Poly(aniline)-poly(acrylate) composite films as modified electrodes for the oxidation of NADH

Article abstract of DOI:10.1039/b001107j

Poly(aniline), electrochemically deposited on an electrode surface in the presence of poly(acrylic acid), forms a film which remains protonated, and conducting, at pH 7. The resulting modified electrode is an electrocatalytic surface for NADH oxidation at +0.05 V vs. SCE in 0.1 M citrate-phosphate buffer at pH 7. The amperometric responses of these composite poly(aniline) films for NADH oxidation were studied in detail and fitted to a kinetic model in which the NADH diffuses into the polymer film and then binds to catalytic sites within the film where it undergoes reduction to NAD⁺. The rate determining process depends on the concentration of NADH present and the polymer film thickness. A comparison of the results presented here for the poly(aniline)-poly(acrylate) films with earlier work on poly(aniline)-poly(vinylsulfonate) films shows that the currents obtained for NADH at these poly(aniline)-poly(acrylate) films are approximately one third of those obtained for the poly(aniline)-poly(vinylsulfonate) films under similar conditions, that the currents saturate at lower NADH concentration and that the response is less stable towards repeated measurements. The poly(aniline)-poly(acrylate) films are, however, less readily inhibited by NAD⁺ and possess the potential advantage that the carboxylate groups can be used as sites for chemical attachment of enzymes or NADH derivatives by using simple coupling reactions.

Full text of DOI:10.1039/b001107j

View Online / Journal Homepage / Table of Contents for this issue
Poly(aniline)–poly(acrylate) composite Ðlms as modiÐed electrodes
for the oxidation of NADH
Philip N. Bartlett* and Evelyne Simon
Department of Chemistry, University of Southampton, Southampton, UK SO17 1BJ
Received 9th February 2000, Accepted 22nd March 2000
Poly(aniline), electrochemically deposited on an electrode surface in the presence of poly(acrylic acid), forms a
Ðlm which remains protonated, and conducting, at pH 7. The resulting modiÐed electrode is an electrocatalytic
surface for NADH oxidation at ]0.05 V vs. SCE in 0.1 M citrateÈphosphate bu†er at pH 7. The
amperometric responses of these composite poly(aniline) Ðlms for NADH oxidation were studied in detail and
Ðtted to a kinetic model in which the NADH di†uses into the polymer Ðlm and then binds to catalytic sites
within the Ðlm where it undergoes reduction to NAD`. The rate determining process depends on the
concentration of NADH present and the polymer Ðlm thickness. A comparison of the results presented here
for the poly(aniline)Èpoly(acrylate) Ðlms with earlier work on poly(aniline)Èpoly(vinylsulfonate) Ðlms shows
that the currents obtained for NADH at these poly(aniline)Èpoly(acrylate) Ðlms are approximately one third of
those obtained for the poly(aniline)Èpoly(vinylsulfonate) Ðlms under similar conditions, that the currents
saturate at lower NADH concentration and that the response is less stable towards repeated measurements.
The poly(aniline)Èpoly(acrylate) Ðlms are, however, less readily inhibited by NAD` and possess the potential
advantage that the carboxylate groups can be used as sites for chemical attachment of enzymes or NADH
derivatives by using simple coupling reactions.
for oxidation of NADH and the detailed analysis of the reac-
Introduction
tion kinetics. There were three motivations for this work.
First, we wished to conÐrm the general validity of the medi-
ation mechanism through the use of another type of
poly(anion). Second, we wished to Ðnd better composite
polymer Ðlms for NADH mediation, both in terms of the
kinetics of the process and the overpotential required and in
terms of the stability of the Ðlms towards potential cycling.
Finally, we wished to use a poly(anion) that allows subsequent
attachment of dehydrogenase dependent enzymes and/or
derivatives of the coenzymes to the electrode for application in
biosensors or biofuel cells. For these reasons, we chose to
study the use of poly(acrylic acid) as a counter anion.
