Reactive sulfur species: Kinetics and mechanism of the oxidation of aryl sulfinates with hypochlorous acid

Article abstract of DOI:10.1021/jp906651n

The mechanism of oxidation of ArSO₂^- (PhSO2 ^- and 5-sulfinato-2-nitrobenzoic acid = TNBO₂ ^1-/2-) with HOCl/OCl^- has been investigated using the kinetic method. In contrast to previous reports for PhSO₂^- (for which it was suggested that OCl^- and not HOCl was the reactant), the reaction proceeds through a conventional pathway: nucleophilic attack by ArSO₂^- on HOCl with concomitant Cl^- transfer to give a sulfonyl chloride intermediate (ArSO₂Cl), which we have identified spectrophotometrically. Remarkably, the rate constant for the reaction of HOCl with ArSO₂^- is on the order of 10 ⁹ M^-1 s^-1, larger than the rate constants for corresponding thiolates, and is nearly diffusion-controlled. In contrast, the rate constant for the reaction of OCl^- with ArSO₂ ^- is approximately 7 orders of magnitude smaller.

Full text of DOI:10.1021/jp906651n

1670
J. Phys. Chem. A 2010, 114, 1670–1676
Reactive Sulfur Species: Kinetics and Mechanism of the Oxidation of Aryl Sulfinates with
Hypochlorous Acid
Hisanori Ueki, Garry Chapman, and Michael T. Ashby*
Department of Chemistry and Biochemistry, UniVersity of Oklahoma, Norman, Oklahoma 73019
ReceiVed: July 14, 2009; ReVised Manuscript ReceiVed: October 30, 2009
-
-
The mechanism of oxidation of ArSO2 (PhSO2 and 5-sulfinato-2-nitrobenzoic acid ) TNBO21-/2-) with
-
HOCl/OCl- has been investigated using the kinetic method. In contrast to previous reports for PhSO2 (for
which it was suggested that OCl- and not HOCl was the reactant), the reaction proceeds through a conventional
-
pathway: nucleophilic attack by ArSO2 on HOCl with concomitant Cl+ transfer to give a sulfonyl chloride
intermediate (ArSO2Cl), which we have identified spectrophotometrically. Remarkably, the rate constant for
the reaction of HOCl with ArSO2- is on the order of 109 M-1 s-1, larger than the rate constants for corresponding
-
thiolates, and is nearly diffusion-controlled. In contrast, the rate constant for the reaction of OCl- with ArSO2
is approximately 7 orders of magnitude smaller.
Introduction1
the reactions of aryl sulfinates with HOCl approach the diffusion
limit, faster than the corresponding reactions of more nucleo-
Hypochlorous acid (HOCl) is considerably more reactive than
its conjugate base, hypochlorite (OCl-), usually by 4 or more
orders of magnitude.2 The most reactive partners with HOCl
tend to have low oxidation potentials and be good nucleophiles.3
In general, the reactions proceed by electrophilic Cl+ transfer.
In some cases, the chlorinated species exhibit relative stability
in water (e.g., for chloramines).4 However, subsequent hydroly-
sis often occurs rapidly in an aqueous environment to yield
effective O-atom transfer, albeit vis-a`-vis a stepwise mechanism
that involves the aforementioned chlorinated intermediates. We
are aware of only one mechanistic study with substantial
evidence for faster oxidation by OCl- versus HOCl: Kice and
Puls reported in 1977 that OCl- reacts with ArSO2- (Ar ) Ph
philic thiolates.
Experimental Methods
General Details. All chemicals were ACS certified grade or
better. Water was doubly distilled in glass. Solutions of NaOH,
mostly free of CO2 contamination, were quantified by titration
with a standardized HCl solution using phenolphthalein as an
indicator. HClO4 solution was standardized against bicarbonate.
Electronic spectra were measured using a HP 8452A diode array
spectrophotometer or the monochromator of a HI-TECH SF-
61 DX2 stopped-flow instrument. For fast reactions, kinetic
measurements were made with a HI-TECH stopped-flow
spectrophotometer using a Xe arc lamp and PMT detector.
During the stopped-flow measurements, the temperature was
maintained at 291 K with a Lauda RC-20 circulator. The
adiabatic temperature increases that were associated with the
pH jump experiments were determined experimentally to be less
than 1 °C. For slow reactions, kinetic data were measured with
the diode-array instrument. The pH values of the buffered
solutions were determined with an Orion Ion Analyzer EA920
using an Ag/AgCl combination pH electrode. Unless otherwise
noted, all kinetic measurements were made using solutions for
which the ionic strengths had been adjusted to 1.0 using NaClO4.
Materials. PhSO2- ·Na+ (98%, Aldrich), PhSO2Cl (99%,
Aldrich), and PhSO3H (>98%, Fluka) were used as received
from the manufacturers. An improved synthesis of 5-mercapto-
2-nitrobenzoic acid (TNB) from 5,5′-dithiobis(2-nitrobenzoate)
(DTNB or Ellman’s reagent), as well as the first syntheses of
5-sulfino-2-nitrobenzoic acid (TNBO2), 5-sulfo-2-nitrobenzoic
acid (TNBO3), and 3-carboxy-4-nitrobenzensulfonyl chloride
(TNBO2Cl) will be described elsewhere. All of the new
compounds that were employed in this study were fully
characterized by 1H NMR, 13C NMR, HRMS, and UV-visible
spectroscopy. The UV-visible spectra of the TNB derivatives
are generally unique (Table 1).
