Electrochemistry of platinum phthalocyanine microcrystals: II. A microelectrode observation of nucleation-growth controlled solid-solid phase transformations in non-aqueous solvent

Article abstract of DOI:10.1016/S0013-4686(00)00700-3

The solid-solid phase transformations and switching reactions occurring in platinum phthalocyanine (PtPc) microcrystals were investigated by chronoamperometry on a microelectrode in acetonitrile containing 0.1 mol dm^-3 of the tetrabutylammonium salt of either BF₄^-, ClO₄^- or PF₆^-. Three different states of the PtPc film (reduced, conductive and over-doped) can be demarcated, depending on its degree of oxidation. The transient response seen is dependent upon the initial and final state of the film. For the reduced film, a nucleation-growth process occurs in the film upon the application of an oxidative potential step to above 0.50 V. If the end-point of the potential step is increased past 0.75 V, a second nucleation-growth process occurs leading to the conductive film. Both of these processes are controlled by a solid-solid phase transformation. At potentials above approximately 1.1 V there is evidence of a further diffusion-controlled reaction, leading to the production of the over-doped film. Oxidation or reduction of the conductive film occurs quickly and appears to be diffusion controlled with no indications of a peak or shoulder in the current-time transients. Reduction of the over-doped film, appears to be controlled by one and possibly two nucleation-growth processes as evidenced by peaks in the chronoamperometric transient. The kinetics of solid-solid phase transformation and the switching reaction is only affected slightly by the nature of the anions present.

Full text of DOI:10.1016/S0013-4686(00)00700-3

Electrochimica Acta 46 (2001) 1223–1231
www.elsevier.nl/locate/electacta
The electrochemistry of platinum phthalocyanine
microcrystals II
A microelectrode observation of nucleation-growth
controlled solid–solid phase transformations in non-aqueous
solvent
Junhua Jiang, Anthony Kucernak *
Department of Chemistry, Imperial College, London SW7 2AZ, UK
Received 1 September 2000; received in revised form 11 October 2000
Abstract
The solid–solid phase transformations and switching reactions occurring in platinum phthalocyanine (PtPc)
−
3
microcrystals were investigated by chronoamperometry on a microelectrode in acetonitrile containing 0.1 mol dm
−
−
−
of the tetrabutylammonium salt of either BF , ClO or PF . Three different states of the PtPc film (reduced,
4
4
6
conducti6e and o6er-doped) can be demarcated, depending on its degree of oxidation. The transient response seen is
dependent upon the initial and final state of the film. For the reduced film, a nucleation-growth process occurs in the
film upon the application of an oxidative potential step to above 0.50 V. If the end-point of the potential step is
increased past 0.75 V, a second nucleation-growth process occurs leading to the conducti6e film. Both of these
processes are controlled by a solid–solid phase transformation. At potentials above ꢀ1.1 V there is evidence of a
further diffusion-controlled reaction, leading to the production of the o6er-doped film. Oxidation or reduction of the
conductive film occurs quickly and appears to be diffusion controlled with no indications of a peak or shoulder in the
current–time transients. Reduction of the o6er-doped film, appears to be controlled by one and possibly two
nucleation-growth processes as evidenced by peaks in the chronoamperometric transient. The kinetics of solid–solid
phase transformation and the switching reaction is only affected slightly by the nature of the anions present. © 2001
Elsevier Science Ltd. All rights reserved.
Keywords: Platinum phthalocyanine; Nucleation-growth; Phase transformation; Switching rate; Microelectrode
switching devices [4] and power sources [5,6]. They can
1
. Introduction
be altered significantly via incorporation of different
metal atoms at the centre of the phthalocyanine ring or
by changing the nature of the substituents on the
periphery of the macrocyclic ring. A phthalocyanine
macrocycle has the tendency to organise itself and form
stacks in which there is interaction between the large
p-system of the adjacent rings in such a way as to give
low-dimensional crystals. These crystals exhibit high
conductivity when excess electrons or holes are intro-
The physical and chemical properties of metalloph-
thalocyanines (MPcs) have led to their applications to
electrochromic devices [1,2], organic solar cells [3],
*
Corresponding author. Tel.: +44-020-5945831; fax: +44-
0
0
20-5945804.
E-mail address: a.kucernak@ic.ac.uk (A. Kucernak).
013-4686/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0013-4686(00)00700-3

1224
J. Jiang, A. Kucernak / Electrochimica Acta 46 (2001) 1223–1231
duced into the conduction or valence band, a process
that is normally carried out by chemical or electro-
chemical doping. Partially oxidised MPc’s are members
of a novel class of charge transfer (CT) salts [7,8].
