New pathways for chlorine dioxide decomposition in basic solution

Article abstract of DOI:10.1021/ic0204676

The product distribution from the decay of chlorine dioxide in basic solution changes as the ClO₂ concentration decreases. While disproportionation reactions that give equal amounts of ClO₂^- and ClO₃^- dominate the stoichiometry at millimolar or higher levels of ClO₂, the ratio of ClO₂^- to ClO₃^- formed increases significantly at micromolar ClO₂ levels. Kinetic evidence shows three concurrent pathways that all exhibit a first-order dependence in [OH^-] but have variable order in [ClO₂]. Pathway 1 is a disproportionation reaction that is first order in [ClO₂]. Pathway 2, a previously unknown reaction, is also first order in [ClO₂] but forms ClO₂^- as the only chlorine-containing product. Pathway 3 is second order in [ClO₂] and generates equal amounts of ClO₂^- and ClO₃^-. A Cl₂O₄ intermediate is proposed for this path. At high concentrations of ClO₂, pathway 3 causes the overall ClO₃^- yield to approach the overall yield of ClO₂^-. Pathway 2 is attributed to OH^- attack on an oxygen atom of ClO₂ that leads to peroxide intermediates and yields ClO₂^- and O₂ as products. This pathway is important at low levels of ClO₂.

Full text of DOI:10.1021/ic0204676

Inorg. Chem. 2002, 41, 6500−6506
New Pathways for Chlorine Dioxide Decomposition in Basic Solution
Ihab N. Odeh, Joseph S. Francisco, and Dale W. Margerum*
Department of Chemistry, Purdue UniVersity, West Lafayette, Indiana 47907-1393
Received July 19, 2002
The product distribution from the decay of chlorine dioxide in basic solution changes as the ClO concentration
2
decreases. While disproportionation reactions that give equal amounts of ClO - and ClO - dominate the stoichiometry
2
3
at millimolar or higher levels of ClO , the ratio of ClO - to ClO - formed increases significantly at micromolar ClO
2
2
3
2
levels. Kinetic evidence shows three concurrent pathways that all exhibit a first-order dependence in [OH-] but
have variable order in [ClO ]. Pathway 1 is a disproportionation reaction that is first order in [ClO ]. Pathway 2, a
2
2
previously unknown reaction, is also first order in [ClO ] but forms ClO - as the only chlorine-containing product.
2
2
Pathway 3 is second order in [ClO ] and generates equal amounts of ClO - and ClO -. A Cl2O intermediate is
2
2
3
4
proposed for this path. At high concentrations of ClO , pathway 3 causes the overall ClO - yield to approach the
2
3
overall yield of ClO -. Pathway 2 is attributed to OH- attack on an oxygen atom of ClO that leads to peroxide
2
2
intermediates and yields ClO - and O as products. This pathway is important at low levels of ClO .
2
2
2
Introduction
OH-. Disproportionation rate expressions with both a first-
order and a second-order dependence in ClO2 concentration
have been cited several times.10-13 However, the rate
constants are not in agreement and it is not clear how the
kinetic data were fit to the rate expression.
Recent use of chlorine dioxide for the disinfection of
anthrax-contaminated premises1 as well as its growing use
in water disinfection,2 wastewater treatment,3 pulp bleaching,4
chemosterilization,5 and food preservation6 shows the need
for a better understanding of the chemical reactions of this
free radical. The rate of decomposition of ClO2 in neutral
aqueous solutions is quite slow,7 but its decay is accelerated
by base.8 The disproportionation reaction in eq 1 has been
proposed by many investigators.8-13 Bray8 proposed a third-
order rate expression for this disproportionation (eq 2) with
a rate constant of 15.3 M-2 s-1 at 19 °C. Halperin and Taube9
studied the general features of ClO2 decay in base with 18O-
labeling techniques at millimolar ClO2 levels and concluded
that a dimer, Cl2O4, forms which subsequently reacts with
2ClO2 + 2OH- f ClO2- + ClO3- + H2O
-d[ClO2]
(1)
(2)
) k[OH-][ClO2]2
dt
In this work we show that an integrated rate expression
can be used to determine reliable first-order and second-
order rate constants for the mixed-order decay of ClO2 over
a wide concentration range. Furthermore, we show that at
-
low ClO2 concentrations the yield of ClO2 is greater than
-
the yield of ClO3 . We attribute this behavior to the presence
* To whom correspondence should be addressed. E-mail: margerum@
chem.purdue.edu.
of the decay reaction in eq 3, where ClO2 oxidizes the
solvent. The mixed-order integrated rate expression and the
(1) Ritter, S. K. Chem. Eng. News 2001, 79 (48), 24-26.
(2) Katz, A.; Narkis, N. Wat. Res. 2001, 35, 101-108.
(3) Stevens, A. A. EnViron. Health Perspect. 1982, 46, 101-110.
(4) Hart, P. W.; Hsieh, J. S. Chem. Eng. Commun. 1993, 126, 27-41.
