Acid-catalyzed oxidation of iodide ions by superoxo complexes of rhodium and chromium

Article abstract of DOI:10.1021/ic049629j

Superoxometal complexes L(H₂O)MOO²⁺ (L = (H ₂O)₄, (NH₃)₄, or N ₄-macrocycle; M = Cr^III, Rh^III) react with iodide ions according to the stoichiometry L(H₂O)MOO²⁺ + 3l^- + 3H⁺ → L(H₂O)MOH²⁺ + 1.5l₂ + H₂O. The rate law is -d[L(H₂O)MOO ²⁺]/dt = k[L(H₂O)MOO²⁺][l^-][H ⁺], where k = 93.7 M^-2 s^-1 for Cr _aqOO²⁺, 4O2 for ([14]aneN₄)(H ₂O)CrOO²⁺, and 888 for (NH₃)₄(H ₂O)RhOO²⁺ in acidic aqueous solutions at 25 °C and 0.50 M ionic strength. The Cr_aqOO²⁺/l^- reaction exhibits an inverse solvent kinetic isotope effect, k_H2O/k _D2O = 0.5. In the proposed mechanism, the protonation of the superoxo complex precedes the reaction with iodide. The related Cr_aqOOH ²⁺/l^- reaction has k_H2O/k_D2O = 0.6. The oxidation of (NH₃)₅Rupy²⁺ by Cr _aqOO²⁺ exhibits an [H⁺]-dependent pathway, rate = (7.0 × 10⁴ + 1.78 × 10⁵[H ⁺])[Ru(NH₃)₅py²⁺][Cr _aqOO²⁺]. Diiodine radical anions, l₂ ^.-, reduce Cr_aqOO²⁺ with a rate constant k = 1.7 × 10⁹ M^-1 s^-1.

Full text of DOI:10.1021/ic049629j

Inorg. Chem. 2004, 43, 5416−5421
Acid-Catalyzed Oxidation of Iodide Ions by Superoxo Complexes of
Rhodium and Chromium
Andreja Bakac,* Chunkai Shi, and Oleg Pestovsky
Iowa State UniVersity, Ames, Iowa 50011
Received March 19, 2004
2+
III
III
Superoxometal complexes L(H O)MOO (L ) (H O)4, (NH ) , or N -macrocycle; M ) Cr , Rh ) react with iodide
2
2
2+
3 4
4
ions according to the stoichiometry L(H O)MOO + 3I- + 3H f L(H O)MOH + 1.5I2 + H O. The rate law is
+
2+
2
2
2
2+
2+
+
2+
−d[L(H O)MOO ]/dt ) k [L(H O)MOO ][I-][H ], where k ) 93.7 M-2 s-1 for CraqOO , 402 for ([14]aneN )(H O)-
2
2
4
2
2+
2+
CrOO , and 888 for (NH ) (H O)RhOO in acidic aqueous solutions at 25 °C and 0.50 M ionic strength. The
3 4
2
CraqOO /I- reaction exhibits an inverse solvent kinetic isotope effect, kH O/kD O ) 0.5. In the proposed mechanism,
2+
2
2
the protonation of the superoxo complex precedes the reaction with iodide. The related CraqOOH /I- reaction has
2+
kH O/kD O ) 0.6. The oxidation of (NH ) Rupy2+ by CraqOO exhibits an [H ]-dependent pathway, rate ) (7.0 × 104
2+
+
3 5
2
2
+ 1.78 × 105[H ])[Ru(NH ) py2+][CraqOO ]. Diiodine radical anions, I2•-, reduce CraqOO with a rate constant k )
+
2+
2+
3 5
1.7 × 109 M-1 s-1.
Introduction
substrates, iodide ions. The guiding question was one of
mechanism. There are no convincing examples of O atom
transfer in the chemistry of superoxometal ions except in
the reactions with •NO, where radical coupling precedes the
O-O bond cleavage,8,9 eq 2.