Platinum phthalocyanine (PtPc) with higher thermal
and electrochemical stability than first-row MPcs,
shows metal-like conductivity in the partially oxidised
state [9–11]. Anion insertion and removal for charge
compensation during the doping/undoping process
causes solid–solid phase transformations in the MPc’s
microcrystals [12]. But the kinetics of the solid–solid
phase transformations has received little attention.
MPcs have extensively conjugated aromatic chro-
mophores. The switching of their redox states generates
new or different visible region bands, and they may be
used as electrochromic materials. The mechanics of the
reversible and irreversible electrochromism of first-row
MPc’s has been discussed by Toshima et al. [13], yet
few attempts have been made to examine the elec-
trochromic effect in precious MPc’s film.
have simultaneously investigated the kinetics of solid–
solid phase transformations and the switching reaction
of PtPc microcrystals using a mechanically rubbed mi-
croelectrode. The unique advantage of the microelec-
trode is that current–time transients free of iR errors
can be recorded. Our new results show that the kinetics
of the solid–solid phase transformations and the
switching rate strongly depend on the degree of doping
of the microcrystals.
2. Experimental
2.1. Chemicals and instrumental
Tetrabutylammonium
perchlorate
(TBAClO4,
Fluka), Tetrabutylammonium
tetrafluoroborate
(TBABF , Aldrich) and tetrabutylammonium hexa-
4
fluorophosphate (TBAPF , Fluka) were of electrochem-
6
ical grade and used as received. Acetonitrile (AN,
Analar) was distilled under vacuum into a flask filled
Because of the low solubility of MPcs in organic
solvent even at high temperature, the investigation of
their redox properties has been a challenging field of
research. A recent experimental method of attaching
microcrystals to electrode surface, powder abrasion, has
successively been adapted to study such insoluble solids
with 4 A molecular sieve. PtPc was synthesized by
,
heating finely crushed phthalonitrile and PtCl at 200°C
2
in the absence of solvent for one hour [27]. Two
separate sublimation steps purified the product, each
step repeated three succesive times on the product of
the previous step. In the first, unreacted phthalonitrile
was sublimed from the crude product at 140°C and 0.1
mbar pressure, and in the latter, PtPc was sublimed
from the crude product at 550°C and 0.1 mbar
pressure.
[14–18]. We have studied the electrochemical behaviour
of PtPc microcrystals in acetonitrile electrolyte using
this method [12]. Similar to conductive polymers, an
obvious first-scan discrepancy and large capacitance are
observed during electrochemical oxidation of such sys-
tems. Reversible phase transformations are coupled to
the redox process of PtPc microcrystals, evidenced by
sharp peaks and unusually large peak potential differ-
ence. These phenomena were also observed on polymer
bound PtPc microcrystal electrode [9,11]. The unusual
peak potential difference has recently been considered
as one criterion of solid–solid phase transformation
under nucleation-growth control by Bond et al. [14,15].
But obtaining direct chronoamperometric evidence of
nucleation–growth kinetics in systems showing unusu-
ally large peak potential difference has been challeng-
ing. PtPc produces more than two redox states and in
acetonitrile electrolyte shows an invariant voltammetric
response after a number of conditioning scans. Our
in-situ UV-visible investigation of the PtPc thin film,
reported in a following paper, indicates that it may be
a prospective polyelectrochromic material.
Voltammetric and chronoamperometric measure-
ments were performed at room temperature (ꢀ20°C)
using an Autolab General Purpose Electrochemical Sys-
tems (GPES, Ecochemie, Netherlands). Electrochemical
measurements were performed in an oxygen-free three-
compartment electrochemical cell. The reference half-
−
3
−3
−3
cell was Ag/10
mol dm
AgNO +0.1 mol dm
3
TBAClO +AN, and the counter electrode was a plat-
4
inum plate. Solutions were deoxygenated with high
purity argon.
2.2. The preparation of PtPc microcrystal rubbed
microelectrode
The carbon fibre microelectrode (IJ Cambria Scien-
tific, UK) contained an 11 mm diameter carbon fibre
sealed into a soda glass tube. It was polished and
cleaned before use. Microcrystals of PtPc were attached
to the electrode surfaces as follows. First, a small
amount of microcrystalline powder was placed on a
smooth glass plate. The electrode was then rubbed on
the powder-covered glass until the electrode surface was
coated by a thin-layer of PtPc. Such a microelectrode
gave a very poor CV response. To improve its perfor-
mance the microelectrode was lightly rubbed on a clean
Chronoamperometry is a powerful technique for
studying nucleation-growth controlled phase transfor-
mation [19–22]. It has provided definitive evidence for
the solid–solid phase transformations controlled by
nucleation and growth in TCNQ microcrystals [14–16].
