A bifunctional molecule that receives two electrons sequentially through only one of its two reducible groups

Full text of DOI:10.1021/ja990155a

J. Am. Chem. Soc. 1999, 121, 2941-2942
2941
portion is the “slow” end, reflecting the peculiarly sluggish
electron-transfer kinetics of nitroalkanes, particularly when studied
using bulky cations in the supporting electrolyte.
A Bifunctional Molecule that Receives Two Electrons
Sequentially through Only One of Its Two Reducible
Groups
Zi-Rong Zheng and Dennis H. Evans*
Department of Chemistry and Biochemistry
UniVersity of Delaware, Newark, Delaware 19716
ReceiVed January 19, 1999
The expected properties of 3 manifest themselves in fragments
4 and 5. The cyclic voltammograms of 4 (Figure 1A) indicate
that it is reduced in a reversible process (ks > 0.1 cm/s) with a
formal potential of -2.07 V vs ferrocene.5 By contrast, nitroketone
5 is reduced irreversibly with a very large separation between
cathodic and anodic peaks (Figure 1B). The standard rate constant
for reduction of 5 is 2 × 10-3 cm/s which is typical6 for a tertiary
nitroalkane (cf. 4.1 × 10-3 cm/s for 2-methyl-2-nitropropane
under the same conditions6c). The rate constant was evaluated by
fitting digital simulations7 to the experimental voltammogram
from which the formal potential was also evaluated to be -2.06
V, almost identical to that of 4. Because of the sluggish electron-
transfer kinetics of 5, very little reduction occurs near its formal
potential.
A typical cyclic voltammogram for 3 is also included in Figure
1. There is a single reduction peak accompanied by one oxidation
peak on the return scan. The peak height corresponds to an overall
two-electron reduction that is reversible and occurs near the
expected formal potential of the fast end of the molecule. We
infer from these results the following reaction sequence:
We report here the first example of a molecule, containing two
electroactive groups, one on either end, that sequentially and
reversibly accepts two electrons at the same potential with the
electrons being inserted and removed through one end of the
molecule.
There are many examples of organic and organometallic species
that can be reduced (or oxidized) in a series of two or more steps.
In most cases, the introduction or removal of each electron occurs
with greater difficulty than that for the previous electron so that,
when carried out electrochemically, the reduction (or oxidation)
steps occur at distinctly different potentials. An example is the
reduction of aromatic hydrocarbons in nonaqueous solvents, which
occurs in two steps separated by several hundred millivolts.1 There
are some systems in which two electrons can be introduced at
essentially the same potential, a well-known example being the
1,ω-bis(4-nitrophenyl)alkanes 1 (n ) 2-4).2 The ease of reduction
of the two nitrophenyl groups is very similar by virtue of the
fact that they are identical and sufficiently separated so as to
interact only very weakly. Obviously, it is not meaningful to ask
which group is the first to react. In the cyclohexyl bromide
derivatives 2, the first step of reduction occurs at the substituted
benzoate group followed by internal electron transfer with
concerted carbon-bromine bond cleavage.3 The resulting alkyl
radical then accepts a second electron to form the carbanion. For
2 the order of reduction is clear, but the overall process is
irreversible owing to the internal dissociative electron-transfer
reaction.
Here F denotes the fast end of the molecule (cyanobenzoate group)
and S is the slow end (nitroalkyl group). Recall that the electron
affinity of both groups will be almost identical based on the study
of fragments 4 and 5. The reaction is initiated by the introduction
of an electron into the fast end forming the radical anion of the
cyanobenzoate group (reaction 1). This reaction is followed by
We have prepared the 4-cyanobenzoate ester 3 with the
intention of creating a molecule with two reducible groups with
almost identical electron-accepting properties but which differ
dramatically in the rate of electron transfer to the group.4 The
cyanobenzoate functionality may be considered the “fast” end of
the molecule because its electron-transfer kinetics are exceedingly
facile as is typical for an aromatic system. The tertiary nitroalkane
(5) Cyclic voltammetry was carried out with EG&G Princeton Applied
Research (PAR) model 283 electrochemistry system, a PAR 303 static mercury
drop working electrode (smallest drop size), acetonitrile as solvent with 0.10
M Bu4NPF6 electrolyte. Resistance compensation was applied at the level of
310 Ω. The standard heterogeneous electron-transfer rate constant, ks, is a
measure of the reversibility of the electron-transfer reaction. The value given
for 4 is approximate because no attempt was made to account accurately for
the effects of solution resistance. The potentials are actually the half-wave
potentials which differ from the reversible formal potential by a small term
(usually less than 10 mV), involving the diffusion coefficients of reactant
and product. The potentials have been referred to the reversible half-wave
potential of the ferrocenium/ferrocene couple measured under the same
conditions as used for 3-5.