The acid-catalyzed oxidation of halides, phosphines, and
sulfides by hydroperoxo metal complexes in aqueous
solutions1-5 takes place by oxygen atom transfer, as dem-
onstrated in several examples by isotopic labeling.2,4 Unlike
the parent hydrogen peroxide, which exhibits both acid-
independent and acid-catalyzed pathways in the reactions
with these substrates,1,6,7 most of the metal hydroperoxides
react exclusively through the acid-catalyzed path, eq 1, where
L represents the equatorial ligand(s) (L ) (H2O)4, (NH3)4,
or N4-macrocycle), S is the substrate, and M is the metal
ion (CrIII, RhIII, or CoIII). The same is true for the anionic
complex (CN)5CoOOH3-.5
L(H2O)MOO2+ + •NO f [LMOONO2+] f
L(H2O)MO2+ + •NO2
LM ) Craq, Rh(N4) (2)
Direct O atom transfer to iodide, a nonradical species,
would thus be unprecedented. Electron transfer, however,
would initially generate iodine atoms in a thermodynamically
unfavorable reaction, eq 3. Surely, rapid follow-up steps, eqs
4 and 5, could drive the reaction to completion, but the uphill
thermodynamics of the first step should make the overall
reaction rather slow.
-d[L(H2O)MOOH2+]
) kS[L(H2O)MOOH2+][S][H+] (1)
dt
L(H2O)MOO2+ + I- f L(H2O)MOO+ + I•
L(H2O)MOO+ + H+ a L(H2O)MOOH2+
(3)
fast (4)
fast (5)
We have now turned to the reactions of the corresponding
superoxometal complexes, L(H2O)MOO2+, with one of these
* Author to whom correspondence should be addressed. E-mail: bakac@
ameslab.gov.
(1) Bakac, A. Prog. Inorg. Chem. 1995, 43, 267-351.
(2) Lemma, K.; Bakac, A. Inorg. Chem. 2004, 43, 4504-4510.
(3) Bakac, A.; Assink, B.; Espenson, J. H.; Wang, W.-D. Inorg. Chem.
1996, 35, 788-790.
(4) Pestovsky, O.; Bakac, A. J. Am. Chem. Soc. 2003, 125, 14714-14715.
(5) Mirza, S. A.; Bocquet, B.; Robyr, C.; Thomi, S.; Williams, A. F. Inorg.
Chem. 1996, 35, 1332-1337.
(6) Mohammad, A.; Liebhafsky, H. A. J. Am. Chem. Soc. 1934, 56, 1680-
1685.
•-
I• + I- a I2
Another option begins with a complex or L(H2O)MOO2+/
I- ion pair followed by a rate-determining attack by the
second iodide to produce I2•-, a species less energetic than
(8) Nemes, A.; Pestovsky, O.; Bakac, A. J. Am. Chem. Soc. 2002, 124,
421-427.
(9) Pestovsky, O.; Bakac, A. J. Am. Chem. Soc. 2002, 124, 1698-1703.
(7) Adzamli, I. K.; Deutsch, E. Inorg. Chem. 1980, 19, 1366-1373.
5416 Inorganic Chemistry, Vol. 43, No. 17, 2004
10.1021/ic049629j CCC: $27.50 © 2004 American Chemical Society
Published on Web 07/27/2004

Acid-Catalyzed Oxidation of Iodide Ions
I•. Such a mechanism, eq 6, adopted, for example, by the
reaction between Fe3+ and I-,10,11 would exhibit a second-
order dependence on iodide concentration.
L(H2O)MOO2+ + I- {-HHO2O} L(I)MOO+
8
-
I
H+, H2O
2
I2•- + L(H2O)MOOH2+ (6)
The results of a detailed kinetic and mechanistic study of
the oxidation of iodide with three L(H2O)MOO2+ complexes
are presented herein.
Figure 1. Plot of kobs vs the product [I-] × [H+] for the reaction of Craq-
OO2+ (0.010-0.035 mM) with iodide ions in 0.10 M HClO4.
Experimental Section
The superoxometal complexes L(H2O)MOO2+ were prepared as
described previously.8,9,12 Their concentrations were determined
spectrophotometrically2,8,9 immediately before the reaction with
iodide. Except in a few experiments that utilized an argon or
dioxygen atmosphere, most kinetic solutions contained 0.4 mM O2,
the “natural” concentration that resulted from mixing the O2-
saturated stock solutions of L(H2O)MOO2+, air-saturated solutions
of dilute HClO4 and NaClO4, and argon-saturated solutions of
iodide. The results were unaffected by the variations in the
concentration of O2.
Stock solutions of iodide (0.10 M) were prepared by dissolving
KI (certified A.C.S., Fisher) or NaI (99.999%, Aldrich) in H2O.
