J. Electroanal. Chem.

, 144 (1983) 59-67 59

Elsevier Sequoia S.A., Lausanne - Printed in The Netherlands

MECHANISM ANALYSIS OF ELECTROCHEMICAL REACTIONS

INVOLVING HOMOGENEOUS CHEMICAL STEPS

THE ELECTRODIMERIZATION OF 4-METHOXYBIPHENYL

C. AMATORE and J.M. SAV~kNT

Laboratoire d'Electrochimie de l'Universitb de Paris 7, 2, place Jussieu, 75251 Paris Cedex 05 (France)
(Received 3rd June 1982)

ABSTRACT

With a few exceptions, the use of electrochemical kinetic methods has long been confined to the
analysis of limiting mechanisms in which the kinetics of the overall reaction is controlled by a single
rate-determining step. More recently, however, rigorous methods have been developed for analyzing the
mechanism of reactions involving two or more competing rate-determining steps, leading to a single or to
several reaction products. They allow the investigation and prediction of the influence of operational
parameters, such as diffusion rate and concentrations of reactants on the transition between limiting
mechanisms. The alternative use of "reaction orders, without calculation" approach, which overlooks
these available methods, may lead to erroneous mechanistic conclusions. This is illustrated by analysis of
the reaction mechanism of the anodic electrodimerization of 4-methoxybiphenyl in acetonitrile, a typical
example of dimerization processes involving cation radicals. Raising the reactant concentration triggers
the passage from one limiting mechanism to another. Rigorous analysis of the kinetic data shows that the
recently proposed competition between a radical-radical coupling and a radical-substrate coupling
pathway is incompatible with the experimental data. The latter can be rationalized in terms of pure
radical-substrate coupling mechanism, with the coupling step being rate determining at high concentra-
tions and the further homogeneous electron-transfer step being rate determining at low concentrations.
Implications in the chemistry of electrodimerization are discussed.

INTRODUCTION

H o m o g e n e o u s c h e m i c a l steps c o u p l e d w i t h e l e c t r o d e e l e c t r o n t r a n s f e r s i n t e r f e r e in
e l e c t r o c h e m i c a l r e a c t i o n s in a n h e t e r o g e n e o u s m a n n e r in t h e sense t h a t t h e y a r e
c o u p l e d w i t h the d i f f u s i o n o f r e a c t a n t s , i n t e r m e d i a t e s a n d p r o d u c t s to a n d f r o m the
e l e c t r o d e surface. T h e k i n e t i c a n a l y s i s of s u c h r e a c t i o n s is t h u s s o m e w h a t m o r e
c o m p l e x t h a n for p u r e l y h o m o g e n e o u s p r o c e s s e s : the t i m e - d e p e n d e n t d i f f e r e n t i a l
e q u a t i o n s d e s c r i b i n g t h e h o m o g e n e o u s k i n e t i c s h a v e to b e r e p l a c e d b y t i m e - a n d
s p a c e - d e p e n d e n t p a r t i a l d e r i v a t i v e e q u a t i o n s ; a c c o r d i n g l y , b o u n d a r y c o n d i t i o n s at
t h e e l e c t r o d e s u r f a c e a n d in the b u l k o f the s o l u t i o n h a v e to b e t a k e n i n t o a c c o u n t .
T h i s e x p l a i n s why, in a first stage, analysis o f this t y p e o f e l e c t r o c h e m i c a l k i n e t i c s
a n d a p p l i c a t i o n to m e c h a n i s m d i a g n o s i s a n d r a t e c o n s t a n t d e t e r m i n a t i o n h a v e b e e n

0022-0728/83/0000-0000/$03.00 © 1983 Elsevier Sequoia S.A.

