms the basis for many
Correspondence to: J. M. Antelo; e-mail: qfmmpm@usc.es.
2005 Wiley Periodicals, Inc.
c
ꢀ

244
AGULLA ET AL.
analytical methods for determining chlorine (for ex-
ample, the iodometric method). Molecular iodine (I2)
hydrolyzes readily to hypoiodous acid [12].
This study examines the reactivity of a chloramine,
N-chlorosuccinimide (NCS), compared with the io-
dide ion, comparing it with the reactivity against the
bromide ion (see Scheme 1).
Inaddition, giventhatoneofthemaincharacteristics
of the NCS is its instability in an aqueous solution, per-
haps due to its radical decomposition in light [13], we
have carried out a kinetic study of its reactivity against
different bases, such as the hydroxide, trifluorethoxide,
and hexafluoroisopropanoxide ions.
ter Kontron UVIKON 930, both equipped with a ther-
mostated cell carrier, which allows the temperature of
the solutions to remain constant at 25.0 0.1◦C.
pH measurements were made with a radiometer
PHM82 pHmeter using a combined electrode radiome-
ter GK2401C. The pHmeter was calibrated with buffer
solutions of pH 4.01 and 7.00 (Crison) and 10.01 and
12.45 (Beckman).
All reactions were carried out using the isolation
method; the concentration of the iodide ion, hydrox-
ide ion, or alkoxides being at least ten times that of
the NCS. The observed first-order rate constants were
obtained by nonlinear regression of the absorbance-
time data; report values are the mean of 6–10 kinetic
runs.
EXPERIMENTAL SECTION
The stoichiometry of the reaction between N-chlo-
rosuccinimide and iodide ion can be represented by the
following equation:
N-chlorosuccinimide (Aldrich) has been recrystallized
prior to use [14]. A stock solution of concentration
0.015 M was freshly prepared and stored in a UV-
opaque flask in order to avoid decomposition.
Ionic strength was kept constant (I = 0.5 M) using
NaClO4 (Aldrich), and a solution of perchloric acid
3.0 M was prepared from commercial perchloric acid
at 70% (Merck).
All sodium iodide solutions were prepared daily,
by direct weighing of commercial product, Aldrich, to
minimize the formation of triiodides.
The different buffer solutions have been prepared
from the corresponding acid or alcohol, by direct
weight of the commercial products (Aldrich), adding
the necessary amount of sodium hydroxide solution,
until the desired pH value is reached.
R1R2NCl + H+ + I − → R1R2NH + ICl
−
ICl + I−
I2 + Cl
−→
←−
In the presence of iodide ion, the iodine formed dur-
ing the reaction can be found to be in equilibrium with
the triiodide ion according to the reaction:
−
I2 + I−
I3
−→
←−
The formation of the triiodide ion is less rapid than
the reaction which is the object of the study, taking
place as soon as iodine appears in the reaction medium.
The spectrum of the triiodide ion presents two clear
absorption maximum at 287 nm and 353 nm, with very
high molar absorptivity coefficients [15], ε = 38,800
The concentration of alkoxide in the reaction
mixture was determined from
M
−1 cm−1 and ε = 25,750 M−1 cm−1, respectively, so
[B]T
(1 + 10(pK−pH)
that with very low concentrations of triiodide ion we
can obtain high absorbance values. In the study carried
out in acidic solutions, the reactions were monitored
by the formation of the triiodide ion at 287 nm.
The stoichiometry of the reaction between N-
chlorosuccimide and hydroxide ion can be represented
by the equation:
[B] =
)
where [B] and [B]T are the concentrations of alkoxide
and total alcohol, respectively, and the pK corresponds
to the pKa of the alcohol.
The kinetic experiments for the most rapid reactions
were carried out in a sequential stopped-flow Applied
Photophysics DX-17MV, while for the slower reac-
tions we have used a double-beam spectrophotome-
R1R2NCl + OH − → R1R2N − + ClOH
In aqueous solution hypochlorous acid is found to be
dissociated with a pKa of 7.5 to 25◦C [16].
Reactions were followed by spectrophotometrically
monitoring the development of a band at 292 nm due
to the sodium hypochlorite (ε = 350 M−1 cm−1 [17]).
The reaction with alkoxides takes place in a similar
way, given that a similar stoichiometry treatment can
be applied.
