Thermodynamics, kinetics, and dynamics of the two alternative aniomesolytic fragmentations of C-O bonds: An electrochemical and theoretical study

Article abstract of DOI:10.1021/ja012444g

Fragmentation reactions of radical anions (mesolytic cleavages) of cyanobenzyl alkyl ethers (intramolecular dissociative electron transfer, heterolytic cleavages) have been studied electrochemically. The intrinsic barriers for the processes have been esta

Full text of DOI:10.1021/ja012444g

Published on Web 04/03/2002
Thermodynamics, Kinetics, and Dynamics of the Two
Alternative Aniomesolytic Fragmentations of C-O Bonds: An
Electrochemical and Theoretical Study
Luisa Pisano,† Maria Farriol, Xavier Asensio, Iluminada Gallardo,
Angels Gonza´lez-Lafont, Jose´ M. Lluch,* and Jordi Marquet*
Contribution from the Departament de Qu´ımica, UniVersitat Auto`noma de Barcelona,
08193 Bellaterra, Barcelona, Spain
Received October 26, 2001
Abstract: Fragmentation reactions of radical anions (mesolytic cleavages) of cyanobenzyl alkyl ethers
(intramolecular dissociative electron transfer, heterolytic cleavages) have been studied electrochemically.
The intrinsic barriers for the processes have been established from the experimental thermodynamic and
kinetic parameters. These values are more than 3 kcal/mol lower as an average than the related homolytic
mesolytic fragmentations of radical anions of 4-cyanophenyl ethers. In the particular case of isomers
4-cyanobenzyl phenyl ether and 4-cyanophenyl benzyl ether, the difference in intrinsic barriers amounts to
5.5 kcal/mol, and this produces an energetic crossing where the thermodynamically more favorable process
(homolytic) is the kinetically slower one. The fundamental reasons for this behavior have been established
by means of theoretical calculations within the density functional theory framework, showing that, in this
case, the factors that determine the kinetics are clearly different (mainly present in the transition state)
from those that determine the thermodynamics and they are not related to the regioconservation of the
spin density (“spin regioconservation principle”). Our theoretical results reproduce quite well the experimental
energetic difference of barriers and demonstrate the main structural origin of the difference.
Scheme 1
1. Introduction
Unimolecular fragmentation of radical ions to yield radicals
and ions (mesolytic cleavages) constitutes an elementary step
of many electron-transfer-initiated processes of chemical and
biochemical interest.1 The rates of these reactions are usually
significantly faster than those observed for the homolytic
cleavage of the same bonds in neutral substrates.2,3 In aryl and
benzyl radical anions, especially those substituted with electron-
withdrawing groups, bond fragmentation requires that the
SOMO electron density be transferred from the π* system of
the aryl ring to the region between the two atoms of the scissile
bond. Some authors have considered mesolytic cleavages as
intramolecular electron transfers and have modeled their dynam-
ics in the framework of Marcus theory.4 The energy barriers
associated with mesolytic cleavages have been proposed to be
the result of electron redistribution and/or solvent reorganization,
allowing them to be used as experimental probes of the kinetics
and thermodynamics of these elementary reactions, especialy
the identification of the factors contributing to the intrinsic
barriers.5,6
In the fragmentation of radical ions wherein the unpaired
electron resides initially in a π orbital on one side of the scissile
bond, electron reapportionment can occur with transfer of charge
across the scissile bond (heterolytic mode, Scheme 1A), or the
charge may remain localized on the original moiety (homolytic
mode, Scheme 1B). Maslak et al. proposed the empirical “spin
regioconservation principle” from studies on aryl benzyl ethers7
* Corresponding authors: fax (+34)935811265; e-mail jordi.marquet@
uab.es.
† On leave from the Universita` di Sassari (Italy).
(1) (a) Savea´nt, J.-M. Tetrahedron 1994, 50, 10117. (b) Maslak, P. Top. Curr.
Chem. 1993, 168, 1. (c) Save´ant, J.-M. Acc. Chem. Res. 1993, 26, 455. (d)
Sancar, A. Biochemistry 1994, 33, 2.
(2) Chanon, M.; Rajzmann, M.; Chanon, F. Tetrahedron 1990, 46, 6193.
(3) (a) Maslak, P.; Narvaez, J. N.; Vallombroso, T. M., Jr. J. Am. Chem. Soc.
1995, 117, 12373. (b) Maslak, P.; Narvaez, J. N. Angew. Chem., Int. Ed.
Engl. 1990, 29, 283.
(4) (a) Marcus, R. A. J. Phys. Chem. 1965, 43, 679. (b) Kojima, H.; Bard, A.
J. J. Am. Chem. Soc. 1975, 97, 6317. (c) Eberson, L. Electron-Transfer
Reactions in Organic Chemistry; Springer: Berlin, 1987.
(5) (a) Save´ant, J.-M. Dissociative Electron Transfer. In AdVances in Electron-
Transfer Chemistry; Mariano, P. S., Ed.; JAI Press: New York, 1994; Vol.
4, pp 53-116. (b) Save´ant, J.-M. J. Am. Chem. Soc. 1987, 109, 6755. (c)
Save´ant, J.-M. J. Phys. Chem. 1994, 98, 3716.
(6) (a) Maslak, P. Top. Curr. Chem. 1993, 168, 1. (b) Maslak, P.; Vallombroso,
T. M., Jr.; Chapman, W. H., Jr.; Narvaez, J. N. Angew. Chem., Int. Ed.
Engl. 1994, 33, 73. (c) Maslak, P.; Narvaez, J. N.; Kula, J.; Malinski, D.
S. J. Org. Chem. 1990, 55, 4550.
(7) (a) Maslak, P.; Guthrie, R. D. J. Am. Chem. Soc. 1986, 108, 2628. (b)
Maslak, P.; Guthrie, R. D. J. Am. Chem. Soc. 1986, 108, 2637.
