Kinetics and mechanisms of S(IV) reductions of bromite and chlorite ions

Article abstract of DOI:10.1021/ic020475m

The reaction of bromite with aqueous S(IV) is first order in both reactants and is general-acid catalyzed. The reaction half-lives vary from 5 ms (p[H⁺] 5.9) to 210 s (p[H⁺] 13.1) for 0.7 mM excess S(IV) at 25°C. The proposed mechanism includes a rapid reaction (k₁ = 3.0 × 10⁷ M^-1 s^-1) between BrO₂^- and SO₃^2- to form a steady-state intermediate, (O₂BrSO₃)^3-. General acids assist the removal of an oxide ion from (O₂BrSO₃)^3- to form OBrSO₃-, which hydrolyzes rapidly to give OBr^- and SO₄^2-. Subsequent fast reactions between HOBr/OBr^- and SO₃^2- give Br^- and SO₄^2- as final products. In contrast, the chlorite reactions with S(IV) are 5-6 orders of magnitude slower. These reactions are specific-acid, not general-acid, catalyzed. In the proposed mechanism, ClO₂^- and SO₃H^-/SO₂ react to form (OClOSO₃H)^2- and (OClOSO₂)^- intermediates which decompose to form OCl^- and SO₄^2-. Subsequent fast reactions between HOCl/OCl^- and S(IV) give Cl^- and SO₄^2- as final products. SO₂ is 6 orders of magnitude more reactive than SO₃H^-, where k₅(SO₂/ClO₂^-) = 6.26 × 10⁶ M^-1 s^-1 and k₆(SO₃H^-/ClO₂^-) = k₆(SO₃H^-/ClO₂^-) = 5.5 M^-1 s^-1. Direct reaction between ClO₂^- and SO₃^2- is not observed. The presence or absence of general-acid catalysis leads to the proposal of different connectivities for the initial reactive intermediates, where a Br-S bond forms with BrO₂^- and SO₃^2-, while an O-S bond forms with ClO₂^- and SO₃H^-.

Full text of DOI:10.1021/ic020475m

Inorg. Chem. 2003, 42, 78−87
Kinetics and Mechanisms of S(IV) Reductions of Bromite and
Chlorite Ions
Kara E. Huff Hartz, Jeffrey S. Nicoson, Lu Wang, and Dale W. Margerum*
Department of Chemistry, Purdue UniVersity, West Lafayette, Indiana 47907-1393
Received July 22, 2002
The reaction of bromite with aqueous S(IV) is first order in both reactants and is general-acid catalyzed. The
+
+
reaction half-lives vary from 5 ms (p[H ] 5.9) to 210 s (p[H ] 13.1) for 0.7 mM excess S(IV) at 25 °C. The proposed
7
-1 -1
-
2-
mechanism includes a rapid reaction (k
intermediate, (O ) . General acids assist the removal of an oxide ion from (O
which hydrolyzes rapidly to give OBr and SO
1
) 3.0 × 10 M s ) between BrO
2
and SO
BrSO
. Subsequent fast reactions between HOBr/OBr and SO
3
to form a steady-state
)3- to form OBrSO -,
give
3
-
2
BrSO
3
2
3
3
-
2-
-
2-
4
3
-
2-
Br and SO
4
as final products. In contrast, the chlorite reactions with S(IV) are 5−6 orders of magnitude slower.
-
H-/SO
These reactions are specific-acid, not general-acid, catalyzed. In the proposed mechanism, ClO
2
and SO
2-. Subsequent
is 6 orders of magnitude
H-/ClO -) ) 5.5 M-1 s-1.
is not observed. The presence or absence of general-acid catalysis
3 2
-
-
-
react to form (OClOSO
3
H)2 and (OClOSO
2
) intermediates which decompose to form OCl and SO
4
-
-
2-
fast reactions between HOCl/OCl and S(IV) give Cl and SO
4
as final products. SO
2
-
-
) ) 6.26 × 10 M s-1 and k
6
-1
more reactive than SO
3
H , where k
5
(SO
and SO
3
2
2-
/ClO
2
6
(SO
3
2
-
Direct reaction between ClO
2
leads to the proposal of different connectivities for the initial reactive intermediates, where a Br−S bond forms with
2 3 2 3
and SO and SO H .
2-, while an O−S bond forms with ClO
-
-
-
BrO
Introduction
The mechanisms of the hypohalous acid (HOX, X ) Cl
has set the maximum contaminant level for drinking water
-
1
- 4
3
at 10 µg L BrO .
5
Lee and Lister reported the kinetics of the reaction
between bromite and sulfite in basic solutions. They observed
kinetics that were first order in bromite and in sulfite and
proposed that the reaction proceeds by two stages (eqs 1
and 2). They found the rate expression in eq 3 for the reaction
-
-
2-
or Br) reduction with many nucleophiles (I , NO
2
, SO
3
,
-
-
-
Br , CN , ClO
2
) proceed by halogen(I) transfer to the
1
nucleophile with halide release. However, little is known
about the reaction mechanisms for the reduction of halite
-
ions (XO
2
). This is due in part to the instability of aqueous
-
-1 -1
where [OH ] ) 0.016-0.046 M, k
)
a
) 3.73 M
s
and k
b
halite ions (chlorite and bromite are stable only in basic
solution, and iodite is a transient species) and due to the
lack of commercially available bromite salts. Halites are often
-
1
0.121 s at 25.0 °C, and µ ) 1.0 M. They assigned the
-
2-
k
a
pathway to a direct reaction of BrO
pathway to a reaction between HBrO
2
and SO
3
and the
-
k
b
2
and SO 2
3
and
2
encountered in oscillating reactions, and bromite is an
proposed that the reaction proceeds by a simple oxygen-
atom transfer mechanism. In our work, we measure the
intermediate for bromate ion formation in the disinfection
3
of drinking water. Bromate is a carcinogen, and the USEPA
-
+
2
kinetics of BrO /S(IV) reactions over an extended p[H ]
range. We show the strong effects of buffers and pH on the
*
To whom correspondence should be addressed. E-mail: margerum@
purdue.edu.
1) (a) Kumar, K.; Day, R. A.; Margerum, D. W. Inorg. Chem. 1986, 25,
-
BrO
2
/S(IV) reaction rates and propose a mechanism to
(
account for these effects.
4
344-4350. (b) Kumar, K.; Margerum, D. W. Inorg. Chem. 1987,
2
6, 2706-2711. (c) Fogelman, K. D.; Walker, D. M.; Margerum, D.
W. Inorg. Chem. 1989, 28, 986-993. (d) Gerritsen, C. M.; Margerum,
D. W. Inorg. Chem. 1990, 29, 2757-2762. (e) Troy, R. C.; Margerum,
D. W. Inorg. Chem. 1991, 30, 3538-3543. (f) Johnson, D. W.;
Margerum, D. W. Inorg. Chem. 1991, 30, 4845-4850. (g) Gerritsen,
C. M.; Gazda, M.; Margerum, D. W. Inorg. Chem. 1993, 32, 5739-
5748. (h) Furman, C. S.; Margerum, D. W. Inorg. Chem. 1998, 37,
4321-4327. (i) Jia, Z.; Margerum, D. W.; Francisco, J. S. Inorg. Chem.
2000, 39, 2614-2620.
(2) (a) Orb a´ n, M.; Epstein, I. R. J. Phys. Chem. 1995, 99, 2358-2362.
(b) Faria, R. D.; Lengyel, I.; Epstein, I. R. J. Phys. Chem. 1993, 97,
1164-1171. (c) De Kepper, P.; Boissonade, J.; Epstein, I. J. Phys.
Chem. 1990, 94, 6525-6536.
(3) von Gunten, U.; Oliveras, Y. EnViron. Sci. Technol. 1998, 32, 63-
70.
(4) Fed. Regist. 1998, 63, 69390.
(5) Lee, C. L.; Lister, M. W. Can. J. Chem. 1979, 57, 1524-1530.
