Journal of
Electroanalytical
Chemistry
Journal of Electroanalytical Chemistry 599 (2007) 52–58
www.elsevier.com/locate/jelechem
Fluorescence of polyaniline films on platinum surfaces. Influence
of redox state and conductive domains
P. Soledad Antonel, Fernando V. Molina, Estela M. Andrade
*
INQUIMAE, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, C. Universitaria, Pabellon II,
Buenos Aires C1428EHA, Argentina
Received 12 April 2006; received in revised form 29 June 2006; accepted 22 August 2006
Available online 11 October 2006
Abstract
The photoluminescence of polyaniline films onto platinum electrodes has been studied as a function of the polymer oxidation state
between the fully reduced (leucoemeraldine) and half oxidized (emeraldine) forms. The fluorescence emission spectra have been corrected
for absorption and reflection from the underlying metal surface. From these results the quantum yield relative to the reduced state as a
function of electrode potential has been obtained. This quantum yield is maximum for the reduced state and decays between ca. 0.3 and
0.4 V in the standard hydrogen electrode scale, reaching a minimum value. This behavior is interpreted in terms of three oxidation states:
leucoemeraldine, intermediately oxidized protoemeraldine and emeraldine, each having different absorption and emission behavior. In
the emeraldine form, the presence of conductive, quasi-crystalline domains is considered to fully quench emission from excitons located
both inside the domains and within a range equal to the exciton delocalization length. A good agreement with the experimental data is
found.
2006 Elsevier B.V. All rights reserved.
Keywords: Conducting polymer; Metal surface; Photoluminescence; Quantum yield; Quenching
1. Introduction
Conducting polymers show interesting optical properties (such as fluorescence, electrochromism, etc.) leading
to a number of proposed applications such as sensors, displays and other optoelectronic devices [1–9]. Aryl amine
polymers, such as polyaniline (PANI), show photoluminescence in their reduced state, which decreases when the polymer is oxidized [10–14]. Even when the quantum yield is
small, the study of these properties is interesting because
it will give insight into the electronic structure of these
materials. There are few studies concerning the photoluminescence of PANI and similar polymers. Son et al. [10]
studied the fluorescence quenching of PANI upon oxida-
*
Corresponding author. Tel.: +54 11 4576 3378/80x119/230; fax: +54
11 4576 3341.
E-mail address: [email protected] (E.M. Andrade).
0022-0728/$ - see front matter 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jelechem.2006.08.007
tion, and developed a simple model assuming that charge
carriers (polarons) are responsible for quenching, allowing
for overlapping between quenching centers. Thorne et al.
[11] studied the time-resolved fluorescence of the reduced
form of PANI, leucoemeraldine (LE), and explained the
time decay curves as due to the diffusion in one dimension
of randomly located traps, postulated to be short segments
of the oxidized, emeraldine (EM), form of PANI. Ram
et al. [12] applied fluorescence microscopy to study the ageing of PANI films in room atmosphere, and explained the
observed decay in terms of a diffusion phenomenon.
Recently [14] the fluorescent emission of PANI in different
redox states was studied, measuring the quantum yield of
the base forms of LE and EM in solution with N-methylpirrolidinone as solvent. The quantum yield for both
forms were found to be /LE = 1.2 · 103 and /EM =
1.0 · 103. Also, the fluorescent behavior of PANI films
supported on platinum electrodes was studied qualitatively;
it was found a complex thickness dependence of the fluo-
P.S. Antonel et al. / Journal of Electroanalytical Chemistry 599 (2007) 52–58
rescent emission, and a potential dependence similar to that
previously reported [10].
The redox behavior of aryl amine polymers such as
PANI is usually interpreted in terms of three redox states,
the reduced leucoemeraldine, half oxidized emeraldine and
fully oxidized pernigraniline, the conversions between them
involving two-electron reactions [1–3,15–23]. The conversion between LE and EM (known as redox switching) is
the most important because pernigraniline is unstable in
aqueous solution. It involves an intermediate state, usually
known as radical cation or polaron state, which shows
unpaired spins in EPR measurements [24–26] and is often
considered to be a transient one. However, one of the first
studies on PANI chemistry [27] described it as a distinct,
stable, oxidation state called protoemeraldine (PE), and
recently it was also considered to be a stable state, explaining the temperature dependence of the PANI magnetic susceptibility in terms of the LE–PE–EM equilibrium [26].
