V. Jouikov, J. Simonet / Electrochemistry Communications 12 (2010) 1262–1265
1263
carefully polished with silicon carbide paper (Struers) or with Norton
polishing paper (types 02 and 03). Before use, the electrodes were
thoroughly rinsed with water, alcohol and acetone and finally dried in
a hot air stream.
Macro-electrolyses performed at a GC electrode in PC at the
threshold of step I (i.e., −1.6 V) result in total disappearance of ArI
−1
(after passing ≈1.5 F mol ). The analysis (GC–MS) of the electrol-
ysis solution shows the addition of the aryl anion onto the carbonate
(for PC, two isomers are obtained according to Eq. 2). Additionally, it
has been checked that parent methyl and ethyl esters of 1-naphthyl
carboxylic acid Ar–C(O)OR give a reversible system at a potential very
close to steps III and IV (Fig. 2). As shown in Fig. 2 B, a pair of
voltammetric steps is obtained after macro-reduction. In the course of
the first sweep, only steps III and IV (assigned with INp to the
Electrolyses were performed in small two-compartment cells on
2
large sheets of GC (3 to 4 cm ). Solutions were extracted with diethyl
ether and analyzed by GC/MS. Yields correspond to percentages of
soluble extracts and do not take into account possible polymer
deposition, which was found to be very small.
reversible reduction of the two isomers of 2, E
p
=−1.80 V and
3
. Results
−
1.90 V) are seen, while an additional scan affords a new minor peak
V (E
p
=−1.38 V) fully reversible if there is no ArI left in solution. This
3
.1. Behavior of ArIs at carbon electrodes
strongly supports that this new reduction step corresponds to the
formation of an unsaturated compound by action of the electro-
generated bases possibly produced in the course of adduct(s) 2
reduction (Eq. (3) Scheme 1). Adding a strong base (e.g., tetrame-
thylammonium hydroxide) to the electrolysis solution also leads to
step V. After many experiments of this kind achieved with ArIs in PC
and EC, it appears that several side products (3 but also 2 in a minor
way) may complicate the overall process by acting as redox medi-
ators. This is strongly reinforced by the fact that standard reduction
potentials of ArIs are expected to be located at about −1.0 V [7], case
of 4-iodobiphenyl, therefore much less negative than data relative to
some aryl bromides found around −2 V [8]. In such a case, the driving
force of the redox catalysis is due to standard potential differences.
In order to check the occurrence of redox processes provoked by the
presence of hydroxyethyl benzoate 2 and vinyl benzoate 3, ArI was
added to the electrolysis solution (Fig. 2 C). The first scan strongly
resembles the scan in Fig. 1 A, while the second scan reveals a large
peak at the potential of V (Ep5 =−1.46 V) together with steps III and
It is very unexpected that many aryl iodides such as 1-
iodonaphthalene (INp), 9-iodophenanthrene (IPA), 2-iodofluorene
IFl), 4-iodobiphenyl (IBp) and others, usually reduced in two two-
electron steps in solvents like acetonitrile and DMF, present a very
different cathodic process when reductions are achieved in cyclic
carbonates like PC and ethylene carbonate (EC). As exhibited in Fig. 1,
the feature (two-electron reduction of the carbon–iodine bond,
(
reaction 1 in Scheme 1, for INp, E =−1.90 V vs. SCE) dramatically
p
changed in the course of repetitive scans. Thus, when the scan starts
from a potential less negative than −1.0 V vs. SCE, the main peak I
may present a very small pre-peak II that may almost totally vanish
during the second scan (Fig. 1 A). Yet, after a rather large number of
scans (B), the pre-peak II (Ep2 =−1.59 V) increases regularly while
step I decays, progressively disappears, and is replaced by a reversible
step. A limit is usually reached after 10scans.
IV then restored. This supports the formation of 3 at the end of the
−
first scan and the triggering of a redox catalysis process by 3•
.
Independently, the response of INp at a smooth silver electrode gives,
under identical conditions, step VI at −1.41 V. It is then very
intriguing why two modes of catalysis, in principle so different, may
lead to such similar potential shifts (ΔE≈+0.4 V at bare carbon).
3.2. Macro-electrolyses at GC
The crucial question is whether step I (absence of redox catalysis)
and step II (strong catalytic effect with a shift of about +0.5 V) as
exhibited in Fig. 1 both yield ArH compounds in high yield, as found at
GC in DMF or acetonitrile. Surprisingly, the answer is no. As a matter
of fact, the reduction of INp at GC in PC at −1.7 V (after a hold at −2 V
to trigger the redox reaction) yields, beside NpH (8%) and Np–PC–OH
(
2 isomers of hydroxyethyl benzoate, 3%), the coupling product Np–
Np (77%). All the INp was reduced. Additionally, the electrode is
covered by a red-brownish deposit, insoluble in acetone, probably
arising from the polymerization of vinyl benzoate. Its structure is
assigned to a redox polymer (polyaryl ester according to Eq. 7) and
this is supported by the existence of a slightly reversible reduction
step (E
p
=−1.89 V). On the contrary, the electrolysis accomplished at
−
2.1 V affords NpH (52%), Np–Np (43%) and Np–Pc–OH (4%). This
behavior is rather general for other aryl iodides but the progressive
deposition of polymer may slow down the reduction process, and a
certain amount of the starting substrate may be recovered.
3.3. Boosting the redox catalytic process
There are at least two additional ways to set up a fast redox
catalytic process as presented above. First, one can add to the solution
an “external” redox mediator the E° of which should be less negative
−
(
i.e., ≈0.4 V) than step I; then, Ar derived from acenaphthylene
Fig. 1. Voltammetry of INp (8.2 mmol L−1) in PC. GC electrode (0.8 mm ); v=50 mV s−1.
2
(E°=−1.60 V) is a very efficient catalyst towards INp. Second, one
can dope the carbon electrode with very small amounts of transition
(A) Two first scans. (B) Repetitive scans.