One- versus two-electron reaction pathways in the electrocatalytic reduction of benzyl bromide at silver cathodes

Article abstract of DOI:10.1016/j.tetlet.2006.08.123

The electrocatalytic reduction of benzyl bromide at a silver cathode has been investigated in acetonitrile in the absence and presence of acids, using cyclic voltammetry (CV) and controlled-potential electrolysis (CPE). CV gives rise to two reduction waves, which represent the dissociative 1e^- reduction of PhCH₂Br to PhCH₂^{{radical dot}} and Br^- followed by a further reduction of the benzyl radical to PhCH₂^- at more negative potentials. The charge stoichiometry (1e^- vs 2e^-/molecule) and product distribution depend on the applied potential and reaction medium. In the absence of added acids, the reduction of PhCH₂Br at potentials of the first wave is a 1e^- process mainly yielding bibenzyl, whereas toluene becomes the principal product at potentials beyond the second wave. The addition of acids strongly modifies the dependence of selectivity on the applied potential. The presence of a strong acid changes the mechanism of the process, which now becomes a 2e^- reduction to toluene, even at potentials corresponding to the first reduction wave.

Full text of DOI:10.1016/j.tetlet.2006.08.123

Tetrahedron Letters 47 (2006) 7735–7739
One- versus two-electron reaction pathways in the
electrocatalytic reduction of benzyl bromide at silver cathodes
Abdirisak A. Isse, Alessio De Giusti and Armando Gennaro*
Department of Chemical Sciences, University of Padova, via Marzolo 1, 35131 Padova, Italy
Received 21 July 2006; revised 22 August 2006; accepted 25 August 2006
Available online 18 September 2006
Abstract—The electrocatalytic reduction of benzyl bromide at a silver cathode has been investigated in acetonitrile in the absence
and presence of acids, using cyclic voltammetry (CV) and controlled-potential electrolysis (CPE). CV gives rise to two reduction
waves, which represent the dissociative 1eꢀ reduction of PhCH2Br to PhCH and Brꢀ followed by a further reduction of the benzyl
Å
2
radical to PhCH2 at more negative potentials. The charge stoichiometry (1eꢀ vs 2eꢀ/molecule) and product distribution depend on
ꢀ
the applied potential and reaction medium. In the absence of added acids, the reduction of PhCH2Br at potentials of the first wave is
a 1eꢀ process mainly yielding bibenzyl, whereas toluene becomes the principal product at potentials beyond the second wave. The
addition of acids strongly modifies the dependence of selectivity on the applied potential. The presence of a strong acid changes the
mechanism of the process, which now becomes a 2eꢀ reduction to toluene, even at potentials corresponding to the first reduction
wave.
ꢀ 2006 Elsevier Ltd. All rights reserved.
1. Introduction
The voltammetric response as well as the chemistry
underlying the reduction process depend on the relative
values of E1 and E2. When E1 > E2, two successive 1eꢀ
reduction waves are observed and the electrochemical
process can be easily directed to radical or anionic chem-
istry by applying potentials corresponding to E1 or E2,
respectively. Unfortunately, E2 is often more positive
than E1, so that RÅ is reduced as soon as it is formed.
Thus, most organic halides, especially chlorides and bro-
mides, exhibit a single 2eꢀ reduction wave in voltam-
metry and give products based on carbanion chemistry
in preparative-scale electrolyses. A very limited number
of examples of systems exhibiting two 1eꢀ reduction
waves have also been reported.4 In such cases, the distri-
bution of the reduction products in electrolysis can be
modulated by varying the applied potential.4b–d
The electrochemical activation of the C–X bond in
organic halides is one of the most important reactions
in organic electrochemistry.1 The process has attracted
much attention both from mechanistic and synthetic
points of view. The electroreduction of RX to the corre-
sponding hydrocarbon is also investigated as a promis-
ing method of treating halogenated wastes.2 The
mechanism of the reductive cleavage of the C–X bond
is well established. It involves a pair of one-electron
transfers occurring in two steps followed by various pos-
sible reactions of the ensuing carbanion, for example,
protonation and nucleophilic substitution.
