Intramolecular, intermolecular, and heterogeneous nonadiabatic dissociative electron transfer to peresters

Article abstract of DOI:10.1021/ja010799u

The electron transfer to peresters was studied by electrochemical means in N,N-dimethylformamide. The reduction was carried out by three independent methods: (i) heterogeneously, by using glassy carbon electrodes, (ii) homogeneously, by using electrogenerated radical anions as the donors, and (iii) intramolecularly, by using purposely synthesized donor-spacer-acceptor (D-Sp-A) systems. Convolution analysis of the heterogeneous data led to results in excellent agreement with the dissociative electron transfer theory. The homogeneous redox catalysis data also confirmed the reduction mechanism. The cyclic voltammetries of the D-Sp-A molecules could be simulated, leading to determination of the corresponding intramolecular dissociative rate constants. Analysis of the results showed that, regardless of the way by which the acceptor is reduced, the investigated dissociative electron transfers are strongly nonadiabatic and, particularly, that the experimental rates are several orders of magnitude smaller than the adiabatic limit. A possible mechanism responsible for the observed behavior is discussed.

Full text of DOI:10.1021/ja010799u

J. Am. Chem. Soc. 2001, 123, 9577-9584
9577
Intramolecular, Intermolecular, and Heterogeneous Nonadiabatic
Dissociative Electron Transfer to Peresters
Sabrina Antonello,† Fernando Formaggio,‡ Alessandro Moretto,‡ Claudio Toniolo,‡ and
Flavio Maran*,†
Contribution from the Department of Physical Chemistry, UniVersity of PadoVa, Via Loredan 2,
35131 PadoVa, Italy, and the Biopolymer Research Center, C.N.R., Department of Organic Chemistry,
UniVersity of PadoVa, Via Marzolo 1, 35131 PadoVa, Italy
ReceiVed March 28, 2001
Abstract: The electron transfer to peresters was studied by electrochemical means in N,N-dimethylformamide.
The reduction was carried out by three independent methods: (i) heterogeneously, by using glassy carbon
electrodes, (ii) homogeneously, by using electrogenerated radical anions as the donors, and (iii) intramolecularly,
by using purposely synthesized donor-spacer-acceptor (D-Sp-A) systems. Convolution analysis of the
heterogeneous data led to results in excellent agreement with the dissociative electron transfer theory. The
homogeneous redox catalysis data also confirmed the reduction mechanism. The cyclic voltammetries of the
D-Sp-A molecules could be simulated, leading to determination of the corresponding intramolecular dissociative
rate constants. Analysis of the results showed that, regardless of the way by which the acceptor is reduced, the
investigated dissociative electron transfers are strongly nonadiabatic and, particularly, that the experimental
rates are several orders of magnitude smaller than the adiabatic limit. A possible mechanism responsible for
the observed behavior is discussed.
In the last 15 years the study of intramolecular electron
transfer (ET) reactions in D-Sp-A molecules, in which a donor
(D) and an acceptor (A) are separated by a molecular spacer
(Sp), has provided a variety of information on how electrons
are transferred through bonds and space.1 Conversely, almost
nothing is known about dissociative electron transfers (DETs),
i.e., those reactions in which ET and bond cleavage are
concerted.2-4 Indeed, there is a relatively large amount of studies
on the decay of radical anions occurring by fragmentation of a
σ bond.2,4,5 In most cases, however, the mixing of the π* (donor
side) and σ* (acceptor side) orbitals is so strong that analysis
of the rate data, e.g., in terms of the relevant thermodynamic
parameters, may require several approximations. In fact, it has
been argued that describing the reductive cleavage of these
systems in terms of electron uptake followed by intramolecular
transfer from the π* orbital to the σ* orbital is not a realistic
model.6 Generally speaking, the study of intramolecular ET
processes, whether dissociative or not, is best performed by
using systems in which the D and A functional moieties are
spatially separated. Very recently, as a first step in this direction,
we reported data on the ∆G° dependence of intramolecular
DETs, using systems in which A was C-Br, D was a series of
ring-substituted benzoates, and Sp was cyclohexyl.7 The in-
tramolecular rate constants were found to be more sensitive to
variation of ∆G° than the corresponding intermolecular data.
This was attributed to the substituent-dependent variation of the
effective distance between the orbitals involved in the transfer.
Concerning the distance effect caused by variation of the
spacer’s length, we expect the rate to decrease with distance
along similar lines as reported for a variety of nondissociative
systems. This is because of reduced electronic coupling between
the reactant and product states, leading to nonadiabatic pro-
cesses. In fact, there are some interesting results on the
dissociative electron attachment to chloronorbornenes in which
the efficiency of chloride ion production was related to the
electronic coupling between localized CsCl σ* and CdC π*
orbitals.8 Some data on the intramolecular DET in D-Sp-A
systems in which Sp is a variable-length alkyl chain have also
been reported.9 The absence of rigidity in the investigated
molecules appears to be responsible for the small distance effect
on the intramolecular DET rates. The role of orbital symmetry
restrictions on the efficiency of π*/σ* coupling has been stressed
for the intramolecular reduction of halides.8a,10 Besides distance
or symmetry effects, however, we obtained some evidence
* Corresponding author: (tel) +39 (049) 827-5147; (fax) +39 (049) 827-
5135; (e-mail) f.maran@chfi.unipd.it.
† Department of Physical Chemistry.
‡ Department of Organic Chemistry.
(1) For example, see: (a) Closs, G. L.; Miller, J. R. Science 1988, 240,
440. (b) Paddon-Row, M. N. Acc. Chem. Res. 1994, 27, 18. (c) Electron
Transfer-From Isolated Molecules to Biomolecules; Jortner, J., Bixon, M.,
Eds.; Wiley: New York, 1999; Part 1.
(2) (a) Save´ant, J.-M. In AdVances in Electron-Transfer Chemistry;
Mariano, P. S., Ed.; JAI Press: Greenwich, CT, 1994; Vol. 4, p 53. (b)
Save´ant, J.-M. AdV. Phys. Org. Chem. 2000, 35, 117.
(3) Eberson, L. Acta Chem. Scand. 1999, 53, 751.
(4) Maran, F.; Wayner, D. D. M.; Workentin, M. S. AdV. Phys. Org.
Chem. 2001, 36, in press.
(5) (a) Maslak, P. In Topics in Current Chemistry; Mattay, J., Ed.;
Springer-Verlag: Berlin, 1993; Vol. 168, p 1. (b) Save´ant, J.-M. J. Phys.
Chem. 1994, 98, 3716.
(7) Antonello, S.; Maran, F. J. Am. Chem. Soc. 1998, 120, 5713.
(8) (a) Pearl, D. M.; Burrow, P. D.; Nash, J. J.; Morrison, H.; Jordan, K.
D. J. Am. Chem. Soc. 1993, 115, 9876. (b) Pearl, D. M.; Burrow, P. D.;
Nash, J. J.; Morrison, H.; Nachtigallova, D.; Jordan, K. D. J. Phys. Chem.