60

restricted to simple kinetic behavior, i.e. those in which the overall kinetics is
controlled by a single rate-determining step. The latter does not necessarily im-
mediately follow or precede the electrode electron transfer, but may be separated
from it by other chemical steps remaining at equilibrium. The corresponding
homogeneous rate law is introduced into the diffusion partial derivative equations,
the resolution of which provides the description of the kinetics of the overall
electrochemical process.
Considering, for example, reaction schemes in which the initial electron-transfer
step A + e ~ B is followed by homogeneous chemical steps involving B, situations of
particular interest are met when the lifetime of B is so short, with respect to
diffusion, that "pure kinetic" conditions are achieved. These imply that a stationary
state exists resulting from mutual compensation of the diffusion and chemical
processes. The variations of the B concentration with time can then be neglected and
the concentration profile of B is confined within a "reaction layer" vanishingly thin
compared to the diffusion layer. Under such conditions, the characteristics of the
polarization curves are related in a quite simple manner to those of the homogeneous
kinetics, namely reaction orders and overall rate constant, provided the initial
electron transfer is sufficiently fast so as not to interfere kinetically. In cyclic
voltammetry, these "pure kinetic" conditions give rise to totally irreversible waves.
The peak potential, Ep, is then a linear function of the A / B standard potential and
of the logarithms of the overall rate constant (k), sweep rate (v), initial substrate
concentration (c °) and, accordingly, of the concentration of electroinactive reactants
(Z°). The OEp/Olog v °, ~Ep/~ log c °, ~Ep//Olog z ° are then simply related to the
reaction orders of the homogeneous kinetic law. These expressions are obtained by
integration of the space-dependent, time-independent, differential equation relative
to B which does not require any numerical calculation. Mechanism analysis is then
based upon the comparison between the experimental slopes and the values predic-
ted for the limiting, single-rate-determining step, kinetic behavior. An early example
of this type of mechanism analysis concerned radical-radical dimerization as
illustrated experimentally by the reduction of benzaldehyde in a protic medium [1].
Since then, the method has been extended to a number of other reaction schemes [2]
and applied to various electrochemical reactions such as electrodimerization of
tropylium [3a] and immonium cations [3b], carbonyl compounds [2f], imines [3c],
activated olefins [3d,3e], polyenic compounds [3f,3g], electrocyclizations [4], electro-
hydrogenations [2f,3c,5] etc. Systematic mechanism analysis thus involves the com-
parison of the measured slopes with those predicted for all the limiting kinetic
behaviour of all the possible mechanisms (see e.g. refs. 3d, 3e). Numerical calcula-
tions are necessary for deriving the overall rate constant from the peak potential,
knowing the A - B standard potential independently. They are also required if one
wants to use the information contained in the whole polarization curve rather than
in the peak potential only. This allows more efficient discrimination between the
various types of limiting kinetic behavior. Measurement of the peak width or fitting
of the whole polarization curve may be used in this connection, as well as other
procedures such as convolution of the current with the linear diffusion characteristic
 61

