The electrochemical reduction of 1,4-dichloroazoethanes: Reductive elimination of chloride to form aryl azines

Article abstract of DOI:10.1016/j.electacta.2010.04.080

A series of 1,4-dichloroazoethanes (1-X/Y, X and Y = 4-NO₂, 4-CN, 4-CH₃ or 4-H) were studied in N,N-dimethylformamide using cyclic voltammetry, constant potential sweep voltammetry (CPSW) and constant potential electrolysis. The voltammograms of 1-X/Y exhibit an irreversible two-electron wave corresponding to dissociative electron transfer (DET) reduction of the carbon-chlorine bond resulting in formation of the azines 2-X/Y in quantitative yield. Additional redox waves correspond to the reversible reduction of the azines to the 2-X/Y^?- radical anion and 2-X/Y²-dianion consecutively, with the exception of 1-NO ₂/NO₂ where both NO₂ groups are reduced simultaneously in a two-electron reversible wave. Thermodynamic and kinetic parameters were determined from CPSW: the standard reduction potentials (E ^o) vary between-0.7 and-1.3V versus SCE as a function of electron-withdrawing substituent; the heterogeneous rate constants (khet) are consistent with a slow heterogeneous electron transfer with values ranging from 10^-3 to 10^-5 cms^-1; the transfer coefficients (α) for 1-NO₂/NO₂ and 1-NO₂/H are greater than 0.5, indicative of a stepwise DET mechanism for the C-Cl bond cleavage while the remaining 1-X/Y compounds have α values between 0.35 and 0.5, and the intrinsic barriers are all significantly lower than predicted for a concerted DET, thereby also suggesting a stepwise DET mechanism.

Full text of DOI:10.1016/j.electacta.2010.04.080

Electrochimica Acta 55 (2010) 5584–5591
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
The electrochemical reduction of 1,4-dichloroazoethanes:
Reductive elimination of chloride to form aryl azines
Vittorio A. Sauro, David C. Magri, Jason L. Pitters, Mark S. Workentin∗
Department of Chemistry, The University of Western Ontario, London, ON, Canada N6A 5B7
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 8 March 2010
Received in revised form 19 April 2010
Accepted 21 April 2010
Available online 29 April 2010
A series of 1,4-dichloroazoethanes (1-X/Y, X and Y = 4-NO2, 4-CN, 4-CH3 or 4-H) were studied in
N,N-dimethylformamide using cyclic voltammetry, constant potential sweep voltammetry (CPSW) and
constant potential electrolysis. The voltammograms of 1-X/Y exhibit an irreversible two-electron wave
corresponding to dissociative electron transfer (DET) reduction of the carbon–chlorine bond resulting in
formation of the azines 2-X/Y in quantitative yield. Additional redox waves correspond to the reversible
reduction of the azines to the 2-X/Y•− radical anion and 2-X/Y2− dianion consecutively, with the excep-
tion of 1-NO2/NO2 where both NO2 groups are reduced simultaneously in a two-electron reversible wave.
Thermodynamic and kinetic parameters were determined from CPSW: the standard reduction potentials
(Eo) vary between −0.7 and −1.3 V versus SCE as a function of electron-withdrawing substituent; the het-
erogeneous rate constants (khet) are consistent with a slow heterogeneous electron transfer with values
ranging from 10−3 to 10−5 cm s−1; the transfer coefficients (˛) for 1-NO2/NO2 and 1-NO2/H are greater
than 0.5, indicative of a stepwise DET mechanism for the C–Cl bond cleavage while the remaining 1-X/Y
compounds have ˛ values between 0.35 and 0.5, and the intrinsic barriers are all significantly lower than
predicted for a concerted DET, thereby also suggesting a stepwise DET mechanism.
Keywords:
1,4-Dichloroazoethanes
Azines
Cyclic voltammetry
Electron transfer
Radical ion
© 2010 Elsevier Ltd. All rights reserved.
1. Introduction
radical-anion intermediate species, in which case the mechanism
proceeds by a stepwise DET mechanism. Alkyl halides and aryl
The azo (N N) bond is a relatively labile group often accompa-
nied by the loss of N2 under a variety of thermal and photochemical
conditions [1–7]. The fragmentation of the C–N bond is rational-
ized to occur by homolytic cleavage upon photolysis via a radical
pathway [4–8]. Conversely, the electrochemical reduction of azo
compounds in aprotic solvents generally involves two sequen-
tial one-electron reductions of the N N to the radical anion and
dianion, respectively [5]. In the presence of a proton donor, the elec-
trochemical reduction results in the transformation of the N N to
the hydrazone NH–NH by a two-electron mechanism [5,9].
Homolytic cleavage is also typical of carbon–halide (C–X) bonds.
The electrochemical reduction of C–X bonds has long been used as
model systems for developing dissociative electron transfer (DET)
theory [11–14]. Dissociative reduction of a molecule with a weak
␴ bond results when electron transfer (ET) is accompanied by
bond cleavage. Provided ET and bond breaking occur in a single
step, within the timeframe of a vibration of a bond, concerted DET
results on acceptance of the electron in the ␴* orbital. However, ET
and bond cleavage may occur in two successive steps involving a
halides are the classic examples of compounds to undergo both
concerted and stepwise DET, respectively [15,16].
The first family of compounds shown to undergo both concerted
and stepwise DET was the benzyl halides [11,12]. Benzyl halides
have an accessible ␲* orbital available in addition to the ␴* orbital
of the C–X bond. The nitro-substituted benzyl halides (bromides
or chlorides) react via a stepwise reductive cleavage while the less
electron-withdrawing substituents (including CN, H, CH3) follow
a concerted pathway. An important parameter for distinguishing
between a stepwise and concerted DET mechanism is the intrinsic
barrier (ꢀGo=/ ), that is the activation free energy when the driving
force for the reaction is zero. In the limiting case of concerted DET,
the (ꢀGo=/ ) is much larger as it contains contributions from both
the reorganization energy (ꢁo) and the bond dissociation energy
(BDE) from the cleaving bond. However, the (ꢀGo=/ ) is significantly
smaller when the bond cleavage step is an intramolecular DET
from an initially formed radical ion of the antenna group to the
␴* of the breaking bond [12]. Many other bonds have been shown
in recent years to undergo DET including molecular systems with
O–O [12,17] and S–S [12a,18] bonds. This latter body of work has
validated and further assisted in developing Savéant’s DET theory
[11].