9
4708
J. AM. CHEM. SOC. 2002, 124, 4708-4715
10.1021/ja012444g CCC: $22.00 © 2002 American Chemical Society

Two Aniomesolytic Fragmentations of C−O Bonds
A R T I C L E S
and thioethers,8 arguing that an extra intrinsic barrier exists for
the homolytic cleavage, which does not regioconserve spin
density. The difference in reactivity between the two modes of
fragmentation has been attributed in the literature (on the basis
of studies on C-C and C-S bond fragmentations) to the
delocalization of charge in the transition state8,9 and to the
influence of solvent reorganization.10
Although there are extensive quantitative data on the mech-
anisms of carbon-halogen bond cleavage by electrochemical
means,11 and significant contributions to the mechanistic
knowledge on aniomesolytic C-C3,6, O-O9, and C-S8,12
cleavages, to our knowledge no reports exist in the literature
about the fundamental reasons for the very different behavior
of the homolytic and heterolytic modes of aniomesolytic
fragmentation for highly polarized strong bonds such as the
C-O bond in ethers (where first this distinct behavior was
experimentally observed). Indeed, even though preparative
useful examples of carbon-oxygen bond reductive cleavage in
ethers have been reported,13 very few mechanistic studies exist,
either chemical14 or electrochemical,15 and the available kinetic
data are very scarce. On the basis of a kinetic isotope effects
study on naphthyl ethers, the involvement of a π* transition
state in the heterolytic mesolytic cleavage, and of a σ* transition
state in the homolytic mesolytic cleavage, has been proposed.14b
Very recently some of us have described the thermodynamics
and kinetics of the homolytic cleavage of carbon-oxygen bonds
in anion radicals obtained by electrochemical reduction of
cyanophenyl ethers.10b We present here the thermodynamics and
kinetics of the heterolytic electroreductive cleavage of related
cyanobenzyl ethers and show that, for mesolysis of C-O bonds,
the intrinsic barriers of the heterolytic mode are substantially
lower (more than 3 kcal/mol) than those of the homolytic mode.
To get some insight on the fundamental reasons for the
differences in kinetic behavior between the homolytic and the
heterolytic mesolytic fragmentation modes, we have carried out
a theoretical study in the gas phase, within the framework of
density functional theory (DFT) for the 4-cyanophenyl benzyl
Figure 1. Voltammogram of 4-cyanobenzyl phenyl ether (1a) in DMF
(10 mM) and 0.1 M TBATFB, at 0.1 V/s. Glassy carbon disk electrode (3
mm diameter). Scan potential range: 0.00/-2.75/+1.50/0.00 V.
ether and 4-cyanobenzyl phenyl ether, isomers that only differ
structurally in the orientation of the C-O bond but that show
a very significant difference in intrinsic barrier for mesolytic
cleavage (Scheme 1C,D). Our theoretical results confirm the
main structural origin of the experimental energetic difference
of barriers.
It is very surprising that despite the importance of the
processes that involve the cleavage of formal three-electron
bonds in aromatic derivatives, very few theoretical studies have
been performed on them. A series of well-established empirical
rules, and the qualitative use of the “reactive mixed-valence
approach” are the customary tools in predicting the organic
reactivity in this particular field.2,4c,16,17 Theoretical molecular
orbital (MO) ab initio investigation of the reductive C-Cl bond
cleavage in nonsubstituted benzyl derivatives led to the conclu-
sion that the corresponding radical anions show a dissociative
behavior.18a However, recent density functional theory (DFT)
calculations reveal the formation of radical anions in these
systems.18b DFT has been applied also very recently to the study
of the photoenzymic repair mechanism that includes a step of
cleavage of a radical anion.19 As far as we know, DFT
methodology has not previously been applied to the study of
the reductive cleavage of highly polar and strong C-X bonds
such as C-O bonds.
(8) Maslak, P.; Theroff, J. J. Am. Chem. Soc. 1996, 118, 7235.
(9) (a) Donkers, R. L.; Maran, F.; Wayner, D. D. M.; Workentin, M. S. J. Am.
Chem. Soc. 1999, 121, 7239. (b) Sabrina, A.; Martin, M.; Wayner, D. D.
M.; Maran, F. J. Am. Chem. Soc. 1997, 119, 9541. (c) Workentin, M. S.;
Maran, F.; Wayner, D. D. M. J. Am. Chem. Soc. 1995, 117, 2120.
(10) (a) Andrieux, C. P.; Save´ant, J.-M.; Tallec, A.; Tardivel, R.; Tardy, C. J.
Am. Chem. Soc. 1996, 118, 9788. (b) Andrieux, C. P.; Farriol, M.; Gallardo,
I.; Marquet, J. J. Chem. Soc., Perkin Trans. 2, in press.
(11) (a) Save´ant, J.-M. Electron Transfer, Bond Breaking and Bond Formation.
In AdVances in Physical Organic Chemistry; Tidwell, T. T., Ed.; Academic
Press: New York, 2000; Vol. 35, p 117. (b) Andrieux, C. P. Organic
Electrochemical Mechanisms. In Encyclopedia of Analytical Chemistry;
Wiley: Chichester, U.K., 2000; p 9983.
2. Experimental Results
(12) (a) Screttas, C. G.; Screttas, M. M. J. Org. Chem. 1978, 43, 1064. (b)
Cohen, T.; Buphaty, M. Acc. Chem. Res. 1989, 22, 152. (c) Severin, M.
G.; Arevalo, M. C.; Farnia, G.; Vianello, E. J. Phys. Chem. 1987, 91, 466.
(13) (a) Maercker, A. Angew. Chem., Int. Ed. Engl. 1987, 26, 972 and references
therein. (b) Azzena, U.; Denurra, T.; Melloni, G.; Piroddi, A. M. J. Org.
Chem. 1990, 55, 5386. (c) Azzena, U.; Melloni, G.; Pisano, L.; Sechi, B.
Tetrahedron Lett. 1994, 35, 6759. (d) Marquet, J.; Cayo´n, E.; Mart´ın, X.;
Casado, F.; Gallardo, I.; Moreno, M.; Lluch, J. M. J. Org. Chem. 1995,
60, 3814.
(14) (a) Lazana, M. C. R. L. R.; Franco, M. L. T. M. B.; Herold, B. J. J. Am.
Chem. Soc. 1989, 111, 8640. (b) Guthrie, R. D.; Shi, B. J. Am. Chem. Soc.
1990, 112, 3136. (c) Cayo´n, E.; Marquet, J.; Lluch, J. M.; Martin, X. J.
Am. Chem. Soc. 1991, 113, 8970. (d) Azzena, U.; Denurra, T.; Melloni,
G.; Fenude, E.; Rassu, G. J. Org. Chem. 1992, 57, 1444. (e) Gonzalez-
Blanco, R.; Bourdelande, J. L.; Marquet, J. J. Org. Chem. 1997, 62, 6910.
(f) Saa´, J. M.; Ballester, P.; Deya´, P. M.; Capo´, M.; Garc´ıas, X. J. Org.
Chem. 1996, 61, 1035. (g) Casado, F.; Pisano, L.; Farriol, F.; Gallardo,
G.; Marquet, J.; Melloni, G. J. Org. Chem. 2000, 65, 322.