(~rt)-1/2 function [6]. Numerical calculation then consists in the computation of the
convolution integral from the experimentally determined current [6].
The relationship between the homogeneous kinetics and the electrochemical
response is somewhat more complex when, still dealing with a single rate-determin-
ing step, the chemical process is not sufficiently fast as compared to diffusion for the
" p u r e kinetic" conditions to be achieved. The partial derivative equations can no
longer be simplified into space-dependent, time-independent, derivative equations.
The dependence of the electrochemical response upon substrate concentration,
electroinactive reactant concentration, time-scale (i.e. sweep rate in cyclic voltamme-
try) can, however, be predicted for any given limiting behavior by dimension
analysis of the system [2a 2c,2f-2h]. It thus appears that the interplay between
diffusion and chemical processes is dependent upon a single dimensionless parame-
ter )k. If, for example, the reaction orders of the homogeneous process are a for A, ,8
for B and ~" for an electroinactive reactant, Z, )t = (RT/F)k(c°) I'~+~ ~c~/v in
cyclic voltammetry [2a-2c,2f 2h] [or, equivalently, X = k(c°) (~+#- ll(c°);t in poten-
tial step chronoamperometry] where k is the overall rate constant, v the sweep rate
and t the time. When looking for the reaction mechanism, a first test is thus to check
that the electrochemical response does follow the concentration and sweep rate (or
time) dependence predicted, through X, for the hypothesized limiting behavior: since
the electrochemical response depends only upon ~., it should remain the same when
varying c °, c z0 and v so that (c°) (~+# I)(c°)~/v remains constant. This is, however,
not sufficient in most cases, since different types of limiting kinetic behavior may
well give rise to the same concentration and sweep rate (or time) dependence. A
further step will be then to compare the experimental and predicted shapes of the
electrochemical r e s p o n s e = f ()t) relationship. This requires the calculation of the
corresponding working curve by algebraic or numerical integration of the partial
derivative equation system. For one-electron processes, such as electrodimerizations,
adequate electrochemical responses are obtained from current ratios in cyclic
techniques such as cyclic voltammetry [7a], derivative cyclic voltammetry [8] or
double-potential-step chronoamperometry [7b,7c].
An early extension of the capabilities of mechanism analysis beyond the cases
described above, was provided by the treatments in which the reversibility of the
coupled chemical reaction was taken into account, involving the participation of
both the forward and the backward reactions to kinetic control [2a,2b,9]. More
recently, a number of more complex reaction schemes have been analyzed involving
the competition between two or even three rate-limiting steps. Methods were thus
described which allow one to predict the influence of the operational parameters,
sweep rate (or-time, or diffusion-layer thickness), concentrations, on the overall
kinetics, particularly how their variations shift the system toward limiting behaviors.
This has been applied to a variety of electrochemical processes, under various micro-
and macroelectrolysis regimes: two-electron processes involving competing heteroge-
neous and homogeneous electron transfers [10], electrocatalysis of chemical reactions
[11], homogeneous catalysis of electrochemical reactions [12], competition between
H-atom and electron transfers [13], between radical-radical and radical-substrate
62

dimerization and, between dimerization and further electron transfer [14]. These
treatments can be extended to other reaction schemes a n d / o r other electrochemical
techniques. Ample background is thus available for analyzing quite complex mecha-
nisms.
This reminder of the previously acquired knowledge in the field of electrochemi-
cal mechanism analysis seemed necessary in view of the recently proposed "reaction
order" approach to the problem [8,15]. As far as limiting kinetic behavior is
concerned, this approach is essentially the same as the previously described methods
referred to above. Being actually based on the rigorous relationships established in
previous work, its application to practical situations is anticipated to lead to correct
conclusions as far as their range of applicability is respected, i.e. as far as single
rate-determining step kinetic behavior is dealt with. However, the use of this
approach has also been repeatedly advocated for more complex reaction schemes
and applied to such experimental situations [8,15c-15j]. Overlooking the abundant
previous work relative to this type of situations [10-14], the kinetic interplay
between diffusion and homogeneous chemical reactions was then either ignored, as if
the homogeneous process was functioning by itself [15f-15h], or taken into account
on questionable intuitive grounds [8]. It is anticipated that this may lead to
erroneous mechanism diagnostics. This has already been shown to be the case [16]
for the mechanism analysis of the electrodimerization of anthracenes substituted
with electron-withdrawing substituents [15i]. We give below a second example
concerning the anodic electrodimerization of 4-methoxybiphenyl in acetonitrile [8],
which is an illustrative example of the general problem of the competition between
radical-radical and radical-substrate coupling in electrodimerization reactions (see
ref. 16 and refs. cited therein).
RESULTS

The kinetic data obtained using derivative cyclic voltammetry (DCV) [8] are
summarized in Fig. 1 under the form of a log-log plot of the variations of vl/2 with
the substrate concentration c °, vl/2 being the value of the sweep rate necessary for
the ratio of the derivative peaks on the backward and forward scans of the cyclic
voltammogram to equal 0.5. It is seen that vl/2 is proportional to c o at high values of
c °, and to (c°) 2 at low c o values. It was concluded [8] from these observations that
the dimerization mechanism involves the competition between a radical-radical
coupling (rrc) and a radical-substrate coupling (rsc) reaction:
R - e ~ R "+ (1)
kl
(rrc) 2 R -+ ~ + R - R + (2)
k2
rsc) R + R .+ ~ ' R - R + (K 2=k2/k 2) (3)
k-2

