Electrochemical studies of osmium-(pyrrole-methyl) pyridine-co-polymers deposited using the membrane template method

Article abstract of DOI:10.1016/j.electacta.2008.01.049

The electrochemical characterisation of films formed by co-polymerisation of [Os(2,2′-bipyridine)₂XCl] (Os-PMP) where X is 3-(pyrrole-1yl-methyl)pyridine with 3-methyl thiophene or 1,2-diaminobenzene was carried out. The use of the membrane tem

Full text of DOI:10.1016/j.electacta.2008.01.049

Available online at www.sciencedirect.com
Electrochimica Acta 53 (2008) 4550–4556
Electrochemical studies of osmium-(pyrrole-methyl) pyridine-co-polymers
deposited using the membrane template method
Susan Warren1, Rodica Doaga1, Timothy McCormac1, Eithne Dempsey∗,1
Centre for Research in Electroanalytical Technologies (CREATE), Department of Science, Institute of Technology Tallaght, Tallaght, Dublin 24, Ireland
Received 6 July 2007; received in revised form 3 October 2007; accepted 12 January 2008
Available online 31 January 2008
Abstract
The electrochemical characterisation of films formed by co-polymerisation of [Os(2,2ꢀ-bipyridine)2XCl] (Os–PMP) where X is 3-(pyrrole-1yl-
methyl)pyridine with 3-methyl thiophene or 1,2-diaminobenzene was carried out. The use of the membrane template method, which allowed
growth of tubules of redox co-polymer was employed in the case of 1,2-diaminobenzene. Cyclic voltammetry allowed formation of the most stable
3-methylthiophene/Os–PMP films while chronocoulometry was used to co-polymerise Os–PMP with 1,2-diaminobenzene, resulting in stable films
of co-polymer micro-tubules with thin-film behaviour up to 0.5 V s−1 (r2 = 0.9997). Surface analysis was carried out using SEM.
© 2008 Elsevier Ltd. All rights reserved.
Keywords: Osmium complex; Co-polymer; Membrane template method
1. Introduction
examined. The co-polymerisation of an osmium bipyridine com-
plex modified with a pyrrole unit connected via a long spacer
armin thepresenceof pyrrole-3-carboxylicacid andtheenzyme,
glucose oxidase, also resulted in a stable reagentless biosensor
[5].
The electrochemical homopolymerisation of the short chain
[Os(2,2ꢀ-bipyridine)2(3-(pyrrole-1yl-methyl)pyridine)Cl−]
(Os–PMP) has not been possible to date, to the best of our
knowledge, but may be co-polymerised with other monomers
such as pyrrole and N-methylpyrrole [6]. The ruthenium
analogue [Ru(2,2ꢀ-bipyridine)2(PMP)Cl]+ has been used to
generate electronically conducting co-polymer films of pyrrole
[7] and 3-methyl thiophene [8] and their ionic and electronic
conductivity examined.
In this work, co-polymer films of Os–PMP with poly-3-
methylthiophene and 1,2-diaminobenzene co-polymers were
formed, the latter grown using the membrane-templated
method. This procedure entails synthesis or deposition of the
desired material within the pores of nano/micropore mem-
branes resulting in polymeric nano/micro-tubules, the size of
which depends on the membrane pore size. The membrane-
templated method has been used for synthesis of polymers
such as 3-methylthiophene [9], poly-aniline and its derivatives
[10–13], poly-acetylene [14] diaminobenzene derivatives [15]
and polypyrrole [16–19]. The electronic conductivity of poly-
mers synthesised by the templated method was found to be
A polymer/co-polymer may play a highly varied role as
a membrane component of chemical and/or biological sensor
devices. It may serve as a catalytic layer, a redox mediator, or it
may provide for molecular recognition and/or pre-concentration
of analyte [1]. Possibly the most important advantage of poly-
mer coatings is that they provide a three-dimensional reaction
zone at the electrode surface, which generates an increase in the
flux of reactions that occur there, in turn augmenting sensitivity.