and p-MeOC6H4) faster than HOCl reacts with ArSO2- to give
- 5
ArSO3
. Based upon a pH-profile for the reaction (that
suggested OCl-, not HOCl, was the reactant), the stoichiometry,
and the effects of Ar substituents, the authors suggested that
-
OCl- nucleophilically attacks ArSO2 to produce a sulfurane-
like intermediate, which subsequently decomposes to give
- 5
ArSO3
:
Given the novel nature of the proposed pathway, we have
reinvestigated the mechanism. We report here a detailed
-
mechanistic study which evidence that the reactions of ArSO2
(PhSO2 and 5-sulfinato-2-nitrobenzoic acid ) TNBO2
-
1-/2-
)
with HOCl/OCl- proceed via conventional pathways, with
electrophilic Cl+ transfer by HOCl to give sulfonyl chloride
intermediates (ArSO2Cl), which we have identified spectropho-
tometrically. We show that the kinetics of hydrolysis of PhSO2Cl
are consistent with the data that Kice and Puls attributed to the
oxidation reaction. However, it is remarkable that the rates of
HPLC Analysis of DTNB, TNB, TNBO2, TNBO3, and
TNBO2Cl. HPLC analysis was performed using conditions
similar to those employed previously.6 A Zorbax SB C18
column (4.6 × 150 mm, 5 µm) was used for all analyses. The
* Corresponding author. E-mail: MAshby@ou.edu.
10.1021/jp906651n  2010 American Chemical Society
Published on Web 12/29/2009

Aryl Sulfinates and Hypochlorous Acid
J. Phys. Chem. A, Vol. 114, No. 4, 2010 1671
poses relatively quickly under neutral7 to acidic conditions,8
whereas OCl- is relatively stable under strong basic conditions.
The following experiment was designed on the basis of these
chemical properties. Using a hand-mixer or the HI-TECH
instrument, a solution containing 200 µM TNBO2Na2, 0.1 M
HClO4 and buffer (none for pH 1, 13, and 14; phosphate buffer
for pH 2-3, 5.81-8.2, and 10.14-12.31; acetate buffer for pH
3.91-5.1) was mixed in a 1:1 ratio with a NaOCl solution in
0.1 M NaOH. The ensuing reaction was monitored with a diode
array instrument or with the spectrophotometer of the HI-TECH
stopped-flow instrument.
TABLE 1: Molar Absorption Coefficients in Aqueous Media
compound
solution (pH)
λmax (nm)
εmax (M-1 ·cm-1
)
DTNB2-
DTNB0
TNB2-
TNB0
PBS (7.4)
HClO4 (0.0)
PBS (7.1)
HClO4 (0.0)
PBS (7.2)
PBS (7.2)
HCl (0.0)
NaOH (14.0)
NaOH (14.0)
HClO4 (1.0)
324
316
412
330
268
264
234
264
262
234
16330
9800
14150
9140
2-
TNBO22-
5760
TNBO3
5800
TNBO2Cl0
∼7500a
-
PhSO2-
950
PhSO3
380
PhSO2Cl0
>6500a
a The accuracy of ε is affected by hydrolysis.
General Data Analysis. Monochromatic kinetic traces were
modeled by single- or double-exponential equations using
KaleidaGraph 3.6 (Synergy Software). The concentration de-
pendencies of the pseudo-first-order rate constants were obtained
by linear least-squares fits. Polychromatic data were analyzed
using SPECFIT/32 (Spectrum Software Associates), a multi-
variate data analysis program.
Calibration of the HI-TECH SF-61 DX2 Stopped-Flow
Instrument and Analysis of the Data of Figure 6. Measure-
ment of the kinetic trace of Figure 6 required additional
precautions. The instrument was calibrated using the reaction
of 2,6-dichlorophenol-indophenol (DCIP) and ascorbic acid
(AA) at pH ) 1.80 under pseudo-first-order conditions (excess
[AA]).9 The plot of [AA] vs kobs started to deviate from linearity
when kobs was larger than 400 s-1, indicating that the highest
limit of kobs that could be measured for a (pseudo-)first-order
reaction on our HI-TECH SF-61 DX2 stopped-flow instrument
is 400 s-1. The dead time and the pretriggering time of the
instrument were determined on the basis of the following
equations:
flow rate of eluent (45% CH3CN/55% H2O, 0.1% TFA) was
set at 1.5 mL/min for the analyses of DTNB, TNB, and
TNBO2Cl. The retention times were 19.8, 11.2, and 16.1 min,
respectively. For TNBO2 and TNBO3, a more polar eluent (5%
CH3CN/95% H2O, 0.1% TFA) was used, and the flow rate was
set at 1.0 mL/min. The retention times were 16.5 and 10.7 min,
respectively.
-
Kinetic Study of the Oxidation of PhSO2 with HOCl/
OCl-. The oxidation was investigated using conditions that were
comparable to those employed by Kice and Puls.5 Stock
-
solutions of PhSO2 were prepared by dissolving a known
amount of the sulfinic acid in phosphate buffer. HOCl/OCl-
solutions were prepared by diluting standardized aqueous
solutions of NaOCl in phosphate buffer. The sulfinate solution
was mixed with freshly prepared solutions of HOCl/OCl- in a
1/1 ratio using a hand-mixer to give [PhSO2Na] ) 0.35 mM
and [NaOCl] + [HOCl] ) 2.13 mM. The pH value after mixing
was 6.01. The reaction was monitored with a diode-array
instrument.
obstd
Deadtime: Aobs)Atote-k
Kinetic Study of the Hydrolysis of TNBO2Cl. The solubility
of TNBO2Cl in aqueous base is relatively high, although it
hydrolyzes quickly. Conversely, TNBO2Cl hydrolyzes more
slowly in aqueous acid, but it exhibits poor solubility therein.
On the basis of these restrictions, two experimental methods
were designed for measuring the hydrolysis of TNBO2Cl.