Scheme 1

KINETICS AND MECHANISM OF THE REACTION OF N-CHLOROSUCCINIMIDE WITH I− AND OH−
245
EXPERIMENTAL RESULTS
Reaction with Iodide Ion
We studied first the effect of the concentration of NCS
on the rate constant, varying its concentration and keep-
ing all the other parameters constant. The observed rate
constants are not affected by variations of the NCS
concentration.
To investigate the influence of the iodide ion con-
centration on the rate constant, we carried out several
experiments in which its concentration varied between
2.5 × 10−4 M and 1.5 × 10−3 M. The observed rate
constants varied linearly with the iodide ion concentra-
tion (Table I); hence, the reaction is first order in iodide
ion. The plot of kobs against iodide ion concentration is
a straight line with a slope of (2.33 0.10) × 105.
On the basis of spectrophotometric measurements,
Figure 1 Spectrophotometric data for the measurement of
K(I−) in 0.1 M de HClO4 ꢀ = 287 nm, T = 25.0 0.1◦C.
The3solid line shows the fit of Eq. (3) for K1 = 748.7 13.5.
we can determine the value of the equilibrium constant
13.5 M−1 for the equilibrium constant of the triiodide
ion, in accordance with the data found in the literature
[18].
◦
−
(KI ) for the triiodide ion at 25 C and I = 0.5 M. For
3
this, we use the infinite time absorbance values, ob-
tained in the study of the influence of the iodide ion
concentration at the wavelength used, 287 nm.
The characteristic equilibrium constant is shown in
Eq. (1):
We carried out five series of experiments varying the
concentration of iodide ion between 2.5 × 10−4 M and
1.5 × 10−3 M at different acid concentrations (0.005–
0.5 M), keeping the temperature, ionic strength, and
NCS concentration constant. We have found that there
exists a linear relationship without an appreciable in-
tercept between the observed rate constant and iodide
ion concentration. The slopes obtained are shown in
Table II, observing an increase in the slopes as the acid
concentration increases.
[I−3 ]
[I2][I−]
−
KI
=
(1)
3
The relationship obtained between the absorbance
values observed and the concentration of the species
which absorbs is given by Eq. (2):
Given that we have shown in the previous study that
there exists an influence of the proton concentration
on the observed rate constant, we have carried out the
study in a wider range, in an interval of −log [H+] com-
prised between 0.2 and 2.4, keeping the NCS concen-
tration, the iodide ion concentration, the temperature,
and the ionic strength constant.
The experimental results obtained are shown in
Fig. 2, where we can observe that a decrease in the rate
constant occurs as the pH of the medium increases,
K ε287[NCS]o[I−]
−
I3
A287
=
(2)
−
−
1 + KI [I ]
3
in which ε (M−1 cm−1) represents the molar absorp-
tivity coefficient of the reactive species, [NCS]o refers
to the initial concentration of NCS, and A287 is the ex-
perimental absorbance obtained at this wavelength.
Figure 1 shows the variation of absorbance at
287 nm with the concentration of iodide ion. By ad-
justing Eq. (2), we have obtained a value of 748.7
Table II Values of the Slopes Obtained in the Study of
the Variation of the Observed Rate Constant with the
Concentration of Iodide Ion at Different Acid
Concentrations
Table I Variation of the Rate Constant Observed with
the Concentration of Iodide Ion. [NCS] = 2 × 10−5 M,
[H+] = 5 × 10−3 M, T = 25.0 0.1◦C, I = 0.5 (NaClO4)
[H+] (M)
10−5 Slope
10 3 [I−] (M)
kobs (s−1
)
A287
0.25
0.50
0.70
1.0
72.8
122.4
164.8
246.7
362.7
0.108
0.188
0.226
0.297
0.370
5 × 10−3
2.33
2.42
2.46
2.59
2.93
0.02
0.05
0.2
0.5
1.5

246
AGULLA ET AL.
that the rate constant increases as does the temperature.
The results obtained are shown in Fig. 3.
Reaction with Hydroxide Ion
and Alkoxides
As in the study carried out for the reaction between
NCS and iodide ion, we analyzed the possible influ-
ence of the concentration of NCS on the rate constant,
varying that concentration while keeping all the other
parameters constant. The observed rate constants are
not affected by variations of the NCS concentration.
