Reactions of Mn(II) and Mn(III) with alkyl, peroxyalkyl, and peroxyacyl radicals in water and acetic acid

Article abstract of DOI:10.1021/jp910140s

The kinetics of oxidation of Mn(II) with acylperoxyl and alkylperoxyl radicals were determined by laser flash photolysis utilizing a macrocyclic nickel complex as a kinetic probe. Radicals were generated photochemically from the appropriate ketones in the presence of molecular oxygen. In both acidic aqueous solutions and in 95% acetic acid, Mn(II) reacts with acylperoxyl radicals with k = (0.5-1.6) × 10⁶ M^-1 s^-1 and somewhat more slowly with alkylperoxyl radicals, k = (0.5-5) x 10 ⁵ M^-1 s^-1. Mn(III) rapidly oxidizes benzyl radicals, k = 2.3 × 10⁸ M^-1 s^-1 (glacial acetic acid) and 3.7 × 10⁸ M^-1 s^-1 (95% acetic acid). The value in 3.0 M aqueous perchloric acid is much smaller, 1× 10⁷ M^-1 s^-1. The decarbonylation of benzoyl radicals in H₂O has k = 1.2 × 10⁶ s ^-1.

Full text of DOI:10.1021/jp910140s

2136
J. Phys. Chem. A 2010, 114, 2136–2141
Reactions of Mn(II) and Mn(III) with Alkyl, Peroxyalkyl, and Peroxyacyl Radicals in Water
and Acetic Acid
Joo-Eun Jee and Andreja Bakac*
Ames Laboratory, Iowa State UniVersity, Ames, Iowa 50011
ReceiVed: October 23, 2009; ReVised Manuscript ReceiVed: December 20, 2009
The kinetics of oxidation of Mn(II) with acylperoxyl and alkylperoxyl radicals were determined by laser
flash photolysis utilizing a macrocyclic nickel complex as a kinetic probe. Radicals were generated
photochemically from the appropriate ketones in the presence of molecular oxygen. In both acidic aqueous
solutions and in 95% acetic acid, Mn(II) reacts with acylperoxyl radicals with k ) (0.5-1.6) × 106 M-1 s-1
and somewhat more slowly with alkylperoxyl radicals, k ) (0.5-5) × 105 M-1 s-1. Mn(III) rapidly oxidizes
benzyl radicals, k ) 2.3 × 108 M-1 s-1 (glacial acetic acid) and 3.7 × 108 M-1 s-1 (95% acetic acid). The
value in 3.0 M aqueous perchloric acid is much smaller, 1 × 107 M-1 s-1. The decarbonylation of benzoyl
radicals in H2O has k ) 1.2 × 106 s-1.
Introduction
ketone (K3) (both Aldrich), phenyl-2-propylphenyl ketone (K2,
Ryan Scientific), and dibenzyl ketone (K4, Acros) were used
as received. Stock solutions of Mnaq were prepared by
The interaction of transition metal complexes with free
radicals is one of the most widespread occurrences in biology,
environment, and laboratory.1-12 Among the common reactivity
modes are metal-element bond formation; oxidation and
reduction of the metal center; and metal-catalyzed dimerization,
disproportionation, fragmentation, or rearrangement of free or
coordinated radicals. Most prevalent are reactions with carbon13
and oxygen14 radicals owing to the central role that these two
elements hold in chemistry and biology, but sulfur, halogen,
nitrogen, and several other types of radicals have also been
exhaustively studied.1-4,7,15,16
Carbon and oxygen radicals are crucial intermediates in metal-
catalyzed oxidations of organic substrates with molecular oxygen
and peroxides. One such reaction is the industrially important
oxidation of para-xylene to terephthalic acid with O2, catalyzed
by cobalt and manganese salts and HBr in acetic acid. The
mechanistic picture is far from understood, but some key
features have been resolved.17-23 It is now widely accepted that
3+
dissolving manganese(III) acetate (Aldrich) in 3.0 M HClO4
containing 0.20 M Mn(II) perchlorate. High concentrations of
3+
acid and Mn(II) were necessary to stabilize Mnaq against
disproportionation. Solutions of Mn(III) acetate in acetic acid
did not require stabilization by Mn((II). Alkylcobaloximes were
prepared by literature procedures.24,25 Alkylperoxyl and acyl-
peroxyl radicals (1-3 µM) were generated in oxygen-saturated
aqueous or mixed acetic acid/water solvents (95:5, v/v) by laser
flash photolysis of the appropriate ketones K1-K4 (0.05-0.2
mM) or alkylcobalt precursors (0.05 mM), Chart 1. The
concentration of metal complexes in all of the kinetic experi-
ments was always much greater (g0.05 mM) than the concen-
tration of the radicals generated in the flash. When required,
the ionic strength was adjusted with NaClO4.