∗
We were interested in examining the electrochemical reduc-
Corresponding author. Tel.: +1 519 661 2111x86319; fax: +1 519 661 3022.
E-mail address: mworkent@uwo.ca (M.S. Workentin).
tion of 1,1ꢀ-dichloro-1,1ꢀ-diaryl-1,1ꢀazoethanes, more generally
0013-4686/$ – see front matter © 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2010.04.080

V.A. Sauro et al. / Electrochimica Acta 55 (2010) 5584–5591
5585
Scheme 1. The photodissociation of 1,4-dichloroazoethanes via C–N bond cleavage.
referred to as 1,4-dichloroazoethanes. In addition to an azo moiety,
2.2. Instrumentation
these compounds also have two reducible benzyl halide groups. In
previous photochemical studies of 1,4-dichloroazoethanes, azines
(2,3-diaza derivatives of butadienes –C N–N C–) were isolated
as products [1,5,10]. The reaction was also found to yield a sig-
nificant amount of radical coupling and oxygenated products
due to the formation of a carbon radical intermediate with loss
of molecular nitrogen (Scheme 1) [5,10]. Considering that 1,4-
dichloroazoethanes exhibit the loss of dinitrogen upon irradiation
via cleavage of the C–N bonds, we were interested in examin-
ing if under electrochemical conditions a competition would exist
between C–Cl versus C–N reductive cleavage, and whether this
competition could be controlled by a change in the aryl substituent.
In this work, the electrochemical reduction of a series of seven
symmetrical and unsymmetrical 1,4-dichloroazoethanes (1-X/Y,
where X and Y are either 4-NO2, 4-CN, 4-CH3 or 4-H) were studied
in N,N-dimethylformamide using cyclic voltammetry, convolution
potential sweep voltammetry (CPSV) and constant potential elec-
trolysis. These compounds provide an unprecedented example of
systems containing two halides, not on adjacent carbon atoms,
that form azine products upon electrochemical reduction. Bulk
electrolysis experiments revealed that dissociative reduction of
1,4-dichloroazoethanes results in the loss of both chlorine atoms
to yield azines in quantitative yield (Scheme 2). CPSV and appli-
cation of DET theory allowed for the determination of a number
of thermodynamic and kinetic parameters including the standard
reduction potential (Eo) and standard heterogeneous rate constant
(khoet). Irrespective of the substituents, rather low intrinsic barri-
ers were determined throughout the series, thereby suggesting
a stepwise DET mechanism in which the azo group significantly
contributes to lowering the LUMO energy.
NMR spectra were recorded on a Varian Gemini 300 MHz
(300.08 MHz for 1H and 75.46 MHz for 13C) or a Varian Mercury
400 MHz (400.08 MHz for 1H and 100.60 MHz for 13C) instru-
ment. Chemical shifts were measured in deuterated chloroform
and reported in parts per million downfield from tetramethylsilane.
Gas chromatography analysis was performed on a Hewlett-Packard
5890 Series II gas chromatograph equipped with a 10 meter HP-5
column and an FID detector. UV–vis spectra were recorded on a
Varian Cary 100 spectrometer. Melting points were recorded using
a Gallenkamp melting point apparatus and are uncorrected. Ele-
mental analysis was performed by Guelph Chemical Laboratories
Ltd., Guelph, Ontario, Canada.
2.3. Cyclic voltammetry
Experiments were performed using either a Princeton Applied
Research (PAR) 283 or 263 potentiostat interfaced to a personal
computer running PAR M270 software. Measurements were con-
ducted in a water-jacketed glass cell maintained at a constant
temperature of 25 ◦C under an argon atmosphere. The glass cell
was stored overnight in an oven maintained at 110 ◦C and assem-
bled hot while flushing with argon. The working electrode was
a 3 mm glassy carbon rod (Tokai) sealed in a glass tube. It was
polished with 1 ␮m diamond paste, cleaned by sonication in 2-
propanol and dried with a cool stream of air. The counter electrode
was a 1 cm2 platinum flag. The quasi-reference electrode was a sil-
ver wire contained in a glass tube sealed with a porous ceramic
tip and filled with a 0.1 M solution of TEAP. Ferrocene (Eo = 0.470 V
versus SCE in DMF) was added to the solution afterwards as
an internal reference to calibrate the potentials versus the sat-
urated calomel electrode. Positive feedback internal resistance
compensation was applied to minimize the effects of the solution
resistance. Linear sweep voltammograms for convolution poten-
tial sweep voltammetry were recorded using the same set-up.
Convolution analysis was performed using custom-made analysis
software.
2. Experimental
2.1. Materials
All chemicals were purchased from Aldrich and used as
received unless otherwise stated. Spectroscopic grade N,N-
dimethylformamide (DMF) was distilled under a nitrogen atmo-
sphere at reduced pressure over calcium hydride. Tetraethylam-
monium perchlorate (TEAP), from Kodak, was recrystallized three
times from ethanol and dried under vacuum at 60 ◦C and stored in
a vacuum dessicator. Ferrocene was purified by sublimation.
2.4. Constant potential coulometry
100 mg of a 1,4-dichloroazoethane was dissolved in 25 mL of
0.1 M TEAP/DMF solution maintained at a constant temperature
Scheme 2. The reaction equation for the reduction of 1,4-dichloroazoethanes 1-X/Y to 2-X/Y azines.

5586
V.A. Sauro et al. / Electrochimica Acta 55 (2010) 5584–5591
of 25 ◦C. Argon was continuously bubbled into solution with vig-
orous stirring to eliminate oxygen. The working electrode was a
platinum mesh or rotating disk electrode with a 12 mm glassy
carbon tip. The counter electrode was a platinum coil separated
from the bulk solution by a 2 mm thick layer of activated alumina
and a sintered glass disk. The appropriate constant potential was
applied using either a Princeton Applied Research (PAR) 283 or
263 potentiostat. Exhaustive electrolysis was monitored by cyclic
voltammetry. Afterwards, the DMF was removed in vacuo while
keeping the electrolysis mixture under 40 ◦C. The recovered solid
mixture was dissolved in dichloromethane and distilled water and
transferred to a separatory funnel. The aqueous layer was extracted
three times. The organic layers were combined, washed with dis-
tilled H2O twice, dried with MgSO4, filtered and evaporated under
vacuum. The remaining solid residue was analyzed by GC and 1H
NMR.
elemental analysis, calc. %C(61.46), H(4.56), det. %C(61.27),
H(4.37).
