Nucleophilic substitution reactions of N-chloramines: Evidence for a change in mechanism with increasing nucleophile reactivity

Article abstract of DOI:10.1021/jo062356k

(Chemical Equation Presented) Third-order rate constants (k_Nu)H (M^-2 s^-1) for the hydronium ion catalyzed reactions of a range of nucleophiles with N-chlorotaurine (1) in water at 25°C and I = 0.5 (NaClO₄) are reported. The solvent deuterium isotope effects on hydronium ion catalysis of the reaction with 1 of bromide and iodide ion are (k_Br)_H/(k_Br)_D = 0.30 and (k _I)_H/(k_I)_D = 0.54, respectively. The inverse nature of these isotope effects and the absence of general acid catalysis are consistent with a stepwise mechanism involving protonation of 1 in a fast preequilibrium step. The appearance of strong catalysis by general acids for the reaction of the more nucleophilic SO₃^2- and HOCH₂CH₂S^- with the chloramine indicates a change to a concerted mechanism, with protonation of the chloramine at nitrogen and chlorine transfer to the nucleophile occurring in a single step. A rough estimate of the lifetime of the protonated chloramine in the presence of the thiolate anion suggests that the concerted mechanism is enforced by the absence of a significant lifetime of the protonated substrate in contact with the nucleophile. Theoretical calculations provide evidence against an electron-transfer mechanism for chlorination of the nucleophiles by protonated 1.

Full text of DOI:10.1021/jo062356k

Nucleophilic Substitution Reactions of N-Chloramines: Evidence for
a Change in Mechanism with Increasing Nucleophile Reactivity
Paula Calvo, Juan Crugeiras,* Ana R´ıos, and Miguel A. R´ıos
Departamento de Qu´ımica F´ısica, Facultad de Qu´ımica, UniVersidad de Santiago,
15782 Santiago de Compostela, Spain
qfjuancm@usc.es
ReceiVed NoVember 15, 2006
Third-order rate constants (kNu)H (M-2 s-1) for the hydronium ion catalyzed reactions of a range of
nucleophiles with N-chlorotaurine (1) in water at 25 °C and I ) 0.5 (NaClO4) are reported. The solvent
deuterium isotope effects on hydronium ion catalysis of the reaction with 1 of bromide and iodide ion
are (kBr)H/(kBr)D ) 0.30 and (kI)H/(kI)D ) 0.54, respectively. The inverse nature of these isotope effects
and the absence of general acid catalysis are consistent with a stepwise mechanism involving protonation
of 1 in a fast preequilibrium step. The appearance of strong catalysis by general acids for the reaction of
2-
the more nucleophilic SO3 and HOCH2CH2S- with the chloramine indicates a change to a concerted
mechanism, with protonation of the chloramine at nitrogen and chlorine transfer to the nucleophile
occurring in a single step. A rough estimate of the lifetime of the protonated chloramine in the presence
of the thiolate anion suggests that the concerted mechanism is enforced by the absence of a significant
lifetime of the protonated substrate in contact with the nucleophile. Theoretical calculations provide
evidence against an electron-transfer mechanism for chlorination of the nucleophiles by protonated 1.
Introduction
suggest that 1 may function as a physiological modulator of
inflammation, inhibiting the production of proinflammatory
agents.3
The large number of reports in the literature on the biological
activity of 1, nowadays a very active area of research, shows
the crucial role that this N-chloro amino acid plays in the human
immune system.4,5 In contrast, there have been few studies of
the chemical reactivity of this compound, despite its importance
in biology and medicine. Although the chemical properties of
1 have been analyzed recently,6 kinetic studies which could
It is well-known that N-chloro compounds are formed in vivo
and play an important role in immunity.1 Activated phagocytes
generate a large number of highly reactive oxidants, including
hypochlorous acid (HOCl), a product of the myeloperoxidase-
H2O2-Cl- system. HOCl is the major microbiocidal agent in
neutrophils, but its excessive production leads to tissue damage.
Taurine, a sulfur-containing amino acid (-O3SCH2CH2NH3
+
)
present in mammalian plasma and cells, acts as a major trap
for HOCl forming the more stable and weaker oxidant taurine
chloramine (-O3SCH2CH2NHCl (1)).2 This species exerts a
protective effect in neutrophils and helps to preserve the
bacteriocidal potential for prolonged periods. Recent studies
(3) Marcinkiewicz, J.; Grabowska, A.; Bereta, J.; Stelmaszyn˜ska, T. J.
Leukocyte Biol. 1995, 58, 667-674. Marcinkiewicz, J.; Grabowska, A.;
Chain, B. M. Immunology 1998, 94, 325-330.
(4) Gottardi, W.; Hagleitner, M.; Nagl, M. J. Pharm. Pharmacol. 2001,
53, 689-697. Koprowski, M.; Marcinkiewicz, J. Cent.-Eur. J. Immunol.
2002, 27, 69-74.
(1) Thomas, E. L.; Learn, D. B. In Peroxidases in Chemistry and Biology;
Everse, J., Everse, K. E., Grisham, M. B., Eds.; CRC Press: Boca Raton,
FL, 1991; Vol. 1, pp 83-103.
(2) Marquez, L. A.; Dunford, H. B. J. Biol. Chem. 1994, 269, 7950-
7956.
(5) Schuller-Levis, G. B.; Park, E. FEMS Microbiol. Lett. 2003, 226,
195-202. Schuller-Levis, G. B.; Park, E. Neurochem. Res. 2004, 29, 117-
126.
(6) Gottardi, W.; Nagl, M. Arch. Pharm. Pharm. Med. Chem. 2002, 335,
411-421.
10.1021/jo062356k CCC: $37.00 © 2007 American Chemical Society
Published on Web 03/31/2007
J. Org. Chem. 2007, 72, 3171-3178
3171

Calvo et al.
provide important information about the mechanism of its
reactions with nucleophilic reagents are scarce. The reactivity
of this and other chloramines toward thiols has been previously
examined,7 but the precise mechanism for this process has not
been established. This information is important for understanding
the potential of N-chloro compounds to oxidize the thiol groups
in proteins and therefore to alter their functional properties.