Peter et al. [23,24] and Lacroix et al. [25,26] have
investigated the fast switching of the electrochromism
of polyaniline using this technique. In this paper, we

J. Jiang, A. Kucernak / Electrochimica Acta 46 (2001) 1223–1231
1225
abrasive strip covered with 11 mm aluminium oxide
Agar Scientific Ltd, UK) until no purple PtPc was
3. Results
(
found on the glass surrounding the electrode surface.
This had the effect of limiting the region over which the
PtPc was deposited to only that of the microdisk. The
PtPc rubbed microelectrode was characterised by mea-
3.1. Cyclic 6oltammetry
Fig. 1 shows the fifth cyclic voltammometric scan for
PtPc rubbed microelectrodes in acetonitrile medium
−
3
−3
suring the steady-state current in 0.02 mol dm
containing 0.1 mol dm
tetrabutylammonium with
−
3
−
4 4
−
−
K Fe(CN) +1 mol dm
KCl solution. The rubbed
BF , ClO or PF6 as anions. The response is stable
3
6
microelectrode produced a well-defined current plateau
and shows no change upon successive scans. Similar to
our previous report [12], the following common features
are observed. The ratio of anodic to cathodic charge in
these plots is close to 1, indicating total charge re-
versibility. PtPc microcrystals produces four redox
states. Peak 1 is reduced at peak 1%, and is accompanied
by a sharp shoulder at peak 2 (reduced at 2%). Peak 3 is
responsible for peak 3%. The first redox process (1,1%) is
electrochemically reversible. It shows a peak potential
difference of ca. 30–60 mV. As the applied potential
exceeds 0.65 V, a sharp peak 2 is seen. The correspond-
ing reduction (2%) is broad, and partially buried under
peak 1%. The unusually large peak–peak potential dif-
ference reaches 0.62 V. Between this peak and the next
oxidation process, a large double layer capacitance is
evident in each scan (0.8–1.0 V). The properties of the
third redox process are strongly dependent upon the
−
1
at a scan rate of 20 mV s
and the average diameter
of the electrode disk was calculated to be 11 mm,
assuming n=1 and a diffusion coefficient of 7.63×
−
6
2
−1
1
0
cm
s
[28]. The amount of PtPc film on the
electrode surface was measured by coulometry. Pro-
vided that all PtPc microcrystals only covered the car-
bon-fibre disk, the thickness of PtPc film was almost
the same as the diameter of the microdisk on the basis
−
3
of a PtPc density of 1.96 g cm
[29,30]. After each
series of experiments, the electrode surface was renewed
by polishing and cleaning with acetone and Millipore
water. The cleaned electrodes were dried in an oven at
8
0°C.
This preparation describes a simple method to attach
microcrystals to the microelectrode surface. A similar
method of attaching TCNQ microcrystals to RAM™
(
carbon–fibre based random assemblies of microelec-
nature of the anion present in solution. Every anion
−
trodes) electrodes has been reported by Bond et al. [14].
RAM™ electrodes provide an advantage in that they
provide a much larger signal but unfortunately may fail
to work in aggressive organic solvents and at high
temperatures because of chemical attack to the epoxy
resin used to seal the carbon fibres.
produces a sharp reduction peak 3%. For PF , peak 3%
6
evolves into three sharp peaks. There is also a large
peak–peak potential difference of about 0.35 V. The
unusual peak-peak potential difference and sharp peaks
are indications of a nucleation-growth controlled solid–
solid phase transformation in the film.
We have examined the properties of PtPc microcrys-
tals with different oxidation states using electrochemical
impedance spectroscopy and compared their electro-
chemical properties as a function of potential and an-
ions [31]. In the following sections the PtPc films
pretreated at 0, 0.8 and 1.35 V are called the reduced,
the conductive and the over-doped film, respectively.
The direct evidence of nucleation-growth controlled
phase transformation has been studied on the different
films by chronoamperometry.
3.2. Chronoamperometry
The PtPc film was changed into different states by
pretreating it for 20 s at 0, 0.8 or 1.35 V before a
potential step was applied to the electrode. Unlike
polyaniline films that requires a long time to reach an
equilibrium state [32,33], the PtPc film can be fully
reduced or oxidised in a short time. Repeatable
chronoamperograms can be achieved on each pre-
treated film.