Laser-flash photolysis experiments utilized a laser system
based on a Lambda Physik Excimer Pro 2 xenon chloride
excimer laser and an Applied Photophysics monitoring system.
The 308-nm laser beam (pulse width 10 ns) was set up
perpendicular to the monitoring beam provided by a pulsed
xenon lamp. The monitoring beam passed through the cell and
grating monochromator to a five-stage photomultiplier tube. The
output signal was digitized with an Agilent Infinium digital
oscilloscope interfaced to an Acorn computer. The kinetic data
were fitted to standard kinetic equations with Kaleidagraph 4.03
software.
•
the reaction generates HBr2 , which abstracts hydrogen atoms
from C-H bonds to generate carbon radicals. In the oxygen
atmosphere of the Mid Century (MC) process, carbon radicals
are captured by oxygen and are converted to alkylperoxyl and
acylperoxyl radicals. Even though these radicals almost certainly
react with the cobalt and manganese catalysts, and are crucial
in determining the course, selectivity, and rates of the remainder
of the oxidation process, the kinetic data for such reactions are
scarce. This is in contrast to the wealth of information on the
corresponding reactions with well-defined macrocyclic com-
plexes, especially those of cobalt.13
•
Organic products of the PhCH2 /Mnaq3+ reaction were deter-
mined by GC-MS after extraction of spent reaction solutions
(3 mL) with 3 mL of CH2Cl2. The concentration of benzyl
alcohol was determined from GC peak areas and a calibration
curve which was constructed from the data that were also
obtained by extraction of 3 mL of known concentration of
PhCH2OH in 3 M HClO4 with 3 mL of CH2Cl2.
Here we report the results of our kinetic and mechanistic study
of the reactions of several oxygen and carbon radicals with
Mn(II) and Mn(III) acetates in aqueous and acetic acid solutions.
Experimental Section
Glacial acetic acid, manganese(II) acetate, manganese(III)
acetate (all Aldrich), phenyl-tert-butyl ketone (K1), di-tert-butyl
Results
2+
Reaction of Mnaq with Alkylperoxyl and Acylperoxyl
Radicals. The chemistry involved in the photochemical genera-
tion of alkylperoxyl radicals from ketones, taking K1 as an
* To whom correspondence should be addressed. E-mail: bakac@
ameslab.gov.
10.1021/jp910140s  2010 American Chemical Society
Published on Web 01/15/2010

Reactions of O and C Radicals with Mn(II) and Mn(III)
J. Phys. Chem. A, Vol. 114, No. 5, 2010 2137
CHART 1: Structures of the Ketones K1-K4 and an Organocobalt Complex Used As Radical Sources in This Work
SCHEME 1: Photochemistry of Phenyl-tert-butyl Ketone
L1Ni2+. The use of UV filters (absorbance >2 below 360 nm)
eliminated the bleach, leaving only the expected exponential
absorbance increase following the flash. The rate constants
obtained with and without the filters were the same, but the
signal was larger in the absence of optical filters. The rest of
the experiments were therefore carried out without the filters.
The rate constants for reaction 1 were determined from
experiments at several different concentrations of LNi2+, typi-
example, is summarized in Scheme 1. The short-lived nπ*
excited state dissociates to benzoyl and tert-butyl radicals with
cally in the range (0.5-8.0) × 10-4 M. Under these conditions,
and taking PhC(O)OO• as an example, the reaction followed
a first-order rate constant of ca. 107 s-1 26
.