Nucleophilic substitution reactions are among the most
studied reaction classes in physical organic chemistry, although
the greatest effort has been put into understanding substitution
at carbon. The lack of information on the mechanism of
nucleophilic reactions involving the attack of a nucleophile on
a halogen atom has recently been pointed out.8 We are interested
in understanding the mechanisms for halogen transfer reactions
and how they are affected by changes in the nature of the
reactants. We report here the results of a study of the reactions
of 1 with several nucleophilic reagents. The transfer of chlorine
from 1 to a nucleophile requires protonation at nitrogen before
or during the transfer step. This avoids the formation of the
unstable nitranion of taurine (pKa ∼ 30). The reaction may be
subject to specific-acid catalysis, with proton transfer to the
nitrogen atom of the chloramine in an initial equilibrium step,
or to general-acid catalysis when there is significant proton
transfer during nucleophilic attack at chlorine.
FIGURE 1. pH-rate profiles of (k2)o (M-1 s-1) for the reaction of
(b) Br-, (1) N3-, (2) SCN-, and (9) I- with N-chlorotaurine at 25 °C
and I ) 0.5 (NaClO4). The solid lines for bromide ion, thiocyanate
ion, and iodide ion were calculated from the values of (kNu)H (M-2 s-1
,
Table 1) using the logarithmic form of eq 1. The line through the data
for azide ion was calculated using eq 2 and the values of (kNu
) and
H
(Ka)NuH that are given in Table 1.
no significant change in (k2)obsd with increasing buffer concen-
tration up to 0.4 M for any of these nucleophiles. Figure 1 shows
pH-rate profiles of (k2)o for the reaction of Br-, N3-, SCN-,
and I- with 1. The values of (k2)o for bromide ion, thiocyanate,
and iodide ion decrease with increasing pH according to eq 1.
In previous studies of the acid-catalyzed oxidation of iodide
by chloramines,9,10 it was proposed that the addition of iodide
to chlorine proceeds through a mechanism involving simulta-
neous chlorine and proton transfer in the rate-determining
transition state. This conclusion was based on the observation
of weak buffer catalysis of this chlorine transfer reaction. In
this work, we have not detected significant catalysis by general
acids of the reaction of N-chlorotaurine with iodide and less
reactive nucleophiles. We present here evidence that the acid-
catalyzed transfer of chlorine to these nucleophiles involves a
stepwise mechanism that proceeds through a N-protonated
chloramine intermediate. We have also found that increasing
the reactivity of the nucleophile results in a change to a
concerted mechanism, which seems to be enforced by the
absence of a significant lifetime for the protonated intermediate
in the presence of the nucleophilic reagent.
(k2)o ) (kNu)H[H3O+]
(1)
The plot (not shown) of (k2)o against [H3O+] for each of these
nucleophiles is linear and the slope gives the third-order rate
constant, (kNu)H (M-2 s-1), for the reaction of the nucleophile
with 1 catalyzed by hydronium ion (Table 1). The lines through
the data in Figure 1 were calculated from the values of (kNu H
using the logarithmic form of eq 1.
)
In studies with azide ion, where it was possible to work at
pH values close to p(Ka)NuH, the value of (kNu)H ,
(M-2 s-1
Table 1) was determined as the average of the values of
(k2)o/[H3O+]fNu, where fNu is the fraction of the nucleophile
present in the basic form. The solid line drawn through the data
in Figure 1 was calculated according to eq 2, using the values
of (kNu)H and (Ka)NuH given in Table 1.
(Ka)NuH
Results
(k2)o ) (kNu)H[H3O+]
(2)
+
[
]
Observed second-order rate constants, (k2)obsd (M-1 s-1), for
the reactions of Br-, N3-, SCN-, I-, SO32-, HOCH2CH2SCH3,
and HOCH2CH2S- with 1, in the presence of buffers at 25 °C
and I ) 0.5 (NaClO4), were determined as the slopes of linear
pseudo-first-order plots of kobsd (s-1) against the total concentra-
tion of the nucleophile (not shown) or as (k2)obsd ) kobsd/[Nu]T.
The rate constants (k2)obsd for the reaction of Br-, N3-, SCN-,
I-, and HOCH2CH2SCH3 with 1 in the presence of 0.02-0.05
M buffer are equal to the rate constants (k2)o (M-1 s-1) for the
reaction catalyzed by solvent species alone, because there is
(Ka)NuH + [H3O ]
Third-order rate constants ((kNu)D, M-2 s-1) for the reaction
of Br- and I- with 1 catalyzed by D3O+ in D2O at 25 °C and
I ) 0.5 (NaClO4) were determined as the average of the values
of (k2)o/[D3O+] (eq 1) and are given in Table 1.
The dependence on pH of the observed second-order rate
constants for the reaction of 1 with sulfite dianion, extrapolated
to zero buffer concentration ((k2)o, M-1 s-1), is shown in Figure
2. The linear region of slope -1.0 below pH 4.5 corresponds
-
to the acid-catalyzed reaction of HSO3 with 1 (Scheme 1),
detected under conditions in which >99% of the nucleophile is
(7) Peskin, A. V.; Winterbourn, C. C. Free Radical Biol. Med. 2001,
30, 572-579. Peskin, A. V.; Winterbourn, C. C. Free Radical Biol. Med.
2003, 35, 1252-1260. Peskin, A. V.; Winterbourn, C. C. Free Radical
Biol. Med. 2006, 40, 45-53.
protonated. The value of the third-order rate constant (kNuH H
)
(M-2 s-1) (Table 1) for this acid-catalyzed reaction was
calculated as the average of the values of (k2)o/[H3O+]fNuH at
pH < 4.5, where fNuH is the fraction of nucleophile present as
the acid form.
Above pH 4.5, the reaction of sulfite dianion with 1 (Scheme
1) becomes significant, and this is the only detectable reaction
at pH > 6. The rate constant (kNu)H (M-2 s-1) (Table 1) for the
(8) Grinblat, J.; Ben-Zion, M.; Hoz, S. J. Am. Chem. Soc. 2001, 123,
10738-10739.
(9) Kumar, K.; Day, R. A.; Margerum, D. W. Inorg. Chem. 1986, 25,
4344-4350.
(10) Antelo, J. M.; Arce, F.; Campos, J.; Parajo´, M. Int. J. Chem. Kinet.