1995, 99, 12379.
(9) (a) Kimura, N.; Takamuku, S. Bull. Chem. Soc. Jpn. 1991, 64, 2433.
(b) Kimura, N.; Takamuku, S. Bull. Chem. Soc. Jpn. 1992, 65, 1668. (c)
Kimura, N.; Takamuku, S. J. Am. Chem. Soc. 1995, 116, 4087. (d) Kimura,
N.; Takamuku, S. J. Am. Chem. Soc. 1995, 117, 8023.
(6) Burrow, P. D.; Gallup, G. A.; Fabrikant, I. I.; Jordan, K. D. Austr. J.
Phys. 1996, 49, 403.
(10) Addock, W.; Andrieux, C. P.; Clark, C. I.; Neudeck, A.; Save´ant,
J.-M.; Tardy, C. J. Am. Chem. Soc. 1995, 117, 8285.
10.1021/ja010799u CCC: $20.00 © 2001 American Chemical Society
Published on Web 09/01/2001

9578 J. Am. Chem. Soc., Vol. 123, No. 39, 2001
Antonello et al.
indicating that there can be kinetically slow intermolecular
DETs, the homogeneous ET to the O-O bond of dialkyl
peroxides,11 for which poor electronic coupling is attributable
to the nature of the acceptor itself.
We are currently studying the distance dependence of DETs
and more generally the problem of nonadiabaticity by using
different types of spacer to control the electronic interaction
between the D and A redox moieties. Thus, we devised the
D-Sp-A systems 1a,b in which A is the dissociative-type tert-
Figure 1. Background-subtracted cyclic voltammetries for the reduction
of 2.1 mM 2a in DMF-0.1 M TBAP at 25 °C. Left to right: 0.1, 0.2,
0.5, 1, 2, 5, 10, 20 V s-1
.
was thus carried out to study in detail the kinetics of the
heterogeneous reduction.4,12,13,15 The i-E curves were trans-
formed into the corresponding convolution current (I) versus E
curves, leading to plots of excellent quality (independence,
within 1-2%, of the limiting value of I on V).16 In the
convolution analysis, the i and I values are then combined to
obtain the potential dependence of both the heterogeneous rate
constant, khet, and the transfer coefficient R.17 By applying the
basic concepts of the Save´ant DET model,18 the standard
potential (E° ) of the dissociative reduction of the perester was
estimated by calculating the value of E corresponding to R )
0.5. This E° value was then used to calculate the standard rate
constant, k°het. The intrinsic barrier of the DET (∆G0*) was
obtained from the slope of the R-E plot.13,18 The consistency
of the E° and k°het values was finally checked by digital
simulation of the cyclic voltammograms, which led to good
reproduction of the experimental curves. For 2a, the values of
E° , k°het, and ∆G0* are -0.30 V, 1 × 10-10 cm s-1, and 12.2
kcal mol-1, respectively. The corresponding values of 2b are
very similar, being -0.24 V, 9 × 10-11 cm s-1, and 13.3 kcal
butyl perester function, CO2-OtBu,12 and Sp is a rigid molecular
framework, CMe2 and 1,4-cyclohexanediyl. In this paper, we
describe the results that we obtained by studying the redox
properties of 1a,b and of the related molecules 2a,b and 3a,b.
Data on the distance effect on the intramolecular DET rate are
provided. Rate data are reported also for the heterogeneous and
intermolecular reduction of the perester O-O bond. Analysis
of the data led to discovery that regardless of the way by which
the reaction is carried out, ET to the perester acceptor is an
exceptionally slow DET.
mol-1
.
To check the actual reduction mechanism, we compared the
results obtained with 2a with those predicted by applying the
DET theory.17 The E° of the concerted DET can be conveniently
expressed through a thermochemical cycle13 which, for the pre-
sent case, is expressed as E° ) E°tBuCOO•/tBuCOO- - BDFE/F. BDFE
is the bond dissociation free energy of the peroxide O-O
bond. The E° of the leaving group, E°tBuCOO•/tBuCOO-, was
estimated to be 0.82 V by analysis and simulation of the
Results and Discussion
Electroreduction of 2 and 3 and Analysis of the DET
Mechanism. Models of both the acceptor (2a,b) and donor
moieties (3a,b) were used to analyze, independently, their redox
behavior both at the electrode and in the bulk. The experiments
were carried out in DMF/0.1 M Bu4NClO4 (TBAP) at 25 °C.
The electrode reduction of model peresters 2a and 2b was
studied by cyclic voltammetry. Both reductions are characterized
by an irreversible and broad peak (at 0.2 V s-1, the peak
potential, Ep, values of 2a and 2b are -1.55 and -1.50 V,
respectively), in agreement with the voltammetric behavior
expected for heterogeneous processes controlled by the activa-
tion barrier. The dependence of the normalized peak current,
ip/V1/2, on the scan rate V is illustrated for 2a in Figure 1. The
decrease of ip /V1/2 together with the fact that the peak broadens
when V increases indicate that the electron-transfer coefficient
R depends on the applied potential E.12-14 Convolution analysis
(15) (a) Imbeaux, J. C.; Save´ant, J.-M. J. Electroanal. Chem. 1973, 44,
169. (b) Save´ant, J.-M.; Tessier, D. J. Electroanal. Chem. 1975, 65, 57.
(16) I and i are related by the convolution integral15
i(u)
(t - u)1/2
t
I ) π-1/2
du
∫
0
The sigmoidal I - E curves are characterized by a diffusion-controlled
limiting value I1, ) nFAD1/2C*, where n is the overall electron consumption,
A the electrode area, D the diffusion coefficient, and C* the substrate
concentration. Provided the electrode process is irreversible, I and i are
related to khet through equation
ln khet ) ln D1/2 - ln {[I1 - I(t)]/i(t)}
(11) Donkers, R. L.; Maran, F.; Wayner, D. D. M.; Workentin, M. S. J.
Am. Chem. Soc. 1999, 121, 7239.
(12) (a) Antonello, S.; Maran, F. J. Am. Chem. Soc. 1997, 119, 12595.
(b) Antonello, S.; Maran, F. J. Am. Chem. Soc. 1999, 121, 9668.
(13) Antonello, S.; Musumeci, M.; Wayner, D. D. M.; Maran, F. J. Am.
Chem. Soc. 1997, 119, 9541.
Being R ) -(RT/F)∂lnkhet/∂E, the apparent value of the transfer coefficient
R can be experimentally determined as a function of E by derivatization of
the lnkhet/D1/2 - E plots.
(17) Save´ant, J.-M. J. Am. Chem. Soc. 1987, 109, 6788.
(14) Conventional analysis of electrochemically irreversible peaks should
provide V-independent ip/V1/2 and peak width (∆Ep/2) values (For example,
see: Bard, A. J.; Faulkner, L. R. Electrochemical Methods, Fundamentals
and Applications, 2nd ed.; Wiley: New York, 2001.). This is because the
Butler-Volmer description of electrode kinetics is employed, which implies
that the transfer coefficient R is constant at any E values. As previously
discussed in detail,12,13,15b the nonconstancy of R with respect to E fully
accounts for experimental trends such as those here described.