More recently, the same scheme was successfully applied
to explain the electrochemomechanical behavior of thick
PANI films [28], considering the presence of a distribution
of formal redox potentials for the PE to EM oxidation [29–
31]. From a structural point of view, it has been proposed
[32] and recently shown [33,34] that in the conductive
(doped emeraldine) state, PANI films are actually composed of small, highly interacting, quasi-crystalline, conductive domains separated by low conductivity regions; it
has been also proposed [35] that conduction is achieved
through resonance tunneling between the conductive
domains.
In this work the potential dependence of the relative
quantum yield of PANI films supported on platinum electrodes is studied. The fluorescent emission is measured,
corrected for absorption and the quantum yield, relative
to the potential of maximum emission, is obtained. We will
present first the correction procedure, then the experimental methods, afterwards the results will be discussed and
finally the conclusions will be enumerated.
2. Relative quantum yield of films supported on metals
In the analysis of photoluminescence of solid films, it
must be considered the absorption of both the excitation
and emission beams by the film material in order to have
quantitatively correct results. When a fluorescent film of
thickness L, supported on a reflective surface, is irradiated
with an excitation beam of intensity I0 (of wavelength k0)
there is, in principle, reflection of both the excitation and
the emitted beams. Fig. 1 illustrates a volume element in
the film in this situation. The emitted intensity measured
at the detector Im, at a given emission wavelength k, will
have two contributions: directly emitted Ied and reflected
Ier. Each of these two will in turn have two contributions:
emission due to the direct excitation beam I0d and that due
to the reflected beam I0r. In general absorption by the film
will be present with extinction coefficients a0 at k0 and ae at
k. Considering the symmetry of the problem, it can be writ-
53
θ 0'
θ0
I0
I 0r
I 0d
Ied
I er
Im
θe
L
θe'
x
Fig. 1. Optical paths in the fluorescence of metal supported emitting films.
See text for details.
ten for the total intensity at a given wavelength dIe(k) emitted from a layer of thickness dx:
a0 x
a0 ð2L xÞ
dI e ¼ uðkÞI 0 exp þ f0 exp nc Adx
cos h0
cos h0
ð1Þ
where u(k) is the fluorescent emission relative to I0 at wavelength k, f0 is the metal reflectance at k0, nc is the chromophore concentration and A is the area covered by the
excitation beam. The intensity reaching the detector dIm
will be:
ae x
ae ð2L xÞ
þ fe exp ð2Þ
dI m ¼ xdI e exp cos he
cos he
where x is a geometrical factor depending on the instrument and fe the metal reflectance at k. The corrected emission intensity per unit thickness, Im,c, can then be obtained
after integrating over the film thickness, assuming that the
quantum yield is independent of thickness, resulting in
Eq. (3):
8
<
1
a0 L
ae L
I m; c ¼ I m
1 exp : a0 þ ae
cos h0 cos he
cos h0
cos he
2ae L fe exp cos
he
aL
aL
1 exp 0 þ e
þ a0
cos h0 cos he
cosaehe
cos h0
2a0 L f0 exp cos
h0
a0 L
ae L
1 exp
þ
cos h0 cos he
cosa0h0 þ cosaehe
91
2a0 L
2ae L =
f0 fe exp cos
h0
cos he
a0 L
ae L
1
þ
exp
þ
a0
;
cos h0 cos he
þ ae
cos h0
cos he
ð3Þ
54
P.S. Antonel et al. / Journal of Electroanalytical Chemistry 599 (2007) 52–58
Using a suitable reference, the absolute fluorescence quantum yield / can be obtained after integrating Im,c over the
emission wavelength range. Because we are interested here
in the potential dependence of / for conducting polymer
films, where Im,c is a function of both k and the applied
potential E, we will instead compute the relative fluorescence quantum yield, /r, for these films using the following
expression:
R
I m;c ðk; EÞ dk
R
/r ðEÞ ¼
I m;c ðk; EC Þ dk
ð4Þ
Here, EC is the cathodic (most negative) potential limit,
where the polymer is fully reduced and its emission is at
a maximum. Thus, /r(E) will have unit value at EC and
decrease as the polymer is oxidized.