1996, 28, 391-396. Antelo, J. M.; Arce, F.; Crugeiras, J.; Miraz, C.; Parajo´,
M. Gazz. Chim. Ital. 1997, 127, 355-360.
3172 J. Org. Chem., Vol. 72, No. 9, 2007

Substitution Reactions of N-Chloramines
TABLE 1. Third-Order Rate Constants for the Acid-Catalyzed Reaction of Nucleophiles with N-Chlorotaurine (1) in Watera
b
d
e
f
nucleophile
p(Ka)NuH
nc
N+
(kNu
)
H (M-2 s-1
)
(kNuH) )
H (M-2 s-1
HOCH2CH2S-
9.7
6.7
6.95h
5.67
4.93
8.87
8.01
4.0
(2.1 ( 0.5) × 1010
(5.6 ( 0.8) × 109
(4.6 ( 0.1) × 109
(5.0 ( 0.1) × 106
2-
SO3
(1.4 ( 0.2) × 108
I-
-12g
(8.5 ( 0.2) × 109 (D2O)
(2.0 ( 0.2) × 109
HOCH2CH2SCH3
SCN-
-5g
0.9g
4.80
3.92
4.02
(1.7 ( 0.1) × 109
N3
4.5
7.54
2.2
(4.8 ( 0.5) × 106
-
Br-
-10g
(7.8 ( 0.4) × 103
(2.6 ( 0.1) × 104 (D2O)
a At 25 °C and I ) 0.5 (NaClO4). b Apparent pKa of the conjugate acid of the nucleophile, determined under the experimental conditions as described in
the Experimental Section, unless noted otherwise. c Swain-Scott n values in water from ref 32, unless noted otherwise. d Ritchie’s N+ parameters taken
from ref 36. e Third-order rate constant for the hydronium ion catalyzed reaction of the basic form of the nucleophile with 1, determined as described in the
text. f Third-order rate constant for the specific-acid-catalyzed reaction of the acid form of the nucleophile with 1. g Data from ref 33b. h Assumed to be the
same as for propanethiolate ion (ref 36), on the basis of the similar reactivity of these two thiol anions toward a simple quinone methide (ref 35).
FIGURE 2. pH-rate profiles of (k2)o (M-1 s-1) for the reaction of
FIGURE 3. Dependence of the normalized observed second-order rate
constant, (k2)obsd/(k2)o, for the reaction of sulfite with N-chlorotaurine
on the total concentration of buffer at 25 °C and I ) 0.5 (NaClO4).
(1) Acetate buffer at pH 4.5. (b) Acetate buffer at pH 5.1. (2)
Phosphate buffer at pH 5.8. (9) Phosphate buffer at pH 6.5.
(9) sulfite, (2) 2-mercaptoethanol, and (O) 2-(methylthio)ethanol with
N-chlorotaurine (1) at 25 °C and I ) 0.5 (NaClO4). The solid line for
sulfite was calculated using eq 3 and the reported values of (kNuH)H,
(kNu)H, kNu, and (Ka)NuH. The line through the data for 2-mercaptoethanol
shows the fit of the data to eq 5, and the line for 2-(methylthio)ethanol
was calculated from the value of (kNu
logarithmic form of eq 1.
)
(M-2 s-1, Table 1) using the
H
(kNu)H [H3O+]fNu]/fNu determined for reactions at pH > 9.0, for
-
which the contribution of the acid-catalyzed reaction of HSO3
with 1 is negligible. The solid line through the data for the
reaction of sulfite with the chloramine in Figure 2 was calculated
using eq 3 and the values of (kNuH)H, (kNu)H, and (Ka)NuH given
in Table 1 and kNu reported above.
SCHEME 1
Figure 3 shows the effect of increasing the concentration of
buffers on the normalized observed second-order rate constants,
(k2)obsd/(k2)o, for the reaction of sulfite with 1, which varies as
a function of pH. Up to pH 4.5, the observed reaction
corresponds solely to the acid-catalyzed nucleophilic attack of
-
HSO3 at chlorine and (k2)obsd is independent of buffer
concentration up to 0.15 M. At higher pH, the reaction of sulfite
dianion with 1 becomes a significant reaction pathway, and the
observed reaction is subject to buffer catalysis (see Figure 3).
Third-order rate constants (k3)buffer (M-2 s-1) for catalysis by
the buffer of the reaction of sulfite dianion with 1 were
calculated as the slopes of linear plots of (k2)obsd against [buffer].
The catalytic constants determined at different buffer composi-
acid-catalyzed reaction of SO32- with 1 was determined as the
slope of a linear plot of [(k2)o - (kNuH)H[H3O+]fNuH]/fNu against
[H3O+], according to eq 3, derived for Scheme 1.
(k2)o ) (kNuH)H[H3O+]fNuH + ((kNu)H[H3O+] + kNu)fNu (3)
tions yielded the same value of (kNu AH
(M-2 s-1), the third-
)
The pH-rate profile for the acid-catalyzed reaction of sulfite
with the chloramine should follow a slope of -1.0 at pH .
(pKa)NuH, as shown by the dashed line in Figure 2. This line
represents values of (k2)o for the reaction of sulfite with 1,
order rate constant for general-acid catalysis of the reaction of
sulfite dianion with 1, after division by the fraction of sulfite
present as the dianion and the fraction of buffer present as the
acid form, according to eq 4. Table 2 summarizes the rate
constants obtained for different general acids.
calculated using eq 3 with kNu ) 0 and the values of (kNuH H
)
and (kNu)H reported in Table 1. The experimental values of (k2)o
in this pH region become pH independent and are consistent
with the existence of an uncatalyzed pathway for the reaction
(k3)buffer ) (kNu)AH AH Nu
f
f
(4)
2-
of SO3 with 1, kNu (M-1 s-1, Scheme 1), which involves a
water molecule acting as an acid. A value of kNu ) 11.0 ( 0.5
Buffer-independent second-order rate constants, (k2)o, for the
reaction between 2-mercaptoethanol and 1 were determined by
M-1 s-1 was calculated as the average of the values of [(k2)o -
J. Org. Chem, Vol. 72, No. 9, 2007 3173

Calvo et al.