Poly(acrylic acid) has the same polymer backbone as
poly(vinylsulfonic acid), but the presence of the carboxylate
groups permits the attachment of biomolecules through
simple peptide coupling reactions.9,10
b-Nicotinamide adenine dinucleotide (NADH) dependent
dehydrogenase enzymes catalyse the oxidation of many
organic molecules.1 Consequently, an efficient electrode for
the oxidation of NADH to enzymatically active NAD` would
make possible the development a wide range of amperometric
enzyme sensors, in addition to the use of dehydrogenase
dependent enzymes in biofuel cells.2,3 NADH oxidation
occurs via a hydride transfer mechanism and requires a large
overpotential at bare metal and unmodiÐed carbon electrodes
in bu†er solution at pH 7. As a result of the high over-
potential, the reaction generally leads to fouling of the elec-
trode surface. An efficient redox mediator for NADH
oxidation should have a site for hydride transfer combined
with the ability to delocalize the charge within its structure.4,5
In previous studies we have shown that poly(aniline) Ðlms, in
their conducting, emeraldine, form can mediate NADH oxida-
tion at much lower overpotentials than required for bare elec-
trodes. Since NADH is rapidly hydrolysed at pH \ 7.0 while,
at the same time, the emeraldine form of poly(aniline) is
usually deprotonated and non-conducting at neutral pH, it is
necessary to use composite poly(aniline) Ðlms formed by depo-
sition of poly(aniline) with high molecular weight polymeric
counter anions in order to maintain the electroactivity of the
Ðlm at pH 7.0. The polymeric counter anions are entrapped
within the poly(aniline) Ðlm so that deprotonation of the
emeraldine form of poly(aniline) can occur only if the egress of
the protons is balanced by the ingress of cations from the
solution. Poly(aniline)Èpoly(vinylsulfonate) and poly(aniline)È
poly(styrenesulfonate) composite Ðlms have been already used
as mediators for NADH oxidation and a kinetic model for the
reaction has been proposed.6h8
Experimental
All experiments were carried out in freshly prepared solutions
using water puriÐed using a Whatman RO 50 and a Whatman
Still Plus system. Aniline was distilled under vacuum and
stored under argon at D4 ¡C before use. Hydrochloric acid
(AnalaR, BDH), potassium chloride (AnalaR, BDH), citric
acid (AnalaR, BDH), disodium hydrogen orthophosphate
dihydrate (AnalaR, BDH), hydrochloric acid (AnalaR, BDH),
b-nicotinamide adenine dinucleotide, reduced form, disodium
salt (NADH, D98%, Sigma) and b-nicotinamide adenine din-
ucleotide (NAD`, D98%, Sigma) were used without further
puriÐcation.
Poly(acrylic acid), sodium salt solution in water, was pur-
chased from Aldrich. Samples of three di†erent molecular
weights were used (1200, 8000 and 15 000 Da). In each case
the commercial solution was diluted with 4.0 M HCl to
In this paper, we present the results of a study of
poly(aniline) Ðlms grown with poly(acrylic acid) as electrodes
DOI: 10.1039/b001107j
Phys. Chem. Chem. Phys., 2000, 2, 2599È2606
This journal is ( The Owner Societies 2000
2599

View Online
emeraldineÈpernigraniline redox transformation. We also note
prepare a 15 wt.% solution of poly(acrylic acid) in 2.7 M HCl.
Solutions of pH 1 and 2 were prepared by mixing solutions of
0.1 M HCl and 0.1 M KCl. For pH from 3 to 7.5, bu†er
solutions were prepared by mixing solutions of 0.1 M citric
acid and 0.2 M disodium hydrogen orthophosphate dihy-
drate.11 All pH values were measured with a pH meter (Model
145 pH ion selective electrode probe, Corning) before use.
UV-Vis spectrophotometry (Hewlett-Packard 8452A UV/Vis
instrument) was used to determine the concentration of all
NADH solutions (e \ 6220 l mol~1 cm~1 at 340 nm12).
All electrochemical measurements were made with an
EG&G Model 263A potentiostat in a conventional three elec-
trode system. For rotating disc studies, an Oxford Electrodes
rotator and electrode were used. The working electrode was a
glassy carbon rotating disc electrode (0.38 cm2) used with a
platinum foil counter electrode and a laboratory-made satu-
rated calomel electrode (SCE) as reference electrode. All the
potentials are reported with respect to SCE. Prior to use, the
working electrode was polished with aqueous slurries of
alumina (1.0 lm, then 0.3 lm).
that, for Ðlms deposited by initially cycling to ]0.9 V fol-
lowed by subsequent cycles to ]0.78 V (see Experimental),
the voltammetry between the two redox peaks is featureless
and shows no signs of redox processes associated with over-
oxidation of the Ðlm.17h19 The cyclic voltammetry of the
leucoemeraldineÈemeraldine system around ]0.10 V is very
stable and the anodic peak current varies linearly with scan
rate, for scan rates up to at least 200 mV s~1 for a Ðlm grown
with a growth charge of 240 mC cm~2, as expected for an
immobilised redox system which is not di†usion limited.
The electrochemical behaviour of the poly(aniline)È
poly(acrylate) composite Ðlms was studied as a function of pH
(Fig. 1 and 2). The two redox systems, clearly distinct at pH 1,
shift with increasing pH and eventually merge in one broad
voltammetric wave at pH [ 5. Accompanying the shift in the
positions of the redox peaks, the total charge for the redox
process decreases as the pH increases although the anodic
peak currents remain proportional to the sweep rate.
Electropolymerisation of aniline was performed at room
temperature from a solution containing 0.5 M aniline, 15
wt.% poly(acrylic acid) and 2.7 M HCl. The electrode poten-
tial was swept from [0.2 to 0.9 V for the Ðrst cycle and then
from [0.2 to ]0.78 V for all subsequent cycles, at 50 mV s~1.
The deposition charge was recorded during the electro-
polymerisation process and when the desired charge was
reached (typically 240 mC cm~2 after about 10 min) polymer
deposition was stopped and the electrode rinsed with water
and 1.0 M HCl.
