General-acid-catalyzed reactions of hypochlorous acid and acetyl hypochlorite with chlorite ion

Article abstract of DOI:10.1021/ic991486r

The rate of oxidation of ClO₂^- by HOCl is first order in each reactant and is general-acid catalyzed. In the initial steps of the proposed mechanism, a steady-state intermediate, HOClOClO^-, forms (k₁ = 1.6 M^-1 s^-1) and undergoes general-acid (HA)-catalyzed reactions (k₂(HA)) to generate a metastable intermediate, ClOClO. Values of k₂(HA)/k_-1 are 1.6 x 10⁴ M^-1 (H₃O⁺), 20 M^-1 (HOAc), and 8.5 M^-1 (H₂PO₄^-). Subsequent competitive reactions of ClOClO with ClO₂^- (k₃) to give 2ClO₂ and with OH^- (k₄(OH)) and other bases (k₅(B)) to give ClO₃^- are very rapid. The relative yields of these products give k₄(OH)/k₃ = 1.3 x 10⁵, k₅(HPO)₄/k₃ = 0.20, and k₅(OAc)/k₃ = 0.06. At low pH and low buffer concentrations, the apparent yield of ClO₂, based on 2ClO₂ per initial HOCl, reaches 140%. This anomaly is attributed to the induced disproportionation of ClO₂^- by ClOClO to give ClO₃^- and additional HOCl. A highly reactive intermediate, ClOCl(O)OClO^-, is proposed that can undergo Cl - O bond cleavage to give 2ClO₂ + Cl^- via one path and ClO₃^- + 2HOCl via another path. The additional HOCl recycles in the presence of excess ClO₂^- to give more ClO₂. Ab initio calculations show feasible structures for the proposed reaction intermediates. Acetic acid has a second catalytic role through the formation of acetyl hypochlorite, which is much more reactive than HOCl in the transfer of Cl⁺ to ClO₂^- to form ClOClO.

Full text of DOI:10.1021/ic991486r

2
614
Inorg. Chem. 2000, 39, 2614-2620
General-Acid-Catalyzed Reactions of Hypochlorous Acid and Acetyl Hypochlorite with
Chlorite Ion
Zhongjiang Jia, Dale W. Margerum,* and Joseph S. Francisco
Department of Chemistry, Purdue University, West Lafayette, Indiana 47907
ReceiVed December 28, 1999
The rate of oxidation of ClO2- by HOCl is first order in each reactant and is general-acid catalyzed. In the initial
-
-1 -1
steps of the proposed mechanism, a steady-state intermediate, HOClOClO , forms (k1 ) 1.6 M s ) and undergoes
HA
HA
general-acid (HA)-catalyzed reactions (k2 ) to generate a metastable intermediate, ClOClO. Values of k2 /k-1
4
-1
+
-1
-1
-
are 1.6 × 10 M (H3O ), 20 M (HOAc), and 8.5 M (H2PO4 ). Subsequent competitive reactions of ClOClO
-
-
OH
B
-
with ClO2 (k3) to give 2ClO2 and with OH (k4 ) and other bases (k5 ) to give ClO3 are very rapid. The
OH
5
HPO
OAc
relative yields of these products give k4 /k3 ) 1.3 × 10 , k5
4
/k3 ) 0.20, and k5 /k3 ) 0.06. At low pH and
low buffer concentrations, the apparent yield of ClO2, based on 2ClO2 per initial HOCl, reaches 140%. This
anomaly is attributed to the induced disproportionation of ClO2 by ClOClO to give ClO3 and additional HOCl.
A highly reactive intermediate, ClOCl(O)OClO , is proposed that can undergo Cl-O bond cleavage to give
ClO2 + Cl via one path and ClO3 + 2HOCl via another path. The additional HOCl recycles in the presence
of excess ClO2 to give more ClO2. Ab initio calculations show feasible structures for the proposed reaction
intermediates. Acetic acid has a second catalytic role through the formation of acetyl hypochlorite, which is
much more reactive than HOCl in the transfer of Cl to ClO2 to form ClOClO.