For many years the electron acceptor properties of osmium
complexes have resulted in their use as electron transfer
mediators for a number of flavoproteins. Osmium bipyridine
complexes have been attached to various conducting polymer
backbones [2,3] which in turn were cross-linked to enzymes,
forming electrically wired reagentless enzyme systems. The
incorporation of the biocatalyst together with a long chain
osmium complex-modified pyrrole derivative with a long flex-
ible spacer arm during the electropolymerisation process was
described by Habermuller et al. [4]. Co-polymerisation of this
new mediator-modified pyrrole monomer with pyrrole was
∗
Corresponding author. Tel.: +353 1 4042862; fax: +353 1 4042700.
E-mail address: eithne.dempsey@ittdublin.ie (E. Dempsey).
ISE members.
1
0013-4686/$ – see front matter © 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2008.01.049

S. Warren et al. / Electrochimica Acta 53 (2008) 4550–4556
4551
greater by an order of magnitude than that of the bulk sam-
ples (powders and thin films) of the same polymers [20]. When
polymers are synthesised within the pores of the track-etched
polycarbonate membranes, the polymer preferentially nucleates
and grows on the pore walls, resulting in polymeric tubules
at short polymerisation times [18]. Due to their finite small
size, high surface area, and porosity, polymer nanotubes have
wide applications ranging from chemical, biological to micro-
electronics, energetics, optics, and medical sensing [11]. The
thickness and length of the polymer nanostructures depends on
polymerisation/co-polymerisation parameters such as polymeri-
sation time, temperature, pH, dopant counterion, etc. [16].
In this research, the synthesis, characterisation and applica-
tion of Os–PMP, co-polymerised with 3-methyl thiophene or
1,2-diaminobenzene was evaluated electrochemically and the
use of polymeric tubules grown using the membrane template
method was employed as a possible method to enhance surface
area and electron transfer capabilities of the film.
with the electrolyte in which the film was to be examined prior
to analysis.
2.2.2. Co-polymerisation of [Os(bpy)2PMP(Cl)]PF6 and
1,2-diaminobenzene (DAB) using the membrane template
method
The co-polymerisation was carried out using chrono-
coulometry, holding the potential at 1.0 V for 500 s. A
polycarbonate membrane (PCM) was held on the GCE using
an O rubber ring and the electrode was then placed in a
solution of 3.5 mM [Os(bpy)2PMP(Cl)]PF6 and 5 mM 1,2-
diaminobenzene, made up in 0.1 M TBAP/CH3CN. Following
polymerisation, the template membrane was dissolved by soak-
ing the electrode for 10 min, under continuous stirring, in
dichloromethane, leaving the co-polymer nanostructures on
the surface of the electrode. The [Os(bpy)2PMP(Cl)]PF6–1,2-
diaminobenzene (Os–PMP–DAB) co-polymer was stabilised by
scanning for 10 cycles at a scan rate of 5 mV s−1
.
2. Materials and methods
2.3. Apparatus
2.1. Materials
Cyclic voltammetry, chronocoulometry and amperometric
experiments were conducted using a CH Instrument 750 poten-
tiostat. A single-compartment electrochemical cell was used
with a platinum counter electrode and Ag | AgCl | KCl (3 M)
reference for aqueous solutions. For non-aqueous solutions the
reference employed was a silver wire in contact with an ace-
tonitrile solution of AgNO3 (0.01 M) and 0.1 M of the same
supporting electrolyte as that employed in the cell (referred
to as Ag | Ag+ throughout the text). All films were prepared
using glassy carbon electrodes which were polished through
three grades of alumina and sonicated in deionised water prior
to use. NMR studies were carried out on a Joel 300 MHz
spectrometer, and UV studies were done using UV-160A (Shi-
madzu). SEM studies were performed using a Hitachi S-2400N
system.