The amperometric response to NADH was studied in a
water jacketed thermostated cell (25.0 ^ 0.1 ¡C), in deoxy-
genated solution under argon or nitrogen at a Ðxed potential.
Prior to each experiment, the poly(aniline) composite Ðlm
coated electrode was preconditioned at ]0.5 V in 1.0 M HCl
for 5 min. The electrode was rinsed with water and pH 7
bu†er solution and then immersed in the electrochemical cell
containing deoxygenated pH 7 bu†er solution. The electrode
was rotated and the potential held at the chosen potential
while aliquots of a stock standard solution of NADH (0.01 M
in bu†er solution) were added to the cell. The resulting steady
state catalytic currents were recorded.
For UV-Vis characterisation, poly(aniline)Èpoly(acrylate)
Ðlms were deposited on an ITO-glass electrode, using the
same procedure as described above. The Ðlms were dried at
40 ¡C and the UV-Vis spectra recorded. For in situ resistance
measurements, a poly(aniline)Èpoly(acrylate) Ðlm was depos-
ited across the gap between two screen printed carbon micro-
band electrodes, as described elsewhere.7
Fig. 1 Cyclic voltammetry of a poly(aniline)Èpoly(acrylate) modiÐed
glassy carbon electrode in bu†er solutions at di†erent pH. The solu-
tions were as follows: pH 1 and 2, mixtures of 0.1 M HCl and 0.1 M
KCl; pH [ 3, 0.1 M citrateÈphosphate bu†er. All voltammograms
were recorded at 50 mV s~1 for a Ðlm deposited with a growth charge
of 90 mC on an electrode of area 0.38 cm2.
Results and discussion
Characterisation of poly(aniline)–poly(acrylate) composite
Ðlms
Poly(aniline)Èpoly(acrylate) composite Ðlms have been pre-
pared in the past by a number of di†erent methods:
electrochemical13 and chemical polymerization from aqueous
solution14,15 and self-assembly of the pre-formed polymer
from non-aqueous solution.16 In the present work, we used
electrochemical polymerisation to deposit the poly(aniline)È
poly(acrylate) composite Ðlms on the electrode surface. The
resulting Ðlms were characterised by electrochemistry and
UV-Vis spectrometry.
In 1.0 M HCl, the electrochemical behaviour of the Ðlm
(not shown) was almost identical with the behaviour of an
HCl doped poly(aniline) Ðlm. Two redox systems are
observed: the Ðrst, centred around ]0.10 V, corresponding to
the leucoemeraldineÈemeraldine redox transformation, and
the second, centred around ]0.78 V, corresponding to the
Fig. 2 Cyclic voltammetry of a poly(aniline)Èpoly(acrylate) modiÐed
glassy carbon electrode in pH 5È8, 0.1 M citrateÈphosphate bu†er
solutions recorded at 50 mV s~1, the Ðlm was grown with a deposi-
tion charge of 90 mC on an electrode of area 0.38 cm2.
2600
Phys. Chem. Chem. Phys., 2000, 2, 2599È2606

View Online
Our studies of NADH oxidation were performed in 0.1 M
citrateÈphosphate bu†er at pH 7. Under these conditions, the
poly(aniline) composite Ðlm shows a single, broad, voltam-
metric wave (Fig. 2). Although the total charge passed in the
redox process is much less than at pH 1 for the same Ðlm (the
charge recorded at pH 7 represents only 9% of the charge
recorded at pH 1), the Ðlm clearly remains electroactive. The
voltammetry of the Ðlm at pH 7 (Fig. 3) is signiÐcantly more
stable towards repeated cycling than that for poly(aniline)È
poly(vinylsulfonate) Ðlms.7 It was noticed that, after contact
with a pH 7 bu†er solution, a poly(aniline) Ðlm, immersed in
1.0
M HCl, is only partially reprotonated. For thin
poly(aniline)Èpoly(acrylate) Ðlms (deposition charge 240 mC
cm~2) about 9% of the charge, measured in 1.0 M HCl, is lost
following the Ðrst cycle from [0.2 to ]0.4 V at pH 7, and
thereafter the charge stabilises with about a 1% further
decrease over the following cycles at pH 7. This initial loss of
activity probably corresponds to the dissolution of some of
the polymer in solution at pH 7. No loss of charge was
observed for thicker Ðlms (deposition charge 470 mC cm~2)
after the Ðrst cycle, although some loss was observed after
several cycles at pH 7.
The conductivity of the poly(aniline)Èpoly(acrylate) com-
posite Ðlms was studied using a dual microband band elec-
trode by electrochemically depositing the polymer across the
20 lm gap between the two electrodes and measuring the
resistance across the Ðlm as a function of the potential of
the Ðlm (see ref. 7 for more details). The plot of log R against
the Ðlm potential (Fig. 4) shows that the minimum resistance
for the poly(aniline)Èpoly(acrylate) composite Ðlm in pH 7.0
solution occurs at [30 mV. This value is slightly anodic of
the mid-peak potential for the voltammetry of the Ðlm and
similar to the corresponding values for poly(aniline)È
Fig. 4 Logarithm of the resistance of a poly(aniline)Èpoly(acrylate)
composite Ðlm deposited across the gap between two screen printed
carbon microband electrodes, measured in 0.1 M citrateÈphosphate
bu†er, pH 7.