To investigate the influence of the concentration of
the sodium hydroxide on the rate constant, we have car-
ried out a series of experiments in which we varied the
concentration of sodium hydroxide between 0.003 and
0.02 M. The experimental results obtained are shown in
Fig. 4, in which we can see that there is a linear increase
in the rate constant as the concentration of sodium hy-
droxide increases, showing an intercept which is com-
patible with zero and a slope with a value of (9.3
0.4) × 103 at 25◦C.
Figure 2 Influence of the pH of the medium on the rate
constant. [NCS] = 2 × 10−5 M, [INa] = 2 × 10−4 M, T =
25.0 0.1◦C, I = 0.5 (NaClO4).
reaching a pH value which is approximately greater
than 2, where the rate constant becomes independent
of the pH value.
Continuing with the study systematics of the influ-
ence of the concentrations of the species which take
part in the reaction on the rate constant, we have stud-
ied the effect of the buffer concentration used to control
the pH of the medium. We have observed that the values
of the rate constant increase with buffer concentration,
which suggests the occurrence of a catalytic effect. In
addition, we have studied the influence of the nature
of the buffer using for this purpose different buffer so-
lutions, observing the catalytic effect on all of them.
Table III shows the results obtained in the presence of
buffers.
In order to calculate the experimental activation pa-
rameters of the reaction, the effect of the temperature
on the rate constant observed was studied. The study
has been carried out in acidic solutions at five different
temperatures: 16, 20, 25, 30, and 40◦C, and it was found
Similar behavior is observed in the presence of hex-
afluoroisopropanoxide ion (HFP) and trifluorethoxide
ion (TFE) (Fig. 5). The observed rate constants in-
crease linearly with increasing concentrations of tri-
fluoroethoxide ion and hexafluoroisopropanoxide ion
in buffer solutions at constant pH. The hydrolysis of
TFE causes changes in the pH value [19], which leads
to modifications in the contribution of the hydroxide
ion catalysis to the value of the observed rate constant.
Thus, the catalytic rate constant for trifluorethoxide ion
was calculated by plotting kobs − kOH[OH−] versus the
trifluorethoxide ion concentration.
Although slight modifications of the pH values take
place when hexafluoroisopropanol buffer is used, no
correction was carried out; the influence of the hydrox-
ide ion catalysis at pH 9 becomes negligible.
Table III Values of the Rate Constants Obtained in the Study of the Influence of the Concentration of Buffer and of
the Slopes Obtained in that Study
kobs (s−1
)
Boric
Phosphate
Acetic
Monochloracetic
Dichloracetic
Trichloracetic
[Buffer] (M)
(pH = 8.87)
(pH = 6.32)
(pH = 4.56)
(pH = 2.86)
(pH = 1.86)
(pH = 0.85)
0.40
0.30
0.20
0.10
0.05
Slope of kobs vs
[Buffer] plot
69.9 0.1
69.0 0.1
68.6 0.1
67.7 0.1
67.5 0.1
6.8 0.1
70.0 0.1
67.5 0.1
64.2 0.1
60.0 0.1
58.5 0.1
33.8 0.1
69.1 0.2
67.1 0.1
64.7 0.1
62.2 0.1
61.0 0.1
23.4 0.2
73.4 0.2
69.4 0.1
66.4 0.1
69.0 0.1
64.1 0.1
60.0 0.1
55.1 0.1
53.0 0.1
45.7 0.1
74.9 0.2
71.4 0.2
65.9 0.1
61.5 0.1
59.0 0.1
46.2 0.2
61.1 0.1
34.9 0.1

KINETICS AND MECHANISM OF THE REACTION OF N-CHLOROSUCCINIMIDE WITH I− AND OH−
247
Figure 3 Influence of the temperature on the rate constant
Figure 5 Influence of the concentration of HFP and TFE
on the observed rate constant [NCS] = 2.5 × 10−4 M, T =
25.0 0.1◦C, I = 0.5 (NaClO4).
[NCS] = 2 × 10−5 M, [INa] = 2.5 × 10−4 M, I = 0.5
(NaClO4).