Fig. 1. Cyclic voltammograms of PtPc microcrystal rubbed 11
Fig. 2 shows current–time transients on reduced PtPc
film upon the step from 0 V (i.e the reduced film) to
different potential values. The transient at 0.5 V (Fig.
mm diameter carbon microelectrodes in acetonitrile medium
−
3
containing 0.1 mol dm
of the tetrabutylammonium salt of
−
−
−
−1
either –, BF4 ; ---, ClO4 ; ···, PF6 w=50 mV s
.

1226
J. Jiang, A. Kucernak / Electrochimica Acta 46 (2001) 1223–1231
peak shift to shorter time. This peak suggests that
another nucleation process is triggered in the film when
it is further oxidised. The currents at long time intervals
−
1/2
have a linear relation with t
. This indicates that the
electrochemical process is completed under diffusion
control. Two separate nucleation processes rather than
successive monolayer formation cause the evolution of
the two peaks. The dependence of oxidation charge
upon potential may be obtained by integrating the
transients. This charge increases with increasing poten-
tial. There is a sudden charge increase when the poten-
tial increases from 0.75 to 0.8 V, yet changes only a
little in the potential range 0.8–1.0 V, i.e. where no
further electrochemical reaction occurs. Further in-
creases in the potential above this limit lead to a further
increase in the oxidation charge. The sudden charge
increase indicates that the latter nucleation process
accelerates the oxidation of the PtPc film.
Fig. 3 shows the current–time transients after the
potential steps are triggered on the conductive film. No
peaks or overlapped peaks are discernible in these
transients, indicating that the phase transformation
caused by further oxidation of the conductive materials
is not under nucleation-growth control. Inset in this
Fig. 2. Potentiostatic transients for the oxidation of a reduced
PtPc film on an 11 mm diameter carbon microelectrode in 0.1
−
3
mol dm
TBAClO +AN. The potential was stepped from 0
4
V to the following values. (a) 0.5 V; (b) 0.6 V; (c) 0.725 V; (d)
.75 V; (e) 0.80 V; (f) 0.85 V; (g) 0.90 V; (h) 1.0 V; (i) 1.1 V;
j) 1.2 V.
0
(
2
(a)) is not diffusion controlled, as the current is not
−
1/2
proportional to t
. At short time there is a high
current due to double layer charging. The decrease of
current is halted by a broad shoulder centred at about
figure shows a linear relationship of these transients
0
.03 s. This overlapped peak becomes more obvious
−1/2
with t
. The relationship indicates that the electro-
and sharper upon increasing the step potential, accom-
panied by a shift of the peak to shorter time. This is
definitive evidence for nucleation and growth in the
film. The potential step to 0.8 V (Fig. 2(e)) produces
another deviation at time intervals later than the peak
mentioned above. This deviation can be considered as a
new peak overlapping the declining part of the current
transient. This peak grows rapidly when higher poten-
tials are applied to the film, accompanied by an obvious
chemical process is under diffusion-control. The further
increase in charge with increasing potential suggests
that the further oxidation of the film is a faradaic
process rather than a capacitance charging–discharging
process.
Fig. 4 shows the current–time transients on the
overdoped film upon the step from a base potential of
1.35 V to different values cathodic of that potential.
The transient at 0.85 V appears to contain the overlap
Fig. 3. Potentiostatic transients for the further oxidation of a conductive PtPc film on an 11 mm diameter carbon microelectrode in
−
3
0
1
.1 mol dm
TBAClO +AN. The potential was stepped from 0.8 V to the following values. –, 1.0 V; ---, 1.1 V; ···, 1.2 V; —··—,
4
−
1/2
.3 V. Inset: Dependence of the transient currents on t
.

J. Jiang, A. Kucernak / Electrochimica Acta 46 (2001) 1223–1231
1227
Fig. 5 shows the current–time transient on the con-
ductive film upon a potential step starting at 0.8 V to
different cathodic final potentials. No indications of a
peak or shoulder are evident in these transients, indicat-
ing that the phase transformation is not nucleation-
growth controlled during further oxidation of the
conductive film. The linear relationship of these tran-
−
1/2
sients at long time intervals with t
inset in Fig. 5,
indicates that electrochemical process is diffusion-
controlled.
Because of the overlapping of different responses in
the potentiostatic transients, we make no attempt to
simulate the kinetics of the solid–solid phase transfor-
mation processes using known 2-D or 3-D nucleation
models. Modelling these transients is difficult, as the
physical nature of the deposit neither fits a 2-D or 3-D
nucleation model. The appearance of nucleation-
growth loop in current–time transients depends upon
many factors, electrolyte composition and concentra-
tion, electrode structure, temperature, and so on.