78 Inorganic Chemistry, Vol. 42, No. 1, 2003
10.1021/ic020475m CCC: $25.00 © 2003 American Chemical Society
Published on Web 11/22/2002

S(IV) Reductions of Bromite and Chlorite Ions
BrO2- + SO32- k
-
-
2-
these effects, and the reactions of oxybromine anions (OBr ,
9
8 OBr + SO4
(1)
(2)
-
-
-
-
BrO
and ClO
2
, and BrO
3
) and oxychlorine anions (OCl , ClO
2
,
-
rapid
) with S(IV) are compared and contrasted.
-
2-
-
2-
3
OBr + SO3
8 Br + SO4
BrO2- + SO32- f OBr + SO
-
2-
EB ) 1.52 V (7)
o
-
4
-
d[BrO ]
kb
OH-]
2
-
2-
3
)
k[BrO ][SO ] k ) k +
(3)
-
2-
-
2-
o
2
a
dt
OBr + SO f Br + SO
4
3
EB ) 1.70 V (8)
[
ClO2- + SO3 f OCl + SO
2-
-
2-
EB ) 1.60 V (9)
o
4
The corresponding chlorite/S(IV) reaction has been studied
somewhat more than the bromite/S(IV) reaction. Halperin
and Taube stated that the reaction is essentially complete
-
2-
-
2-
o
OCl + SO f Cl + SO
EB ) 1.83 V (10)
3
4
6
within 1 min in 0.1 M HCl and at pH 5. They suggested
that the reaction proceeds by oxygen-atom transfer. Adam
and Gordon used basic sulfite solutions (pH > 10.5) to
Experimental Section
Reagents. All solutions were prepared using distilled, dionized
18 MΩ cm) water that was boiled to remove CO and sparged
with Ar(g) to remove dissolved oxygen. NaClO was recrystallized
before use and standardized gravimetrically. Carbonate-free NaOH
was standardized against potassium hydrogen phthalate. Na SO
7
(
2
remove hypochlorite selectively from chlorite solutions at
room temperature for subsequent titration and chromato-
graphic analysis. This indicates that the chlorite/sulfite
4
2
3
8
reaction is relatively slow at this pH. Nagyp a´ l et al. found
solutions (>98.5%) were prepared just before use and were
-
2-
13
the ClO
2
/SO
3
reaction is complete within a few seconds
standardized by iodimetry. Several techniques were utilized to
at 90 °C in a continuous-stirred-tank reactor above pH 12.
Frerichs et al. gave rate constants at 25 °C for the three
pathways in eqs 4-6. These are from the unpublished data
of Rushing and Thompson. In our work, we see no
exclude oxygen from the sulfite solutions: sonication of solutions
for the dissolution of the solid salt; continuous Ar(g) sparging;
9
transfer of solutions by gastight syringes. NaClO
2
was recrystal-
1h,14
lized
and standardized spectrophotometrically, where ꢀ260 ) 154
-1
-1 1h
12
M
cm
was modified (by more careful cleanup methods) to give improved
purity with 63% NaBrO (by weight), 14% NaOH, 1.9% Na SO
0.08% NaNO , 7.0% NaBrO , 1.3% NaBr, and waters of
crystallization. The hydroxide content of the bromite salt was
. The previously reported synthesis of sodium bromite
contribution from rate 1 (eq 4) and find different values for
-
rates 2 and 3 for the ClO
2
/S(IV) reaction. Furthermore, we
2
2
4
,
-
show buffers have no effect on the ClO
rates.
2
/S(IV) reaction
<
3
3
measured by adding Na
the resulting p[H ] change caused by OH reaction with HPO
Sulfate, nitrate, bromate, and bromide have no effect on the
observed kinetics under the conditions of the experiments. Bromite
2
HPO
4
to a bromite stock and measuring
rate 1 ) k ′[ClO ][SO ] k ′ ) 1.2 × 10 M-1 s-1 (4)
-
2-
-2
+
-
2-
.
1
2
3
1
4
rate 2 ) k ′[ClO ][SO H ] k ′ ) 24.7 M s-1 (5)
-
2
-
-1
2
3
2
solutions were standardized spectrophotometrically, where ꢀ295
)
-
1
-1 15
1
15 M cm .
-
+
-
rate 3 ) k ′[ClO ][H ][SO H ]
Methodology and Instrumentation. The measured pH was
3
2
3
+
k ′ ) 2.19 × 10 M s-1 (6)
8
-2
converted to p[H ] on the basis of electrode calibration at µ ) 1.0
3
4
M (NaClO ). Ion chromatographic data were obtained with a Dionex
12
DX-500 instrument as described previously. UV-vis spectra and
kinetic scans were obtained on a Perkin-Elmer Lambda-9 UV-
vis-NIR spectrophotometer. Quartz cells were used and were
thermostated at 25.0(1) °C. Data for faster reactions were obtained
on an Applied Photophysics stopped-flow spectrometer, equipped
with a 0.962 cm path length cell. At least 10 pushes/reaction were
The electrode potentials in basic solution for the two-step
-
-
2-
reduction of BrO
2
and ClO
2
by SO
3
are shown in eqs
10
-
-
7
-10. The BrO
2
/OBr potential is calculated on the basis
-
- 10
- 11
-
12
of the BrO
3
/OBr , BrO
couples. The thermodynamic driving forces for the BrO
and the ClO /SO
2 3
work shows that despite the similarities in the reaction
potentials and products, the BrO
than 5 orders of magnitude faster than the ClO
reactions. The reaction kinetics suggest that BrO
reactions and ClO
2
/BrO
3
,
2 2
and BrO /BrO
-/
2
2-
-
2-
16
averaged for analysis, and SigmaPlot v. 8.0 was used for data
SO
3
reactions are comparable. Our
analysis. Observed first-order rate constants (kobs) greater than 50
-
1
-
s
were corrected for mixing according to eq 11, where kcorr
)
2
/S(IV) reactions are more
-
1 17
-
corrected rate constant and kmix ) 4620(120) s .
2
/S(IV)
/S(IV)
-
2
-1
1
1
-
kcorr
)
-
(11)
2
/S(IV) reactions occur by different
(k
kmix)
obs
pathways. Reaction mechanisms are proposed to account for
The halite/sulfite reaction rates were measured when the [S(IV)]/
(
(
(
6) Halperin, J.; Taube, H. J. Am. Chem. Soc. 1952, 74, 375-380.
7) Adam, L. C.; Gordon, G. Anal. Chem. 1995, 67, 535-540.
8) Nagyp a´ l, I. R.; Epstein,; Kustin, K. Int. J. Chem. Kinet. 1986, 18,
-
[
XO
2
] ratio was greater than 20/1 to maintain pseudo-first-order
conditions. The loss of bromite was followed at 295 nm (ꢀ ) 115
3
45-353.
(
9) Frerichs, G. A.; Mlnarik, T. M.; Grun, R. J.; Thompson, R. C. J. Phys.
Chem. A 2001, 105, 829-837.
10) Standard Potentials in Aqueous Solutions; Bard, A. J., Parsons, R.,
(13) Jeffery, G. H.; Bassett, J.; Mendham, J.; Denney, R. C. Vogel’s
Textbook of QuantitatiVe Chemical Analysis, 5th ed.; Longman
Scientific & Technical: Essex, U.K., 1980; pp 384-398.
(14) Peintler, G.; Nagyp a´ l, I.; Epstein, I. R. J. Phys. Chem. 1990, 94, 2954-
2958.
(
Jordan, J., Eds.; Marcel Dekker: New York, 1985; pp 74-75, 79-
82, 102.
(
11) Stanbury, D. M. In AdVances in Inorganic Chemistry; Sykes, A. G.,
Ed.; Academic Press: San Diego, CA, 1989; Vol. 33, pp 69-137.
12) Wang, L.; Nicoson, J. S.; Huff Hartz, K. E.; Fransisco, J. S.; Margerum,
D. W. Inorg. Chem. 2002, 41, 108-113.
(15) (a) Lee, C. L.; Lister, M. W. Can. J. Chem. 1971, 49, 2822-2826.
(b) Perrone, T. F. Ph.D. Thesis, Purdue University, 1999; pp 89-91.
(16) SigmaPlot V. 8.0 for Windows; SPSS Inc.: Chicago, IL, 2002.
(17) Nicoson, J. S.; Margerum, D. W. Inorg. Chem. 2002, 41, 342-347.
(
Inorganic Chemistry, Vol. 42, No. 1, 2003 79

Huff Hartz et al.