The slopes obtained from Fig. 5 allow us to de-
termine the values for the catalytic constants by HFP,
kHFP = (1.1 0.1) M−1 s−1 and TFE, kTFE = (5.4
It has also been proved experimentally that the ob-
served rate constants are directly proportional to the
iodide ion concentration and to the hydrogen ion con-
centration. Thekobs valuesarenotaffectedbyvariations
in N-chlorosuccinimide concentration. The kobs values
do increase with buffer concentration when hydrogen
ion and iodide ion concentrations are constant. This
behavior is consistent with that other buffer-catalyzed
reactions.
0.1) × 102 M−1 s−1
.
The study of the effect of the temperature has been
carried out on the influence of the concentration of
the hydroxide ion on the rate constant at five different
temperatures: 16, 25, 30, 35, and 40◦C (Fig. 4).
MECHANISM AND DISCUSSION
Reaction with Iodide Ion
Based on the experimental results, we can propose
a reaction mechanism in which Cl+ transfer from the
N-chlorosuccinimide to iodide ion to give an ICl inter-
mediate, which reacts very rapidly with excess iodide
ion to give I2 and I−3 . The mechanism proposed is the
following:
The rate dependence for the reaction of the N-
chlorosuccinimide, in a buffered solution, with excess
iodide, is first order in NCS:
K1
←−
+
R1R2NClH+
R1R2NCl+H
(rapid)
−d[R1R2NCl]/dt = kobs [R1R2NCl]
(3)
−→
k2
−→
R1R2NCl + I−
R R N− + I
1
2
2
k3
−→
R1R2NClH+ + I−
R R NH + I
1
2
2
ICl + I−
I2 + I−
I2 + Cl− (rapid)
−→
←−
−
−→
I
(rapid)
←−
3
−
+
−→
R1R2NH
R1R2N + H
(rapid)
←−
On the basis of the proposed mechanism, we obtain the
following rate equation:
d[R1R2NCl]
v = −
Figure 4 Influence of NaOH concentration on the rate con-
stant at different temperatures [NCS] = 2 × 10−4 M, I = 0.5
(NaClO4).
dt
= k2[I−][R1R2NCl] + k3[I−][R1R2NClH+] (4)

248
AGULLA ET AL.
Taking account of the equilibrium corresponding
with the protonation of the N-chlorosuccinimide and
that the total concentration of N-chlorosuccinimide,
[R1R2NCl]T, is the sum of the concentration of the
protonated N-chlorosuccinimide, [R1R2NClH+], and
theunprotonated N-chlorosuccinimide[R1R2NCl], the
rate equation takes the following form:
cesses subject to general acid–base catalysis. Bearing
this in mind the rate constant will be expressed by
Eq. (8):
ꢀ
k3i
kobs = k2[I−] +
[I−][HA]
(8)
K1
i
Equation (9) expressed in terms of the total buffer
concentration and the dissociation constant of the
buffer would take the form
k2 K1[I−] + k3[H+][I−]
v =
[R1R2NCl]T
(5)
K1 + [H+]
kobs
[I−][H+]
k2
[H+]
k3
kHA [Buffer]T
K1 (Ka + [H+])
Therefore, the rate constant observed is related to
the rate constants of the elemental processes through
the equation:
=
+
+
(9)
K1
On the basis of the representation of kobs/[I−][H+]
against [Buffer]T/(Ka + [H+]), we would obtain
through the slope the value of the catalytic constant
(Table IV) for the different acids studied. The value of
the catalytic constant by H3O+ could be obtained from
the influence of the proton concentration. In Fig. 6, we
can observe the good fit of the experimental results for
the case of the reaction in the presence of the trichlo-
racetic, dichloracetic, and monochloracetic buffers. As
we can observe in Fig. 6, the intercept does not remain
constant, which is due to the fact that term k2 depends
on proton concentration. Taking the values of k2 and
k3/K1 calculated previously, we can obtain the values
for the theoretical intercepts, finding that they coincide
acceptably well, within the experimental error margin,
with those obtained from experimental results.
k2 K1[I−] + k3[I−][H+]
kobs
=
(6)
K1 + [H+]
If we suppose that in our working conditions K1 ꢁ
[H+], the concentration of protons can be considered
negligible compared with term K1, the rate constant
observed will then take the form:
k3
kobs = k2[I−] +
[I−][H+]
(7)
K1
We can see then that this mechanism leads to an
equation in accordance with all the results obtained
experimentally. Hence, the representation of the ob-
served rate constant compared with proton concen-
tration should be a straight line, of ordinate k2 [I−]
and slope k3/K1 [I−]. We have obtained values for
k2 = (2.52 0.02) × 105 M−1 s−1 and for k3/K1 =
As can be seen, kHA values obtained increase with
the acid strength in the following order: Cl3CCOOH >
Cl2CHCOOH > ClCH2COOH > CH3COOH >
−
(1.2 0.1) × 105 M−2 s−1
.