2.1. Electrochemical Measurements. Figure 1 shows a
typical voltammogram of 4-cyanobenzyl phenyl ether (1a), at
slow sweep rate (0.1 V s-1), in DMF. A bielectronic irreversible
reduction wave (Epc(I) ) -2.20 V) can be observed. In these
conditions, another reduction wave, reversible and monoelec-
(16) (a) Symons, M. C. R. Pure Appl. Chem. 1981, 53, 223. (b) Bunnett, J. F.;
Creary, X. J. Org. Chem. 1975, 40, 3740. (c) Villar, H.; Castro, E. A.;
Rossi, R. A. Can. J. Chem. 1982, 60, 2525.
(17) (a) Moreno, M.; Gallardo, I.; Bertra´n, J. J. Chem. Soc., Perkin Trans. 2
1989, 2017. (b) Martin, X.; Marquet, J., Lluch, J. M. J. Chem. Soc., Perkin
Trans. 2 1993, 87. (c) Pierini, A. B.; Duca, J. S., Jr. J. Chem. Soc., Perkin
Trans. 2 1995, 1821. (d) Voityuk, A. A.; Michel-Beyerle, M.-E.; Ro¨sch,
N. J. Am. Chem. Soc. 1996, 118, 9750. (e) Pierini, A. B., Duca, J. S., Jr.;
Vera, D. M. A. J. Chem. Soc., Perkin Trans. 2 1999, 1003.
(15) (a) Thornton, T. A.; Ross, G. A.; Patil, D.; Mukaida, K.; Warwick, J. O.;
Woolsley. N. F.; Bartak, D. E. J. Am. Chem. Soc. 1989, 111, 8640. (b)
Andersen, M. L.; Long, W.; Wayner, D. D. M. J. Am. Chem. Soc. 1997,
119, 6590.
(18) (a) Benassi, R.; Bertarini, C.; Taddei, F. J. Chem. Soc., Perkin Trans. 2
1997, 2263. (b) Dem’yanov, P. I.; Myshakin, E. M.; Boche, G.; Petrosyan,
V. S.; Alekseiko, L. N. J. Phys. Chem. A 1999, 103, 11469.
(19) Durbeej, B.; Eriksson, L. A. J. Am. Chem. Soc. 2000, 122, 10126.
9
J. AM. CHEM. SOC. VOL. 124, NO. 17, 2002 4709

A R T I C L E S
Pisano et al.
Scheme 2
1b, 40% of the starting material was recovered, and 60%
toluonitrile was obtained (methanol was not quantified). Analysis
of the electrolyzed solutions by electrochemical and chromato-
graphic techniques indicated that no cyanobenzyl alcohol was
formed (the only ether bond cleavage observed corresponds to
the cyanobenzyl ether bond). With different solvents (DMF,
ACN, or THF) and different salts (TBABF4, LiClO4) as
supporting electrolytes, no significant changes on the cleavage
rates, standard potentials, and nature and yields of products were
observed. Therefore, in these cases, the effect of solvation and
ion pairing seems negligible.
Table 1. Reactivity Data for the C-O Bond Aniomesolytic
Heterolytic Cleavage of Ethers
compound
D (eV) −E°RX/
E°X /X ∆G° (eV)
log kb
∆Go# (eV)
a
a
•
•-
RX
NCPhCH2OPh, 1a
2.23
2.15
2.15
0.24c -0.363 5.0 ( 0.5 0.655
NCPhCH2OCH3, 1b 2.996d
0.54e
0.013 1.0 ( 0.2 0.673
a Potentials are in volts vs SCE. b Values from digital simulations of
the cyclic voltammetry curves by DigiSim. c Value taken from ref 24a.
Electrochemical reduction results for the cyanobenzyl phenyl,
1a, and cyanobenzyl methyl, 1b, ethers studied are indicative
of a mechanism such as the one described in Scheme 2. The
radical anion resulting from single electron transfer to an
aromatic ether is a frangible species that decomposes readily.
In the radical anion, the unpaired electron must be located
initially in the cyanophenyl moiety of the molecule since its
standard potential is very close to that of benzonitrile. The
d Calculated from DNCPhCH OCH ) DNCPhCH OPh - DPhOH + DCH OH e Value
.
2
3
2
3
taken from ref 24b.
tronic, placed at potentials slightly more negative than the
precedent one (E°(II) ) -2.37 V) appears. This wave shows
the same features than the reduction wave of toluonitrile in
dimethylformamide (DMF) (the same potential, reversible at
any sweep rate, and monoelectronic), and therefore it is assigned
to the reduction of this compound, formed by reductive cleavage
of the alkyl ether bond (see the preparative electrolysis described
below). At positive potentials, and only at low sweep rates, it
is possible to observe after the reduction of 1a, a small oxidation
wave at 0.25 V. This anodic wave would correspond to the
oxidation of the phenoxide anion (0.24 V in acetonitrile).20
Voltammograms of 4-cyanobenzyl methyl ether (1b) are very
similar (Epc(I) ) -2.24 V, E°(II) ) -2.37 V). Analysis of the
first reduction peak intensity indicates the existence of a two-
electron process at low sweep rates and one-electron process at
high sweep rates (>2 × 104 V s-1 for 1a and >2 V s-1 for
1b).
cleavage of the bond leads to NCPhCH2 and RO- and thus
•
involves a “mesolytic heterolytic” dissociation (intramolecular
dissociative electron transfer) of the benzylic C-O bond in the
radical anion •-NCPhCH2OR. The final reduction of NCPhCH2
•
will probably take place in solution by another radical anion
due to the relatively low value of the cleavage rate constant
(DISP 1 mechanism).21,22
2.2. Thermodynamics and Kinetics of Mesolytic Hetero-
lytic Bond Cleavage in Cyanobenzyl Alkyl Ethers. We have
just demonstrated that the bond cleavage between the oxygen
and the benzylic carbon step in the reduction of 4-cyanobenzyl
phenyl ether (1a) and 4-cyanobenzyl methyl ether (1b) leads
to heterolytic dissociation:
The shape of voltammograms (peak width) suggests, in both
cases, a mixed kinetic control by electron transfer and a coupled
chemical reaction.21 The peak potential is not concentration-
dependent (in the range 1-10 mM) and the variation of the
peak potential with the scan rate is 30 mV by unit log scan rate
for low scan rates.21 Therefore, we can conclude that the initially
produced radical anion reacts following a slow first-order
reaction pathway leading to a second electron transfer following
a DISP 1 mechanism (Scheme 2).22 At a low sweep rate, reaction
1 of Scheme 2 is followed by reactions 2 and 3 (two electrons
altogether). At a fast sweep rate, only the reversible formation
of cyanobenzyl ether radical anion would be observed (reaction
1 of Scheme 2; one-electron reversible process). This conclusion
was confirmed by digital simulations of the cyclic voltammetry
curves, with DigiSim software, that allowed a reliable deter-
mination of the standard potentials for the formation of the
radical anions of 1a and 1b and their cleavage rate constants
(Table 1).