Fig. 4. Potentiostatic transients for the reduction of an over-
doped PtPc film on a 11 mm diameter carbon microelectrode in
−
3
0.1 mol dm
TBAClO +AN. The potential was stepped
4
from 0.8 V to the values shown. Inset: Relationship between
−
1/2
the transients current and t
.
3
.3. The effect of anion
of a diffusion-controlled response and a very broad,
unchanging peak at longer time. This broad peak sug-
gests a slow nucleation process and does not change
significantly with shifts in potential. It grows very
slowly as the step potential decreases. Another new
peak evolves at 0.2 s when the potential is stepped to
Fig. 6(a) and (b) shows the current–time transients
for different PtPc films in acetonitrile medium contain-
ing 0.1 mol dm
−3
−
tetrabutylammonium with BF4 and
−
−
PF6 as anions. Compared to the ClO4 anion, the
following common features are observed. The transients
are also strongly dependent upon the oxidised states of
the film. Two nucleation processes occur in the reduced
film when a high potential is applied. Once the film
becomes conductive, its reduction or further oxidation
are diffusion-controlled at long time intervals, no indi-
cation of phase transformation is found from these
0
.7 V. This peak grows with the increase in magnitude
of the potential step, accompanied by a large time shift.
These transients indicate that a solid–solid phase trans-
formation takes place involving one, although possibly
two, separate nucleation-growth steps during the reduc-
tion of the over-doped film.
Fig. 5. Potentiostatic transients for the reduction of a conductive PtPc film on a 11 mm diameter carbon microelectrode in 0.1 mol
−
3
dm
tetrabutylammonium perchlorate in acetonitrile. The potential was stepped from 0.8 V to the following values. –, 0.3 V; ---,
0.2 V; ···, 0.1 V; —··—, 0 V.

1228
J. Jiang, A. Kucernak / Electrochimica Acta 46 (2001) 1223–1231
−
6
The phase transformations of PF -doped film seem to
undergo three nucleation-growth steps, evidenced by
three nucleation waves in the transients.
4. Discussion
4.1. Solid–solid phase transformations
The phenomenon of a solid–solid phase transforma-
tion controlled by nucleation and growth in the micro-
crystals of TCNQ has been analysed by Bond et al.
[14,15]. They developed a model to explain the princi-
pal experimental features of the solid–solid phase
transformations [15]. In this model shown in Fig. 7(a),
a microcrystal (phase 1) is assumed to be a hemisphere
whose volume does not change during the phase trans-
formation to its salt (phase 1%). The electrolyte solution
and electrode constitute phase 2 and 3, respectively.
Nucleation is most likely to occur at the three-phase
boundary (1,2,3) since this is the location that min-
imises the emergent area of the high-energy (1,1%) inter-
face in the critical nucleus, and this is the point at
which iR effects are minimized. Furthermore, consider-
ation of conductivity and transport pathways also sug-
gests that the three-phase boundary is the region in
which an electrochemical reaction is most likely to
occur.
Fig. 6. Potentiostatic transients for PtPc films with different
states on a 11 mm diameter carbon microelectrode in acetoni-
−
3
trile medium containing 0.1 mol dm
of the tetrabutylammo-
−
−
nium salt of either: ···, BF4 ; –, PF6
.
They obtained a relationship between the nucleation
rate h, the overpotential p, and the specific interfacial
free energy k1−1% of the two-phase boundary between
the microcrystal and its salt during the transformation
3
ꢀ
−4yk1−1´
ꢁ
h=h exp
(1)
0
2
3
kT(nFz p)
m
where h is the maximum possible nucleation rate, z
0
m
the molar density of the micro-crystal, and the other
terms have their usual meanings. The critical over-
potential
−
1/2 3/2
ꢀ
4y
ꢁ
k
1−1´
pcrit
=
(2)
3
kT ln(h /)
nFzm
0
where  is the minimum detectable nucleation rate.
Eq. (1) shows that the nucleation rate is determined
by the specific interfacial free energy k1−1% of the two-
phase boundary in the critical nucleus at a given over-
potential. A large k1−1% leads to a decrease in the
nucleation rate. The rate is improved by an increase of
overpotential if k1−1% remains constant. Eq. (2) demon-
strates that the greater k1−1%, the greater the critical
overpotential for the nucleation process. k1−1% is
strongly dependent upon the composition of the salt
phase 1%. The properties of phase 1% are determined by
the degree of oxidation of phase 1. The applied poten-
tial controls the oxidation state of phase 1 and affects
Fig. 7. Schematic representation of solid–solid phase transfor-
mations occurred in PtPc microcrystals on the microelectrode
in acetonitrile electrolyte. (a) Bond model [14]; (b) five-phase
system model; (c) nucleation triggering on the conductive film.