The iodide content was determined spectrophotometrically (λmax
) 226 nm, ꢀ ) 1.46 × 104 M-1 cm-1) after diluting to appropriate
concentration levels. Stock solutions were deaerated with a stream
of argon and kept in the dark.
concentration, 0.10 M H+, the measured rate constant was
9.44 ( 0.22 M-1 s-1.
The reaction is first order in [H+]. Figure 1 shows a plot
of kobs against the product [H+] × [I-] at 0.50 M ionic
strength. The line has a zero intercept and a slope kCr ) 93.7
( 1.0 M-2 s-1.
-
The yield of [I2 + I3 ] was calculated from the final
absorbance at 350 nm, using KI ) 830 M-1 (eq 7)15 and
-
ꢀ(I3 ) ) 26 000 M-1 cm-1. These data, combined with the
initial concentration of CraqOO2+, yielded the stoichiometric
-
ratio R ) [I2 + I3 ]∞/[CraqOO2+]0 ) 1.47 ( 0.05, consistent
with the reaction in eq 8 followed by rapid I2/I3- equilibration
of eq 7. This stoichiometry defines the rate constant kCr as
in eq 9.
Kinetics of the L(H2O)MOO2+/I- reactions were determined at
-
the 350- or 290-nm maximum in the absorption spectrum of I3
I2 + I- a I3
KI
(7)
-
by using a UV-vis Shimadzu 3101 PC spectrophotometer. The
kinetic study of the I2•-/CraqOO2+ reaction utilized an LKS 50
Applied Photophysics laser flash photolysis instrument.13 The
radicals were generated from I2/I3- in the presence of excess Craq-
OO2+ by a 355-nm flash from a Nd:YAG laser. The reaction was
CraqOO2+ + 3I- + 3H+ f CraqOH2+ + 1.5I2 + H2O (8)
-d[CraqOO2+]
dt
d[I2]
) 0.667
) kCr[CraqOO2+][I-][H+] (9)
dt
•-
monitored at 380 nm, where I2 exhibits a maximum, ꢀ ) 9.4 ×
103 M-1 cm-1 14 All of the experiments utilized 0.6 mM I-, 0.030
.
On occasion, the kinetic traces would come to an unusual,
abrupt stop, especially at high iodide concentrations (g2
mM), as shown in Figure 2a. Such behavior would persist
for several stock solutions of CraqOO2+, but eventually the
system would return to normal and yield traces such as that
shown in Figure 2b.
Prior to the break, the traces were strictly exponential. The
fit to first-order kinetics with floating Abs∞ yielded the same
rate constant as that obtained from well-behaved traces that
exhibited no break. The stoichiometry calculated from the
final measured absorbance remained the same as that in eq
8. Clearly, the stoichiometry calculated from the fitted Abs∞
was greater than R ) 1.5. In the worst case, the deviation
was about 10%, but more typically it was within 1-5%.
We looked for the potential role of several species that
are known or suspected to be present in our solutions, but
most had no effect; that is, they neither improved the
“broken” traces nor introduced a break for solutions that gave
well-behaved kinetics. In solutions containing O2, the fol-
lowing additives at concentrations up to 1 mM were without
mM I2, and 0.10 M HClO4. For the oxidation of Ru(NH3)5py2+ by
the superoxometal complexes, we used an Olis RSM 1000 stopped-
flow spectrophotometer.
The ionic strength was maintained in some experiments at the
0.50 M level with NaClO4 + HClO4. Laboratory-distilled water
was purified further by passage through a Millipore Milli-Q water
purification system. All of the kinetic experiments were performed
at 25.0 ( 0.2 °C.
Results
CraqOO2+/I- Reaction. In the presence of a large excess
of iodide, the build-up of I2/I3- followed exponential kinetics
in most of the experiments (see below) and yielded identical
rate constants at 350 and 290 nm. The reaction is first order
in iodide as shown by the linear plot of kobs against [I-]
(Figure S1, Supporting Information). At constant acid
(10) Fudge, A. J.; Sykes, K. W. J. Chem. Soc. 1952, 119-124.
(11) Amankwa, L.; Cantwell, F. F. Anal. Chem. 1989, 61, 2562-2566.
(12) Bakac, A. J. Am. Chem. Soc. 1997, 119, 10726-10731.
(13) Huston, P.; Espenson, J. H.; Bakac, A. J. Am. Chem. Soc. 1992, 114,
9510-9516.