measurements [36],the relationship between them is dependent on polymerization conditions, thus here the charge
will be reported. The fluorometric measurements were
performed using a PTI Quantamaster stationary spectrofluorometer. Fluorescence emission spectra in the range
330–600 nm were recorded potentiostatically at increasing
electrode potentials between 0.1 and 0.6 V. The excitation
wavelength was in most cases k0 = 310 nm; in some experiments it was varied in order to examine the number of
chromophores, with the emission range adjusted accordingly. The incidence angle was set at 30, to avoid direct
reflection of the excitation beam into the photomultiplier.
At each applied potential, the emission spectrum was
recorded when the current (absolute value) was below
1 lA cm2. The time required was about 15 min at each
potential, and it was determined by chronoamperometric
measurements. The response of the bare Pt surface was also
observed.
3. Experimental
3.2. Absorption spectra of PANI films on gold
3.1. Fluorescence of polymer films on platinum electrodes
PANI films were synthesized electrochemically from
0.1 M aniline in 1 M sulfuric acid. AR grade chemicals
and high purity water from a Milli Q system were
employed. Aniline was distilled under reduced pressure
and reducing conditions shortly before use. A Teq-03
potentiostat under computer control was employed. The
electropolymerization was performed by potential cycling
between 0.05 and 1.00 V at a scan rate, v, of 0.025 V s1.
Both the electropolymerization and the fluorescence experiments were conducted in a specially constructed cell,
allowing to perform electrochemical processes with simultaneous fluorescence or absorbance measurements [14].
The working electrode was a platinum plate mounted on
a rotatory holder in order to adjust the incidence angle.
The electrode was polished with alumina slurries of successively smaller particle sizes down to 0.05 lm, washed in an
ultrasonic bath, and rinsed with ultrapure water. Before
each experiment, it was soaked in sulfonitric mixture and
rinsed thoroughly. A platinum auxiliary electrode was
placed in a compartment separated by a porous glass,
whereas the reference electrode was a reversible hydrogen
one (RHE) in the same solution. The potentials were
converted to and are presented on the standard hydrogen
electrode scale (SHE).
The polymerization was monitored through the voltammetric reduction charge, Qtot, in the full potential range,
and was stopped at different Qtot values, in order to obtain
films of different thicknesses. After polymerization, the
working electrode was extracted from the cell and carefully
rinsed with high-purity water. The cell was filled with 1 M
sulfuric acid, and voltammetric cycling was performed
between 0.05 and 0.6 V, at 0.025 V s1. The voltammetric
reduction charge (from EM to LE), Q, in these conditions
was recorded as a measure of film thickness; although the
thickness could be estimated from Q through ellipsometric
For the absorbance measurements, semitransparent
gold films 40 nm thick were deposited by evaporation onto
quartz windows, and their absorption spectra in the range
310–600 nm were recorded. Then, PANI films were electropolymerized as described above, and their absorption
spectra were recorded potentiostatically at potentials
between 0.1 and 0.6 V, in the same conditions as for the
fluorometric experiments. In all cases the absorbance of
the bare gold film was subtracted. All the absorbance spectra were obtained using a Hewlett-Packard 8452A diode
array spectrophotometer with the same cell described
previously [14].
3.3. Reflectance spectra of platinum electrodes
The total reflectance of the platinum electrodes, polished
as described above, was measured in the range 310–600 nm
using a Shimadzu PC-3101 spectrophotometer equipped
with a ISR-260 reflectance accessory, using BaSO4 as
standard.
