Kinetics of dissociative electron transfer to ascaridole and dihydroascaridole - Model bicyclic endoperoxides of biological relevance

Article abstract of DOI:10.1002/1521-3765(20010917)7:18<4012::AID-CHEM4012>3.0.CO;2-A

The homogeneous and heterogeneous electron transfer (ET) reduction of ascaridole (ASC) and dihydroascaridole (DASC), two bicyclic endoperoxides, chosen as convenient models of the bridged bicyclic endoperoxides found in biologically relevant systems, were studied in aprotic media by using electrochemical methods. ET is shown to follow a concerted dissociative mechanism that leads to the distonic radical anion, which is itself reduced in a second step by an overall two-electron process. The kinetics of homogeneous ET to these endoperoxides from an extensive series of radical anion electron donors were measured as a function of the driving force of electron transfer (ΔG°_ET). The kinetics of heterogeneous ET were also studied by convolution analysis. Together, the heterogeneous and homogeneous ET kinetic data provide the best example of the parabolic nature of the activation-driving force relationship for a concerted dissociative ET described by Saveant; the data is particularly illustrative due to the low bond-dissociation enthalpy (BDE) of the O-O bond and hence small intrinsic barriers. Analysis of the data allowed the dissociative reduction potentials (E°_diss) to be determined as -1.2 and -1.1 Vagainst SCE for ASC and DASC, respectively. Unusually low pre-exponential factors measured in temperature-dependent kinetic studies suggest that ET to these O-O bonded systems is nonadiabatic. Analysis of ET kinetics for ASC and DASC by the Saveant model with a modification for nonadiabaticity allowed the intrinsic free energy for ET to be determined. The use of this approach and estimates for the BDE provide approximations of the reorganization energies. We suggest the methodology described herein can be used to evaluate the extent of ET to other endoperoxides of biological relevance and to provide thermochemical data not otherwise available.

Full text of DOI:10.1002/1521-3765(20010917)7:18<4012::AID-CHEM4012>3.0.CO;2-A

FULL PAPER
Kinetics of Dissociative Electron Transfer to Ascaridole and
DihydroascaridoleÐModel Bicyclic Endoperoxides of Biological Relevance
Robert L. Donkers and Mark S. Workentin*[a]
Abstract: The homogeneous and hetero-
geneous electron transfer (ET) reduc-
tion of ascaridole (ASC) and dihydroas-
caridole (DASC), two bicyclic endoper-
oxides, chosen as convenient models of
the bridged bicyclic endoperoxides
found in biologically relevant systems,
were studied in aprotic media by using
electrochemical methods. ET is shown
to follow a concerted dissociative mech-
anism that leads to the distonic radical
anion, which is itself reduced in a second
step by an overall two-electron process.
The kinetics of homogeneous ET to
these endoperoxides from an extensive
series of radical anion electron donors
were measured as a function of the
driving force of electron transfer
(DGꢀET†. The kinetics of heterogeneous
ET were also studied by convolution
analysis. Together, the heterogeneous
and homogeneous ET kinetic data pro-
vide the best example of the parabolic
nature of the activation ± driving force
relationship for a concerted dissociative
respectively. Unusually low pre-expo-
nential factors measured in tempera-
ture-dependent kinetic studies suggest
that ET to these O O bonded systems is
nonadiabatic. Analysis of ET kinetics
Â
for ASC and DASC by the Saveant
Â
ET described by Saveant; the data is
model with a modification for nonadia-
baticity allowed the intrinsic free energy
for ET to be determined. The use of this
approach and estimates for the BDE
provide approximations of the reorgan-
ization energies. We suggest the meth-
odology described herein can be used to
evaluate the extent of ET to other
endoperoxides of biological relevance
and to provide thermochemical data not
otherwise available.
particularly illustrative due to the low
bond-dissociation enthalpy (BDE) of
the O O bond and hence small intrinsic
barriers. Analysis of the data allowed
the dissociative reduction potentials
(Eꢀdiss† to be determined as 1.2 and
1.1 Vagainst SCE for ASC and DASC,
Keywords: electrochemistry ´ elec-
tron transfer
´ endoperoxides ´
O O activation ´ radical ions
artemisinin (ART), and related semisynthetic derivatives.[4±21]
It has only recently been generally agreed that the mechanism
of antimalarial action involves an initial hemin Fe(ii)-medi-
ated electron transfer (ET), which leads to fragmentation of
the endoperoxide O O bond. This results in the formation of
oxygen-centered radical intermediates, which subsequently
lead to the toxic intermediates.[20, 22±29]
Introduction
Endoperoxides are compounds whose oxygen ± oxygen
(O O) bond often plays a key role in the activity of a number
of chemically and biologically relevant substances.[1, 2] Recent
reviews summarize the numerous marine and terrestrial
sources of endoperoxides that exhibit wide-ranging bioactiv-
ity, with many targeted as drug candidates.[1, 2] An important
example of an endoperoxide found in vivo is the prostaglan-
din endoperoxide (PGH2), a key intermediate in the biosyn-
thesis of fatty acids to prostaglandins, prostacyclins, throm-
boxanes and leukotrienes.[3] Another example that has been
well publicized is the potent antimalarial 1,2,4-trioxane,
[a] Prof. M. S. Workentin, Dr. R. L. Donkers
Department of Chemistry, University of Western Ontario
London, Ontario N6A 5B7 (Canada)
Fax : (‡1)519-661-3022
ET reduction of O O bonds in acyclic peroxides, a-peroxy
lactones, a-pyrone endoperoxides, and dioxetanes is also
recognized in connection with the chemically initiated elec-
tron-exchange luminescence (CIEEL) mechanism.[30±37] Al-
though a generally accepted mechanism is yet to be estab-
lished,[38±42] the ET to the O O bond in CIEEL is described
essentially as a concerted dissociative process. However, little
E-mail: mworkent@uwo.ca
Supporting information for this article is available on the WWW under
http://www.wiley-vch.de/home/chemistry/ or from the author. A com-
plete description of the temperature dependent rate studies, including
plots of the data and tables of the experimental and calculated
activation parameters.
4012
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0718-4012 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 18

4012±4020
Â
other insight into the fine detail of the ET mechanism has
been provided. In contrast, and with few exceptions,[1, 3, 43, 44]
ET is not generally considered in the mechanism of action of
endoperoxides; this is due, in part, to the lack of accurate
thermodynamic and kinetic information on these types of
systems. Proper consideration of the role of ET in the possible
modes of action requires knowledge of the driving force for
ET between the donor and the endoperoxide. Without
accurate values of the standard reduction potential of each,
the feasibility of ET cannot be determined accurately. This
represents the most difficult challenge with these molecular
systems.
Based on work by others[45, 46] and ourselves[47, 48] on the
reduction of acyclic peroxides, the ETreduction mechanism of
the O O bond in endoperoxides should be concerted
dissociative, in which bond fragmentation and electron uptake
occur in a single step. For endoperoxides,[49±51] ET and O O
cleavage lead to the concerted formation of an intermediate
that we refer to as a distonic radical anion [Eq. (1)], in which
charge and spin are separated.