•
• -(NCPhCH2OR) f NCPhCH2 + RO-
(1)
To obtain the thermodynamic and kinetic details of this process
we follow the model for cleavage of radical anions described
by Save´ant5 in the framework of the Marcus theory. Thus, the
standard free energy, ∆G°, for the overall reaction (eq 1) is
∆G° ) D(RX) + E°(RX/RX•-) - E°(X•/X ) - T∆S
(2)
-
with R ) NCPhCH2 and X ) R′O in our case.
The activation free energy, ∆G# (in electronvolts), is calcu-
lated by the Eyring equation (eq 3), taking the preexponential
factor equal to 5 × 1012 s-1 1a
:
log k ) log A - (F/2.303RT)∆G#
(3)
The activation free energy, ∆G#, is related quadratically to the
standard free energy of the reaction, ∆G°:
Electrolysis of 4-cyanobenzyl phenyl ether (1a) and 4-cyano-
benzyl methyl ether (1b) were carried out until 2F, at a potential
slightly more negative than the peak potential for each one. In
the case of 1a, 25% of the starting material was recovered, and
24% toluonitrile and 40% phenol were obtained. In the case of
∆G# ) ∆G0# (1 + ∆G°/4∆G0 )
(4)
# 2
#
The term ∆G0 is the intrinsic barrier free energy.
The cleavage rate constant of the radical anions and the
standard potential for the formation of the radical anions of 1a
and 1b were derived from our cyclic voltammetric data by
digital simulation of the curves with DigiSim. The standard
redox potentials of the leaving anions were taken from the
literature20 (see Table 1). The value corresponding to the
(20) (a) Hapiot, P.; Pinson, J.; Yousfi, N. New J. Chem. 1992, 16, 877. (b)
Eberson, L. Acta Chem. Scand. 1984, B38, 439.
(21) (a) Nadjo, L. Save´ant, J.-M. J. Electroanal. Chem. 1973, 48, 113. (b)
Andrieux, C. P.; Save´ant, J.-M. Electrochemical Reactions. In InVestigation
of Rates and Mechanism of Reactions; Techniques of Chemistry; Ber-
nasconi, C. F., Ed.; Wiley: New York, 1986; Vol 6, Chapter 2.1, p 305.
(22) Amatore, C.; Save´ant, J.-M. J. Electroanal. Chem. 1980, 107, 353.
9
4710 J. AM. CHEM. SOC. VOL. 124, NO. 17, 2002

Two Aniomesolytic Fragmentations of C−O Bonds
A R T I C L E S
Table 2. Comparison of the Thermodynamic and Kinetic
Parameters for the Two Alternative C-O Bond Aniomesolytic
Fragmentations of Alkyl Ethersa
processes in gas phase. In this section we will first provide the
calculational details and then the theoretical results.
3.1. Calculational Details. The quantum mechanical calcula-
tions have been carried out within the framework of density
functional theory (DFT),26 which is reaching a widespread use
in the calculation of quite sizable organic and inorganic
molecules. The spin-unrestricted formalism has been used in
solving the Kohn-Sham DFT equations.27 The particular
functional used has been the Becke’s three-parameter hybrid
method with Lee, Yang, and Parr’s correlation functional
(B3LYP),28 nowadays one of the most-used functionals. The
basis set chosen for the calculations had to be flexible enough
to describe anionic species and, therefore, the split valence
6-31+G basis set,29 which includes a diffuse sp shell on the
heavy atoms, has been used. The diffuse functions provide more
ample space allowance for the additional electron in the radical
anions. It has been shown30 that the B3LYP method with a
moderately large basis set yields electronic spin density ratios
in good agreement with experiment.
type of cleavage
compound
∆G°
log k
∆G# ∆Go#
“heterolytic” NCPhCH2OPh, 1a
NCPhCH2OCH3, 1b
-10.45 5.0 ( 0.5 10.3 15.1
0.30 1.0 ( 0.2 15.7 15.5
-16.67 2.9 ( 0.2 13.1 20.6
-13.67 3.1 ( 0.2 12.9 19.1
-2.83 1.0 ( 0.2 15.7 17.2
“homolytic”b NCPhOCH2Ph, 2
NCPhOCH2CHCH2, 3
NCPhOCH3, 4
NCPhO(CH2)2CHCH2, 5 -2.37 0.7 ( 0.2 16.1 17.2
a Energies are given in kilocalories per mole; k is given in reciprocal
seconds. b Values from ref 10b.
variation of entropy in eq 2 has been taken as 1 meV/K.23 The
values used, and the results obtained, for the thermodynamic
and kinetic parameters for compounds 1a and 1b are shown in
Table 1.
2.3. Discussion of the Experimental Results: Mesolytic
Heterolytic vs Mesolytic Homolytic Fragmentations. As
indicated in the Introduction, some of us have very recently10b
described the thermodynamics and kinetics of the mesolytic
homolytic cleavage of a series of cyanophenyl alkyl ethers, and
it was our intention to compare these reported thermodynamic
and kinetic parameters with the ones described here for the
mesolytic heterolytic cleavage of cyanobenzyl methyl and
phenyl ethers. In Table 2, the thermodynamic and kinetic
parameters for the reductive cleavage of compounds 1a and 1b
are compared with the corresponding values for some selected
cyanophenyl alkyl ethers (compounds 2-5).10b From the table,
it comes that despite the variability of thermodynamic driving
forces for the studied reactions (spanning from ∆G° ) +0.30
to -16.67 kcal/mol), a relatively narrow range of intrinsic barrier
values is obtained for each series of compounds. Compounds
that show heterolytic mesolytic cleavages (1a and 1b) have a
lower intrinsic barrier (15.3 kcal/mol as an average) than the
ones that show homolytic mesolytic fragmentation (2, 3, 4, and
5; 18.53 kcal/mol as an average). This differential behavior
(more than 3 kcal/mol) is independent of the thermodynamics
and seems to be due to fundamental differences between both
types of cleavages. A very related behavior has been described
for C-S fragmentations.8
Full geometry optimization and direct location of stationary
points (minima and transition-state structures) have been carried
out by means of the Schlegel gradient optimization algorithm
by using redundant internal coordinates.31 Diagonalization of
the potential energy analytical second-derivative matrix (Hes-
sian) has been done to disclose the nature of the stationary point
of the potential energy surface: no negative eigenvalues indicate
a potential energy minimum, whereas one negative eigenvalue
identifies a transition-state structure. In this second case, the
eigenvector (transition vector) associated with the negative
eigenvalue shows the direction along which the potential energy
lowers. When it has been required, the minimum energy path
(MEP) has been calculated by following the Gonzalez-Schlegel
mass-weighted internal coordinates reaction-path algorithm.32
The MEP goes downhill from the transition-state structure to
the two minima at both sides of it. The MEP has been built up
with a step size of 0.02 bohr.