Potassium hexachloroosmate-K2OsCl6 (Aldrich), 97% 3-
(pyrrol-1yl-methyl)pyridine/PMP (Aldrich), 99+% ethylene
glycol (Aldrich), 95+% ammonium hexafluorophosphate
NH4PF6 (Aldrich), [Os(bpy)2PMP(Cl)]PF6 was synthesised
as described below, tetrabutylammonium perchlorate-TBAP
(Sigma), 99+% 3-methylthiophene (Acros Organics), HPLC
Grade Acetonitrile-CH3CN (Aldrich) used for electrochemical
experiments was previously dried using magnesium sulphate
˚
and then 4 A molecular sieves, 8–12 mesh (Aldrich), HPLC
Grade dichloromethane (Labscan), deuterated acetonitrile-d2
99.5 atom%D, phosphate-bufferedsalinetablets-PBS(Aldrich),
sodium dihydrogen phosphate ACS (Merck), potassium
chloride-KCl (Fluka), dimethylformamide/DMF (Aldrich). 1,2-
Phenylenediamine (Sigma–Aldrich) was purified three times by
recrystallisation from water and polycarbonate track etch mem-
branes with pores of 1 ␮m PCM (polycarbonate membrane)
were obtained from Sterlitech Corporation.
2.4. Synthesis
Os(bpy)2Cl2 was prepared as described previously [21].
Preparation of Os(bpy)2PMP(Cl)]PF6 was adapted from syn-
thetic methods previously described by Warren et al. [22]
and Serroni et al. [23]. Os(bpy)2Cl2 (0.175 mmol/100 mg) and
3-(pyrrol-1yl-methyl) pyridine (0.185 mmol/29.23 mg) were
stirred in the dark in ethylene glycol (16 cm3) for 20 min to
ensure the solids had dissolved. The solution was then heated
to reflux temperature for 30 min. The solution was allowed to
cool to room temperature before addition of deionised water
(16 cm3) and filtration to remove any unreacted starting materi-
als. Excess NH4PF6 was added to the filtrate and this solution
was then placed at 4 ◦C for 30–60 min prior to filtration of the
product to ensure that complete precipitation had occurred. The
product was dried under vacuum and then recrystallised from
a 2:1 acetone/water mixture. The purified product was filtered,
washed with copious amounts of water and dried overnight in a
vacuum oven at 60 ◦C, 200 mbar pressure. Percentage yield of
product obtained was 93% (136.9 mg).
2.2. Procedures
2.2.1. Co-polymerisation of [Os(bpy)2PMP(Cl)]PF6 and
3-methylthiophene
The optimum ratio for co-polymerisation was found to
be 3.5 mM [Os(bpy)2PMP(Cl)]PF6 and 20 mM 3-methyl-
thiophene. The osmium complex was dissolved in 0.1 M
TBAP/acetonitrile, before adding the 3-methylthiophene and
all solutions were made up to a final volume of 3 cm3. Solu-
tions were degassed prior to co-polymerisation. Co-polymer
films formed using cyclic voltammetry at a glassy carbon elec-
trode, were cycled between −0.2 → +1.4 V vs. Ag | Ag+ using a
scan rate of 0.1 V s−1. Films formed by chronocoulometry were
formed by holding the potential at +1.4 V vs. Ag | Ag+. A fresh
solution was used for each film to minimise errors. After for-
mation of each film the modified electrode was rinsed carefully

4552
S. Warren et al. / Electrochimica Acta 53 (2008) 4550–4556
2.6. SEM
Following film preparation electrodes were mounted and sta-
bilised using parafilm. A strip of aluminum foil, with a hole
pierced in the centre to allow access to the sample surface, was
placed over the electrode to minimise charging effects of the
electron beam on the electrode surface. Alternatively sputtering
of a nanometer thick layer of gold was performed in order to
minimise or avoid effects of electron beam onto the electrode
surface and charging of the sample.
3. Results and discussion
3.1. Characterisation of [Os(bpy)2PMP(Cl)]+
Spectroscopic characterisation using UV–visible spec-
troscopy of [Os(bpy)2PMP(Cl)]PF6 (in acetonitrile) resulted
in λmax = 294 nm (ε = 8,185 M−1 cm−1). A cyclic voltammo-
gram of the [Os(bpy)2PMP(Cl)]PF6 complex (Fig. 1) shows
the reversible Os2+/3+ couple at +0.035 V vs. Ag | Ag+, an irre-
versible oxidation wave at +1.092 V vs. Ag | Ag+ attributed to
the pyrrole ligand, and two reversible peaks at −1.790 and
−2.088 V vs. Ag | Ag+ pertaining to the successive reductions
of the bipyridine ligands, bpy0/−1 and bpy−1/−2, respectively.