Both carbon radicals
the rate law of eq 2, where k0 represents the sum of all of the
first and pseudofirst order rate constants for the loss of radicals
in reactions not involving L1Ni2+, and kNi is the desired second-
order rate constant for the L1Ni2+/radical reaction. The slope
of the linear plot of kobs against [L1Ni2+] for the reaction with
PhC(O)OO• in aqueous solutions, Figure 2, yielded the rate
are rapidly (k g 3 × 109 M-1 s-1 27
) intercepted by molecular
oxygen and converted to peroxyl radicals, PhC(O)OO• and
(CH3)3COO•. Except for partial decarbonylation of PhCH2C(O)•,
which loses CO in several microseconds in both nonaqueous28,29
and aqueous (see below) solutions, none of the acyl radicals in
this work undergo measurable decarbonylation on the time scale
of the radical/O2 reaction.28,29 Potential partial decarbonylation
of PhCH2C(O)• would diminish the yield of the desired
acylperoxyl radicals, but would not interfere with the measure-
ment of the metal/PhCH2C(O)OO• kinetics since the resulting
alkylperoxyl radicals react with metal complexes much more
slowly, see below.
constant kNi ) (1.0 ( 0.1) × 109 M-1 s-1
.
-d[PhC(O)OO·]/dt ) (k0 + kNi[L1Ni2+])[PhC(O)OO·]
(2)
2+
In the next series of experiments, both Mnaq and LNi2+
Other radicals were generated in similar reactions, as sum-
marized in Table 1. In some cases, several different sources
were used to generate the same radical to check the reliability
of the method and to confirm that no other chemistry interfered
under the experimental conditions. Owing to the strong absor-
bance of the ketones K1-K3 and the organocobalt complexes
in the UV, it was not feasible to directly monitor the growth of
Mn(III), which absorbs strongly only below 300 nm. Instead,
L1Ni2+ (L1 ) 1,4,8,11-tetraazacyclotetradecane) was used as a
radical probe, and the formation of the Ni(III) product in reaction
1 was monitored at 370 nm.
were present and competing for the radical. The concentration
2+
of Mnaq was varied as much as was experimentally feasible
while keeping [L1Ni2+] constant. Under such conditions, an
additional term appears in the rate law, eq 3:
-d[PhC(O)OO·]/dt ) (k0 + kNi[L1Ni2+] +
kMn[Mn2+])[PhC(O)OO·] (3)
A plot of kobs against [Mnaq2+] yielded a straight line that,
for the reaction with PhC(O)OO•, had a slope kMn ) (1.0 (
0.1) × 106 M-1 s-1, Figure 3.
kNi
L1Ni2+ + ROO· 98 L1NiIIIOOR2+
(1)
2+
In view of the large and variable concentrations of Mnaq
in kinetic runs, the effect of ionic strength on the kinetics was
explored in several experiments. For the reaction between
PhC(O)OO• (generated from K1) and Mnaq2+, the variation of
ionic strength in the range 0.10-0.50 M had a minimal effect
(<10%) on the rate constant, as expected for a reaction with an
uncharged reactant. Subsequent experiments were run at variable
ionic strength determined by the concentration of H+ and
reactants.
Experimental observations in 95% acetic acid were similar
to those in 0.10 M aqueous HClO4, except that the intercepts
in the plots of kobs versus [L1Ni2+] for acylperoxyl radicals were
significantly larger than for aqueous solutions, Figure S1. Under
such conditions, an additional pathway, perhaps oxidation of
As shown in the examples of the kinetic traces in Figure
1, there was an initial bleach followed by exponential
absorbance increase. The bleach was unexpected because kinetic
solutions did not absorb measurably at 370 nm prior to laser-flash
experiments. Additional tests showed that small amounts of a new
radical source, presumably L1NiOO(O)CR2+ and L1NiOOR2+,
were generated photochemically prior to the laser shot. Most
likely, the photolysis of ketones with the analyzing beam during
the short time (about 100 ms) between the opening of the shutter
and the flash was responsible for initiating (prematurely) the
photochemistry of Scheme 1 and capture of the radicals by

2138 J. Phys. Chem. A, Vol. 114, No. 5, 2010
Jee and Bakac
solvent
b
TABLE 1: Summary of Rate Constants for Reactionsa of Peroxyl Radicals with L1Ni2+ and Mn(II)
c
c
radical source
radical
kNi (108 M-1 s-1
)
kMn (106 M-1 s-1
)
K1 and K2
PhC(O)OO•
tBuC(O)OO•
CH3C(O)OO•
PhCH2C(O)OO•
tBuC(O)OO•
PhC(O)OO•
CH3OO•
10
10
9.7
9.1
2.2
2.1
0.40
0.27
0.37
0.14
0.13
0.025
1.0
1.1
water, pH 1.0
water, pH 1.0
water, pH 1.0
water, pH 1.0
95% AcOH
95% AcOH
95% AcOH
water, pH 1.0
95% AcOH
95% AcOH
95% AcOH
water, pH 1.0
K3
b,d
2+
L1CoC(O)CH3
K4
K3
K1
0.55
1.5
1.6
(dmgH)2(H2O)CoCH3
b,d
2+
L1CoCH3
CH3OO•
(dmgH)2(H2O)CoCH2Ph
K3
K2
K1 and K3
PhCH2OO•
tBuOO•
0.45
0.18
0.049
PhCMe2OO•
tBuOO•
a At 25 °C, λirr ) 308 nm, λdet ) 370 nm, O2-saturated. b L1 ) 1,4,8,11-tetraazacyclotetradecane. c Estimated standard deviation is 10%.