Examining the kinetics of ET and utilizing Saveantꢁs model
can provide important thermochemical parameters. The
model was successfully applied to studying the dissociative
reduction of the C X bonds of alkyl and benzyl halides,[59±63]
N X,[64] C S,[65, 66] and S S[67, 68] bonds. Considerably fewer
studies have been done on the reduction of the O O bonds of
peroxides.[46±48, 51]In a preliminary communication we reported
the only known estimates of the standard dissociative
reduction potentials (Eꢀdiss† of any endoperoxide, specifically
two bicyclic endoperoxides, as-
caridole (ASC) and dihydroas-
caridole (DASC) using this
approach.[51] ASC is a natural
product with known activity as
an anthelmintic,[69] but it was
chosen, along with DASC, as a
suitable model system for the broader class of bioactive
endoperoxides, such as PGH2. Our analysis in the preliminary
report involved examining the kinetics of the homogeneous
electron-transfer reduction of these endoperoxides as a
function of the potential of the electron donor.
‡e
.
R-O-O-R
O-R-R-O
(1)
!
We now report the complete details of this kinetic study,
including an investigation of the temperature dependence of
the rate constants, and the determination of heterogeneous
rate constants from electrochemical data by using convolution
potential sweep voltammetry. Together, the heterogeneous
and homogeneous ET data provide the most comprehensive
study of the kinetics of ET to endoperoxides. Our results show
that ET to the O O bond is nonadiabatic. These kinetic and
mechanistic details have important ramifications to under-
standing ET to the O O bonds in endoperoxides in vivo.
More fundamentally, ASC and DASC also provide unique
and interesting cases for studying dissociative ET because,
unlike previous examples, the anion product in the dissocia-
tive reduction remains in the same molecule. The systems
described here provide an unprecedented example of the
parabolic activation ± driving force relationship observed for
dissociative ET systems, presumably because of the low O O
BDE and thus low intrinsic barrier.
In Equation (1) the electron (e ) source can either be an
electrode (heterogeneous) or a homogeneous reactant. Re-
duction potentials for concerted dissociative processes are not
easily determined by using simple heterogeneous electro-
chemical methods, since the direct reduction is subject to a
large over-potential due to the slow, rate-determining hetero-
geneous ET. When the standard reduction potentials cannot
be conveniently measured, they can often be estimated by
using thermochemical cycles.[52] However, unlike many acyclic
peroxides, the data required in the thermochemical cycles for
endoperoxides are not readily available from thermolysis
studies.[53±55] In order to determine these thermochemical
parameters that aid in understanding the mechanism of action
of endoperoxides, and to better understand the chemistry that
could be initiated by ET, we began a detailed investigation
into the kinetics of ET chemistry to endoperoxides using
electrochemical techniques.
The rate constant for ET, kET, is related to the reaction
activation free energy, DG=, by Equation (2), in which kel is
the transmission coefficient and Z is a pre-exponential term.
!
Results and Discussion
DG=
ket ˆ kelZ exp
(2)
RT
Cyclic voltammetry and coulometry: Cyclic voltammetry was
used to study the reduction of ASC and DASC in both
acetonitrile (MeCN) and N,N-dimethylformamide (DMF)
containing 0.1m tetraethylammonium perchlorate (TEAP) at
258C with a glassy carbon cathode. A representative cyclic
voltammogram of ASC measured under these conditions is
shown in Figure 1. The voltammograms of both ASC and
DASC are essentially identical and show a broad, irreversible
cathodic peak at all the potential scan rates investigated (0.1
Â
Over the last decade, Saveant and co-workers have
provided a framework that allows the kinetics of ET to be
predicted for concerted dissociative reductions.[56±58] Saveantꢁs
Â
model relates the activation free energy, DG=, to the reaction
free energy, DG8, by an equation similar to the well-known
Marcus equation [Eq. (3)], except that the intrinsic barrier,
DG=0 , contains contributions from the bond dissociation
enthalpy (BDE) of the fragmenting bond in addition to the
reorganization energy, l [Eq. (4)].
1
1
to 50 Vs ). At 0.1 Vs , the peak potentials (Ep) of the
reduction of ASC and DASC are 1.85 and 1.90 V against
SCE in MeCN, respectively, and 1.88 and 1.93 V against
SCE in DMF. On glassy carbon the voltammograms are
reproducible, and the position and shape of the reduction
waves do not change in the presence of non-nucleophilic acids
ꢀ
ꢁ
DGꢀ
4DG=0
2
DG=ˆ DG=0 1 ‡
(3)
(4)
ꢀ
ꢁ
DGꢀ
2
l ‡ BDE
DG=ˆ
1 ‡
4
l ‡ BDE
Chem. Eur. J. 2001, 7, No. 18
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0718-4013 $ 17.50+.50/0
4013

M. S. Workentin and R. L. Donkers
FULL PAPER
behavior further indicates that the ET reduction of these
endoperoxides is governed by the kinetics of the ET process
and, thus, cannot be described by Butler± Volmer kinetics.
Since the ET is rate determining, the observed Ep is more
negative than the thermodynamic standard potential (E0diss†.
The determination of the Eꢀdiss is one of the key aspects of the
present study, and will be discussed below.
Preparative electrochemical studies were also done, and the
coulometric results of the electrolyses of ASC and DASC on a
carbon electrode are summarized in Table 1. Electrolysis
1
leads to the consumption of approximately 2 Fmol , or two
electron equivalents, in both solvents in the presence and
absence of acid. Work-up and isolation after electrolysis yield
quantitatively the corresponding cis-diols, namely cis-1-iso-
propyl-4-methylcyclohexane-1,4-diol and cis-1-isopropyl-4-
methylcyclohex-2-ene-1,4-diol, in both solvents. The results
of the voltammetric and product studies are consistent with
the dissociative reduction of the O O bond by the overall
two-step, two-electron process illustrated for ASC in Equa-
tions (5) and (6).
1
1
Figure 1. Cyclic voltammograms at 0.2 Vs in ACN with 0.1 molL
1
TEAP for 2 mmolL ASC. Dashed lines show the presence of excess
2,2,2-trifluoroethanol. Inset: plot of the dependence of the peak potential
as a function of scan rate.
such as 2,2,2-trifluoroethanol (TFE; shown in Figure 1) or
acetanilide. Irreversible anodic peaks are also observed for
ASC and DASC; the peak current depending on the scan rate.
The position of these peaks and the observation that they
disappear in the presence of acid indicate that they are due to
the oxidation of alkoxides formed in the initial reduction and/
or other electro-generated bases. Increasing the potential scan
rate, n, causes the Ep to become more negative by an average
1
value of 123 mV (logn) (Figure 1, inset) for both ASC and
DASC. The peak width, defined as the difference between the
potential at the half-current height and peak potential (i.e.,
DEp/2 ˆ Ep/2 Ep), of the reduction wave also varies by scan
1
rate. For ASC, the DEp/2 is 182 mV at 0.2 Vs , and increases
to 203 mVat 10 Vs . A summary of the voltammetric data of
these two compounds appears in Table 1 and indicates a
potential dependence on the transfer coefficient, a.[56, 70]
1
According to this mechanism, the first electron results in a
concerted dissociative fragmentation affording O-R-R-O , a
.
species that we term the distonic radical anion. Formation of a
similar intermediate on reduction of ASC in the presence of
Fe(ii) was proposed previously.[72] In this study, the resulting
radical reacted to give high yields of b-scission products. Such
competing radical processes do not occur in our system under
Table 1. Voltammetric data for the reduction of ASC and DASC in DMF and
MeCN (0.1m TEAP) at 258C measured by cyclic voltammetry at a glassy carbon
working electrode.