Reduced PtPc, 1; lightly doped PtPc, 1%; Conductive PtPc, 1%%;
Electrolyte, 2; Electrode substrate, 3.
transients. For the reduction of the over-doped film, the
phase transformations show anion-dependence. The re-
−
4
duction of the BF -doped film triggers two nucelation-
growth controlled solid–solid phase transformations.
the nucleation kinetics in the microcrystal. The k1−1%-

J. Jiang, A. Kucernak / Electrochimica Acta 46 (2001) 1223–1231
dependent critical overpotential pcrit is determined by
4.2. Switching reaction
1229
the oxidation state. The response is quite different to
that expected from normal electrodeposition theory.
Anions must be incorporated into the film for charge
compensation during oxidation of the film. The char-
acteristics of phase 1% may be anion-dependent, and it
appears as if the nucleation kinetics are also affected
by the doping anion.
This model can be used to analyse our experimen-
tal results. The partially oxidised PtPc microcrystal
film has three quasi-states: lightly doped, conductive
and over-doped [31]. When a small oxidation poten-
tial is applied to the undoped film, the film becomes
lightly doped undergoing a solid–solid phase transfor-
mation which can be analysed using the model in
Fig. 7(a). The characteristics of the lightly doped film
are expected to change slightly and do not lead to a
The study of the switching reaction of MPc films has
received less attention although they are prospective
electrochromic materials. In conductive polymers, the
switching reaction of electrochromic polyaniline film
has been widely investigated. Peter and coworkers per-
formed cyclic voltammetric and chronoamperometric
experiments using a fast operational amplifier poten-
tiostat equipped with positive feedback iR compensa-
tion to study the rate of switching in thin polyaniline
−
3
film in 0.1 mol dm
H SO4 on a microelectrode
2
[23,24]. They found that the switching process is con-
trolled by the RC time constant during reductive steps,
whereas oxidative steps were slower and kinetically
controlled. This observation is different from other
published work in which the switching is assumed to be
determined by the kinetics of anion insertion and re-
moval [34,35]. Lacroix and coworkers argued that, if
the switching reaction is controlled by mass transport,
then it is limited by a process involving a moving front
across the thickness of the film and not by the diffusion
of anions [25,26]. Their constant velocity of an advanc-
ing boundary may be the result of a chemical reaction
or a phase transition. They considered the effect of cell
resistance and the film capacitance as an alternative
explanation. In this case, the current response [25]
significant change in k1−1%. The nucleation rate is de-
termined by the applied overpotential, and increases
with increasing p. When the applied potential is fur-
ther increased, a large amount of anions are incorpo-
rated into the film and the film becomes highly
conductive. There is a substantial change in the film
character, resulting in a significant increase in k1−1%
.
Due to the solid–solid phase transformation within
the film, it undergoes another nucleation-growth step.
At these potentials, the solid–solid phase transforma-
tion can be described using a model shown in Fig.
ꢀ
E
ꢁ ꢂ −t
n
I=
exp
(3)
RNC
RNCC
7
(b). The nucleation of a new phase 1%% is also most
likely to occur at the three-phase boundary (1%, 2, 3)
according to Bond’s analysis. In this nucleation, the
rate is found to depend greatly on the applied poten-
where E is the applied potential step, R the non-com-
NC
pensated cell resistance, and C the film capacitance.
This current will decline to 1% of its original value in a
time of five RNCC.
tial. This fact confirms that k1%−1%% is larger than k1−1%
.
In our case, the switching reaction of the PtPc film is
found to depend greatly on the initial state of the film.
On the reduced film, the switching is under nucleation-
growth control.
Once the film undergoes this nucleation process, the
electrochemical reaction is accelerated and gives a
sharp peak in cyclic voltammograms. After these pro-
cesses the film will be in the conductive state.
As a conductive film, the electrochemical transfor-
mation of the film can be triggered from the two-
phase interface (2,1%%) rather than the three-phase
Once the film become conductive, the kinetics of the
switching process seems to change both on the reducing
step and on further oxidation steps. The transient cur-
−
1/2
rent has a linear relation with t
at long time
(
1%,2,3) region because there is no significant iR drop
intervals. The charging and discharging of the film
capacitance through the uncompensated cell resistance
is expected to be completed in less than one millisec-
ond. But the decline in current occurs over a much
longer time, and thus the possibility that the RC time
constant is responsible for the switching can be ruled
out. The front velocity is expected to be much faster in
MPc’s film than in conductive polymer films. If a
moving front across the thickness of the film is consid-
ered as an explanation, the switching should have been
completed much more quickly than observed based on
within the film. The model is shown in Fig. 7(c). The
nucleation process will be triggered from all directions
and quickly completed in the absence of mass-trans-
port limitations, and no peak will be found during its
current–time transients. The over-doping of the film
leads to a decrease in the conductivity of the film.