(5) Rosenblatt, D. H. U.S. Patent 4 504 442, 1983.
(6) Tsai, L. S.; Huxsoll, C. C.; Roberston, G. J. Food Sci. 2001, 66, 472-
477.
-
use of ion chromatographic measurements of ClO2 and
ClO3- yields permit the identification of three pathways for
the decay of ClO2. The reaction orders and product yields
are different for the proposed pathways, but all three
mechanisms have base-assisted electron-transfer steps.14
(7) Von Heijne, G.; Teder, A. Acta Chem. Scand. 1973, 27, 4018-4019.
(8) Bray, W. C. Z. Anorg. Allg. Chem. 1906, 48, 217-250.
(9) Halperin, J.; Taube, H. J. Am. Chem. Soc. 1952, 74, 375-380.
(10) Granstrom, M. L.; Lee, G. F. Public Works 1957, 88, 90-92.
(11) Gordon, G.; Keiffer, R. G.; Rosenblatt, D. H. In Progress in Inorganic
Chemistry; Lippard, S. J., Ed.; Wiley-Interscience: New York, 1972;
pp 201-287.
4ClO2 + 4OH- f 4ClO2- + O2 + 2H2O
(3)
(12) Emerich, D. E. Ph.D. Dissertation, Miami University, 1981.
(13) Gordon, G. Pure Appl. Chem. 1989, 61, 873-878.
(14) Wang, L.; Nicoson, J. S.; Huff Hartz, K. E.; Francisco, J. S.; Margerum,
D. W. Inorg. Chem. 2002, 41, 108-113.
6500 Inorganic Chemistry, Vol. 41, No. 24, 2002
10.1021/ic0204676 CCC: $22.00 © 2002 American Chemical Society
Published on Web 11/05/2002

Chlorine Dioxide Decomposition in Basic Solution
Experimental Section
Reagents. All solutions were prepared with doubly deionized,
distilled water. Chlorine dioxide was prepared as described
elsewhere14 and was protected from light and stored in a refrigerator.
The stock ClO2 solution was standardized spectrophotometrically
at 359 nm (ꢀ ) 1230 M-1 cm-1).15 Commercially available NaClO2
was recrystallized using a previously described procedure,16 and
its purity was determined by ion chromatography. Stock solutions
of NaClO2 were standardized spectrophotometrically at 260 nm (ꢀ
) 154.0 M-1 cm-1).15 Ionic strength, µ, was adjusted to 1.0 M
with recrystallized NaClO4.
pH Measurement. An Orion model 720A digital pH meter
equipped with a Corning combination electrode was used for pH
measurements. The electrode was calibrated with previously
standardized HClO4 and NaOH to correct pH to p[H+] when pKw
is 13.60 (25.0 °C, 1.0 M NaClO4).17
Kinetics. Kinetic traces for ClO2 decay in base were acquired
with a Perkin-Elmer Lambda 9 UV-vis-NIR spectrophotometer.
The kinetics of the ClO2 reaction with HO2- were followed by use
of an Applied PhotoPhysics SX18 MV stopped-flow spectropho-
tometer (APPSF, optical path length ) 0.962 cm). These reactions
were followed by measuring the loss of ClO2 at 359 nm. SigmaPlot
8.018 was used for regression analyses.
Figure 1. Kinetic trace and regression results for the decomposition of
ClO2 in basic solution. Conditions: 0.37 mM ClO2; 0.40 M NaOH; 25.0-
(2) °C; µ ) 1.0 M. Circles are the experimental data. The mixed-order
dependence in [ClO2] regression line is based on eq 5. Insert shows the
decomposition of ClO2 at 25 µM levels, with all other conditions as above
(circles). The mixed-order fit is obtained using eq 5. ka,obsd ) 6.19(3) ×
Product Analysis. A Dionex DX-500 chromatograph was used
to measure yields of ClO2- and ClO3- by a method similar to EPA
300.1.19 Samples were injected via an autosampler (AS 40) through
a 25 µL injection loop onto quaternary amine anion-exchange guard
(AG9 HC) and separation (AS9 HC) columns. Analytes were eluted
with 9 mM Na2CO3 at a flow rate of 1.0 mL/min. Gas-assisted
suppressed-conductivity detection (ED 40), with an ASRS-Ultra
suppressor in the self-regenerating mode and a current of 100 mA,
was used to detect the analytes. Residual ClO2 was purged with
Ar prior to injection onto the column. Ionic strength was not
adjusted in the stoichiometric analyses. The use of polypropylene
as opposed to glass containers12 and alternative preparation methods
of ClO2 synthesis20 did not affect the product distribution.