800 nm were no longer present. Chemically prepared
poly(aniline)Èpoly(acrylate) composites have been shown to
give similar UV-Vis spectra.15
The UV-Vis spectra of a poly(aniline)Èpoly(acrylate) Ðlm
(deposition charge D3 mC cm~2) deposited on an ITO-glass
electrode are shown in Fig. 5. Similar spectra were obtained
for dry Ðlms and for Ðlms in solution. For the acid treated Ðlm
we see three bands at 340, 440 and 750 nm, as expected. On
treating the Ðlm at pH 7, the bands at 340 and 440 nm do not
change signiÐcantly but there is a blue shift in the long wave-
length band at D700 nm which causes the Ðlm to change in
appearance from green to blue. From these results we con-
clude that much of the poly(aniline)Èpoly(acrylate) composite
in the emeraldine state remains protonated at pH 7. This is
consistent with our resistance measurements. Since the
poly(aniline)Èpoly(acrylate) composite Ðlm remains electro-
active and conducting at pH 7, it can be used as a mediator in
NADH oxidation.
poly(vinylsulfonate)
poly(styrenesulfonate) ([25 mV) Ðlms.6,7
([50
mV)
and
poly(aniline)È
The poly(aniline)Èpoly(acrylate) composite Ðlm was studied
by UV-Vis spectroscopy. According to the literature,20 the
protonated emeraldine form of poly(aniline) has an adsorption
peak at 340 nm, corresponding to the pÈp* transition of the
benzene rings, a broad adsorption at 440 nm corresponding to
the radical cation (semiquinone structure) and the very broad
band at 750 nm due to imine moieties (quinone structure). The
same authors found that when poly(aniline) deposited from
HCl was placed in 1 M Na SO , so that it was deprotonated
2
4
and consequently not conducting, the absorptions at 440 and
Fig. 3 Cyclic voltammetry of a poly(aniline)Èpoly(acrylate) modiÐed
glassy carbon electrode in 0.1 M citrateÈphosphate bu†er at pH 7.0.
The Ðrst 20 cycles are shown, recorded at a scan rate of 20 mV s~1.
The Ðlm was grown with a deposition charge of 150 mC on an elec-
trode of area 0.38 cm2.
Fig. 5 UV-Vis spectra of a poly(aniline)Èpoly(acrylate) Ðlm depos-
ited on an ITO-glass electrode. Solid line, poly(aniline)Èpoly(acrylate)
Ðlm held at ]0.5 V vs. SCE in 0.1 M HCl, then dried at 40 ¡C; dotted
line, poly(aniline)Èpoly(acrylate) immersed in 0.1 M citrateÈphosphate
bu†er, pH 7.
Phys. Chem. Chem. Phys., 2000, 2, 2599È2606
2601

View Online
Kinetic model for NADH oxidation
constant for adsorption and k the rate constant for oxida-
cat
tion of the adsorbed NADH; site represents a catalytic site
within the Ðlm; and K is the inhibitor constant for reversible
inhibition by NAD`. According to this mechanism, the
NADH oxidation at poly(aniline)Èpoly(acrylate) composite
coated electrodes was studied under the same conditions as
used in the studies of poly(aniline)Èpoly(vinylsulfonate) and
poly(aniline)Èpoly(styrenesulfonate) coated electrodes in order
to determine whether the same kinetic model could be used to
describe the reaction of NADH at these Ðlms and in order to
facilitate a direct comparison of the behaviour of the di†erent
Ðlms.
i
current for NADH oxidation is given by7
i \ nFAK D
y/L
(1)
M
S, film
with
C
e1@2a
(1 ] a)M2[a [ ln(1 ] a)]N1@2
D
According to our kinetic model the reaction of NADH at
the electrode can be described in terms of the following
mechanism:
y \ M2e[a [ ln(1 ] a)]N1@2 tanh
(2)
k{D,S
NADH
bulk
A8B NADH
L2[site]k
0
k{D,S
cat
M
e \
(3)
D
K
solution mass transport
S, film
and
KS
NADH A8B NADH
0
film
a \ K [NADH ]/K
(4)
S
0
M
partition into the Ðlm
where L is the Ðlm thickness and [site] is the concentration of
active catalytic sites within the Ðlm. Assuming that the Ðlm
thickness is linearly related to the charge passed during Ðlm
KM
NADH ] MsiteN A8B MNADHN
film
adsorption at a site within the Ðlm
growth, Q
,
growth
L \ Q
p
(5)
kcat
growth growth
MNADHN ÈÈÈ
Õ
MNAD`N ] H` ] 2e~
where p
is a constant.