TABLE 2. Third-Order Rate Constants for General-Acid
Catalysis of the Reaction of Sulfite Dianion with N-Chlorotaurine
(1) in Watera
SCHEME 2
b
c
catalyst
pKa
(kNu) )
AH (M-2 s-1
CH3COOH
H2PO4
4.6
6.5
8.9
9.8
(1.1 ( 0.1) × 106
(1.43 ( 0.05) × 103
(3.5 ( 0.6) × 102
(2.1 ( 0.1) × 105
-
boric acid
-
HCO3
a At 25 °C and I ) 0.5 (NaClO4). b Apparent pKa’s of the general acids
under the experimental conditions, given by pKa ) pH - log([A-]/[AH]),
determined from the pH of the buffer solutions and the concentration ratios
[A-]/[AH]. c Third-order rate constant for general-acid catalysis of the
reaction of sulfite dianion with 1, determined as described in the text.
constant for reaction of this nucleophile with 1 catalyzed by
hydronium ion (Table 1) was calculated as the slope of a linear
plot of (k2)o against [H3O+] (not shown). The line through the
data in Figure 2 was calculated from the value of (kNu)H using
the logarithmic form of eq 1.
Discussion
The literature on the chemistry of N-chloramines shows that
in aqueous solution they transfer a chlorine to nucleophiles such
as amines,11 iodide ion,9,10 sulfite dianion,12 and thiols.7
However, some reports on the reaction of chloramine with
amines in aqueous solution conclude that this reaction proceeds
through two competitive pathways: (1) chlorine transfer to the
amine leading to a new chloramine and (2) nucleophilic attack
at nitrogen to form a substituted hydrazine.13 Although these
studies suggest that nucleophilic substitution at nitrogen may
be significant for these N-chlorocompounds under certain
conditions, we have found no evidence for reaction at nitrogen
in our study of the reactions of different nucleophiles with
N-chlorotaurine. Furthermore, the observed acid catalysis of the
addition of nucleophiles to 1 reported in this work is consistent
with the requirement that protonation at nitrogen takes place
before or during chlorine transfer. Such catalysis would not be
expected for nucleophilic attack at nitrogen with chloride ion
as the leaving group.
FIGURE 4. Catalysis of the reaction of 2-mercaptoethanol with 1 by
phosphate buffers at 25 °C and I ) 0.5 (NaClO4). (2) 80% free base,
pH 7.10. (9) 50% free base, pH 6.49. (b) 20% free base, pH 5.86.
extrapolation of the observed second-order rate constants to zero
buffer concentration. The pH-rate profile for this reaction at
pH , p(Ka)NuH shows two distinct regions (see Figure 2). The
profile follows a slope of -1.0 below pH 5, and the observed
reaction is due to the acid-catalyzed reaction of RSH with 1
(Scheme 1). The third-order rate constant (kNuH)H )
(M-2 s-1
Mechanism of the Addition of Nucleophiles to 1. The pH-
rate profiles of Figures 1 and 2 show that the reactions of Br-,
N3-, SCN-, I-, and HOCH2CH2SCH3 with 1 proceed through
the specific-acid-catalyzed reaction of these nucleophiles with
the chloramine. The failure to detect catalysis by general acids
together with the inverse solvent deuterium isotope effects on
reported in Table 1 for this nucleophile was calculated as the
slope of a linear plot of (k2)o against [H3O+] according to eq 5,
derived for Scheme 1 when [H3O+] . (Ka)NuH and kNu is not
significant. The pH-independent region above pH 7 represents
the acid-catalyzed reaction of the thiolate anion with the
chloramine, (kNu)H in Scheme 1. The value of (kNu)H given in
these acid-catalyzed processes, (kNu)H/(kNu)D ) (0.30 ( 0.02)
Table 1 was calculated from the average of the values of (k2)o
- (kNuH)H [H3O+] at pH > 6 and the reported value of (Ka)NuH
(Table 1).
for bromide ion and (0.54 ( 0.02) for iodide ion (Table 1),
indicate that these reactions occur by a mechanism involving
protonation of 1 in a fast preequilibrium step (Scheme 2).14
According to this mechanism, the correlations shown for Br-,
SCN-, and I- in Figure 1 and for HOCH2CH2SCH3 in Figure
2 should break to a slope of zero at pH ) p(Ka)1-H, as a result
of complete substrate protonation. Saturation of acid catalysis
has not been observed for bromide ion up to 0.5 M perchloric
acid, and this is consistent with a recently reported value of
(Ka)1-H ) (1.14 ( 0.09) M for the acidity constant of protonated
1 in water at 25 °C and I ) 0.5 (NaClO4), determined in a
(k2)o ) (kNuH)H[H3O+] + (kNu)H(Ka)NuH
(5)
There is no detectable catalysis of the reaction of mercap-
toethanol with 1 by acetate buffers between pH 4-5. However,
there is strong buffer catalysis for this reaction at pH > 5.5,
where the observed reaction is essentially due to the nucleophilic
attack of the thiolate anion. Figure 4 shows plots of (k2)obsd
against the total concentration of phosphate buffers for the
reaction of 1 with the thiol at different buffer ratios. The slopes
of these plots are (k3)buffer, the third-order rate constants for
(11) Snyder, M. P.; Margerum, D. W. Inorg. Chem. 1982, 21, 2545-
2550. Yoon, J.; Jensen, J. N. EnViron. Sci. Technol. 1993, 27, 403-409.
(12) Yiin, B. S.; Walker, D. M.; Margerum, D. W. Inorg. Chem. 1987,
26, 3435-3441.
(13) Ferriol, M.; Gazet, J.; Rizk-Ouaini, R.; Saugier-Cohen Adad, M.-
T. Inorg. Chem. 1989, 28, 3808-3813. Elkhatib, M.; Marchand, A.; Peyrot,
L.; Counioux, J. J.; Delalu, H. Int. J. Chem. Kinet. 1997, 29, 89-98.
Elkhatib, M.; Peyrot, L.; Metz, R.; Tenu, R.; Elomar, F.; Delalu, H. Int. J.
Chem. Kinet. 2002, 34, 575-584.
(14) Keeffe, J. R.; Kresge, A. J. In Techniques of Chemistry: InVestiga-
tion of Rates and Mechanisms of Reactions; Bernasconi, C. F., Ed.; Wiley:
New York, 1986; pp. 747-790.
catalysis of the reaction by the buffer. A value of (kNu AH )
)
(4.0 ( 0.5) × 106 M-2 s-1 was calculated as an average of the
values of (k3)buffer/fAHfNu, according to eq 4.