Computation. Calculations were performed using the GAUSS-
IAN 98 program21 to determine plausible structures for the reaction
intermediates. Equilibrium geometries were optimized using the
Becke three-parameter hybrid functional combined with the Lee,
Yang, and Parr correlation (B3LYP) density functional theory
method. Initial geometry optimizations were done at the B3LYP
level of theory using the 6-311G(d) basis set. The calculations with
the large 6-311++G(3df,3pd) basis set were also performed and
included. All equilibrium geometries were fully optimized to better
than 0.001 Å for bond distances and 0.1° for bond angles. All
energies were corrected for zero-point energies as determined from
harmonic vibrational frequency calculations.
10-4 s-1, and kb,obsd ) 8.28(3) M-1 s-1
.
a mixed-order (combination of first-order and second-order)
dependence in [ClO2] as given in eq 4. This is the case over
a wide range of ClO2 concentrations (see insert of Figure
1). An integrated rate expression (eq 5) gives excellent fits
for all the kinetic data, where [ClO2]0 is the initial concentra-
tion of ClO2 and [ClO2]t is the concentration of ClO2 at any
time during the course of the reaction.
-d[ClO2]
) ka,obsd[ClO2] + kb,obsd[ClO2]2
(4)
(5)
dt
a,obsdt
k
a,obsd[ClO2]0e-k
[ClO2]t )
ka,obsd + kb,obsd([ClO2]0 - [ClO2]0e-k
)
a,obsdt
This expression permits resolution of ka,obsd and kb,obsd values
as the OH- concentration varies from 0.05 to 0.45 M (Figure
2) and shows that there is a first-order dependence in [OH-]
for both ka,obsd and kb,obsd (eq 6). The resolved values are ka
) 1.38(8) × 10-3 M-1 s-1 and kb ) 21.8(4) M-2 s-1 at 25.0
°C, µ ) 1.0 M. The ka value is a factor of 2.3 smaller and
the kb value is 1.3 larger than the corresponding values of
Granstrom and Lee.10 However, our stoichiometric studies
show that ka represents two first-order pathways rather than
one. The data show, as will be detailed later, that ClO2
decomposition proceeds by three concurrent pathways.
Results and Discussion
Kinetics. The decay of ClO2 in basic solution (with excess
[OH-]) does not fit the second-order dependence in [ClO2]
that is given in eq 2, nor does it fit a simple first-order
dependence in [ClO2]. As shown in Figure 1, the decay fits
-d[ClO2]
) ka[OH-][ClO2] + kb[OH-][ClO2]2 (6)
dt
(15) Furman, C. S.; Margerum, D. W. Inorg. Chem. 1998, 37, 4321-4327.
(16) Jia, Z.; Margerum, D. W.; Francisco, J. S. Inorg. Chem. 2000, 39,
2614-2620.
(17) Molina, M.; Melios, C.; Tognolli, J. O.; Luchiari, L. C.; Jafelicci, M.
J. Electroanal. Chem. Interfacial Electrochem. 1979, 105, 237-246.
(18) SigmaPlot 8.0 for Windows; SPSS Inc.: Chicago, IL, 2002.
(19) EPA Method 300.1; U.S. EPA: Cincinnati, OH, 1997.
(20) Granstrom, M. L.; Lee, G. F. J. AWWA 1958, 50, 1453-1466.
(21) Frisch, M. J.; et al. GAUSSIAN 98; Gaussian Inc.: Pittsburgh, PA,
1998.
Stoichiometry. Ion chromatography shows that ClO2- and
ClO3- are the only chlorine-containing products formed from
the decomposition of ClO2 in basic solution. However, the
ratio of ClO2 to ClO3 is not 1:1 as required for the
disproportionation reaction (eq 1) and reported by several
-
-
Inorganic Chemistry, Vol. 41, No. 24, 2002 6501

Odeh et al.
Scheme 1. Proposed Mechanism for the Decomposition of Chlorine
Dioxide in Basic Solution
Figure 2. Dependence of the observed rate constants on [OH-]: (a) ka,obsd
;
(b) kb,obsd. Conditions: 0.37 mM ClO2; 0.05-0.45 M NaOH; 25.0(2) °C; µ
) 1.0 M. ka ) 1.38(8) ×10-3 M-1 s-1, and kb ) 21.8(4) M-2 s-1
.
formed during ClO2 decomposition. A flow-through ap-
paratus allowed a continuous stream of ClO2(g) to bubble
slowly through a flask containing 0.20 M NaOH. This pro-
cedure allowed the ClO2 that decomposed to be replenished
and maintained a ClO2 concentration of 0.10(2) mM in solu-
tion. Spectrophotometric analysis at two wavelengths (molar
absorptivities (M-1 cm-1) at 260 nm are ꢀOCl- ) 87.9 and
ꢀClO2- ) 154, and at 295 nm ꢀOCl- ) 356 and ꢀClO2- ) 125)
was used to determine whether OCl- was present. No OCl-
could be detected (<1 µM) under conditions where 0.57(1)
Figure 3. Effect of increasing [ClO2]0 on the [ClO2-]F/[ClO2]0 ratio due
to ClO2 decomposition in basic solution (circles). Conditions: 0.05-1.58
mM ClO2; 0.20 M NaOH. Predicted [ClO2-]F/[ClO2]0 is based on the
relative contribution from pathways 1 and 2 following eq 15 (solid line). R
) 0.60(3), and â ) 0.40(2). The dashed line is the predicted [ClO2-]F/
[ClO2]0 ratio if only pathways 2 and 3 existed (eq 6, where ka corresponds
to pathway 2 only and kb corresponds to pathway 3). The subscript F in
[ClO2-]F refers to the final levels of chlorite formed upon the completion
of the ClO2 decay reaction.