Method A (for slow reactions at pH 0-8): A known amount
of TNBO2Cl was dissolved in 1 mL of dioxane. An aliquot was
diluted 100 times with aqueous buffer (phosphate buffer for
pH 1.89-3.05 and 5.79-8.15; acetate buffer for pH 3.82-5.27)
or nonbuffered (pH 0-1) solutions. Immediately thereafter, the
hydrolysis reaction was monitored using a diode-array instrument.
Method B (for fast reactions at pH 10.6-14): A known
amount of TNBO2Cl was dissolved in 1 mL of dioxane. An
aliquot was diluted 100 times with 0.5 M HClO4. The freshly
prepared TNBO2Cl solution in 0.5 M HClO4 and 1% dioxane
solution was placed in a reservoir syringe of the stopped-flow
instrument. Buffered (phosphate buffer for pH 10.61-12.11)
or nonbuffered (pH > 11) solutions containing 0.5 M NaOH
were placed in another reservoir syringe of the stopped flow
instrument. The hydrolysis was measured in single-mixing mode
with the HI-TECH stopped-flow spectrophotometer.
Kinetic Study of the Hydrolysis of TNBO2Cl in the
Presence of HOCl/OCl-. Using a hand-mixer, a freshly
prepared TNBO2Cl solution in 0.5 M HClO4 with 1% dioxane
was mixed in 1:1 ratio with a freshly prepared HOCl/OCl-
phosphate-containing solution containing 0.5 M NaOH. Then,
the reaction was monitored with a diode-array instrument.
Kinetic Study of the Oxidation of TNBO2 with HOCl/
OCl-. TNBO2 is soluble and relatively stable in aqueous media
regardless of the pH. However, TNBO2 is slowly oxidized by
O2 in aqueous solutions to give TNBO3. HOCl/OCl- decom-
obsts
Pre-triggering time: Atot)Afite-k
where Aobs is the observed absorbance maximum; Atot is the
effective absorbance change associated with the reaction; Afit is
the fitted absorbance change (extrapolated to t ) 0); td is the
dead time and ts is the pretriggering time of the instrument. It
has been previously shown mathematically that it should be
necessary to correct for a concentration gradient within the
stopped-flow observation cell for fast second-order reactions,
where the aging of the reaction mixture has to be taken into
account and the observed rate constants have to be corrected
using the geometrical parameters of the observation cell).10 The
half-life for the reaction of Figure 6 is ca. 6 ms, close to the
mixing time of the instrument. However, as has been observed
previously,10 standard second-order treatment (which ignores
the existence of a concentration gradient) yielded a rate constant
that was statistically identical to one computed when the gradient
was taken into account. Furthermore, we investigated the effect
of the mixing rate constant (kmix) on kobs and we have found
that no correction was necessary for the data collected with our
instrument for the reaction conditions of Figure 6.
Results
Products of the Reaction of HOCl/OCl- with ArSO2- and
Hydrolysis of ArSO2Cl. The eventual products of the oxidation
of ArSO2- (PhSO2- and TNBO21-/2-) with HOCl/OCl- in aqueous
medium were the ArSO3- derivatives. The ArSO3- derivatives
were also observed as the only organic products of the hydrolysis

1672 J. Phys. Chem. A, Vol. 114, No. 4, 2010
Ueki et al.
Figure 2. Log-log plot of kobs (s-1) vs [H+] for the oxidation (circles,
with exponential fit illustrated with a solid line) of TNBO2 (100 µM)
with NaOCl (100 µM) and hydrolysis (squares, with exponential fit
illustrated with a dashed line) of TNBO2Cl (88.5-117 µM) for 0 e
pH e 8.3. All the data sets were fitted to a first-order rate law with
SVD analysis of the entire UV-vis spectra. Buffer ) (none, phosphate,
or acetate) and I ) 1.0 M (buffer + NaClO4).
Figure 1. Absorbance decrease at 234 nm (λmax for TNBO2Cl)
observed when TNBO2 (100 µM) is reacted with NaOCl (100 µM) in
0.3 M PBS at pH ) 7.29 and I ) 1.0 M (PBS + NaClO4). Twenty
percent of the data are illustrated (for clarity) with a first-order fit (k )
0.0355 ( 0.0001 s-1). Inset: UV-vis spectra recorded at 1 s intervals
(data for 12 s intervals illustrated for clarity). SVD modeling of these
data with a first-order model yielded k ) 0.0358 ( 0.0001 s-1
.
1
of ArSO2Cl. The ArSO3- products were identified by H NMR,
HPLC, and UV-vis using authentic samples (data not shown).
Characterization of an Intermediate Observed During the
Reaction of HOCl/OCl- with ArSO2 . ArSO2Cl (PhSO2Cl and
-
TNBO2Cl) were observed spectrophotometrically as intermedi-
ates during the reaction of ArSO2 with HOCl/OCl-. The
-
sulfonyl chlorides were identified by their unique UV-vis
spectra using authentic samples (Table 1). Figure 1 illustrates
2-
a typical time-resolved experiment for the reaction of TNBO2
(100 µM) with HOCl/OCl- (100 µM) at pH 7.3. Note that λmax
) 268 nm for TNBO22-, but the first spectrum that is observed
(Figure 1, inset) exhibits an absorbance maximum at 234 nm,
which is characteristic of TNBO2Cl (Table 1). During the time
interval of 300 s, the spectrum of TNBO2Cl is replaced with
2-
the spectrum for TNBO3 (with a λmax ) 264 nm). A similar
-
observation was made when PhSO2 (λmax ) 262 nm) was
oxidized with HOCl/OCl- to give PhSO3 (λmax ) 264 nm)
-
vis-a`-vis PhSO2Cl (λmax ) 234 nm).
Kinetics of the Hydrolysis of ArSO2Cl. Given the observa-
tion of ArSO2Cl as an intermediate (vide supra), in preparation
for measuring the rates of oxidation, we investigated the
hydrolysis of ArSO2Cl (PhSO2Cl and TNBO2Cl). The rate of
hydrolysis of TNBO2Cl was studied between pH 0 and 13.7.