The dependence on pH of the buffer-independent observed
second-order rate constants, (k2)o (M-1 s-1), for the reaction of
1 with 2-(methylthio)ethanol is shown in Figure 2. The values
of (k2)o increase with [H3O+] according to eq 1, and a value of
(kNu H
)
) (2.0 ( 0.2) × 109 M-2 s-1 for the third-order rate
3174 J. Org. Chem., Vol. 72, No. 9, 2007

Substitution Reactions of N-Chloramines
TABLE 3. Second-Order Rate Constants, kNu, for the Reaction of
SCHEME 3
Nucleophiles with N-Protonated Chlorotaurine, 1-H, in Watera
kNu (M-1 s-1
)
nucleophile
1-Hb
2c
I-
5.2 × 109
2.3 × 109
1.9 × 109
1.6 × 108
5.7 × 106
5.5 × 106
8.9 × 103
68
HOCH2CH2SCH3
SCN-
SCHEME 4
SCHEME 5
-
HSO3
HOCH2CH2SH
0.12d
-
N3
5.5 × 105
Br-
1.4
a At 25 °C and I ) 0.5 (NaClO4). b Second-order rate constant for the
reaction of the nucleophile with 1-H, determined as described in the text.
c Second-order rate constant for addition of the nucleophile to the quinone
methide 4-[bis(trifluoromethyl) methylene]cyclohexa-2,5-dione (2), taken
from ref 36, unless noted otherwise. d Data from ref 35.
proton transfer from the catalyst to the nitranion (∆pKa ∼ 20)
within the intermediate complex of the stepwise mechanism is
negligible. If there is no barrier to this proton transfer, the
intermediate cannot exist and the reaction must occur by a
concerted mechanism, where protonation of the chloramine at
nitrogen and chlorine transfer takes place in a single step
(Scheme 4). A Brønsted plot (not shown) for general-acid
catalysis of the reaction of sulfite dianion with 1 yields a
Brønsted coefficient of R ) (0.74 ( 0.07). This suggests that
proton transfer to nitrogen is ca. 70% complete at the transition
state.
study of the disproportionation of this chloramine in aqueous
solution.15 Table 3 gives the values of the second-order rate
constant, kNu (M-1 s-1), for the reactions of Br-, N3-, SCN-,
I-, and HOCH2CH2SCH3 with 1-H, calculated from the
observed values of (kNu)H (Table 1) and (Ka)1-H ) 1.14 M using
the relationship kNu ) (kNu)H(Ka)1-H (Scheme 2).
Several studies of the nucleophilic addition of thiocyanate
anion to alkyl halides16 and carbocations17 show the preferential
reaction of this ambident nucleophile at sulfur, with kS/kN ratios
ranging from 102 to 104. The observation of a large difference
in reactivity between the S and N nucleophilic centers of
thiocyanate in these reactions suggests that this nucleophile
reacts with 1-H through the sulfur atom to give the unstable
chlorine thiocyanate, ClSCN, which rapidly decomposes to form
thiocyanogen.18
The data in Table 3 show that the value of kNu approaches
the expected limit of 5 × 109 M-1 s-1 for a diffusion-controlled
bimolecular reaction for the most reactive nucleophiles. The
value of kNu for the reaction with iodide ion suggests that the
reaction of this nucleophile with 1-H is diffusion-limited and
is consistent with a rate constant kNu′ for reaction within an
encounter complex larger than k-d for diffusional separation of
the ion pair (Scheme 5). The values of Kas ) kd/k-d e 1.0 M-1
for the weak association of the two ions in water to form an
encounter complex19 and kd ) 5 × 109 M-1 s-1 for the diffusion-
limited reaction of iodide ion give k-d g 5 × 109 s-1 and,
therefore, kNu′ g 1011 s-1 for this nucleophile. This suggests
that the rate constant kNu′ for the reaction of 1-H with more
nucleophilic species could reach the value of 1013 s-1 for a bond
vibration frequency. A correlation of log kNu for the activation-
limited reactions of 1-H with log kNu for the attack of the same
nucleophiles at the fluorine atom in Selectfluor (1-chloromethyl-
4-fluoro-1,4-diazoniabicyclo [2.2.2]octane bis(tetrafluoro-
borate))20 gives a good linear fit with a slope of 1.6. According
to this correlation, we can predict that the rate constants for the
reaction of 1-H with I- and HOCH2CH2S- should be 30 and 1
× 105-fold larger, respectively, than that for SCN-. The
estimated rate constant for collapse of the ion pair, [RNH2-
Cl+‚Nu-], is kNu′ g 1011 s-1 for iodide ion and kNu′ g 1014 s-1
for the thiolate anion. The latter value suggests that there is no
chemical barrier to the collapse of 1-H in the presence of this
strong nucleophile. This is consistent with a change to a
concerted mechanism, which is enforced by the absence of a
significant lifetime for the protonated chloramine when it is in
contact with the thiolate ion.21
The rate profiles in Figure 2 show that the reaction of each
of sulfite and 2-mercaptoethanol with the chloramine occurs
by two pathways. The linear regions of the profiles at low pH
represent the acid-catalyzed reaction of the protonated form of
-
these nucleophiles (HSO3 and HOCH2CH2SH) with 1. The
absence of catalysis by general acids of both reactions in this
pH region is consistent with preequilibrium substrate protonation
followed by the reaction of the acid form of the nucleophiles
with 1-H (Scheme 2). Second-order rate constants for this last
step, kNu (M-1 s-1), determined as indicated above, are given
in Table 3. At higher pH, the dominant pathway for the reaction
of these sulfur nucleophiles with 1 is the acid-catalyzed reaction
2-
of SO3 and HOCH2CH2S-. These reactions show strong
general-acid catalysis, which suggests that a proton is being
transferred in the rate-determining step.
Nucleophilic attack at the neutral chloramine generates a
strongly basic nitranion (pKa ∼ 30). Therefore, general acid
catalysis of the reaction of either SO32- or HOCH2CH2S- with
1 cannot be attributed to trapping of the initially formed nitranion
by the buffer acid. The reaction must proceed through a
preassociation mechanism in which the attack of the nucleophile
at chlorine must take place in the presence of the acid catalyst.
This mechanism could be either concerted or stepwise (Scheme
3). It is likely that the activation barrier for the largely favorable
Our experimental observations are not consistent with the
conclusions of earlier studies of the reaction of iodide ion with
(15) Antelo, J. M.; Arce, F.; Calvo, P.; Crugeiras, J.; R´ıos, A. J. Chem.
Soc., Perkin Trans. 2 2000, 2109-2114.
(16) Fava, A.; Iliceto, A.; Bresadola, S. J. Am. Chem. Soc. 1965, 87,
4791-4794.