-
mM ClO2 formed. Ion chromatographic results for these
conditions show that 0.560(4) mM ClO2- and 0.436(5) mM
-
ClO3 formed. The values obtained by the ion chromato-
graphic method are more precise than those determined using
spectrophotometry and are in agreement within the error.
-
Low concentrations of ClO2 favor ClO2 as a product
-
-
compared to nearly equimolar levels of ClO2 and ClO3
formed at high ClO2 concentrations. This suggests that the
second-order pathway, which is preferred at higher concen-
trations, is responsible for the products in eq 1. However,
the second-order pathway is not sufficient to account for all
the chlorate formed. This indicates the presence of two
pathways with a first-order dependence in ClO2, one produces
authors.8,10,12 As shown in Figure 3, the percent yield of
ClO2- becomes greater than that of ClO3- as the concentra-
-
tion of ClO2 decreases. At 74 µM ClO2, the ClO2 yield is
-
67.5% of the chlorine-containing products while the ClO3
yield is only 32.5%. As will be shown, there is a need to
introduce three pathways to account for all products formed.
It is clear that something must be oxidized to balance the
-
-
equal amounts of ClO2 and ClO3 and the other forms
-
ClO2 and O2. The product distribution is not affected by
-
excess reduction to ClO2 , and oxidation of water is the only
changes of OH- concentration because all pathways have
first-order dependence in [OH-].
possibility. Therefore, we propose the formation of O2 (eq
3), a reaction that is thermodynamically very favorable in
basic solution. The oxidation of water would proceed via
Mechanism. Scheme 1 shows the proposed mechanism
with three concurrent pathways needed to account for the
kinetic and stoichiometric results. Pathway 1 generates a
species where OH- adducts to the Cl atom of ClO2 to form
-
peroxide intermediates, and it is known that HO2 reacts
- 22
rapidly with ClO2 to form O2 and ClO2 . To eliminate other
possible pathways, we also tested to see if any OCl- is
an (HOCl(O)O)- intermediate. This is similar to adducts
2- 25
(22) Hoigne´, J.; Bader, H. Water Res. 1994, 28, 45-55.
6502 Inorganic Chemistry, Vol. 41, No. 24, 2002
proposed for ClO2 reactions with HO2-,23 I-,24 and S2O3
.

Chlorine Dioxide Decomposition in Basic Solution
Figure 4. Equilibrium geometries (bond distances in Å) and atomic charges
for the steady-state species (a) (OClOOH)- and (b) (OClOH)-. The numbers
are reported for the B3LYP/6-311++G(3df,3pd) level of theory.
Figure 5. Effect of increasing [H2O2]T on the rate of the ClO2/HO2- reac-
tion in the p[H+] range 10.89 to 12.60. The reaction was followed at 359
nm, 25.0 °C, µ ) 1.0 M, and 0.050 mM ClO2. k7a ) 1.6(3) × 105 M-1 s-1
.
An ab initio calculation for the (HOCl(O)O)- adduct14 shows
a weak, but significant, interaction between Cl and the OH
group. The subsequent electron-transfer step (k2) is very rapid
and generates ClO2- and HOClO2 (which reacts rapidly with
ClO214 and Br2O426 intermediates. Ab initio calculations have
been reported previously16 for Cl2O4. The reaction of Cl2O4
with OH- in step k10 is another example of base-assisted
electron transfer, in this case between two weakly associated
ClO2 molecules.
-
OH- (k3) to give ClO3 ). Hence, the reaction is first order
in [ClO2] and [OH-] and forms equimolar ClO2- and ClO3 .