H2PO4 > H3BO3, in accordance with the results ob-
tained in previous studies [20,21].
These constants can also be calculated on the ba-
sis of the representation of the observed rate constant
compared with iodide ion concentration, obtaining an
intercept zero and an slope equal to (k2 + k3/K1 [H+]).
By carrying out the study at different proton concen-
trations, we can represent the value of each one of the
slopes obtained against each proton concentration (the
experimental data are shown in Table II), obtaining
of the intercept a value for k2 = (2.36 0.01) × 105
Table IV Values of the Catalytic Constants Obtained
for the Reaction of Iodide Ion with NCS
Species
pKa
p
q
kHA (M−2 s−1
)
H3O+
−1.74a
0.66b
1.26b
2.60b
4.50c
6.80b
8.97d
3
1
1
1
1
2
1
2
2
2
2
2
3
4
(1.20 0.10) × 105
(4.72 0.05) × 105
(3.66 0.05) × 105
(2.80 0.05) × 105
(1.87 0.02) × 105
(1.80 0.02) × 105
(5.44 0.05) × 104
Cl3CCOOH
Cl2CHCOOH
ClCH2COOH
CH3COOH
H2PO−4
M
−1 s−1 and of the slope a value for k3/K1 = (1.27
0.03) × 105 M−2 s−1
.
We can show that the values for the constants k2
and k3/K1 obtained separately agree acceptably, not
exceeding an error of 6%.
H3BO3
a After Bell, R. P. The Proton in Chemistry; Chapman and Hall:
London, 1973.
b After Martell, A. E.; Smith, R. M. Critical Stability Constants;
Plenum: New York, 1982; Vol. 5, p. 287.
c After Deneux, M.; Meillous, R.; Benoit, R. L. Can J Chem 1968,
46, 1383–1388.
d After Owen, B. B.; King, E. J. J Am Chem Soc 1943, 65, 1612–
1620.
Bro¨nsted Relationship. In the experimental study of
the reaction of the N-chlorosuccinimide with iodide
ion, we had found that the buffer-catalyzed rates in-
crease with the strength of the general acids that are
present in the solution. This behavior is typical of pro-

KINETICS AND MECHANISM OF THE REACTION OF N-CHLOROSUCCINIMIDE WITH I− AND OH−
249
Figure 6 Fit of kobs/([H+][I−]) against the total concentra-
tion of the buffer.
Figure 7 Representation of the Bro¨nsted equation, log k
against the pKa of the different acids studied, for the reaction
of NCS with iodide ion. ((•) = kH O).
The acid dissociation constants (Ka) and the rate
constants for the buffer-catalyzed reactions (kHA) can
be correlated by using the Bro¨nsted relationship [22],
which is shown in Eq. (10):
2
It is important to realize that in these processes
the proton from HA cannot be transferred to N-
chlorosuccinimide in a pre-equilibrium step, because
the rates involved would result in specific catalysis,
where the rates are dependent only on hydrogen ion
concentration and not on the HA concentration. This
means that the transition state must contain three
species, HA, R1R2NCl, and I−. The nitrogen would
thus accept a proton from HA, while it transfers Cl+
to I−. To reach this transition state it is not nec-
essary for there to be a trimolecular collision, it is
enough for the species AH to approach a weakly asso-
ciated (R1R2NClI−) species, or perhaps I− to approach
(AHNR1R2Cl) species. In any case, a complete pro-
ton transfer prior to the reaction with iodide ion would
be impossible in a general acid–base catalytic process
such as the present one.
ꢁ
ꢂ
α
kHA
p
Kaq
p
= GA
(10)
in which p is the number of equivalent acidic positions
in HA, q is the number of equivalent basic positions in
the conjugated base, A, Ga is the Bro¨nsted proportion-
ality constant, and α is the degree of proton transfer in
the transition state. The reasoning consists of the inter-
pretation of whether the transition state resembles the
reactants or the products. The values of α, for differ-
ent reactions in aqueous solution, are usually found in
the interval 0 < α < 1. This equation is a linear-free
energy relationship, between the free energies of ac-
tivation for the proton transfer reactions and the free
energies of acid ionizations. A plot of log (kHA/p) vs
log (Kaq/p) should give a line with a slope equivalent
to the α value (Fig. 7).