At all values of pH, a decrease in the absorbance at 234 nm
(λmax for TNBO2Cl) was observed with a proportional increase
in the absorbance at 264 nm (λmax for TNBO3). All of the kinetic
traces were described by single exponential equations, which
indicate that the reaction proceeds with pseudo-first-order
kinetics with a first-order dependence upon [TNBO2Cl]0. Figure
2 illustrates that the rate of oxidation for [TNBO2]0 ) [HOCl]0
+ [OCl-]0 ) 100 µM is identical to the rate of hydrolysis of
TNBO2Cl for the approximate pH range that was investigated
by Kice and Puls.5 For each pH, TNBO2Cl was observed as
the primary TNB derivative (as identified by λmax ) 234 nm)
in the first spectrum that was collected. Similar data were
Figure 3. Top: effect of [OH-] on the hydrolysis of TNBO2Cl in
unbuffered solution. The inset is an expansion of the same data. All
the data sets are well fitted to a first-order rate law with SVD analysis
of the entire UV-vis spectra. Based on the slope and intercept of the
linear fits (illustrated), k0 ) 5.5123 [s-1] and kOH ) 373.48 [M-1 s-1].
Conditions: [TNBO2Cl]0 ) 96 µM, 1% dioxane, I ) 1.0 M (NaOH +
NaClO4). Bottom: pH-kobs profile for the hydrolysis of TNBO2Cl in
the 0-12 pH range. All time traces were fit to first-order rate laws
with SVD analysis of the entire UV-vis spectra. Conditions: 88.5-117
µM TNBO2Cl, 0.3 M buffer (none, phosphate, or acetate for pH 0-12),
0.5-1% dioxane, I ) 1.0 M (buffer + NaClO4). Based on a fit of the
entire data set to the equation kobs ) k0 + kOH[OH-] (illustrated), k0 )
collected for the reaction of PhSO2 with HOCl/OCl- under
-
0.008 ( 0.011 s-1 and kOH ) 438 ( 13 M-1 s-1
.
the same conditions employed by Kice and Puls, and PhSO2Cl
was the first observable species (data not shown). The rate of
hydrolysis of TNBO2Cl- was also investigated between pH 10
and 14 in unbuffered medium. No change in pH was observed
for these reactions (because [OH-] was in large excess with
respect to [TNBO2Cl]0). The pseudo-first-order rate constants
were determined by SVD analysis of polychromatic data. The
combined data sets are summarized in Figure 3. The rate of
hydrolysis of PhSO2Cl has been investigated previously,11 albeit
not above pH 11, and not for I ) 1. For the purpose of the

Aryl Sulfinates and Hypochlorous Acid
J. Phys. Chem. A, Vol. 114, No. 4, 2010 1673
Figure 4. Effect of [OH-] on the rate of hydrolysis of PhSO2Cl (50
µM) Conditions: 50 µM PhSO2Cl, unbuffered, 0.5% dioxane, I ) 1.0
M (NaOH + NaClO4). First-order fits are illustrated (20% of the data
shown). Inset: relationship between kobs (s-1) and [OH-], fit with a linear
function to give kOH ) 27 ( 1 M-1 s-1
.
present study (vide infra), it was necessary to measure the rate
of hydrolysis of PhSO2Cl for pH 13-14 and I ) 1 (Figure 4).
Kinetics of the Reaction of ArSO2- with HOCl/OCl-. The
-
rate of reaction of ArSO2 with HOCl/OCl- exceeds the
capabilities of the stopped-flow method below pH 9. The rate
of hydrolysis of TNBO2Cl- becomes faster than the rate of
oxidation of TNBO22- above pH 12. However, over the narrow
pH range of 10.5-11.5 when [TNBO22-]0 ) [OCl-]0 (an excess
of one or the other would accelerate the oxidation reaction rate),
we observe both reactions. Figure 5 illustrates kinetic traces
for the reaction of TNBO22- (100 µM) with OCl- (100 µM) at
pH 10.5, as monitored at two wavelengths: 230 nm (near λmax
for TNBO2Cl-) and 260 nm (near λmax of TNBO32-). Note that
Kice and Puls employed higher reactant concentrations ([Ph-
SO2-]0 ) 300-600 µM and [HOCl]0 + [OCl-]0 ) 2-3 mM)
and lower pH (5-9) than the conditions we employed to collect
the data of Figure 5. The traces of Figure 5 exhibit biphasic
kinetics, as expected for a stepwise reaction (Table 2). The upper
trace of Figure 5 for 230 nm (λmax for TNBO2Cl is 234 nm)
exhibits an initial increase in absorption, followed by a decrease,
as expected for the formation of an intermediate, TNBO2Cl-,
that subsequently hydrolyzes to give TNBO32-. As expected,
the lower trace at 260 nm (λmax for TNBO32- is 264 nm) mirrors
this behavior, with an initial decrease in absorption, followed
by an increase in absorption (as expected for an initial decrease
in [TNBO22-] followed by the formation of TNBO32-). At lower
pH, only the second reaction is observed (vide infra).
Figure 5. Biphasic absorption changes that occur at 230 nm (λmax for
TNBO2Cl is 234 nm) and at 260 nm (λmax for TNBO2 and TNBO3 are
268 and 264 nm, respectively), observed when TNBO2 (100 µM) is
reacted with NaOCl (100 µM) in 0.3 M PBS at pH ) 10.5 and I ) 1.0
M (PBS + NaClO4). A fit using the model A + B f C f D is
illustrated. The insets are the same data with expanded abscissa axes.