oxidation of bound NADH
growth
In these equations, [NADH] refers to the concentration of
the NADH at the outside of the polymer Ðlm. For the rotat-
ing disc electrode, this concentration is related to the bulk
concentration by
0
Ki
MNAD`N A8B NAD` ] MsiteN
film
desorption of product
[NADH ] \ [NADH ] [ i/nFAk@
bulk D, S
(6)
0
KP
NAD` A8B NAD`
film
0
The e†ects of NAD` inhibition may be taken into account7
partition into the Ðlm
by replacing K and k by
M
cat
k{D,P
k@ \ k /(1 [ K /K )
(7)
(8)
(9)
cat
cat
M
i
NAD` A8B NAD`
0
bulk
k{D,P
and
solution mass transport
G
1 ] (K [NAD`] ] K [NADH ])/K
H
where the subscripts bulk, 0 and Ðlm refer to the bulk solu-
tion, the interface between the polymer and the solution and
the inside of the Ðlm, respectively; k@
coefficient for the species X at the rotating disc electrode
p
0
S
0
i
K@ \ K
M
M
(1 [ K /K )
M
i
is the mass transport
D, X
respectively, where
(k@ \ 1.554D2@3l~1@6W 1@2, where D is the di†usion coeffi-
[NAD `] \ [NAD `] ] i/nFAk@
X
X
X
0
bulk D, P
cient, l the kinematic viscosity and W the rotation speed in
hertz21); K and K are the partition coefficients for NADH
The di†erent limiting cases and the corresponding expres-
sions for the current for NADH oxidation arising from eqn.
S
P
and NAD`, respectively, into the Ðlm, K is the equilibrium
M
Fig. 6 Case diagram derived from eqn. (1) for the kinetic model. The di†erent equations represent the di†erent limiting expressions for the
current in each case and across the boundaries between each pair of cases.
2602
Phys. Chem. Chem. Phys., 2000, 2, 2599È2606

View Online
(1) are summarised in the case diagram given in Fig. 6. In the
following analysis we shall use these expressions to Ðt the
amperometric response of our Ðlms to NADH. To do this, we
Ðrst identify the appropriate initial and Ðnal cases for a given
set of experimental data by considering the dependence of the
current on the experimental variables of NADH concentra-
tion, [NADH] , and Ðlm thickness, L . This allows us to select
0
the appropriate expression for the current from Fig. 6 since
each of the four cases shows a unique set of dependences on
these experimental variables. The experimental data are then
Ðtted to the appropriate expression using a commercial non-
linear least squares Ðtting routine (SigmaPlot, Jandel
ScientiÐc).
Oxidation of NADH
Poly(acrylic acid) is available commercially with di†erent
molecular weights, so samples with three di†erent molecular
weights, 1200, 8000 and 15 000 Da, were used for comparison
(Fig. 7). The three Ðlms were grown under identical conditions
and the charge passed in deposition and in cycling in each
case was the same. The catalytic currents observed for the
composite Ðlms grown with the lowest molecular weight
poly(acrylic acid) sample (1200 Da) were signiÐcantly smaller
than those found for the other two Ðlms. For the Ðlms grown
with the 8000 and 15 000 Da poly(acrylic acid) samples, the
catalytic currents for NADH oxidation are almost identical.
In all of the subsequent studies we chose to work with the
8000 Da poly(acrylic acid) sample. In addition, we used thin
Ðlms (deposition charge 240 mC cm~2) for the studies of
potential dependence and the e†ects of rotation speed in order
to reduce the time taken to attain the steady state response
[the response time for a thin Ðlm is about 50 s; for a thick Ðlm
(deposition charge 395 mC cm~2), it increases to more than 10
min].
Fig. 8 Currents measured at ]0.05 V for the oxidation of NADH at
poly(aniline)Èpoly(acrylate) modiÐed glassy carbon electrodes
(deposition charge 90 mC, geometric area 0.38 cm2) plotted as a func-
tion of NADH concentration recorded at a rotation speed of 9 Hz, in
0.1 M citrateÈphosphate bu†er, pH 7, under argon. Results for four
replicate Ðlms prepared using 8000 Da poly(acrylate) under nominally
identical conditions are shown.
of the Ðlms where 9% of the charge for cycling in acid was lost
on cycling from [0.2 to ]0.4 V, at pH 7.
We now turn to the inÑuence of the potential on the cata-
lytic current for NADH oxidation. Results obtained for six
potentials using six di†erent Ðlms deposited under the same
conditions (deposition charge 240 mC cm~2) are shown in
Fig. 9. During these experiments it was noticed that the fastest
responses were obtained at O]50 mV and that at more
anodic potentials the response times become very long (more
than 20 min). These observations are consistent with the
resistance measurement since at higher potentials the Ðlms
become less conducting and consequently charge propagation
through the Ðlm is slow. The points in Fig. 9 are the experi-
The results for NADH oxidation at replicate poly(aniline)È
poly(acrylate) Ðlms show reasonably good reproducibility
(Fig. 8), although there is some scatter as the concentration of
NADH increases above 0.5 mM. In addition, when the same
Ðlm was used for more than one experiment, a drop in the
catalytic currents was observed. This decrease was consistent
with the behaviour described above for the cyclic voltammetry
Fig. 7 Currents measured at ]0.05 V for the oxidation of NADH at
poly(aniline)Èpoly(acrylate) modiÐed glassy carbon electrodes
(deposition charge 90 mC, geometric area 0.38 cm2) plotted as a func-
tion of NADH concentration recorded at a rotation speed of 9 Hz, in
0.1 M citrateÈphosphate bu†er, pH 7, under argon. Results for Ðlms
prepared with samples of three di†erent molecular weights of
poly(acrylate) are shown: =, 1200; …, 8000; and >, 15 000 Da. In
each case the Ðlms were prepared under identical conditions.