1,1ꢀ-Dichloro-1-(4-nitrophenyl)-1ꢀ-phenyl-1,1ꢀ-azoethane
(1-
NO2/H): [20] m.p. = 83–84 ◦C decomposition with gas evolution;
1H NMR (400 MHz, CDCl3: ıH 2.19(s, 3H), 2.23(s, 3H), 7.39(m,
3H), 7.54(d, 2H), 7.80(d, 2H), 8.25(d, 2H); 13C NMR (100 MHz,
CDCl3): ıC 147.9, 140.4, 129.0, 128.6, 127.7, 126.3, 123.7, 94.7, 93.0,
30.6, 30.1; UV–vis (MeCN): ꢁmax = 245 nm, ε = 6150 cm−1 M−1
;
elemental analysis, calc. %C(54.56), H(4.30), det. %C(54.21),
H(4.40).
2.5. Syntheses of the 1,4-dichloroazoethanes
All 1-X/Y were synthesized in a fumehood by chlorination of
the corresponding acetophenone azines (2-X/Y) and purified by
recrystallization [19–21]. Excess chlorine gas was condensed into a
round-bottom flask containing neat 2-X/Y or a suspension of 2-X/Y
in CH2Cl2. The suspension was stirred at −60 ◦C for 0.5 h and then
slowly warmed to room temperature. During this time, most of the
Cl2 gas evaporates from the reaction mixture. Any remaining gas
was removed under vacuum. The crude 1-X/Y was recrystallized
from a hexanes/ethyl acetate mixture with slight warming (below
30 ◦C). The recrystallization solution of 1-X/Y was cooled in dry ice
and the resulting solid filtered and dried under vacuum. The purity
of the 1-X/Y was confirmed by melting point determination and
elemental analysis.
1,1ꢀ-Dichloro-1,1ꢀ-diphenyl-1,1ꢀ-azoethane
(1-H/H):
[19]
m.p. = 98–99 ◦C decomposition with gas evolution; 1H NMR
(300 MHz, CDCl3): ıH 2.22(s, 6H), 7.37(m, 6H), 7.58(m, 4H); 13C
NMR (75 MHz, CDCl3): ıC 141.0, 128.7, 128.6, 126.4, 94.5, 30.1;
UV–vis (MeCN): ꢁmax = 235 nm, ε = 1980 cm−1 M−1
; elemental
analysis, calc. %C(62.55), H(5.26), det. %C(62.54), H(5.70).
1,1ꢀ-Dichloro-1,1ꢀ-bis(4-methylphenyl)-1,1ꢀ-azoethane (1-CH3/
CH3): [10] m.p. = 118–119 ◦C decomposition with gas evolution;
1H NMR (300 MHz, CDCl3): ıH 2.21(s, 6H), 2.35(s, 6H), 7.19(d, 4H),
7.44(d, 4H); 13C NMR (75 MHz, CDCl3): ıC 140.0, 131.8, 128.2, 123.2,
93.9, 30.2; UV–vis (MeCN): ꢁmax = 234 nm, ε = 12400 cm−1 M−1
;
elemental analysis, calc. %C(64.48), H(6.02), det. %C(64.58), H(6.24).
1,1ꢀ-Dichloro-1,1ꢀ-bis(4-nitrophenyl)-1,1ꢀ-azoethane
(1-NO2/
NO2): [10] m.p. = 156–159 ◦C decomposition with gas evolution;
1H NMR (300 MHz, CDCl3): ıH 2.20(s, 6H), 7.81(d, 4H), 8.25(d, 4H);
13C NMR (75 MHz, CDCl3): ıC 147.4, 130.8, 127.6, 123.8, 93.4, 30.7;
UV–vis (MeCN): ꢁmax = 278 nm, ε = 17284 cm−1 M−1; elemental
analysis, calc. %C(48.38), H(3.56), det. %C(50.07), H(3.57). The
discrepancy in %C is due to trace recrystallization solvent in the
sample, evidenced by the NMR spectrum. This solid was dried
under vacuum and its purity was verified by NMR spectroscopy
prior to use.
1,1ꢀ-Dichloro-1,1ꢀ-bis(4-cyanophenyl)-1,1ꢀ-azoethane (1-CN/CN):
m.p. = 114–117 ◦C decomposition with gas evolution; 1H NMR
(300 MHz, CDCl3): ıH 2.17(s, 6H), 7.73(m, 8H); 13C NMR (75 MHz,
CDCl3): ıC 145.7, 132.4, 127.3, 118.1, 112.9, 93.5, 30.5; UV–vis
(MeCN): ꢁmax = 266 nm, ε = 3150 cm−1 M−1; elemental analysis,
calc. %C(60.52), H(3.96), det. %C(60.70), H(3.99).
1,1ꢀ-Dichloro-1-(4-cyanophenyl)-1ꢀ-phenyl-1,1ꢀ-azoethane
(1-
CN/H): m.p. = 79–80 ◦C decomposition with gas evolution; 1H
NMR (400 MHz, CDCl3): ıH 2.17(s,3H), 2.22(s,3H), 7.38(m, 3H),
7.54(d, 2H), 7.72(m, 4H); 13C NMR (100 MHz, CDCl3): ıC 146.1,
140.4, 132.4, 128.9, 128.6, 127.3, 126.3, 118.3, 112.7, 94.7, 93.2,
Fig. 1. Cyclic voltammograms of 2.0 mM (a) 1-H/CH3, (b) 1-CH3/CH3, (c) 1-H/H, (d)
1-CN/H, (e) 1-CN/CN, (f) 1-NO2/H and (g) 1-NO2/NO2 in 0.1 M TEAP/DMF at 0.2 V s−1
with a glassy carbon electrode. The arrows indicate the potential sweep direction in
all the CVs.