.
the electrochemical conditions. At the potential where O-R-
MeCN
DMF
.
R-O is generated (or O-R-R-OH), its reduction to the
ASC
1.85
1.97
DASC
1.90
2.00
2.11
ASC
1.88
1.98
DASC
1.93
2.06
2.23
dialkoxide (or O-R-R-OH) is strongly favored, by using the
reasonable assumption that the reduction potential of the
distonic radical anion is similar to that of the tert-butyl alkoxy
radical (E8 ˆ 0.30 V vs. SCE in MeCN and 0.23 V vs. SCE
in DMF).[47, 48] In principle, even if intermediate alkoxy
radicals abstract hydrogen atoms from solvent, particularly
DMF, there are no complications to our analysis described
below, since the resulting solvent radical will be at a potential
where reduction is also a driven process.[45] The net result is
still a two electron reduction, in which the first ET is rate
determining.
1
1
Ep (V versus SCE)
0.1 Vs
1.0 Vs
10 Vs
1
2.12
2.16
aapp
ˆ
(RT/nF)(dlnn/dE)
0.24
0.24
0.24
0.24
1
1
a ˆ 1.85RT/F(Ep/2 Ep)
0.1 Vs
1.0 Vs
10 Vs
0.267
0.253
0.234
1.99
0.270
0.254
0.236
2.06
0.285
0.259
0.227
2.05
0.270
0.253
0.225
1.96
1
1
n [Fmol
]
The transfer coefficient (a) is defined as the rate of change
of the free energy of activation as a function of the free energy
of ET, namely (dDG=/dDGEꢀ T). It can be calculated from the
experimental data from either the variation of Ep with n
through the slope of a plot of logn against Ep or from DEp/2
data.[56, 71] For ASC and DASC, a decreases with increasing
scan rate over the potential scan rates investigated (Table 1),
and average values of a ˆ 0.24 are found in both MeCN and
DMF. The low a and its scan rate dependence indicate that
the heterogeneous rate constant, khet, for ET from electrode
to ASC and DASC is rate determining. The voltammetric
Homogeneous electron-transfer reduction: The reduction of
ASC and DASC by homogeneous electron-transfer techni-
ques was studied. In these experiments an extensive series of
.
±
electro-generated donor radical anions (D ) were used as the
electron source, and react according to Equations (7) ± (9).
Part of this work was reported briefly in our preliminary
communication and involved using both the method of
homogeneous redox catalysis[73±75] and, for measurement of
4014
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0718-4014 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 18

Electron Transfer to Endoperoxides
4012±4020
some of the slower reactions, the method outlined by
Pedersen and Daasbjerg.[76, 77] In the former, the catalytic
increase in current for the reduction of the donor, measured as
a function of varying concentrations of ASC or DASC at
different scan rates, was compared with data obtained by
simulation of the two electron mechanism outlined in
ˆ 1.55‡0.808E and a ˆ 1.22‡0.561E for ASC and DASC,
respectively. An a value of 0.5 corresponds to the dissociative
reduction potentials, Eꢀdiss, for ASC and DASC and gives
values of 1.30 V and 1.28 V, respectively. These values
show that the observed Ep values of the heterogeneous
reduction overestimate the true standard potential by at least
0.6 V.
Equations (8) and (9) to obtain kET
.
.
±
D ‡ e > D
(7)
(8)
(9)
Heterogeneous ET reduction: The kinetics of ET to ASC and
DASC were studied by heterogeneous reduction, in which the
electron source in Equation (5) was a glassy carbon electrode.
A convolution analysis approach was used to study the
heterogeneous electron-transfer kinetics of the dissociative
reduction of ASC and DASC. This method, for which the
experimental and theoretical treatments have been described
in detail previously,[46, 59, 78, 79] has only recently proven to be
powerful for the determination of the standard potentials for
systems that undergo dissociative ET. In particular, it has been
successfully used to study the direct reduction of a number of
acyclic dialkyl peroxides[46, 48] and perbenzoates.[78, 80] We have
used this method to determine the standard reduction
potential of the antimalarial trioxane, Artemisinin.[49]
kET
.
.
±
D
D
‡ ASC/DASC
D ‡ O-R-R-O±
!
.
.
±
‡
O-R-R-O± ! D ‡ ±O-R-R-O±
In none of the cases was there evidence for any competing
reactions between the donor and endoperoxides or between
the donor radical anion and distonic radical anion. The
involvement of reactions of the mediator radical anion with
endoperoxide (or with solvent) was tested by monitoring the
1
extent of recovery of mediator after electrolysis of 2 Fmol
with one half equivalent of the endoperoxide. At least 98% of
the mediator was recovered, even in the cases for slow
dissociative ET, where competing reactions of the radical
anion with electrophilic impurities could occur.
Kinetic data obtained from an average of at least three
independent experiments in both MeCN and DMF are
summarized in Table 2. Data corresponding to the slowest
The collection of high-quality cyclic voltammograms is
essential for meaningful thermochemical information to be
extracted from the acquired data. A sample of background-
subtracted linear-scan voltammograms measured between
1
1
reactions that define the parabolic shape of a plot of ln(khom
)
0.1 Vs
and 10 Vs
for ASC is shown in Figure 2a.
against the E8 of the donors have been performed more often
and with a variety of donors with similar E8 to verify the value
of the rate constants with donors at that potential. In general,
the reaction rate constants increase as the standard potential
of the mediator becomes more negative (this corresponds to
an increase in driving force for reaction Equation (8)). The
kinetic data are independent of the solvent system (DMF or
MeCN).
Convolution of the voltammetric curves yields limiting
current values, Ilim, that are independent of scan rate. The
voltammetric curves correspond to the uptake of two
electrons per molecule; this is consistent with the two-step
fragmentation mechanism in Equations (5) and (6). I is
related to the actual current i through the convolution
integral.[71, 81] The I against E plot is a wave-like curve and
the plateau value is reached when the applied potential is
sufficiently negative. The result of this analysis for ASC is
shown in Figure 2b. Under these conditions (diffusion con-
The homogeneous ET kinetic data can be related to the
transfer coefficient, a, through a ˆ (RT/F)d(lnkET)/dE.
Thus, taking the derivative of the quadratic fit of the
trol), I reaches its limiting value, Ilim, defined as Ilim
ˆ
combined ln(khom) versus E8 data gives equations a
nFAD1/2C*, where n is the overall electron consumption, A
.
Table 2. Rate constants for electron transfer (logkET) from electrochemically generated donors (D ±) with varying standard potentials (E8) to ASC and
DASC in 0.1m TEAP/DMF and MeCN solutions at 258C.