(6) (a) Save´ant, J. M.; Tessier, D. J. Electroanal. Chem. 1975, 65, 57-66.
(b) Save´ant, J. M.; Tessier, D. J. Phys. Chem. 1977, 81, 2192-2197. (c)
Corrigan, D. A.; Evans, D. H. J. Electroanal. Chem. 1980, 106, 287-304.
(d) Petersen, R. A.; Evans, D. H. J. Electroanal. Chem. 1987, 222, 129-150.
(e) Gilicinski, A. G.; Evans, D. H. J. Electroanal. Chem. 1989, 267, 93-104.
(f) Evans, D. H.; Gilicinski, A. G. J. Phys. Chem. 1992, 96, 2528-2533.
(7) Simulations were performed using the software package DigiSim
(Version 2.1, Bioanalytical Systems, West Lafayette, IN). In the simulations
a stepsize of 1 mV and an exponential expansion factor of 0.5 were used.
(1) (a) Perichon, J. In Encyclopedia of Electrochemistry of the Elements;
Bard, A. J.; Lund, H., Eds.; Dekker: New York, 1978; vol XI, pp 71-161.
(b) For a discussion of the factors affecting the separation in potentials, see:
Evans, D. H.; Hu, K. J. Chem. Soc., Faraday Trans. 1996, 92, 3983-3990.
(2) Ammar, F.; Save´ant, J. M. J. Electroanal. Chem. 1973, 47, 115-125.
(3) Antonello, S.; Maran, F. J. Am. Chem. Soc. 1998, 120, 5713-5722.
(4) (a) Compound 5 was prepared by the base-catalyzed addition of
2-nitropropane to methyl vinyl ketone. This nitroketone was reduced to the
corresponding alcohol with sodium borohydride. Finally, ester 3 was formed
by reaction of the alcohol with 4-cyanobenzoyl chloride. We also prepared
2,4,4-trimethylpentyl 4-cyanobenzoate as a model of 3 by reaction of the
commercially available alcohol with 4-cyanobenzoyl chloride. This ester 6,
whose size, shape, and molecular weight are almost identical to those of 3,
was used to confirm our measurement of the diffusion coefficient of 3 by
steady-state microelectrode voltammetry or chronocoulometry4b with 9,10-
anthraquinone (D ) 2.07 × 10-5 cm2/s at 298 K in acetonitrile) as standard.
The results were D6 ) 1.35 × 10-5 cm2/s and D3 ) 1.38 × 10-5 cm2/s. (b)
Oliver, E. W.; Evans, D. H.; Caspar, J. V J. Electroanal. Chem. 1996, 403,
153-158.
10.1021/ja990155a CCC: $18.00 © 1999 American Chemical Society
Published on Web 03/11/1999

2942 J. Am. Chem. Soc., Vol. 121, No. 12, 1999
Communications to the Editor
the radical anion of 4 with neutral 5 was 6 × 106 M-1s-1 and, in
the simulations to be described below, it was found that when
this value was used for kf,5, the effect on the simulated voltam-
mogram was negligible. Disproportionation reaction 6 also
involves electron transfer to the slow end of 3 and is only slightly
downhill based on the formal potentials of 4 and 5. Its rate
constant, kf,6, should be similar to kf,5, so that reaction 6 will also
make a negligible contribution to the rate of reduction of the slow
end of the molecule. Thus, it is concluded that intramolecular
electron transfer (reaction 2) is the most important pathway for
getting electrons into the nitroalkane functionality of 3.
What is the rate constant for the intramolecular electron
transfer? Clearly it is a rather rapid process because complete,
overall two-electron reduction (reactions 1-3) is seen even at
40 V/s (Figure 1C). The separator between the cyanobenzoate
group and the aliphatic nitro group is rather short, essentially a
four-carbon saturated chain. In addition, the separator is quite
flexible, allowing easy attainment of conformations in which the
nitro group and the π-system of the cyanobenzoate group are very
close to one another. Thus, 3 has been designed for very rapid
intramolecular electron transfer.