30.4, 30.1; UV–vis (MeCN): ꢁmax = 266 nm, ε = 2000 cm−1 M−1
;

V.A. Sauro et al. / Electrochimica Acta 55 (2010) 5584–5591
5587
1,1ꢀ-Dichloro-1-(4-methylphenyl)-1ꢀ-phenyl-1,1ꢀ-azoethane (1-
CH3/H): m.p. = 92–93 ◦C decomposition with gas evolution; 1H
NMR (400 MHz, CDCl3): ıH 2.20(s, 6H), 2.34(s, 3H), 7.17(d, 2H),
7.39(m, 5H), 7.55(m, 2H); 13C NMR (100 MHz, CDCl3): ıC 141.0,
138.7, 137.9, 129.2, 128.7, 128.5, 126.4, 126.3, 94.5, 94.4, 30.1,
Table 2
Electrochemical data for the initial irreversible reduction wave for 1-X/Y in 0.1 M
TEAP/DMF solution.
−∂Ep/∂log ꢂc
nd
a,b
c,b
1-X/Y
Ep/2
|Ep/2 − Ep|
NO2
NO2
CN
CN
H
NO2
H
CN
H
H
CH3
H
−0.735
−0.763
−0.975
−1.061
−1.145
−1.164
−1.166
60
69
99
100
105
109
109
38
47
87
86
86
82
82
1.92
1.94
2.02
2.01
2.09
2.16
2.01
29.9, 21.1; UV–vis (MeCN): ꢁmax = 218 nm, ε = 8220 cm−1 M−1
;
elemental analysis, calc. %C(63.55), H(5.66), det. %C(63.36),
H(5.57).
CH3
CH3
3. Results
a
V versus SCE in DMF/0.1 M TEAP.
At 0.1 V s−1 scan rate.
mV.
b
c
The syntheses of the 1-X/Y compounds were performed by
chlorination of the corresponding acetophenone azines [19–21].
The details and spectral characterization appear in Section 2. The
electrochemistry of the series of seven 1-X/Y compounds were
investigated in N,N-dimethylformamide (DMF) containing 0.1 M
tetraethylammonium perchlorate (TEAP) at a glassy carbon elec-
trode. Fig. 1 illustrates the reductive cyclic voltammograms for the
complete series of compounds at 0.2 V s−1. Each voltammogram is
characterized by an initial two-electron irreversible cathodic wave
A, followed by a second reversible wave B (See Fig. 1c). This second
wave corresponds to a one-electron reduction of the product 2-X/Y
compounds, with the exception of 1-NO2/NO2 where this reversible
wave corresponds to a two-electron reduction of the correspond-
ing 2-NO2/NO2. With the exception of 1-NO2/NO2, a third reduction
wave is observed at more negative potentials. When at least one of
the substituents is a cyano or nitro group, wave C is a reversible cou-
ple (Fig. 1d–f). Table 1 summarizes the observed peak potentials for
wave A, the standard potentials for wave B, and the peak potentials
(Fig. 1a–c) and standard potentials (Fig. 1d–f) for wave C. An iden-
tifiable trend is that as the substituents attached to the phenyl ring
become more electron-donating, the waves corresponding to A and
B are observed at more cathodic potentials.
Table 2 lists specific data corresponding to wave A for the 1-
X/Y series. In each case, the first reduction wave is irreversible at
all scan rates investigated (up to 25 V s−1). The peak potential (Ep)
of wave A was observed to shift to more negative potentials upon
increasing the scan rate (ꢂ). The scan rate normalized peak currents
(ip/ꢂ1/2) were found to decrease with increasing scan rate, typi-
cal for systems undergoing DET controlled by slow heterogeneous
kinetics. The peak width at half current height |Ep/2 − Ep| follows a
trend of increasing from 60 to 109 mV as the substituents become
more electron-donating. The variation of Ep with the logarithm of
the scan rate (log ꢂ) is linear with the slope ranging from 38 to 87.
Furthermore, addition of a non-nucleophilic proton donor, such as
trifluoroacetic acid (TFE), has no effect on wave A [22].
d
Number of electrons determined by constant potential electrolysis in F mol−1
.
Product studies and comparison with authentic samples con-
firmed that waves B and C correspond to the formation of the
acetophenone azines (2-X/Y) [21]. Quantitative formation of 2-
X/Y was observed by constant potential electrolysis experiments,
where solutions of 1-X/Y were electrolyzed at a potential slightly
negative to the Ep of wave A until no trace of 1-X/Y remained (see
Fig. 2). In each case, 2 F mol−1 was consumed (Table 2). The cyclic
voltammograms after exhaustive electrolysis were identical to the
cyclic voltammograms of authentic 2-X/Y [21]. The products were
isolated from the electrolysis mixture and confirmed to be 2-X/Y
using 1H NMR spectroscopy, GC and MS analysis. Hence, the sec-
ond and third reductive waves observed in the CVs of 1-X/Y are
due to the one-electron reductions of the corresponding 2-X/Y rad-
ical anions and dianions, respectively, with the exception of the
2-NO2/NO2 where the second wave is a two-electron reduction
[21].
The symmetry factors (˛ values) for wave A, which is the initial
heterogeneous ET to the 1,4-dichloroazoethanes, were evaluated
to gain insight into the reaction mechanism. Three independent
approaches were employed including the peak width at each scan
rate (Eq. (1)), the linear regression of Ep over a range of scan rates
to obtain an average value (Eq. (2)), and the apparent value from
the potential dependence of ln khet from CPSV analysis (Eq. (3))
[11,24]. The first two methods are simpler in that only a data point
Table 1
Reduction potentials of the 1,4-dichloroazoethanes in 0.1 M TEAP/DMF solution at
a glassy carbon electrode.
E(oB)
E(oC)
a,b
a,b,c
a,b,d
Fig. 1, label
1-X/Y
Ep(A)
X
Y
e
g
f
NO2
NO2
CN
CN
H
NO2
H
CN
H
H
CH3
H
−0.795
−0.832
−1.074
−1.161
−1.250
−1.273
−1.275
−0.937
−0.972
−1.377
−1.559
−1.902
−1.987
−1.941
–
−1.526
−1.597
−1.925
−2.295f
−2.394f
−2.337f
e
d
c
b
a
CH3
CH3
a
V versus SCE in DMF/0.1 M TEAP.
Peak potentials at 0.1 V s−1
b
c
d
e
f
.