-
-
-
-
-
2
-
+
-
Introduction
By using radioactive chlorine as a tracer, Taube and Dodgen12
proposed an unsymmetrical intermediate, ClClO2 (or ClOClO),
which is common to the reactions in eqs 1 and 2. Imagawa,
Chlorine dioxide is used as an alternative to chlorine in water
1
-3
4
treatment and pulp bleaching because it is a strong oxidant
and does not produce trihalomethanes and other organic
1
3
-
Fukagawa, and Yoshie reported that HOCl reacts with ClO2
-
2
3
to form ClO2 (d[ClO2]/dt ) k[ClO2 ] [HOCl], k ) 3.2 × 10
5
halogens as does chlorine. In water treatment plants, chlorine
-2
-1
7
M
s ). Subsequently, Peintler, Nagyp a´ l, and Epstein inves-
dioxide is generated by mixing aqueous solutions of chlorine
tigated this reaction in 0.5 M acetate buffer at pH 5-6 by
and sodium chlorite.1 Above pH 4, chlorine hydrolyzes to
-3
monitoring the formation of ClO2. They found only a first-order
-
form hypochlorous acid. The reaction of HOCl with ClO2
-
-
dependence in [ClO2 ]. In excess ClO2 , eq 1 was the
stoichiometric reaction, and they reported the rate expression
forms ClO2 (eq 1) and ClO3- (eq 2). The overall stoichiometry
+
-
-
+
-
to be d[ClO2]/dt ) 2k[HOCl][H ][ClO2 ], where k ) 1.1 ×
HOCl + 2ClO2 + H f 2ClO + Cl + H O (1)
2
2
6 -2 -1
1
0 M s . In excess HOCl, however, eq 3 was proposed to
HOCl + ClO2- f ClO - + Cl + H
-
+
(2)
3
HOCl + ClO2 f ClO3- + Cl + H O
-
2
(3)
2
2
of the reaction varies with the initial concentration ratio of the
reactants and the pH. The yield of ClO2 increases with increasing
explain the decrease in ClO2 yield with increasing [HOCl]0
-
-
2
-
because of ClO formation, where -d[ClO ]/dt ) k[HOCl] -
initial concentration of ClO2 and decreases with increasing
3
2
-
3
-2 -1
6-8
[ClO ], k ) 2.1 × 10 M s . (The abstract of that paper
pH. The USEPA has proposed a maximum contaminant level
2
-
3
-2 -1
of 1.0 mg/L for chlorite ion9,10 and has suggested chlorate ion
erroneously gave a value of 2.1 × 10
M
s
for this rate
8
-
11
constant.) Gordon and Tachiyashiki studied the HOCl/ClO
as a candidate for future regulation.
2
-
reaction in 0.1 M phosphate buffer at pH 6-10 and [OCl ]0/
-
(
1) Vigneswaran, S.; Vigneswaran, C. Water Treatment Processes: Simple
Options; CRC Press: Boca Raton, FL, 1995; pp 191-208.
[ClO2 ]0 ratios of 0.5-2.0. The minor amount of ClO2 that was
produced gradually decreased over a period of 50 h. The
(
2) Cheremisinoff, N. P.; Cheremisinoff, P. N. Water Treatment and Waste
RecoVery: AdVanced Technology and Applications; PTR Prentice
Hall: Englewood Cliffs, NJ, 1993; pp 64-68.
-
formation of ClO3 was the dominant reaction, and the proposed
(
3) Aieta, E. M.; Boyes, W.; Grasso, D.; Longley, K. E.; Robson, C. M.;
Rogers, S. E.; Scheible, O. K.; Williams, R. E.; White, G. C.; Zeh, T.
G. In Wastewater Disinfection; Chou, L., Sinozich, P., Eds.; Water
Pollution Control Federation: Alexandria, VA, 1986; pp 65-69.
4) Howard, W.; Strumila, G. B. In The Bleaching of Pulp, 3rd ed.; Singh,
R. P., Ed.; Tappi Press: Atlanta, GA, 1979; pp 113-157.
(9) USEPA. National Primary Drinking Water Regulations; Disinfectants
and Disinfection Byproducts; Proposed Rule. Fed. Regist. 1994, 59,
38668-38829.
(10) USEPA. National Primary Drinking Water Regulations; Disinfectants
and Disinfection Byproducts Notice of Data Availability; Proposed
Rule. Fed. Regist. 1998, 63, 15674-15692.
(
(5) Werdehoff, K. S.; Singer, P. C. J. AWWA 1987, 79, 107-113.
(6) Tang, T. F.; Gordon, G. EnViron. Sci. Technol. 1984, 18, 212-216.
(7) Peintler, G.; Nagyp a´ l, I.; Epstein, I. R. J. Phys. Chem. 1990, 94, 2954-
(11) USEPA. Priority List of Substances Which May Require Regulation
under the Safe Drinking Water Act. Fed. Regist. 1991, 56, 1470-
1474.
2958.
(12) Taube, H.; Dodgen, H. J. Am. Chem. Soc. 1949, 71, 3330-3336.
(13) Imagawa, H.; Fukagawa, M.; Yoshie, T. Nippon Kagsku Kaishi 1974,
2, 238-243.
(
8) Gordon, G.; Tachiyashiki, S. EnViron. Sci. Technol. 1991, 25, 468-
474.
1
0.1021/ic991486r CCC: $19.00 © 2000 American Chemical Society
Published on Web 05/04/2000

Reactions of Hypochlorous Acid with Chlorite Ion
Inorganic Chemistry, Vol. 39, No. 12, 2000 2615
-
mechanism is shown in eqs 4 and 5, where Cl2O2 was treated
as a steady-state species.