(14) Devonshire, R.; Weiss, J. J. J. Phys. Chem. 1968, 72, 3815-3820.
(15) Eigen, M.; Kustin, K. J. Am. Chem. Soc. 1962, 84, 1355-1361.
Inorganic Chemistry, Vol. 43, No. 17, 2004 5417

Bakac et al.
Figure 2. Kinetic plots and fits to the first-order rate equation for the reaction of CraqOO2+ with iodide ions obtained for different stock solutions under
identical conditions: [I-] ) 1.0 mM, [H+] ) 0.10 M, [O2] ) 0.4 mM, and [CH3OH] ) 0.01 M; [CraqOO2+]0 ) 0.021 mM (a) and 0.020 mM (b).
Table 1. Summary of Kinetic Data for Acid-Catalyzed Reactions of
L(H2O)MOO2+ and L(H2O)MOOH2+ with Iodide Ions in Acidic
Aqueous Solutions at 25 °C
3+
effect: Cl-, Feaq (both potential impurities in solutions
2+
containing transition metals and perchlorate ions), Mnaq
(scavenger for CraqO2+), Cuaq2+ (scavenger for superoxide),
Znaq2+ (formed during Zn/Hg reduction of Craq3+), and CH2O
(produced from CraqO2+ and methanol during the preparation
of CraqOO2+). Also, the behavior of the system was
unchanged when methanol was replaced with ethanol as a
scavenger for CraqO2+ in the preparation of CraqOO2+.
The break in kinetic traces could be eliminated or at least
made much less prominent if the reactions were carried out
in the presence of large amounts of methanol (2 M) or
2-PrOH (1 M). Small amounts of chromate (e20 µM)
removed the breaks while keeping the rate constant and the
stoichiometry unchanged. Copper(II) ions had the same
effect, but only if the kinetic solutions were deaerated by
purging with argon prior to the reaction.
complex
k/M-2 s-1
source
CraqOO2+
93.7
185a
402
888
this work
this work
this work
this work
CraqOO2+
L1(H2O)CrOO2+ b
(NH3)4(H2O)RhOO2+
CraqOOH2+
988, 1120
1840a
530
536
8800
ref 3, this work
this work
ref 2
ref 2
ref 2
CraqOOH2+
L2(H2O)RhOOH2+ c
L1(H2O)RhOOH2+ b
(NH3)4(H2O)RhOOH2+
a In D2O, from data at 0.10 M DClO4. b L1 ) [14]aneN4. c L2 ) Me6-
[14]aneN4.
-
Craq3+, and traces of HCrO4 . Because the unusual traces
were observed only occasionally and gave the same kinetics
as the fully exponential ones, they cannot be attributed to
the main CraqOO2+/I- reaction and will be ignored in further
discussion. Moreover, the reactions of two other superoxo-
metal complexes, L1(H2O)CrOO2+ (L1 ) [14]aneN4) and
(NH3)4(H2O)RhOO2+, which always yielded clean exponen-
tial traces (see below) exhibited in their reactions with iodide
ions kinetic dependencies similar to those of CraqOO2+. This
finding strengthens our confidence in the results obtained
with CraqOO2+, despite an occasional unusual trace in this
reaction.
Superoxide ions appear to be responsible for the observed
behavior, although neither the chemistry leading to HO2 /
•
O2•-, nor the reason why HO2 /O2 itself would cause the
observed distortion of the kinetic traces, is obvious.
Perhaps the most convincing evidence for the involvement
of the superoxide is the effect of Cuaq2+, a powerful catalyst
for the dismutation of superoxide under air-free conditions
•
•-
where the two oxidation states, Cuaq2+ and Cuaq+, alternately
oxidize and reduce HO2 /O2
O2-containing solutions, where Cuaq+ generates both free and
bound superoxide in the reaction with O2.17 Consistent with
this picture, Cuaq2+ had no effect on the CraqOO2+/I- reaction
in air-saturated solutions but did clean up the kinetics in the
absence of O2.
Similarly, chromate reduces superoxide ions rapidly, which
may be the source of its effect on the CraqOO2+/I- reaction.
The alcohols, however, are not expected to react with HO2 /
•
.