" R - R + + R -+ k~ + R - R + + R (4)

+R-R+~ 2 H + + final dimer

 63

log(vii2/V.S -1 )
'3

/I
/z
///

¢ rsc2
Iog(c°/M )
-4 -3 -2

Fig. 1. Anodic electrodimerization of 4-methoxybiphenyl. Comparison between experimental kinetic data

(e) (from Table 1 in ref. 8) and predicted kinetics for a rrc-rsc2 competitive mechanism ( - - - - - - ) and a
rsc mechanism ( ); vl/2: sweep rate value necessary for the ratio of the derivative peaks on the
backward and forward scans to equal 0.5. cO: 4-methoxybiphenyl bulk concentration.

(R = 4-methoxybiphenyl). More precisely, the forward reaction (1) was regarded as

the rds in the rrc reaction which fits the c o proportionality found at high c o 's, while
the forward reaction (3) was regarded as the rds of the rsc reaction with reaction (2)
remaining at equilibrium, which fits the (c°) 2 proportionality observed at low c°'s.
Intuitively, it is predicted that the corresponding limiting rrc behavior will pre-
dominate at low c o's and the rsc behavior at high c o 's, just the opposite to what is
found experimentally concerning the variations of v 1/2 with c °. This discrepancy was
explained by difficult-to-verify qualitative reasonings on the differences existing
between the purely homogeneous process and the electrochemical process where the
homogeneous reactions are coupled with diffusion. What is the actual answer to the
question resulting from a rigorous treatment of the hypothesized competition
between the two mechanisms in the context of diffusion to and from the electrode?
The problem can be formulated in dimensionless terms as follows:

Oa ()2a Ob 32b 2RT kLC°b2[l + K2kacoa I

()'r ~)y2 ' O"r ~)y2 F v ~ k1 ]
~-=O,y>~O and y~oo,~->O:a=l, b=O
64

y = O , ~-> O" ( ~ a / ~ y ) + ( ~ b / ~ y ) = O , a=bexp[-(F/RT)(E-E°)]

E=Ei+vt for O<~t<~O, E=Ef-v(t-O) for t>O
the current-time function being obtained as
i = - F S D 1/2c° ( F v / R r ),/2 (a a / O y )0
with