Eo for the reversible redox couple of 1-X/Y.
Eo for the third reduction wave of the 1-X/Y unless otherwise noted.
Fig. 2. Cyclic voltammograms before (dashed) and after (solid) the addition of
2 F mol−1 of charge during the constant potential electrolysis of 1-CN/CN in 0.1 M
No third reduction wave observed.
Irreversible at all scan rates up to 25 V s−1
.
TEAP/DMF at 0.2 V s−1
.

5588
V.A. Sauro et al. / Electrochimica Acta 55 (2010) 5584–5591
Table 3
Data derived from convolution sweep potential voltammetry and estimates of the ˛ values for the 1-X/Y series of compounds at a glassy carbon electrode.
E1o-X/Y
D1-X/Y
ꢀGo=/
log khoet
−4.4
a
b
c
d,e
f
g
1-X/Y
˛
˛
avg
˛
app
0.79h
0.79i
0.74j
NO2
NO2
H
−0.599
−0.545
−0.904
−0.981
−1.084
−1.111
−1.075
0.75
0.62
0.34
0.34
0.34
0.36
0.35
0.84
0.71
0.39
0.37
0.35
0.34
0.35
8.3
2.0
0.69h
0.69i
0.68j
NO2
CN
CN
H
−5.4
−3.3
−3.3
−3.1
−3.1
−3.3
9.6
3.8
7.1
6.4
5.8
4.0
4.4
0.48h
0.45i
0.38g
CN
H
8.2
0.47h
0.43i
0.41g
9.2
0.45h
0.43i
0.40g
H
11.0
9.5
0.44h
0.41i
0.39g
CH3
CH3
H
0.44h
0.42i
0.38g
CH3
10.1
a
Determined by Eq. (1) at various scan rates. See footnotes h–j.
Determined by Eq. (2).
Determined by Eq. (3).
b
c
d
e
f
From extrapolation of ˛ data to 0.5 from several experiments.
V versus SCE in DMF/0.1 M TEAP.
10−6 cm2 s−1
.
Determined from the experimental ␣–E data by Eq. (9) in kcal mol−1; for comparison the theoretical stepwise ꢀGo=/ are 3.1, 3.8 and 6.2 kcal mol−1 by the Marcus (Eq.
g
(10)), Hush (Eq. (11)) and Bard (Eq. (12)) approaches, respectively.
h
0.1 V s−1
0.5 V s−1
1.0 V s−1
.
.
.
i
j
or two are examined from the cyclic voltammograms whereas the
last method examines the potential dependence over the entire
set of data. A summary of the ˛ data from all three methods are
presented in Table 3. The 1-NO2/NO2 and 1-NO2/H have ˛ values
ranging from 0.62 to 0.84 depending on the method of evaluation.
The fact that the values are greater than 0.5 suggests the initial
electron transfer occurs via a stepwise mechanism [11]. The other
five 1-X/Y compounds have ˛ values within the range of 0.35–0.50,
which falls within the mixed kinetic region where both a stepwise
and a concerted DET mechanism may be in competition [11,17c].
electrode reaction [24,25]. The log khoet values, estimated from
interpolation of the second order polynomial fit to the heteroge-
neous kinetics are low, but typical for dissociative systems, ranging
from −3.1 to −5.4 [11,12]. Derivatization of the ln khet–E data over
small intervals for each scan rate results in the determination of
the potential dependence of the apparent transfer coefficient, ˛app
(Eq. (3)), uncorrected for the effect of the electrode double-layer.
From the linear ˛–E relationship, the standard reduction poten-
tial was determined by extrapolation to ˛ = 0.5 (Eq. (6)). Despite
the observed reduction being electrochemically irreversible, the
technique allows for the thermodynamically significant E1o-X/Y to
1.85RT
F(Ep/2 − Ep)
be easily determined. The E1o-X/Y, which range from −0.60 to −1.1 V,
˛ =
(1)
(2)
decrease as a function of electron-donating substituent ability.
ꢀ
ꢁ
ꢁ
RT
∂(ln ꢂ)
∂Ep
ꢂ
t
˛
=
−
avg
i(u)
(t − u)1/2
Ilim = ꢃ−1/2
du
(4)
2F
ꢀ
0
RT ∂(ln k)
F
Ilim = nFAD1/2C∗
(5)
(6)
˛
app
= −
(3)
∂E
F
˛ = 0.5 +
(E − Eo)
Convolution potential sweep voltammetry was used to evaluate
thermodynamic and kinetic parameters for the direct reduction of
1-X/Y in 0.1 M TEAP/DMF solution over a range of scan rates from
0.1 to 25 V s−1 [23]. Table 3 also summarizes this data. The tech-
nique involves the subtraction of background currents from the
voltammograms at each corresponding scan rate and the resultant
data transformed into sigmoidal-shaped curves via the convolu-
tion integral (Eq. (4)) [23]. From the limiting plateau current, the
diffusion coefficients were determined to range from 8.3 × 10−6
to 1.1 × 10−5 cm2 s−1 in 0.1 M TEAP/DMF solution (Eq. (5)). The
potential dependence of the heterogeneous rate constant (khet) was
derived from the convoluted current curves. The heterogeneous
kinetics, log khoet, in Table 3 are consistent with an irreversible
8ꢀG0=/
4. Discussion
The standard potentials E1o-X/Y for all the compounds reported
in Table 1 are within 300 mV of the observed peak potential Ep
(wave A). In a stepwise ET mechanism, the peak potential is nor-
mally close to the thermodynamic standard reduction potential
where ˛ = 0.5 by definition [24]. However, in a concerted DET reduc-
tion, the observed peak potential appears at a much more negative
potential due to the large activation overpotential that must be
overcome to stretch and break the cleaving bond, which results in

V.A. Sauro et al. / Electrochimica Acta 55 (2010) 5584–5591
5589
Scheme 3. Possible reaction pathways for the reduction of 1-X/Y to 2-X/Y.
much lower experimental ˛ value, typically below 0.3, for example,
as observed with benzyl chlorides [15].
relatively good agreement with the experimental values, which
range from 2.0 to 7.1 kcal mol−1, thereby suggesting that there is
no apparent contribution from the BDE. Hence, the reasonable fits
with the Marcus model (Eq. (7)) suggests the reduction occurs via
a stepwise DET mechanism throughout the series of compounds.