We have obtained a value for log Ga = 5.47 0.03
and a value for the Bro¨nsted parameter (α) of 0.13
0.01. The value found reflects the fact that the proton
transfer is in the vicinity of 13%, which shows a small
degree of proton transfer in the transition state. The low
α value is indicative of a weakly basic site that influ-
ences but does not dominate the reactivity with iodide
In the studies carried out in the present article and in
similar ones [25], we found that second-order rate con-
stants for the reactions of the type R1R2NCl + X−, in
which X− refers to bromide, iodide, cyanide ions, etc.,
which are found to be related to the nucleophilicity of
these anions by means of the Swain–Scott relationship
[26], shown in Eq. (11), in which n is the nucleophilic-
ity of the different anions (X−) and s is the sensitivity
of the reaction site, compared with C which presents
in the species CH3Br for which s = 1.00
ion. The kH O point was not used in the calculation of
2
α since it obviously does not fit the trend shown in the
other points. It is not clear that H2O actually acts as an
acid in these reactions [23], and this may be the reason
it does not fit the Bro¨nsted relationship. We have found
an effect of the same type in the study of the reaction
of the NCS with bromide ion (Ferna´ndez Ferna´ndez
I., Tesina de Licenciatura, June 1997) which was to be
expected given the similarity of both reactions.
ꢁ
ꢂ
k
log
= sn
(11)
ko
For the reaction of NCS with iodide ion, which presents
a parameter n = 5.04, we have calculated a value for
bimolecular rate constant of 2.52 × 105 M−1 s−1 and
we have found a Bro¨nsted exponent value of α = 0.12.

250
AGULLA ET AL.
For the reaction of NCS with bromide ion, which
presents a parameter n = 3.89, the value calculated for
the bimolecular constant is 0.96 M−1 s−1, whereas the
Bro¨nsted exponent value is α = 0.40.
of the reagents remain constant. Similar behavior has
been observed in the presence of trifluorethoxide and
hexafluoropropanoxide ions, in which a linear increase
occurs in the rate constant with the concentration of
alkoxide in buffered solutions at a constant pH value.
These results show the existence of a general-base cat-
alytic process.
For this reaction, bearing in mind that the study has
been carried out also with other bases other than the
hydroxide ion, we could propose a mechanism:
If we relate the values of log k against the nucle-
ophilicity (n) of the anions, I− and Br−, we can see
that an increase occurs in the bimolecular constant as n
increases. We can also calculate an approximate value
for the slope of that relationship, using the value calcu-
lated for the bimolecular constant for the reaction with
hydroxide ion, k = 9.3 × 103 M−1 s−1, which presents
a parameter n = 4.02. The value of the slope is 3.5,
which is the value which will be assigned to parameter
s.
+
R1R2NClH+
R1R2NCl + H
(rapid)
−→
←−
kOH
−→
R1R2NCl + OH−
R R N− + ClOH
1
2
The Swain–Scott relationship was obtained on the
basis of the reaction which corresponds with a nucle-
ophilic attack at carbon, where the typical values for s
are 1.0 0.5. On the basis of the value of s obtained
for our reaction (s = 4.0) and the extremely high value
for s found in the reaction with HOCl [27], in which
s = 4.7, we can show that the nucleophilic reactions
at chlorine are far more sensitive to the nucleophilic
strength of the anions than the reactions at carbon and
they are also much more rapid. This can be attributed
in part to the much easier expansion of the number of
valence electrons around chlorine.
−
+
−→
←−
ClOH
ClO + H
(rapid)
−
+
−→
←−
R1R2NH
R1R2N + H
(rapid)
On the basis of the proposed mechanism, we obtain the
rate equation (12), in which kOH refers to the catalytic
rate constant for hydroxide ion and kB is the catalytic
rate constant for any other base
d[R1R2NCl]
−
= (kOH[OH−] + kB[B])[R1R2NCl]T
dt
In addition, we must bear in mind that the Bro¨nsted
α values for Br− and I− decrease as the nucleophilicity
of X− increases, α = 0.40 for Br− and α = 0.12 for
I−. When the n values are large, the reactions are very
favorable and the degree of proton transfer from HA
is small because there is less need for acid assistance.