r = F v t / R T , y = x ( F v / R T D ) t/2, a = [ R ] / c °, b = [R .+] / c o

t is the time, x the distance from the electrode, c o the bulk conc. of R, E the
electrode potential, E ° the R / R + standard potential, E i the initial potential, E r the
scan inversion potential, 0 the scan inversion time, S the electrode surface area and
D the diffusion coefficient of R and R .+. The calculations were carried out for
( F / R T ) ( E ° - Ei) = 10, E i then being sufficiently negative to E ° to have practically
no influence on the cyclic voltammogram. Here, Ef was taken 300 mV positive to E °,
at 22°C.
Since it is assumed that the rrc mechanism predominates at high c o 's, k I is found
equal to 2 . 7 6 × l 0 4 M - 1 s - 1 (right-hand side asymptote in Fig. 1). Similarly,
K z k 3 = 1.28 × l0 s M -2 s -1 as derived from the rsc behavior assumed to pre-
dominate at low c ° ' s (left-hand-side asymtote in Fig. 1). The above system and thus
the ratio of the derivative peaks are thus functions of only two parameters, v and c0.
For each value of c °, the value of v, Vl/2, for which this ratio equals 0.5 was
numerically obtained from an explicit finite difference resolution of the above
system [17]. The resulting v t / 2 - c ° working curve is shown in Fig. 1 (dashed line) in
comparison with the experimental data.
It is clearly seen that the predicted kinetics is in complete disagreement with the
experimental data, showing that the proposed rrc-rsc competitive mechanism is
certainly wrong. This casts grave doubts on the contention that the process taking
place at high concentrations involves the radical-radical coupling of two cation
radicals [8]. Important consequences concerning the chemistry of electrodimerization
in general ensue. The dimerization reaction occurring at high concentration in the
oxidation of 4-methoxybiphenyl was indeed taken as the prototype of radical-radi-
cal coupling in the ion-radical series, a reaction claimed to occur very seldom [8,15i]
in contrast with the conclusions of past investigations. In this connection, the
activation energy of the reaction 44.4 kJ mol-~ (10.6 kcal mol-1), was regarded as a
minimum value for the rrc of all ion radicals, serving as evidence against the reality
of previously postulated rrc processes for which the activation energies are smaller [8
1- All these reasonings now become very doubtful since the very existence of a rrc
process in the oxidation of 4-methoxybiphenyl in the high concentration range itself
appears doubtful. On the other hand, rrc of ion radicals may well involve activation
energies significantly smaller than 63 kJ mol-1 (15 kcal m o l - t ) as discussed recently
[161.
Since the rrc-rsc competitive mechanism is clearly ruled out by proper analysis of
the experimental data, the question arises of the actual nature of the mechanism of
 65

the oxidative electrodimerization of 4-methoxybiphenyl. Let us consider that the rrc

pathway is negligible, i.e. that dimerization totally proceeds along the rsc mecha-
nism. In this context, we remove the restriction that reaction (3) is the rds, with
reaction (2) remaining at equilibrium. In other words, it is considered that both steps
may participate in the kinetic control of the overall rsc process. There are then two
types of limiting kinetic behavior, one in which reaction (2) is the rds and another in
which reaction (3) is the rds with reaction (2) acting as a pre-equilibrium. These
types of limiting behavior will be named rscl and rsc2 respectively. For rscl, v~/2 is
proportional to c °, while for rsc2 it is proportional to (c°) 2. Passing from rsc2 to
rscl implies that reaction (3) becomes more and more efficient as opposed to
backward reaction (2). An increase of c o is anticipated to produce this effect since
reaction (3) is second order and backward reaction (2) is first order. This is similar to
that which occurs in disproportionation processes where the chemical step inter-
posed between the electrode electron transfer and the solution electron transfer is a
first-order reaction in both directions [10a]. In more quantitative terms, the overall
process is described by the following system which involves the stationary-state
assumption concerning "R-R+:
Oa ~2a Ob O2b 2 R T k2c ° ab 2
~'r ~y2 ' ~'r ~y2 F v b + k 2/k3 c°

with the same initial and boundary conditions as for the rrc-rsc2 mechanism and
the same symbolism. The rscl behavior is obtained for k3 c° >> k z, i.e. at the high
concentration limit. It ensues that k 2 can be estimated as k 2 = 3.55 x 104 M-1 s-1
The rsc2 behavior is reached when k3 c° << k 2, i.e. at the low concentration limit.
We then obtain an estimation of k 2 k 3 / k 2, k 2 k 3 / k _ 2 = 1.28 × 108 M 2 s - ~. The
system and thus the ratio of the derivative peaks are thus a function of the two
parameters v and c °. For each value of c °, the value of v, vt/2, for which this ratio
equals 0.5, was then numerically obtained according to the same computation
procedure as for the rrc-rsc2 case. The resulting v ~/2-c ° working curve is shown in
Fig. 1 (solid line) in comparison with the experimental data; the best fitting with the
experimental data leads to the following values: k 2 = 4.07 × 104 M - l s -1 and
k3kz//k_2 = 1.65 × 108 M -2 s - i .
It is seen that the agreement between the predicted and the experimental kinetic
behavior is now excellent. The rsc mechanism is thus compatible with the DCV data.
For the following reasons, some caution should, however, be exerted as to the
conclusion that this is the actual mechanism of the reaction. We attempted to check
that the linear sweep voltammetric peak variations with sweep rate and concentra-
tion predicted for such a mechanism are indeed followed. In the low range of
concentrations (c °--0.1 m M ) this implied the use of rather low sweep rates
(v < 0.01 V s - i ) in order that the wave becomes irreversible, i.e. that "pure kinetic"
conditions be achieved. It was also of interest to raise the concentration so as to
attempt to reach the region where the rscl behavior predominates, which corre-
sponds to ~Ep//~ log v = - 30 mV and ~ E p / 3 tog c o = 30 mV, as opposed to - 20
66