TABLE 2: SVD Analyses of the Polychromatic Data That
Exhibit Biphasic Kinetics between pH 10 and 11.5 for the
2-
Oxidation of TNBO2 (100 µM) by NaOCl (100 µM) Using
the Model A + B f C f Da
kobs
k′obs
kHOCl
b
pH
(M-1 s-1)b × 10-4
(s-1
)
(M-1 s-1)c × 108
10.14
10.47b
10.77
10.84
11.01
11.29
11.5
59.4 ( 0.7
18.5 ( 0.4
10.3 ( 0.1
5.76 ( 0.08
4.15 ( 0.06
3.77 ( 0.07
1.05 ( 0.02
0.1414 ( 0.0004 3.2 ( 0.3
0.2323 ( 0.0006 2.18 ( 0.02
0.2391 ( 0.0007 2.43 ( 0.01
0.377 ( 0.001
0.462 ( 0.002
0.660 ( 0.005
0.93 ( 0.01
1.17 ( 0.02
1.7 ( 0.1
2.92 ( 0.06
3.32 ( 0.09
X ) 2.4 ( 0.8
b
a Two traces of this data set are illustrated in Figure 5. kobs
)
Reaction Conditions Where Oxidation Is Faster Than
Hydrolysis. The rate of oxidation of ArSO2- by HOCl can only
be monitored over a very narrow pH range of ca. 9.0-9.5 (at
very low reactant concentrations) and above pH 13. The
oxidation reaction is too fast to monitor using the stopped-flow
method below pH 9 (for concentrations of reactants that produce
sufficient changes in absorption). Between pH 10 and 13, the
rate of hydrolysis is competitive with the rate of oxidation (e.g.,
Figure 5). Above pH 13, oxidation can become rate-limiting
(if the reactant concentrations are sufficiently high). Since we
were interested in extrapolating to the pH range that was studied
by Kice and Puls (pH 5-9), we investigated the rate of reaction
the rate constants for the formation of TNBO2Cl and k′obs ) the rate
constants for the hydrolysis of TNBO2Cl. kHOCl ) kobsKaHOCl/[H+].
c
investigating the first reaction, oxidation of ArSO2- with OCl-,
because the subsequent reaction of ArSO2Cl with borate is
relatively slow at pH 9. The reaction for [PhSO2-]0 ) [OCl-]0
) 25 µM at pH 9 is illustrated in Figure 6. Equimolar reaction
conditions were chosen to minimize the reaction rate (of the
hetero-second-order reaction) while employing the minimum
concentration that yielded sufficient absorbance change to
accurately determine the rate. The final absorption (0.065 au)
corresponds to the absorption at 234 nm of the (transient)
spectrum of PhSO2Cl (Figure 6, inset). The difference in
of PhSO2 with OCl- at pH 9. One of the few buffer systems
-
that is effective near pH 9 is borate. Unfortunately, we have
observed that sulfonyl chlorides react with the borate buffer
(unpublished results). However, this is not problematic for
-
absorption between the starting PhSO2 reactant and the
PhSO2Cl (∆Abs234nm ) 0.065 - 0.010 ) 0.054 au), as well as

1674 J. Phys. Chem. A, Vol. 114, No. 4, 2010
Ueki et al.
Figure 6. Absorption change that occurs at 234 nm (λmax for PhSO2Cl)
-
that is observed (an average of 20 mixing cycles) when PhSO2 (25
µM) is reacted with NaOCl (25 µM) in 50 mM borate buffer at pH )
9.0 and I ) 1.0 M (NaH2BO3 + NaClO4). A hetero-second-order fit is
illustrated (kobs ) 2.32(9) × 107 M-1 s-1). Inset: the absorption spectrum
-
of 25 mM PhSO2 and 1 min after reaction with 25 mM NaOCl.
the change that is observed in the time trace (∆Abs234nm ) 0.065
- 0.030 ) 0.035 au), suggests that ca. 45% of the reaction has
occurred in the mixing time of the stopped-flow experiment.
This was taken into consideration when fitting the data set of
Figure 6 to a second-order rate function.
Reaction Conditions Where Oxidation Is CompetitiWe with
(or Slower Than) Hydrolysis. The data of Figure 5 (and related
conditions) illustrate both the oxidation and hydrolysis reactions.
However, it is difficult to accurately measure the rate of
oxidation under such conditions because the oxidation reaction
remains fast, and because it overlaps with the hydrolysis reaction
from pH 10-13. Ideally, the [OCl-] would be raised to the
point that its disappearance could be measured directly
(ε292nm(OCl-) ) 350 M-1 cm-1). This is not possible below pH
13, as the reaction becomes too fast to measure by stopped-
flow. However, it is possible to make such measurements above
pH 13 (Figure 7). The data of Figure 7 that were collected at
234 nm (λmax for PhSO2Cl, cf. the analogous data for TNBO2Cl
of Figure 5) illustrate the biphasic kinetics that we interpret as
the formation of PhSO2Cl and its subsequent hydrolysis (vide
supra). More PhSO2Cl is produced at [OH-] ) 0.1 M than at
[OH-] ) 1.0 M, as expected for faster oxidation (vide infra)
and slower hydrolysis at lower pH. In contrast to the kinetic
traces at 234 nm, the kinetic traces at 292 nm (λmax for OCl-)
exhibit simple first-order behavior. Thus, the ∆Abs at 292 nm
is proportional to -d[OCl-]/dt and the rate of the oxidation
reaction. The data of Figure 7 were collected under the pseudo-
Figure 7. Top: absorption change that occurs at 292 nm (λmax for OCl-)
that is observed when PhSO2- (1 mM) is reacted with NaOCl (1 mM)
in unbuffered medium with 0.1 e [OH-] e 1.0 M and I ) 1.0 M
(NaOH + NaClO4). First-order fits are illustrated (5% of the data
shown). Bottom: absorption change at 234 nm (λmax for PhSO2Cl) that
occurs under the same reaction conditions. Double exponential fits are
illustrated (20% of the data shown). Inset: plot of kobs computed from
both sets of data that corresponds to the oxidation reaction versus
1/[OH-] (circles are for λ ) 292 nm and squares are for λ ) 234 nm).