(19) Ion Association; Davies, C. W.; Butterworth & Co.: London, 1962;
pp 77-87.
(20) Crugeiras, J.; R´ıos, A.; Riveiros, E. Universidad de Santiago de
Compostela, 2005. Unpublished results.
(21) Jencks, W. P. Chem. Soc. ReV. 1981, 10, 345-375. Richard, J. P.;
Jencks, W. P. J. Am. Chem. Soc. 1984, 106, 1383-1396.
(17) Loos, R.; Kobayashi, S.; Mayr, H. J. Am. Chem. Soc. 2003, 125,
14126-14132.
(18) Barnett, J. J.; McKee, M. L.; Stanbury, D. M. Inorg. Chem. 2004,
43, 5021-5033.
J. Org. Chem, Vol. 72, No. 9, 2007 3175

Calvo et al.
different N-chloramines.9,10 It has been proposed, on the basis
of the observation of weak buffer catalysis for this reaction,
that the mechanism involves proton transfer from the general
acid to the developing nitrogen anion as the chlorine is being
transferred to iodide ion. However, most of the chloramines
that have been investigated previously are less reactive than
N-chlorotaurine, since they have more basic leaving groups.
Therefore, for these less reactive chloramines, there should be
a significant barrier for chlorine transfer to iodide ion from the
protonated chloramine. Our results suggest that the concerted
mechanism is followed only when it is enforced by the
nonexistence of the protonated chloramine in the presence of
the nucleophile, and therefore the reaction of iodide ion with
chloramines, which are less reactive than chlorotaurine, should
proceed through the stepwise mechanism shown in Scheme 2.
We are currently carrying out a more extensive study of the
reaction of chloramines with iodide ion to determine whether
the weak catalytic effect observed previously can in fact be
attributed to general acid catalysis of the addition reaction.
Feasibility of an Electron-Transfer (ET) Mechanism.
Recent studies on the mechanism by which N-fluoro compounds
transfer the fluorine atom to nucleophiles have focused on the
distinction between nucleophilic substitution and electron-
transfer pathways.22-25 Differding and Ruegg have looked at
the mechanism of electrophilic fluorinations by using a cycliz-
able radical probe and found that free radicals are not involved
in the formation of fluorinated products.22 They have also shown
that the observed rate constants for fluorination of several
nucleophiles by N-fluoro-saccharin sultam are much faster than
the rate constants for the hypothetical electron-transfer reaction,
estimated using electron-transfer theory, and concluded that
nucleophilic substitution at fluorine is the only pathway that
leads to fluorination.23 In addition, the one-electron electro-
chemical reduction of this N-fluoro compound has been found
to follow a dissociative mechanism with simultaneous electron
transfer and breaking of the N-F bond, to give a nitrogen-
centered sultam radical and fluoride ion.26 This supports the
conclusion that fluorination of nucleophiles by this N-F reagent
proceeds by an SN2 mechanism. We have previously suggested
that fluorination of nucleophiles such as iodide ion, thiocyanate
ion, and bromide ion by N-fluorobenzenesulfonimide in aqueous
solution occurs by direct nucleophilic attack at fluorine, with
nucleophilic reactivity at this center being largely determined
by the ease of one-electron oxidation of the nucleophile.27 We
would like to know whether the related reaction of nucleophiles
at chlorine proceeds by a direct substitution or by initial electron
transfer from the nucleophile to the electrophile to form radical
intermediates followed by a radical-coupling step.
SCHEME 6
6). The latter pathway can only be followed if 1-H• has a
lifetime in aqueous solution longer than a bond vibration.
Independently of the mechanism by which reduction takes place,
the reductive cleavage of 1-H may lead to formation of the
chlorine radical, which would then couple with Nu• to give
ClNu. Alternatively, a nitrogen-centered radical and chloride
ion could be the preferred products, in which case chlorination
of the nucleophile can only occur by direct substitution at
chlorine.
To investigate whether an electron-transfer mechanism can
lead to chlorinated products, we have carried out a series of
quantum chemical calculations to analyze the stability in
aqueous solution of the radical 1-H• formed by electron transfer
to 1-H and the nature of the products formed after reduction.
Figure 5 shows the potential energy profile for the cleavage of
the corresponding radical in the gas phase, obtained from density
functional theory (DFT) calculations using the UB3LYP func-
tional in combination with the 6-31+G(d) basis set, together
with the optimized geometry of the minimum.
There is only a shallow energy minimum on this surface at
a nitrogen-chlorine distance of 2.47 Å. This is much larger
than the expected bonding distance for these two atoms and
shows that the radical 1-H• is an unstable species. The long-
distance energy minimum corresponds to a complex with a
dissociation energy of 58.8 kJ/mol, in which the electron spin
density is distributed almost equally between nitrogen (0.45)
and chlorine (0.52) and the latter carries about half of the
negative charge. The distance of 1.909 Å between one of the
oxygen atoms of the SO3 group and a hydrogen bound to
nitrogen is consistent with the existence of a hydrogen bond.
Similar complexes between the fragments formed upon dis-
sociative one-electron transfer have been experimentally detected
in the reduction of haloalkanes in polar solvents.28 At longer
N-Cl distances, the nitrogen atom tends to accumulate negative
The reduction of 1-H by transfer of an electron from the
nucleophile involves the cleavage of the N-Cl bond, and the
reaction may proceed by a concerted mechanism or in two
successive steps through the intermediate radical 1-H• (Scheme
(22) Differding, E.; Ruegg, G. M. Tetrahedron Lett. 1991, 32, 3815-
3818.
(23) Differding, E.; Wehrli, M. Tetrahedron Lett. 1991, 32, 3819-3822.
(24) Umemoto, T.; Fukami, S.; Tomizawa, G.; Harasawa, K.; Kawada,
K.; Tomita, K. J. Am. Chem. Soc. 1990, 112, 8563-8575.
(25) Vincent, S. P.; Burkart, M. D.; Tsai, C. Y.; Zhang, Z.; Wong, C.-
H. J. Org. Chem. 1999, 64, 5264-5279. Nyffeler, P. T.; Gonzalez-Duro´n,
S.; Burkart, M. D.; Vincent, S. P.; Wong, C. H. Angew. Chem., Int. Ed.
2005, 44, 192-212.
(26) Andrieux, C. P.; Differding, E.; Robert, M.; Save´ant, J. M. J. Am.