The third bipyridine reduction was expected but not observed as
it occurred outside the electrochemical window of our system.
Fig. 1. [Os(bpy)2PMP(Cl)]PF6 at GCE. in 0.1 M tetrabutylammonium perchlo-
rate/acetonitrile at scan rate of 0.1 V s−1, vs. Ag | Ag+.
1H NMR: (300 MHz, acetonitrile-d2) values given in ppm
relative to SiMe4 (see Fig. 1 inset).
3-(Pyrrole-1-yl-methyl) pyridine unit
• Ring A: δ, 6.45 (s, H2, H5), 6.04 (d, H3, H4).
• Ring B: δ, 4.93 (s).
• Ring C: δ, 8.47 (d, H2), 8.32 (t, H6), 7.47 (dd, H4), and 7.27
(m, H5).
3.1.1. Co-polymer film studies of [Os(bpy)2PMP(Cl)]PF6
and 3-methylthiophene
• Bipyridine ring D: δ, 8.45 (d, H2), 8.33 (t, H4), 7.85 (m, H5),
All films were grown as described above both by cyclic
voltammetry and chronocoulometry, and analysed in 0.1 M
TBAP/acetonitrile or 0.1 M LiClO4/H2O. Table 1 shows the
electrochemical data for films formed using chronocoulometry.
Films analysed in 0.1 M TBAP/MeCN showed a consistent E1/2
value of + 0.059 V vs. Ag | Ag+ for films with charges below
1 × 10−4 C. Films analysed in 0.1 M LiClO4/H2O showed a
higher E1/2 of +0.312 V vs. Ag | AgCl, and much larger ꢀEp and
FWHM values, indicating that it was more difficult to oxidise the
film in aqueous electrolytes. The increase in ꢀEp and FWHM
values with increasing film thickness could be related to the
increasing hydrophobicity of the film as its thickness increases,
resulting in an increasing resistive effect within the film towards
the hydrophilic electrolyte and solvent.
and 7.38 (m, H4).
• Bipyridine ring E: δ, 7.61 (t, H2), 7.27 (m, H5), 7.85 (m, H4),
and 6.93 (m, H3).
• Bipyridine ring F: δ, 7.61 (t, H2), 7.85 (m, H5), 7.85 (m, H4),
and 6.93 (m, H3).
• Bipyridine ring G: δ: 8.33 (t, H2), 7.85 (m, H4), 7.22 (m, H5),
and 6.93 (m, H3).
2.5. Electrochemical characterisation
Characterisation of starting materials and the [Os(bpy)2
PMP(Cl)]PF6 (Os–PMP) product was carried out in 0.1 M
tetrabutylammonium perchlorate (TBAP) in acetonitrile.
[Os(bpy)2PMP(Cl)]Cl was analysed in a 0.1 M KCl/PBS solu-
tion vs. Ag | AgCl | KCl (3 M). All solutions were degassed for
a minimum of 10 min prior to analysis.
Table 2 shows the electrochemical data for films formed by
cyclic voltammetry and analysed in organic and aqueous elec-
trolytes. Once again the films studied in aqueous electrolyte
Table 1
Electrochemical data for the Os2+/3+ osmium/3-methylthiophene films formed by chronocoulometry analysed in 0.1 M TBAP/acetonitrile or 0.1 M LiClO4/H2O
Charge passed (mC)
Film Q × 105 (C)
E1/2 (V)
ꢀEp (mV)
FWHMOX (mV)
FWHMRED (mV)
MeCN
H2O
MeCN
H2O
MeCN
H2O
MeCN
H2O
MeCN
H2O
2.5
5.0
10.0
15.0
0.987
1.276
1.470
2.491
0.317
0.429
0.608
0.059
0.059
0.059
0.070
0.310
0.314
0.313
27
27
32
33
75
87
86
122
120
130
136
180
204
220
−102
−104
−108
−100
−148
−146
−144
a
a
a
a
a
–
–
–
–
–
a
Films formed for greater than 10 mC showed no electrochemistry in aqueous electrolyte.