(18) (a) The DET model17 leads to the same relationship between the
activation free energy ∆G* and free energy ∆G° obtained by Marcus for
outer-sphere ETs (For example, see: Marcus, R. A.; Sutin, N. Biochim.
Biophys. Acta 1985, 811, 265.); i.e., ∆G* ) ∆G0* [1 + ∆G°/4∆G0*]2. (b)
Under electrochemical conditions and by neglecting the double-layer effect,
the rate/driving force relationship becomes ∆G* ) ∆G0* [1 + F(E - E°)/
4∆G0*]2. Since R is ∂∆G*/∂∆G° ) -(RT/F) ∂ ln khet/∂E, it follows that
∆G0*) F/(8 ∂R/∂E).

DissociatiVe Electrom Transfer to Peresters
J. Am. Chem. Soc., Vol. 123, No. 39, 2001 9579
Table 1. Reduction of 2a by Radical Anions in DMF at 25˚C
E°
(V)
∆G°
(eV)
log khom
donor
(M-1 s-1
)
3a
-1.41
-1.26
-1.10
-1.05
-0.86
-0.85
-0.80
-1.11
-0.96
-0.80
-0.75
-0.56
-0.55
-0.50
4.25
3.68
2.86
2.50
1.73
1.67
1.49
3-fluoro-3a
nitrobenzene
nitronaphthalene
4-nitroacetophenone
anthraquinone
4-nitrobenzonitrile
formation of delocalized radical anions. In fact, convolution
analysis of the reduction of 3a gave evidence of a very small
intrinsic barrier, which is consistent with a process ruled mainly
by solvent reorganization. The analysis was carried out at low
temperatures and relatively high scan rates to slow the hetero-
geneous kinetics. The data were thus obtained under conditions
in which the ET is quasi-reversible.15b The temperature effect
on k°het, studied for 3a in the range 233-278 K by using both
mercury and glassy carbon electrodes, led to preexponential
Arrhenius factor in the range 5 × 102-5 × 103 cm s-1, in
agreement with an adiabatic outer-sphere ET.
Homogeneous DET and Comparison with the Heteroge-
neous Results. The ET rate between free diffusing D and A
was determined for 2a by the homogeneous redox catalysis
approach.24 In this method, radical anion donors (D-•) are
electrogenerated and used to reduce homogeneously the accep-
tor. Thus, the reversible reduction peak of the donor (or
mediator) is transformed into a chemically irreversible, catalytic
peak upon addition of the peroxide (eqs 3-5). The current of
Figure 2. Comparison between homogeneous (O) and heterogeneous
(•) rate constants for the reduction of pivaloyl peroxide 2a in DMF at
25 °C. The dashed line is the second-order fit to the data (see text).
voltammetric oxidation peak of tBuCO2-, as previously de-
scribed for other carboxylates.12b,19,20 The bond dissociation
energy (BDE) of 2a, 30.6 kcal mol-1, was corrected for the
entropy by adding -6 kcal mol-1 21
.
∆G0 was estimated
*
through equation ∆G0 ) (BDE + λS)/4,17 where λS is the
solvent reorganization energy. The latter was calculated to be
20.5 kcal mol-1 by using the empirical equation λS ) 55.7/r,13
where λS is given in kilocalories per mole and the molecular
radius r in angstroms; r, 2.72 Å, was obtained by using the
experimental diffusion coefficient and the Stokes-Einstein
equation.22 By using this procedure, we obtained for the
concerted DET to 2a the theoretical values E° ) -0.25 V and
*
∆G0 ) 12.8 kcal mol-1, in excellent agreement with the
*
D + e a D-•
-• + RCO2-OtBu f D + RCO2 + tBuO•
(3)
(4)
corresponding convolution results, -0.30 V and 12.2 kcal mol-1
.
-
The above analysis thus ensures that the reductions of 2a
and 2b proceed by concerted ET/O-O bond cleavage (eq 1).
D
D
-• + tBuO• f D + tBuO-
(5)
RCO2-OtBu + e f RCO2- + tBuO•
tBuO• + e f tBuO-
(1)
(2)
the catalytic peak depends on V and the concentration of the
acceptor. The rate constant (khom) values (Table 1) were
determined by simulation of the experimental voltammograms
obtained using scan rates in the range 0.2-2 V s-1 and at least
three concentrations of the acceptor. Besides steps 3-5, the
simulations included the set of side reactions already used for
other peroxides.11,23,25
The first reduction step is rapidly followed by the reduction of
the tBuO• radical (eq 2), which has E° ) -0.23 V.23 The
reduction of the investigated peresters is also remarkably slow,
k°het being 1 × 10-10 and 9 × 10-11 cm s-1 for 2a and 2b,
respectively, which causes the peak to be more negative than
the estimated E° by ∼1.3 V. On the other hand, the reduction
of the model donors is reversible (3a, E° ) -1.41 V; 3b, E° )
-1.45 V) and fast (3a, k°het(GC) ) 0.13 cm s-1, k°het(Hg) )
0.12 cm s-1; 3b, k°het(GC) ) 0.12 cm s-1), as expected for the
(19) The oxidation peak of the carboxylate tBuCO2- is characterized by
the following parameters (0.2 Vs-1): Ep ) 0.96 V, peak width ∆Ep/2 ) 85
mV, ∂Ep/∂ log V ) 50 mV/decade. The carboxylate was generated
electrochemically through reduction of 2a (eqs 1 and 2).
(20) Isse, A. A.; Gennaro, A.; Maran, F. Acta Chem. Scand. 1999, 53,
1013.
In Figure 2 the homogeneous and the heterogeneous ET data
are combined together by adjusting the two vertical scales to
fit the same parabola. The latter was generated by using the
experimental heterogeneous ∆G0*. Although the heterogeneous
and the homogeneous ∆G0*s are not exactly the same, the latter
term can be estimated to be 11.5 kcal mol-1 and thus very
similar to the heterogeneous one. The homogeneous ∆G0* was
estimated as one-fourth of the sum of the BDE, 30.6 kcal
mol-1 21a
and the solvent reorganization energy λS, 15.4 kcal
,
mol-1. The λs value was estimated by using the empirical
equation λs ) 95[(2rD)-1 + (2rA)-1 - (rD + rA)-1],11 where rD
and rA are the donor and acceptor radii; an average value of
(21) (a) Bartlett, P. D.; Hiatt, R. R. J. Am. Chem. Soc. 1958, 80, 1398.