and 20 mV respectively for the rrc mechanism. When we used these experimental
conditions they led to poor accuracy and reproducibility of peak potential measure-
ments, resulting most probably from product adsorption. It was suspected that the
deprotonated dimer could itself be oxidized in the same potential region as the
starting compound, possibly leading to polycondensated products..Whether or not
these reactions are a part of the process taking place under DVC conditions is not
clear at present. More generally, the exact nature of the reaction products formed
under these conditions is not ascertained, compared to what is obtained under
preparative scale conditions [18].

CONCLUSIONS

In conclusion, we would again like to emphasize the dangers of using "reaction

order, without calculations" approaches to mechanism analysis outside their actual
range of validity, i.e., in the case of reactions that are kinetically controlled by more
than one step. On the plea of avoiding mathematical complexity, these approaches
thus ignore the availability of a sound theoretical background which would allow the
correct mechanistic conclusions to be drawn. In this sense, they amount to a
regression with regard to the present capabilities of mechanism analysis of electro-
chemical processes.

REFERENCES

1 J.M. Sav6ant and E. Vianello, C.R. Acad. Sci. (Paris), 256 (1963) 2597.
2 (a) J.M. Sav6ant and E. Vianello, Electrochim. Acta, 12 (1967) 629; (b) ibid., 1545; (c) M.
Mastragostino, L. Nadjo and J.M. Sav6ant, ibid., 13 (1968) 721; (d) J.M. Sav6ant, Rev. Claim. Min., 5
(1968) 477; (e) C.P. Andrieux, L. Nadjo and J.M. Sav~ant, J. Electroanal. Chem., 26 (1970) 147; (f) L.
Nadjo and J.M. Sav~ant, ibid., 33 (1971) 419; (g) C.P. Andrieux, L. Nadjo and J.M. Sav~ant, ibid., 41
(1973) 137; (h) ibid., 42 (1973) 223; (i) L. Nadjo and J.M. Sav~ant, ibid., 44 (1973) 327; (j) C.P.
Andrieux and J.M. Sav~ant, Ibid., 53 (1974) 165.
3 (a) A.M. Khopin and S.I. Zhdanov, Elektrokhimiya, 4 (1968) 228; (b) C.P. Andrieux and J.M.
Sav6ant, J. Electroanal. Chem., 26 (1970) 223; (c) ibid., 33 (1971) 453; (d) E. Lamy, L. Nadjo and J.M.
Sav6ant, ibid., 42 (1973) 189; (e) ibid., 50 (1974) 141; (f) L.A. Powell and R.M. Wightman, ibid., 106
(1980) 377; (g) ibid., 117 (1981) 321.
4 (a) F. Ammar, C.P. Andrieux and J.M. Say,ant, J. Electroanal. Chem., 53 (1974) 407; (b) I.
Tabakovi~, M. Lacan and S. Damoni, Electrochim. Acta, 21 (1976) 621.
5 C. Amatore and J.M. Sav6ant, J. Electroanal. Chem., 107 (1980) 353.
6 (a) J.C. Imbeaux and J.M. Sav6ant, J. Electroanal. Chem., 44 (1973) 169; (b) L. Nadjo, J.M. Sav6ant
and D. Tessier, ibid., 64 (1975) 143; (c) C.P. Andrieux, J.M. Sav6ant and D. Tessier, ibid., 63 (1975)
429.
7 (a) M.L. Olmstead, R.G. Hamilton and R.S. Nicholson, Anal. Chem., 41 (1969) 260; (b) M.L.
Olmstead and R.S. Nicholson, ibid., 41 (1969) 851; (c) V.W. Childs, J.T. Maloy, C.P. Keszthelyi and
A.J. Bard, J. Electrochem. Soc., 118 (1971) 874.
8 B. Aalstad, A. Ronlan and V.D. Parker, Acta Chem. Scand., B35 (1981) 649.
9 (a) J.M. Sav6ant and E. Vianello, Electrochim. Acta, 8 (1963) 905; (b) R.S. Nicholson and I. Shain,
Anal. Chem., 36 (1964) 706.
10 (a) C. Amatore and J.M. Savb,ant, J. Electroanal. Chem., 85 (1977) 27; (b) ibid., 86 (1978) 227; (c)
ibid., 102 (1979) 21; (d) ibid., 107 (1980) 353; (e) C. Amatore, D. Lexa and J.M. Savbant, ibid., 111
(1980) 81; (f) ibid., 123 (1981) 189; (g) ibid., 203.
 67