Figure 1. Sulfite dependence of the halite/S(IV) reaction where kobs )
slope[SO3]T (y-intercepts are zero within the error): (∆) bromite data where
-
1
-1
slope from fit equals 20.9(5) M
s ; (o) chlorite data where slope from
-
3
-1 -1
-
fit equals 2.5(1) × 10
M
s . Conditions: 0.659 mM BrO2 , 30.0 mM
+
-
+
[
PO4]T, p[H ] 11.55(3); 4.0 mM ClO2 , 50 mM [CO3]T, p[H ] 9.53(3), µ
1.0 M, 25.0(1) °C. Insert shows kobs dependence on [SO3]T for the bromite/
)
-
2-
-
+
5
-1 -1
Figure 2. Buffer dependence of BrO2 /SO
reaction. (a) H PO data
sulfite reaction (0) at lower p[H ]. Slope ) 2.56(5) × 10
M
s
.
3
2
4
-
-
+
-
-
+
3 rxn a 3 a 3
are corrected for [SO H ] by k (K (SO H ) + [H ])/K (SO H ). Condi-
Conditions: 0.075 mM BrO2 , 40 mM [PO4]T, p[H ] 7.10(1).
-
+
tions: 0.105 mM BrO2 , 0.921 mM S(IV), p[H ] 6.59(3), slope ) 3.4(7)
6
-2 -1
×
10 M
s . Error bars reflect the standard deviation in determination.
M- cm
1
-1 15
)
or 240 nm (ꢀ ) 620 M cm-1). The loss of chlorite
-1
(
b) B(OH)3 data are corrected for boric acid/sulfite complex by eq 21.
-
1
-1 1h
was followed at 260 nm (ꢀ ) 154 M cm
)
or 290 nm (ꢀ ) 97
-
+
Conditions: 0.50 mM BrO2 , 6.08(4) mM S(IV), p[H ] 8.98(2), slope )
-1
-1
2-
5
-2 -1
-
-
M
cm ; this wavelength was used when SO
3
absorbance
1.06(5) × 10 M
s . (c) HCO3 conditions: 0.96 mM BrO2 , 30.8(2)
+
4
-2 -1
2-
-
mM S(IV), p[H ] 9.53(1), slope ) 1.50(6) × 10 M
s . (d) HPO4
overwhelms the ClO
2
signal). The rates of loss of bromite and
-
+
conditions: 0.80 mM BrO2 , 30.7(2) mM S(IV), p[H ] 10.48(2), slope )
chlorite were first order (eq 12), and the data fit an integrated rate
expression (eq 13), where kobs ) observed rate constant, A )
absorbance, t ) time, i ) initial, and ∞ ) final.
3
-2 -1
1
.5(2) × 10 M
s .
3 T
in [SO ] (Figure 1), and the rate expression for the halite/
S(IV) reaction is shown in eq 15.
-
d[XO2-]
)
kobs[XO2-]
(12)
(13)
-
dt
-d[XO ]
kobs
2
-
)
k [XO ][SO ]
3 T
krxn
)
(15)
-
kobst
rxn
2
dt
[SO ]
3 T
A ) A + (A - A )e
t
∞
i
∞
The first-order dependences in [XO
that the rate-determining steps involve the reduction of halite
-] and [SO
] indicate
2 3 T
Results and Discussion
Stoichiometry and Reaction Order with Respect to
Halite and Sulfite. Ion chromatography and UV-vis ab-
sorbance spectrometry were used to verify the reaction
stoichiometries. Bromide and sulfate are the only products
observed from the bromite/S(IV) reaction, and chloride and
-
1c
to hypohalite (eq 16). The HOCl/OCl reactions with sulfite
-
1e
and the HOBr/OBr reactions with sulfite are known to
be much faster than the measured reaction rates. Therefore,
subsequent hypohalite reduction (eq 17) is not observed
kinetically.
sulfate are the only products observed from the chlorite/S(IV)
-
reaction. For the BrO
2
/S(IV) reaction, the average ratio of
XO2- + SO3 f OX + SO
2-
-
2-
(16)
(17)
4
-
-
-
initial [BrO
2
] to final [Br ] is 1.01(2) (after [Br ] from the
-
-
2- rapid
-
2-
bromite stock is subtracted). For the ClO
the average ratio of initial [ClO
The stoichiometry for the four-electron reduction of halite
by sulfite is given by eq 14.
2
/S(IV) reaction,
] to final [Cl ] is 1.03(5).
OX + SO
8 X + SO
3
4
-
-
2
-
2-
Buffer Dependence of the BrO /SO Reaction. The
2
2-
3
-
rate constants for the BrO /SO
2
3
reaction increase with
increasing phosphate ([PO
([Bor] ) buffer concentration, where acidic forms of the
buffers accelerate the reaction. Plots of krxn vs the acidic form
4 T 3 T
] ), carbonate ([CO ] ), and borate
XO2- + 2SO3 f X + 2SO4
2-
-
2-
(14)
T
2-
The sulfite concentration was varied to establish the
of the buffer (HA) are given in Figure 2, where the HPO
4
,
-
2-
-
-
reaction order in [SO
]
3 T
where [SO
3
]
T
) [SO
3
H ] + [SO
3
]
HCO
3 2 4
, and H PO concentrations are calculated from the
2
-
+
[SO ]. The reactions of halites with sulfite are first order
2
corresponding acid dissociation constants (K
a
(HPO
4
) )
80 Inorganic Chemistry, Vol. 42, No. 1, 2003

S(IV) Reductions of Bromite and Chlorite Ions
0-
M).
11.08
-) ) 10-9.57 M, K -) ) 10-6.26
a 2 4
(H PO
The reaction rate increases linearly as the con-
1
M, K
a
(HCO
3
1
h,18,19a
centration of the general acid increases.
Borate buffer also affects the rate of BrO
-/S(IV) reaction.
2
Boric acid is known to form complexes with several
-
-
- 20
oxyanions, including OCl , OBr , and NO
2
.
For this
reason, bromite and sulfite solutions were examined for boric
acid complex formation. Spectral scans showed no shift in
+
the bromite spectrum upon addition of borate buffer (p[H ]
8.6) so that bromite/boric acid complex formation appears
to be negligible. However, boric acid causes the sulfite
absorbance to decrease, due to the formation of a complex.
BS
A boric acid/sulfite complex equilibrium constant (K ) is
defined as shown in eq 18.
KBS
2
-
2-
B(OH) + SO3
{
} (SO B(OH) )
(18)
Figure 3. [H+] dependence of the BrO -/SO 2- reaction in absence of
3
3
3
2
+
3
-
14
-11
buffers, 7.7 × 10 -1.4 × 10
M H , µ ) 1.0 M, and 25.0(1) °C:
S
2-
HS
-
-1 -1
12
-2 -1
intercept ) 4.50(3) M
s
; slope ) 3.03(4) × 10
M
s .
A ) b(ꢀ [SO ] + ꢀ [SO H ] +
T
3
3
ꢀBS[(SO B(OH) ) ]) (19)
2-
-
no more than 1% of BrO
2
would be complexed to B(OH)
3
.
3
3
-
This may explain why no BrO
served spectrophotometrically.
2
/B(OH)
3
complex is ob-
The KBS value is calculated from absorbance data on the basis
of eq 19 for total absorbance (A ) at any wavelength, where
b ) 0.10 cm path length and ꢀ ) molar absorptivities of S
T
Borate buffer accelerates the bromite/sulfite reaction but
also simultaneously decreases the rate because a fraction of
(
sulfite), HS (hydrogen sulfite), and BS (boric acid/sulfite
2-
[
3 T 3 3
SO ] is sequestered as the unreactive (SO B(OH) )
-
complex), respectively. The absorbance of SO
3
H is negli-
BS
species. For this reason, krxn must be corrected for K as
gible above 230 nm,21 and the measured absorbances were
corrected for the small borate buffer and NaClO absorbance.
Equation 20 is derived by incorporating the equation for total
-
well as K
a
(SO
3
H ) (eq 21). With this correction, krxn increases
4
linearly with increasing [B(OH)
3
] (Figure 2).