Thermodynamic magnitudes have been computed by using
the statistical thermodynamic formulation33 of partition functions
within the ideal gas, rigid rotor, and harmonic oscillator models.
A pressure of 1 atm and a temperature of 298.15 K have been
assumed in the calculations. The analytical second derivatives
of the potential energy with respect to the Cartesian coordinates
have been used for the determination of vibrational frequencies.
The imaginary frequency is neglected in the thermodynamic
evaluation for transition-state structures.
It is especially interesting to compare the behavior of
compounds 4-cyanobenzyl phenyl ether (1a) and 4-cyanophenyl
benzyl ether (2), which are isomers. The difference in intrinsic
barrier (5.5 kcal/mol in this case) provokes a crossing in the
reaction energetics, such that the thermodynamically more
spontaneous process (2, homolytic), by a difference of 6.2 kcal/
mol, is the slower one by more than 2 orders of magnitude.
The Gaussian 94 and 98 packages34,35 have been used to carry
out all these electronic calculations.
3. Theoretical Calculations
As has been shown above, the radical anion of the 4-cyano-
benzyl phenyl ether (1a) undergoes a faster C-O alkyl ether
bond cleavage than the radical anion of the 4-cyanophenyl
benzyl ether (2) in solution, despite a clear thermodynamic
advantage for the latter reaction. To understand this different
kinetic behavior and to assess the weight on that of the factors
just related to the radical anions themselves (that is, depending
on the solute but excluding the contribution of the solvent
reorganization), we have theoretically studied both C-O fission
(24) Maran, F.; Celadon, D.; Severin, M. G.; Vianello, E. J. Am. Chem. Soc.
1991, 113, 9320.
(25) Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456.
(26) Kohn, W.; Becke, A. D.; Parr, R. G. J. Phys. Chem. 1996, 100, 12974.
(27) (a) Hohenberg, P.; Kohn, W. Phys. ReV. B 1964, 136, 864. (b) Kohn, W.;
Sham, L. J. Phys. ReV. A 1965, 140, 1133.
(28) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (b) Becke, A. D. J. Chem.
Phys. 1996, 104, 1040. (c) Becke, A. D. In Modern Electronic Structure
Theory; Yarkony, D. R., Ed.; World Scientific: Singapore, 1995.
(29) Clark, T.; Chandrasekhar, J.; Spitznagel, G. W.; Shleyer, P. v. R. J. Comput.
Chem. 1983, 4, 294.
(30) Qin, Y.; Wheeler, R. A. J. Chem. Phys. 1995, 102, 1689.
(31) Peng, C.; Ayala, P. Y.; Schlegel, H. B.; Frisch, M. J. J. Comput. Chem.
1996, 17, 49.
(32) Gonzalez, C.; Schlegel, H. B. J. Phys. Chem. 1990, 94, 5523.
(33) McQuarrie, D. A. Statistical Thermodynamics; University Science Books:
Mill Valley, CA, 1973.
(23) Andrieux, C. P.; Save´ant, J.-M.; Tardy, C. J. Am. Chem. Soc. 1998, 120,
4167.
9
J. AM. CHEM. SOC. VOL. 124, NO. 17, 2002 4711

A R T I C L E S
Pisano et al.
Figure 3. Occupied π* molecular spin orbital of the radical anion of the
4-cyanophenyl benzyl ether (2).
molecular plane as indicated by the dihedral angles in Table 3,
and a π* molecular spin orbital (HSOMO) becomes occupied
(Figure 3).
From the geometrical point of view, the extra electron
shortens the scissile C-O bond from 1.460 Å at the neutral
molecule to 1.441 Å at the radical anion. Then, rather surpris-
ingly, we observe that the extra electron compresses the scissile
bond, which initially evolves in the opposite sense to what could
be expected at first glance.
Figure 2. Stationary points in the C-O alkyl ether bond fragmentation of
the radical anion of the 4-cyanophenyl benzyl ether (2) potential energy
surface.
Table 3. Main Geometrical Parameters of the Five Stationary
Points Located for the C-O Alkyl Ether Bond Fragmentation of the
Radical Anion of 4-Cyanophenyl Benzyl Ether (2), along with the
Corresponding Potential Energies and Gibbs Free Energiesa
c
d
C8O7b C5C4O7C8c C4O7C8C10 O7C8C10C9c
∆V
∆G°
Our theoretical results indicate that the electronic ground state
for the radical anion of the 4-cyanophenyl benzyl ether (2) is a
π* state with the extra electron concentrated on the cyanophenyl
part of the molecule and on the oxygen atom. This state is not
appropriate to produce the C-O alkyl ether bond fragmentation.
Then an intramolecular electron transfer from the π* ground
state to a suitable excited state has to occur prior to the
experimentally observed C-O bond cleavage. This will be a
σ* electronic excited state (at the reactant) that involves the
occupation of a σ* antibonding C-O alkyl ether molecular spin
orbital. Assuming a classical frame, the radiationless electron
transfer must take place at the intersection region of the diabatic
potential energy surfaces corresponding to the π* and σ*
diabatic states. Random thermal fluctuations in the nuclear
configurations of the radical anion in the π* state, involving
especially the lengthening of the C-O alkyl ether bond, occur
until that intersection region is reached; then the energies of
both diabatic states become equal and the electron jump happens.
After the intramolecular electron transfer the C-O bond, already
in the σ* state, dissociates directly. The appearance of the proper
fluctuations costs free energy. It is this free energy that
determines the rate of the fragmentation process. Likewise, the
classical energy barrier arises from the potential energy required
to deform the radical anion in the π* state up to the minimum
energy structure of the intersection region (the transition-state
structure).