Chem. Soc. 1993, 115, 6592-6599.
(27) Antelo, J. M.; Crugeiras, J.; Leis, J. R.; R´ıos, A. J. Chem. Soc.
Perkin Trans. 2 2000, 2071-2076.
FIGURE 5. Gas-phase potential energy of the reduced form of 1-H,
calculated at the UB3LYP/6-31+G(d) level, versus the N-Cl distance.
Inset: Optimized geometry of 1-H + e- in the gas phase.
3176 J. Org. Chem., Vol. 72, No. 9, 2007

Substitution Reactions of N-Chloramines
charge and this, together with the low spin density on this atom,
supports the conclusion that the amine (taurine), and not a
nitrogen-centered radical, is formed upon one-electron reduction
of 1-H in the gas phase (see Table S2 in the Supporting
Information). The calculated energy of the two possible sets of
products, obtained by adding the energies of the two fragments,
shows that the reductive cleavage of 1-H in the gas phase to
give chloride ion and a nitrogen-centered radical is less favorable
than the cleavage to chlorine radical and the amine by ca. 38
kJ/mol. Geometry optimization of the amine radical results in
the transfer of a proton from the amino to the sulfonate group,
which is then hydrogen bonded to nitrogen. However, the results
of calculations using a continuum solvation model show that
the behavior in water is different. The interactions that give
rise to the long-distance complex in the gas phase disappear in
aqueous solution and the absence of energy minima suggests
that the reduction of 1-H in this solvent takes place by a
concerted mechanism. The results of single-point calculations,
using the optimized gas-phase geometries, indicate that at long
N-Cl distances the negative charge accumulates on the chlorine
whereas essentially all the spin density is located on the nitrogen.
This observation is consistent with formation of the amine
radical and chloride ion by the reduction of 1-H in aqueous
solution. This is in good agreement with recent experimental
studies which show that N-chloro compounds undergo reductive
cleavages to yield a nitrogen-centered radical and chloride ion.29
Our attempts to fully optimize the geometry of this system in
aqueous solution were not successful, but intermediate results
suggest that a proton moves from nitrogen to oxygen and a
hydrogen bond is formed between this hydrogen and the nitrogen
atom. In fact, geometry optimization in both the gas phase and
in aqueous solution of the hypothetical amine radical, -O3SCH2-
CH2NH2•+, results in the transfer of a proton from the amino
to the sulfonate group to give HO3SCH2CH2NH•, in a confor-
mation that allows the formation of a hydrogen bond to nitrogen.
In aqueous solution, the nitrogen radical and chloride ion are
more stable than the amine and chloride radical by ca. 40 kJ
mol-1. In conclusion, the fact that electron transfer to 1-H in
aqueous solution results in the formation of chloride ion provides
evidence that chlorination by 1-H proceeds by nucleophilic
substitution at the chlorine atom.
the rate constant for the reaction of HOCH2CH2SCH3 with 1-H
is 400-fold larger than that for the reaction of HOCH2CH2SH,
and this effect is similar to the observed 600-fold difference in
the reactivities of CH3SCH3 and CH3SH toward the quinone
methide 4-[bis(trifluoromethyl) methylene]cyclohexa-2,5-di-
one.35 Furthermore, the estimated 3 × 107-fold increase in
reactivity toward 1-H on moving from HOCH2CH2SH to
HOCH2CH2S-, on the basis of a comparison of nucleophile
reactivities toward 1-H and Selectfluor,20 is not significantly
different from the 2 × 107-fold difference between the rate
constants for addition of these two nucleophiles to the same
quinone methide.35 These data suggest similar nucleophile
selectivities for displacement by sulfur nucleophiles on chlorine
and for addition of sulfur nucleophiles to carbocations. However,
bromide ion reacts with 1-H 600-fold slower than HOCH2CH2-
SH, whereas this same nucleophile is ca. 10 times more reactive
than HOCH2CH2SH toward carbocations.35,36 Also, the observed
reactivity difference between iodide ion and bromide ion of at
least 6 × 105-fold for reaction with 1-H is much larger that the
50-fold difference in reactivity for the two nucleophiles toward
carbocations. Therefore, the large nucleophile selectivities
observed for nucleophilic substitution at halogen, on the basis
of the reactivities of a very small number of sulfur nucleophiles
and halide ions, could be due to the anomalous reactivity of
halide ions. We are currently studying the reaction of some
N-halo compounds with a wide range of nucleophilic reagents
to fully characterize the reactivity of these substrates toward
nucleophilic substitution at chlorine. Azide ion is less reactive
than iodide ion and thiocyanate anion, showing a reactivity
pattern toward 1-H similar to that found for alkyl halides.
However, its reaction with 1-H is ∼2900-fold faster than what
would be expected from the n value for this nucleophile. Larger
positive deviations have been found for the addition of this
species to carbocations, but the origin of the enhanced nucleo-
philic reactivity shown by azide ion in certain reactions is still
not clear.36
Experimental Section
Materials. The sodium salt of N-chlorotaurine (1) was prepared
by reaction of taurine with chloramine-T in ethanol. Deuterium
oxide and deuterium chloride (35% w/w) were 99.9% D and 99.5%
D, respectively. Inorganic salts and organic chemicals were reagent
grade from commercial sources and were used without further
purification. Stock solutions of sodium sulfite were prepared daily,
and their concentrations were determined immediately after their
use in a kinetic run by titration with starch iodine.37
Buffer Solutions and pH Measurements. The following buffers
(0.05 M) were used to maintain constant pH in studies of the
reaction of 1 with nucleophiles: methoxyacetate, pH 3.2-4.0;
chloroacetate, pH 2.5-3.2; acetate, pH 4.5-5.2; phosphate, pH
5.8-7.2; N-2-hydroxyethyl-piperazine-N′-2-ethanesulfonic acid
(HEPES), pH 7.6; borate, pH 8.2-9.6; taurine, 9.0; 1,1,1,3,3,3-
hexafluoroisopropanol (HFIP), pH 8.6-9.9; carbonate, pH 9.4-
10.3. In some experiments with azide ion, the nucleophile also
served as the solution buffer (pH 5.0). Solution pH was measured
at 25 °C using a Radiometer PHM82 pH-meter equipped with a
GK3401C combined glass electrode. In reactions monitored by
Structure-Reactivity Correlations. Previous reports on the
reactions of electrophilic chlorine and fluorine compounds have
shown a very large sensitivity of N-halo compounds to the
strength of the attacking nucleophile, and the observed slopes
of Swain-Scott correlations for these reactions are in the range
s ) 4-5.27,30 The observed increase in kNu for the reaction of
nucleophiles with N-protonated chlorotaurine (Scheme 2) from
8.9 × 103 M-1 s-1 for bromide ion to 5.2 × 109 M-1 s-1 for
iodide ion (Table 3) is also much larger than predicted by the
Swain-Scott and Ritchie nucleophilicity scales.31-34 However,
(28) Pause, L.; Robert, M.; Save´ant, J. M. J. Am. Chem. Soc. 2000, 122,
9829-9835. Pause, L.; Robert, M.; Sav’eant, J.-M. J. Am. Chem. Soc. 2001,
123, 11908-11916. Cardinale, A.; Isse, A. A.; Gennaro, A.; Robert, M;
Save´ant, J. M. J. Am. Chem. Soc. 2002, 124, 13533-13539.