•- 16 The situation is different in
Deuterium Isotope Effect. Two sets of experiments, one
in H2O and one in D2O (95-96% D content), were carried
out for the CraqOO2+/I- reaction. At 0.10 M H(D)ClO4, the
rate constants were 9.35 M-1 s-1 (H2O) and 18.5 (D2O),
yielding an inverse solvent kinetic isotope effect, kH/kD )
0.5. Similar experiments with the hydroperoxo complex, Craq-
OOH2+, in 0.10 M H(D)ClO4 yielded kH ) 112 and kD )
184 M-1 s-1, and thus kH/kD ) 0.6. The kH value agrees well
with the one calculated from our earlier data, 98.8 M-1 s-1.3
All of the kinetic data are summarized in Table 1.
•
O2•- under the conditions of our work but may be involved
at some stage preceding the formation of superoxide-
generating intermediates in the Craq2+/O2 reaction. As noted
earlier,1 this chemistry is complex, and the rates of oxygen
L1(H2O)CrOO2+/I- Reaction. The reaction stoichiometry
2+
is [I2 + I3 ]∞/[L1(H2O)CrOO2+]0 ) 1.42 ( 0.05, consistent
-
bubbling or Craq addition may change the proportion of
various intermediates before they are converted to CraqOO2+,
with the chemistry analogous to that shown for CraqOO2+ in
eqs 8 and 9. In the presence of a large excess of iodide and
H+, the kinetic traces were exponential and yielded values
of kobs that were linearly dependent on the concentration of
(16) Bielski, B. H. J.; Cabelli, D. E.; Arudi, R. L.; Ross, A. B. J. Phys.
Chem. Ref. Data 1985, 14, 1041-1100.
(17) Mi, L.; Zuberbuhler, A. D. HelV. Chim. Acta 1991, 74, 1679-1688.
5418 Inorganic Chemistry, Vol. 43, No. 17, 2004

Acid-Catalyzed Oxidation of Iodide Ions
Figure 3. Plot of kobs vs [CraqOO2+] for the reaction of CraqOO2+ with
I2•-. Conditions: [CraqOO2+]0 ) 0.05-0.16 mM, [I-] ) 0.6 mM, [I2] )
0.03 mM, and [H+] ) 0.10 M.
Figure 4. Plot of kobs vs [H+] for the reaction between CraqOO2+ and
Ru(NH3)5py2+
.
radical reactions,18 but an outer-sphere mechanism cannot
be entirely ruled out. Self-exchange rate data are not reliably
known for either partner19 in reaction 12, which prevents us
from making a Marcus estimate for the rate constant.
iodide and H+, Supporting Information, Figures S2 and S3.
At 0.50 M ionic strength, the third-order rate constant kLCr
,
defined in the same way as kCr in eq 9, is 402 ( 7 M-2 s-1.
(NH3)4(H2O)RhOO2+/I- Reaction. Solutions of (NH3)4-
(H2O)RhOO2+, prepared photochemically from (NH3)4(H2O)-
RhH2+ in oxygen-saturated solutions,12 contained large
amounts of (NH3)4(H2O)RhOOH2+ as a natural impurity.
Because both species react with iodide, the kinetic traces
were biexponential. The data analysis by consecutive kinetic
treatment yielded two sets of rate constants. The faster of
the two agreed with the independently known rate constant
for the (NH3)4(H2O)RhOOH2+/I- reaction, 8.80 × 103 M-2
s-1, eq 10, although the precision of the data was low. The
reason is the choice of concentrations in the present work,
where we sought to optimize conditions for the slower
(NH3)4(H2O)RhOO2+/I- reaction so that a large portion (up
to 80%) of the hydroperoxo stage was completed during the
manual mixing of the reagents. To obtain precise rate
constants for the (NH3)4(H2O)RhOO2+/I- reaction, we fixed
the rate constant for the faster stage in the calculations at its
known value.
hν, I
-
•-
I3
-8 2I2
(11)
(12)
+
H
I2•- + CraqOO2+ 8 I2 + CraqOOH2+
In support of the chemistry in eq 12, the final absorbance
at 380 nm was consistent with the formation of 1 equiv of
-
I2/I3 . Also, the absorbance of the reaction solutions grew
between successive laser shots, consistent with the additional,
-
slower formation of I2/I3 from CraqOOH2+ and I-. As
written, reaction 12 has a driving force of 0.8 V.