-
The reactions in the k1 and k2 steps can be considered as an
example of base-assisted electron transfer.14
Hydrogen Peroxide Reaction with ClO2. Hoigne´22 as-
signed a rate constant of 1.3(2) × 105 M-1 s-1 for the reaction
between HOO- and ClO2 in the pH range 5-11 (temperature
and other conditions were not specified). We measured the
rate by stopped-flow methods from p[H+] ) 9.4 to 12.6 by
observing the loss of ClO2 under pseudo-first-order condi-
tions where [H2O2]T . [ClO2]. Figure 5 shows the depen-
In pathway 2 we propose OH- forms an adduct to one of
the oxygen atoms of ClO2 to give (OClOOH)- (k4) as a
reactive intermediate. An ab initio calculation of this species
(Figure 4a) indicates a favorable intermediate where an OCl
segment with a net charge of -0.447 is weakly bound to an
OOH segment with net charge of -0.553. This adduct can
undergo rapid electron transfer with a second ClO2 (k5) to
give ClO2- and OClOOH. The latter species reacts favorably
with OH- to generate HOClO and HOO- (k6). The reaction
-
dence of k7,obsd ()k7a[HO2 ]) on [H2O2]T at several p[H+]
values. The rate expression is given in eq 9, where [H2O2]T
) [H2O2] + [HOO-] and k7a is the rate-determining step for
the reaction sequence given in eqs 10-12. Our value of 1.6-
(3) × 105 M-1 s-1 for k7a, at 25.0 °C and 1.0 M ionic strength,
between HOO- and ClO2 is known to give ClO2 and O2,
-
where k7 represents a series of rapid steps.22 The stoichiom-
where KaH O is 10-12.13 M,27 is in good agreement with
Hoigne´’s value. The reaction is independent of carbonate
and phosphate (up to 80 mM) buffer concentrations. The k7a
rate constant for the HOO- reaction is 9 orders of magnitude
greater than the k4 rate constant in pathway 2 and therefore
does not affect the reaction rate between ClO2 and OH-.
2
2
etry for the overall reaction in pathway 2 is given in eq 3,
-
where no ClO3 is formed. The reactions in the k4 and k5
steps can be considered as another example of a base-assisted
electron-transfer process. Alternative steps for the reaction
of (OClOOH)- that lead to the same products are given in
eqs 7 and 8. Ab initio calculations show that (OClOH)- is
a possible intermediate (Figure 4b).
2
2
-d[ClO2]
dt
2k7aKaH O
)
[H2O2]T[ClO2]
(9)
k5′
H2O2
+
(
)
(OClOOH)- + OH- fast8 (OClOH)- + HOO-
(7)
(8)
Ka
+ [H ]
k7a
9
HOO- + ClO2
8 (HOOCl(O)O)-
(10)
(11)
k6′
(OClOH)- + ClO2 8 OClOH + ClO2
-
fast
k7b
-
(HOOCl(O)O)- + ClO2 8 HOOCl(O)O + ClO2
fast
Pathway 3 is second-order in ClO2 and proceeds via a
Cl2O4 intermediate that is in rapid preequilibrium with two
ClO2 molecules (k9/k-9). This intermediate was proposed by
Halperin and Taube9 and is similar to the proposed BrO2-
k7c
HOOCl(O)O + OH- 8 H2O + ClO2- + O2 (12)
fast
(26) Buxton, G. V.; Dainton, F. R. S. Proc. R. Soc. A 1968, 304, 427-
439.
(27) Smith, R. M.; Martell, A. E. Critical Stability Constants. Volume 4:
Inorganic Complexes; Plenum Press: New York, 1979; p 75.
(23) Ni, Y.; Wang, X. Can. J. Chem. Eng. 1996, 75, 31-36.
(24) Fabian, I.; Gordon, G. Inorg. Chem. 1997, 36, 2494-2497.
(25) Horva´th, A. K.; Nagypa´l, I. J. Phys. Chem. A 1998, 102, 7267-7272.
Inorganic Chemistry, Vol. 41, No. 24, 2002 6503

Odeh et al.
-
Table 1. Summary of Rate Constants and Activation Parameters for
the Decomposition of ClO2 in Basic Solution
(pathways 1 and 3), where half of the ClO2 lost forms ClO2
-
and half forms ClO3 . The regression analysis gives R )
basic
nucleophile
∆Hq,
∆Sq,
0.60(3) and â ) 0.40(2). The solid line in Figure 3 shows
the fit of the experimental data to eq 15 (after multiplying
by 100/[ClO2]0) and corresponds to k1 ) 4.1(4) × 10-4 M-1
s-1 and k4 ) 1.4(1) × 10-4 M-1 s-1 (Table 1). The dashed
line shows that pathways 2 and 3 alone are not sufficient to
rate consta
kJ mol-1 J mol-1 K-1
OH-
ka ) 1.38(8) × 10-3 M-1 s-1
60(2)
32(2)
-101(6)
-119(7)
kb ) 21.8(4) M-2 s-1
k1 ) 4.1(1) × 10-4 M-1 s-1
k4 ) 1.4(1) × 10-4 M-1 s-1
(k9/k-9)k10 ) 10.9(2) M-2 s-1
-
account for the observed [ClO2 ]F/[ClO2]0 ratio (eq 6, where
ka corresponds to pathway 2 only and kb corresponds to
2-
CO3
CO3
ka ) 8.6(4) × 10-4 M-1 s-1
CO3
kb ) 6.7(2) M-2 s-1
3-
PO4
ka ) 2.8(6) × 10-5 M-1 s-1
pathway 3).
PO4
PO4
kb ) 24(2) M-2 s-1
-
a Conditions: 25.0 °C; µ ) 1.0 M NaClO4; λ ) 359 nm.