measurements, and we find no evidence of appreciable ClO
2
-ClO
2
complex formation. Detailed experimental procedures for the determi-
-
nation of ClO
3
by capillary electrophoresis were described in the
previous paper.14
ka
HOCl + HClO { } Cl O + H O
2
2
2
2
k-a
k ) 1.00 × 10 M s-1 (4)
5
-1
a
Results and Discussion
kb
-
+
Cl O + HOCl
98 ClO3 + Cl + H
2
2
2
Kinetics and ClO2 Yield. The rate expression is defined in
eq 6, where [HOCl]T ) [HOCl] + [OCl ]. In terms of the
formation of ClO2, the integrated rate expression is given in eq
7. The yield of ClO2 is calculated on the basis of the limiting
-
5
k /k ) 2.29 × 10 M (5)
-
-a
b
In the previous studies,7,8 the effect of buffer concentrations
was ignored and the mechanism for eq 5 with Cl2 formation
was not clear. Furman and Margerum14 observed a strong buffer
d[HOCl]T
-
effect on the rates of the HOBr/ClO2 reaction at pH 5-9. A
-
) kobsd[HOCl]T
(6)
(7)
dt
-
mechanism was proposed for this reaction with HOBrOClO
-
(
or HOBrClO2 ) as a steady-state species that undergoes
[ClO2]∞
ln
) kobsd
t
general-acid-catalyzed reactions to form a highly reactive
intermediate, BrOClO (or BrClO2). The subsequent rapid
reactions of BrOClO (or BrClO2) with OH yield ClO3 , while
competitive reactions with ClO2 lead to ClO2 formation. From
studies of the Cl2/ClO2 reaction in strong acidic solutions,
(
)
[
ClO ] - [ClO ]
2
∞
2 t
-
-
-
reagent [HOCl]T and the stoichiometry in eq 1. The observed
-
-
+
first-order rate constants (kobsd) vary with [ClO2 ], [H ], and
[buffer]T. In phosphate buffer, the kobsd values have a first-order
Emmenegger and Gordon15 reported that greater than 100%
yields of ClO2 were obtained on the basis of the stoichiometry
of eq 1. They suggested catalyzed disproportionation of ClO2-
might take place, but no mechanism was given. In the present
work, we show that phosphate and acetate buffers have large
effects on the kinetics of the reaction between HOCl and ClO2-
and the yield of ClO2. Mechanisms are proposed to explain the
anomaly of greater than 100% yields of ClO2. The reactivities
-
dependence on [ClO2 ], while the yields of ClO2 increase with
-
[ClO2 ] and approach 100% (Figure 1). Therefore, the observed
first-order rate constants do not depend on the yield of ClO2,
-
which indicates that the formation of ClO and ClO3 occurs
2
after the rate-determining steps. A mechanism is needed in
which an intermediate is formed from the interaction of HOCl,
-
ClO , and general acids in the rate-determining steps. This
2
-
-
and mechanisms are compared for the HOCl/ClO2 and the
intermediate must react very rapidly either with another ClO
2
-
HOBr/ClO2 reactions.
-
-
to give 2ClO2 or with OH to give ClO3 . The proposed
mechanisms are based both on detailed kinetic data that define
the rate-determining steps and on the yields of products which
depend on competitive reactions after the rate-determining steps.
Experimental Section
Proposed Mechanism for the HOCl/ClO2- Reaction
Reagents. Doubly distilled, deionized water was used for solution
preparations. “Chloride-free” hypochlorite solution was prepared by
(General-Acid Path). A mechanism in eqs 8-13 is proposed
-
-
the reaction of yellow HgO with Cl
2
(99.9+%) in CCl
2
4
at 0 °C with
for the HOCl/ClO2 reaction that is similar to the HOBr/ClO2
1
6
stirring to form Cl
2
O. The resulting Cl O solution was filtered, and
-
the filtrate was extracted with ice-cooled NaOH solution to give “Cl -
free” NaOCl solution. The hypochlorite solution was standardized
spectrophotometrically at 292 nm (ꢀOCl ) 362 M cm ). Sodium
chlorite was recrystallized by literature procedures.
KHOCl
a
-
+
HOCl {
} OCl + H
(8)
(9)
-
1
-1 14
7
,14,17
k1
The purity of
-
-
HOCl + ClO2 {k-1} HOClOClO
the recrystallized NaClO
titrimetry. Dilute solutions of ClO
2
was determined to be 98.7% by iodometric
-
2
-
were standardized spectrophoto-
k 2H A
1
-1 14
-
-
metrically at 260 nm (ꢀ ) 154 M cm ). Solution ionic strength
HOClOClO + HA 8 ClOClO + H O + A (10)
2
(
µ) was adjusted with NaClO
studies were carried out at 25.0 °C and µ ) 1.0 M.
Instrumental Methods. The measured pH was corrected to p[H ]
4
that was recrystallized from water. All
k3
ClOClO + ClO2-
2
-
+
8 2ClO + Cl
(11)
(12)
rapid
1
4
at 25.0 ( 0.1 °C and µ ) 1.0 M. Kinetic studies were performed on
an Applied PhotoPhysics SX.18MV stopped-flow spectrophotometer
k 4O H
-
-
-
+
ClOClO + OH rapid8 ClO3 + Cl + H
(
APPSF, optical path length ) 0.962 cm). The progress of the HOCl/
-
ClO
3
with ClO
2
reactions was followed by observing the formation of ClO
2
at
kA
5
-
-
-
+
-
1
-1 14
60 nm (ꢀClO2 ) 1230 M cm
)
under pseudo-first-order conditions
ClOClO + A + H O rapid8 ClO3 + Cl + H + HA (13)
2
-
2
in large excess over HOCl. All rate constants are an average
1
8
of three to five runs. Gordon and Emmenegger reported that ClO
and ClO
and molar absorptivity of 900 M cm at 360 nm. However, the
absorbance of ClO
2
-
-1
reaction.14 A steady-state species, HOClOClO , is formed (eq
9) which undergoes general-acid-catalyzed reactions to form
ClOClO (eq 10). Scheme 1 illustrates the proposed general-
acid-catalyzed path to form ClOClO. In subsequent rapid
-
2
can form a complex with a formation constant of 1.6 M
-
1
-1
-
2
at 360 nm was not taken into account in their
(
(
(
14) Furman, C. S.; Margerum, D. W. Inorg. Chem. 1998, 37, 4321-4327.