Fig. 9 Currents for the oxidation of NADH recorded at six poten-
tials at a poly(aniline)Èpoly(acrylate) modiÐed glassy carbon electrode
(deposition charge 90 mC, geometric area 0.38 cm2) plotted as a func-
tion of NADH concentration recorded at a rotation speed of 9 Hz, in
0.1 M citrateÈphosphate bu†er, pH 7, under argon: L, [30; +, 0;
…, 25; >, 50; =, 75; and K, 100 mV. The solid lines represent the
best Ðts of the experimental data to the expression for the case IÈIII
boundary, the resulting kinetic parameters are given in Table 1.
Phys. Chem. Chem. Phys., 2000, 2, 2599È2606
2603

View Online
Table 1 Best Ðt parameters from the analysis of the currents for
NADH oxidation at poly(aniline)Èpoly(acrylate) modiÐed glassy
carbon electrode at di†erent potentials
k
[site]p
cat
E vs. SCE/V
mol cm~2 C~1 s~1
K /K /mM
M
s
[0.030
0.000
]0.025
]0.050
]0.075
]0.100
(0.86 ^ 0.04) ] 10~9
(1.56 ^ 0.03) ] 10~9
(2.06 ^ 0.02) ] 10~9
(3.69 ^ 0.07) ] 10~9
(3.88 ^ 0.07) ] 10~9
(4.24 ^ 0.05) ] 10~9
1.055 ^ 0.0708
0.558 ^ 0.195
0.635 ^ 0.138
0.775 ^ 0.025
0.685 ^ 0.021
0.569 ^ 0.012
The results are for the data in Fig. 9 Ðtted to the expression for the
case IÈIII boundary.
mental data and the lines represent the best Ðts to the kinetic
model for the IÈIII boundary. The values of the rate constants
obtained from the Ðts are given in Table 1. As the potential
becomes more anodic, the parameter k [site]p
increases
cat
acid
and there is a corresponding slight decrease in K /K .
M
s
For all subsequent experiments NADH oxidation was
studied at ]0.05 V. The e†ect of Ðlm thickness is shown in
Fig. 10 for eight Ðlm thicknesses and six NADH concentra-
tions. In plotting the data in Fig. 10 we have assumed that the
Ðlm thickness is proportional to the charge passed to deposit
the Ðlm [see eqn. (5)]. Based on scanning electron microscopy
studies of these Ðlms, we estimate that for our electrodes a
growth charge of 90 mC corresponds to a thickness of about 5
lm. From Fig. 10 we can see that the catalytic currents
increase with increasing Ðlm thickness and NADH concentra-
tion, conÐrming that NADH oxidation occurs throughout the
Ðlm. The data in Fig. 10 were Ðtted to the expression for the
current across the IÈII case boundary (thin Ðlm to thick Ðlm,
under unsaturated conditions, Fig. 6). In all cases satisfactory
Ðts were obtained and the corresponding kinetic parameters
obtained from the Ðts are given in Table 2.
Next we turn to the inÑuence of the rotation speed on the
catalytic current, Fig. 11. In this case we Ðnd no signiÐcant
e†ect on the currents on increasing the rotation speed from 2
to 25 Hz, indicating that mass transport of NADH or NAD`
in the bulk solution is not rate limiting, even at low NADH
concentrations. The best Ðt parameters obtained from Ðtting
the data in Fig. 11 to the expression for the case IÈIII bound-
ary (Fig. 6) are given in Table 3.
Fig. 10 Currents for the oxidation of NADH at poly(aniline)È
poly(acrylate) modiÐed glassy carbon electrodes, geometric area 0.38
cm2, coated with Ðlms of di†erent thickness plotted as a function of
the deposition charge. Results for eight NADH concentrations are
shown, recorded at ]0.05 V at a rotation speed of 9 Hz, in 0.1 M
citrateÈphosphate bu†er, pH 7, under argon. NADH concentration:
L, 0.05; È, 0.1; K, 0.15; ), 0.2; …, 0.3; @, 0.45; =, 0.5; and + 0.6
mM. The solid lines represent the best Ðts of the experimental data to
the expression for the case IÈII boundary; the resulting kinetic param-
eters are given in Table 2.