[ClO2 ]F )
n
0.5(Rka + kb([ClO2]20m + [ClO2]20m-20)/2) + âka
Detailed Rate Expressions and Resolved Rate Con-
stants. As the mechanism in Scheme 1 indicates, two ClO2
radicals are consumed in pathways 1 and 3 while four are
consumed in pathway 2. The rate expression in eq 13
×
∑
(
)
ka + kb([ClO2]20m + [ClO2]20m-20)/2
m)1
([ClO2]20m-20 - [ClO2]20m) (15)
-
-d[ClO2]
dt
k1k2[ClO2]
[ClO3 ]F )
) 2[OH-][ClO2]
+
(
k-1 + k2[ClO2]
n
0.5(Rka + kb([ClO2]20m + [ClO2]20m-20)/2)
×
∑
2k4k5[ClO2]
k9k10[ClO2]
(
)
ka + kb([ClO2]20m + [ClO2]20m-20)/2
m)1
+
(13)
-
)
k-4 + k5[ClO2]
k-9 + k10[OH ]
([ClO2]20m-20 - [ClO2]20m) (16)
-d[ClO2]
dt
) 2(k1 + 2k4)[OH-][ClO2] +
Temperature Dependence. The decomposition reactions
of 0.10 mM ClO2 in 0.20 M NaOH at µ ) 1.0 M were
measured from 0.0 to 25.0 °C to resolve activation param-
eters for the composite rate constants, ka and kb (eq 6). The
∆Hq values, 60(2) kJ mol-1 for ka and 32(2) kJ mol-1 for kb,
are in substantial agreement with previous values.10 The ∆Sq
values are -101(6) J mol-1 K-1 for ka and -119(7) J mol-1
K-1 for kb. The large negative ∆Sq values for ka ()2k1 +
4k4) and for kb ()2k9k10/k-9) are consistent with the need to
bring together several species in these composite rate
constants. Eyring plots are given in Figure S3 (Supporting
Information).
2k9k10
[OH-][ClO2]2 (14)
k-9
accounts for these stoichiometric factors and is derived on
the basis of the assumption that (HOClO2)-, (OClOOH)-,
and Cl2O4 are each steady-state intermediates. If k-1 , k2-
[ClO2], k-4 , k5[ClO2], and k-9 . k10[OH-], the rate
expression is simplified to eq 14 that agrees with our
experimental observations. The agreement between the
experimental data and the fit down to micromolar levels of
ClO2 (Figure 1, insert) suggests that values for k2/k-1 and
k5/k-4 are larger than 106 M-1. As seen from eqs 6 and 14,
the first-order rate constant (ka) obtained from kinetic data
equals 2(k1 + 2k4) and the second-order rate constant (kb)
Buffer Effects. Basic carbonate and phosphate buffers
accelerate the rate of ClO2 decay, and the corresponding rate
expressions are given in eq 17, where B ) CO32- or PO4
3-
.
-
-
equals 2k9k10/k-9. Yields of ClO2 and ClO3 as a function
of [ClO2] are needed to resolve the k1 and k4 values. A total
of 15 sets of reactions with [ClO2]0 varying from 0.074 to
-d[ClO2]
) kaB[B][ClO2] + kb [B][ClO2]2
(17)
B
dt
We found no contribution from HCO3- or HPO4- to the rate
of ClO2 decay. Figure 6 shows the significant contribution
-
1.58 mM in 0.20 M [OH-] were analyzed for ClO2 and
-
ClO3 after 99.9% completion. Figure 3 shows the percent
of CO32- to both the first-order path (kaCO ) 8.6(4) × 10
-
-4
3
[ClO2 ]F/[ClO2]0 as a function of the initial ClO2 concentra-
CO3
tion. To resolve k1 and k4 values, kinetic traces were
generated on the basis of eq 6 for the conditions in Figure
3. These traces were divided into 20 s intervals, and the
amounts of ClO2- and ClO3- formed at the midpoint of each
interval were calculated by successive approximation for
3000 data points. The summation of ClO2 formed at each
time interval to give [ClO2 ]F is expressed by eq 15, where
R and â are the relative contributions of pathways 1 and 2,
respectively, m is the interval number, and n is the total
number of intervals. Similarly, the [ClO3 ]F is given by eq
M-1 s-1) and the second-order path (kb ) 6.7(2) M-2 s-1),
where ka,obsd ) kaB[B] and kb,obsd ) kb [B]. By contrast,
B
B
B
previous evaluation10 gave kaCO ) 7 × 10-14 M-1 s-1 (a
3
CO3
value too low to be measured) and kb
) 2.45 M-2 s-1.