Donor
E8 (MeCN)
versus SCE[a]
E8 (DMF)
versus SCE[a]
ASC
DASC
[b]
[b]
[b]
[b]
logkET
logkET
logkET
logkET
MeCN
DMF
MeCN
DMF
perylene
acenaphthylene
1.674
1.651
1.640
4.00
3.64
3.86
4.06
3.43
3.78
1.629
3.42
3.45
[d]
[d]
[d]
1,4-dicyanobenzene
tetracene
4,4'-dimethoxyazobenzene
4-methyl-4'methoxyazobenzene
4-methoxyazobenzene
azobenzene
3,3'-dimethoxyazobenzene
2-nitrobiphenyl
nitrobenzene
1.576
3.44
3.20
[c]
[d]
[d]
1.548
1.531
1.478
1.424
1.316
1.258
1.186
1.100
1.078
1.040
3.43
3.08
2.97
2.60
1.94
1.62
3.20
2.78
2.49
2.17
1.47
1.32
1.526
1.478
1.426
1.335
1.278
1.157
2.91
2.75
2.47
1.75
1.39
0.157
2.75
2.51
2.05
1.81
1.42
0.240
0.822
0.204
1.100
0.477[e]
0.767[e]
0.647[e]
1.030[e]
0.976[e]
0.899[e]
[d]
[d]
[d]
[d]
4-nitrobiphenyl
1-nitronaphthalene
[d]
1.041
0.602[e]
1.230[e]
‡
[a] E8 measured versus an internal standard (ferrocene) and corrected: Fc/Fc ˆ 0.449 V against SCE in MeCN and 0.475 V against SCE in DMF.
[b] Measured by homogeneous redox catalysis (ref. [73 ± 75]). [c] Not soluble. [d] Not measured. [e] Measured by using methods outlined in refs. [76] and
[77].
Chem. Eur. J. 2001, 7, No. 18
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0718-4015 $ 17.50+.50/0
4015

M. S. Workentin and R. L. Donkers
FULL PAPER
the standard dissociative reduction potential of ASC and
DASC. Doing this yields Eꢀdiss values of 1.0 and 0.97 V,
respectively. These values are 0.3 V less negative than those
estimated by the homogeneous approach.
It is interesting to compare the homogenous and heteroge-
neous approaches. In the former, the amount of measurable
data is small and limited to the number and range of donors
that can be used. In the latter, the free energy of the reaction is
varied continuously and, thus, the amount of data in a
particular free energy range is large, although the range of
free energy over which data can be measured is limited by the
scan rates that can be investigated. While we somewhat prefer
the Eꢀdiss values obtained by homogeneous data, because of the
limited potential range over which the heterogeneous data
could be measured, the Eꢀdiss values we use below in DGꢀET
calculations are the average values from the two approaches
to better reflect the uncertainty in the estimate; thus, Edꢀiss
ˆ
1.2 Æ 0.2 and 1.1 Æ 0.2 V for ASC and DASC, respectively.
It must be stressed that these average Eꢀdiss values provide a
more accurate determination of the standard reduction
potential then those estimated from the peak potential.
Both the heterogeneous and homogeneous ET rate-con-
stant data can thus be plotted against the free energy of ET, by
taking into account that DGꢀET ˆ F(EDꢀ =D
Eꢀdiss) and using
.
the average Eꢀdiss values for ASC and DASC; these plots are
shown in Figure 3a and b for ASC and DASC, respectively.
The homogeneous kinetic data from Table 2 shows the fit to
Equations (2) ± (4) as described below.
Figure 2. a) Background-subtracted linear-scan voltammograms for
1
1
2 mmolL ASC in ACN with 0.1 molL TEAP and excess 2,2,2-tri-
fluoroethanol at a glassy carbon electrode. The voltammetric data is in its
1
n-normalized form. The scan rates were 0.1, 0.5, 1, 2, 5, and 10 Vs
.
b) Corresponding convolution curves at scan rates 0.1, 0.2, 0.5, 1, 2, 5, 8, and
10 Vs 1. c) Potential dependence of the apparent transfer coefficient for
the reduction.
the electrode area, D the diffusion coefficient, and C* the
substrate concentration. Since I is related to the surface
concentration of reactant and product, an equation relating
khet to i and I can be derived. For a totally irreversible process,
as is the case for the reduction of ASC and DASC, at constant
Ilim the limiting current can be related to khet by using the
expression:
!
‰Ilim Iꢁt†Š
iꢁt†
lnkhet ˆ lnD1/2 ln
For ASC and DASC the diffusion coefficients were
1
determined to be 2.36 Â 10 5 cm2 s in MeCN and 1.08 Â
1
10 5 cm2 s in DMF, respectively, by using the Tayler dis-
persion method.[82, 83] When corrected for the viscosity change
of solvent with 0.1m TEAP according to the Stokes ± Einstein
1
relation, the D values become 2.16 Â 10 5 cm2 s and 9.93 Â
1
10 6 cm2 s , respectively. Convolution analysis of 20 sets of
data measured at 1 mV intervals provide lnkhet data that can
be plotted against the driving force, E. By using the usual
activation ± driving force relationship lnkhet versus E data is
related to the apparent transfer coefficient aapp. Each curve
from the lnkhet data is fitted to the above equation in 21 mV
segments. The result of this analysis for ASC is shown in
Figure 2c; similar results were determined for DASC and are
not shown. Alpha values determined in this way agree with
those summarized in Table 1. In principle, these values can be
used similarly to the homogeneous data to obtain estimates of
Figure 3. Composite plot of logkhom [m 1 s 1] versus DGEꢀ T (left) with plots
of logkhet [cm2 s 1] versus DGEꢀ T (right) for a) DASC and b) ASC. The
heterogeneous plot is a composite of 25 separate experiments where the
scan rate was varied between 0.1 and 10 Vs 1. The solid line through the
homogeneous data is fitted according to Equations (2) and (4) with the
parameters summarized in the text and by using the nonadiabatic Z term.
The solid line through the heterogeneous data is to a simple second-order
polynomial.
4016
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0718-4016 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 18

Electron Transfer to Endoperoxides
4012±4020
Temperature dependence of the rate constants: Activation
parameters for the reduction of ASC and DASC were
measured from a temperature study by using a pair of donors,
acenaphthylene and 4,4'-dimethoxyazobenzene. These studies
provided logA values of 7.1 ± 7.5 and Ea values of 4.9 ±
Values for the BDE of ASC and DASC can be estimated from
gas-phase bond-dissociation enthalpy for the homolytic
cleavage of 1,4-dimethyl-2,3-dioxabicyclo[2.2.2] octane. The
heat of formation of the endoperoxide was estimated to be
1
50.53 kcalmol from ab initio energies, following a group
1
6.0 kcalmol for ASC and logA values of 7.6 ± 7.8 and Ea
contribution procedure analogous to that of Schleyer and
Ibrahim, with the following changes: energies were calculated
at the B3LYP/6 ± 31G**//RHF/6 ± 31G** level and groups
1
values of 5.7 ± 6.8 kcalmol for DASC. (See Supporting
Information for complete details.) Interestingly, the pre-
exponential factors are 2.5 ± 3 orders of magnitude lower than
those calculated for an adiabatic ET process by the methods
previously described.[48] Therefore, the adiabatic dissociative
ET model does not account for the experimental kinetic data
for these endoperoxides. The unusually low, measured pre-
exponential factors are, however, predicted by the theory for
nonadiabatic dissociative ET[84] and consistent with our recent
results for ET to acyclic peroxides, which is nonadiabatic due
to poor electronic coupling between the reactant and product
electron potential energy surfaces.[48]
.
were defined by Benson. The heat of formation of the O-R-
.