3.4. Chronoamperometric measurements
Chronoamperometric measurements with PANI films
electropolymerized as described before were performed in
1 M H2SO4 at different applied potentials between 0.1
and 0.6 V. In each experiment, the potential was held at
the cathodic limit (0.1 V) until the polymer film was fully
reduced and relaxed as described by Rodriguez Presa
et al. [37]. Then, the potential was varied in increasing steps
up to 0.6 V, then in decreasing steps to the cathodic limit.
In each case, the current transient was recorded until a stationary, nearly zero value was reached (15–20 min, approximately, depending on thickness). Charge values at each
potential were obtained by integrating and accumulating
the current transients.
P.S. Antonel et al. / Journal of Electroanalytical Chemistry 599 (2007) 52–58
55
4. Results and discussion
Im,c (A.U.)
0.1 V
0.2 V
0.3 V
0.4 V
0.5 V
0.6 V
Im (A.U.)
300
400
λ /nm
500
600
Fig. 3. Emission spectra of PANI film (16.5 mC cm2) on platinum in
1 M H2SO4 at different excitation wavelengths. (a) E = 0.1 V; (b) E =
0.6 V. (—) k0 = 310 nm; (- - -) k0 = 380 nm; (- Æ - Æ -) k0 = 400 nm; ( )
k0 = 420 nm.
1.0
0.8
0.6
0.4
0.2
0.0
0.1
0.2
0.3 0.4
E/V
0.5
0.6
Fig. 4. Relative fluorescence quantum yield for PANI films in 1 M H2SO4
as a function of the applied potential for different thicknesses. Points,
experimental data; lines, fitting to Eq. (15). (s —) 9.8 mC cm2; (h - - -)
20.3 mC cm2; (n ) 30.9 mC cm2.
a
b
1.5
A
1.0
0.5
0.0
300
b
φr
Fig. 2 shows the corrected fluorescent emission spectra
of a PANI film on Pt at different potentials among with
the corresponding absorption spectra. These were obtained
from the experimental spectra through Eq. (3), after
subtracting the metal dispersion background, using the
absorption spectra of Fig. 2b and the platinum reflectance
spectrum (not shown); because PANI films are porous and
incorporate electrolyte [28], it is assumed that its refractive
index is approximately equal to that of the external medium, thus a00 ¼ a0 and a0e ¼ ae (see Fig. 1). The excitation
wavelength was fixed at 310 nm, which falls inside de
p–p* absorption envelope of the benzenoid units [38]. The
main characteristics of the emission spectra have been
described before, including the presence of small peaks
attributed to Raman dispersion of the underlying metal
surface [14]. It is observed a decrease in emission intensity
as the polymer film is oxidized (going from 0.1 to 0.6 V),
leaving a small but not null emission when the polymer is
in the emeraldine state. This behavior has been attributed
to quenching by charge carriers [10–14].
In order to check the presence of other quenching mechanisms, the excitation wavelength was varied in the region
380–420 nm, recording the emission spectra at different
applied potentials. Fig. 3 shows representative results. No
essential changes in the emission spectra are observed,
other than the displacement of the peaks due to Raman
dispersion from the metal surface. Changes of k0 within
the p–p* band (i.e., around 310 nm, spectra not shown)
did not show changes, either. These results indicate that
there is essentially only one chromophore at each redox
state of the PANI film (see below).
Fig. 4 shows the relative fluorescence quantum yield of
PANI films, /r, as a function of the applied potential.
Im (A.U.)
a
400
λ /nm
500
600
Fig. 2. (a) Emission spectra for a PANI film (9.8 mC cm2) on platinum
in 1 M H2SO4 at different electrode potentials (k0 = 310 nm). (b) Corresponding absorption spectra for a 9.8 mC cm2 PANI film on gold at the
same potentials.
The fluorescence quantum yields were calculated using
Eq. (4). As it was observed for the fluorescent emission
spectra at different potentials, the relative quantum yield
decays as the polymer is oxidized. That decrease mainly
takes place between 0.3 and 0.4 V, along with PANI oxidation. Beyond 0.4 V, /r remains almost constant but not
null. This limiting quantum yield is dependent on the thickness, being higher for the thinnest films.