RX þ eꢀ ! RÅ þ Xꢀ E1
ð1Þ
ð2Þ
RÅ þ eꢀ ¢ Rꢀ
E2
The possibility of creating conditions for which the
reduction of RÅ occurs beyond that of RX, that is,
E2 < E1, is of fundamental importance in electrosynthe-
sis. As pointed out by Fry and Powers,4d the relative
values of E1 and E2 depend on both the nature of the
halogen atom and the structure of the compound. How-
ever, predicting the dependence of the values of E1 and
E2 on the structure of RX is a difficult task. An alterna-
tive approach is to work on the choice of the cathode
material. In fact, if catalytic surfaces, capable of signifi-
cantly lowering the reduction overpotential for RX, are
The dissociative electron transfer (ET) (Eq. 1) may pro-
ceed via a stepwise mechanism with an intermediate for-
mation of RXÅꢀ or it may occur in a single step, the
cleavage of the C–X bond being concerted with the ET.3
Keywords: Electrocatalysis; Benzyl bromide; Silver cathode; Dehydro-
halogenation; Electrocarboxylation.
*
Corresponding author. Tel.: +39 0498275132; fax: +39
0498275239; e-mail: armando.gennaro@unipd.it
0040-4039/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.08.123

7736
A. A. Isse et al. / Tetrahedron Letters 47 (2006) 7735–7739
used as cathode materials, E1 may become more positive
than E2. It has been recently shown that silver has
extraordinary electrocatalytic properties towards the
reduction of organic halides in aprotic solvents.5,6 We
have recently shown that the 2eꢀ single wave observed
for the reduction of benzyl bromide and benzyl iodide
at an inert electrode is split into two waves at an Ag
cathode.6 In this letter, we wish to describe the results
of a study on the potentialities of an Ag cathode in
promoting either radical or anionic chemistry in the
reduction of benzyl bromide. The process was investi-
gated by varying both the applied potential and reaction
medium. In particular, attention was focussed on the
role of acids on the selectivity and charge stoichiometry
of the process.
The results of the preparative-scale electrolyses of benzyl
bromide in the absence and presence of proton donors
are summarised in Table 1. As proton donors, we used
dichloroacetic acid (DCA), acetic acid (AcOH), phenol
(PhOH) and 2,2,2-trifluoroethanol (TFE). All experi-
ments were conducted under potentiostatic conditions
and were interrupted after the current decreased to ca.
5% of its initial value. In all cases, PhCH2Br conversion
was greater than 95%. As can be seen, a mixture of
toluene and bibenzyl, with a distribution strongly
depending on both reaction medium and applied poten-
tial, Eapp, is obtained.
The data obtained without added acid (Table 1, entries
1–3) can be rationalised according to the following reac-
tion mechanism:
ꢀ
PhCH2Br þ eꢀ ! PhCH2 þ Br
ð3Þ
ð4Þ
ð5Þ
ð6Þ
Å
2. Results and discussion
Å
2PhCH2 ! PhCH2CH2Ph
As shown in Figure 1, the cyclic voltammetry of benzyl
bromide at Ag gives rise to two irreversible reduction
peaks with Ep values of ꢀ1.08 and ꢀ1.40 V versus
SCE at v = 0.2 V sꢀ1.7 The addition of proton donors
such as acetic acid increases the first peak at the expense
of the second one, which decreases up to disappearing,
while the peak potentials are not significantly affected.
Notice that the reduction of the acids at the electrode
surface starts at potentials beyond the reduction peaks
of PhCH2Br and, hence, does not contribute to the
observed CV response. As we have shown in an earlier
PhCH2 þ eꢀ ¢ PhCH2
Å
ꢀ
PhCH2ꢀ þ HA ! PhCH3 þ Aꢀ
where the proton donor HA is the residual water or the
solvent itself.8 Reactions 3–6 may also involve adsorbed
species, especially the starting halide and the products of
the first ET. The process at ꢀ1.0 V involves 1eꢀ reduc-
tion of PhCH2Br to a benzyl radical, which then dime-
rises to give bibenzyl (Eqs. 3 and 4). The reduction
Å
potential of PhCH2 at a gold electrode has been re-
paper,6 PhCH2Br is sequentially reduced to PhCH2
ported to be ꢀ1.43 V versus SCE in acetonitrile.9 Thus,
at potentials more negative than this value, the reduc-
tion of the radical becomes fast, resulting in the forma-
tion of toluene as the principal product. It is important
to stress that, even under such conditions (entry 3), only
a modest yield of toluene can be obtained together with
a significant quantity of bibenzyl. The latter is more
likely to be formed via an SN2 reaction between
Å
and Brꢀ and then to PhCH2 at the first and second
ꢀ
peaks, respectively. This process involves the adsorption
of the starting compound as well as its reduction prod-
ucts, especially the bromide ion. The presence of a
strong proton donor appears to modify the reaction
mechanism. Now, a very sharp peak, which is typical
of a process involving adsorbed species, is observed.