S. Warren et al. / Electrochimica Acta 53 (2008) 4550–4556
4553
resulted in larger ꢀEp and FWHM values in comparison to films
studied in organic media. All films were prepared using 0.1 M
TBAP/MeCN, and analysed in that same solvent system for
organic film studies and in 0.1 M LiClO4/H2O for aqueous film
studies. It would be assumed that the film would have improved
behaviour in a solvent and electrolyte system from which it
was formed, considering that both were employed in the poly-
merisation process and become an integral part of the resulting
polymer film [24]. ꢀEp and FWHM values were again observed
to increase with film thickness. The E1/2 values appeared to be
independent of film thickness for both electrolytes and it was
observed that the film was more difficult to oxidise in aqueous
electrolyte.
Films formed using both methods with similar charge values
were evaluated using cyclic voltammetry and it was noticeable
that the film formed by chronocoulometry had a much larger
background current (data not shown), which was accredited to
the conducting polymer backbone. This could be due to the
differing timescales of the two techniques employed. In chrono-
coulometry, the potential was stepped just once from −0.2 V vs.
Ag | Ag+ to the oxidation potential of the 3-methylthiophene,
+1.4 V vs. Ag | Ag+ and held there until the appropriate charge
has passed. The timescale of this experiment was in the region
of seconds, e.g. 15 mC film takes 25 s. In the case of cyclic
voltammetry the potential was continually swept from −0.2 to
1.4 V vs. Ag | Ag+, a process that took minutes. From this per-
spective it is possible that films formed by CV contained less
conductive polymer backbone as the timescale of the experi-
ment leads to degradation of the conductive properties of the
polymer. When cyclic voltammograms of films prepared using
each method were compared in aqueous electrolyte there was
a much smaller Faradaic current when compared to the films
studied in acetonitrile. This phenomenon may be attributed to
the known hydrophobicity of polythiophenes, which can result
in less efficient electrochemical behaviour in aqueous media [1].
Fig. 2 showscyclicvoltammogramsfortwofilms, onestudied
in organic electrolyte and the other studied in an aqueous elec-
trolyte. Firstly the film in aqueous media has a more positive
E1/2, indicating that this film is more difficult to oxidise/reduce
Fig. 2. Cyclic voltammograms for a film formed by CV and analysed in 0.1 M
TBAP/MeCN (—) and 0.1 M LiClO4/H2O (
) at υ = 0.010 V s−1
.

4554
S. Warren et al. / Electrochimica Acta 53 (2008) 4550–4556
in aqueous media. Secondly, although the charge for both
films is somewhat similar ∼2 × 10−5 C, the shapes and current
responses of the two CVs are quite different. The film studied in
organic media was more typical of a thin film with ꢀEp = 32 mV.
The aqueous film however, showed a larger ꢀEp = 72 mV and
the oxidation and reduction peaks appeared broader than those
of the organic film—suggesting slower charge transport proper-
ties. Plots of ꢀEp vs. scan rate (υ) for a thin (one cycle) and a
thick(sevencycles)film werecarriedoutin0.1 MTBAP/MeCN.
The thin film showed little dependence on scan rate but for thick
films ꢀEp had an exponential dependence on scan rate (data not
shown).