L(H2O)MOO2+/Ru(NH3)5py2+ Reactions. The first-order
dependence on [H+] in the reactions of L(H2O)MOO2+
complexes with iodide raised questions about timing and the
site of proton attack. Specifically, does a protonated form
of the superoxo complex L(H2O)MOOH3+ exist and play a
kinetic role? To help provide some answers and to find out
whether [H+] catalysis is a general feature of the reductions
of L(H2O)MOO2+, we examined the effects of H+ on the
reactions of two superoxometal complexes, CraqOO2+ and
L2(H2O)RhOO2+ (L2 ) Me6-[14]aneN4), with a typical outer-
sphere reductant, Ru(NH3)5py2+.
+
H , I
(NH3)4(H2O)RhOOH2+ + I- + H+
-8
(NH3)4(H2O)RhOH2+ + I2/I3 (10)
-
The (NH3)4(H2O)RhOO2+/I- reaction exhibited a first-
order dependence on [H+] and [I-], as shown in Supporting
Information, Figure S4. The value of the third-order rate
constant is kRh ) 888 ( 15 M-2 s-1.
At a constant acid concentration, mixed second-order
kinetics were observed for both reactions, as expected. As
shown in Figure 4, the dependence on [H+] for CraqOO2+ is
well described by eq 13, where k0 ) 7.0 × 104 M-1 s-1 and
kH ) 1.78 × 105 M-2 s-1 at 0.30 M ionic strength.
CraqOO2+/I2•- Reaction. There is a reasonable possibility
(see Discussion section) that the L(H2O)MOO2+/I- reactions
generate I• and I2•-, which might dimerize or react with Craq-
OO2+. To determine the likelihood of the latter reaction under
the conditions of the CraqOO2+/I- reaction, we determined
-d[Ru(NH3)5py2+]
)
dt
(k0 + kH[H+])[Ru(NH3)5py2+][CraqOO2+] (13)
•-
the kinetics of the CraqOO2+/I2 reaction by laser flash
photolysis.
The observation of a pronounced acid-dependent term lends
credence to our assertion that protonated superoxometal
complexes are involved, especially when the reduction of
L(H2O)MOO2+ to L(H2O)MOO+ is thermodynamically
unfavorable; see Discussion section.
Immediately after the flash, the absorbance at 380 nm
increased, owing to the formation of I2•-, and then decreased
exponentially with [CraqOO2+]-dependent kinetics, eqs 11 and
12 followed by eq 7. A plot of kobs against the average Craq-
OO2+ concentration, Figure 3, yielded k12 ) (1.7 ( 0.1) ×
109 M-1 s-1. The reaction probably takes place by initial
radical coupling, as observed earlier for other CraqOO2+/
(18) Bakac, A. AdV. Inorg. Chem. 2004, 55, 1-59.
(19) Stanbury, D. M. AdV. Chem. Ser. 1997, 253, 165-182.
Inorganic Chemistry, Vol. 43, No. 17, 2004 5419

Bakac et al.
On the other hand, the L2(H2O)RhOO2+/Ru(NH3)5py2+
reaction showed no dependence on [H+] (k ) (4.96 ( 0.13)
× 105 M-1 s-1 at 0.050 e [H+] e 0.30 M).
I- faster than CraqOO2+ does. The same is true for the other
two L(H2O)MOO2+/L(H2O)MOOH2+ couples. We were not
able to find a specific, selective reagent for I2 in the
•-
presence of the other reaction components.
Discussion
As an alternative to the one-electron path of eq 14, the
reaction might take place in a two-electron oxygen atom
transfer step, as shown in eqs 19 and 20. From an upper
limit for the CraqOO2+/CraqO2+ potential (<1.5 V in 1 M
HClO4)1, reduction potentials for the couples I•/I- (1.33 V)
and I2•-/2I- (1.03 V), and equilibrium constants for reactions
5 and 21, we estimate the driving force for reaction 19 in
1.0 M HClO4 to be e0.5 V. Even though this is only an
upper limit, it would seem that no reasonable correction of
the estimated CraqOO2+/CraqO2+ potential could make reac-
tion 19 thermodynamically less favorable than reaction 14,
for which ∆E ) - 0.33 V at 1.0 M H+.
One-electron oxidation of I- by CraqOO2+, as in eq 14, is
thermodynamically uphill by >0.3 V.1,20,21 Potential data for
the other L(H2O)MOO2+ complexes in this work are not
available, but they are not expected to differ dramatically
from those for CraqOO2+.