During reduction of the over-doped film, nucleation
of the reduced phase at the three-phase boundary be-
comes more likely again and the conversion proceeds
along similar lines to that shown in Fig. 7 (b). Nucle-
ation processes occurring during reduction of the
over-doped film show anion-dependence, especially at
higher overpotentials.
−
1
the reported front velocity of 0.2–6 mm s
in conduc-
tive polymers [36–38]. Thus, it is most likely that the
switching is controlled by the kinetics related to anion

1230
J. Jiang, A. Kucernak / Electrochimica Acta 46 (2001) 1223–1231
diffusion. This transport may be enhanced by a signifi-
cant contribution of migration to ion transport within
the internal structure of the film. On the over-doped
film, the switching reaction becomes nucleation-growth
controlled again.
Acknowledgements
We thank Dr N. Long and Mr Mongay-Batalla for
providing PtPc samples. We wish to thank Johnson
Matthey for the loan of precious metals. This project
has been funded by the United Kingdom EPSRC and
MOD/DERA under grant GR/L 57920.
The diffusion coefficient of anions in the film can be
determined from the slope s of the linear part of i
−
1/2
versus t
plots in Figs. 3 and 5 by
/2 −1/2
1
s=NFAD cy
(4)
where c is the concentration of the anions in the film.
References
As c within the film is difficult to determine accurately,
1
/2
1/2
the results are presented in D c form. The D
c
[1] G.C.S. Collins, D.J. Schiffrin, J. Electrochem. Soc. 132
(1985) 1835.
−
5
−2 −1/2
values range from 1–3×10
mol cm
s
, de-
pending on the stepping potential. These values have
[2] M. Kimura, T. Horai, K. Hanabusa, H. Shirai, Chem.
Lett. 7 (1997) 653.
[3] G. Guilloud, J. Simon, Chem. Phys. Lett. 219 (1994) 123.
1/2
the same order of magnitude as D c values reported
−
4
−
4
−
for BF , ClO and PF6 in conducting polymer sys-
[
[
[
[
4] M.M. Nicholson, Ind. Eng. Chem. Prod. Res. Div. 21
1982) 261.
5] J.A.R. Van Veen, C. Visser, Electrochim. Acta 24 (1979)
21.
6] C. Coutanceau, P. Crouigneau, J.M. Leger, C. Lamy, J.
Electroanal. Chem. 379 (1994) 389.
7] T. Inabe, S. Nakumura, W.B. Liang, T.J. Marks, R.C.
Burton, C.R. Kannewurf, K.I. Imeada, J. Am. Chem.
Soc. 107 (1985) 7224.
tems [39], strongly suggesting that the switching process
is controlled by anion diffusion within the PtPc film.
(
9
5
. Conclusions
Solid–solid phase transformations and switching re-
action occurring in the redox process of PtPc micro-
crystals have been investigated by chronoamperometry
on a rubbed microelectrode in acetonitrile medium
[8] M.Y. Ogawa, J. Martinsen, S.M. Palmer, J.L. Stanton, J.
Tanaka, R.L. Greene, B.M. Hoffman, A.J. Ibers, J. Am.
Chem. Soc. 109 (1987) 1115.
[9] I.L. Kogan, K. Yakushi, J. Mater. Chem. 7 (1997) 2231.
10] H. Yamakado, K. Yakushi, N. Kosugi, H. Kuroda, A.
Kawamoto, J. Tanaka, T. Sugarto, M. Kinoshita, S.
Hino, Bull. Chem. Soc. Jpn. 62 (1989) 2267.
−
3
containing 0.1 mol dm
tetrabutylammonium with
−
−
−
BF , ClO or PF6 as anions. Chronoamperometry
4
4
[
[
proves to be a useful technique for providing simulta-
neously information about both solid–solid phase
transformation and the switching reaction occurring in
these electrochromic microcrystals. The unique advan-
tage of the rubbed microelectrode is that well-defined
voltammograms and current-time transients free of iR
errors can be obtained. Both the solid–solid phase
transformation kinetics and switching reaction are de-
pendent upon the degree of oxidation of the film. For
the reduced film, the first nucleation-growth process
occurs at low potentials (B0.70 V). An increase in
applied potential triggers the second nucleation-growth
process while at the same time accelerating the first
nucleation-growth process. Once the film becomes con-
ductive, no shoulder is observed during the current–
time transient for either reduction or further oxidation
of the conductive film and the phase transformations
are quickly completed in a short time. The switching
process is controlled by kinetics related to anion trans-
port within the film. For the over-doped film, nucle-
ation-growth controlled phase transformations are also
triggered during the reduction of the film, and a phase
transformation process controls the reduction.