These rate data are summarized in Table 3. All of the data points were
included in the linear fit except for the square that lies off the line
(which corresponds to λ ) 234 nm when the rate of hydrolysis is
comparable to the rate of oxidation).
TABLE 3: Comparison of the Pseudo-First-Order Rate
Constants for Hydrolysis (khyd) and Oxidation (kox) as a
Function of [OH-]
a
b
c
b
[OH-] khyd (s-1
)
khyd (s-1
)
kox (s-1
)
kox (s-1
)
0.10
0.25
1.84 ( 0.01 3.63 ( 0.07 1.117 ( 0.003 0.87 ( 0.01
5.07 ( 0.02 6.50 ( 0.07 0.545 ( 0.003 0.525 ( 0.005
0.50 11.43 ( 0.05 17.3 ( 0.6 0.304 ( 0.001 0.313 ( 0.001
1.00 29.3 ( 0.2
30 ( 1
0.179 ( 0.001 0.1884( 0.0009
first-order reaction conditions [PhSO2 ]0 ) 100 µM and [OCl-]0
-
a Measured directly using PhSO2Cl and pH ump (Figure 4).
b Deconvoluted from the biphasic data of Figure
7 (bottom).
) 1 mM (although an analogous set of data were also collected
c Measured directly at λ ) 292 (Figure 7, top).
for [PhSO2 ]0 ) 1 mM and [OCl-]0 ) 100 µM, data not shown).
-
The kinetic traces at 292 nm are well-described by a first-order
model (Figure 7, top). The values of kobs (s-1) that were
determined at 292 nm are plotted versus 1/[OH-] in the inset
of Figure 7. The significance of the slope and intercept of the
line that describes the data will be discussed later. We attribute
the pH dependence to the reaction of HOCl (not OCl-), even
at the very high pH of Figure 7 (pH 13-14). The reason for
the dominance of HOCl at pH 13-14 (5.6-6.6 orders of
magnitude from the pKa of HOCl) is the fact that kHOCl is near
the diffusion limit (109-1010 M-1 s-1) and is about 7 orders of
magnitude greater than kOCl (vide infra). The biphasic data of
Figure 7 were fit to a double exponential function that describes
sequential (pseudo-)first-order reactions. In general, when the
two rate constants (k1 and k2) of sequential reactions are of the
same magnitude, the data are equally well described using rate
laws where k1 < k2 and k1 > k2. However, in the present case,
we have independently measured the rates of oxidation of
PhSO2- by HOCl (Figure 7, top) and the rates of hydrolysis of
PhSO2Cl (Figure 4). Accordingly, we are able to attribute the
data of Figure 7 (bottom) to the oxidation and hydrolysis steps
(Table 3). Note that the oxidation step of Figure 7 (top) and
Table 3 (last two columns) exhibit an inverse dependency on
[OH-] (as expected for reaction by HOCl), whereas the rate of

Aryl Sulfinates and Hypochlorous Acid
J. Phys. Chem. A, Vol. 114, No. 4, 2010 1675
hydrolysis is proportional to [OH-]. Importantly, since the rate
constant for the reaction of HOCl with PhSO2- (kHOCl) must be
extrapolated by 5.6-6.6 orders of magnitude to the pKa of
HOCl, the data of Figure 7 are not necessarily expected to yield
a precise value for kHOCl. However, the actual objective of
-
-d[OCl-]
-d[PhSO2 ]
Rate )
)
)
dt
dt
kOCl ·KHa OCl + kHOCl ·[H+]
KHa OCl + [H+]
-
·[HOCl]T ·[PhSO2 ]
(2)
(
)
collecting the data of Figure 7 was to determine the value of
-
the rate constant for the reaction of OCl- with PhSO2 (kOCl
,
where [HOCl]T ) [HOCl] + [OCl-] ≈ [OCl-] (for pH > 9).
For the data of Figure 6, the second term of the numerator
dominates and KHa OCl . [H+ ], so eq 2 simplifies to give the
observed pseudo-second-order rate constant:
the rate constant that Kice and Puls thought they were
measuring), which is contained in the intercept of the line of
the inset of Figure 7. The values of kHOCl and kOCl will be
evaluated when the rate law is derived in the Discussion.
kHOCl ·[H+]
Discussion
kobs
)
(3)
KHa OCl
Mechanism of the Hydrolysis of ArSO2Cl. The kinetics and
mechanism of the hydrolysis of aliphatic and aromatic sulfonyl
chlorides have been previously investigated in detail.11,12 Our
objectives here were simply to establish whether TNBO2Cl
behaves as a typical aryl sulfonyl chloride and (in the case of
PhSO2Cl) to extend the range of reaction conditions to include
the conditions we employed in our kinetic studies of the
oxidation reaction (namely to employ I ) 1.0 and to extend
the pH above pH 11). The rate of hydrolysis of ArSO2Cl can
be described by the following rate law:
Solving for kHOCl and inserting [H+] ) 10-9 M and KaHOCl
)
10-7.4 M gives a value of
kobs ·KHa OCl
2.32 × 107 M-1 -1 ·10-7.4
s M
kHOCl
)
)
)
[H+]
10-9
M
9.2 ( 0.3 × 108 M-1 s-1
(4)
The value for kHOCl from eq 4 is consistent with values we have
j
-d[ArSO2Cl]/dt ) kobs[ArSO2Cl] ) (k0 +
determined by SVD analysis of data at higher pH (Table 2, X
) (2.4 ( 0.8) × 108 M-1 s-1), where the rate of hydrolysis is
competitive with the rate of oxidation. Thus, kHOCl is very close
to the diffusion limit. It is also possible to extrapolate a value
of kHOCl using the data of Figure 7. The linear fit of the data of
Figure 7 (inset) suggests that kHOCl still influences the rate above
pH 13. Applying eq 3 to the slope of the line, multiplying by
[OH-] (because plot is 1/[OH-] vs kobs), multiplying by [HOCl]0
) 1 mM (pseudo-first-order-conditions), and solving for kHOCl
(while substituting Kw ) [H+][OH-] ) 10-13.8 M2)13 gives
kOH[OH-])[ArSO2Cl]
(1)
We have interpreted the rate law as hydrolysis by H2O (k0) and
by OH- (kOH). The data we have measured for TNBO2Cl are
described by eq 1. While we did not reinvestigate the rate of
hydrolysis of PhSO2Cl below pH 13, our value of kOH ) 27 (
1 M-1 s-1 for I ) 1 (Figure 4) is comparable to the value of
kOH ) 28.2 M-1 s-1 that has been reported previously for pH <
11 and I ) 50 mM.11
Mechanism of the Reaction of PhSO2- with HOCl/OCl-.