As previously mentioned, the log khoet values are low ranging
from −3.1 to −5.4. Unexpectedly, the measured kinetics for 1-
NO2/NO2 and 1-NO2/H are the lowest in the series. We believe that
the khet values should be similar to the other 1-X/Y compounds. The
window of accessible ˛ data was quite narrow for 1-NO2/NO2 and
1-NO2/H, covering only 100 mV of data, which results in a greater
degree of error during the determination of the E1o-X/Y from the
ꢀ
ꢁ
ꢃ
ꢄ
e2
1
1
Ds
1
4a
ꢁoMarcus
=
−
(10)
(11)
4ꢃεo Dop
ꢀ
ꢁ
ꢃ
ꢄ
e2
1
1
1
ꢁoHush
=
−
extrapolation of the data and accounts for the more than one order
of magnitude discrepancy.
4ꢃεo Dop
Ds
2a
The application of DET theory was also used to determine the
intrinsic barrier (ꢀGo=/ ), the solvent reorganization energy (ꢁo) and
bond dissociation energy (BDE). The intrinsic barrier is diagnostic
for distinguishing between a concerted and stepwise mechanis-
tic pathway. In the stepwise case, ꢀGo=/ only has a contribution
from ꢁo so that the classic Marcus–Hush model applies (Eq. (7)). For
2.08
ꢁoBard
=
(12)
(13)
a
aCl− (2a5−X/Y − aCl− )
a =
a5-X/Y
Scheme 3 illustrates two possible heterogeneous reduction
mechanisms for the series of 1,4-dichloroazoethanes. The diagnos-
tic thermodynamic parameters suggest the initial 1-X/Y reduction
proceeds via formation of a radical-anion intermediate involving
acceptance of an electron into a delocalized ␲* orbital energy fol-
lowed by an intramolecular dissociative electron transfer to a C–Cl
bond. Afterwards, a number of other elementary reaction steps
occur ultimately resulting in the formation of 2-X/Y and chlo-
ride anions. The unsymmetrical substituted 1-X/Y are expected
to undergo reductive cleavage of the C–Cl bond nearest the most
electron-withdrawing substituent as similarly observed in the
reduction of the C N bond in the 2-X/Y [21].
the concerted case, Savéant’s model is used to evaluate the ꢀGo=/
,
which includes a contribution for the BDE (Eq. (8)). ꢀGo=/ was deter-
mined from the slope of the ␣–E convolution data (Eq. (9)). The
determined ꢀGo=/ (Table 3) values are significantly smaller than
those determined for other C–Cl systems including benzyl chlorides
[15]. The results suggest 1-X/Y undergoes a stepwise reductive
cleavage for all the compounds investigated.
ꢁo
ꢀGo=/
ꢀGo=/
=
=
(7)
(8)
(9)
4
ꢁo + BDE
4
After dissociative reduction of a C–Cl bond, a benzylic carbon
radical is formed. The mechanism could then precede either by a
rapid radical rearrangement leading to the expulsion of a chlorine
atom (route 1) which is subsequently reduced, or reduction of the
carbon radical followed by an anionic rearrangement followed by
elimination of chloride anion (route 2). The mechanistic pathway
depends on whether the carbon radical formed upon C–Cl fragmen-
tation is immediately reduced to an anion intermediate depending
on the overpotential, as in route 2, and the activation barrier that
must be overcome for reorganization of the molecular structure and
•
∂˛
=
1
8ꢀGo=/
∂E
A major difference between the Marcus and Savéant ET models
is the magnitude of ꢀGo=/ . Hence, estimates for ꢁo can be helpful
in distinguishing between a concerted or stepwise DET mecha-
nism [18]. Several approaches exist for estimating ꢁo, as shown
by Eqs. (10–12), based on the molecular and solvent parameters
(where Dop and Ds are the optical and static dielectric constants)
[24]. These approaches provide reasonable estimates of ꢁo when
there is negligible contribution from the internal reorganization
energy. To account for ET at only a portion of the acceptor 1-X/Y
molecule, an effective radius approach was used (Eq. (13)) [11a–b,
17g–h]. The theoretical ꢀGo=/ values are obtained by substituting
the results from Eqs. (10–12) into Eq. (7), which yields 3.1, 3.8 and
6.2 kcal mol−1 by the Marcus (Eq. (10)), Hush (Eq. (11)) and Bard
(Eq. (12)) approaches, respectively. These estimated ꢀGo=/ are in
expulsion of Cl•, which is then readily reduced (ECl
= +1.81 V
−
/Cl
versus SCE) [26] as in case 1. The reduction potentials of the car-
bon radicals can be approximated from the analogous substituted
benzyl radicals [27], although the Eo of each radical is expected to
be substituent dependent. For example, for 1-CN/CN the Ep is at
−1.074 V whereas the Eo for the 4-CN benzyl radical is −0.77 V so
that heterogeneous reduction of the carbon radical is energetically
feasible [27]. Conversely, for 1-H/H the Ep is at −1.250 V whereas

5590
V.A. Sauro et al. / Electrochimica Acta 55 (2010) 5584–5591
the Eo for the benzyl radical is −1.45 V versus SCE [27]. In this case,
reduction of the 1-H/H• is an uphill process with a lower khet. The
inductive effect of an azo group on the Eo of benzyl radicals is not
known, but it is anticipated to shift Eo to more positive potentials
due to its electron-withdrawing properties (ꢄI = 0.30 for –N N–Ph)
[29]. From the mechanistic pathway we would expect to observe
coupling products or other products derived from the radical inter-
mediate (route 1). The absence of this observation suggests the
mechanism proceeds by route 2, although neither can be ruled
out. A third possible mechanism, involves the double simultane-
ous (or successive but at similar potentials) dissociative reduction
with loss of both chlorides to yield the dibenzylic diradical, which
is a minor resonance structure better written as the azine. A sim-
ilar mechanism was ruled out by Savéant and coworkers in the
reduction of a number of vicinal dihalides, otherwise known as 1,2-
dichloro-1,2-diphenylethanes [16f]. Given the stepwise nature of
the dissociative reduction for 1-X/Y such a mechanism is consid-
ered unlikely.