On the other hand, if the nucleophilicity is small, the
reaction needs larger degree of assistance by proton
transfer, as reflected by larger α values.
The reactivity of the bromide ion is less than that
of the iodide ion given that the bromide is less nucleo-
philic than the iodide. The similarity of the reactions of
the NCS with iodide and bromide suggests that there
should exist a parallel with the transfer mechanisms of
Cl+.
(12)
The rate constant observed will then take the form
kobs = kOH[OH−] + kB[B]
(13)
We will suppose that the predominant species in the
reaction medium for such high pH values is the NCS
which is found entirely in a unprotonated form.
Bro¨nsted Relationship. It is well known that strongly
basic alkoxide and hydroxide ions show negative de-
viations from Bro¨nsted plots, hence we have carried
out these correlations with the data obtained experi-
mentally in our study. Although curvature cannot be
established from three points, Bro¨nsted plot of log kOH
against the pKa of the attacking base (Fig. 8) shows that
the system presents the characteristic curvature of these
ions [28]. The curvature in Bro¨nsted plots reflects some
type of nonlinear relationship between the changes in
basicity and reactivity.
This curvature in the structure-reactivity correla-
tions may represent a change in rate-determining step
or a change in transition state structure. It is important
therefore to examine critically the reasons for nonlinear
structure-reactivity correlations. Although there exists
an explanation for the Bro¨nsted curvature based on the
Reaction with Hydroxide Ion
and Alkoxides
When introducing thesolution of N-chlorosuccinimide
in a very alkaline medium, in the presence of OH− ions,
a decomposition reaction occurs. We have found a first-
order dependence on the reaction rate with regard to the
N-chlorosuccinimide, showing that the values of kobs
are not affected by variations in its concentration. The
experimental studies carried out show that an increase
occurs in the rate constant observed as the concen-
tration of sodium hydroxide increases, when the rest

KINETICS AND MECHANISM OF THE REACTION OF N-CHLOROSUCCINIMIDE WITH I− AND OH−
Table V Values of the Activation Parameters
251
Calculated as Much for the Reaction with Iodide Ion as
for the Reaction with the Hydroxide Ion
Reaction with
Iodide Ion
Reaction with
Hydroxide Ion
k2
20
k3/K1
kOH
Ea (kJ mol−1
ꢀH# (kJ mol−1
ꢀS# (J mol−1
)
2
2
29
26
2
2
57
54
5
5
)
K
19
−1) −153 20 −130 10 −49 12
on the basis of the experimental results that the rate
constants comply with the Arrhenius equation and the
equation derived from the theory of the transition state,
using these constants to calculate the energy values,
enthalpy values, and entropy activation values for the
processes of the reactions studied, which are shown in
Table V.
The enthalpy values calculated show that the pro-
cess corresponds with a chemical control of the reac-
tion. With regard to the entropies, they present negative
values, which show that the transition state is more or-
derly than the reactives. Besides, the high values found
for the reaction with iodide ion show that the transition
state is highly solvated.
Figure 8 Bro¨nsted correlation obtained for the reaction of
NCS with different bases.
hypothesis of solvation effects of the alkoxides, it is
important to emphasize that this effect can be used to
explain almost anything. Hupe and Wu’s [29] conclu-
sion is that most of the curvatures in Bro¨nsted plots of
oxygen anions depends on the basicity of the oxygen
bases rather than the difference in energies of the reac-
tants and products. The same curvature is observed for
correlations of several reactions of oxygen anions and
elimination reactions [28,30].
If we observe the resulting correlation, it seems that
the data for bases of pKa between 8 and 12.45 fall
close to a straight line with a greater slope than those
which are between the pKa values of 12.45 and 16,
which would form another line with a less pronounced
slope. This curvature might be interpreted as a dramatic
example of the Hammon postulate in which the transi-
tion state structure becomes the reactant-like for higher
pKa.
On the other hand, given that the hydroxide ion
seems to have a smaller reactivity than we would ex-
pect, we could also interpret these results as an “anoma-
lous” behavior of this ion. However, given that there
exist studies carried out with similar systems in which
it has been found that the hydroxide ion correlates
with bases with a similar pKa, we could discard this
interpretation.