To locate the transition-state structure we have stretched the
C-O alkyl ether bond of the radical anion of the 4-cyanophenyl
benzyl ether (2). For each value of the C-O distance, the rest
of the geometrical parameters of the molecule have been
optimized in order to minimize the energy. The energy grows
monotonically with the C-O distance, in such a way that no
maximum energy point (the transition-state structure) appears,
at least before a C-O distance of 1.85 Å. However, a careful
analysis of the calculations reveals that the set of geometrical
structures associated with that energy profile keeps the planarity
of the molecule. As a consequence, two kind of geometrical
distortions are required to achieve the fragmentation: the C-O
alkyl ether stretching and the twisting of the two rings away
from planarity. Taking all this into account, we have followed
reactant
intermediate 1.446
TS
complex
products
1.441
180.0
177.3
129.5
1.1
180.0
-78.5
-64.6
180.7
0.0
12.9
40.5
0.0
0.0
1.50
2.41
0.0
1.03
1.21
1.481
3.240
∞
-29.1 -33.4
-21.7 -34.3
a The bond distances, dihedral angles, and energies are given in
angstroms, degrees, and kilocalories per mole, respectively. The numbers
labeling the nuclei correspond to the ones in Figure 2. b Bond distances.
c Dihedral angles. d Potential energy.
3.2. Theoretical Results. To begin with, we will first focus
on the C-O alkyl ether bond fragmentation of the radical anion
of the 4-cyanophenyl benzyl ether (2). We have located five
stationary points on the corresponding potential energy surface
(Figure 2). With regard to the C-O cleavage, their more relevant
geometrical parameters are the scissile C-O bond distance and
the three dihedral angles that reflects the relative twisting of
the two aromatic rings. They are presented in columns 2-5 of
Table 3 (the numbers labeling the nuclei are indicated in the
first structure of Figure 2). The last two columns in Table 3
give the potential energy and the Gibbs free energy, respectively,
associated with each stationary point, the reactant structure being
always taken as the origin of energies.
When one electron is added to neutral 4-cyanophenyl benzyl
ether (2) to form the corresponding radical anion (the reactant
of the C-O fragmentation), the two aromatic rings keep on the
(34) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.; Johnson, B.
G.; Robb, M. A.; Cheeseman, J. R.; Keith, T. A.; Petersson, G. A.;
Montgomery, J. A.; Raghavachari, K.; Al-Laham, M. A.; Zakrzewski, V.
G.; Ortiz, J. V.; Foresman, J. B.; Cioslowski, J.; Stefanov, B. B.;
Nanayakkara, A.; Challacombe, M.; Peng, C. Y.; Ayala, P. Y.; Chen, W.;
Wong, M. W.; Andres, J. L.; Replogle, E. S.; Gomperts, R.; Martin, R. L.;
Fox, D. J.; Binkley, J. S.; Defrees, D. J.; Baker, J.; Stewart, J. P.; Head-
Gordon, M.; Gonzalez, C.; Pople, J. A. Gaussian 94; Gaussian: Pittsburgh,
PA, 1995.
(35) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M.
A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A.; Stratmann,
R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin,
K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi,
R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.;
Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick, D. K.;
Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz,
J. V.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Kamaromi, I.;
Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng,
C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.;
Johnson, B. G.; Chen, W.; Wong, M. W.; Andres, J. L.; Head-Gordon,
M.; Replogle, E. S.; Pople, J. A. Gaussian 98; Gaussian: Pittsburgh, PA,
1998.
9
4712 J. AM. CHEM. SOC. VOL. 124, NO. 17, 2002

Two Aniomesolytic Fragmentations of C−O Bonds
A R T I C L E S
the evolution (to reactant or to products) of some of the planar
structures when they are allowed to relax after some slight
deviation from planarity has been introduced. Direct location
of the transition state from the planar structure with a scissile
C-O bond distance of 1.55 Å led to the structure whose main
geometrical parameters are given in the third row of Table 3.
At this transition-state structure the C-O bond becomes 0.04
Å longer than in the reactant. On the other hand, the molecule
is now clearly far from planarity (see Figure 2). In particular,
the C5C4O7C8 dihedral angle (129.5°) carries the scissile C8-O7
bond to a position orthogonal to, rather than coplanar with, the
cyanophenyl ring, in this way trying to maximize the overlap
between the π system of the cyanophenyl ring and the σ system
of the C-O alkyl ether bond. As a result of all these distortions,
the classical energy barrier turns out to be 2.41 kcal/mol.
The transition vector (that is, the eigenvector of the Hessian
associated with the unique negative eigenvalue that, as a
consequence, indicates the direction in which the potential
energy goes downhill) turns out to be a mixing of the variation
of the C8-O7 bond distance and the twisting of the dihedral
angles, showing the kind of nuclear motions that make possible
the C-O fragmentation.
Figure 4. NPA net atomic charges (in au) for neutral 4-cyanophenyl benzyl
ether (2), the corresponding radical anion, the intermediate, and the transition
state of the C-O fragmentation and NPA spin distribution (in au) for the
radical anion of the 4-cyanophenyl benzyl ether (2), the intermediate, and
the transition state of the C-O fragmentation.
From the transition-state structure the MEP toward the
reactant leads to a minimum energy structure (Figure 2) that is
an intermediate of the C-O scission and whose main geo-
metrical parameters are given in the second row in Table 3. At
this intermediate the cleavage of the C-O alkyl ether bond has
not begun yet; its main geometrical difference with regard to
the reactant consists of the change of the C4O7C8C10 dihedral
angle, which raises the benzyl ring well above the plane of the
rest of the molecule. As a consequence, the intermediate appears
only 1.5 kcal/mol above the reactant. Conversely, the MEP
toward the products does not converge, probably due to the jump
of the molecule between the two electronic states, which are
very close near the transition-state structure. Anyway, if the
molecule is allowed to relax after the C8-O7 bond has been
lengthened 0.1 Å, the C-O scission takes place, reaching a
complex (fourth row in Table 3, Figure 2), 29.1 kcal/mol below
the reactants, composed of two fragments: the cyanophenolate
anion and the benzyl radical. In this complex the C8-O7 alkyl
ether bond is already broken (3.240 Å). On going to the final
separated products, the potential energy rises 7.4 kcal/mol.
Figure 5. Occupied π* molecular spin orbital of the radical anion of the
4-cyanobenzyl phenyl ether (1a).
the difference between the R and â NPA net atomic charges) is
also given. It can be seen that the additional negative charge in
the radical anion lies for the most part on the cyanophenyl atoms
and the oxygen atoms and that, on going from the planar radical
anion to the transition state, a transfer of charge from the phenyl
rings to the scissile bond atoms (C-O) is observed. This is even
clearer considering the spin distribution change. In this case,
the spin density at the benzyl carbon increases significantly,
and this agrees with the proposed population of the σ* spin
molecular orbital at the transition state. It is well-known that in
a C-O σ bond, the C atom provides the most important
contribution to the antibonding σ* molecular orbital. Once the
cleavage has occurred, in the hydrogen-bonded complex the
benzyl radical is practically neutral (the total NPA net charge
is 0.03 au), but it contains already the whole unpaired electron
(the total NPA spin distribution is 1.00 au).