Recently, it was proposed [28] that the leucoemeraldineemeraldine conversion takes place in two steps and that the
quinoidic, conductive structure is developed in the second
one. The redox units of PANI are assumed to be in one
of these redox states: fully reduced LE, intermediate PE
and half oxidized EM. The reactions are:
þ
þ
þ
½–B–NHþ
2 –B–NH2 –B–NH2 –B–NH2 n
þ
þ
þ
þ
½–B–NH2 –B–NH2 –B–NH –B–NHþ
2 n þ ne þ nH
ð5Þ
56
P.S. Antonel et al. / Journal of Electroanalytical Chemistry 599 (2007) 52–58
þ
þ
þ
½–B–NHþ
2 –B–NH2 –B–NH –B–NH2 n
þ
þ
þ
½–B–NH2 –B–NH2 –B–NH @Q@NHþ n þ ne þ nHþ
ð6Þ
+
where –NHþ
2 -represents protonated amine groups, –NH @
protonated imine groups, B benzenoid rings, Q quinoidic
rings and B–NH+–B stands for the radical cation (polaron)
protonated protoemeraldine state. Furthermore, the quasiequilibrium redox behavior was successfully interpreted by
introducing a formal potential distribution for reaction (6).
In this scheme, the fraction of polymer in the PE state, xPE,
and that in the EM state, xEM, are given by:
1
1 þ ðK 2 wÞm þ ð1=K 1 wÞ
ðK 2 wÞm
¼
1 þ ðK 2 wÞm þ ð1=K 1 wÞ
xPE ¼
ð7Þ
xEM
ð8Þ
with xLE + xPE + xEM = 1, xLE being the fraction of
polymer in the leucoemeraldine state. In these equations,
K 1 ¼ expðeEo01 =kT Þ; K 2 ¼ expðeEo0
2 =kT Þ; w = exp(eE/kT);
Eo01 is the formal potential of the first polymer redox
change, Eo0
2 is the mean value of the distribution for the
formal potentials of the second polymer redox change, m
(0 < m 6 1) is related to the width of the distribution
(m = 1 gives a Dirac’s delta function), E is the applied
potential and e, k and T have their usual meaning. The
charge spent in oxidizing the polymer is given by:
x
PE
Q ¼ Qm
þ xEM
ð9Þ
2
where Qm is the charge spent to fully oxidize the polymer
from LE to EM. Chronoamperometric measurements allow
the parameters involved in Eqs. (7)–(9) to be obtained.
Fig. 5 shows representative results for the Q vs. E curves
obtained; the data points shown are average of the increasing and decreasing potential runs. The data can be satisfactory fitted (line) to Eqs. (7)–(9) with Eo01 ¼ 0:32 0:01V;
Eo02 ¼ 0:46 0:02V, m = 0.52 ± 0.08 and the Qmax values
indicated in the figure; it should be noted that these values
are very similar to the voltammetric charges Q. The agreement is very good, with Eo01 ; Eo02 and m independent of the
film thickness within experimental error. The inset in
Fig. 5 shows the fraction of redox units in the LE, PE
and EM states as a function of the applied potential.
Knowing the polymer speciation at each potential it
should be possible to decompose the absorption spectra
as a combination of the spectra of the three forms:
A ¼ xLE ALE þ xPE APE þ xEM AEM
ð10Þ
where the Ai are the absorbances of the separate forms.
This was verified, and the individual spectra obtained are
shown in Fig. 6. The absorbance spectra at different potentials are found to be, within experimental error, linear combinations of these individual spectra. It should be noted
also that this description is qualitatively consistent with
earlier spectroscopic studies of PANI such as that of
Stilwell and Park [39] where, in the potential range studied
here, three different absorbing species were observed.
The fluorescence decay on polymer oxidation has been
previously attributed to quenching due to either the mobility of the quinoidic domains within chains [11], or of the
excited states (excitons) [12] or both [10,14]. However, the
presence of three different species with different absorption
spectra indicates that there should be three emitting species, not necessarily with the same emission characteristics,
contributing to the observed fluorescence. Moreover, the
structure of the polymer film has to be taken into account.