-
reactions, ClOClO reacts with ClO2 to form ClO2 (eq 11) and
15) Emmenegger, F.; Gordon, G. Inorg. Chem. 1967, 6, 633-635.
16) (a) Johnson, D. W.; Margerum, D. W. Inorg. Chem. 1991, 30, 4845-
-
-
reacts with OH and H2O, assisted by a general base (A ), to
-
4851. (b) Cady, G. H. Inorg. Synth. 1957, 5, 156-165.
form ClO3 (eqs 12 and 13) as shown in Scheme 2.
(
17) Fabian, I.; Gordon, G. Inorg. Chem. 1991, 30, 3785-3787.
The pKa value of HClO2 is 1.72 at 25.0 °C and µ ) 1.0 M.17
(
18) Gordon, G.; Emmenegger, F. Inorg. Nucl. Chem. Lett. 1966, 2, 395-
-
398.
At pH 4.7-7.5, almost all Cl(III) is in the form of ClO2 . The

2616 Inorganic Chemistry, Vol. 39, No. 12, 2000
Jia et al.
Scheme 2. Pathways for the Formation of ClO2 and ClO3-
from ClOClO
Figure 1. Dependence of observed first-order rate constants (kobsd) and
-
ClO
2
yields on [ClO
2
+
] in phosphate buffer. Conditions: [HOCl]
T
)
-
5
7
1
.17 × 10 M, p[H ] ) 6.44 ( 0.01, [PO
4 T
] ) 0.10 M, 25.0 °C, µ )
.0 M.
Scheme 1. General-Acid (HA)-Assisted Path To Form
ClOClO
M)19 is used to correct for OCl at high pH because it is not a
-
reactive species.
The yield of ClO2 (eq 16) depends on the competition among
-
-
the reactions of ClOClO with ClO2 (eq 11), OH (eq 12), and
H2O (eq 13). The reactions in eqs 11-13 only affect the yield
-
2
k [ClO ]
3
2
%
ClO )
× 100
2
(16)
2
A
5
-
OH
4
-
-
k [A ] + k [OH ] + 2k [ClO ]
3
of ClO2, not the rate. The yield of ClO2 increases with increasing
-
[
ClO2 ] (Figure 1) and decreases with increasing phosphate
buffer concentration (Figure 2) and with increasing pH (Figure
OH
3
). Curve-fitting the yields of ClO2 to eq 16 gives k4 /k3 )
5
HPO
4
(1.3 ( 0.1) × 10 and k5
/k3 ) 0.20 ( 0.02.
+
rate expressions (eqs 14 and 15) are derived from the above
d[HOCl]T
p[H ] and Phosphate Buffer Dependence on kobsd. The
observed first-order rate constants increase with decreasing
+
-
p[H ] (Figure 3). This is because acid converts OCl to the
reactive form of HOCl (eq 8) and also reacts with the
-
)
dt
-
HA
+
intermediate, HOClOClO , to form ClOClO and H2O (eq 10).
k
[H ]
2
-
In addition, our results show that the observed first-order rate
k1
[HA] [ClO ][HOCl]
∑k-1
2
T
[H ] + Ka
+
HOCl
-
(
)
[
]
constants increase with increasing [H2PO4 ] (Figure 2). General
-
(14)
acids such as H2PO4 can assist the reaction by transferring a
HA
2
k
-
proton to HOClOClO to form ClOClO and H2O (eq 10). The
1
+
[HA]
∑
7,8
k-1
buffer effect has been ignored in the previous studies. Least-
+
squares fitting of the observed rate constants with [H ] and
HA
+
-
-1 -1
H PO
2 4
k
[H ]
[H2PO4 ] to eq 15 gives k1 ) 1.6 ( 0.1 M s , k2
/k-1 )
2
-
k1
[HA] [ClO ]
2
-1
H
4
-1
∑k-1
8.5 ( 0.9 M , and k2 /k-1 ) (1 ( 1) × 10 M . A more
+
HOCl
(
)
[
]
[H ] + Ka
H
accurate value of k2 /k-1 is obtained from the studies in acetate
kobsd
)
(15)
HA
H O
buffer. The value of k2
HOCl/ClO2 Reactions in Acetate Buffer. A strong acetate
buffer dependence for the kobsd values was observed as shown
2
/k-1 is negligible.
k
2
1
+
[HA]
-
∑k-1
+
-
mechanism, in which HA is the general acid (H3O , H2PO4 ,
HOAc, or H2O). The pKa
(
19) Gerritsen, C. M.; Margerum, D. W. Inorg. Chem. 1990, 29, 2757-
HOCl
value of 7.47 (25.0 °C, µ ) 1.0
2762.