d Reference 31.
Figure 1. Kinetic traces for the reaction of benzoylperoxyl radical from K1 with L1Ni2+ (left) and with Mn(OAc)2 (right). Conditions: [K1] ) 1.6
× 10-4 M, 95% acetic acid, O2-saturated, λirr ) 308 nm, λdet ) 370 nm. Experiment in the right-hand panel had 0.16 mM L1Ni2+ as a kinetic probe.
Figure 2. Plot of kobs against the concentration of L1Ni2+ for the
2+
Figure 3. Plot of kobs against the concentration of Mnaq for the
reaction with PhC(O)OO• generated by laser flash photolysis of K1
(open circles) and K2 (filled circles) in aqueous 0.10 M HClO4.
reaction with PhC(O)OO• generated by laser flash photolysis of K1
(open circles) and K2 (filled circles) in the presence of 0.080 mM
L1Ni2+ as a radical probe in aqueous 0.10 M HClO4.
the solvent, appears to contribute to the disappearance of
RC(O)OO• radicals. The slopes yielded kNi ) (2.1 ( 0.2) ×
108 M-1 s-1 (Figure S1) and kMn ) (1.6 ( 0.2) × 106 M-1 s-1
(Figure S2). Several other radicals were also examined in both
solvents. All the kinetic data are summarized in Table 1.
As expected on the basis of Scheme 1, two successive
reactions were observed for each metal complex with each
ketone precursor. The faster step, described above, was assigned
to the reactions with acylperoxyl radicals, and the slower to
alkylperoxyl radicals, consistent with the greater reduction
potential30 of RC(O)OO• and also with our earlier observations
typically 1-3 mM L1Ni2+ and 0.02-0.10 M Mn(II). Under
these conditions, the reactions of acylperoxyl radicals were
complete in the first few microseconds, and the slower observed
reactions were those of alkylperoxyl radicals.
Decarbonylation of PhCH2CO•. The rate constant for this
reaction was determined by observing the growth of benzyl
radicals at 317 nm (ꢀ ) 5500 M-1 cm-1 32
upon laser flash
)
photolysis of the ketone K4. At the monitoring wavelength of
317 nm, the signal-to-noise ratio was less than optimal, and
the precision of the data in our estimate was no better than about
(30%. The more intense maximum at 258 nm (ꢀ ) 14000 M-1
cm-1) was not useful because of the interfering background
absorbance by K4.
in the reactions of CH3OO• and CH3C(O)OO• with L1Ni2+ 31
. In
every case, the two reactions were sufficiently separated in time
to allow the use of simple first-order kinetic treatment for each
step. The rate constants for the reactions of alkylperoxyl radicals
required the use of higher concentrations of metal complexes,
The sudden absorbance increase during the flash, correspond-
•
ing to the instantaneous formation of PhCH2 in the photochemi-

Reactions of O and C Radicals with Mn(II) and Mn(III)
J. Phys. Chem. A, Vol. 114, No. 5, 2010 2139
cal step of eq 4 was followed by an exponential growth in
absorbance that we associate with reaction 5, in agreement with
previous studies in nonaqueous solutions.28,29,33 At longer times,
the bimolecular self-decay of benzyl radicals caused the
absorbance to decrease again.