CraqOO2+ + I- + H+ f CraqOOH2+ + I•
(14)
If the reaction adopts the one-electron path, then the
unfavorable thermodynamics must be overcome in part by
•-
the rapid follow-up formation of I2 and its irreversible
oxidation with CraqOO2+, eqs 5 and 12.
The first-order dependence on [H+] clearly rules out an
outer-sphere oxidation of I- by CraqOO2+, for which ∆G
would be greater than that for reaction 14 by an amount
determined by the pKa of the product CraqOOH2+. This pKa
is probably around 7-8, similar to the value estimated for a
macrocyclic cobalt compound, L1(H2O)CoOOH2+.22 The
equilibrium constant for reaction 15 at 1 M H+ is thus only
∼10-12 Μ-1. Even if the reverse of reaction 15 proceeds at
a diffusion-controlled rate, the forward path could have a
rate constant of no more than 10-2 M-1 s-1, 4 orders of
magnitude less than that for the experimentally observed
reaction.
-
CraqOO2+ + I- + H+ f CraqO2+ + IOH ( I , H+8 I2) (19)
- +
I , H
CraqO2+ + I-
8 CraqOH2+ + I2
(20)
•-
I2 + H2O a IOH + I- + H+
K ) 2 × 10-13 M-2 (ref 23) (21)
Despite the thermodynamic considerations, there is no
experimental evidence for such a path. Tests for CraqO2+
proved to be inconclusive. Mnaq2+ and Ceaq3+, both of which
are rapidly oxidized by CraqO2+, had no effect on the reaction.
This result, however, does not rule out the involvement of
CraqO2+ because the oxidized forms of both scavengers react
rapidly with I- and would simply replace CraqO2+ in that
role.
L1(H2O)CrOO2+ is unique among the three superoxides
in this work. This ion gives rise to a hydroperoxide that
undergoes an intramolecular transformation to a strongly
oxidizing Cr(V) complex.24 Under carefully selected experi-
mental conditions (low concentrations of iodide and acid),
the one-electron reduction of L1(H2O)CrOO2+ with iodide
should initiate a rapid chain reaction, provided that L1Cr(V)
reacts with I- in one-electron steps. Under such circum-
stances, the limiting step in the overall reaction should be
the hydroperoxo-to-oxo conversion, which has k ) 0.19 s-1.25
No chain reaction was observed, suggesting that either the
L1(H2O)CrOO2+/I- reaction proceeds by a two-electron route
that bypasses the hydroperoxide or, more likely, that L1Cr-
CraqOO2+ + I- f CraqOO+ + I•
(15)
Most likely, a rapid acid-base equilibrium generates small
amounts of the reactive, protonated superoxometal complex,
which oxidizes iodide in the next step, eqs 16 and 17,
followed by the known chemistry of I•, I2•-, and CraqOOH2+
in eqs 5, 12, and 18.
CraqOO2+ + H+ a CraqOOH3+
(16)
(17)
CraqOOH3+ + I- f CraqOOH2+ + I•
-
I , H
CraqOOH2+ + I-
+8 CraqOH2+ + I2 + H2O (18)
The equilibrium constant for reaction 14 taking place by
any mechanism, including the sequence shown in eqs 16 and
17, is ∼10-5 M-1. Again, setting the limit for k at 1010 M-1
s-1 for the reverse step, the forward reaction must have k <
105 M-2 s-1. The experimental value, 93.7 M-2 s-1, satisfies
this requirement. Thus, the one-electron process of eq 14 or
its equivalent is consistent with the data.
A good test for the chemistry in eq 14 (or eqs 16 and 17)
would be the observation of either CraqOOH2+ or I2•- as an
intermediate. Unfortunately, measurable amounts of Craq-
OOH2+ did not and could not accumulate even if the reaction
took place as in eq 14 because this hydroperoxide reacts with
(V) reacts with iodide in a two-electron process that bypasses
•-
I•/I2
.
The experimental data for the reactions of I- with all three
superoxides are consistent with both electron transfer and O
atom transfer, as shown for CraqOO2+ in eqs 14 and 19,
respectively. At this stage, the one-electron path appears
preferable despite the less-favorable thermodynamics (avail-
able only for the aquachromium case). Our mechanistic
preference is based on the similarities of the rate constants,
(23) Cotton, F. A.; Wilkinson, G. AdVanced Inorganic Chemistry, 5th ed.;
Wiley-Interscience: New York, 1988.
(24) Bakac, A.; Wang, W.-D. J. Am. Chem. Soc. 1996, 118, 10325-10326.