Reactions of Hypochlorous Acid with Chlorite Ion
Inorganic Chemistry, Vol. 39, No. 12, 2000 2617
Figure 2. Phosphate buffer effect on the rate constants and yields of
-
-4
ClO
2
+
. Conditions: [ClO
2
] ) 0.0444 M, [HOCl]
T
) 1.11 × 10 M,
p[H ] ) 6.45 ( 0.02, 25.0 °C, µ ) 1.0 M.
2
Figure 4. Acetate buffer effect on the rate constants and ClO yields.
-
-5
+
Conditions: [ClO
)
2
] ) 0.0101 M, [HOCl]initial ) 4.71 × 10 M, p[H ]
4.70 ( 0.01, 25.0 °C, µ ) 1.0 M.
+
Figure 3. Effect of p[H ] on the observed first-order rate constants
-
(
0
1
k
obsd) and yields of ClO
2
in phosphate buffer. Conditions: [ClO
2
] )
-
4
.0475 M, [HOCl]
.0 M.
T
) 1.01 × 10 M, [PO
4 T
] ) 0.10 M, 25.0 °C, µ )
in Figure 4. From the intercept of Figure 4 and the k1 value
2
Figure 5. Dependence of the rate constants and the yields of ClO on
H
-
-5
determined in the phosphate buffer studies, the value of k2 /
[ClO
2
] in acetate buffer. Conditions: [HOCl]initial ) 5.56 × 10 M,
+
T
4
-1
k-1 is calculated to be (1.6 ( 0.5) × 10 M , in accord with
p[H ] ) 4.71 ( 0.01, [HOAc] ) 0.2034 M, 25.0 °C, µ ) 1.0 M.
H O
eq 15. The value of k2
2
/k-1 is negligible. The yields of ClO2,
which appear to be greater than 140% at low acetate concentra-
tions, decrease to 102% with increasing total acetate buffer
concentration (from 0.02 to 0.6 M) at p[H ] 4.70 (Figure 4).
first-order rate constants in high concentrations of acetate buffer
are much greater than expected from the general-acid path (eqs
8-13). The extent of this deviation is shown in Figure 4. Then
we will discuss reasons that the apparent yields of ClO2 at low
pH can become much larger than 100%.
+
The anomalously high ClO2 yields are not due to [HOAc]T.
Indeed, we propose the ClO2 yields decrease in Figure 4 because
-
-
-
OAc assists the hydrolysis of ClOClO to form ClO3 and Cl
(
Proposed Acetyl Hypochlorite Mechanism for the HOCl/
eq 13). Figure 5 shows that the rate of the HOCl/ClO2- reaction
-
ClO2 Reaction in Acetate Buffer. Acetyl hypochlorite, CH3-
in acetate buffer does not have a simple first-order dependence
C(O)OCl (or ClOAc), has been prepared by the reaction of Cl2
-
20
on [ClO2 ]. The yield of ClO2 first increases (to a maximum
with mercuric or silver acetate in glacial acetic acid. Rate
-
-1 -1
of 116%) and then decreases to 101% as [ClO2 ] increases.
constants of 0.1-0.24 M
s
have been reported for the
-
Two aspects of the behavior of these HOCl/ClO2 reactions
need to be addressed. We will first consider why the observed
(20) Anbar, M.; Dostrovsky, I. J. Chem. Soc. 1954, 1105-1108.

2618 Inorganic Chemistry, Vol. 39, No. 12, 2000
Jia et al.
reaction of HOCl and HOAc to form ClOAc.21-23 An acetyl
hypochlorite path (eqs 17-20) is proposed for the HOCl/ClO2
-
Table 1. Rate Constants for the Reactions of ClO
HOBr at µ ) 1.0 M and 25.0 °Ca
2
with HOCl and
-
-
b
- c
2
rate const
HOBr/ClO
2
HOCl/ClO
-
1
k
k
k
k
k
k
k
k
k
k
k
1
2
2
2
4
5
5
6
7
7
9
, M s-1
97 ( 6
1.6 ( 0.1
H
-1
5
4
5
/k-1, M
(3.1 ( 0.5) × 10
8.3 ( 0.6
(1.6 ( 0.5) × 10
HOAc
H PO
/k-1, M-1
/k-1, M
20
8.5 ( 0.9
(1.3 ( 0.1) × 10
0.20 ( 0.02
0.06
-1
2
4
OH
3
/k
3
(6.83 ( 0.05) × 10
HPO
OAc
4
/k
/k
3
3
1
-
-1
, M
s
4.5 ( 0.3
H
/k-6, M-2
K
5
K
C
HOAc
(2.8 ( 0.6) × 10
-2
C
/1k-6, M
46 ( 4
1 × 10
-
4
/k
8
, M
a
-
pK ) 7.47 for HOCl, 6.26 for H
a
2
PO
4
, 4.61 for HOAc; pK )
w
1
3.60. b Reference 14. c This work.
reaction in acetate buffer in addition to the general-acid path
-
and q is the number of equivalent basic sites on A . The slope
(
eqs 8-13). Acetyl hypochlorite is formed from HOCl and
(
R) of 0.41 indicates a significant degree of proton transfer from
the general acid (HA) to HOClOClO (Scheme 1) in the
/k-1 in the general-acid path
is obtained from the fit of data to eqs 15 and 21 at low
concentrations of acetate buffer. This value (20 M ) agrees
with that estimated from the Brønsted-Pedersen relationship.