1
R-O diradical was estimated to be 21.4 kcalmol . Using
1
these values provides O O BDE of 29 kcalmol . The BDE
for ASC is significantly lower than the value 37 kcalmol for
1
the BDE of acyclic peroxides.[48] This difference is expected
based on the added molecular strain and repulsive forces due
to the eclipsing interactions of the lone pairs on the oxygen
atoms in ASC and DASC.
Interestingly, this value of the BDE is the same as that
Â
estimated from an expression derived by Saveant that relates
the BDE to the Ep of the dissociative process and the
ꢀ
.
Â
oxidation potential of the anion fragment, namely ERO =RO . If
Determination of thermochemical parameters: Saveantꢁs
ꢀ
.
one makes the assumption that EHORRO =HORRO is equal to the
theory of dissociative ET can be applied to the logkET versus
DGꢀET data for ASC and DASC [Eqs. (2) ± (4)] to determine
valuable thermochemical information, such as the intrinsic
barrier DG0=.[56, 57, 60, 61] The kinetic plots shown in Figure 3 are
constructed from the homogeneous data summarized in
Table 2, the heterogeneous kinetic data obtained by convo-
ꢀ
.
standard potential of the tertiary alkoxy radical EtBuO =tBuO
,
then the DBDE between di-tert-butylperoxide and ASC
can be estimated from DBDE ˆ 2/3[Ep(di-tert-butyl
peroxide) Ep(ASC)]. Since the BDE of the O O bond in
1
di-tert-butyl peroxide is 37 kcalmol and its Ep ˆ 2.50
lution analysis and our subsequent determination of Edꢀiss
.
under the same conditions, the BDE of ASC is estimated to
1
be 28 kcalmol . It should be stressed that this latter relation-
ship of Saveantꢁs is uncertain, and the fact that our theoretical
These figures convincingly illustrate the parabolic nature of
and provides support for the activation ± driving force rela-
tionship described by Equations (2) ± (4). The amount of
curvature in the data depends on DG=0 , which itself depends
on the reorganization energy and BDE for the dissociative
reduction. Both sets of data can be fitted according to
Equations (2) ± (4) to provide an estimate for DG=0 by
using an estimate of the nonadiabatic pre-exponential factor
(kelZ).
An estimate of kelZ for the present work was determined by
using the homogeneous ET kinetics of di-tert-butylperoxide
determined from the previous study.[47, 48] Using a value of
DG=0 ˆ 13.1 kcalmol 1 leads to a value of 4.3  108 m 1 s for
Â
estimation is the same may be fortuitous.
By using the above estimate for BDE and the DG0=
obtained from the fit of the data, the homogeneous reorgan-
ization energies (lhom) were determined to be 19.4 and
1
24.3 kcalmol for ASC and DASC, respectively. It should
be noted that lhom values obtained from the fits by using the
usual adiabatic Z term (3 Â 1011 s ) are 9 ± 12 kcalmol
1
1
higher and do not agree with estimates below. Estimates for
the homogeneous solvation energy can also be obtained by
using expressions by Marcus and Kojima ± Bard[48, 85] and a
similar expression with a modified term suggested by Sa-
1
veant.[56] Accurate values for radius of the donor, rD, and the
Â
the kelZ for di-tert-butylperoxide. To give an estimated kelZ
for the nonsterically encumbered ASC/DASC in this study, we
added a further correction of 0.8 log units to remove the
contribution of the steric effect imposed by the bulky tert-
butyl groups of di-tert-butylperoxide. This latter value was
estimated by comparison of the difference in khet values
between di-tert-butylperoxide and di-n-butylperoxide at sim-
ilar driving forces; these two endoperoxides have the same
radius of the acceptor, rAB, are required for a reasonable
estimate of lhom; these were estimated from the following
procedure. The radius of ASC was determined from the
density of the compound by using r(Š) ˆ 108[(3M/4pNA1)1/3]
and from the diffusion coefficient calculated by the Stokes ±
Einstein equation. It is expected that most of the charge in the
distonic radical anion is localized on the oxygen atom and that
it is shielded from the rest of the molecule. Thus we use an
effective radius that better accounts for this smaller area
where the charge is localized. The effective radii, reff, was
determined by using reff ˆ [rA(2rAB rA)/rAB], where AB refers
the starting endoperoxide and A the tert-butyl alkoxide anion,
similar to the anion that is generated in the ET.[48, 56] An
effective radius for the tert-butyl alkoxide anion was deter-
mined similarly with AB as the radius of tert-butyl alkoxide
and B the radius of an oxygen atom.[46] It was assumed that
ASC and DASC have the same radius. An average rD of
3.80 Š was chosen for the donors. The calculated effective
BDE.[48] This leads to an estimated kelZ of 2.7 Â 109 m 1 s .
1
The logkhom versus DGꢀET data for ASC and DASC were then
fitted to Equations (2) ± (4) by using this kelZ value to
determine the intrinsic barriers, DG=0 . The fits are shown in
Figure 3 as solid lines through the data and provide values for
1
DG=0 of 11.9 and 9.6 kcalmol
for ASC and 13.2 and
1
11.0 kcalmol for DASC from the homogeneous and hetero-
geneous data, respectively.
Values for DG=0 can be related to the BDE and the
solvation energy (l) since DG=0 ˆ (l ‡ BDE)/4. We assume
that the BDE for ASC and DASC is the same within error.
Chem. Eur. J. 2001, 7, No. 18
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0718-4017 $ 17.50+.50/0
4017

M. S. Workentin and R. L. Donkers
FULL PAPER
radii is 2.76 Š. This radius leads to a lhom of 15.2 kcalmol 1 by
using the Kojima ± Bard approach and 15.9 kcalmol with
determine BDE and other thermochemical parameters
for other endoperoxides, especially those of biological rele-
vance.
1
Â
Saveantꢁs approach.
The slope of the regression analysis of the a versus E plots is
directly linked to the intrinsic barrier through the expression
da/dE ˆ F/8DG=0 . With the BDE for ASC and DASC known,
1
the heterogeneous solvation energy, lhet is 9.9 kcalmol for
Experimental Section
1
ASC and 15.5 kcalmol for DASC. These can be compared
Materials: Ascaridole and dihydroascaridole were synthesized by standard
literature procedures[87, 88] and purified by distillation with a Kugelrohr
apparatus at 408C (0.025 mm). CAUTION: Peroxides are potentially
explosive; heat with care. The homogeneous ET mediators 9,10-diphenyl-
anthracene, fluoranthene, perylene, acenaphthylene, 1,4-dicyanobenzene,
tetracene, azobenzene, 2-nitrobiphenyl, nitrobenzene, 4-nitrobiphenyl, and
1-nitronaphthalene were used as received from Aldrich. Other mediators
were synthesized: 4-methoxyazobenzene (m.p. 55 568C ) and 4-methoxy-
4'methylazobenzene (m.p. 100 ± 100.58C) were made by treatment of the
corresponding hydroxy-substituted azobenzenes with methyl iodide and
potassium carbonate in acetone. 4,4'-dimethoxyazobenzene,[89] 3,3'-dimeth-
oxyazobenene[90] and 4-hydroxy-4'-methylazobenzene[91] were made by
following literature procedures. All of the substituted azobenzenes were
recrystallized with an appropriate solvent or purified by column chroma-
tography according to the corresponding literature procedure before use.