PhCH2COCH2Ph
9
hν8 PhCH2CO· + PhCH2·
(4)
(5)
PhCH2CO· f PhCH2· + CO
Figure 4. Kinetic traces at 317 nm for the self-reaction of 9.0 µM
PhCH2 radicals (left) and oxidation of PhCH2 with 9.6 × 10-5
M
•
•
Mn(III) (right) in glacial acetic acid. Conditions: [K4] ) 0.16 mM, 25
°C, λexc ) 308 nm, λdet ) 317 nm.
The experiments were carried out in three solvent systems
and yielded the following rate constants for reaction 5: 1.2 ×
106 s-1 (H2O), 2.0 × 106 (3.0 M aqueous HClO4), and 2.3 ×
106 (glacial acetic acid). The latter value is in good agreement
with the reported 2.0 × 106 s-1 in this solvent.34
Similar experiments in glacial acetic acid yielded traces such
as those illustrated in Figure 4 from which we obtained kself
)
(4.3 ( 0.3) × 109 M-1 s-1. The plot of kobs against [Mn(III)],
Oxidation of Benzyl Radicals with Mn(III). The kinetic
data were obtained by laser flash photolysis of K4 under air-
free conditions (argon) in the presence of Mnaq3+ by monitoring
the disappearance of benzyl radicals at 317 nm. The signal-to-
noise ratio was again small for reasons mentioned above, so
that the precision on these rate constants is also no better than
(30%.
shown in Figure S6, yielded kMn ) (2.3 ( 0.3) × 108 M-1 s-1
.
In 95% acetic acid, the values are kself (4.1 ( 0.3) × 109 M-1
s-1 and kMn (3.7 ( 0.4) × 108 M-1 s-1
.
Organic products were analyzed by GC-MS and NMR. For
this purpose, benzyl radicals were generated by steady-state
photolysis of K4 with the 313 nm light. In aqueous solutions
(3.0 M HClO4) in the absence of Mn(III), bibenzyl was the
exclusive product. When the radicals were generated in the
presence of 2 mM Mn(III), benzyl alcohol and benzaldehyde
were produced, as shown in Figure S7. No bibenzyl was found,
The spectrum of Mnaq3+ in 3.0 M HClO4 is shown in Figure
S3. The high [H+] and Mnaq2+ in these solutions were necessary
3+
to stabilize Mnaq against disproportionation, see the Experi-
mental Section. The use of L1Ni2+ as a kinetic probe in reactions
•
3+
as expected from the large rate constants for the Mn(III)/PhCH2
of Mnaq with radicals is ruled out by the rapid oxidation of
L1Ni2+ by Mnaq3+. Initial concentrations of the radicals were
calculated from the maximum absorbance at 317 nm, reached
after the chemistry in eqs 4-5 was completed (2-3 microsec-
onds after the flash).
reaction.
In 95% acetic acid, the results were similar to those in
aqueous solution, but somewhat unexpectedly, small amounts
of bibenzyl were observed even in the presence of Mn(III),
Figure S8. This result can be rationalized by the greater steady
state concentrations of PhCH2• in acetic acid, as judged by the
much stronger signals in this solvent in the kinetic experiments
and by the known inefficiency of radical formation by ketone
photolysis in water.
•
Under the reaction conditions ([PhCH2 ]0 ) 1-3 µM,
[Mnaq3+] ) 0.2-0.9 mM), radical dimerization, eq 6, competes
•
3+
with the PhCH2 /Mnaq reaction of eq 7.