As mentioned in the introduction, several studies [33–35]
have shown the presence of conductive domains in PANI
films. Fig. 7 depicts schematically the polymer structure
as proposed by Prigodin and Epstein [35], which is consistent with the recent experimental findings of Wu and
Chang [33] and Krinichnyi et al. [34]. Small ordered
domains, where the chains are strongly coupled, are separated by disordered regions. These domains become conductive upon polymer oxidation through the introduction
of energy levels within the bandgap; this will result in an
efficient quenching of the excited states appearing in these
regions. According to Wu and Chang [33], even in a fully
oxidized and doped PANI film disordered, nonconductive
regions remain, which can have dimensions in the order
of 100 nm or more, thus excitons may not be quenched if
they are far away from a metallic domain. The relevant
parameter is the exciton delocalization length; at present
there are no measurements of this length for PANI or
1.0
α
0.5
xLE
xPE
xEM
1.6
1.2
0.0
0.3
0.6
E/V
10
Α
Q /m C cm-2
20
0.8
0.4
0
0.1
0.2
0.3
0.4
0.5
0.6
E/V
Fig. 5. Stationary charge vs. potential data for PANI films in 1M H2SO4.
Points, experimental data; lines, fitting to Eqs. (5)–(7). (s—) Qmax =
23.2 mC cm2; (n - - -) Qmax = 11.4 mC cm2; (h ) Qmax =
7.7 mC cm2. Inset: predicted speciation diagram obtained.
0.0
300
400
λ /nm
500
600
Fig. 6. Absorption spectra for the three redox states of PANI
(Q = 9.8 mC cm2). (—) leucoemeraldine; (- - -) protoemeraldine; ( )
emeraldine.
P.S. Antonel et al. / Journal of Electroanalytical Chemistry 599 (2007) 52–58
57
Then, the quantum yield relative to the reduced form
(xLE = 1) can be written as
/r ¼
PE
xLE ALE þ xPE APE //LE
þ y F xEM AEM //EM
LE
xLE ALE þ xPE APE þ xEM AEM
ð13Þ
As it was observed in Fig. 4, there is a limiting quantum
yield when the polymer is fully in its EM state (xEM = 1).
This limiting quantum yield is given by:
/lim ¼ y F
/EM
/LE
ð14Þ
The relative quantum yield can be rewritten as follows:
/r ¼
Fig. 7. The structure of a polyaniline film. It is composed of small
domains, highly ordered with strong interchain coupling separated by
disordered regions. An exciton initially formed near an ordered domain
will be quenched when the polymer becomes conductive.
similar polymers. Recently, some studies for poly(phenylenevinylene), either in solution or in nonoriented films,
have estimated it to be between 6 and 17 repeat units
(about 5–15 nm) [40,41]. Theoretical calculations [42] suggest that in PANI this distance could be smaller, due to
the larger torsion angles in this polymer. Thus, it can be
expected that a fraction of the excitons can be too far from
the metallic domains to be quenched during their lifetime,
as depicted schematically in Fig. 7.
Taking into account the above discussion, we propose
the following description for the photoluminescence of
PANI films. Each of the three PANI species have fluorescent emission with different quantum yield /i; when conductive domains are present (in amount proportional to
xEM), the emission from excitons located either inside these
domains, or near to them, is efficiently quenched. These
excitons can thus be regarded as effectively nonemitting.
If yF is the fraction of far excitons (that is, the fraction
of sites in the polymer film which are separated from the
conductive domains by a distance greater than the exciton
delocalization length), the fluorescence flux, IF, can be
written as
ALE
APE
AEM
/LE þ xPE
/PE þ y F xEM
/EM
I F ¼ I a xLE
ð11Þ
A
A
A
where Ia is the absorbed flux, Ai are the absorbances at the
excitation wavelength of the respective forms, and A is
given by Eq. (10).
The fluorescence quantum yield can be calculated as
/¼
IF
Ia
ð12Þ
PE
xLE ALE þ xPE APE //LE
þ /lim xEM AEM
xLE ALE þ xPE APE þ xEM AEM
ð15Þ
As it is shown in Fig. 4, Eq. (15) fits satisfactory the experimental data, with /PE = 0.3/LE for PANI films of different thicknesses; /lim ranges from 0.12 to 0.29 for films of
Q from 30.9 to 9.8 mC cm2, respectively. The /lim values
are directly obtained from the experimental data, so that
/PE//LE is the only adjustable parameter. It should also
be noted that these results could be not fitted satisfactorily
considering only two species.