Let us turn our attention to the C-O alkyl ether bond
fragmentation of the radical anion of the 4-cyanobenzyl phenyl
ether (1a). The occupied π* molecular spin orbital in the radical
anion, which is also planar, is depicted in Figure 5. By
comparison of Figures 3 and 5, it can be noted that the position
of the cyano substituent entirely determines in which ring the
HSOMO concentrates.
Let us describe in short the movements during the fragmenta-
tion of the radical anion of the 4-cyanophenyl benzyl ether. First,
from the planar reactant to the intermediate, the scissile C8-
O7 bond does not lengthen and it remains coplanar with the
cyanophenyl ring, whereas the benzyl ring rises up to an almost
perpendicular conformation. At the transition-state structure the
C8-O7 bond becomes slightly stretched and it goes far from
the plane of the cyanophenyl ring in order to permit the HSOMO
to have a significant C-O σ* contribution, which will lead to
the C-O alkyl ether scission. In Figure 4, the evolution of the
net atomic charges along the reaction coordinate is described.
These charges derive from the natural atomic orbitals and are
obtained from a natural population analysis (NPA) according
to the procedure developed by Weinhold and co-workers.36 In
Figure 4, the evolution of the NPA spin distribution (that is,
(36) (a) Reed, A. E.; Weinstock, R. B.; Weinhold, F. J. Chem. Phys. 1985, 83,
735. (b) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. ReV. 1988, 88
(8), 899.
From the geometrical point of view, the extra electron
lengthens the scissile C-O bond from 1.451 Å at the neutral
9
J. AM. CHEM. SOC. VOL. 124, NO. 17, 2002 4713

A R T I C L E S
Pisano et al.
provides the most important contribution to an antibonding
molecular orbital. Then, what is probably happening in the case
of 1a is that when the electron is captured, the scissile C-O
bond in the planar radical anion is weakened by increasing the
C-O antibonding contribution in the occupied molecular
orbitals (spin density in the C atom). As explained above, the
molecule dissociates directly when the σ* spin molecular orbital
of the C-O alkyl ether bond becomes populated. The planar
radical anion is just placed at the intersection region, thus, a
slight deviation of planarity already introduces components of
the antibonding C-O interaction in the σ* spin molecular
orbital, and the molecule fragments directly. Therefore, the
radical anion of 1a shows purely dissociative behavior, with
no Gibbs free energy barrier.
Figure 6. NPA net atomic charges (in au) for neutral 4-cyanobenzyl phenyl
ether (1a) and the corresponding radical anion and NPA spin distribution
(in au) for the radical anion of the 4-cyanobenzyl phenyl ether (1a).
In the case of the cyanophenyl benzyl ether (2), when the
electron is captured, the scissile C-O bond in the planar radical
anion is strengthened by the increased C-O bonding contribu-
tion in the occupied molecular orbitals (spin density in the O
atom). To achieve the cleavage, the σ* spin molecular orbital
of the C-O alkyl ether bond must be populated, and this would
occur at the intersection region, provided that some coupling
takes place between the two diabatic states. The scissile C-O
bond stretching moves the molecule toward the intersection
region, and the dihedral twisting away from planarity provides
the coupling. The radical anion of the cyanophenyl benzyl ether
(2) needs to lengthen the scissile C8-O7 bond from 1.441 Å
up to 1.481 Å to reach that intersection region. Then, a Gibbs
free energy barrier of 1.21 kcal/mol is required to reach the
transition state.
molecule to 1.469 Å at the radical anion. The analysis of the
eigenvalues of the Hessian of the radical anion of the 4-cyano-
benzyl phenyl ether provides a rather surprising conclusion: this
planar structure is not a minimum energy structure but a
transition-state structure. The transition vector is a mixture of
the twisting of the dihedral angles that breaks the planarity of
the molecule, but it does not contain any component corre-
sponding to the variation of the scissile C-O bond distance. In
other words, this radical anion is a minimum along the direction
corresponding to the C-O bond internal coordinate, but a
maximum along the direction that twists the two rings away
from planarity. When the geometry is displaced some few
degrees along this last direction and then it is allowed to relax,
the C-O bond fragments and a complex 13.1 kcal/mol below
the transition-state structure is reached. This complex consists
of two fragments, the cyanobenzyl radical (the total NPA net
charge is still -0.22 au, with a total NPA spin distribution of
0.80 au) and the phenolate anion, with two hydrogen bonds
between the oxygen atom and the hydrogen atoms 15 and 16
(the two hydrogen bond lengths are 2.59 Å). The C-O alkyl
ether bond length is already 2.65 Å. Separation of the two
fragments to reach the products costs 12.5 kcal/mol in terms of
potential energy or 3.9 kcal/mol in terms of Gibbs free energy
(the difference of magnitude is due to the entropic term, which
favors the formation of fragments).
Why is the intersection region energetically more difficult
to reach in the case of 2 (homolytic mesolytic cleavage) than
in the case of 1a (heterolytic mesolytic cleavage)? The additional
unpaired electron is mainly located in the cyanophenyl moiety
of the molecule, and whatever migration of the electronic density
toward the scissile C-O σ* region requires potential energy,
the farther away from the cyanophenyl ring the migration, the
bigger the increment of potential energy. Considering the C atom
provides the most important contribution to an antibonding σ*
molecular orbital, the migration of the unpaired electron to
populate the C-O σ* region (which is located essentially around
the C atom) in the case of 2 happens to a somewhat farther
away region than in the case of 1a, where the C atom is directly
attached to the cyanophenyl ring. In addition, our calculations
(Figure 6) indicate that for 1a the transition state can be
reasonably represented as [•-NCArCH2+-OPh]. Therefore, in
this case, and according to Guthrie and Shi,14b the bond-
polarized contributions to the transition state can make the
intersection region easier to reach by allowing the extra electron
to remain more localized in the π system than in the case of 2.
In Figure 6, the net NPA atomic charges for the neutral
4-cyanobenzyl phenyl ether (1a) and its radical anion and the
NPA spin distribution for the radical anion are shown. In
agreement with the finding that the planar radical anion is
already a transition state that leads to cleavage of the scissile
bond upon a slight twisting of the structure, the benzylic carbon
shows a significant spin density (0.043 au; compare with the
transition state of compound 2, Figure 4).
Let us rationalize now the different behavior of the two radical
anions. The first striking difference is the fact that the uptake
of one electron (planar radical anion) causes a shortening of
the C-O scissile bond in the case of cyanophenyl benzyl ether
(2) and the oposite effect in cyanobenzyl phenyl ether (1a).