2PhCH2· f PhCH2CH2Ph kself
(6)
Discussion
+
H2O, -H
PhCH2· + Mn3+
8 PhCH2OH + Mn
kMn
The data in Table 1 follow a clear pattern. Acylperoxyl
radicals react much faster than alkylperoxyl radicals with Mn(II)
and L1Ni2+ in both water and 95% acetic acid. This result is
consistent with the greater reduction potential of acylperoxyl
radicals.30 Of the two metal complexes, L1Ni2+ is oxidized more
rapidly, again in agreement with the reduction potentials,
2+
aq
aq
(7)
The data were fitted to an expression for mixed first and second
order kinetics, eq 8, where Abs0, Abst, and Absinf represent absorbances
at times zero, t, and after completion of the reaction; kψ is the pseudo-
first-order rate constant kMn[Mnaq3+] for reaction 7; and kself is the rate
constant for radical self-reaction. The value of the latter was determined
in the absence of added Mn3+, kself ) 2.0 × 109 M-1 s-1, in good
agreement with literature values which range from 1.2 × 109 to 3.2
× 109 M-1 s-1.13
E(L1Ni3+/2+) ) 1.0 V35 in acidic aqueous solutions, E(MnIII/II
)
∼ 1.5 V in aqueous solutions,36 and 1.77 V in acetic acid.37
The great reactivity of L1Ni2+ toward free radicals has been
already demonstrated in earlier studies.31,38-41
Overall, the solvent effect on the rate constants in Table 1 is
quite modest. This might be surprising in view of the chemical
differences between the species involved, and different polarity
of the two solvents. In aqueous solutions, manganese ions are
present as cationic aqua complexes, but in 95% acetic acid,
acetato and/or mixed acetato-aqua species are involved. On the
other hand, the reduction potential of Mn(III)/Mn(II) is only
moderately higher in acetic acid (1.77 V)37 than in water (∼1.5
V),36 which accounts for similar kinetics. Also, the lack of major
effect of solvent polarity agrees with a kinetic step that involves
a metal-radical bond formation, as opposed to an outer-sphere
electron transfer that would be facilitated in the more polar
solvent.
kψe-k t(Abs0 - Abs∞)
ψ
Abst ) Absinf
+
•
kψ + 2kself[PhCH2]0(1 - e-k t
)
ψ
(8)
Representative traces for benzyl radical self-reaction and the reaction
3+
with Mnaq are shown in Figure S4. A plot of kψ against the
concentration of Mnaq3+, Figure S5, yielded kMn ) (1.0 ( 0.3) × 107
M-1 s-1.

2140 J. Phys. Chem. A, Vol. 114, No. 5, 2010
Jee and Bakac
The rate constants obtained for the reaction of Mn(II) with
higher temperatures, where the solubility of O2 is significantly
reduced, Mn(III)-radical reactions may become competitive.
The kinetics of decarbonylation of benzoyl radicals have been
studied earlier by other groups in several solvents,28,29,33 but no
data were available for aqueous solutions. Our value in water,
1.2 × 106 s-1, is smaller than any of the values in other solvents.
Admittedly, the range is quite narrow, but there is a clear trend
of increasing rate constants with decreasing polarity of the
solvent.
alkylperoxyl radicals in 95% acetic acid, (2-4) × 105 M-1 s-1
,
are close to the (extrapolated) value of 3 × 105 M-1 s-1 that
was reported for the analogous reaction of cyanoisopropylper-
oxyl radicals, (CN)(CH3)2COO•, at 70 °C.22,42 That study of a
thermal reaction initiated by homolysis of azo-bis-isobutyroni-
trile (AIBN) in the presence of oxygen and manganese ions
employed spectrophotometric product analysis and chemilumi-
nescence to derive the rate constant. The agreement between
that work and ours lends credence to both approaches, one based
on small, steady-state concentrations of the radicals,22,42 and the
other, reported here, utilizing bursts of radicals generated by
laser flash photolysis and L1Ni2+ as a kinetic probe. The latter
method does offer an advantage in that it can generate a variety
of radicals from appropriate precursors for kinetic studies.
The proposed scheme22,42 for the ROO•/Mn(II) reaction is
shown in eqs 9 and 10.
Acknowledgment. We are thankful to Dr. Pestovsky for
useful comments and discussions. The support for this project
from BP Amoco is gratefully acknowledged. The research was
carried out in the facilities of the Ames Laboratory [under
contract No. DE-AC02-07CH11358 with the U.S. Department
of Energy-Basic Energy Sciences].
Supporting Information Available: Figures S1-S8. This
information is available free of charge via the Internet at http://
pubs.acs.org.