(29) Rees, M. D.; Davies, M. J. J. Am. Chem. Soc. 2006, 128, 3085-
3097. Hawkins, C. L.; Davies, M. J. J. Chem. Soc., Perkin Trans. 2 1998,
1937-1945. Hawkins, C. L.; Davies, M. J. Biochem. J. 1998, 332, 617-
625.
(34) Ritchie, C. D. Can. J. Chem. 1986, 64, 2239-2250.
(35) Toteva, M. M.; Richard, J. P. J. Am. Chem. Soc. 2000, 122, 11073-
11083.
(30) Gerritsen, C. M.; Margerum, D. W. Inorg. Chem. 1990, 29, 2757-
2762.
(31) Swain, C. G.; Scott, C. B. J. Am. Chem. Soc. 1953, 75, 141-147.
(32) Koskikallio, J. Acta Chem. Scand. 1969, 23, 1477-1489.
(33) (a) Pearson, R. G. J. Org. Chem. 1987, 52, 2131-2136. (b) Pearson,
R. G. J. Am. Chem. Soc. 1986, 108, 6109-6114.
(36) Richard, J. P.; Toteva, M. M.; Crugeiras, J. J. Am. Chem. Soc. 2000,
122, 1664-1674.
(37) Kolthoff, I. M.; Belcher, R.; Stenger, V. A.; Matsuma, G. In
Volumetric Analysis; Interscience Publishers Ltd.: London, 1957; Vol. III.
J. Org. Chem, Vol. 72, No. 9, 2007 3177

Calvo et al.
conventional UV spectroscopy, the pH was determined at the end
of the reaction. In kinetic experiments involving fast reactions that
were monitored using a stopped-flow device, the pH was measured
for control solutions identical to the solutions used in the stopped-
flow experiments. Values of pD were obtained by adding 0.4 to
the observed reading of the pH meter. The apparent pKa’s of the
Quantum Chemical Calculations. All calculations were per-
formed using the Gaussian 03 package.38 Molecular geometries were
optimized at the DFT/UB3LYP level using the 6-31+G(d) basis
set. Frequency calculations were carried out to characterize all
stationary points as minima or transition states on the potential
energy surface. DFT/UB3LYP calculations with different basis sets
(6-311G**, CEP121g or aug-cc-pVTZ) and MP2/6-31G* calcula-
tions led to essentially the same results. The effects of solvent were
simulated at 300.0 K by the PCM solvent model using Pauling
radii for the solute atoms with a scaling coefficient of 1.21 and
mapping the internal surface of the cavity with 70 elements.
nucleophiles at 25 °C and I ) 0.5 (NaClO4), given by p(Ka)NuH
)
pH - log([Nu]/[NuH]), were determined from the solution pH and
the stoichiometric concentrations of [Nu] and [NuH].
Kinetic Studies. All reactions were carried out in water at 25
°C and at constant ionic strength of I ) 0.5 maintained with sodium
perchlorate. Kinetic experiments always employed at least a 10-
fold excess of nucleophile over substrate. Reactions of 1 with
halftimes of less than 15 s were studied by using the DX17MV
stopped-flow device from Applied Photophysics. An aqueous
solution of the substrate and a buffered solution of the nucleophile
were mixed in a ratio of 1:1 to give a final reaction mixture
containing a concentration of 1 of 1 × 10-5 to 1 × 10-3 M. The
slower reactions of 1 were monitored using a conventional UV
spectrophotometer and were initiated by making a 100-fold dilution
of a solution of substrate into the reaction mixture to give a final
substrate concentration between 1 × 10-5 and 1 × 10-3 M. The
reactions with bromide and iodide ions were followed by monitoring
the appearance of Br3- at 266 nm and I3- at 287 nm, respectively.
Reactions with azide ion were difficult to follow because of the
large background absorbance of this nucleophile and were studied
Acknowledgment. This research was supported by a grant
from the Ministerio de Ciencia y Tecnolog´ıa (BQU2001-2912).
We thank John P. Richard (University at Buffalo, SUNY, United
States) and Howard Maskill (University of Newcastle, United
Kingdom) for helpful discussions of this work. We also
acknowledge computer time at CESGA.
Supporting Information Available: Rate constants (k2)obsd (M-1
s-1) for the reaction of nucleophiles with N-chlorotaurine (1) in
water at 25 °C and I ) 0.5 (NaClO4). Calculated electronic energies,
electron spin densities, and atomic charges of the reduced form of
N- protonated chlorotaurine (1-H•) as a function of the N-Cl
distance. Calculated energies of the possible products formed upon
reduction of N-protonated chlorotaurine (1-H) in the gas phase and
in aqueous solution. This material is available free of charge via
the Internet at http://pubs.acs.org.
in the presence of a fixed concentration of iodide ion by monitoring
-
the increase in absorbance at 287 nm because of formation of I3
.
Other reactions of 1 were followed by monitoring the disappearance
of the substrate at the following wavelengths: SCN-, 270 nm;
SO32-, 287 nm; HOCH2CH2SCH3, 240 nm; HOCH2CH2S-, 250
nm. First-order rate constants, kobsd (s-1), were obtained from the
fit of the absorbance data to a single-exponential function and were
reproducible to (5%.
JO062356K
(38) Frisch, M. J. et al. Gaussian 2003; Gaussian, Inc.: Pittsburgh, PA,
2003.
3178 J. Org. Chem., Vol. 72, No. 9, 2007