Some hints come from the data of spin distribution gathered in
Figures 4 and 6. Thus, in both cases, the planar radical anion
shows a significant spin density in the atom directly linked to
the cyanophenyl ring (where the main part of the extra electronic
density is located). On the other hand, it is well-known that in
a C-O bond, the main contribution to a bonding molecular
orbital comes from the oxygen atom, whereas the C atom
4. General Discussion
Our electrochemical studies indicate that heterolytic mesolytic
fragmentations of radical anions of 4-cyanobenzyl ethers have
lower intrinsic barriers (more than 3 kcal/mol on average) than
the related homolytic mesolytic fragmentations of radical anions
of 4-cyanophenyl ethers. In the particular case of isomers
4-cyanobenzyl phenyl ether (1a) and 4-cyanophenyl benzyl ether
(2), this difference (∆∆G0 ) amounts to 5.5 kcal/mol and
produces an interesting energetic crossing since the thermody-
namically more favorable process (cleavage of the radical anion
#
9
4714 J. AM. CHEM. SOC. VOL. 124, NO. 17, 2002

Two Aniomesolytic Fragmentations of C−O Bonds
A R T I C L E S
of product 2, homolytic mesolytic cleavage, ∆∆G° ) -6.2 kcal/
mol) is the kinetically slower one (∆∆G# ) 2.8 kcal/mol). The
fundamental reasons for this behavior have been established by
means of theoretical calculations.
atom is directly attached to the cyanophenyl ring (because this
scission will lead to the cyanophenolate), but it favors kinetically
the cleavage of the radical anion in which the carbon atom,
which has the most important contribution to the σ* C-O alkyl
ether bond, is directly attached to the cyanophenyl ring.
Concerning the “spin regioconservation principle” (see Intro-
duction), our results show that it can work as a mnemonic rule
but that the real reasons for it to work are not related to the
conservation of the spin density.
Our results suggest that the kinetic preference for heterolytic
mesolytic fragmentation will be higher the more polarized the
scissile bond is, and that the “spin regioconservation principle”
will only hold for structures with highly polarized scissile bonds,
where the kinetic advantage for the heterolytic mesolytic
cleavage is able to overcome the thermodynamic tendency
(homolytic mesolytic cleavage).
Even taking into account that the theoretical study has been
performed in the gas phase, we have to underline that our
theoretical calculations qualitatively agree and explain the
experimental results in solution. Theoretically, both C-O alkyl
ether cleavages turn out to be exergonic (∆G° < 0), with a
clear thermodynamic advantage for the fragmentation of the
radical anion of 4-cyanophenyl benzyl ether (2). Here also, the
radical anion of 4-cyanobenzyl phenyl ether (1a) fragments
faster than the radical anion of 4-cyanophenyl benzyl ether (2).
The agreement between theory and experiment is qualitatively
quite good when we compare the differences in Gibbs free
energy barriers between 2 and 1a (∆∆G#): 2.8 kcal/mol from
experiment and 1.21 kcal/mol from theoretical calculation
(assuming that the Gibbs free energy barrier for the fragmenta-
tion of the radical anion of 1a is zero). Indeed, this indicates
that any systematic error in the theoretical methodology has
been canceled in the comparison of two very similar substrates
(isomers). Certainly, the individual Gibbs free energy barriers
in solution are significantly larger than what we obtain in the
gas phase. However, the fact that ∆∆G# is very similar in the
gas phase (theoretical calculations) and in solution (experiment)
indicates that, in this particular case, the rate differences between
the two fragmentation modes (and therefore the differences in
intrinsic barriers) arise essentially from the fundamental features
of the radical anion structures.
6. Experimental Section
Chemicals. All chemicals were purchased from Aldrich and were
of the highest purity available. They were used as received. 4-Cyano-
benzyl phenyl ether37 (1a) and 4-cyanobenzyl methyl ether38 (1b) were
prepared and identified following previously described procedures.39
Instruments and Procedures. Cyclic voltammetry and electrolysis
instruments and procedures have been previously described.40 The
electrochemical experiments were carried out at 20 °C. Digital
simulations of the cyclic voltammetry curves were performed by using
the DigiSim 2.0 software by Bioanalytical Systems Inc. The products
from preparative electrolyses, toluonitrile, and phenol, were identified
by comparison with commercial samples and the reactions were
quantified by gas chromatography.
Acknowledgment. Dedicated to Professor Marcial Moreno-
Man˜as on the occasion of his 60th birthday. Financial support
from DGI (MCyT of Spain) through Project BQU2000-0336,
from DGESIC through Project PB98-0915, and from “Gener-
alitat de Catalunya” through Project 1999SGR00090 is gratefully
acknowledged. The use of computational facilities of CESCA
and the CEPBA coordinated by the C4 is also gratefully
acknowledged.
5. Conclusions
We have demonstrated that the clear thermodynamic advan-
tage of the fragmentation of the radical anion of 2, both in gas
phase and in solution, is due to the fact that the factors that
determine the kinetics of the aniomesolytic fragmentations are
clearly different from those that control their thermodynamics.
We have shown that kinetics depends on the features of the σ*
C-O alkyl ether region, which have no influence on the
properties of the reactants or products. However, thermodynam-
ics rather depends on the stabilizing effect of the electron-
acceptor cyano substituent at the products, which indeed is
clearly larger in the negatively charged cyanophenolate (one
of the fragments coming from 2) than in the neutral cyanobenzyl
radical (one of the fragments resulting from 1a). In other words,
the introduction of the cyano substituent favors thermodynami-
cally the fragmentation of the radical anion in which the oxygen
JA012444G
(37) (a) Penn, J. H.; Lin, Z. J. Org. Chem. 1990, 55, 1554. (b) Wagner, G.;
Horn, H. Pharmazie 1975, 30, 353.
(38) (a) Tzeng, D.; Weber, W. P. J. Org. Chem. 1981, 46, 265. (b) Eliel, E. L.;
Badding, V. G.; Rerick, M. N. J. Am. Chem. Soc. 1962, 84, 2371.
(39) Kornblum, N.; Lurie, A. P. J. Am. Chem. Soc. 1959, 81, 2705.
(40) (a) Andrieux, C. P.; Larumbe, D.; Gallardo, I. J. Electroanal. Chem. 1991,
304, 241. (b) Andrieux, C. P.; Batlle, A.; Esp´ın, M.; Gallardo, I.; Jiang,
Z.; Marquet, J. Tetrahedron, 1994, 50, 6913. (c) Gallardo, I.; Guirado, G.,
Marquet, J. J. Electroanal. Chem. 2000, 488, 64. (d) Gallardo, I.; Guirado,
G.; Marquet, J. Chem. Eur. J. 2001, 7, 1759.
9
J. AM. CHEM. SOC. VOL. 124, NO. 17, 2002 4715