) 6.30(5)) also acts as a
-
23
Hydrogen sulfite (SO
3
H , pK
a
general acid and has an acid strength similar to that of
S
BS BS
-
(
ꢀ + ꢀ K [B(OH) ])bK (SO H )[SO ]
3 a 3 3 T
-
H
2
PO
4
a
(pK ) 6.26). Under our conditions, however, the
AT )
(20)
-
+
-
BS
-
-
K (HSO ) + [H ] + K (SO H )K [B(OH) ]
concentration of SO
and so the SO
3
H is low relative to that of H
H contribution to the general-acid-catalyzed
2
PO
4
,
a
3
a
3
3
-
3
) [SO 2
-] + [SO
-
sulfite species ([SO
3
]
T
3
3
3
H ] + [(SO B-
rate is negligible.
2
-
(
OH)
3
-
) ]), eq 19, and the equilibrium constant between
2-
krxn
)
SO
3
H and SO
3
. The concentration of B(OH)
3
is calculated
after correction for the
formation of polyborates (see Supporting Information for
-
8.85
22a
-
+
-
BS
using K
a
(B(OH)
3
) )10
M
kobs (K (SO H ) + [H ] + K (SO H )K [B(OH) ])
a
3
a
3
3
-
[
SO3]
T
K (SO H )
2
2b
a
3
details). The measured absorbances at three wavelengths
240, 245, and 250 nm) were curve fit to eq 20, where the
calculated sulfite/boric acid equilibrium complex value is K
7.6(5) M . Midgely developed a linear free-energy
relationship for boric acid complexes with univalent ions (K)
on the basis of the pK value of the conjugate acid of the
univalent ion (log K ) -2.15(58) + 0.45(5)pK
predicted by this equation is 4.8(4) M , which is in
reasonable agreement with the value obtained here. Midg-
ely’s equation also predicts (on the basis of pK
.59(5), measured under the conditions of this experiment;
(21)
(
BS
+
-/SO
2-
[
H ] Dependence of the BrO
2
3
Reaction. The
-
1
20
)
+
[
H ] concentration was varied by adding different amounts
-
of NaOH to BrO
2
solutions and observing the reaction with
a
2-
+
SO
3
. A plot of krxn vs [H ] is shown in Figure 3. The plot
BS
a
). The K
-14
-11
+
is linear in the range of 7.7 × 10 -1.4 × 10
M H .
s
-
1
-1 -1
The intercept and slope of this plot are 4.50(3) M
and
12
-2 -1
+
3
.03(4) × 10 M s , respectively. The [H ] dependences
2- 3- - 2-
4 4 3 3
a
(HBrO
2
) )
were also resolved in HPO
/PO
, HCO
/CO
, and
3
borate buffers, and the slopes were in agreement, after
-
see Figure S2) that the BrO
2
/B(OH)
3
complex formation
-12
correction for the buffer acid contribution, from 3.7 × 10
BB
-1
constant (K ) would be 0.3(5) M . If this value were valid,
-8
+
to 1.6 × 10 M H .
5
The intercept in Figure 3 is equivalent to Lee and Lister’s
(
(
18) Tesfai, T. M.; Margerum, D. W. Unpublished work.
19) Martell, A. E.; Smith, R. M. Critical Stability Constants; Plenum
Press: New York, 1982; Vol. 5, (a) p 403, (b) p 411.
-
13.60
2
definition of k
µ ) 1.0 M NaClO
a
, and the slope multiplied by K
w
(10
M ,
2
4
+
4
)
corresponds to k
b
in eq 3. Our [H ]
(
20) (a) Bousher, A.; Brimblecombe, P.; Midgeley, D. J. Chem. Soc., Dalton
5
range is larger by many orders of magnitude than they used,
Trans. 1987, 943-946. (b) Midgley, D. J. Chem. Soc., Dalton Trans.
1
991, 1585-1587.
(
(
21) Liu, Q. Ph.D. Thesis, Purdue University, 2000; p 99.
(23) Van Eldik, R.; Harris, G. Inorg. Chem. 1980, 19, 880-886.
(24) Molina, M.; Melios, C.; Tognolli, J. O.; Luchiari, L. C.; Jafelicci, M.,
Jr. J. Electroanal. Chem. Interfacial Electrochem. 1979, 105, 237-
246.
22) (a) Pa a´ l, T.; Barcza, L. Acta Chim. Acad. Sci. Hung. 1975, 78, 33-
4
0. (b) Baes, C. F. Jr.; Mesmer, R. E. The Hydrolysis of Cations;
John Wiley & Sons: New York, 1976; pp 104-111.
Inorganic Chemistry, Vol. 42, No. 1, 2003 81

Huff Hartz et al.
1
-1
and we find k
fraction of protonated bromite can be neglected because the
pK of bromous acid is 3.59(5).
BrO
dence on BrO
a
) 4.50 M- s-1 and k
b
) 0.076 s . The
The expression for krxn on the basis of this mechanism, in
the absence of buffers, is shown in eq 30, where K
1
) k
1
/
+
a
k-1 and K ) k /k . The zero-order and first-order [H ]
2
2
-2
-
2-
2
/SO
3
Reaction Mechanism. The kinetic depen-
dependences of krxn are resolved from the data in Figure 3.
-, SO
2-
, and HA requires reaction intermedi-
+
2
3
The first-order dependence in [H ] is a combination of two
H
H O
12
ates that lead to the reduction of Br(III) and the oxidation
of S(IV). Electron transfer processes via Br(II) and S(V)
would be extremely unfavorable and are ruled out. Hence, a
intermediate must form that can react with
general acids to cause Br(III) reduction and S(IV) oxidation.
However, a preequilibrium between HA and bromite or HA
, which complexes SO
acceptable because this mechanism would lead to specific-
acid, not general-acid, catalysis. Therefore, bromite and
H , are the reactive species in
/S(IV) reactions. The mechanism of eqs 22-29 is
proposed. Bromite acts as an electrophile and reacts rapidly
with sulfite, a nucleophile, to form (O
pathways where (k K + k
2
K K /K ) ) 3.03(4) × 10
3
1
4
1
2
w
-
2
-1
M
s . The intercept corresponds to a water-catalyzed
H O
-1
-1
pathway, where k3
2
K1 ) 4.50(3) M s . The rate
3
-
(
BrO
2
‚SO
3
)
expression given by eq 30 includes a second-order depen-
+
H
4
dence in [H ] (due to k ) that is not observed in the Figure
H
3
×
4 1 2 w
. Therefore, the value for k K K /K must be less than 5
2
-
and sulfite (except B(OH)
3
3
) is not
17
-3 -1
10
M
s . This pathway is not significant relative to
+
the other paths when p[H ] > 6.0.
-
sulfite, not HBrO
2
or SO
3
H2O
k4 K K
H2O
H
1
2
+
-
BrO
2
krxn ) k
K + k K +
[H ] +
3
1
(
3
1
)
[
Kw
H
-
3-
k K K
K (SO H )
a 3
2
-
BrSO
3
) . This species
4
1
2
+
2
[
H ]
(30)
+
2
can protonate to form (HO
data indicate that the concentrations of the (O
2
BrSO
3
) . Kinetic and spectral
Kw
]
-
K (SO H ) + [H ]
a
3
3
-
2
BrSO
3
)
and
2-
(
HO
2
BrSO
3
)
are not appreciable. Reactions in carbonate,
borate, and phosphate buffers (where p[H ] > 7.2) indicate
[HA] and k-2[OH ] . k
the presence of buffers at p[H ] > 7.2, (O
The presence of two bromite/sulfite intermediates, (O
2
Br-
+
3
-
2-
SO
3
)
2 3
and (HO BrSO ) , and two general-acid-catalyzed
HA
-
HA
that k-1 . k
3
4
[HA]. Hence, in
+
3-
paths are kinetic requirements. The composite rate constant
2
BrSO
3
)
and
+
12
-2
2-
for the first-order dependence in [H ] is 3.03 × 10
M
(HO
2
BrSO
3
)
can be treated as preequilibrium species.