The influence of scan rate on current response was analysed
for films grown using CV. A general trend that occurred between
all films studied in organic electrolyte was that the slope for Ip,c
was slightly less than that obtained for Ip,a. This suggests that the
−
insertion of the ClO4 anion into the film occurred faster than
expulsion of the anion from the film. As the co-polymer film
was polymerised from 0.1 M TBAP/MeCN, this electrolyte and
solvent are an integral part of the co-polymer matrix, and thus
would have an affinity for that particular electrolyte. For films
studied in aqueous electrolytes, this trend was seen to reverse,
whereby the Ip,c slopes were found to be larger than those for
the Ip,a values. Once again this can be explained in terms of
anion affinity of the film, the larger Ip,c values indicating that
the anion expulsion process required less time to occur relative
to the anion inclusion step. In terms of the anion expulsion, as the
co-polymer film has a poor affinity for the aqueous anion it will
repelitwith moreease relativetothatof theanion which partially
√
consists of its makeup, resulting in a larger slope and
DCTCM
value (where DCT = charge transport diffusion co-efficient and
CM = concentration of redox sites) for Ip,c relative to the Ip,a
For films analysed in both organic and aqueous media, thin-
film behaviour was observed up to four formation cycles, up
to a scan rate of 0.3 V s−1. Fig. 3(a) and (b) is graphical rep-
resentations of the Randles–Sevcik plots for thick and thin
Fig. 3. (a) Randles–Sevcik plots for thin and thick films studied in 0.1 M
TBAP/MeCN. (b) Randles–Sevcik plots for thin and thick films studied in 0.1 M
LiClO4/H2O.
.
plate method. The co-polymerisation was performed in the
presence of the polycarbonate membrane (PCM) with pores
of θ ≈ 1 ␮m. Following dissolution of the membrane the
GCE/Os(bpy)2PMP(Cl)]PF6–DAB nanostructures system were
examined electrochemically and using SEM.
This technique allows formation of microrods of polymer
grown in the pores of a polycarbonate membrane. The co-
polymer films formed with and without the membranes were
analysed at different scan rates in the 0.1 M KCl/PBS pH 7.00.
The Os couple had a more positive potential in an aqueous
electrolyte, indicating that this film was more difficult to oxi-
dise/reduce in aqueous media (similar to Os–PMP/3-MT films).
This was the case for films prepared normally and for those
where the polycarbonate membrane (PCM) was employed to
generate redox co-polymer tubules.
films in organic and aqueous electrolytes, respectively. It was
√
observed that
D
CTCM values increased (from 2 × 10−9 to
15 × 10−9 mol cm−2 s−0.5) with increasing cycle number (one
to five cycles) as expected. Values are given for films formed up
to five cycles as in the case of greater loadings a large shift in
E1/2 and ꢀEp with respect to scan rate was observed. As film
thickness was unknown it was not possible to isolate individual
DCT and CM values.
Films grown using two cycles were analysed for 100 cycles at
a scan rate of 0.1 V s−1, and stability described as the % decrease
in anodic current. It was observed as expected that Os/3-MT co-
polymer films were more stable in the organic electrolyte (1.3%)
than the aqueous system (2.8%).
3.1.2. Co-polymerisation of [Os(bpy)2PMP(Cl)]PF6 and
1,2-diaminobenzene (DAB) using the membrane template
method
The electrochemical co-polymerisation of Os(bpy)2
PMP(Cl)]PF6 with 1,2-diaminobenzene was carried out
using chronocoulometry in 0.1 M TBAP/CH3CN, holding
the potential at 1.0 V for 500 s, using the membrane tem-
Table 3 shows a comparison of electrochemical data for
Os–PMP–DAB films grown normally (i) and using the mem-
brane templating method (ii). In terms of surface coverage (Γ )
(measured at 5 mV s−1), the co-polymer films showed similar
values2.5 × 10−10 mol cm−2 and2.9 × 10−10 mol, respectively.