(b) Theoretical calculations provided a very similar value, 31 kcal mol-1
:
Benassi, R.; Taddei, F. J. Mol. Struct. 1994, 303, 101. (c) The entropy
correction at 25 °C was estimated by taking into account both the commonly
used fragmentation contribution (Benson, S. W. Thermochemical Kinetics,
2nd ed.; Wiley: New York, 1976.)11 and the activation entropy for the
thermal decomposition of 2a.21a
*
3.8 Å was used for rD.26 Therefore, by assuming the ∆G0
difference negligible, the difference between the two ordinates
of Figure 2 would therefore reflect that between the two
preexponential factors. The observed log Zhom - log Zhet value,
(22) The diffusion coefficients of 2a and 2b, D ) 1.0 × 10-5 and 9 ×
10-6 cm2 s-1, respectively, were calculated from the limiting values of the
convolution current.16 Since the convolution analysis leads to ln(khet/D1/2)-
E plots, these D values were also used to calculate the pertinent khet and
k°het values. The radius of 2a was calculated also by using the effective
radius approach2a or molecular models, leading to 2.64 and 2.59 Å,
respectively; in either case, the increase of the value of λS is within 1 kcal
(23) Workentin, M. S.; Maran, F.; Wayner, D. D. M. J. Am. Chem. Soc.
1995, 117, 2120.
(24) (a) Andrieux, C. P.; Blocman, C.; Dumas-Bouchiat, J. M.; M’Halla,
F.; Save´ant J.-M. J. Electroanal. Chem. 1980, 113, 19. (b) Andrieux, C.
P.; Save´ant J.-M. J. Electroanal. Chem. 1986, 205, 43.
(25) Kjær, N. T.; Lund, H. Acta Chem. Scand. 1995, 49, 848.
mol-1
.

9580 J. Am. Chem. Soc., Vol. 123, No. 39, 2001
Antonello et al.
Figure 4. Potential dependence of R for the reduction of 1a in DMF-
0.1 M TBAP at 25 °C.
Figure 3. Background-subtracted cyclic voltammetries for the reduction
of 2.3 mM 1a in DMF-0.1 M TBAP at 25 °C. Left to right: 0.1, 0.2,
0.5, 1, 2, 5, 10, 20 V s-1
.
respectively. Convolution analysis of the first peak data led to
evidence of a wavelike potential dependence of R (Figure 4),
which is very similar to those previously reported for the
reduction of some ring-substituted perbenzoates.12 This is, in
fact, the typical R-E pattern predicted for a DET occurring by
mixed concerted/stepwise mechanism (eqs 6 and 7).12b Reaction
7.2, is indeed close to 7.8, the difference calculated by using
the commonly accepted2a,11,26a,27 adiabatic preexponential factors
Zhom ) 3 × 1011 M-1 s-1 and Zhet ) 4.8 × 103 cm s-1 (Zhet
)
(RT/2πM)1/2, where M is the molar mass of 2a).
Because of their large intrinsic barriers, DETs are intrinsically
slow ET processes. However, inspection of the data reveals that
for the specific cases investigated both the khom and khet values
are surprisingly small. By using an Eyring-type rate constant
expression (k ) Z′ exp(-∆G*/RT), where Z′ includes the
transmission coefficient κ, i.e., Z′ ) κZ), the quadratic DET
rate/driving force relationship,18a the reaction free energy ∆G°
) -F(E°2a - E°D/D-•), and the experimental khom values, the
average Z′hom results to be 1 × 106 M-1 s-1. Even by
considering11 the latest modifications to the DET theory,27 this
value is still 4-5 orders of magnitude below the adiabatic limit.
Therefore, we can conclude that the homogeneous DET to 2a
is strongly nonadiabatic, even more pronouncedly than previ-
ously observed with dialkyl peroxides such as di-tert-butyl
peroxide and dicumyl peroxide.11 Because of the empirical
relationship between the homogeneous and the heterogeneous
preexponential values that we commented on above, it follows
that the electrode DET is also strongly nonadiabatic. In fact,
D-Sp-A + e f D-Sp- + A•
(6)
(7)
D-Sp-A + e a •-D-Sp-A f D-Sp- + A•
6 corresponds to the direct reduction of the slow end (A) of the
system (concerted DET). Conversely, reaction sequence 7
corresponds to electron injection into the fast end (D), followed
by intramolecular DET from D to A (stepwise DET). The second
mechanism has been discussed in detail for both dissociative-7
and nondissociative-29 type acceptors. In graphs a and b of
Figure 5, the experimental curves and those simulated for the
pathways 6 and 7 are compared for two different scan rates. It
is evident that the first peak arises from the overlapping of the
two components. On the other hand, because of the very
different intrinsic barriers involved, the stepwise pathway is
favored over the concerted DET at larger driving forces and
thus when E becomes more negative.2a,12 Since the peak is
irreversible and thus scan rate dependent, the concerted-to-
stepwise mechanism transition is conveniently induced by
increasing V, as illustrated very nicely in graphs a and b of Figure
5. The practical result is that for E values more negative than
-1.3 V the mechanism is essentially a pure stepwise process.
As such, the R values obtained for E < -1.3 V should reflect
the dependence on E typical of the fast end of 1a. In fact, the
slope estimated in this E region (cf. Figure 4) is in good
agreement with that expected for an outer-sphere ET process
ruled solely by solvent reorganization and is essentially the same
of that obtained with the model donor 3a.
Full simulation of the competitive mechanisms 6 and 7 is
illustrated in graph c of Figure 5. The simulations of Figure 5
were carried out by using the information obtained with 2 and
3 and by taking into account the possible intermolecular side
reactions, as described in detail elsewhere.7,29 Because of the
presence of the very short spacer -CMe2- (poor spatial
separation between D and A) and the fact that 2a has no amino
substituents R to the carbonyl, the optimized E° values of the
A and D ends of 1a are 0.18 and 0.06 V more positive than
those of 2a and 3a, respectively. For 1a, the intramolecular rate
constant (kintra) was thus estimated to be 1.3 × 104 s-1. The
corresponding preexponential factor (Z′intra) was calculated from
*
by using the experimental k°het and ∆G0 values of 2a, we
calculated Z′het ) 0.1 cm s-1, a value that is almost 5 log units
smaller than that expected for an adiabatic reduction. For 2b,
Z′het is small as well, being 0.5 cm s-1. These reductions are
thus the first characterized cases of heterogeneous nonadiabatic
DETs.
Intramolecular DET. The reduction of the D-Sp-A system
1a exhibits an irreversible peak (Ep ) -1.29 V, at 0.2 V s-1
)
followed by a reversible component (E° ) -1.70 V). As verified
in comparison with an authentic sample, the second peak is due
to the reduction of the phthalimide moiety in the carboxylate
anion D-Sp-, formed together with tBuO• in the DET. The
dependence of the normalized peak current ip/V1/2 on V is
illustrated in Figure 3. Although the figure is the equivalent of
Figure 1, the pattern of the main reduction peak is pretty
different. In fact, when V is progressively increased, ip/V1/2 first
increases and then decreases.28 At the same time, the peak first
sharpens and then broadens; representative values of the peak
width ∆Ep/2 are 99, 78, and 99 mV at 0.1, 2, and 50 V s-1
,
(26) (a) Kojima, H.; Bard, A. J. J. Am. Chem. Soc. 1975, 97, 6317. (b)
Eberson, L. AdV. Phys. Org. Chem. 1982, 18, 79.