BIBLIOGRAPHY
1. Brodtmann, N. V.; Russo, P. J. J Am Water Works Assoc
1979, 71(1), 40–42.
2. Vogt, C.; Regli, S. J Am Water Works Assoc 1981, 73(1),
33–40.
3. Rook, J. J. Environ Sci Technol 1977, 11, 478.
4. Cooper, W. J.; Meyer, L. M.; Bofill, C. C.; Cordal, E.
Water Chlorin 1983, 4, 285–296.
5. Minear, R. A.; Bird, J. C. Water Chlorin 1980, 3, 151–
160.
6. Thomas, R. F.; Weisner, M. J.; Brass, H. J. Water Chlorin
1980, 3, 161–168.
7. Karpel Vel Leitner, N.; Vesella, J.; Dore, M.; Legube, B.
Environ Sci Technol 1998, 32, 1680–1685.
8. Bruchet, A.; N’guyen, K.; Mallevialle, J.; Anselme,
C. In Proceedings of AWWA Annual Conference, Los
Angeles, CA, June 18, 1989; pp. 125–141.
9. Gittelman, T. S.; Yohe, T. L.; Staudte. P. B. Proc Water
Qual Technol Conf 1989, 17, 101–115.
10. Bunn, W. W.; Haas, B. B.; Deane, E. R.; Kloepfer, R. D.
Environ Lett 1975, 10, 205–213.
The simplest explanation of the above observation
is that OH− and RO− act as nucleophiles (analogously
to I− and Br− in acidic solution) to give succinimide
and ROCl. So the “curved” Bro¨nsted plot would then
simply result from sterical hindrance.
11. Lister, M. W.; Rosenblum, P. Can J Chem 1963, 41,
3016–3020.
Activation Parameters. We have carried out the study
of the influence of the temperature on the rate constant
observed with the purpose of calculating the activa-
tion parameters, for the reaction studied. We have seen
12. Burger, J. D.; Liebhafsky, H. A. Anal Chem 1973, 45(3),
600–602.
13. Skell, P. S.; Day, J. C. J Am Chem Soc 1970, 100(6),
1951–1953.

252
AGULLA ET AL.
14. Paquette, L. A. Encyclopaedia of Reagents for Organic
Synthesis, 1995; Vol. 2, 1205.
23. Nagy, J. C.; Kumar, K.; Margerum, D. W. Inorg Chem
1988, 27, 2773–2780.
15. Palmer, D. A.; Ramette, R. W.; Mesmer, R. E. J Solution
Chem 1984, 13(9), 673–683.
16. Morris, J. C. J Phys Chem 1966, 70, 3798–3805.
17. Anbar, A.; Dostrovsky, I. J Chem Soc 1954, 1105–1108.
18. Ruasse, M. F.; Aubard, J.; Gallard, B.; Adenier, A. J Phys
Chem 1996, 90, 4382–4388.
19. Gandler, J. R.; Storer, J. W.; Ohlberg, A. A. J Am Chem
Soc 1990, 112, 7756–7762.
20. Antelo, J. M.; Arce, F.; Campos, J.; Parajo´, M. Int J Chem
Kin 1996, 28(5), 391–396.
24. Gillespie, R. J. J Chem Educ 1970, 47, 18–23.
25. Wang, T. X.; Margerum, D. W. Inorg Chem 1994, 33,
1050–1055.
26. Swain, C. G.; Scott, C. B. J Am Chem Soc 1953, 75,
141–147.
27. Gerritsen, C. M.; Margerum, D. W. Inorg Chem 1990,
29, 2757–2762.
28. Hupe, D. J.; Jencks, W. P. J Am Chem Soc 1977, 99,
451–464.
29. Hupe, D. J.; Wu, D. J Am Chem Soc 1977, 99, 7653–
7659.
30. Jencks, W. P.; Brant, S. R.; Gandler, J. R. Fendrich,
G.; Nakamura, C. J Am Chem Soc 1982, 104, 7045–
7051.
21. Antelo, J. M.; Arce, F.; Crugeiras, J.; Miraz, C.; Parajo´,
M. Gazz Chim Ital 1997, 127, 355–360.
22. Maskill, H. The Physical Basis of Organic Chemistry;
Oxford Science Publications, 1985; Ch. 10.