The % decrease in current intensity (Ipa and Ipc % loss), were
9.7% and 5.9% for the film grown normally and 3.7% and 2.9%

S. Warren et al. / Electrochimica Acta 53 (2008) 4550–4556
4555
Table 3
Electrochemical characterisation for the osmium co-polymer Os–PMP–DAB films in 0.1 M KCl/PBS (pH 7.00) following formation by chronocoulometry in the
absence (i) of and in the presence of (ii) polycarbonate membrane
Scan rate (mV s−1
)
E1/2 (V)
ꢀEp (V)
FWHMOX (mV)
FWHMRED (mV)
(i)
(ii)
(i)
(ii)
(i)
(ii)
(i)
(ii)
10
50
100
500
0.346
0.343
0.342
0.343
0.338
0.335
0.335
0.335
0.052
0.050
0.048
0.063
0.044
0.041
0.038
0.052
192
194
196
222
178
176
178
184
166
170
172
174
162
166
166
178
for film grown in the presence of the polycarbonate membrane,
demonstrating an increase in stability for the nanostructured
film. ꢀE, E1/2 and full width at half-maximum both for oxi-
dation and reduction peaks (FWHM) were evaluated and films
formed in the presence of polycarbonate membrane had slightly
lower values for each parameter demonstrating that films were
thinner and easier to oxidise.
template membrane. The prevention of the aggregation process
may be possible by applying an ultrasonication step in the pres-
ence of a surfactant solution, but this phenomenon could damage
the larger nanotubes and would also remove the nanostructures
from the electrode resulting in formation of a suspension [16]. A
future approach could be based on the application of templated
method in order to synthesise polymers using the polycarbon-
ate membranes by firstly converting one side of the membrane
into a working electrode. This has been accomplished by elec-
troplating the membrane using Au or Pt solutions, resulting in
Au or Pt microelectrodes at the base of each of the pores [17]
Fig. 4 shows the dependence of current vs. scan rate
for the Os(bpy)2PMP(Cl)]PF6–DAB co-polymer analysed in
0.1 M TBAP/CH3CN, after the membrane was dissolved in
CH2Cl2, demonstrating thin-film behaviour with linearity up to
0.5 V s−1 (r2 = 0.9992and0.9997forip(c) andip(a), respectively).
SEM images of 1,2-DAB alone growth through the pores
of PCM (Fig. 5(A)) showed the rod like structures of dimen-
sions approximately 5–10 ␮m. The image shown in Fig. 5(B)
was obtained upon dissolution of the template membrane show-
ing the aggregation of co-polymer nanostructures into structures
with dimensions ranging from 20 ␮m to approximately 80 ␮m
spaced at regular intervals across the surface of the electrode.
Hulteen and Martin [25] reported that in developing template
synthetic methods two typical concerns need to be taken into
consideration, a solvophobic component (because the polycar-
bonate membranes are more hydrophobic than hydrophilic) to
the interaction between the co-polymer and the pore wall or/and
an electrostatic component if the polymer is cationic because
there are anionic sites on the pore walls of the polycarbonate
membranes. The fact that the individual nanostructures aggre-
gated forming clusters (Fig. 5(B)) may be due to the advent of
capillary forces between the nanotubes upon dissolution of the
Fig. 5. (A) SEM image of GCE/polydiaminobenzene tubule network, magnifi-
cation 3000×, accelerating voltage 10 kV, resolution 10 ␮m; (B) SEM image of
Os(bpy)2PMP(Cl)]PF6–diaminobenzene fibrilar co-polymer aggregate, magni-
fication 2000×, accelerating voltage of 10 kV, resolution 20 ␮m.
Fig. 4. Effect of scan rate, between 5 and 500 mV s−1, for GCE/Os(bpy)2
PMP(Cl)]PF6–DAB sensor in 0.1 M TBAP/CH3CN.

4556
S. Warren et al. / Electrochimica Acta 53 (2008) 4550–4556
which were then employed to electrochemically synthesise the
polymer tubules into each pore.
References
[1] G.G. Wallace, G.M. Spinks, L.A.P. Kane-Maguire, P.R. Teasdale, Con-
ductive Electroactive Polymers, Intelligent Materials Systems, CRC Press,
2003, p. 179.
4. Conclusions
[2] Y. Degani, A. Heller, J. Am. Chem. Soc. 111 (1989) 2357.
[3] A. Heller, Acc. Chem. Res. 23 (1990) 128.
[4] K. Habermuller, A. Ramanavicius, V. Laurinavicius, W. Schuhmann, Elec-
troanalysis 12 (2000) 1383.
[5] S. Reiter, K. Habermuller, W. Schuhmann, Sens. Actuators B 79 (2001)
150.