Furthermore, the second peak due to the reduction of
ꢀ
PhCH2 and PhCH2Br, rather than by radical–radical
Å
ꢀ
PhCH2 to PhCH2 disappears. It is likely that, in the
presence of an acid, the process becomes a 2eꢀ reduction
of PhCH2Br to toluene.
coupling. Thus, the selectivity of the process for toluene
production is not very high, even at potentials as nega-
tive as ꢀ1.8 V versus SCE. Also the overall yield de-
creases with decreasing Eapp and the process consumes
ca. 1eꢀ/molecule of PhCH2Br, even at the most negative
potential. This is due to side reactions consuming the
starting halide, for example, the nucleophilic attack at
ꢀ
ꢀ
PhCH2Br by PhCH2 and conjugate bases NCCH2
and OHꢀ, arising from the protonation step (Eq. 6).10
When good proton donors are added, the dependence of
the product distribution on applied potential changes
drastically. In the case of dichloroacetic acid, the process
at ꢀ1.0 V becomes a 2eꢀ reduction of PhCH2Br to tol-
uene. A remarkable increase of the yield of toluene at
ꢀ1.0 V is also obtained with acetic acid and phenol.
With these two acids, the process approaches a 2eꢀ
reduction at potentials slightly more negative than
ꢀ1.0 V. For instance at ꢀ1.4 V, toluene yields as high
as 89% and 83% can be obtained with CH3CO2H and
PhOH, respectively. Trifluoroethanol is less efficient
than the above acids, but its effect is well evident if
compared with the experiments performed in the
absence of added proton donors. When Cl2CHCO2H
Figure 1. Cyclic voltammetry of 3.02 mM benzyl bromide in
CH3CN + 0.1 M (C2H5)4NClO4 recorded at v = 0.2 V sꢀ1 in the
absence (a) and presence of acetic acid, (b) cAcOH = 1.5 mM, (c)
cAcOH = 2.9 mM, (d) cAcOH = 5.7 mM.

A. A. Isse et al. / Tetrahedron Letters 47 (2006) 7735–7739
7737
Table 1. Electrolysis of benzyl bromide (20 mM) in CH3CN + 0.1 M (C2H5)4NClO4 at a silver cathode
RHb (%)
RRb (%)
Totalb (%)
a
Entry
Acid
cacid (mM)
Eapp (V)
n (F/mol)
1
2
3c
4
5
6
7
8
ꢀ1.0
ꢀ1.5
ꢀ1.8
ꢀ1.0
ꢀ1.0
ꢀ1.2
ꢀ1.4
ꢀ1.6
ꢀ1.0
ꢀ1.4
ꢀ1.8
ꢀ1.0
ꢀ1.4
ꢀ1.8
1.0
1.0
1.1
2.0
1.6
1.8
1.9
2.0
1.5
1.7
1.7
1.1
1.3
1.4
7
27
48
89
68
86
89
92
60
83
80
15
58
65
83
46
9
7
26
9
5
1
33
6
3
90
73
57
96
94
95
94
93
93
89
83
95
83
69
Dichloroacetic acid
Acetic acid
Acetic acid
Acetic acid
Acetic acid
Phenol
Phenol
Phenol
Trifluoroethanol
Trifluoroethanol
Trifluoroethanol
40
41
41
41
41
40
40
40
41
41
41
9
10
11
12
13
14
80
25
4
a Applied potential versus SCE.
b Yield calculated with respect to converted PhCH2Br.
c Hydrocinnamonitrile (19%) was also detected.