Spectroscopic grade N,N-dimethylformamide (Lancaster) was distilled
over CaH2 under nitrogen at reduced pressure. Spectroscopic grade
acetonitrile (BDH) was distilled over CaH2 and run through a column of
activated alumina prior to storage. Both solvents were stored over alumina,
which had been activated at 3508C while under vacuum (0.02 mmHg), for
48 hours. Tetraethylammonium perchlorate (TEAP, Aldrich) was recrystal-
lized three times from ethanol and dried under vacuum at 608C.
with calculated values by using the approaches of Marcus
1
1
(13.9 kcalmol ) and Kojima ± Bard (20.2 kcalmol )[85]
.
The values obtained for DG=0 and l from the homogeneous
and heterogeneous data are comparable to previously re-
ported values obtained for other O O systems, including the
trioxane, artemisinin, and a number of acyclic perox-
ides.[46, 48, 49] In general, the intrinsic free energies are similar
for all systems, within error (an error of 0.1 V in Eꢀdiss manifests
1
itself as an error of 1 kcalmol in DG=0 †. Since the BDE of
the endoperoxides is much lower than that of the acyclic
peroxides, the result is a somewhat higher average value for l.
Keeping in mind that an error of 1 kcalmol 1 in DG0= leads to
more uncertainty in l, it is tempting to think that the higher l
values in the endoperoxides may reflect additional internal
reorganization energy in these strained systems. Such con-
tributions are generally considered to be small in dissociative
processes with large BDEs but, with the much lower BDEs in
these endoperoxides, they may contribute to the intrinsic
barrier. Also of interest is the observation that the l is larger
for DASC than ASC. DASC, lacking the alkene moiety, is
somewhat more floppy that ASC. This floppiness is reflected
in the C-O-O-C dihedral angles (08 in ASC against 158 in
DASC).[86]
Nuclear magnetic resonance spectra were recorded on a Varian XL300
spectrometer with tetramethylsilane as the internal reference standard. IR
spectra were recorded on Aviv-17 IR spectrometer and mass spectra were
recorded on a MAT8200 Finigan high resolution mass spectrometer.
Electrochemical apparatus and procedures: Cyclic voltammetry was
performed by using a PAR283 or 263 potentiostat interfaced to a personal
computer with PAR270 electrochemistry software. A three-electrode
arrangement was used in the electrochemical cell.[71] The cell ohmic drop
was compensated for by using a positive feedback approach and adjusted to
at least 98% of the oscillation value. The electrodes and a stir-bar were
assembled in a water-jacketed electrochemical cell, while in an oven at
1008C, and placed in a Faraday cage to cool under a continuous flow of
high-purity dry argon. An argon atmosphere was maintained throughout
the experiment to protect the contents of the cell from atmospheric oxygen
and moisture. After cooling, a temperature of 258C was maintained
through the jacketed cell by using a circulating water bath. For the
temperature-dependence studies, the temperature was controlled by a
VWR Scientific Model 1150A constant-temperature circulator; the cell
was allowed to equilibrate for 30 minutes at each temperature. The
appropriate solvent and 0.1m electrolyte were added to the cell, and oxygen
was purged from the solution with argon. Throughout the experiment, an
argon atmosphere was maintained. The working electrode was a 3 mm
diameter glassy carbon rod (Tokai, GC-20) sealed in glass tubing. Prior to
each experiment, the working electrode was freshly polished with 0.25 mm
diamond paste and ultrasonically rinsed in 2-propanol for fifteen minutes.
Electrochemical activation of the electrode was carried out in the
background solution by cycling several times between 0 and 2.7 V versus
SCE at a 0.2 Vs 1 scan rate. The surface area was calculated with reference
to the diffusion coefficient of ferrocene in ACN and DMF with TEAP
(0.1m). The counter electrode was a platinum plate placed symmetrically
under the working electrode. The reference electrode was made of a silver
wire immersed in a glass tube with a fine sintered bottom containing TEAP
(0.1m) in the desired solvent. Its potential remained constant for the
duration of the experiment and was then calibrated by adding ferrocene. In
these studies, the standard potential of the ferrocene/ferrocenium couple
was 0.449 V and 0.475 V versus SCE for ACN and DMF, respectively. All
reported potentials are referenced to SCE.
Conclusion
ASC and DASC undergo a concerted dissociative reduction
by an ET to the s* orbital of the O O bond. The systems
reported here provide interesting cases for testing the theories
of dissociative ET since, on reduction, both the charged and
the radical fragments remain in the same molecule. We
provide the only reliable and most comprehensive kinetic
data to date for ET to these types of cyclic O O systems and
use them to determine accurate standard reduction potentials
for model endoperoxides. Further, our temperature-depend-
ent rate studies provide support for the theory that the
concerted dissociative ET to endoperoxides is nonadiabatic.
Like the O O bond in acyclic peroxides, the nonadiabaticity
and thus poor electronic coupling between reactant and
product surfaces may be the result of the symmetry of the
O O bond and the lack of change in dipole moment during
bond stretching in the transition state. In addition to providing
the best example of the quadratic nature of the activation ±
driving relationship for dissociative ETs, fits of the homoge-
neous and heterogeneous ET kinetic data to nonadiabatic ET
theory provide an estimate of DG=0 . The use of a reasonable
BDE for this system provides an understanding of the
reorganization energies. Now that the thermochemical prop-
erties of O O reduction are better understood, the electro-
chemical approaches described herein may be used to
Standard potentials for the homogeneous donors were determined in a
standard voltammetry cell. Cyclic voltammograms at scan rates ranging
from 0.1 to 5 Vs 1 of a 1 ± 3 mmolL 1 solution of donor were obtained, then
4018
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0718-4018 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 18

Electron Transfer to Endoperoxides
4012±4020
ferrocene was added to the cell and a series of ferrocene/ferrocenium
voltammograms were acquired. The standard potentials for the reversible
homogeneous donors and ferrocene were determined as the average of the
anodic and cathodic peak potentials. Simulation of homogeneous redox
catalysis data was performed with Digisim 2.1ꢂ software.
[13] C. W. Jefford, M. G. H. Vicente, Y. Jacquier, F. Favarger, J. Mareda, P.
Millasson-Schmidt, G. Brunner, U. Burger, Helv. Chim. Acta 1996, 79,
1475.
[14] R. K. Haynes, S. C. Vonwiller, Acc. Chem. Res. 1997, 30, 73.
[15] M. Hamzaoui, O. Provot, B. Camuzat-Dedenis, H. Moskiwitz, J.
Â
Mayrargue, L. Ciceron, F. Gay, Tet. Lett. 1998, 39, 4029.
Preparative electrolysis: All preparative electrolysis was carried out in
argon-purged solutions containing TEAP (0.1m, 25 mL), with the reference
and glassy carbon electrode described for cyclic voltammetry. The working
electrode was a reticulated vitreous-carbon or platinum gauze and the
anode a platinum wire that was separated by a glass tube with a sintered
[16] J. N. Cumming, D. Wang, S. B. Park, T. A. Shapiro, G. H. Posner, J.
Med. Chem. 1998, 41, 952.
[17] G. H. Posner, M. H. Parker, J. Northrop, F. S. Elias, P. Ploypradith, S.
Xie, T. A. Shapiro, J. Med. Chem. 1999, 42, 300.