The rate constants for PO4 (Figure S4) are kaPO ) 2.8(6)
3-
4
× 10-5 M-1 s-1 and kb ) 24(2) M-2 s-1. In these studies,
-
PO4
contributions from the OH- path were also observed with
[OH-] ) 6.3 × 10-3 M and were taken into account. Product
studies in both buffers (Table 2) show that the percent of
-
-
-
ClO2 formed increases as the concentration of ClO2
16, which needs only the R value. The value of 0.5 in these
equations corresponds to the disproportionation reactions
decreases. This suggests that pathways analogous to pathway
2 also occur with carbonate and phosphate forming adducts
6504 Inorganic Chemistry, Vol. 41, No. 24, 2002

Chlorine Dioxide Decomposition in Basic Solution
Table 3. Comparison of the Rate Constants for Cl2O4, ClO2BrO2, and
Br2O4 Reactions with OH-
reacn
K9k10, M-2 s-1 a
2ClO2 + OH- f ClO2- + HOClO2
ClO2 + BrO2 + OH- f BrO2- + HOClO2
2BrO2 + OH- f BrO2- + HOBrO2
10.9b
1.25 × 108 c
1.3 × 1013 d,e
.
a K9 ) k9/k-9 b This work. c Reference 14. d Reference 26. e Reference
28.
Figure 7. Equilibrium geometries (bond distances in Å) and atomic charges
(italics) for OBrOBr(O)O. The numbers are reported for the B3LYP/6-
311++G(3df,3pd) level of theory.
favorable thermodynamically as the hydroxide ion concentra-
tion increases as shown by the redox potentials in eqs 18
and 19. The reactions of nucleophiles (Nu) with ClO2/BrO2
CO3
CO3
Figure 6. Dependence of (a) ka,obsd
and (b) kb,obsd
on [CO32-] for
ClO2 decomposition in basic solution. Conditions: 0.2 mM ClO2; 20-250
mM CO32-; p[H+] ) 9.4-10.3; 25.0 °C; µ ) 1.0 M. ka
) 8.6(4) ×
CO3
10-4 M-1 s-1, and kb
) 6.7(2) M-2 s-1
.
CO3
+
-
have been proposed14 to first generate NuClO2 and BrO2
(eq 20). These NuClO2+ species then react with OH- to give
Table 2. Products of ClO2 Decomposition in the Presence of (a)
-
Carbonate and (b) Phosphate Buffer
ClO3 and release Nu (eq 21). A similar process for Cl2O4
[ClO2]0, µM [ClO2-]F, µM [ClO3-]F, µM 100[ClO2-]F/[ClO2]0, %
would correspond to eq 22, which would be followed by a
step analogous to eq 21. However, the reaction in eq 22 is
thermodynamically less favorable than that in eq 20 because
(a) Decomposition of ClO2 in Carbonate Buffera-c
43.0(6)
69.6(6)
83.0(4)
124.2(7)
167.4(3)
217(1)
26.4(3)
39.6(3)
47.2(2)
68.8(2)
90.4(3)
115.4(2)
138.4(3)
16.6(5)
30.0(5)
35.8(4)
55.4(7)
77.00(8)
101(1)
61(1)
-
56.9(7)
56.9(4)
55.4(4)
54.0(2)
53.2(3)
54.0(2)
the reduction potential for BrO2/BrO2 is favored over that
of ClO2/ClO2- by 0.28 V. The intermediate NuClO2+ in eq
22 appears to be more difficult to form unless Nu is a strong
base.
256.4(5)
118.0(4)
(b) Decomposition of ClO2 in Phosphate Bufferb-d
2ClO2 + 2OH- f ClO2- + ClO3- + H2O EB° ) 1.449 V
50.6(7)
76.4(9)
90.0(6)
131.8(7)
164(1)
32.6(1)
46.2(2)
57.8(2)
75.6(5)
91(2)
18.0(6)
30.2(8)
32.2(5)
56.2(5)
72.6(6)
91.4(8)
113.4(9)
64.4(9)
60.5(7)
64.2(4)
57.4(5)
56(1)
(18)
ClO2 + BrO2 + 2OH- f BrO2- + ClO3- + H2O
EB° ) 1.726 V (19)
210(1)
248(1)
119.0(7)
135.0(6)
56.6(4)
54.3(4)
+
ClO2BrO2 + Nu f BrO2- + NuClO2
(20)
a [CO3]T ) 0.10 M; p[H+] ) 10.2(1); pKa(HCO3-) ) 9.48; 25.0 °C.
b The error represents the standard deviation of at least three runs for each
sample analyzed. c The subscript F in [ClO2-]F and [ClO3-]F refers to the
final levels of chlorite and chlorate formed upon the completion of the ClO2
decay reaction. d [PO4]T ) 0.10 M; p[H+] ) 11.8(1); pKa(HPO42-) ) 11.08;
25.0 °C.
NuClO2+ + 2OH- f ClO3- + Nu + H2O
(21)
(22)
Cl2O4 + Nu f ClO2- + NuClO2
+
with ClO2 (i.e. (OClOOCO2)2- and (OClOOPO3)3- species).