-
1
H O
H
12
s . If the k
4
2
path did not exist, then k
3
K
1
) 3.03 × 10
k1
-2
s-1. The H
+
3-
H
BrO2- + SO3 { } (O BrSO )
2-
3-
(22)
M
3
O /(O
2
3
BrSO )
rate constant (k
3
) can be
2
3
k-1
no greater than the diffusion control limit in aqueous solution,
1
0
-1 -1
1
0
M
s . With this constraint, the lower limit for the
k2
3
-
2-
-
3-
-1
(
O BrSO ) + H O { } (HO BrSO ) + OH (23)
2 3 1
formation constant of (O BrSO ) , K , is 303 M . With a
2
3
2
2
3
k-2
formation constant this large, we would expect to observe a
significant fraction of bromite complexed to sulfite. However,
the kinetic spectra show no changes in the characteristic
k3HA
3
-
-
-
-
(
O BrSO ) + HA
8 OBrSO3 + A + OH
(24)
2
3
3-
k4HA
bromite peak shape, indicating that the (O
2
BrSO
3
)
con-
2
-
-
-
-
3
-1
(
HO BrSO ) + HA
8 OBrSO3 + A + H O (25)
centration is negligible and K
the second reaction pathway, k
e 5 × 10 M . Therefore,
2
3
2
1
H O
2
, must occur. This path
4
rapid
OBrSO3- + H O
8 OBr + H + HOSO
-
+
-
(26)
+
dominates the first-order dependence in [H ], so that
2
3
H O
12
-2 -1
H
k
4
2
1
K K
2
/K
w
) 3.03 × 10
M
s , where k
3
K
1
e 5 ×
rapid
07 M
-2 -1
-
-
1
s
OBr + H O { } HOBr + OH K /K (HOBr) (27)
2
w
a
Two general-acid-assisted pathways can be extracted from
rapid
2
-
-
-
HA
HA
the experimental data, k
demonstrated for HCO
3
K
1
and k
4
1 2 w
K K /K , and this is
HOBr + SO3
8 BrSO3 + OH
(28)
(29)
-
3
. Equation 31 shows the rate
+
-
BrSO3- + OH- rapid8 Br + HOSO3-
expression in carbonate buffer. At a fixed [H ], the equation
becomes linear (eq 32) and the rate constant depends on
-
-
We propose that general acids react with both adducts;
[HCO
3
]. The slope of the krxn vs [HCO
3
] plot increases
2-
3-
-
+
HA removes O from (O
2
BrSO
3
)
and releases OH ,
BrSO
with increasing [H ], where the slope is the sum of the two
-
2-
-
whereas HA abstracts OH from (HO
2
3
)
and releases
HCO
3
-dependent paths (eq 33).
2
H O. The result in both cases would be an intermediate,
-
-
-
H2O
OBrSO
3
, that rapidly hydrolyzes to form OBr , HOSO
3
,
k4 K K
H2O
1
2
+
HCO3-
-
+
krxn ) k3 K +
[H ] + k
K [HCO ] +
1 3
and H . In this process, Br(III) is reduced to Br(I) and S(IV)
is oxidized to S(VI) during either the formation or hydrolysis
1
3
Kw
HCO3-
-
-
k4
K K
1 2
of OBrSO
3
. The OBr released is in rapid equilibrium with
-
+
[
HCO ][H ] (31)
3
HOBr. The resulting HOBr reacts very rapidly with an
Kw
2-
-
additional SO
3
to form BrSO
3
by a Br-atom transfer
-
2-
9
-1
-1 1e
krxn ) intercept + slope[HCO ]
(32)
(33)
mechanism (k(HOBr/SO
3
) ) 10 M s ). Finally,
3
-
-
-
BrSO
3
rapidly hydrolyzes to form Br and HOSO
3
which,
HCO3-
k4
K K
1 2
2-
+
HCO3-
+
under the experimental conditions, gives SO
4
and H . The
slope ) k3
K +
[H ]
1
-
19b
Kw
pK
a
of HOSO
3
is 1.10.
82 Inorganic Chemistry, Vol. 42, No. 1, 2003

S(IV) Reductions of Bromite and Chlorite Ions
Table 1. Summary of Equilibrium and Rate Constantsa
constant
value (std dev)
pKa(H3O+)
pKa(HClO2)
pKa(HBrO2)
pKa(SO2)
-1.7425c
1.72(2)
3.59(5)
1.90(5)
6.30(5)
6.26(1)
7.47(1)
8.59(3)
8.85(1)
9.57(3)
11.08(1)
13.60(1)
15.34(1)
2
6
b
2
3
-
23
1h
pKa(SO3H )
-
pKa(H2PO4 )
1
1
2
d
pKa(HOCl)
pKa(HOBr)
pKa(B(OH)3)
g
2a
-
19a
18
pKa(HCO3 )
2
-
pKa(HPO4
pKw
)
2
4
c
pKa(H2O)
BS d
-1 b
-1 e
K
K
K
7.6(5) M
0.3(5) M
BB d
S f
28
4.9(1)
-
2-
BrO2 /SO3 Reaction Rate Constants
-
2-
3.0(5) × 10 M s-1 b
7
-1
k1(BrO2 /SO3
)
+
H O
-1 -1 b
Figure 4. [H ] dependence of bromite/sulfite reaction rate constant, kobs/
k3
k4
k3
k3
k4
k4
k4
2
K1
4.50(3) M
s
-
-8
H O
2
12
-2 -1 b
[SO3]T (o), at 0.106 mM BrO2 , 0.91-1.21 mM [SO3]T, 6.3 × 10 -1.3
K1K2/Kw
3.03(4) × 10
M
s
-6 +
HPO42-
3
-2 -1 b
× 10 M H , and 30 mM [PO4]T. The solid line shows a curve fit to
K1
K1
1.5(2) × 10 M
s
+ + - + + -
HCO3
HCO3
-
3
-2 -1 b
kobs/[SO3]T ) (d(e[H ] + f[H ][H2PO4 ]/d)/(1 + e[H ] + f[H ][H2PO4 ]/
7
9(2) × 10 M
s
-
-
+
-1
-
13
-3 -1 b
d))Ka(SO3H )/(Ka(SO3H ) + [H ]), where d ) k1 ) 3.0(5) × 10 M
K1K2/Kw
K1K2/Kw
K1K2/Kw
1.8(3) × 10
M
s
s
s
-
-
1
H O
5
-1
H PO
2 4
s
, e ) k4
2
K /k
M
2
-1
-3
K
w
) 1.5(3) × 10 M , and f ) k4
K K /K
w
1
2
)
B(OH)3
H PO4
13
-3 -1 b
8.1(5) × 10
M
M
13
s-1. Error bars reflect the propagation of error in
-
13
-3 -1 b
1.3(3) × 10
2
1.3(3) × 10
measurement.
-
ClO2 /S(IV) Reaction Rate Constants
-
6
-1 s-1 b
k5(ClO2 /SO2)
6.26(4) × 10 M
-
23
-
-
-1 -1 b
(pK
the rate of SO
comparison to the BrO
a 3
(SO H ) ) 6.30). Under our experimental conditions,
k6(ClO2 /SO3H )
5.5(3) M
s
-
-
3
H
reaction with BrO
2
is negligible in
a
b
c
2
5.0 °C, µ ) 1.0 M (NaClO4). This work. Calculated on the basis of
-
2-
d
2
/SO
3
reaction rate. However, this
pKa(H2O) ) -log(Kw/55.5). Equilibrium constant for the formation of boric
acid/sulfite complex (K ) and boric acid/bromite complex (K ). Calcu-
lated on the basis of pKa (HBrO2) and ref 20. K represents the ratio between
hydrogen sulfite isomers, [HOSO2 ]/[HSO3 ].