(27) Andrieux, C. P.; Save´ant, J.-M.; Tardy, C. J. Am. Chem. Soc. 1998,
120, 4167.
(28) The small decrease of the ip/V1/2 ratio of the second peak is related
to the relatively slow reduction of the carboxylate. In fact, the reduction of
k
intra and the quadratic DET rate/driving force relationship. The
intrinsic barrier was obtained by considering that the O-O BDE
the phthalimide moiety in the carboxylate D-Sp- has k°het ) 0.055 cm s-1
,
as determined by the Nicholson method (Nicholson, R. S. Anal. Chem. 1965,
37, 1351).
(29) Zheng, Z.-R.; Evans, D. H. J. Am. Chem. Soc. 1999, 121, 2941.

DissociatiVe Electrom Transfer to Peresters
J. Am. Chem. Soc., Vol. 123, No. 39, 2001 9581
be significantly different. In fact, all other factors being the same,
the strength of the π*/σ* coupling and thus kintra depends not
only on the separation but also on the relative orientation of
the D and A groups. Concerning 1a and 1b, we drawn this
preliminary conclusion by comparing the X-ray crystal structure
of 1b32 with a molecular model of 1a, built according to the
X-ray structure of a similar molecule, R-(phthalimido)isobutyric
anhydride.33 The result obtained with 1b, however, allows us
to make another useful comparison with previous work. It is
worth noting that the kintra of 1b is 2.2 orders of magnitude
smaller than that of a bromide system of similar geometry, i.e.,
cis-1-methyl-4-benzoyloxycyclohexyl bromide, in which analo-
gously to 1b32 the donor is equatorial and the acceptor axial,
and driving force ∆G° ) -1.21 eV, logkintra ) 5.5.7 This result
would be surprising on the basis of a simple comparison between
the two nuclear factors because the C-Br bond of tertiary alkyl
bromides is ∼35 kcal mol-1 stronger than the O-O bond of
peresters and thus the bromides have much larger intrinsic
barriers than peroxides. Therefore, since the DET to tertiary
alkyl bromides is adiabatic, as checked11 by applying the
nonadiabatic DET theory,34 the observed small kintra of 1b also
points to the remarkable slowness of perester reduction.
Nonadiabatic DET and Conclusions. The nonadiabaticity
issue is emerging as an important but also intriguing aspect of
DETs. Previously, we found that the homogeneous reduction
of dialkyl peroxides is slower than expected on the basis of the
adiabatic DET theory by ∼2 orders of magnitude.11 More
specifically, we used the nonadiabatic DET theory34 and found
that for these peroxides the electronic coupling energy between
Figure 5. Comparison between background-subtracted cyclic volta-
mmetries (solid lines) for the reduction of 2.3 mM 1a at different scan
rates, in DMF-0.1 M TBAP at 25 °C, and corresponding simulations.
In graphs a and b, the curves were simulated by considering either the
concerted (eq 6, dotted lines) or the stepwise (eq 7, dashed lines)
mechanism. For clarity, the second peak was not included in the
simulation of the concerted mechanism. Graph c illustrates the result
of full simulation (O) of a typical experimental curve.
the reactant and product states, HRP, is of the order of 15 cm-1
,
a value distinctly below the commonly accepted adiabatic limit
of 200 cm-1 (≈ RT).35 In another study,12b we found that in
order to account for the heterogeneous reduction kinetics of
some ring-substituted perbenzoates, which proceeds by a
competitive concerted/stepwise mechanism (cf. eqs 6 and 7), it
was necessary to assume that the preexponential factor of the
concerted DET component is ∼2 orders of magnitude smaller
than that for the transient formation of the radical anion. In
addition, by using our previous convolution results on, for
example, the reduction of tert-butyl perbenzoate, it is possible
of 1a is essentially the same of that of 2a30 and by taking into
account, as previously described,7 that the presence of the spacer
increases the D/A distance and thus the solvent reorganization
term.1a The experimental Z′intra was thus estimated to be ∼7
orders of magnitude lower than the adiabatic limit Zintra ) kBT/
h. Although this difference includes some distance effect brought
about by the short spacer, this finding points, once again, to
the remarkable intrinsic slowness of the perester reduction.
The voltammetric behavior of 1b is quite similar to that of
1a. The main difference is that the reduction of D-Sp- is now
easier (E° ) -1.50 V) because the cyclohexyl spacer of 1b
greatly diminishes the effect caused by the charge of the
carboxylate on the E° of D. Simulation of the experimental
curves, carried out as for 1a, led to kintra ) 2 × 103 s-1. As
expected for a system in which the spacer is sufficiently long
to separate efficiently the electron exchanging moieties, the A
and D groups are now characterized by the same E° values
determined with the models 2b and 3b.
Although the kintra of 1b is smaller than that of 1a, as expected,
the decrease appears to be somewhat small compared to
nondissociative-type systems.1 This relatively small distance
effect is not caused by a significant difference between the two
DET driving forces, which are 1.23 and 1.20 eV for 1a and 1b,
respectively. On the other hand, the spacer of 1a is probably
too short to put much weight on this comparison. We are
currently investigating this aspect,31 which is complicated by
the fact that the relative D/A orientations of 1a and 1b seem to
to calculate k°het ) 2 × 10-9 cm s-1 and Z′het ) 8 cm s-1
.
Very recently, Workentin and co-workers reported evidence
indicating that the DET to endoperoxides artemisinin, ascaridole,
and dihydroascaridole is nonadiabatic.36 Therefore, it appears
that the homogeneous or heterogeneous DET to peroxides is,
as a rule, inherently nonadiabatic. Although for the peresters
investigated here the decrease of the preexponential factors with
respect to the corresponding adiabatic values may be easily in
error by 1 order of magnitude (because of the uncertainty
*
associated with the ∆G° and ∆G0 estimates), analysis of the
reduction results provides compelling evidence for the slowest
DET so far characterized.37
(31) Crisma, M.; Antonello, S.; Formaggio, F.; Moretto, A.; Maran, F.;
Toniolo, C. Work in progress.
(32) X-ray crystallography indicated that 1b has a cis(cyclohexane)-
equatorial(phthalimido)-axial(perester) conformation. Full details of the
structural analysis of 1b and related compounds will be published
elsewhere.31
(33) Valle, G.; Toniolo, C.; Jung, G. Liebigs Ann. Chem. 1986, 1809.
(34) (a) German, E. D.; Kuznetsov, A. M. J. Phys. Chem. 1994, 98, 6120.
(b) German, E. D.; Kuznetsov, A. M.; Tikhomirov, V. A. J. Phys. Chem.
1995, 99, 9095.
(35) Newton, M. D.; Sutin, N. Annu. ReV. Phys. Chem. 1984, 35, 437.
(36) (a) Magri, D. C.; Donkers, R. L.; Workentin, M. S. J. Photochem.
Phytobiol. A: Chem. 2001, 138, 29. (b) Donkers, R. L.; Workentin, M. S.