Vicinal dihalides have been also studied by Lund and cowork-
ers [30]. The CVs exhibited an irreversible two-electron reduction
wave followed by a more negative reversible one-electron wave.
The first wave was attributed to reductive cleavage of the C–X
bonds followed by elimination of the second halogen to yield elec-
troactive E-stilbene products. Although they did not apply Savéant
DET theory, they did suggest a concerted DET mechanism. In con-
trast, we show that 1,4-dichloroazoethanes react by a stepwise DET
mechanism. The key difference between the two classes of com-
pounds is the azo group, which appears to contribute to lowering
the ␲* orbital energy level of all the 1-X/Y compounds. Other fam-
ilies of compounds have been shown to have unexpectedly low
intrinsic barriers regardless of the substituent on the aryl moiety.
In the case of diaryl disulfides, the ET mechanism was within the
mixed kinetic region between the two mechanistic extremes of
stepwise and concerted [12a,18]. In our recent study of 3,3,6,6-
tetraaryl-1,2-dioxanes, we discovered that the intrinsic barriers
were inconsistent with a purely concerted DET to the O–O bonds
[17k]. Density functional theory calculations illustrated that the
LUMO was not localized on the O–O bond, but rather delocalized
over a large molecular framework including the equatorial aryl
rings, the adjacent C–O bond and the O–O bond.
ORCDF, and the University of Western Ontario (ADF) is greatly
appreciated. VAS thanks the OGS program for a scholarship. DCM
thanks the OGSST program for a scholarship.
References
[1] P.S. Engel, Chem. Rev. 80 (1980) 99.
[2] P.S. Engel, S. Duan, G.B. Arhancet, J. Org. Chem. 62 (1997) 3537.
[3] P.S. Engel, S.L. He, W.B. Smith, J. Am. Chem. Soc. 119 (1997) 6059.
[4] S.K. Danek, D.P. Kelly, A.K. Serelis, J. Org. Chem. 52 (1987) 2911.
[5] S. Patai, in: S. Patai (Ed.), The Chemistry of the Hydrazo, Azo and Azoxy Groups,
John Wiley & Sons, New York, 1975.
[6] C.F. Billera, T.B. Dunn, D.A. Barry, P.S. Engel, J. Org. Chem. 63 (1998) 9763.
[7] P.S. Engel, L. Pan, K.H. Witmire, I. Guzman-Jimenez, J. Org. Chem. 65 (2000)
1016.
[8] S. Engel, L. Pan, Y. Ying, L.B. Alemany, J. Am. Chem. Soc. 123 (2001) 3706.
[9] J.P. Stradins, V.T. Glezer, in: A.J. Bard, H. Lund (Eds.), Azo, Azoxy and Diazo
Compounds, vol. 13, Marcel Dekker Inc., New York, 1979.
[10] N. Levi, D.S. Malament, J. Chem. Soc. Perkin Trans. II (1976) 1249.
[11] (a) J.-M. Savéant, in: D. Bethel (Ed.), Advances in Physical Organic Chemistry,
vol. 26, Academic Press, New York, 1990, p. 1;
(b) J.-M. Savéant, in: P.S. Mariano (Ed.), Advances in Electron Transfer Chem-
istry, vol. 4, JAI Press, Greenwich, CT, 1994, p. 53;
(c) J.-M. Savéant, Acc. Chem. Res. 26 (1993) 455;
(d) J.-M. Savéant, in: T. Tidwell (Ed.), Advances in Physical Organic Chemistry,
vol. 35, Academic Press, New York, 2000, p. 117;
(e) C. Costentin, M. Robert, J.-M. Savéant, Chem. Phys. 324 (2006) 40;
(f) C. Costentin, L. Donati, M. Robert, Chem. Eur. J. 15 (2009) 785.
[12] (a) F. Maran, D.D.M. Wayner, M.S. Workentin, in: T.T. Tidwell (Ed.), Advances
in Physical Organic Chemistry, vol. 36, Academic Press, New York, 2001, p. 85;
(b) S. Antonello, F. Maran, Chem. Soc. Rev. 34 (2005) 418.
[13] R.A. Rossi, A.B. Pierini, A.B. Penenory, Chem. Rev. 103 (2003) 71.
[14] A. Houmam, Chem. Rev. 108 (2008) 2180.
[15] Examples of aryl halides:
(a) C.P. Andrieux, I. Gallardo, J.M. Savéant, K.B. Su, J. Am. Chem. Soc. 108 (1986)
638;
(b) C.P. Andrieux, A.L. Gorande, J.-M. Savéant, J. Am. Chem. Soc. 114 (1992)
6892;
(c) C.P. Andrieux, C. Comgellas, J. Kanoufi, J.-M. Savéant, A. Thiebault, J. Am.
Chem. Soc. 119 (1997) 9527;
(d) C.P. Andrieux, E. Differding, M. Robert, J.-M. Savéant, J. Am. Chem. Soc. 115
(1993) 6592;
(e) C.P. Andrieux, J.-M. Savéant, A. Tallec, R. Tardivel, C. Tardy, J. Am. Chem. Soc.
119 (1997) 2420;
(f) R. Fuhlendorff, D. Occhialini, S.U. Pedersen, H. Lund, Acta. Chem. Scand. 43
(1989) 803;
(g) D. Occhialini, S.U. Pedersen, H. Lund, Acta. Chem. Scand. 44 (1990) 715;
(h) D. Occhialini, J.S. Kristensen, K. Daasbjerg, H. Lund, Acta. Chem. Scand. 46
(1992) 474.
[16] Examples of alkyl halides:
(a) L. Pause, M. Robert, J.-M. Savéant, J. Am. Chem. Soc. 122 (2000) 9829;
(b) A. Cardinale, A.A. Isse, A. Gennaro, M. Robert, J.-M. Savéant, J. Am. Chem.