Greater Than 100% Yields of ClO2 and ClOClO-Induced
Disproportionation of ClO2 . If eq 1 gives a 100% yield of
ClO2, then ClO3 formation (eq 2) would not be expected
-
HOAc (eq 17), and it reacts rapidly with ClO2 to form an
adduct, AcOClOClO (eq 18). This adduct can undergo general-
-
-
HOAc
transition state. The value of k2
acid-catalyzed reactions to form ClOClO (eqs 19 and 20), which
-
-
-
reacts rapidly with ClO2 , OH , and OAc (eqs 11-13).
-
1
Equation 18 is proposed as a rapid equilibrium unlike eq 9
because ClOAc is a better electrophile than HOCl. The
equilibrium formation constant of ClOAc (measured in acetic
acid with low H2O concentrations) is only 0.0025. Therefore,
acetyl hypochlorite and its ClO2 adduct are treated as combined
2
4
-
25
-
-
because these two reactions are in competition with one another.
However, our experimental results (Table 2) show that signifi-
cant amounts of ClO3 are generated even when yields of ClO2
steady-state species. The observed first-order rate constant is
derived in eq 21.
-
-
H
7
+
HOAc
7
are greater than 100% (based on the initial limiting [HOCl]T).
k K [HOAc][ClO ](k [H ] + k
[HOAc])
6
C
2
-
The sums of the yields of ClO2 and ClO3 average 115% from
kobsd
)
(21)
H
+
HOAc
7
-
2
-
k-6 + K (k [H ] + k
[HOAc])[ClO ]
pH 6.2 to 4.7. Since ClO2 is in excess in these experiments,
C
7
any disproportionation (eq 23) to give additional HOCl and
In the proposed mechanism (eqs 17-20), HOAc has two
2ClO2- + H f HOCl + ClO
+
-
(23)
+
roles. One is to form a very strong Cl -transfer agent (eq 17),
3
-
and the other is to assist the breakup of the AcOClOClO adduct
+
-
-
4
ClO2- + 2H f 2ClO + ClO + Cl + H O (24)
(
eq 20). This gives a much larger acceleration to the rate than
2 3 2
the “normal” role of HOAc as a general acid (eq 10). The kinetic
data from acetate buffer dependence (Figure 4) and [ClO2 ]
-
-
ClO3 would account for this behavior, because the HOCl and
ClO2- would react further to give ClO2 (eq 1) with the overall
dependence (Figure 5) are fit to the sum of eq 15 for the general-
acid path and eq 21 for the ClOAc path. Figure 4 shows the
domination of the ClOAc path compared to the general-acid
-
stoichiometry in eq 24. However, the rate of uncatalyzed ClO2
disproportionation
effect. Hence, we propose the mechanism in eqs 25-27 to
28-30
is much too slow to explain the observed
path. Previous estimates2
1-23
of the k6 value have not considered
the importance of a much larger k-6 value (i.e., reversibility).
ClOClO + ClO2- f ClOCl(O)OClO-
(25)
(26)
(27)
-1
-1
Our steady-state treatment gives a k6 value of 4.5 M s , which
is 20 times larger than the previous estimates. The resolved rate
constants are summarized in Table 1.
-
-
ClOCl(O)OClO f OCl + Cl O
2
3
The Brønsted-Pedersen relationship26 for the k2HA/k-1 term
-
+
Cl O + H O f HOCl + ClO + H
2
3
2
3
(
eq 22) gives log Ga ) 3.14 ( 0.04 and R ) 0.41 ( 0.01,
HA
+
27
where the pKa values are -1.72 (H3O ), 4.61 (HOAc), and
account for a ClOClO-induced reaction that causes HOCl to be
generated and thus leads to more ClO2 than expected. The first
step (eq 25) has the same reactants shown in eq 11, but we
HA
HA
k2
Ka q
log
) log G + R log
(22)
a
(
)
(
)
k p
p
-
-1
now suggest ClOCl(O)OClO as another intermediate that can
dissociate via Cl-O bond cleavage to give different products.
In one path (k8 in Scheme 2), loss of a terminal Cl leads to
-
14
6
.26 (H2PO4 ); p is the number of equivalent protons on HA,
-
OCl(O)OClO, an unstable dimer of ClO2 that undergoes rapid
(
(
(
(
21) Israel, C. C. J. Chem. Soc. 1950, 1286-1289.
homolytic bond cleavage to give 2ClO2. In the other path (k9
22) Chung, A.; Israel, C. C. J. Chem. Soc. 1955, 2667-2673.
23) Anbar, M.; Dostrovsky, I. J. Chem. Soc. 1954, 1094-1104.
24) de la Mare, P. B. D.; Hilton, I. C.; Varma, S. J. Chem. Soc. 1960,
-
-
in Scheme 2), loss of ClO from ClOCl(O)OClO generates
HOCl and Cl2O3 (eq 26). The latter species can react with H2O
4044-4054.
-
to give another HOCl and ClO3 (eq 27). This second path
(
(
(
25) de la Mare, P. B. D.; Hilton, I. C.; Vernon, C. A. J. Chem. Soc. 1960,
provides additional HOCl that ultimately leads to higher yields
4039-4044.
26) Bell, R. P. The Proton in Chemistry, 2nd ed.; Cornell University:
Ithaca, NY, 1973; p 198.