Chem. Eur. J., in press.
(30) Ru¨chardt, C.; Hamprecht, G. Chem. Ber. 1968, 101, 3957.

9582 J. Am. Chem. Soc., Vol. 123, No. 39, 2001
Antonello et al.
Intrinsic nonadiabaticity can be confirmed also by carrying
out the calculation of the electronic coupling matrix element
HRP, along lines similar to those previously described in detail
for the intermolecular DET to dialkyl peroxides.11 Thus, we
used the German-Kuznetsov nonadiabatic DET theory in its
semiclassical limit, in which the intramolecular modes and the
solvent polarization are treated classically (harmonic approxima-
tion).34 A Morse potential was used for the motion along the
O-O bond coordinate, whereas the energy curve of the
fragmented products was taken as the repulsive part of the
reactant Morse curve, as in the original Save´ant theory.17 Under
these conditions, the rate constant for DET between donor and
acceptor at encounter distance (k*) can be expressed by eq 8 in
be an intrinsic reason for observing such a slow DET rate. Very
recently, it has been argued on the basis of both experimental
data and theoretical calculations that there are reactions whose
rate is markedly reduced because of the failure of the Born-
Oppenheimer approximation near the transition state; a very nice
and mind-provoking account on the matter has been published
very recently by Butler, who contributed significantly to this
topic.40 In these reactions, such as for example some photodis-
sociations, the electronic wave function does not instantaneously
adjust along the reaction coordinate near the transition state. In
fact, the dominant bonding electronic configuration on the
reactant side must change to become repulsive past the transition
state. If the dynamics of the electronic rearrangement is
sufficiently slow, the crossing between the reactant and product
curves is only narrowly avoided, causing the reaction rate to
drop significantly. An analogous situation may hold for
concerted DETs, in which the donor injects one electron into
the σ* orbital of the frangible bond. In this framework, it would
not be unreasonable to start considering DETs as reactions that
are particularly susceptible to proceed nonadiabatically. This
hypothesis, however, calls for theoretical and further experi-
mental studies.
2π
k* ) (HRP)2{16πRT∆G*0 exp[-â(r* - r0)]}-1/2
p
exp(-∆G*/RT) (8)
which r* - r0 is the bond elongation at the transition state, â
is the Morse exponential factor, and ∆G* is the activation free
energy. Therefore, HRP can be calculated once the first-order
rate constant k* is related to the experimental rate constant kinter
and all the other quantities on the right-hand side of eq 8 are
calculated. kinter can be corrected to k* by using the equilibrium
constant (Kd) for the diffusion-controlled formation of reacting
D/A complex, i.e., kinter ) Kdk*. To calculate Kd, we used
equation Kd ) [(4πNr2δr)/1000],38 which takes into account
that the peroxide acceptor is uncharged and thus that no electric
work is required to form the encounter complex. According to
this model, the DET is viewed as occurring significantly only
between the contact distance r (taken as the van der Waals
distance rD + rA) and r + δr, in which δr ranges from ∼2 to
∼0.3 Å for adiabatic and nonadiabatic reactions, respectively.38
â, 2.84 Å-1, was calculated from the O-O stretching frequency
By taking into account the present results as well as those
obtained previously with other dissociative-type acceptors, it
also appears that the weaker the cleaving bond the smaller the
electronic coupling matrix element HRP. In fact, the BDE and
average HRP values (the comparison is for δr ) 0.3 Å) are as
follows: tert-butyl bromide, 66 kcal mol-1 and 190 cm-1; di-
tert-butyl peroxide, 37.3 kcal mol-1 and 15 cm-1; dicumyl
peroxide, 36.2 kcal mol-1 and 17 cm-1 11 2a, 30.6 kcal mol-1
;
and 0.3 cm-1. On the other hand, within the description of the
reactant and product energy curves in terms of the Morse
equation, it is worth reminding that, for a given cleaving bond,
stretching frequency ν0, and ∆G°, bond elongation at the
transition state is minimized by a decrease of BDE. On these
grounds, one might be tempted to speculate that a more reactant-
like transition state also implies a larger degree of electronic
wave function reorganization. It follows that lowering the
driving force would increase the electronic coupling between
the reactant and product energy curves at the transition state
and thus the DET rate. In keeping with this hypothesis, we note
ν0 using the relationship â ) ν0(2π2µ/BDE)1/2, where µ is
39
the reduced mass. The bond elongation r* - r0 is related to the
reaction free energy ∆G°34 and becomes larger for less
exoergonic reactions. Because of the weakness of the perester
O-O bond, r* - r0 is rather small. In particular, we estimated
that r* - r0 ranges from 0.09 to 0.17 Å for ∆G° ) -1.11 and
-0.50 eV, respectively. ∆G*, ∆G0*, and ∆G° were calculated
as already described. By using this procedure, we found that
for the homogeneous reduction of 2a HRP is indeed remarkably
small, being in the range 0.1-0.3 cm-1 throughout the series
of mediators employed. As for the Z′ estimates, these HRP values
also suffer from the uncertainty of the input estimated quantities
and the limits of the theory. Nevertheless, it is interesting to
observe that the calculated HRP values correspond to a rate drop
of ∼6 orders of magnitude with respect to an adiabatic process.
An aspect that is particularly worth noting about the present
investigation is that the same nonadiabaticity outcome could
be observed by using electrode, solution, or intramolecular
donors. This implies that the D/A orientation cannot be an
important issue because both the heterogeneous and the
homogeneous intermolecular DET rates are the result of random
distance and orientation distributions between D and A. In
keeping with this reasoning, it thus appears that there should
that for the nonadiabatic reduction of di-tert-butyl peroxide the
2
Arrhenius preexponential factor (which is proportional to HRP
,
as shown in eq 8) indeed increases as the driving force
decreases.11 A similar, although less pronounced trend, was
reported in the temperature study of the reduction of tert-butyl
bromide by aromatic radical anions.41a The trend observed with
other halides, however, led to ambiguous results.41 On the other
hand, we expect that nonadiabatic rather than intrinsically fast
DETs would be more prone to be affected by this rate-enhancing
mechanism at low driving forces. This is an intriguing possibility
that would suggest a new way of looking at the long-debated
problem of why some DET systems give rise to almost linear
activation/driving force relationships.42,43
It is evident that the nonadiabatic issue is at its very beginning
in the area of DETs. More experimental data on carefully
selected acceptor molecules and specific theoretical calculations
are definitely needed to better understand the reasons why some
(37) Nonadiabaticity cannot be attributed to steric hindrance; as a matter
of fact, we showed that screening of the O-O σ* orbital by the bulky tert-
butyl groups of di-tert-butyl peroxide is responsible for a rate drop of only
0.8 log unit.11
(40) See: Butler, L. J. Annu. ReV. Phys. Chem. 1998, 49, 125 and
pertinent references therein.
(41) (a) Daasbjerg, K.; Pedersen, S. U.; Lund, H. Acta Chem. Scand.
1991, 45, 424. (b) Balslev, H.; Daasbjerg, K.; Lund, H. Acta Chem. Scand.