[18] G. H. Posner, J. N. Cumming, S.-H. Woo, P. Ploypradith, S. Xie, T. A.
Shapiro, J. Med. Chem. 1998, 41, 940.
[19] J. Bhisutthibhan, X.-C. Pan, D. J. Hossler, D. J. Walker, C. A. Yowell,
J. Carlton, J. B. Dame, S. R. Meshnick, J. Biol. Chem. 1998, 273, 16192.
[20] Y. Wu, Y. Y. Zheng, Y.-L. Wu, Angew. Chem. 1999, 111, 2730 ± 2733;
Angew. Chem. Int. Ed. 1999, 38, 2580.
bottom and
a 3 mm layer of neutral alumina. Constant potential
electrolysis was carried out in an identical cell to that described for
constant current electrolysis, except that the electrodes were connected to a
potentiostat. Electrolysis was carried out at a potential of approximately
100 mV past the peak maximum and continued until the current was
reduced to the initial background level.
[21] W.-M. Wu, Y. Wu, Y.-L. Wu, Z.-J. Tao, C.-M. Zhou, Y. Li, F. Shan, J.
Am. Chem. Soc. 1998, 120, 3316.
[22] A. Robert, B. Meunier, Chem. Eur. J. 1998, 4, 1287.
A cyclic voltammogram on a small 3 mm glassy carbon electrode was used
as a probe for the concentration of the substrate in the electrolysis cell in
both constant potential and constant current experiments. After electrol-
ysis, water (40 mL) was added to the cell, and the contents were extracted
with dichloromethane (4 Â 20 mL). The organic extracts were then washed
with water (2 Â 20 mL), dried over sodium sulfate, filtered and evaporated.
In the preparative scale electrolysis of ASC and DASC, only one product
was formed in each; this corresponded to the following: electrolysis of
ASC: cis-1-isopropyl-4-methylcyclohex-2-ene-1,4-diol. 1H NMR (CDCl3):
d ˆ 0.87 (d, J ˆ 6.9 Hz, 3H), 0.93 (d, J ˆ 6.9 Hz, 3H), 1.23 (s, 3H), 1.62 ± 1.87
(m, 7H), 5.53 (dd, J ˆ 10.1,1.3 Hz, 1H), 5.69 (dd, J ˆ 10.1, 1.3 Hz, 1H);
13C NMR (CDCl3): d ˆ 16.72, 18.74, 27.38, 35.19, 37.52, 69.88, 72.18, 137.54,
132.36; IR: 3353 cm 1 (br), 3024, 2969, 2880, 1606, 1374, 1381, 1138, 1004;
MS: m/z (%): 155 (12), 127 (62), 109 (100). Deuterium exchange verified
that there were two alcoholic protons.
Â
[23] G. H. Posner, D. Wang, L. Gonzalez, X. Tao, J. N. Cumming, D.
Klinedinst, T. A. Shapiro, Tetrahedron Lett. 1996, 37, 815.
[24] A. Robert, B. Meunier, Chem. Soc. Rev. 1998, 27, 273.
[25] R. K. Haynes, S. C. Vonwiller, Tetrahedron Lett. 1996, 37, 253.
[26] R. K. Haynes, S. C. Vonwiller, Tetrahedron Lett. 1996, 37, 257.
[27] A. Robert, G. Meunier, J. Am. Chem. Soc. 1997, 119, 5968.
Â
[28] G. H. Posner, S. B. Park, L. Gonzalez, D. Wang, J. N. Cumming, D.
Klinedinst, T. A. Shapiro, M. D. Bachi, J. Am. Chem. Soc. 1996, 118,
3537.
[29] P. M. OꢁNeill, L. P. Bishop, N. J. Searle, J. L. Maggs, S. A. Ward, P. G.
Bray, R. C. Storr, B. K. Park, Tetrahedron Lett. 1997, 38, 4263.
[30] W. Adam, A. Schonberger, Chem. Ber. 1992, 125, 2149.
[31] J. Koo, G. B. Schuster, J. Am. Chem. Soc. 1978, 100, 4496.
[32] G. B. Schuster, Acc. Chem. Res. 1979, 12, 366.
[33] G. B. Schuster in Advances in Electron Transfer Chemistry Vol. 1 (Ed.:
P. S. Mariano), JAI Press, Greenwich, 1991, p. 163.
[34] J. J. Zupancic, K. A. Horn, G. B. Schuster, J. Am. Chem. Soc. 1980,
102, 5279.
Electrolysis of DASC: cis-1-isopropyl-4-methylcyclohexane-1,4-diol.
1H NMR (CDCl3): d ˆ 0.85 (d, J ˆ 6.9 Hz, 6H), 1.15 (s, 3H), 1.32 ± 1.86
(m, 10H); 13C NMR (CDCl3): d ˆ 16.76, 25.79, 31.89, 35.78, 36.11, 38.76,
1
70.25; MS: m/z (%): 154 (8), 139 (10), 129 (15), 111 (100); IR: 3353 cm
(br), 2956, 2875, 1371, 1379, 1123, 998. Deuterium exchange verified that
there were two alcoholic protons.
[35] W. Adam, I. Erden, Angew. Chem. 1978, 90, 223; Angew. Chem. Int.
Ed. Engl. 1978, 17, 210.
[36] W. Adam, O. Cueto, J. Am. Chem. Soc. 1979, 101, 6511.
[37] W. Adam, K. Zinner, A. Krebs, H. Schmalstieg, Tetrahedron Lett.
1981, 22, 4567.
[38] Y. Takano, T. Tsunesad, H. Isobe, Y. Yoshioka, K. Yamaguchi, I. Saito,
Bull. Chem. Soc. Jpn. 1999, 72, 213.
[39] T. Wilson, Photochem. Photobiol. 1995, 62, 601.
[40] W. Adam, A. V. Trofimov, J. Org. Chem. 2000, 65, 6474.
[41] W. Adam, M. Matsumoto, A. V. Trofimov, J. Org. Chem. 2000, 65,
2078.
[42] W. Adam, M. Matsumoto, A. V. Trofimov, J. Am. Chem. Soc. 2000,
122, 8631.
Acknowledgements
The financial support of the Natural Sciences and Engineering Research
Council of Canada (NSERC), the Canadian Foundation for Innovation,
ORCDF, and the University of Western Ontario (ADF) is gratefully
acknowledged. R.L.D. thanks NSERC for a PGS scholarship. Professor
Flavio Maran is thanked graciously for numerous and fruitful discussions,
and for providing analysis software.
[43] S. Yamada, K. Nakayama, H. Takayama, J. Org. Chem. 1983, 48, 3477.
[44] M. G. Zagorski, R. G. Salomon, J. Am. Chem. Soc. 1982, 104, 3498.
[45] N. T. Kjñr, H. Lund, Acta Chem. Scand. 1995, 49, 848.
[46] S. Antonello, M. Musumeci, D. D. M. Wayner, F. Maran, J. Am. Chem.
Soc. 1997, 119, 9541.
[47] M. S. Workentin, F. Maran, D. D. M. Wayner, J. Am. Chem. Soc. 1995,
117, 2120.
[48] R. L. Donkers, F. Maran, D. D. M. Wayner, M. S. Workentin, J. Am.
Chem. Soc. 1999, 121, 7239.