ROO· + Mn(II) a ROOMn(III)
(9)
+
References and Notes
H
ROOMn(III) 8 ROOH + Mn(III)
(10)
(1) Nitroxide Radicals and Nitroxide Based High-Spin Systems;
Landolt-Boernstein: Numerical Data and Functional Relationships in Science
and Technology-New Series, Group 2; Molecules and Radicals, Magnetics
Properties of Free Radicals, Subvolume D.; Alberti, A., Editor; 2005.
(2) N-Centered Radicals; Alfassi, Z. B. Ed.; 1998.
(3) S-Centered Radicals; Alfassi, Z. B. Ed.; 1999.
(4) Chatgilialoglu, C. Organosilanes in Radical Chemistry; 2004.
(5) Gansaeuer, E., Ed.; In Radicals in Synthesis I: Methods and
Mechanisms; Top. Curr. Chem. 2006, 263.
(6) Halliwell, B.; Gutteridge, J. M. C. Free Radicals Biol. Med. 1988.
(7) Ingold, K. U.; Roberts, B. P. Landolt-Boernstein Numerical Data
and Functional Relationships in Science and Technology, Group II: Atomic
and Molecular Physics, Vol. 13: Radical Reaction Rates in Liquids. Pt. C:
Radicals Centered in N, S, P and Other Heteroatoms. Nitroxyls; 1983.
(8) Free Radicals; Kochi, J. K. Ed.; 1973; Vol. 2.
(9) Free Radicals in Biology; Pryor, W. A. Ed.; 1980; Vol. 4.
(10) Togo, H. AdVanced Free Radical Reactions for Organic Synthesis;
2003.
In the previous work it was noticed,22,42 however, that the
kinetics of formation of Mn(III) and ROOH depended inversely
on the concentration of Mn(II) in the 0.1-1 mM range. The
reported value of 3 × 105 M-1 s-1 was obtained by extrapolation
to [Mn(II)] ) 0. The dependence on [Mn(II)] was explained
by an additional “deactivation” step, eq 11, whereby Mn(II)
catalyzes the dissociation of ROOMn(III) back to ROO• and
Mn(II). Our data show that the initial formation of ROOMn(III)
is first order in each reactant, which supports the proposal that
later steps, such as that in eq 11, are responsible for the inverse
dependence on [Mn(II)] in the earlier work.22,42
(11) Focused Issue. Oxygen/Nitrogen Radicals: Cell Injury and Disease.
(Proceedings of a Conference held in 2001.) In Mol. Cell. Biochem.
Vallyathan, V.; Castranova, V.; Shi, X. Eds.; 2002; Vol. 234/235, (1&2).
(12) Free Radicals in Chemistry, Biology and Medicine; Yoshikawa,
T., Toyokuni, S., Yamamoto, Y., Naito, Y. Eds.; 2000.
(13) Neta, P.; Grodkowski, J.; Ross, A. B. J. Phys. Chem. Ref. Data
1996, 25, 709–1050.
(14) Alfassi, Z. B. Peroxyl Radicals, Wiley: Chichester, 1997.
(15) Neta, P.; Huie, R. E.; Ross, A. B. J. Phys. Chem. Ref. Data 1988,
17, 1027–1284.
(16) Symons, M. C. Phil. Trans. Royal Soc. London, Series B, Biological
sciences 1985, 311, 451–472.
(17) Ishii, Y.; Sakaguchi, S.; Iwahama, T. AdV. Synth. Cat. 2001, 343,
393–427.
ROOMn(III) + Mn(II) f ROO· + 2Mn(II) (11)
The reaction of benzyl radicals with Mn(III) is fast in aqueous
solutions (1 × 107 M-1 s-1) and even faster in acetic acid where
the observed rate constant was 2.3 × 108 M-1 s-1 (glacial acetic
acid) or 3.7 × 108 M-1 s-1 (95% acetic acid). Our study43 of
the reaction between Mn(III) and bromide in acetic acid yielded
biexponential kinetics that led to the conclusion that commercial
Mn(III) acetate exists in at least two forms in an approximate
1:1 ratio. The same source of Mn(III) was used in the present
study, and we must assume that the manganese reactant is again
a mixture of two forms. The kinetic traces were exponential
because Mn(III) was used in large excess over the concentration
of the radicals but we cannot be certain whether only one or
both forms of Mn(III) reacted. If the former is true, then the
actual second-order rate constant for that form is up to a factor
of 2 greater than the values listed above, which are based on
the total (analytical) concentrations of Mn(III). We presume that
(18) Jones, G. H. J. Chem. Res. (S) 1981, 207.