Soc. 124 (2002) 13533;
5. Conclusions
(c) K. Daasbjerg, S.U. Pedersen, H. Lund, Acta Chem. Scand. 43 (1989) 876;
(d) T. Lund, H. Lund, Acta Chem. Scand. B 41 (1987) 93;
(e) L. Pause, M. Robert, J.-M. Savéant, J. Am. Chem. Soc. 121 (1999) 7158;
(f) D. Lexa, J.-M. Savéant, K.B. Su, D.L. Wang, J. Am. Chem. Soc. 109 (1987) 6464;
(g) H. Jensen, K. Daasbjerg, J. Chem. Soc. Perkin Trans. 2 (2000) 1251;
(h) K. Daasbjerg, T.B. Christensen, Acta Chem. Scand. 49 (1995) 128;
(i) X.-M. Zhang, J. Chem. Soc. Perkin Trans. 2 (1993) 2275;
(j) C. Costentin, J.-M. Savéant, J. Am. Chem. Soc. 122 (2000) 2329;
(k) C. Costentin, P. Hapiot, M. Medebielle, J.-M. Savéant, J. Am. Chem. Soc. 122
(2000) 5623.
The present study on the electrochemical reduction of a series
of substituted 1,4-dichloroazoethanes, 1-X/Y, provides an unprece-
dented illustration of molecular systems containing two halides,
separated by three bonds, that react through a dissociative electron
transfer mechanism to yield azine products. The initial reduction
proceeds by DET to cleave a C–Cl bond. The low intrinsic barriers
suggest the reduction proceeds by a stepwise DET throughout the
series in contrast to benzyl halides and vicinal dihalides. We pos-
tulate the azo moiety facilitates the stepwise pathway by lowering
the energy of the LUMO over the bulk of the molecular architec-
ture, although a concerted DET may also be competitive in some
cases. The stable carbon radical formed after dissociation is then
reduced directly by the electrode as the reduction potentials of
the radicals are generally expected to be more positive than the
applied electrode potential [17g,27], resulting in an anionic inter-
mediate, and the expulsion of chloride to form 2-X/Y azines. Unlike
the photochemical reactions, no coupling products were observed
from carbon radical intermediates. Hence, we provide an efficient
route to obtaining pure azines in quantitative yield.
[17] (a) M.S. Workentin, F. Maran, D.D.M. Wayner, J. Am. Chem. Soc. 117 (1995)
2120;
(b) S. Antonello, F. Maran, J. Am. Chem. Soc. 119 (1997) 12595;
(c) S. Antonello, F. Maran, J. Am. Chem. Soc. 121 (1999) 9668;
(d) S. Antonello, M. Musumeci, D.D.M. Wayner, F. Maran, J. Am. Chem. Soc. 119
(1997) 9541;
(e) M.S. Workentin, R.L. Donkers, J. Am. Chem. Soc. 120 (1998) 2664;
(f) R.L. Donkers, F. Maran, D.D.M. Wayner, M.S. Workentin, J. Am. Chem. Soc.
121 (1999) 7239;
(g) R.L. Donkers, M.S. Workentin, Chem. Eur. J. 7 (2001) 4012;
(h) D.C. Magri, M.S. Workentin, Org. Biomol. Chem. 1 (2003) 3418;
(i) D.C. Magri, M.S. Workentin, Org. Biomol. Chem. 18 (2008) 3354;
(j) D.C. Magri, M.S. Workentin, Chem. Eur. J. 14 (2008) 1698;
(k) D.L.B. Stringle, D.C. Magri, M.S. Workentin, Chem. Eur. J. 16 (2010) 178.
[18] (a) T.B. Christensen, K. Daasbjerg, Acta Chem. Scand. 51 (1997) 307;
(b) K. Daasbjerg, H. Jensen, R. Benassi, F. Taddei, S. Antonello, A. Gennaro, F.
Maran, J. Am. Chem. Soc. 121 (1999) 1750;
Acknowledgements
(c) S. Antonello, R. Benassi, G. Gavioli, F. Taddei, F. Maran, J. Am. Chem. Soc. 124
(2002) 7529;
(d) S. Antonello, K. Daasbjerg, H. Jensen, F. Taddei, F. Maran, J. Am. Chem. Soc.
125 (2003) 14905.
The financial support of the Natural Science and Engineering
Research Council of Canada, Canadian Foundation for Innovation,

V.A. Sauro et al. / Electrochimica Acta 55 (2010) 5584–5591
5591
[19] D.S. Malament, J.M. McBride, J. Am. Chem. Soc. 92 (1970) 4586.
[20] A.F. Hegarty, J.A. Kearney, J. Org. Chem. 40 (1975) 3529.
[21] V.A. Sauro, M.S. Workentin, J. Org. Chem. 66 (2001) 831.
[22] 2,2,2-Trifluoroethanol did cause the two one-electron waves, B and C, to merge
into a single two-electron wave with increasing acid concentrations.
[23] (a) J.C. Imbeaux, J.-M. Savéant, J. Electroanal. Chem. 44 (1973) 169;
(b) J.-M. Savéant, D. Tessier, J. Electroanal. Chem. 61 (1975) 251.
[24] A.J. Bard, L.R. Faulkner, Electrochemical Methods. Fundamentals and Applica-
tions, John Wiley & Sons, New York, NY, 1980.
[26] D.D.M. Wayner, in Handbook of Organic Chemistry, Redox Properties, vol. 2, p.
363.
[27] (a) D.D.M. Wayner, D.J. McPhee, D. Griller, J. Am. Chem. Soc. 110 (1988) 132;
(b) B.A. Sim, P.H. Milne, D. Griller, D.D.M. Wayner, J. Am. Chem. Soc. 112 (1990)
6635;
(c) P. Hapiot, V.V. Konovalov, J.-M. Savéant, J. Am. Chem. Soc. 117 (1995) 1428.
[29] C. Hansch, A. Leo, R.W. Taft, Chem. Rev. 91 (1991) 165.
[30] (a) H. Lund, E. Hobolth, Acta Chem. Scand. B 30 (1976) 895;
(b) T. Lund, C. Bjorn, H.S. Hansen, A.K. Jensen, T.K. Thorsen, Acta Chem. Scand.
47 (1993) 877;
[25] J.C. Eklund, A.M. Bond, J.A. Alden, R.G. Compton, Perspectives in Modern
Voltammetry: Basic Concepts and Mechanistic Analysis, vol. 32, Academic
Press, 1999.
(c) H. Lund, H.L. Kristensen, Acta Chem. Scand. B 33 (1979) 495;
(d) H. Balslev, K. Daasbjerg, H. Lund, Acta Chem. Scand. 47 (1993) 1221.