and CH3CO2H are used, the total yield is very high
(ca. 95%) and does not depend on Eapp or on the yield
of toluene. Conversely, when the alcohols PhOH and
F3CCH2OH are used as proton donors, the total yield
decreases with decreasing Eapp, that is, increasing
toluene yield. This trend is similar to that observed when
no acids are added and can be attributed to the nucleo-
philic reaction between the conjugate bases of the
proton donors and benzyl bromide.
in Figure 3 in which the yields of toluene obtained at
ꢀ1.0 and ꢀ1.4 V versus SCE are plotted versus pKa of
the acids in acetonitrile.11
The most striking result of the electrolyses with added
acids is that the 2eꢀ reduction of PhCH2Br to PhCH3
can be achieved at potentials that are more than 0.4 V
more positive than the reduction potential of the benzyl
radical. In principle, the added acids may facilitate the
Å
reduction of PhCH2 by increasing the rate of protonation
ꢀ
The influence of the acids on the product distribution is
shown in Figure 2, which compares data obtained in the
absence and presence of CH3CO2H and PhOH. The
data obtained using F3CCH2OH as a proton donor
show a similar trend, but were omitted for the sake of
clarity. As can be clearly seen, high yields (>75%) of tol-
uene can be obtained with both acids at Eapp ꢁ ꢀ1.1 V,
whereas very negative potentials should be applied to
obtain a modest yield of toluene if no acid is added.
The variation of the product distribution as a function
of Eapp depends on the nature of the acid, stronger acids
favouring formation of toluene. This is better illustrated
of PhCH2 (Eq. 6). There are, however, at least two
observations against such anÅinterpretation. Firstly, the
reduction potential of PhCH2 is not affected by the addi-
tion of strong acids as evidenced by CV investigations
with CH3CO2H. Secondly, the maximum effect of a chem-
ical reaction following an ET (EC process) is known to be
a positive shift of Ep of ca. 0.3 V.12 Moreover, the reduc-
Å
tion of PhCH2 follows an EC mechanism with a fast
chemical reaction, even without added acids.8 Therefore,
added acids would not cause significant shift of the reduc-
tion potential of the raꢀdical if their role were merely that
of protonating PhCH2 in place of H2O or CH3CN.
Figure 2. Distribution of electrocatalytic reduction products of benzyl
bromide at Ag in the absence ((r) RH, (Æ) RR) and presence of acetic
acid ((d) RH, (s) RR) or phenol ((j) RH, (h) RR). The lines were
drawn to show the trends.
Figure 3. Electrocatalytic reduction of benzyl bromide at Ag in the
presence of acids: dependence of toluene yield, obtained at ꢀ1.0 V (d)
and ꢀ1.4 V (j), on pKa of the acids in CH3CN.

7738
A. A. Isse et al. / Tetrahedron Letters 47 (2006) 7735–7739
Table 2. Electrolysis of benzyl bromide (20 mM) in CO2-saturated CH3CN + 0.1 M (C2H5)4NClO4 at a silver cathode
RHb (%)
RRb (%)
RCO2Hb (%)
a
Entry
Eapp (V)
n (F/mol)
1
2
3
4
5
ꢀ1.0
ꢀ1.2
ꢀ1.4
ꢀ1.5
ꢀ1.7
1.1
1.4
1.8
1.8
2.0
10
15
10
15
14
80
60
32
21
13
6
20
55
60
71
a Applied potential versus SCE.
b Yield calculated with respect to converted PhCH2Br.
It appears, therefore, that in the presence of a strong
acid a new reaction pathway comes into play, in compe-
tition with the reaction sequence 3–6. The foÅrmation of
toluene may be achieved by an ET to PhCH2 concerted
with a proton transfer (Eq. 7), instead of two separate
steps. Although concerted proton–electron transfers
are not very common in the electrochemical literature,
a few examples have recently been reported.13 The rate
of reaction 7 depends on pKa of the acid HA. It is there-
fore in competition with reactions 5 and 6 and becomes
important only with strong acids capable of releasing
easily protons.
principal reaction pathway involves a 1eꢀ reduction
with radical–radical coupling. As Eapp becomes more
negative, the charge consumption tends towards the
2eꢀ/molecule of PhCH2Br with a high production of
phenylacetic acid. The electrocarboxylation of organic
halides is an interesting method of synthesis of carbox-
ylic acids. In this respect, Ag is a good cathode material,
even if Eapp values more negative than the reduction po-
tential of the radical are to be applied in order to obtain
satisfactory yields of carboxylate.