Unfortunately, because yF cannot at present be obtained
independently, the separate emission spectra for the three
species could not be obtained. The fitting results indicate
that the PE form has a quantum yield noticeably smaller
than LE. This can be attributed to the introduction of additional energy levels upon formation of the radical cation, as
it can be observed by the presence of a new absorption
band in its spectrum around 425 nm. Again, because yF
is not known, it is not possible to obtain /EM; it can be
expected, in principle, to be smaller than /PE. Even, with
these limitations, it is clear that the fluorescence behavior
of PANI films is consistent with the presence of three
redox species, each with its own absorption and emission
behavior. As already noted, /lim decreases as the film
thickness increases; this is in agreement with the
results of Wu and Chang [33], which indicate that in
thicker films the polyaniline chains are packed more closely
and ordered. As the film becomes more ordered, a higher
fraction of the polymer is involved in conductive domains
and, consequently, the fraction of far excitons will be
lower.
5. Conclusions
The behavior of the fluorescent emission of polyaniline
films onto platinum electrodes as a function of the polymer
redox state is consistent with the presence of three redox
species, with different absorption and emission properties.
The presence of quasi-crystalline conductive domains cause
strong fluorescence quenching of excitons located inside or
near these domains in the emeraldine state.
58
P.S. Antonel et al. / Journal of Electroanalytical Chemistry 599 (2007) 52–58
Acknowledgements
The authors are deeply grateful to Dr. E. San Roman
for helpful discussions and suggestions. The authors gratefully acknowledge financial support from the Universidad
de Buenos Aires (UBACYT 2004–2007 X105), the Consejo
Nacional de Investigaciones Cientı́ficas y Técnicas (CONICET, PIP 05216) and the Agencia Nacional de Promoción
Cientı́fica y Tecnológica (Grant No. 06-12467). F. V. M. is
a member of the Carrera del Investigador Cientı́fico of
CONICET.
References
[1] G.P. Evans, in: H. Gerischer, C.H. Tobı́as (Eds.), Advances in
Electrochemical Science and Engineering, vol. 1, VCH, New York,
1990.
[2] E. Samuelsen (Ed.), Science and Applications of Conducting Polymers, Hilger, Bristol, UK, 1991.
[3] G. Inzelt, in: A.J. Bard (Ed.), Electroanalytical Chemistry, vol. 18, M.
Dekker, New York, 1994, p. 89.
[4] P.G. Pickup, in: R.E. White (Ed.), Modern Aspects of Electrochemistry, vol. 33, Kluwer Academic Publishers/Plenum Press, New York,
1999, p. 549.
[5] P. Chandrasekhar, in: Conducting Polymers and Applications,
Kluwer Academic Publishers, Boston, MA, 1999 (Chapter 21).
[6] C. Marcel, J.-M. Tarascon, Solid State Ionics 143 (2001) 89–101.
[7] E.H.L. Falcao, W.M. de Azevedo, Synth. Met. 128 (2002) 149–151.
[8] O.S. Wolfbeis, Anal. Chem. 74 (2002) 2663–2678.
[9] S. Christie, E. Scorsone, K. Persaud, F. Kvasnik, Sens. Actuators B
90 (2003) 163–169.
[10] Y. Son, H.H. Patterson, C.M. Carlin, Chem. Phys. Lett. 162 (1989)
461–466.
[11] J.R.G. Thorne, J.G. Masters, S.A. Williams, A.G. MacDiarmid,
R.M. Hochstrasser, Synth. Met. 49 (1992) 159–165.
[12] M.K. Ram, G. Mascetti, S. Paddeu, E. Maccioni, C. Nicolini, Synth.
Met. 89 (1997) 63–69.
[13] J.Y. Shimano, A.G. MacDiarmid, Synth. Met. 123 (2001) 251–262.