on poly(aniline)Èpoly(vinylsulfonate) and poly(aniline)È
poly(styrenesulfonate) composites.6,7 In all three cases the
mechanism of the reaction is the same, with the NADH react-
ing at sites throughout the polymer Ðlm and with clear evi-
dence for reversible inhibition by the product, NAD`. The
oxidation process occurs around 50 to 100 mV vs. SCE in the
region where the poly(aniline) composite is in its conducting
state. The currents for oxidation of NADH at thin Ðlms of
poly(aniline)Èpoly(acrylate) are about one third of those at
poly(aniline)Èpoly(vinylsulfonate) Ðlms and about the same as
those for poly(aniline)Èpoly(styrenesulfonate) Ðlms under the
same conditions. In addition, the currents at the poly(aniline)È
Finally, the e†ects of NAD` inhibition were investigated by
adding NAD` to the bulk solution. Fig. 12 shows the cata-
lytic current as a function of NADH concentration both with
and without added NAD`. It is clear that the NAD` inhibits
the reaction. From the data in Fig. 12, we estimate the value
of K /K as 8.2 mM.
Table 3 Best Ðt parameters from the analysis of the currents for
NADH oxidation at a poly(aniline)Èpoly(acrylate) modiÐed glassy
carbon electrode, at di†erent rotation speeds
k
[site]p/
cat
i
P
W /Hz
2È25
mol s~1 C~1 cm~2
K /K /mM
M
S
Comparison results for di†erent Ðlms
(3.68 ^ 0.05) ] 10 ] ~9
0.76 ^ 0.02
It is instructive to compare the results obtained here for
poly(aniline)Èpoly(acrylate) composite Ðlms with earlier work
The results are for the data in Fig. 11 Ðtted to the IÈIII boundary.
Table 2 Best Ðt parameters from the analysis of the currents for NADH oxidation at a poly(aniline)Èpoly(acrylate) modiÐed glassy carbon
electrode for di†erent Ðlm thicknesses
[NADH] /mM
(K2D [site]k [site]/K )1@2/cm s~1
cat
(k [site]p2/K D )1@2/C~1
0
S
s
M
cat
M s
0.05
0.10
0.15
0.20
0.30
0.45
0.50
0.60
(9.9 ^ 0.3) ] 10~4
(9.8 ^ 0.3) ] 10~4
(9.2 ^ 0.4) ] 10~4
(8.9 ^ 0.4) ] 10~4
(8.0 ^ 0.3) ] 10~4
(7.2 ^ 0.3) ] 10~4
(6.8 ^ 0.2) ] 10~4
(6.7 ^ 0.3) ] 10~4
4.9 ^ 0.4
4.4 ^ 0.3
4.2 ^ 0.4
4.0 ^ 0.4
4.2 ^ 0.4
3.9 ^ 0.3
4.1 ^ 0.3
3.7 ^ 0.3
The results are for the data in Fig. 10 Ðtted to the IÈII boundary.
2604
Phys. Chem. Chem. Phys., 2000, 2, 2599È2606

View Online
polymers at pH 7. Poly(acrylic acid) is a weaker acid than
poly(vinylsulfonic acid) and although acetic acid has been
shown to dope poly(aniline) e†ectively,22,23 because the pK
a
of poly(acrylic acid) (D6.0)24 is higher than that of acetic acid
(4.75), it is likely that doping with poly(acrylic acid) is less
e†ective. Consequently, we believe that the concentration
of active catalytic sites within the Ðlm is smaller when
using poly(aniline)Èpoly(acrylate) than in poly(aniline)È
poly(vinylsulfonate) composite Ðlms. This is consistent with
the di†erences in the resistance and voltammetry of the two
composite polymer Ðlms at pH 7.
Despite being slightly poorer than poly(aniline)È
poly(vinylsulfonate) Ðlms for the electrocatalysis of NADH
oxidation, the poly(aniline)Èpoly(acrylate) composite Ðlms do
have some advantages, notably that the carboxylate groups
can be used as sites for the attachment of enzymes or deriv-
atives of NADH within the Ðlms in order to construct self-
contained systems for use in biosensors or biofuel cells. Also,
for this type of application the reduced sensitivity towards
product inhibition may prove to be a distinct advantage.
Fig. 11 Currents for the oxidation of NADH at poly(aniline)È
poly(acrylate) modiÐed glassy carbon electrodes (deposition charge 90
mC, geometric area 0.38 cm2) plotted as a function of NADH concen-
tration, recorded at ]0.05 V in 0.1 M citrateÈphosphate bu†er, pH 7,
under argon. Results for Ðve rotation speeds are shown: …, 2; =, 4;
>, 9; @, 16; and +, 25 Hz.
Conclusion
Poly(aniline)Èpoly(acrylate) Ðlms are electrocatalysts for
NADH oxidation, at 0.05 V, in bu†er solution at pH 7. Our
study shows that the reaction occurs at sites throughout
the polymer and that it is reversibly inhibited by NAD`.
This is the same mechanism as found for the reaction
at poly(aniline)Èpoly(vinylsulfonate) and poly(aniline)È
poly(styrenesulfonate) Ðlms.