(19) Metelski, P. D.; Espenson, J. H. J. Phys. Chem. A 2001, 105, 5881–
5884.
(20) Partenheimer, W. Catal. Today 1995, 23, 69–158.
(21) Partenheimer, W.; Gipe, R. K. ACS Symp. Ser 1993, 523, 81–88.
(22) Zakharov, I. V. Kinet. Cat. 1998, 39, 485–492.
(23) Zakharov, I. V.; Kumpan, Y. V. Kinetika i Kataliz 1993, 34, 1026–
1031.
(24) Schrauzer, G. N. Inorg. Synth. 1968, 11, 61–70.
(25) Yamazaki, N.; Hohokabe, Y. Bull. Chem. Soc. Jpn. 1971, 44, 63–
69.
(26) Turro, N. J. Modern Molecular Photochemistry; Benjamin/Cum-
mings Publishing Co.: Menlo Park, 1978; Ch 13.
(27) Marchaj, A.; Kelley, D. G.; Bakac, A.; Esenson, J. H. J. Phys. Chem.
1991, 95, 4440–4441.
(28) Fischer, H.; Paul, H. Acc. Chem. Res. 1987, 20, 200–206.
(29) Tsentalovich, Y. P.; Fischer, H. J. Chem. Soc., Perkin Trans. 1994,
2, 729–733.
3+
in aqueous solutions all of the Mn(III) is present as Mn(H2O)6
and that the observed rate constant applies to the reaction of
•
this ion with PhCH2 .
The large rate constant for the Mn(III)/PhCH2 reaction in
•
acetic acid suggests that the reactions of Mn(III) with carbon
radicals could be involved in the catalytic oxidations of
hydrocarbons if the concentrations of Mn(III) approach or
exceed those of molecular oxygen. The latter reacts with benzyl
and related radicals with diffusion controlled kinetics, but at
(30) Merenyi, G.; Lind, J.; Engman, L. J. Chem. Soc., Perkin Trans.
1994, 2, 2551–2553.
(31) Bakac, A. J. Am. Chem. Soc. 2002, 124, 9136–9144.
(32) Christensen, H. C.; Sehested, K.; Hart, E. J. J. Phys. Chem. 1973,
77, 983–987.

Reactions of O and C Radicals with Mn(II) and Mn(III)
J. Phys. Chem. A, Vol. 114, No. 5, 2010 2141
(33) Tsentalovich, Y. P.; Kurnysheva, O. A.; Gritsan, N. P. Russ. Chem.
Bull., Int. Ed. 2001, 50, 237–240.
(34) Kurnysheva, O. A.; Gritsan, N. P.; Tsentalovich, Y. P. Phys. Chem.
Chem. Phys. 2001, 3, 3677–3682.
(35) Haines, R. I.; McAuley, A. Coord. Chem. ReV. 1981, 39, 77–119.
(36) Bard, A. J.; Parsons, R.; Jordan, J. Standard Potentials in Aqueous
Solution; Marcel Dekker: New York and Basel, 1985; p 430.
(37) Jiao, X.-D.; Espenson, J. H. Inorg. Chem. 2000, 39, 1549–1554.
(38) Sauer, A.; Cohen, H.; Meyerstein, D. Inorg. Chem. 1988, 27, 4578–
4581.
(39) Kelley, D. G.; Marchaj, A.; Bakac, A.; Espenson, J. H. J. Am. Chem.
Soc. 1991, 113, 7583–7587.
(40) Kelly, C. A.; Mulazzani, Q. G.; Venturi, M.; Blinn, E. L.; Rodgers,
M. A. J. J. Am. Chem. Soc. 1995, 117, 4911–4919.
(41) Zilbermann, I.; Maimon, E.; Cohen, H.; Meyerstein, D. Chem. ReV.
2005, 105, 2609–2625.
(42) Adamyan, V. A.; Geletii, Y. V.; Popova, O. G.; Zakharov, I. V.
Kinet. Catal. 1991, 31, 1307–1310.
(43) Jee, J.-E.; Pestovsky, O.; Bakac A., manuscript in preparation.
JP910140S