1993, 47, 1221.
(42) (a) Lund, H.; Daasbjerg, K.; Lund, L.; Pedersen, S. U. Acc. Chem.
Res. 1995, 28, 313. (b) Lund, H.; Daasbjerg, K.; Lund, T.; Occhialini, D.;
Pedersen, S. U. Acta Chem. Scand. 1997, 51, 135.
(38) Sutin, N. Prog. Inorg. Chem. 1983, 30, 441.
(39) Our own measurements led to 854 cm-1, a value that is within the
very narrow range of frequencies reported for the O-O vibration of an
extensive series of peroxides: Vacque, V.; Sombret, B.; Huvenne, J. P.;
Legrand, P.; Suc, S. Spectrochim. Acta A 1997, 53, 55.

DissociatiVe Electrom Transfer to Peresters
J. Am. Chem. Soc., Vol. 123, No. 39, 2001 9583
compound was isolated by flash chromatography (CH2Cl2-EtOH 96:
4). Yield, 90%: mp 191-192 °C (EtOAc-petroleum ether); Rf1 ) 0.80;
Rf2 ) 0.55; Rf4 ) 0.80; IR (KBr) 1773, 1704 cm-1; 1H NMR (DMSO-
d6) δ 7.83 (s, 4H, Pht CH), 3.90 (m, 1H, cyclohexane CH), 2.6 (m,
1H, cyclohexane CH), 2.20 (m, 4H, cyclohexane CH2), 1.55 (m, 4H,
cyclohexane CH2).
of these DET reactions are inherently very slow. Most probably,
the study of intramolecular DET rates in well-defined D-Sp-A
molecules will provide a particularly efficient tool to gain new
insights into the fine details of the dynamics of these dissociative
processes.
cis-4-Phthaloylaminocyclohexanecarboxylic Acid-OOtBu (1b).
This compound was prepared from cis-4-phthaloylaminocyclohexan-
ecarboxylic acid (0.15 g, 0.55 mmol) and a 5.5 M solution of tert-
butyl hydroperoxide in n-decane (0.10 mL, 0.55 mmol) as described
above for 1a and purified by flash chromatography (EtOAc-petroleum
ether 1:9). Yield, 79%: mp 105-106 °C (EtOAc-petroleum ether);
Rf1 ) 0.95; Rf2 ) 0.95; Rf3 ) 0.30; IR (KBr) 1769, 1723, 1703 cm-1;1H
NMR (CDCl3) δ 7.78 and 7.67 (2m, 4H, Pht CH), 4.15 (m, 1H,
cyclohexane CH), 2.85 (m, 1H, cyclohexane CH), 2.50 (m, 4H,
cyclohexane CH2), 1.70 (m, 4H, cyclohexane CH2), 1.38 (s, 9H, OOtBu
CH3); 13C NMR (CDCl3) δ 171.13 (perester CO), 168.06 (Pht CO),
133.76 (phenyl CH4 and CH5), 131.95 (phenyl CH1 and CH2), 123.02
(phenyl CH3 and CH6), 83.48 (OOtBu quaternary C), 50.10, 36.84,
26.95, 26.41, 26.24 (cyclohexane CH and CH2 and OOtBu CH3).
Pht-Aib-OMe (3a). To an ice-cold solution of Pht-Aib-OH (1 g,
4.2 mmol) in CH2Cl2 (10 mL), DMAP (0.5 g, 4.2 mmol) and EDC‚
HCl (0.79 g, 4.2 mmol) were added, followed by methanol (0.17 mL,
4.2 mmol). After stirring for 4 h at room temperature, the reaction
mixture was evaporated to dryness and the residue dissolved in EtOAc.
The organic solution was washed with 10% KHSO4, H2O, 5% NaHCO3,
and H2O, dried over Na2SO4, and concentrated under reduced pressure.
A flash chromatography step (CH2Cl2-EtOH 98:2) was required to
purify the title compound. Yield, 80%: mp 70 °C (EtOAc-petroleum
ether); Rf1 ) 0.90; Rf2 ) 0.90; Rf3 ) 0.30; IR (KBr) 1765, 1734, 1710
Experimental Section
Synthesis and Characterization. Melting points were determined
using a Leitz model Laborlux 12 apparatus and are not corrected. Thin-
layer chromatography (TLC) and column chromatography were per-
formed on Merck Kieselgel 60 F254 plates and on Merck Kieselgel 60
(0.040-0.063 mm), respectively. The following eluants were used for
TLC analysis: (1) chloroform-ethanol, 9:1; (2) toluene-ethanol, 7:1;
(3) ethyl acetate-petroleum ether, 1:8; (4) 1-butanol-acetic acid-
water, 3:1:1. The TLC chromatograms were visualized by UV
fluorescence (254 nm) or developed by chlorine-starch-potassium
iodide chromatic reaction, as appropriate. Selective visualization of the
-O-O- bond was achieved by developing the TLC plates with a
solution of NH4SCN (0.63 g) and FeSO4‚(NH4)2SO4‚6H2O (Mohr’s salt,
0.88 g) in 1% H2SO4 (12 mL). All compounds were obtained in a
chromatographically homogeneous state. The solid-state IR absorption
spectra (KBr disk technique) were recorded with a Perkin-Elmer model
1720X FT-IR spectrophotometer, nitrogen-flushed, equipped with a
sample shuttle device, at 2-cm-1 nominal resolution, averaging 50 scans.
1H NMR and 13C NMR spectra were obtained by using a Bruker model
AC 250 spectrometer. Deuteriochloroform (99.96%, d; Aldrich) and
deuterated dimethyl sulfoxide (99.96%, d6; Acros Organics), with
tetramethylsilane as the internal standard, were used as solvents.
The mixed alkyl-acyl peroxides (peresters) 2a21a and 2b44 were
prepared by using a literature procedure.45 The mediators used for the
homogeneous redox catalysis experiments were commercially available
or synthesized as described below.
1
cm-1; H NMR (CDCl3) δ 7.80 and 7.70 (2m, 4H, Pht CH), 3.75 (s,
3H, OMe CH3), 1.85 (s, 6H, Aib CH3).
3-FPht-Aib-OH (3-FPht, 3-fluorophthaloyl). This derivative was
prepared from H-Aib-OH (0.46 g, 4.5 mmol) and the symmetrical
anhydride (3-FPht)2O (1 g, 5.4 mmol) according to the procedure
described for Pht-Aib-OH.30 Yield, 79%: mp 145-146 °C (EtOAc-
petroleum ether); Rf1 ) 0.75; Rf2 ) 0.50; Rf4 ) 0.80; IR (KBr) 1785,
1717 cm-1; 1H NMR (CDCl3) δ 7.70 and 7.35 (2m, 3H, Pht CH), 1.87
(s, 6H, Aib CH3).