Although poly(aniline)Èpoly(acrylate) Ðlms are not as good
electrocatalysts as poly(aniline)Èpoly(vinylsulfonate) Ðlms for
NADH oxidation, they do have the advantage that the inhibi-
tion by NAD` is less and it should be possible to attach
enzymes or modiÐed cofactors to the polymer through linking
to the carboxylate groups. By growing poly(aniline) Ðlms with
mixtures of poly(vinylsulfonate) and poly(acrylate) counter
ions, it may be possible to combine the advantages of the two
systems.
poly(acrylate) Ðlms saturate at lower concentrations of NADH
than for either of the other two composite Ðlms, and are
less stable towards repeated measurements. Interestingly,
the poly(aniline)Èpoly(acrylate) composite Ðlms are less
easily inhibited by NAD` than the poly(aniline)È
poly(vinylsulfonate) Ðlms. These di†erences can be attributed
to two separate e†ects. First, di†erences in the morphology of
the Ðlms are likely to a†ect the partition and di†usion of
NADH within the Ðlm. Poly(aniline)Èpoly(vinylsulfonate) and
poly(aniline)Èpoly(acrylate) prepared electrochemically have
di†erent morphologies: poly(aniline)Èpoly(acrylate) is formed
of extended needles whereas poly(aniline)Èpoly(vinylsulfonate)
forms a Ðbrous network.13 Second, the di†erent polymeric
counter ions lead to di†erences in the electroactivity of the
Acknowledgement
This work was supported by the US Office of Naval Research.
References
1
2
3
4
M. Dixon and E. C. Webb, Enzymes, Longman, London, 1979,
3rd edition.
G. T. R. Palmore, H. Bertschy, S. H. Bergens and G. M. White-
sides, J. Electroanal. Chem., 1998, 443, 155.
I. Willner, G. Arad and E. Katz, Bioelectrochem. Bioenenerg.,
1998, 44, 209.
P. N. Bartlett, P. Tebbut and R. G. Whitaker, Prog. React. Kinet.,
1991, 16, 55.
L. Gorton, J. Chem. Soc., Faraday T rans. 1., 1986, 82, 1245.
P. N. Bartlett and E. N. K. Wallace, J. Electroanal. Chem., in the
press.
5
6
7
8
P. N. Bartlett, P. R. Birkin and E. N. K. Wallace, J. Chem. Soc.,
Faraday T rans, 1997, 93, 1951.
P. N. Bartlett and E. N. K. Wallace, in Comprehensive Chemical
Kinetics, ed. R. G. Compton and G. Hancock, Elsevier, Amster-
dam, 1999, vol. 37, p. 35.
9
F. Albericio, J. M. BoÐll, A. El-Faham and S. A. Kates, J. Org.
Chem., 1998, 63, 9678.
10 Y. Degani and A. Heller, J. Phys. Chem., 1987, 91, 1285.
11 R. M. C. Dawson, D. C. Elliott and K. M. Jones, Data for Bio-
chemical Research, Oxford Science Publications, Oxford, 1986.
12 B. W. Carlson and L. L. Miller, J. Am. Chem. Soc., 1985, 107, 479.
13 J. H. Hwang and S. C. Yang, Synth. Met., 1989, 29, E271.
14 H. Hu, J. M. Saniger and J. G. Banuelos, T hin Solid Films, 1999,
347, 241.
Fig. 12 Currents for the oxidation of NADH at poly(aniline)È
poly(acrylate) modiÐed glassy carbon electrodes (deposition charge 90
mC, geometric area 0.38 cm2) plotted as a function of NADH concen-
tration, recorded at ]0.05 V at a rotation speed of 9 Hz in 0.1 M
citrateÈphosphate bu†er, pH 7, under argon. Results recorded in the
absence (L) and in the presence (…) of 2.1 mM NAD` are shown.
15 S. A. Chen and H. T. Lee, Macromolecules, 1995, 28, 2858.
Phys. Chem. Chem. Phys., 2000, 2, 2599È2606
2605

View Online
16 D. Li, Y. Jiang, Z. Wu, X. Chen and Y. Li, Polymer, 1999, 40,
21 W. J. Albery, Electrode Kinetics, Oxford University Press,
7065.
Oxford, 1975.
17 J. Desilvestro and W. Scheifele, J. Mater. Chem., 1993, 3, 263.
22 F. Zuo, R. P. McCall, J. M. Ginder, M. G. Roc, J. M. Leng, A. J.
Epstein, G. E. Asturias, S. P. Ermer, A. Ray and A. G. MacDia-
mird, Synth. Met., 1989, 29, E445.
23 M. Angelopoulos, A. Ray, A. G. MacDiamird and A. J. Epstein,
Synth. Met., 1987, 21, 21.
24 H. F. Mark and N. G. Gaylord, Encyclopedia of Polymer Science
and T echnology, Wiley, New York, 1964, vol. 1, pp. 217È219.
18 E. M. Genies and C. Tsintavis, J. Electronal. Chem., 1985, 195,
109.
19 C. Mailhe and J. Desilvestro, J. Electroanal. Chem., 1989, 262,
289.
20 D. E. Stillwell and S. M. Park, J. Electrochem. Soc., 1989, 136,
427.
2606
Phys. Chem. Chem. Phys., 2000, 2, 2599È2606