[14] P.S. Antonel, E.M. Andrade, F.V. Molina, Electrochim. Acta 49
(2004) 3687–3692.
[15] E.M. Genies, C.T. Tsintavis, J. Electroanal. Chem. 195 (1985) 109–
128.
[16] W. Huang, B. Humphrey, A. MacDiarmid, J. Chem. Soc. Faraday
Trans. I 82 (1986) 2385–2400.
[17] D. Orata, D.A. Buttry, J. Am. Chem. Soc. 109 (1987) 3574–3581.
[18] G. Horanyi, G. Inzelt, Electrochim. Acta 33 (1988) 947–952.
[19] C. Odin, M. Nechtschein, P. Hapiot, Synth. Met. 47 (1992) 329–350.
[20] M.C. Miras, C. Barbero, R. Kötz, O. Haas, J. Electroanal. Chem. 369
(1994) 193–197.
[21] E.M. Andrade, F.V. Molina, M.I. Florit, D. Posadas, J. Electroanal.
Chem. 419 (1996) 15–21.
[22] A. Neudeck, A. Petr, L. Dunsch, Synth. Met. 107 (1999) 143–158.
[23] G. Inzelt, Electrochim. Acta 45 (2000) 3865–3876.
[24] S.H. Glarum, J.H. Marshall, J. Electrochem. Soc. 134 (1987) 2160–
2165.
[25] R. Patil, Y. Harima, K. Yamashita, K. Komaguchi, Y. Itagaki, M.
Shiotani, J. Electroanal. Chem. 518 (2002) 13–19.
[26] A. Petr, A. Neudeck, L. Dunsch, Chem. Phys. Lett. 401 (2005) 130–
134.
[27] A.G. Green, A.E. Woodhead, J. Chem. Soc. London 101 (1912)
1117–1123.
[28] L. Lizarraga, E.M. Andrade, M.I. Florit, F.V. Molina, J. Phys.
Chem. B 109 (2005) 18815–18821.
[29] D. Posadas, M. Fonticelli, M.J. Rodriguez Presa, M.I. Florit, J. Phys.
Chem. B 105 (2001) 2291–2296.
[30] D. Posadas, M.J. Rodriguez Presa, M.I. Florit, Electrochim. Acta 46
(2001) 4075–4081.
[31] D. Posadas, M.I. Florit, J. Phys. Chem. B 108 (2004) 15470–15476.
[32] F. Lux, G. Hinrichsen, V.I. Krinichnyi, I.B. Nazarova, S.D.
Cheremisow, M.M. Pohl, Synth. Met. 55 (1993) 347–352.
[33] C.-G. Wu, S.-S. Chang, J. Phys. Chem. B 109 (2005) 825–832.
[34] V.I. Krinichnyi, S.V. Tokarev, H.-K. Roth, M. Schrödner,
B. Wessling, Synth. Met. 152 (2005) 165–168.
[35] V.N. Prigodin, A.J. Epstein, Europhys. Lett. 60 (5) (2002) 750–756.
[36] C.M.G.S. Cruz, E.A. Ticianelli, J. Electroanal. Chem. 428 (1997)
185–192.
[37] M.J. Rodriguez Presa, D. Posadas, M.I. Florit, J. Electroanal. Chem.
482 (2000) 117–124.
[38] M.K. Ram, E. Maccioni, C. Nicolini, Thin Solid Films 303 (1997)
27–33.
[39] D.E. Stilwell, S.-M. Park, J. Electrochem. Soc. 136 (1989) 427–433.
[40] G.H. Gelinck, J.J. Piet, B.R. Wagewijs, K. Müllen, J. Wildeman, G.
Hadziioannou, J.M. Warman, Phys. Rev. B 62 (3) (2000) 1489–1491.
[41] A. Ruseckas, P. Wood, I.D.W. Samuel, G.R. Webster, W.J. Mitchell,
P. Burn, V. Sundström, Phys. Rev. B 72 (2005) 115214-1–115214-5.
[42] K. Harigaya, Chem. Phys. Lett. 281 (1997) 319–324.