The ratios of ka/kb rate constants for these bases are 1.28 ×
10-4 M for CO32-, 6.33 × 10-5 M for OH-, and 1.1 × 10-6
M for PO43-. Thus, the first-order pathways are more
Third-order rate constants (M-2 s-1) for reactions of the
halogen dioxides with OH- (Table 3) increase by 12 orders
of magnitude for the following sequence: ClO2 + ClO2;16
ClO2 + BrO2;14 BrO2 + BrO2.26,28,29 Much of this enormous
change in reactivity can be attributed to the relative stabilities
of Cl2O4,16 ClO2BrO2,14 and Br2O4 (Figure 7, Table S6),
where the association constants of the halogen dioxide
increase by a factor of 3 × 105 from Cl2O4 to Br2O4 on the
basis of our ab initio calculations. The relative rate of OH-
reaction with the halogen dioxide dimers increases greatly
from Cl2O4 to Br2O4.
favorable for CO32- than for OH- and are less favorable for
3-
PO4
.
Comparison of the Reactions of ClO2/ClO2, ClO2/BrO2,
and BrO2/BrO2 in Base. The decomposition kinetics of ClO2
require basic nucleophiles (CO32-, PO43-, and OH-) for both
the first-order and the second-order pathways, whereas a
much broader group of nucleophiles (that include Cl-, Br-,
-
2-
2-
ClO2 , SO42-, CH3COO-, and HPO4 as well as CO3
,
PO43-, and OH-) assists the electron-transfer reactions
between ClO2 and BrO2.14 Both the ClO2/ClO2 and ClO2/
BrO2 reactions consume base and become much more
(28) Field, J. R.; Fo¨rsterling, H.-D. J. Phys. Chem. 1986, 90, 5400-5407.
(29) Nicoson, J. S.; Wang, L.; Becker, R. H.; Huff Hartz, K. E.; Muller,
C. E.; Margerum, D. W. Inorg. Chem. 2002, 41, 2975-2980.
Inorganic Chemistry, Vol. 41, No. 24, 2002 6505

Odeh et al.
Deviation from Marcus Theory. An alternative possibil-
responsible for the large deviation from the traditional
Marcus theory in the case of small molecules.31 We propose
that the large deviation from the Marcus theory is an
indication that the measured rate constants represent adduct
formation reactions rather than actual electron-transfer
processes. The mechanism proposed in Scheme 1 shows that
the adduct in k1 has to form first for the electron transfer in
k2 to take place. Pathways 2 and 3 also involve adduct
formation followed by electron transfer. In the case of HOO-,
a similar mechanism is shown in eqs 10 and 11 and is in
agreement with the proposed mechanism by Ni and Wang.23
Recent work in the Margerum group14 has shown that when
ClO2 reacts with BrO2, an adduct formation with a nucleo-
phile is needed first to allow an electron-transfer process to
occur.
ity for pathway 1 is a direct electron transfer between ClO2
-
and OH- to form ClO2 and OH. A subsequent reaction of
OH and ClO2 would give HClO3. Table S3 gives rate
constants determined for the reaction of ClO2 with OH- and
HO2- (kmeas) as well as the electron-transfer reactions of ClO2
- 30
- 31
2- 32
with NO2 , N3 , SO3
,
and SCN-33 on the basis of
Marcus theory.34 Self-exchange rate constants35 and redox
potentials36 have been compiled by Stanbury. The measured
rate constant for the ClO2/OH- reaction (kmeas ) k1) is 2 ×
104 times larger than expected by Marcus theory. Previous
studies have discussed inner-sphere mechanisms, orbital
overlap of reactants, and solvent nonadditivity as factors
(30) Stanbury, D. M.; Martinez, R.; Tseng, E.; Miller, C. E. Inorg. Chem.
1988, 27, 4277-4280.
(31) Awad, H. H.; Stanbury, D. M. J. Am. Chem. Soc. 1993, 115, 3636-
3642.
Acknowledgment. This work was supported by National
Science Foundation Grants CHE-9818214 and CHE-0139876.
(32) Suzuki, K.; Gordon, G. Inorg. Chem. 1978, 17, 3115-3118.
(33) Figler, J. N.; Stanbury, D. M. J. Phys. Chem. A 1999, 103, 5732-
5741.
(34) Espenson, J. H. Chemical Kinetics and Reaction Mechanisms, 2nd
ed.; McGraw-Hill: New York, 1995; pp 243-247.
(35) Stanbury, D. M. AdVances in Chemistry Series: Electron-Transfer
Reactions; American Chemical Society: Washington, DC, 1997; pp
165-182.
(36) Stanbury, D. M. AdVances in Inorganic Chemistry; Academic: New
York, 1989; Vol. 33, pp 69-138.
Supporting Information Available: Tables and figures with
supplemental kinetic and product data, the derivation of eq 5, and
the detailed method for use of eqs 15 and 16. This material is
available free of charge via the Internet at http://pubs.acs.org.
IC0204676
6506 Inorganic Chemistry, Vol. 41, No. 24, 2002