BS
BB
e
does not completely account for the rate behavior. Instead,
f
S
+
these data show that as [H ] increases the preequilibrium
-
-
conditions are no longer valid and a steady-state approxima-
3
-
tion for the (O
2
BrSO
3
)
species is needed. The equation
, eq 34)
-
A plot of the slopes of the HCO
3
dependence data vs
for the total steady-state concentrations ([SS]
T
+
-
HCO3-
[
)
M
H ] resolves the two HCO
3
-assisted paths, where k
3
K
1
3
-2 -1
HCO3-
13
9(2) × 10 M
s
and k
4
K
1
K
2
/K
w
) 1.8(3) × 10
[
SS] )
T
-
3
-1
s . These data support the presence of two general-
-
2-
-
acid-assisted paths (eqs 24 and 25).
k [BrO ][SO ](K + [OH ])
1
2
3
2
A similar treatment was applied to B(OH)
3
dependence
H2PO4-
-
H2O
-
H2O
-
k [OH ] + (k
[OH ] + k
K + k
K [H PO ])
B(OH)
) 8.1(5) × 10 M-3 s . The
13
-1
-1
3
4
2
4
2
2
4
data, where k
4
3
1 2 w
K K /K
+
y-intercept for the plot of the slopes vs [H ] was zero within
(34)
B(OH)
the error of the fit, which means that the k
negligible under these conditions. As [H ] increases, the (O
3
3
1
K term is
+
3-
2
-
is derived by assuming that [SS]
+ [(HO BrSO ) ], and the ratio of the bromite/sulfite
2 3
intermediates is related to K
T
is equal to [(O
2
3
BrSO ) ]
3
-
2-
2-
BrSO
3
) /(HO
2
BrSO
3
2
)
equilibrium constant (K
2
) favors
path does not
-
B(OH)
3
higher [(HO
2
BrSO
3
) ]. Thus, the k
3
2
(eq 35). (A detailed derivation
contribute significantly to the rate constant.
is given in the Supporting Information).
The general-acid-catalyzed rate constants resolved from
the data are shown in Table 1. As noted above, as the [H ]
2
-
-
+
[(HO BrSO ) ][OH ]
2 3
K )
(35)
HA
2
3-
increases, the k
3
term becomes negligible. For this reason,
[
(O BrSO ) ]
2 3
H PO -
we can assume that the k
3
2
4
K
1
term is negligible, and
H PO -
the k
4
2
4
K
1
K
2
/K
w
term is obtained from the curve fit of
The rate expression for bromite/sulfite reaction in the
absence of buffers for p[H ] < 7.2 is shown in eq 36, where
-
13
-3 -1
the krxn vs H
2
PO
4
data (1.3(3) × 10 M s ). By a similar
+
+
HPO 2-
4
argument, at lower [H ], the k
negligible, so that k
obtained from the curve fit of the krxn vs HPO
4
K
1
K
2
/K
w
term is
H O
-
H O
2
k
3
2
[OH ] is negligible relative to k
4
K
2
. The curve fit of
HPO 2-
3
-2 -1
4
K
(1.5(2) × 10 M s ) can be
-
3
1
the data (Figure 4) shows that k
1
, which represents BrO
2
/
2-
plot.
2-
7
-1 -1
4
SO
direct reaction, is 3.0(5) × 10 M s . By compari-
3
-/SO
2-
Reaction Inter-
Steady-State Behavior of BrO
2
3
son, sulfite reacts rapidly with hypohalous acids to form
+
+
-
2-
9
mediates at p[H ] 7.2-5.9. Above p[H ] 7.2, the BrO
2
+
/
(HOXSO ) intermediates with rate constants of 5.0 × 10
3
2
-
-1 -1
1c,1e
SO
3
reaction rate has a first-order dependence in [H ].
However below p[H ] 7.2, the krxn values level off as [H ]
M
s
for HOCl and HOBr.
Our data show that bromite
+
+
is also very reactive but its rate constant is 2 orders of
magnitude less. This indicates that BrO
electrophile than HOCl and HOBr.
-
increases (Figure 4). This occurs in part because the fraction
2
is a weaker
+
of protonated sulfite ion increases as [H ] increases
Inorganic Chemistry, Vol. 42, No. 1, 2003 83

Huff Hartz et al.
kobs
)
[
SO3]T
H2O
+
H2PO4-
4
-
+
k1(k4 K [H ] + k
K [H PO ][H ])
2 2 4
2
×
H2O
+
H2PO4-
4
-
+
k K + k
K [H ] + k
K [H PO ][H ]
2 2 4
-
1
w
4
2
-
K (SO H )
a
3
(36)
-
+
K (SO H ) + [H ]
a
3
+
-/SO
2-
The [H ] dependence of the BrO
2
3
reaction rate
+
below p[H ] 7.2 also provides further evidence that any
-
SO
3
H and HBrO
expressions for BrO
2
reaction paths are negligible. The rate
-
-
2-
2
/SO
3
H , HBrO
2
/SO
3
2
, and HBrO /
-
3
SO H
reactions do not fit the data using known acid
dissociation constants.
Brønsted-Pedersen Relationship for the BrO
Reaction. General-acid-catalyzed reaction mechanisms often
follow the Brønsted-Pedersen relationship,25a eq 37. This
describes the correlation between the strength of the assisting
-
Figure 5. Brønsted-Pedersen plot for general-acid catalysis of BrO2 /
-/SO
2-
2-
2-
HA
SO3
reaction: (HO2BrSO3)
+ HA path (k4 K1K2/Kw) (0) and
2
3
-
HA
HA
(
O2BrSO3)3 + HA path (k3 K1) (o). Data are fit to eq 37 where k
represents the composite rate constant for general-acid catalysis.
HA
acid in relation to the rate constant, where k represents
the composite general-acid-catalyzed rate constant, p is the
number of equivalent acidic sites in HA, q is the number of
-
equivalent basic sites in A , G
a
is a constant, and R describes
the sensitivity of the reaction to the acid strength of HA.
Plots of the experimental data are shown in Figure 5. Both
general-acid-catalyzed pathways (eqs 24 and 25) follow the
HA
Brønsted-Pedersen relation, where R(k
3
1
K ) ) 0.88(7) and
HA
R(k
4
1 2 w
K K /K ) ) 0.3(1). This suggests there is a higher
3
-
degree of proton transfer from HA to (O
2
3
BrSO )
(Figure
2-
6
a) than to (HO
2
3
BrSO )
(Figure 6b).
HA
k
qKa
log
) log G + R log
a
(37)
p
p
Figure 6. Proposed halite/S(IV) transition state species: (a) BrO2 /SO32-/
-
HA
-
2-
+
HA
The behavior of boric acid is somewhat surprising because
HA (k3 K1 path, eq 24); (b) BrO2 /SO3 /HA/H (k4 K1K2/Kw path, eq
-
-
-
boric acid is a Lewis acid. However, Bell25b suggests that
boric acid may act also as a Brønsted acid because it can
25); (c) ClO2 /SO3H (k6 path, eq 42); (d) ClO2 /SO2 (k5 path, eq 41).
To obtain the rate parameters for this reaction, the data
were fit in smaller p[H ] ranges. The expression eq 38 is
2 3
coordinate a water to form a H OB(OH) complex. This
+
complex could donate a proton to the transition state.
9
-
based on Frerichs’ work but includes the constant for SO
2
/
Lack of Buffer Dependence for the ClO
2
/S(IV) Reac-
-
23
SO
the data occurs when the y-intercept (y
the y-intercept represents direct chlorite/sulfite reaction, this
3
H equilibrium (pK
a
(SO
2
) ) 1.90(5)). The best fit to
tion. The chlorite/S(IV) reaction shows no buffer dependence
+
0
) value is zero. Since
on the rate constant for 30-75 mM [PO
4
]
T
at p[H ] 10.42-
+
(
3), 25-100 mM [CO
at p[H ] 8.75(2), 125-200 mM [PO
and 40-80 mM [OAc]
3
]
T
at p[H ] 9.52(1), 16-80 mM [Bor]
at p[H ] 6.38(1),
at p[H ] 4.33(5). As the buffer
T
-
2-
+
+
indicates that the direct reaction between ClO
is negligible.
2 3
and SO
4
]
T
+
T
concentration increases, the rate of reaction does not change.
+
kobs
+ 2
+
-
However, the rate increases with [H ]. Thus, the acid
(
[H ] + K (SO )[H ] + K (SO H )K (SO )) )
a 2 a 3 a 2
-
[SO ]
3 T
dependence for the ClO
2
/S(IV) reaction is due to specific-
+
+ 2
acid assistance, and not general-acid assistance.
y + a[H ] + b[H ] (38)
0
+
-/S(IV) Reaction. The
[
H ] Dependence of the ClO
2
second-order rate constant for the reaction of chlorite with
excess sulfite was measured as a function of [H ] (p[H ]
+
A plot of krxn vs p[H ] is shown in Figure 7. The data fit
the rate expression generated by eq 38 where a ) 5.7(3) ×
+
+
+
3
.8-11.1). As [H ] increases, an increase in the reaction
-
2
-1
6
-1 -1
1
0
s
and b ) 6.26(4) × 10 M s . No higher order
-
rate is observed beginning near the pK
which indicates that SO
a
value of SO
3
H ,
than
+ 3
paths were found (such as [H ] ), which shows that, for
-
-
3
H is more reactive with ClO
2
+
26
2 a
p[H ] > 3.8, the contribution of HClO (pK ) 1.72) to
the reaction rate constant can be neglected.