In conclusion, the results obtained from the electrocata-
lytic reduction of benzyl bromide at an Ag cathode show
that it is possible to control the selectivity of the reduc-
tion process by varying parameters such as Eapp and
composition of the reaction medium. Thanks to the
remarkable electrocatalytic properties of the Ag surface,
the reduction of PhCH2Br occurs at potentials well po-
sitive of the reduction potential of the benzyl radical.
Thus, 1eꢀ or 2eꢀ reduction of the halide to bibenzyl
or toluene, respectively, can be achieved by appropri-
ately choosing the Eapp value. The radical process at
ꢀ1.0 V is quite selective (RH to RR yield ratio is smaller
than 0.1) and results in a high yield of dimer. In con-
trast, neither the yield nor the selectivity of the 2eꢀ
reduction process at more negative potentials is so good.
However, a 2eꢀ process with a very high yield of toluene
can be achieved, even at the least negative potentials, if a
good proton donor such as a carboxylic acid is added.
ꢀ
Å
PhCH2 þ eꢀ þ HA ! PhCH3 þ A
ð7Þ
Another possible reaction pathway explaining the role
of acids may be proposed in analogy with the behaviour
of some organometallic complexes, which undergo
hydrolysis with the heterolytic rupture of carbon–metal
bonds.14 If we assume that the dissociative ET reaction 3
involves adsorbed intermediates, an adsorbed benzyl-sil-
ver species, which is liable to protonation, would be
formed. The protonation step involves a heterolytic rup-
ture of the silver–carbon bond and an ET to the posi-
tively charged Ag site on the electrode surface. The
process may be considered to occur either in a stepwise
manner or in a single step. The overall hydrolysis reac-
tion may be represented as follows:
(Ag–CH2Ph)surf + eꢀ + HA!Agsurf + PhCH3 + Aꢀ
ð8Þ
where the subscript surf stands for species bound to the
electrode surface.
The electrocarboxylation of benzyl bromide at Eapp
<
ꢀ1.50 V versus SCE gives phenylacetic acid with satis-
factory yields. Overall, the results reported in this paper
put into evidence the importance of the catalytic role of
silver surface in the electroreduction of benzyl bromide,
which makes Ag a promising cathode material for dehy-
drohalogenation of organic halides as well as electrosyn-
thesis processes involving such compounds.
Although the above two reaction pathways are formally
equivalent, both involving protonation accompanied by
an ET, they are very different from the mechanistic
viewpoint. However, it is not possible to discriminate
between these two possibilities on the basis of the data
so far available.
Acknowledgement
This work was financially supported by the Ministero
The role of a good anion scavenger such as CO2 was
also investigated. Controlled-potential electrolyses were
performed in an undivided cell, using an Ag cathode and
an Al sacrificial anode, and the results are reported in
Table 2. Unlike the strong proton donors, CO2 does
not significantly change the reaction mechanism. The
data obtained at different Eapp values are in line with
the reaction sequence 3–6 with an additional reaction
between CO2 and PhCH2ꢀ. At potentials more positive
than the reduction potential of the benzyl radical, the
`
dell’Istruzione, dell’Universita e della Ricerca (MIUR).
References and notes
1. Peters, D. G. In Organic Electrochemistry, 4th ed.; Lund,
H., Hammerich, O., Eds.; Marcel Dekker: New York,
2001; pp 341–377.
2. Rondinini, S.; Mussini, P. R.; Specchia, M.; Vertova, A. J.
Electrochem. Soc. 1998, 148, D102–D107; Liu, Z.; Better-

A. A. Isse et al. / Tetrahedron Letters 47 (2006) 7735–7739
7739
ton, E. A.; Arnold, R. G. Environ. Sci. Technol. 2000, 34,
804–811.
diately before use, by immersing in HNO3 for a few
seconds followed by washing in double-distilled water and
acetone. The products of the electrolyses were analysed by
using a JASCO HPLC model PU-2080, equipped with a
JASCO UV detector and a 25 cm · 4.6 mm Discovery HS
C18 Supelco column. The eluent was a mixture of CH3CN
and H2O containing 5 g/L of CH3CO2H.