27) Portanova, R.; Bernardo, P. D.; Cassol, A.; Tondello, E.; Magon, L.
Inorg. Chim. Acta 1974, 8, 233-240.
(28) Hong, C. C.; Rapson, W. H. Can. J. Chem. 1968, 46, 2053-2060.
(29) Kieffer, R. G.; Gordon, G. Inorg. Chem. 1968, 7, 235-239.
(30) Kieffer, R. G.; Gordon, G. Inorg. Chem. 1968, 7, 239-244.

Reactions of Hypochlorous Acid with Chlorite Ion
Inorganic Chemistry, Vol. 39, No. 12, 2000 2619
-
-
8
Table 2. Determination of ClO
3
by CZE from the HOCl/ClO
2
results disagree with the previous explanation for the formation
a
Reaction in 0.02 M [HOAc]
T
-
of ClO3 from the reaction of Cl2O2 with HOCl (eq 5).
Structures of the Proposed Intermediates. To help under-
p[H+] [HOCl]
, mM % ClO
% ClO
-
∑(% ClO
+ % ClO
-)
3
T
2
3
2
stand if the proposed intermediates in Schemes 1 and 2 are
6
5
5
5
5
5
4
4
.22
.42
.26
.19
.15
.05
.98
.72
3.01
1.50
1.50
3.01
1.177
1.05
0.525
1.177
90
105
101
96
102
109
121
106
16 ( 1
11 ( 2
106
3
2,33
116
109
104
115
115
127
125
plausible, ab initio calculations were used.
Equilibrium
7.6 ( 0.4
8 ( 2
geometries were determined from both B3LYP/6-31G* and
B3YLP/6-311++G(3df, 3pd) calculations. The latter results give
the molecular structures of the intermediates, bond distances,
and atomic charges shown in Figure 6. One proposed intermedi-
13 ( 2
6 ( 2
6 ( 1
3
4
19 ( 2
ate, Cl2O3, has been observed in the gas phase, and ClOCl-
O)O was calculated to be the most stable structure by Clark
(
a
ClO
2
-] ) 0.01 M; pressure injection 25 mbar, 10 s; separation
[
33
and Francisco. The relative energetics of the reaction pathways
are given in the Supporting Information. Aqueous solvation will
undoubtedly have a large effect on their relative stabilities and
reactivities. However, the calculated structures serve as useful
models for the proposed intermediates.
voltage -25 kV.
of ClO2 even though ClO3- is also formed. Scheme 2 shows
the proposed pathways to account for the effects of pH and
buffer on the observed yields. It must be emphasized that these
steps all occur after the rate-determining steps and are based
on logical explanations of how the yields are affected by changes
-
-
Comparison between the HOCl/ClO2 and HOBr/ClO2
1
4
-
Reactions. Previous studies showed that HOBr and ClO2
OH
in conditions. The k4 path accounts for the decreased yield
react by the same mechanism as given in Scheme 1 for the HOCl
of ClO2 as the pH increases, and the k5A path explains why
high acetate ion concentrations reduce the yield of ClO2.
At low pH and low buffer concentrations, the k4 and k5A
pathways are negligible. If the fraction of the ClOClO/ClO2-
reaction by the k9 path is x, then the fraction of this reaction by
-
and ClO2 reaction. The HOBr reactions are faster, where the
k1 step is a factor of 60 times larger than that for HOCl. The
OH
experimentally observed rate constants with acid-assisted con-
+
-
3
tributions from H3O and H2PO4 vary from 60 to 10 for
kobsd /kobsd
HOBr
HOCl
, depending on the pH and buffer concentra-
k8 path is (1 - x). The overall stoichiometry is described in eq
tions.
-
2
8. The yields of ClO2 and ClO3 in terms of HOCl are given
There are several reasons why we prefer to designate the
initial steady-state intermediates as chainlike structures, i.e.,
in eqs 29 and 30. When x ) 0.2, the predicted yield of ClO2 is
-
-
HOBrOClO and HOClOClO rather than Y-shaped structures
-
-
-
withhalogen-halogenbonds,i.e.,HOBrCl(O)O andHOClCl(O)O .
First, studies by Perrone and Margerum of HOCl reactions
with BrO2 require a chainlike adduct (HOClOBrO ) to account
for the formation of ClO2 as a product. Second, ab initio
calculations show the chainlike structure is preferred. Third,
(
1 - 2x)HOCl + 2ClO2
f
3
5
-
-
2
(1 - x)ClO + xClO3 + (1 - x)Cl (28)
-
-
2
1
- x
%
ClO )
× 100
× 100
(29)
(30)
36
2
1
- 2x
the proposed mechanism in Scheme 2 with Cl2O3 as an
ClO3-
)
x
intermediate is more favorable for ClOCl(O)O than for ClClO3.
33
%
1
- 2x
We cannot rule out the possibility of some competing paths
with halogen-halogen bonding for the intermediates, but we
suggest that the chainlike structures are preferred.
1
4
33%, which is close to the experimental result of 142% at pH
.7 and [HOAc]T ) 0.02 M in Figure 4. As the [HOAc]T is
Although the k1 step corresponds to the formation of an adduct
increased to 0.5 M (Figure 4), the yield of ClO2 drops to 105%.