We conclude that the mechanism of Scheme 1 is operative
because (1) the reaction of PhSO2- with HOCl/OCl- quantita-
tively yields PhSO2Cl at pH < 8, conditions under which
hydrolysis is relatively slow (cf. Figure 1); (2) PhSO2Cl is
produced as a transient under conditions where the rate of
hydrolysis is competitive with rate of oxidation (Figure 7); and
(3) the rate of oxidation of PhSO2- by HOCl/OCl- exhibits an
inverse dependence on [OH-], even under very alkaline condi-
tions (Figure 7, inset). Assuming that the Brønsted acid/base
reaction is fast (diffusion-controlled), the following rate law is
readily derived (for pH g 9, where OCl- dominates over HOCl):
slope·KHa OCl
kHOCl
)
)
Kw ·[HOCl]0
0.1034 ( 0.003 M-1
(10-13.8 M)·(0.001 M)
3.3 ( 0.1 × 108 M-1 s-1
s
-1 ·10-7.4
M
)
(5)
The values of kHOCl that were determined from the data of Figure
6, Table 2, and Figure 7 (inset) are self-consistent (within a
factor of 4 over a ∆[H+] of 4 orders of magnitude). In addition,
the intercept of the line of Figure 7 (inset, which is proportional
to kOCl) appears to be significant; we estimate kOCl to be 98 (
-
SCHEME 1: Kinetics and Mechanism of the Oxidation of PhSO2 with HOCl/OCl-
HOCl h H+ + OCl-
pKa ) 7.4
-
PhSO2H h H+ + PhSO2
pKa ) 1.4
+
H
-
PhSO2 + OCl- 8 PhSO2Cl + OH-
kOCl ) 98 ( 9 M-1 s-1
-
PhSO2 + HOCl f PhSO2Cl + OH-
kHOCl ) 9.2 ( 0.3 × 108 M-1 s-1
-
PhSO2Cl + H2O f PhSO3 + 2H+ + Cl- k0 ) 1.93 × 10-3 s-1
-
PhSO2Cl + OH- f PhSO3 + H+ + Cl- kOH ) 27 ( 1 M-1 s-1

1676 J. Phys. Chem. A, Vol. 114, No. 4, 2010
Ueki et al.
9 M-1 s-1. Therefore, about a 7 orders of magnitude difference
exists between kHOCl and kOCl, and 6.6 decades of difference
exists between the pKa of HOCl (OCl- would be protonated
below pH 7.4 and the rate of HOCl would reach a maximum)
and pH 13. Thus, at about pH 13, the reaction of OCl- is
competitive with the reaction of HOCl. Scheme 1 summarizes
the acid dissociation of HOCl7 and PhSO2H,14 the rate data we
report for the oxidation of PhSO2- with HOCl/OCl- (using the
value of kHOCl that was measured at the lowest possible pH)
and the subsequent hydrolysis of PhSO2Cl (the value of k0 in
Scheme 1 is from the literature).11
M-1 s-1, similar to that of the corresponding thiolates, and nearly
diffusion-controlled. In contrast, the rate constant for the reaction
-
of OCl- with ArSO2 is approximately 7 orders of magnitude
smaller.
Acknowledgment. This research was funded by the National
Science Foundation (CHE-0503984).
Supporting Information Available: UV-vis spectra of
-
-
2-
2-
PhSO2
, PhSO3 , PhSO2Cl, TNBO2 , TNBO3 , and
TNBO2Cl. This material is available free of charge via the
Internet at http://pubs.acs.org.
Previous Mechanistic Studies. Besides the work of Kice
and Puls that has been the subject of the present paper,5 there
have been few previous mechanistic studies of the oxidation of
sulfinates.15 It has been previously suggested that the oxidation
of cysteine by ClO2 under acidic condition involves a
cysteinyl-ClO2 adduct that subsequently hydrolyzes to give
cysteine sulfinic acid and HOCl in a rate-limiting step, with
subsequent oxidation of the sulfinate by HOCl to give cysteine
sulfonate.16 However, the kinetics of the reaction of cysteine
sulfinate with HOCl to give the sulfonate were not investigated.
References and Notes
(1) References to compounds without charges are inclusive of all acid/
base derivatives (e.g., TNB). Reference to specific proton states are indicated
with a charge (e.g., TNB0, TNB-1, and TNB-2).
(2) Nagy, P.; Ashby, M. T. J. Am. Chem. Soc. 2007, 129, 14082.
Gerritsen, C. M.; Margerum, D. W. Inorg. Chem. 1990, 29, 2757. Fogelman,
K. D.; Walker, D. M.; Margerum, D. W. Inorg. Chem. 1989, 28, 986.
Kumar, K.; Margerum, D. W. Inorg. Chem. 1987, 26, 2706.