[1] E. L. Clennan, C. S. Foote in Organic Peroxides (Ed.: W. Ando),
Wiley, Chichester, England, 1992, p. 225.
[2] D. A. Casteel, Nat. Prod. Rep. 1999, 16, 55.
[3] R. G. Salomon, Acc. Chem. Res. 1985, 18, 294.
[4] W.-S. Zhou, X.-X. Xu, Acc. Chem. Res. 1994, 27, 211.
[5] J. A. Vroman, I. A. Khan, M. A. Avery, Tetrahedron Lett. 1997, 38,
6173.
Â
[49] R. L. Donkers, M. S. Workentin, J. Phys. Chem. 1998, 102, 4061.
[50] R. L. Donkers, J. Tse, M. S. Workentin, Chem. Commun. 1999, 135.
[51] M. S. Workentin, R. L. Donkers, J. Am. Chem. Soc. 1998, 120, 2664.
[52] D. D. M. Wayner, V. D. Parker, Acc. Chem. Res. 1993, 26, 287.
[53] C. G. Moore, J. Chem. Soc. 1951, 234.
[54] A. C. Baldwin, The Chemistry of Peroxides, Wiley, New York, 1983.
[55] L. Batt, K. Kristie, R. T. Milne, A. Summers, Int. J. Chem. Kinet. 1974,
6, 877.
[6] G. H. Posner, L. Gonzalez, J. N. Cumming, D. Klinedinst, T. A.
Shapiro, Tetrahedron 1997, 53, 37.
[7] F. Zouhiri, D. Desmaele, J. dꢁAngelo, J. Mahuteau, C. Riche, F. Gay, L.
Ciceron, Eur. J. Org. Chem. 1998, 2897.
[8] E. Van Geldre, A. Vergauwe, E. Van den Eeckhout, Plant Mol. Biol,
1997, 33, 199.
[9] G. H. Posner, H. OꢁDowd, T. Caferro, J. N. Cumming, P. Ploypradith,
S. Xie, T. A. Shapiro, Tetrahedron Lett. 1998, 39, 2273.
[10] H. OꢁDowd, P. Ploypradith, S. Xie, T. A. Shapiro, G. H. Posner,
Tetrahedron 1999, 55.
È
Â
[56] J.-M. Saveant in Advances in Electron Transfer Chemistry (Ed.: P. S.
Mariano), JAI Press, Greenwich, CT, 1994, p. 53.
Â
[57] C. P. Andrieux, I. Gallardo, J.-M. Saveant, K.-B. Su, J. Am. Chem. Soc.
[11] S. R. Meshnick, T. E. Taylor, S. Kamchonwongpaisan, Microbiol. Rev.
1996, 60, 301.
1986, 108, 638.
Â
[58] J.-M. Saveant, Acc. Chem. Res. 1993, 26, 455.
[12] M. Jung, S. Lee, Heterocycles 1997, 45, 1055.
Chem. Eur. J. 2001, 7, No. 18
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0718-4019 $ 17.50+.50/0
4019

M. S. Workentin and R. L. Donkers
FULL PAPER
[59] S. Antonello, F. Maran, J. Am. Chem. Soc. 1998, 120, 5713.
[76] K. Daasbjerg, S. U. Pedersen, H. Lund, Acta Chem. Scand. 1991, 45,
424.
Â
[60] J.-M. Saveant, J. Am. Chem. Soc. 1987, 109, 6788.
Â
[61] J.-M. Saveant, J. Am. Chem. Soc. 1992, 114, 10595.
[77] S. U. Pedersen, K. Daasbjerg, Acta Chem. Scand. 1989, 43, 301.
[78] S. Antonello, F. Maran, J. Am. Chem. Soc. 1998, 119, 12595.
[79] F. Maran, D. D. M. Wayner, M. S. Workentin in Advances in Physical
Organic Chemistry (Ed.: T. Tidwell), Academic Press, in press.
[80] S. Antonello, F. Maran, J. Am. Chem. Soc. 1999, 121, 9668.
[62] T. Lund, H. Lund, Acta Chem. Scand. Ser. B. 1987, 41, 93.
[63] C. Burkholder, W. R. Dolbier, Jr., M. Medebielle, J. Org. Chem. 1998,
63, 5385.
Â
[64] C. P. Andrieux, D. Differding, M. Robert, J.-M. Saveant, J. Am. Chem.
Â
Soc. 1993, 115, 6592.
[81] J. C. Imbeaux, J.-M. Saveant, J. Electroanal. Chem. 1973, 44, 169.
[65] S. Jakobsen, H. Jensen, S. U. Pedersen, K. Daasbjerg, J. Phys. Chem.
1999, 103, 4141.
[82] H. J. V. Tyrrell, K. R. Harris, Diffusion in Liquids, Butterworths,
London, 1984.
Â
[66] M. G. Severin, M. C. Arevalo, F. Maran, E. Vianello, J. Phys. Chem.
[83] C. Erkey, A. Akgerman in Measurements of the Transport Properties
of Liquids (Eds.: W. A. Wakeham, A. Nagashima, J. V. Sengers),
Blackwell, London, 1991, p. 251.
[84] E. D. German, A. M. Kuznetsov, J. Phys. Chem. A 1994, 98, 6120.
[85] H. Kojima, A. J. Bard, J. Am. Chem. Soc. 1975, 97, 6317.
[86] R. S. Brown, Can. J. Chem. 1975, 53, 3439.
1993, 97, 150.
[67] T. B. Christensen, K. Daasbjerg, Acta Chem. Scand. 1997, 51, 307.
[68] K. Daasbjerg, H. Jensen, R. Benassi, F. Taddei, S. Antonello, A.
Gennaro, F. Maran, J. Am. Chem. Soc. 1999, 121, 1750.
[69] The Merck Index, 11th ed. Merck Co. Inc., Rahway, NJ (USA), 1989.
Â
[70] J.-M. Saveant, D. Tessier, Faraday Discuss. Chem. Soc. 1982, 74, 57.
[87] G. O. Schenck, Angew. Chem. 1952, 64, 12.
[71] A. J. Bard, L. R. Faulkner, Electrochemical Methods, Fundamentals
and Applications Wiley, New York, 1980.
[72] D. Brown, B. T. Davis, T. G. Halsall, A. R. Hands, J. Chem. Soc. 1962,
4492.
[88] W. Adam, H. J. Eggelte, J. Org. Chem. 1977, 42, 3987.
[89] R. F. Nystrom, W. G. Brown, J. Am. Chem. Soc 1948, 70, 3738.
[90] E. C. Horning in Synthesis of Azobenzene Vol. 3, Wiley, London, 1955,
p. 103.
[73] C. P. Andrieux, C. Blocman, J. M. Dumas-Bouchait, F. MꢁHalla, J.-M.
[91] W. McPherson, G. W. Stratton, J. Am. Chem. Soc. 1915, 37.
Â
Saveant, Electroanal. Chem. 1980, 113, 19.
Â
[74] C. P. Andrieux, C. Blocman, J. M. Dumas-Bouchait, J.-M. Saveant, J.
Am. Chem. Soc. 1979, 101, 3431.
Received: January 18, 2001
Â
[75] C. P. Andrieux, J.-M. Saveant, Electroanal. Chem. 1986, 205, 43.
Revised: May 21, 2001 [F3008]
4020
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0718-4020 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 18