11 (a) C. Amatore, J. Chaussard, J. Pinson, J.M. Sav~ant and A. Thibbault, J. Am. Chem. Soc., 101 (1979)
6012; (b) C. Amatore, J.M. Sav6ant and A.Thi~bault, J. Electroanal. Chem., 103 (1979) 303; (c) C.
Amatore, J. Pinson, J.M. Sav6ant and A. Thi6bault, ibid., 107 (1980) 59; (d) ibid., 75; (e) ibid., 123
(1981) 231; (f) J.M. Sav6ant, Acc. Chem. Res., 13 (1980) 323.
12 (a) C.P. Andrieux, J.M. Dumas-Bouchiat and J.M. Sav6ant, J. Electroanal. Chem., 87 (1978) 39; (b)
ibid., 55; (c) ibid., 88 (1978) 43; (d) ibid., 113 (1980) 1; (e) C.P. Andrieux, C. Blocman, J.M.
Dumas-Bouchiat, F. M'Halla and J.M. Sav6ant, ibid., 113 (1980) 19; (f) C.P. Andrieux, C. Blocman,
J.M. Dumas-Bouchiat, F. M'Halla and J.M. Sav6ant, J. Am. Chem. Soc., 102 (1980) 3806.
13 (a) F. M'Halla, J. Pinson and J.M. Sav6ant, J. Am. Chem. Soc., 102 (1980) 4120; (b) C. Amatore, F.
M'Halla and J.M. Sav6ant, J. Electroanal. Chem., 123 (1981) 219.
14 (a) C. Amatore and J.M. Sav6ant, J. Electroanal. Chem., 125 (1981) 1; (b) ibid., 23; (c) ibid., 126
(198i) 1; (d) C. Amatore and J.M. Sav~ant, J. Am. Chem. Soc., 103 (1981) 5021; (e) M. Falsig, H.
Lund, L. Nadjo and J.M. Sav6ant, Acta Chem. Scand., B34 (1980) 685.
15 (a) E. Ahlberg and V.D. Parker, Acta Chem. Scand., B35 (1981) 117; (b) V.D. Parker, ibid., 259; (c)
ibid., 147; (d) ibid., 149; (e) ibid., 279; (f) ibid., 295; (g) ibid., 349; (h) ibid., 583; (i) O. Hammerich
and V.D. Parker, ibid., 341; (j) B. Aalstad, A. Ronlan and V.D. Parker, ibid., 247.
16 C. Amatore, J. Pinson and J.M. Sav6ant, J. Electroanal. Chem., 137 (1982) 143.
17 J. Crank, Mathematics of Diffusion, Oxford at the Clarendon Press, London, 1964.
18 A. Ronlan, K. Bechgaard and V.D. Parker, Acta Chem. Scand., 27 (1973) 2375.