(3) Pattison, D. I.; Davies, M. J. Chem. Res. Toxicol. 2001, 14, 1453.
(4) Nagy, P.; Ashby, M. T. Chem. Res. Toxicol. 2005, 18, 919. Nagy,
P.; Ashby, M. T. Chem. Res. Toxicol. 2007, 20, 79.
(5) Kice, J. L.; Puls, A. R. J. Am. Chem. Soc. 1977, 99, 3455.
(6) Landino, L. M.; Mall, C. B.; Nicklay, J. J.; Dutcher, S. K.;
Moynihan, K. L. Nitric Oxide 2008, 18, 11.
(7) Adam, L. C.; Fabian, I.; Suzuki, K.; Gordon, G. Inorg. Chem. 1992,
31, 3534.
-
The rate constants for the reactions of ArSO2 with HOCl of
ca. 109 that have been reported herein are unexpectedly large.
Given the electrophilic nature of HOCl, it follows that the best
nucleophiles should be the most reactive. This is the case for
the amino acids, where the most nucleophilic sulfur donors
cysteine and methionine are most reactive.3 A similar situation
is found for inorganic ions.17 Given the fact that sulfinates are
considerably less nucleophilic than thiolates,18 it is surprising
(8) Wang, T. X.; Margerum, D. W. Inorg. Chem. 1994, 33, 1050. Eigen,
M.; Kustin, K. J. Am. Chem. Soc. 1962, 84, 1355. Lifshitz, A.; Perlmutter-
Hayman, B. J. Phys. Chem. 1962, 66, 701.
(9) Tonomura, B.; Nakatani, H.; Ohnishi, M.; Yamaguchi-Ito, J.;
Hiromi, K. Anal. Biochem. 1978, 84, 370.
-
that ArSO2 reacts with HOCl with rate constants that are
several orders of magnitude larger than those for thiolates.3
As an aside, it has recently become appreciated that sulfinic
acid modifications in proteins link protein function to cellular
oxidative status,19 and that HOCl is produced by the oxidation
of Cl- with H2O2 in a reaction that is catalyzed by human
neutrophilic myeloperoxidase.20 Accordingly, we are particularly
interested in the possible role of the oxidation of sulfinates by
HOCl in the context of biomarkers21 for myeloperoxidase-
induced oxidative stress (that may be linked to inflammatory
diseases22). While we had previously thought it was improbable
that sulfinates could compete kinetically with typical antioxi-
dants in vivo (such as the ubiquitous thiol-containing tripeptide
glutathione) the results of the present study have caused us to
reconsider that position.
(10) Dunn, B. C.; Meagher, N. E.; Rorabacher, D. B. J. Phys. Chem.
1996, 100, 16925.
(11) Rogne, O. J. Chem. Soc. B 1968, 1294.
(12) Kevill, D. N.; Park, B.-C.; Park, K.-H.; D’Souza, M. J.; Yaakoubd,
L.; Mlynarski, S. L.; Kyong, J. B. Org. Biomol. Chem. 2006, 4, 1580. Koo,
I. S.; Yang, K.; Park, J. K.; Woo, M. Y.; Cho, J. M.; Lee, J. P.; Lee, I.
Bull. Korean Chem. Soc. 2005, 26, 1241. Koo, I. S.; Yang, K.; Shin, H. B.;
An, S. K.; Lee, J. P.; Lee, I. Bull. Korean Chem. Soc. 2004, 25, 699. Ivanov,
S. N.; Kislov, V. V.; Gnedin, B. G. Russ. J. Gen. Chem. 2004, 74, 95.
(13) Martell, A. E.; Smith, R. M. Inorganic Complexes; Critical Stability
Constants, Vol. 4; Plenum Press: New York, 1976.
(14) Ogata, Y.; Sawaki, Y.; Isono, M. Tetrahedron 1970, 26, 3045.
(15) Gancarz, R. A.; Kice, J. L. Tetrahedron Lett. 1980, 21, 1697.
(16) Ison, A.; Odeh, I. N.; Margerum, D. W. Inorg. Chem. 2006, 45,
8768.
(17) Nagy, P.; Lemma, K.; Ashby, M. T. Inorg. Chem. 2007, 46, 285.
(18) Ritchie, C. D.; Virtanen, P. O. I. J. Am. Chem. Soc. 1972, 94, 4966.
Ritchie, C. D.; Saltiel, J. D.; Lewis, E. S. J. Am. Chem. Soc. 1961, 83,
4601. Okuyama, T. In The Chemistry of Sulphinic Acids, Esters, and Their
DeriVatiVes; Patai, S., Ed.; John Wiley & Sons: New York, 1990; p 639.
(19) Jacob, C.; Holme, A. L.; Fry, F. H. Org. Biomol. Chem. 2004, 2,
1953.
(20) Klebanoff, S. J. J. Leukocyte Biol. 2005, 77, 598.
(21) Harwood, D. T.; Nimmo, S. L.; Kettle, A. J.; Winterbourn, C. C.;
Ashby, M. T. Chem. Res. Toxicol. 2008, 21, 1011.
Conclusions
The oxidation of ArSO2- (PhSO2- and TNBO22-) with HOCl/
OCl- proceeds via a conventional pathway: nucleophilic attack
by ArSO2 on HOCl with concomitant Cl+ transfer to give a
-
(22) Ashby, M. T. J. Dent. Res. 2008, 87, 900. Pattison, D. I.; Davies,
M. J. Curr. Med. Chem. 2006, 13, 3271.
sulfonyl chloride intermediate (ArSO2Cl), which we have
identified spectrophotometrically. Remarkably, the rate constant
for the reaction of HOCl with ArSO2 is on the order of 109
JP906651N
-