´
3. Andrieux, C. P.; Le Gorande, A.; Saveant, J.-M. J. Am.
Chem. Soc. 1992, 114, 6892–6904; Andrieux, C. P.;
´
Saveant, J.-M.; Tallec, A.; Tardivel, R.; Tardy, C. J.
Am. Chem. Soc. 1997, 119, 2420–2429; Cardinale, A.; Isse,
´
A. A.; Gennaro, A.; Robert, M.; Saveant, J.-M. J. Am.
Chem. Soc. 2002, 124, 13533–13539.
8. Since the pKa of toluene is more than 10 orders of
magnitude greater than that of CH3CN or residual water,
4. (a) Fry, A. J.; Krieger, R. L. J. Org. Chem. 1976, 41, 54–
57; (b) Vieira, K. L.; Peters, D. G. J. Org. Chem. 1986, 51,
1231–1239; (c) Cleary, J. A.; Mubarak, M. S.; Vieira, K.
L.; Anderson, M. R.; Peters, D. G. J. Electroanal. Chem.
1986, 198, 107–124; (d) Fry, A. J.; Powers, T. A. J. Org.
Chem. 1987, 52, 2498–2501; (e) Andrieux, C. P.; Gallardo,
ꢀ
PhCH2 is protonated to toluene as soon as it is formed.
See: Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456–463.
9. Sim, B. A.; Milne, P. H.; Griller, D.; Wayner, D. D. M. J.
Am. Chem. Soc. 1990, 112, 6635–6638.
10. Isse, A. A.; Gennaro, A. Indian J. Chem. A 2003, 42, 751–
757.
´
I.; Saveant, J.-M. J. Am. Chem. Soc. 1989, 111, 1620–1626.
5. Rondinini, S.; Mussini, P. R.; Muttini, P.; Sello, G.
Electrochim. Acta 2001, 46, 3245–3258; Bellomunno, C.;
Bonanomi, D.; Falciola, L.; Longhi, M.; Mussini, P. R.;
Doubova, L. M.; Di Silvestro, G. Electrochim. Acta 2005,
50, 2331–2341; Isse, A. A.; Falciola, L.; Mussini, P. R.;
Gennaro, A. Chem. Commun. 2006, 344–346.
6. Isse, A. A.; De Giusti, A.; Gennaro, A.; Falciola, L.;
Mussini, P. R. Electrochim. Acta 2006, 51, 4956–4964.
7. All electrochemical experiments were performed at 25 ꢁC
in CH3CN + 0.1 M (C2H5)4NClO4, using an EG&G
PARC Model 173 potentiostat equipped with a digital
coulometer and a LeCroy LT322 oscilloscope. A three-
electrode configuration was used with a silver disc (2 mm
diameter) working electrode, a AgjAgIjIꢀ (0.1 M in DMF)
reference electrode and a platinum wire as counter
electrode. The electrolyses were performed at a controlled
potential in a two-compartment cell. The cathode was an
Ag cylinder of 12 cm2 area, which was activated, imme-
11. The pKa values of the acids in acetonitrile were taken from
(a) Kosuke, I. Acid–base dissociation constants in dipolar
aprotic solvents; Blackwell Scientific Publications: Oxford,
1990; For trifluoroethanol, a value in DMSO is reported;8
the pKa value in CH3CN was estimated using the
correlations of data in DMF, DMSO and CH3CN
reported in (b) Daasbjerg, K. Acta Chem. Scand. 1995,
49, 878–887.
12. Bard, A. J.; Faulkner, L. R. Electrochemical Methods, 2nd
ed.; Wiley: New York, 2001; p 496.
13. Haddox, R. M.; Finklea, H. O. J. Phys. Chem. B 2004,
108, 1694–1700; Costentin, C.; Evans, D. H.; Robert, M.;
´
Saveant, J.-M.; Singh, P. S. J. Am. Chem. Soc. 2005, 127,
12490–12491.
14. Zhou, D.-L.; Walder, P.; Scheffold, R.; Walder, L. Helv.
Chim. Acta 1992, 75, 995–1011; Fry, A. J.; Singh, A. H. J.
Org. Chem. 1994, 59, 8172–8177; Isse, A. A.; Gennaro, A.;
Vianello, E. J. Chem. Soc., Dalton Trans. 1996, 1613–1618.