[6] W. Schuhmann, C. Kranz, J. Huber, H. Wohlschla¨ger, Synth. Met. 61 (1993)
31.
[7] J. Ochmanska, P.G. Pickup, J. Electroanal. Chem. 271 (1989) 83.
[8] J. Ochmanska, P.G. Pickup, J. Electroanal. Chem. 297 (1991) 197.
[9] M.A. Sato, S. Tanaka, K. Kaeriyama, Synth. Met. 14 (1986) 279.
[10] S. Mu, Electrochim. Acta 51 (2006) 3434.
[11] L.Y. Xin, X.G. Zhang, G.Q. Zhang, C.M. Shen, J. Appl. Polym. Sci. 96
(2005) 1539.
[12] R.V. Parthasarathy, C.R. Martin, Chem. Mater. 6 (1994) 1627.
[13] L.D. Pra, S.D. Champagne, Thin Solid Films 479 (2005) 321.
[14] W. Liang, C.R. Martin, J. Am. Chem. Soc. 112 (1990) 9666.
[15] F. Curulli, F. Valentini, S. Orlanducci, M.L. Terranova, C. Paoletti, G.
Palleschi, Sens. Actuators B 100 (2004) 65.
[16] S. De Vito, C.R. Martin, Chem. Mater. 10 (1998) 1738.
[17] L.S. Van Dyke, C.R. Martin, Langmuir 6 (1990) 1118.
[18] V.M. Cepak, J.C. Hulteen, G. Che, K.B. Jirage, B.B. Lakshmi, E.R. Fisher,
C.R. Martin, J. Mater. Res. 13 (1998) 3070.
[19] R.M. Penner, C.R. Martin, J. Electrochem. Soc. 133 (1986) 310.
[20] M. Steinhart, S. Senz, R.B. Wehrspohn, U. Go¨sele, J.H. Wendorff, Macro-
molecules 36 (2003) 3646.
[21] S. Warren, T. McCormac, E. Mitkova Mihaylova, E. Dempsey, Electro-
analysis 18 (2006) 1778.
[22] S. Warren, T. McCormac, E. Dempsey, Bioelectrochemistry 67 (2005) 23.
[23] S. Serroni, S. Campagna, G. Denti, T.E. Keyes, J.G. Vos, Inorg. Chem. 35
(1996) 4513.
[24] M. Omastova, S. Kosima, B. Shakalova, D. Jancula, Synth. Met. 53 (1993)
227.
Successful synthesis and characterisation of a pyrrole mod-
ified osmium complex, [Os(bpy)2PMP(Cl)]PF was performed.
To the best of our knowledge the research presented here is the
first report based on Os–PMP immobilisation by polymerisation
with 3-methylthiophene and 1,2-diaminobenzene. Further nov-
elty lies in the use of the membrane template method to form
redox tubules of Os–PMP and DAB on the electrode surface.
The complex was firstly co-polymerised with
3-methylthiophene, using cyclic voltammetry and chrono-
coulometry resulting in a stable redox active film. Secondly,
a co-polymer based on Os(bpy)2PMP(Cl)]PF6 and 1,2-
diaminobenzene, was formed using the membrane templating
method. Electrochemical evaluation was carried out and it was
found that the Os–PMP/DAB film had similar characteristics
to that of the Os-3MT co-polymer film, showing improved
characteristics in organic electrolytes (thin-film behaviour was
demonstrated up to 0.5 V s−1). Those Os–PMP/DAB films
formed using the membrane template method had improved
stability and were easier to oxidise relative to those examined
in the absence of the tubule structure on the surface. It was
expected that an enhanced surface area would be evident
with increased surface coverage, however, due to apparent
aggregation of the tubules into regularly spaced clusters on the
surface the coverage was not significantly improved. Future
work will examine further this process and will also extend
the method to include immobilisation of glucose oxidase with
subsequent evaluation of the mediating properties of the redox
system.
[25] J.C. Hulteen, C.R. Martin, J. Mater. Chem. 7 (1997) 1075.