11] I.L. Kogan, K. Yakushi, Electrochim. Acta 43 (1998)
2
053.
[
[
12] J. Jiang, A. Kucernak, Electrochim. Acta 45 (2000) 2227.
13] N. Toshiwa, T. Tominaga, S. Kawamura, Bull. Chem.
Soc. Jpn. 69 (1996) 245.
[14] A.M. Bond, S. Fletcher, P.G. Symons, Analyst 123 (1998)
1891.
15] A.M. Bond, S. Fletcher, F. Marken, S.J. Shaw, P.G.
Symons, J. Chem. Soc. Faraday Trans. 92 (1996) 3925.
16] M.F. Suarez, A.M. Bond, R.G. Compton, J. Solid State
Electrochem. 4 (1999) 24.
17] J.Q. Chambers, K. Scaboo, C.D. Evans, J. Electochem.
Soc. 143 (1996) 3039.
[
[
[
[
18] S
&
. Komorsky-Lovri c´ , J. Electroanal. Chem. 397 (1995)
2
11.
[19] T. Yea, K.M. Yin, J. Carbajal, R.E. White, J. Elec-
trochem. Soc. 138 (1991) 2869.
[20] T.F. Otero, E. Angulo, Solid State Ionics 63–65 (1993)
8
03.
[
[
21] K. Aoki, J. Electroanal. Chem. 373 (1994) 67.
22] T.F. Otero, H. Grande, J. Rodriguez, J. Electroanal.
Chem. 394 (1995) 211.
23] M. Kalaji, M. Peter, L.M. Abrantes, J.C. Mesquita, J.
Electroanal. Chem. 274 (1989) 289.
[
[
A model of solid–solid phase transformation devel-
oped by Bond and co-workers was used to analyse the
nucleation process of these micro-crystals. Anions ap-
pear to have no substantial influence on the solid–solid
phase transformations and switching reaction.
24] M. Kalaji, L. Nyholm, L.M. Peter, J. Electroanal. Chem.
3
13 (1991) 271.
[25] J.C. Lacroix, K.K. Kanazawa, A. Diaz, J. Electrochem.
Soc. 136 (1989) 1308.

J. Jiang, A. Kucernak / Electrochimica Acta 46 (2001) 1223–1231
1231
[
[
26] J.C. Lacroix, A. Diaz, J. Electrochem. Soc. 135 (1988)
457.
27] A.B.P. Lever, E.R. Milaeva, G.P. Speier, in: C.C.
Leznoff, A.B.P. Lever (Eds.), Phthalocyanines: Properties
and Applications, vol. 3, VCH, New York, 1993.
28] J. Koryta, J. Dvo rˇ a´ k, L. Kavan, Principles of Eletcro-
chemistry, second edition, Wiley, Chichester, UK, 1993.
29] C.J. Brown, J. Chem. Soc. A (1968) 2494.
[33] I. Betova, M. Bojinov, E. Lankinen, G. Sundholm, J.
Electroanal. Chem. 472 (1999) 20.
[34] T. Osaka, K. Naoi, S. Ogano, J. Electrochem. Soc. 124
(1987) 2096.
[35] E.M. Genies, G. Bidan, A. Diaz, J. Electroanal. Chem.
149 (1983) 101.
[36] J.C. Lacroix, K. Fraoua, P.C. Lacaze, J. Electroanal.
Chem. 444 (1998) 83.
[37] Y. Tesuka, K. Aoki, K. Shinozaki, Synth. Met. 30 (1989)
369.
[38] K. Aoki, J. Cao, H. Hoshino, Electrochim. Acta 38
(1993) 1711.
[39] P. Marque, J. Roncali, F. Garnier, J. Electroanal. Chem.
218 (1987) 107.
1
[
[
[
30] M. Ashida, N. Uyeda, E. Suito, Bull. Chem. Soc. Jpn. 39
(
1966) 2616.
31] J. Jiang, A. Kucernak, J. Electroanal. Chem. 490 (2000)
7.
32] M. Grzeszczuk, P. Posk, J. Electroanal. Chem. 387 (1995)
9.
[
[
1
7
.