Pht-Aib-OOtBu (1a) (Pht, phthaloyl; Aib, R-aminoisobutyric acid).30
With respect to the previously described literature procedure,30 an
improved synthetic protocol was employed. To an ice-cold solution of
Pht-Aib-OH30 (0.40 g, 1.6 mmol) in CH2Cl2 (6 mL), 4-(dimethylamino)-
pyridine (DMAP; 0.20 g, 1.6 mmol) and N-ethyl-N′-dimethylamino-
propylcarbodiimide) hydrochloride (EDC; 0.31 g, 1.6 mmol) were
added, followed by a 5.5 M solution of tert-butyl hydroperoxide in
n-decane (0.29 mL, 1.6 mmol). After stirring for 4 h at room
temperature, the reaction mixture was evaporated to dryness and the
residue dissolved in ethyl acetate (EtOAc). The organic solution was
washed with 10% KHSO4, H2O, 5% NaHCO3 and H2O, dried over
Na2SO4, and evaporated to dryness. The title compound was purified
by silica gel column chromatography (EtOAc-petroleum ether 1:8).
It precipitated as a waxy solid from a EtOAc solution upon addition of
petroleum ether. Yield, 80%: Rf1 ) 0.95; Rf2 ) 0.95; Rf3 ) 0.35; IR
3-FPht-Aib-OMe. This derivative was prepared from 3-FPht-Aib-
OH (1 g, 4.0 mmol) as described above for Pht-Aib-OMe. Yield 76%:
mp 95-96 °C (EtOAc-petroleum ether); Rf1 ) 0.90; Rf2 ) 0.90; Rf3
1
) 0.35; IR (KBr) 1776, 1736, 1710 cm-1; H NMR (CDCl3) δ 7.80
and 7.36 (2m, 3H, Pht CH), 3.76 (s, 3H, OMe CH3), 1.83 (s, 6H, Aib
CH3).
cis-4-Phthaloylaminocyclohexanecarboxylic Acid-OMe (3b). This
compound was prepared from cis-4-phthaloylaminocyclohexanecar-
boxylic acid (0.10 g, 0.37 mmol) as described above for 3a and purified
by flash chromatography (CH2Cl2-EtOH 97:3). Yield, 88%: mp 128-
129 °C (EtOAc-petroleum ether); Rf1 ) 0.90; Rf2 ) 0.90; Rf3 ) 0.30;
IR (KBr) 1765, 1723, 1706 cm-1;1H NMR (CDCl3) δ 7.80 and 7.70
(2m, 4H, Pht CH), 4.15 (m, 1H, cyclohexane CH), 3.76 (s, 3H, OMe
CH3), 2.65 (m, 1H, cyclohexane CH), 2.3 (m, 4H, cyclohexane CH2),
1.65 (m, 4H, cyclohexane CH2).
1
(KBr) 1780, 1719 cm-1; H NMR (CDCl3) δ 7.83 and 7.73 (2m, 4H,
Pht CH), 1.91 (s, 6H, Aib CH3), 1.34 (s, 9H, OOtBu CH3);13C NMR
(CDCl3) δ 169.98 (perester CO), 167.96 (Pht CO), 134.14 (phenyl CH4
and CH5), 131.65 (phenyl CH1 and CH2), 123.18 (phenyl CH3 and CH6),
84.33 (OOtBu quaternary C), 60.25 (Aib quaternary C), 26.14 (OOtBu
CH3), 24.67 (Aib CH3).
cis-4-Phthaloylaminocyclohexanecarboxylic Acid. (Pht)2O (0.26
g, 2.5 mmol) and cis-4-aminocyclohexanecarboxylic acid (0.30 g, 2.1
mmol) were finely ground and then melted at 190 °C. After 15 min,
heating was stopped. The cold reaction mixture was dissolved in a
H2O-triethylamine (pH 8) solution and extracted with diethyl ether.
The aqueous layer was acidified to pH 2 with KHSO4 and extracted
twice with EtOAc. The organic phase was washed with H2O, dried
over Na2SO4, and evaporated to dryness. From this residue, the title
Electrochemical Apparatus and Procedures. N,N-Dimethylfor-
mamide (Carlo Erba, 99.8%) and tetrabutylammonium perchlorate
(Fluka, 99%) were treated as previously described.13 Electrochemical
measurements were conducted in an all-glass cell, thermostated at the
required temperature. An EG&G-PARC 173/179 potentiostat-digital
coulometer, EG&G-PARC 175 universal programmer, Nicolet 3091
12-bit resolution digital oscilloscope, and Amel 863 X/Y pen recorder
were used. The feedback correction was applied to minimize the ohmic
drop between the working and the reference electrodes. The glassy
carbon (Tokai GC-20) electrode was prepared and activated before each
measurement as previously described.13 The electrode area was
determined through the limiting convolution current of ferrocene, the
diffusion coefficient of ferrocene being 1.13 × 10-5 cm2 s-1 in DMF.11
Some experiments were carried out on 3a by using a mercury
microelectrode.13 The reference electrode was Ag/AgCl, calibrated after
each experiment against the ferrocene/ferricenium couple. In the
presence of 0.1 M TBAP, we measured E°Fc/Fc+ to be 0.464 V versus
(43) Other mechanisms have been proposed to be responsible for rate
enhancement at low driving forces. One of them considers the interference
of an inner-sphere component when the ET becomes more endergonic.42
Recently, Jensen and Daasbjerg published very interesting results suggesting
that for some halides the rate could be enhanced at low driving forces
because of a more pronounced SN1-like contribution to the transition-state
structure (Jensen, H. K.; Daasbjerg, K. J. Chem. Soc., Perkin Trans 2 2000,
1251.).
(44) Lorenz, P.; Ru¨chardt, C.; Schacht, E. Chem. Ber. 1971, 104, 3429.
(45) Blomquist, A. T.; Berstein, J. J. Am. Chem. Soc. 1951, 73, 5546.

9584 J. Am. Chem. Soc., Vol. 123, No. 39, 2001
Antonello et al.
the KCl saturated calomel electrode (SCE) in DMF. All potential values
are reported versus SCE. The counter electrode was a 1-cm2 Pt plate.
Convolution and voltammetric analyses were carried out on low-
analyzed by using our homemade convolution software or compared
with the corresponding digital simulations. The DigiSim 3.03 package
was used for all simulations, using a step size of 1 mV and an
exponential expansion factor of 0.5.
noise voltammetric curves obtained in the range 0.1-50 V s-1
,
according to a procedure previously described.13 Since the reduction
of peresters, RCO3tBu, eventually yields RCO2 and tBuO-, a small
-
Acknowledgment. This work was financially supported by
the University of Padova (research project A.0EE00.97) and
the Ministero dell′Universita` e della Ricerca Scientifica e
Tecnologica (MURST).
amount of a weak acid, acetanilide, was added to the solution to
protonate tBuO-. Acid addition hampered the father-son reaction
RCO3tBu + tBuO- f RCO2tBu + tBuO2-, which would have affected
the voltammetric curves at low scan rate values.12 The curves were
digitalized, corrected for the background contribution by subtracting
the curves previously obtained in the absence of the substrate, and then
JA010799U