Simulations of voltammograms of 3 based on reactions 1-3
provided adequate agreement with the experimental results for
scan rates from 5 to 40 V/s. Below 5 V/s, some irreversibility
sets in due to the cleavage of nitrite from either F-S-• or •-F-
S-• (probably the latter) as has been observed with the radical
anion of tert-nitrobutane.6c The simulation for 3 at 40 V/s is also
given in Figure 1. The parameters used in the simulation (see
caption to Figure 1) are quite reasonable. In particular, it is
interesting that the simulations are very sensitive to the magnitude
of K2. The value used in the simulation of Figure 1, 0.3, can only
be varied between about 0.2 and 0.5 while maintaining adequate
agreement, meaning that the intramolecular electron-transfer
reaction 2 is unfavorable by 20-40 mV, whereas it is predicted
to be favorable by 10 mV on the basis of the study of fragments
4 and 5. This appears to be quite acceptable agreement because
it is not expected that exact prediction of the properties of intact
3 can be achieved by examining fragments 4 and 5.
The rate constant for intramolecular electron transfer, kf,2,
cannot be determined accurately because the voltammogram at
40 V/s is very close to the overall two-electron limit. The value
used in the simulation, 2 × 105 s-1, may be regarded as an
approximate lower limit. Very precise evaluation of the electrode
area and diffusion coefficient of 3 would be required to place an
error bar on this estimate.
Figure 1. Cyclic voltammograms of 3 and fragments 4 and 5. Scan
rate: 40 V/s. (A) 1.17 mM 4. (B) 1.00 mM 5. (C) 0.98 mM 3 (line).
Digital simulation (circles). Simulation parameters: E°1 ) -2.058 V,
ks,1 ) 0.076 cm/s, R1 ) 0.5; K2 ) 0.3, k2,f ) 2 ×105 s-1, E°3 ) -2.080
V, ks,3 ) 0.076 cm/s, R3 ) 0.5. The area of the mercury drop was 9.8 ×
10-3 cm2, and the temperature was 298 K.
intramolecular electron transfer from the fast to slow end of the
molecule (reaction 2) with subsequent introduction of a second
electron into the fast end to form the dianion final product
(reaction 3).
On the return sweep, oxidation occurs by the reverse of the
above sequence, i.e., an electron is transferred from the dianion
to the electrode through the fast end (reverse of reaction 3), an
intramolecular electron transfer occurs (reverse of reaction 2),
and finally, the second electron is transferred to the electrode
through the fast end.
Reactions 4-6 represent alternative pathways for introducing
electrons into the slow nitroalkyl group. The first of these, reaction
4, is the direct electron transfer from the electrode to the slow
end, a process that is too slow to contribute significantly. In fact,
if one assumes that the formal potentials and rate parameters for
reduction of 3 (reactions 1 and 4) are identical to those of
fragments 4 and 5, fewer than 3% of the electrons will enter 3
through the slow end at the formal potential for fast end
reduction.8
Reaction 5 is the intermolecular transfer of an electron to the
slow end. We assessed the importance of this reaction by
determining the rate of intermolecular electron transfer between
the radical anion of 4 and neutral 5 by the method of homogeneous
redox catalysis.6e,9 The latter reaction should be a good model
for reaction 5. It was found that the rate constant for reaction of
In summary, we have designed and prepared a molecule that
accepts two electrons sequentially by the introduction into only
one of the two electroactive groups. The cyanobenzoate group
serves as a fast and reversible shuttle of electrons into the
kinetically impeded nitroalkyl portion of the molecule. The
generality of the phenomena discussed above is presently being
explored by studies of molecules with both shorter and longer
spacer units as well as conformationally immobile separators. The
results of these studies will be described in the full paper.
Acknowledgment. This work was supported by the National
Science Foundation, Grant CHE-9704211.
(8) This calculation assumes that the forward heterogeneous electron-
transfer rate constant, kf, is given by the Butler-Volmer equation, kf ) ks
exp[-(RF/RT)(E - E°)]. For the fast end, kf,1 ) ks,1 > 0.1 cm/s at E ) E°1
) -2.07 V. For the slow end, kf,4 was computed using ks,4 ) 2 × 10-3 cm/s,
R ) 0.4 (as obtained from simulations of the voltammograms of 5) and E°4
) -2.06 V. The result, kf,1/kf,4 > 40, leads to the conclusion that <3% of the
electrons are entering through the slow end at E ) E°1.
JA990155A
(9) Andrieux, C. P.; Blocman, C.; Dumas-Bouchiat, J. M.; M’Halla, F.;
Save´ant, J. M. J. Electroanal. Chem. 1980, 113, 19-40.