2-
SO
3
.
(
25) Bell, R. P. The Proton in Chemistry, 2nd ed.; Cornell University:
Ithaca, NY, 1973; (a) pp 195-198, (b) p 223, (c) p 200.
(26) Fabian, I.; Gordon, G. Inorg. Chem. 1992, 31, 2144-2150.
84 Inorganic Chemistry, Vol. 42, No. 1, 2003

S(IV) Reductions of Bromite and Chlorite Ions
-
ClO
2
/S(IV) Reaction Mechanism. The mechanism for
-
the ClO
2
/S(IV) reaction must include several features that
-
are different from the BrO
required by the kinetics. No ClO
2
/S(IV) mechanism but are
-
2-
2
/SO
3
direct reaction is
observed because the y-intercept for the plot of kobs/[SO ]
3 T
+
-
2-
vs [H ] is zero. Although the ClO
2
/SO
3
reaction potential
-
2-
is slightly more favorable than the BrO
potential, the ClO
the BrO
s . Instead, ClO
2
/SO
3
reaction
-
2-
2
/SO
reaction rate constant is 3.0 × 10 M
reacts preferentially with SO H , while
H reaction is negligible. Furthermore, if the
3
reaction rate is negligible whereas
-
2-
7
-1
2
/SO
3
-1
-
-
2
3
-
-
the BrO
2
/SO
3
-
ClO
2
/S(IV) reaction mechanism were the same as the
-
BrO
2
/S(IV) reaction mechanism, we would expect strong
general-acid catalysis of the reaction rate by reaction with
-
2-
-
-
ClO
2
/SO
3
and ClO
2
3
/SO H
intermediates. However,
buffers have no effect on the reaction rate. These distinctions
+
Figure 7. p[H ] dependence of chlorite/S(IV) rate constant (krxn, o). Solid
-
require a mechanism for the ClO
fundamentally different from the BrO
2
/S(IV) reaction that is
-2
+
line (-) fits the following equation: krxn ) (5.7 × 10 [H ] + 6.26 ×
-
6
+ 2
+ 2
+
-
2
/S(IV) reaction.
a 2 a 3 a 2
10 [H ] )/([H ] + K (SO )[H ] + K (SO H )K (SO )). The insert shows
-
the increase in rate constant near pKa(SO3H ). Conditions: µ ) 1.0 M;
5.0(1) °C.
The mechanism of eqs 39-45 is developed on the basis
2
-
of the experimental data, where ClO
2
acts as a nucleophile
-
and SO
2
and SO
3
H are electrophiles. Sulfite is in a rapid
-
the rate parameter must be corrected for the HSO
3
isomer.
H reaction mecha-
nism is shown in eq 46, and the resolved rate constants,
preequilibrium with hydrogen sulfite, which is in rapid
preequilibrium with sulfur dioxide. Chlorite and sulfur
-
-
2 3
The rate expression for the ClO /SO
-
dioxide react to form an intermediate (OClOSO
2
) , which
6
obtained from the fit of eq 38, are b ) k
5
) 6.26(4) × 10
-
rapidly dissociates into OCl and SO
the SO
3
. The hydrolysis of
3
product is rapid in aqueous solution. Similarly,
chlorite and hydrogen sulfite react to form an (OClOSO
intermediate, which rapidly dissociates into OCl and
-1 -1
S
S
-1 -1
M
s
and a(1 + K )/(K K
a
(SO
2
)) ) k ) 5.5(3) M s .
6
27
2-
S
3
H)
K K (SO )
+
2
a
2
+
-
k [H ] + k
[H ]
5
6
(
)
S
kobs
1 + K
-
-
HOSO
3
-
. OCl is in rapid equilibrium with HOCl (K
a
(HOCl)
)
+
2
+
-
[
SO ]
7.47
1d
-
3 T
([H ] + K (SO )[H ] + K (SO H )K (SO ))
)
10
M), and HOCl is rapidly reduced to Cl by a
a
2
a
3
a
2
2-
8
-1 -1 1c
(46)
second SO
3
(k ) 7.6 × 10 M s ) .
The rate constant for ClO
of magnitude larger than ClO
The SO
much better electrophile than SO
2
-/SO
2
reaction (k
5
) is 6 orders
rapid
2
-
-
-
-
SO3 + H O { } SO H + OH K /K (SO H ) (39)
2
3
w
a
3
-/SO
-
2
3
H rate constant (k
6
).
is a
H , and the rate constant
+
2
path predominates when p[H ] < 7.5. SO
2
rapid
-
-
SO H { } SO + OH K /K (SO )
(40)
(41)
-
3
2
w
a
2
3
for reaction with chlorite, a nucleophile, increases with
increasing electrophilicity of S(IV). A similar behavior has
k5
ClO2- + SO2
9
8 (OClOSO2)
-
-
been observed for bromate/S(IV) reaction, where the BrO
3
/
k6
SO
SO
2
rate constant is 3100 times greater than the BrO
3
-/
ClO2- + SO H-
98 (OClOSO H)
2-
(42)
(43)
(44)
(45)
-
29
3
3
H rate constant.
An alternate mechanism was also considered (see Sup-
porting Information). This mechanism includes chlorite and
3
rapid
(
OClOSO )-
8 OCl + SO3
-
2
hydrogen sulfite in a rapid preequilibrium with a (OClO-
rapid
2-
2-
-
-
SO
3
H) intermediate. This intermediate would protonate,
which allows for the observed squared [H ] dependence.
Both intermediates would dissociate in the rate-determining
(
OClOSO H)
8 OCl + HOSO3
3
+
rapid
2
HOCl + SO3 -
-
2-
+
8 Cl + SO4 + H
-
-
+
3
steps to form OCl and HOSO . The [H ] dependence for
-,
, where
this mechanism is identical to the mechanism given in eqs
39-45 under our conditions. Thus, the mechanisms are
kinetically indistinguishable. However, we prefer the mech-
Hydrogen sulfite exists in two isomeric forms: HSO
3
-
where the proton is bonded to the sulfur, and HOSO
2
the proton is bonded to an oxygen. The latter isomer
S
-
-
anism given in eqs 39-45 because it accounts for the
predominates, and the ratio K ) [HOSO
2
]/[HSO
3
] is 4.9
isomer
-
28
-
expected electrophilicity difference between SO
2
and SO
3
H .
at 25.0 °C, µ ) 1.0 M. We expect only the HOSO
2
-
SO
2 3 2
is more electrophilic than SO H , and SO reacts faster
to be reactive because a chlorite to sulfur bond is needed
for the reaction mechanism. Since k refers to the average
rate constant for the two isomeric forms of hydrogen sulfite,
-
-
with ClO
2
3
than SO H .
6
Oxygen-Atom Transfer vs Oxyhalogen Transfer. The
-
proposed BrO
2
/S(IV) reaction mechanism includes inter-
(
(
27) Meijer, E. J.; Sprik, M. J. Phys. Chem. A. 1998, 102, 2893-2898.
28) Horner, D. A.; Connick, R. E. Inorg. Chem. 1986, 25, 2414-2417.
(29) Szirovicza, L.; Boga, E. Int. J. Chem. Kinet. 1998, 30, 869-874.
Inorganic Chemistry, Vol. 42, No. 1, 2003 85

Huff Hartz et al.
mediates with Br-S connectivity rather than O-S con-
with O-Cl connectivity, which do not have to react with
-
nectivity. BrO
2
acts as an electrophile in this reaction, and
general acids.