(25) Bakac, A.; Wang, W.-D. Inorg. Chim. Acta 2000, 297, 27-35.
(20) Kang, C.; Anson, F. C. Inorg. Chem. 1994, 33, 2624-2630.
(21) Stanbury, D. M. AdV. Inorg. Chem. 1989, 33, 69-138.
(22) Kumar, K.; Endicott, J. F. Inorg. Chem. 1984, 23, 2447-2452.
5420 Inorganic Chemistry, Vol. 43, No. 17, 2004

Acid-Catalyzed Oxidation of Iodide Ions
which lie within a factor of <10 for the three superoxo
complexes. Such a small spread seems reasonable for one-
electron reactions yielding hydroperoxo complexes.24,25
Oxygen atom transfer, however, would initially produce
LMIVO2+ complexes. Although the thermodynamic data are
not available, we would expect reaction 19 to be more
favorable for the oxophilic chromium than for the rhodium,
opposite from the order of the rate constants in Table 1.
Admittedly, this analysis is complicated by the observed acid
catalysis. In addition to intrinsic reactivity, the differences
in protonation constants will also influence the observed
kinetic trends.
the greater value for K16. The magnitude of the effect (kH/kD
) 0.5) is similar to that observed in the CraqOOH2+/I-/H+
reaction (kH/kD ) 0.6), for which the prior protonation
mechanism is considered most likely.3 Also, the protonation
constant for a macrocyclic oxoruthenium(V) complex has
kH/kD ) 0.46 M-1.28 The similarity between these values
strongly supports the prior protonation mechanism for all
three L(H2O)MOO2+/I-/H+ reactions in this work.
Another argument in favor of such a scheme is the
observation of an acid-dependent path in the CraqOO2+/
(NH3)5Rupy2+ and CraqOO2+/hydrazine29 reactions. In both
cases, the involvement of an ion pair/complex is highly
unlikely. The operation of a parallel, acid-independent path
in the CraqOO2+/(NH3)5Rupy2+ reaction and the lack of any
acid dependence in the L2(H2O)MOO2+/(NH3)5Rupy2+ reac-
tion are consistent with the much greater reducing power of
(NH3)5Rupy2+ (E ) 0.305 V)30 compared to that of I- (1.33
V),21 which is clearly sufficient for the reaction to generate
even the hydrolytically less stable L(H2O)MOO+ ions with
the rate constants exceeding 104 M-1 s-1.
Regardless of whether the reaction takes place by electron
or atom transfer, the involvement of H+ helps the reaction
thermodynamically and kinetically. In our proposed mech-
anism, the superoxo complex accepts a proton prior to the
reaction with iodide, eqs 16 and 17, so that the observed
rate constant is the product K16k17. A different order of events,
for example, the formation of an ion pair or a complex,
followed by the reaction with H+, eq 22, appears less likely
in view of the observed solvent isotope effect.
Acknowledgment. We are grateful to Dr. Lemma for his
help with some of the experiments. This work was supported
by a grant from National Science Foundation, CHE 0240409.
Some of the work was conducted with the use of facilities
at the Ames Laboratory.
CraqOO2+ + I- f
{[CraqOO2+, I-] or ICraqOO+ } +8 products (22)
H
Supporting Information Available: Kinetic plots showing acid
and iodide dependence. This material is available free of charge
via the Internet at http://pubs.acs.org.
In general, the greater strength of the O-D bonds relative
to the O-H bonds causes the equilibrium protonation
constants to be greater in D2O.26,27 Translated to the present
case, the major reason for the increased reactivity in D2O is
IC049629J
(28) Che, C.-M.; Lau, K.; Lau, T.-C.; Poon, C.-K. J. Am. Chem. Soc. 1990,
112, 5176-5181.
(29) Bruhn, S. L.; Bakac, A.; Espenson, J. H. Inorg. Chem. 1986, 25, 535-
(26) Albery, W. J. Prog. React. Kinet. 1967, 4, 353-398.
(27) Albery, W. J. In Proton-Transfer Reactions; Caldin, E., Gold, V., Eds.;
Chapman & Hall: London, 1975.
538.
(30) Bard, A. J.; Parsons, R.; Jordan, J. Standard Potentials in Aqueous
Solution; Marcel Dekker: New York, 1985; p 418.
Inorganic Chemistry, Vol. 43, No. 17, 2004 5421