-
between a Lewis base (OClO ) and a weak Lewis acid (HOCl
-
We suggest that the high OAc concentration causes this 37%
decrease in ClO2 yield, which would mean an 18.5% increase
-1 -1
or HOBr), the rate constants of 1.6 and 97 M
s
are much
smaller than those for most Lewis acid-base reactions. Once
again, ab initio calculations are helpful in understanding why
this is the case. The O-Cl (1.707 Å) bond length in HOCl
increases by 0.14 Å, and the O-Cl (1.576 Å) bond lengths in
-
in ClO3 formation. This would correspond (approximately) to
OAc
-
-
k5 /k3 ) (0.185/0.105)([ClO2 ]/[OAc ]) ) 0.06, which is
OH
5
reasonable compared to values of k4 /k3 ) 1.3 × 10 and
HPO
-
k5
4
/k3 ) 0.20, because OAc is a much weaker base. Much
higher concentrations of HOCl were used in the CZE experi-
ments. The ClO2 and ClO3 yields in Table 2 at low pH
correspond roughly to x ) 0.1 (113% ClO2 and 13% ClO3 ).
(
32) Molecular orbital calculations were performed using the Gaussian 94
program: Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Gill, P. M.
W.; Johnson, B. G.; Robb, M. A.; Cheeseman, J. R.; Keith, T.;
Petersson, G. A.; Montgomery, J. A.; Raghavachari, K.; Al-Laham,
M. A.; Zakrzewski, V. G.; Ortiz, J. V.; Foresman, J. B.; Cioslowski,
J.; Stefanov, B. B.; Nanayakara, A.; Challacombe, M.; Peng, C. Y.;
Ayala, P. Y.; Chen, W.; Wong, M. W.; Andres, J. L.; Replogle, E. S.;
Gomperts, R.; Martin, R. L.; Fox, D. J.; Binkley, J. S.; Defrees, D. J.;
Baker, J.; Stewart, J. P.; Head-Gordon, M.; Gonzales, C.; Pople, J. A.
Gaussian 94, Revision D.2; Gaussian, Inc.: Pittsburgh, PA, 1995.
-
-
Greater than 100% yields of ClO2 were also observed in the
-
+
15
Cl2/ClO2 reaction ([H ] ) 0.2 M) and in the disproportion-
ation31 of ClO2 ([H ] ) 0.01-2 M). The disproportionation
-
+
-
of chlorous acid in the absence of Cl followed the same
stoichiometry as in eq 24, where the limiting step was proposed
to be the slow formation of HOCl (eq 31).2
8-30
We propose
(33) All equilibrium geometrical parameters were fully optimized, using
Schlegel’s method, to better than 0.001 Å for bond distances and 0.10°
for bond angles with a self-consistent field convergence of at least
slow
+
-
-
9
2HClO2
8 HOCl + H + ClO3
(31)
10 on the density matrix: Clark, J.; Francisco, J. S. J. Phys. Chem.
A 1997, 101, 7145-7153.
34) (a) Hayman, G. D.; Cox, R. A. Chem. Phys. Lett. 1989, 155, 1-7. (b)
Burkholder, J. B.; Mauldin, R. L., III; Yokelson, R. J.; Solomon, S.;
Ravishankara, A. R. J. Phys. Chem. 1993, 97, 7597-7605. (c) Friedl,
R. R.; Birk, M.; Oh, J. J.; Cohen, E. A. J. Mol. Spectrosc. 1995, 170,
(
that HOCl or Cl2 will react with ClO2- to form ClOClO (eqs 9
and 10), which again reacts with ClO2 to form an adduct,
ClOCl(O)OClO . At low pH, this adduct can generate more
HOCl, which leads to more ClO2 and ClO3 (Scheme 2). Our
-
-
3
83-396.
-
(
35) Perrone, T. F.; Margerum, D. W. Inorg. Chem. To be submitted for
publication.
(31) Schmitz, G.; Rooze, H. Can. J. Chem. 1981, 59, 1177-1187.
(36) Guha, S.; Francisco, J. S. Chem. Phys. Lett. Submitted for publication.

2620 Inorganic Chemistry, Vol. 39, No. 12, 2000
Jia et al.
-
Figure 6. Equilibrium geometries (bond distances in angstroms) and atomic charges for key intermediates in the HOCl/ClO
2
reactions.
ClO2- change by +0.014 and -0.024 Å when the HOClOClO-
Supercomputer was provided by funding from the NASA
Mission to Planet Earth, Aeronautics, and Space Science
Program.
(
Figure 6a) adduct forms. Similar changes are found for the
-
formation of HOBrOClO , where the O-Br (1.844 Å) bond
length in HOBr increases by 0.139 Å to form the adduct. These
rearrangements will contribute significantly to the activation
barriers in the formation of the adducts.
Supporting Information Available: Tables of kinetic data, equi-
librium geometries, atomic charges, zero-point energies, and relative
energies and figures showing a Brønsted-Pedersen plot and a relative
energy diagram. This material is available free of charge via the Internet
at http://pubs.acs.org.
Acknowledgment. This work was supported by National
Science Foundation Grant CHE-98-18214. We thank the Jet
Propulsion Laboratory at the California Institute of Technology
for providing generous support of supercomputer time. The Cray
IC991486R