-
2-
the Br atom expands its octet to accept an electron pair from
2 3
The thermodynamic driving forces for the BrO /SO
2
-
-
2-
-
the nucleophilic sulfur atom of SO
the removal of an oxide ion from (O
3
. General acids assist
and ClO /SO reactions are comparable. Yet, the ClO /
2
3
2
3-
2
BrSO
3
)
(Figure 6a)
S(IV) reaction is more than 5 orders of magnitude slower
2-
-
-
2 3
or a hydroxide ion from (HO BrSO )
(Figure 6b). If general
2 2
than the BrO /S(IV) reaction in basic solution. The ClO /
acids did not remove the oxide or hydroxide from the
bromine side of the intermediate, incorrect products would
S(IV) reaction must overcome significant kinetic barriers for
reaction to occur. The proposed reaction mechanism proceeds
-
2-
-
form. The data analysis shows that the rate of the BrO
S(IV) reaction decreases as the fraction of protonated [SO
2
/
through (OClOSO H) and (OClOSO2) intermediates,
3
]
3 T
where chlorite donates an electron pair to sulfur. Sulfur can
2-
increases. Protonation of SO
3
reduces the nucleophilicity
expand its number of valence electrons more readily than
3
-
-
of the S atom and reduces the fraction of (O
formed.
2
BrSO
3
)
oxygen. The rate-determining steps for the ClO /S(IV)
2
reaction include oxidation of the sulfur atom, transfer of an
oxygen atom, and reduction of the chlorine atom (eq 47).
The alternative connectivities of (BrO
2
‚SO
3
)3 contain
and (OBrOSO ) . The (O BrO-
-
3-
3-
O-S bonds: (O
2
BrOSO
2
)
3 2
3-
III
IV
2-
I
VI
2-
SO
products so it can be ruled out. The (OBrOSO
conceivable as an oxygen-atom transfer from bromine to
2
)
species does not provide a path that leads to the
(OCl OS O H) f (OCl OS O H)
(47)
3
3
3
-
3
)
species is
+
If only the dependence in [H ] were considered, it may
-
2-
sulfur, which would give OBr and SO
4
. However, there
-
be reasonable to assume that the BrO
2
/S(IV) reaction and
is no obvious way that general acids would catalyze the
-
the ClO
2
/S(IV) reaction occur through the same oxygen-
3-
3
decomposition of (OBrOSO ) . If HA reacts with the
atom transfer mechanism. However, rate expressions that
oxybromine side of the intermediate, then the HA would
-
-
include SO
3
H or SO
2
reaction with BrO
2
do not fit the
-
remove an OBr so that the oxygen atom could be transferred
data (using known values the acid dissociation constants).
to the sulfur. This is not a characteristic general acid function.
Furthermore, the observed general-acid catalysis of the
2
-
-
Instead, general acids typically react with O or OH to
-
BrO
2
/S(IV) reaction and the lack of a general-acid catalysis
-
form OH or H
2
O. HA could react with the sulfur side of
-
-
2 2
for the ClO /S(IV) reaction shows that the BrO /S(IV)
the halite/S(IV) intermediate. This would result in an
intermediates must have a different connectivity than the
-
(OBrOSO
2
) intermediate that would have to hydrolyze at
-
ClO
2
/S(IV) intermediates. General-acid catalysis is required
-
-
the sulfur to produce OBr and HOSO
3
. It is illogical that
3-
-
2-
for (O
2
BrSO
3
)
to produce OBr and SO
4
, while a
-
the function of the general acid is to remove the OH in the
first step, only to have it return in the subsequent steps. In
fact, the general acid reaction is completely unnecessary to
2-
general-acid catalysis is not needed for (OClOSO
produce OCl and OSO
3
H) to
-
-
3
H . For these reasons, we cannot
-
assign the BrO
2
/S(IV) reaction the same mechanism as the
-
-
2-
produce OBr and HOSO
3
from (OBrOSO
3
H) . Of the
-/S(IV) reaction.
ClO
2
3-
possible (BrO
2 3
‚SO )
connectivities, only the adduct with
In comparison of chlorite/sulfite and bromite/sulfite reac-
tion mechanisms, it is known that the reactions of hypochlo-
rous acid and hypobromous acid with sulfite occur through
Br-S bonds requires general-acid catalysis. For these
reasons, we use the general-acid catalysis as the rationale
for the Br-S connectivity of the bromite/sulfite intermedi-
ates.
1c,1e,30
halogen-sulfur intermediates.
Others have studied the
29,31
kinetics and mechanisms of bromate/S(IV) reaction
and
The ClO
2
-/S(IV) reaction mechanism includes intermedi-
32-34
chlorate/S(IV) reaction.
halates increase as [SO
The reaction rates of these
becomes more protonated. Despite
favorable thermodynamics (eqs 48 and 49),
chlorate show virtually no reaction with SO
the bromate/S(IV) reaction rate is not affected by buffers,
which is similar to the chlorite/S(IV) kinetics. In light of
these observations, we propose that, for S(IV)/oxyhalogenate
reactions, reductions of Cl(III), Cl(V), and Br(V) proceed
by oxygen/sulfur intermediates, while Br(III), Br(I), and Cl-
ates with O-S connectivity rather than Cl-S connectivity.
This reaction shows a specific-acid, not a general-acid,
catalysis. The Cl atom does not expand its octet as easily as
the Br atom and cannot accept the electrons from sulfur as
]
3 T
10,11
bromate and
2-
3
. Furthermore,
31
easily as the Br atom. Furthermore, the Cl-O bond is less
-
polar than the Br-O bond. Therefore, the Cl atom in ClO
2
-
is not as electropositive as the Br atom in BrO
2
and is not
reacts through
-
a good electrophile. To overcome this, ClO
2
the O atom, which is more nucleophilic. As a consequence,
(I) prefer halogen/sulfur intermediates. The arrangement of
-
ClO
2
acts as a nucleophile and reacts preferentially with
the oxyhalogen/S(IV) intermediate and whether the oxy-
halogen anion acts as a nucleophile or electrophile relative
to S(IV) determines the reaction rate. We propose that the
-
2-
SO
SO
2
and SO
/SO
3
H , which are better electrophiles than SO
3
.
-
-
2
3
H and ClO
2
form adducts with expanded valence
around the S atom via oxygen/sulfur connectivity (Figure
-
2-
6
c,d). These adducts, (OClOSO
2
-
) and (OClOSO
3
H) , form
/S(IV) mechanism were the
2
/S(IV) mechanism, then we would expect
(30) Yiin, B. S.; Margerum, D. W. Inorg. Chem. 1988, 27, 1670-1672.
(31) Williamson, F. S.; King, E. L. J. Am. Chem. Soc. 1957, 79, 5397-
-
-
3 2
OCl and HOSO . If the ClO
same as the BrO
5
400.
-
(
32) Gleason, E. H.; Mino, G.; Thomas, W. M. J. Am. Chem. Soc. 1957,
61, 447-450.
intermediates with Cl-S connectivity that would react with
(
33) Dobrynin, N. A.; Dymarchuk, N;. P.; Mishchenko, K. P. Zh. Obsch.
Khim. 1969, 39, 2157-2164.
general acids. However, the lack of general-acid catalysis
-
shows that the ClO
2
/S(IV) mechanism prefers intermediates
(34) Men e´ dez, S. M. ReV. Fac. Cienc. 1968, 9, 119-213.
86 Inorganic Chemistry, Vol. 42, No. 1, 2003

S(IV) Reductions of Bromite and Chlorite Ions
presence or absence of general-acid catalysis and the
and an Emerson Kampen Foundation Fellowship to Purdue
University (K.E.H.H.).
2-
-
oxyhalogen reactivity toward SO
3 3
/SO H is a consequence
of the connectivity of the oxyhalogen/S(IV) reaction inter-
mediate.
Supporting Information Available: Tables and figures of
kinetic data, the derivation of the rate expression derivation for
BrO3- + SO32- f BrO - + SO4
2-
EB ) 1.81 V (48)
o
2
-/SO
2-
-/S(IV)
BrO
2
3
reaction, an alternative mechanism for ClO
2
reaction, and a description of polyborate distribution calculation.
This material is available free of charge via the Internet at
http://pubs.acs.org.
ClO3- + SO32- f ClO - + SO42- EB ) 1.68 V (49)
o
2
Acknowledgment. This work was supported by National
Science Foundation Grants CHE-9818214 and CHE-0139876
IC020475M
Inorganic Chemistry, Vol. 42, No. 1, 2003 87