View Article Online / Journal Homepage / Table of Contents for this issue

Analyst
Published on 31 August 2010. Downloaded by Universidade de Pernambuco on 19/02/2018 19:39:58.

Interdisciplinary detection science

www.rsc.org/analyst Volume 135 | Number 11 | November 2010 | Pages 2745–3012

ISSN 0003-2654

CRITICAL REVIEW
Gordon G. Wallace et al.
Printing conducting polymers 0003-2654(2010)135:11;1-H

CRITICAL REVIEW
Printing conducting polymers
B. Weng,a R. L. Shepherd,a K. Crowley,b A. J. Killardb and G. G. Wallace*a
Received 11th May 2010, Accepted 26th July 2010
DOI: 10.1039/c0an00302f
Recent developments in both materials science and printing technologies have led to a rapid expansion
Published on 31 August 2010. Downloaded by Universidade de Pernambuco on 19/02/2018 19:39:58.
in the field of printed conducting polymers. This review provides an overview of the most common
printing methods currently in use and the material requirements of each. Examples of printed devices
fabricated from a range of conducting polymers are given with an emphasis on the development of
sensors.
1. Introduction
Until about 30 years ago, all carbon based polymers were
rigidly regarded as insulators. In 1977, it was found that poly-
acetylene could be rendered conductive by doping.1,2 Since then,
a series of conductive polymers, including polyaniline (PANi),
polypyrrole (PPy) and polythiophenes such as poly(3,4-ethyl-
enedioxythiophene) (PEDOT), have been widely investigated
and applied in many fields, including sensors,3–5 actuators,6,7
battery electrodes,8,9 and platforms for tissue engineering.10–13
Their use as the active element in a range of sensing technol-
ogies can be attributed to their inherent characteristics wherein
the electronic and/or electrochemical properties of the conduct-
ing polymer can be rendered responsive to selected elements
of the environment. Sensitivity and selectivity can be tuned by
the use of an appropriate conducting polymer backbone, intro-
duction of appropriate molecular dopants, or, for a given
composition, application of an appropriate electrochemical
potential.14
Many applications of conducting polymers have been limited
by a lack of processability, since these materials are known to be
insoluble and infusible.15 Consequently, electrodeposition has
a
ARC Centre of Excellence for Electromaterials Science, Intelligent
Polymer Research Institute, University of Wollongong, Northfields
Avenue, Wollongong, NSW, 2522, Australia
b
Biomedical Diagnostics Institute, National Centre for Sensor Research,
Dublin City University, Dublin 9, Ireland
Bo Weng obtained her BSc
(2006) in Materials Science and
Engineering from Beihang
University, China, and is now
a PhD candidate at the Univer-
sity of Wollongong, Australia.
Bo Weng
This journal is ª The Royal Society of Chemistry 2010
View Article Online
www.rsc.org/analyst | Analyst
been the usual method of choice to produce conducting polymers
in the research laboratory. While this provides a convenient and
controllable method for coating conductive substrates, it is
limited in the ability to produce thin (<1 mm) uniform layers with
many polymers of interest and it is difficult to achieve control
over spatial distribution of conducting polymers on a range of
substrates – conductive or non-conductive, porous or non-
porous. In cases where conducting polymers can be rendered
soluble (usually by functionalising the polymer backbone) or
when stable nanodispersions can be formed, electro-spinning,16,17
wet spinning18,19 spin coating20,21 and printing22,23 have been used
for processing. Only printing provides a convenient route to the
deposition of conducting polymers with spatial resolution in the
x, y plane in the order of tens of microns and makes layer
thicknesses of the order of 100 nm feasible. Printing also provides
a rapid, low cost method which can potentially be used
commercially for mass production.24
Challenges in the development of printing technologies for
conductive polymers include the ability to produce soluble,
dispersible materials of sufficient conductivity and in the devel-
opment of ink formulations that will provide appropriate
delivery and curing characteristics to provide adherent, compact
features with appropriate resolution and conductivity. These
desirable characteristics are as much a function of the solvents
used as they are the conducting polymer material.
The primary role of the solvent is to provide compatibility
with the conducting polymer to be printed. Some degree of
Rod Shepherd received his
BEnvSc (Hons) and PhD from
the University of Wollongong,
Australia in 1997 and 2002
respectively. He is currently
a Research Fellow at the Intel-
ligent Polymer Research Insti-
tute, University of Wollongong,
Australia. His research interests
include chemical sensors, inher-
ently conducting polymers,
colorimetric optical sensors,
robotics, electrochromics and
Roderick Shepherd wireless sensor networking.
Analyst, 2010, 135, 2779–2789 | 2779
 View Article Online

solvent–polymer interaction is required to ensure that the ink
produced is stable for an adequate period of time. The solvent
used will also dictate the rheological and surface energy char-
acteristics of the ink produced. For example, the use of low
molecular weight solvents (e.g. water or low molecular weight
organic solvents) will generally provide low viscosity. If higher
viscosity is required polymer additives (thickeners) or higher
molecular weight solvents may be utilised. Some additives that
Published on 31 August 2010. Downloaded by Universidade de Pernambuco on 19/02/2018 19:39:58.
temperature substrate platforms can be used to accelerate the
process.
In recent years, growing efforts have been focused on the
fabrication of conducting polymer based devices by printing.
Here we review progress towards the use of printing to produce
sensors and associated electrochemical devices that utilise
conducting polymers including polyanilines, polypyrroles and
polythiophenes. Printing is a reproduction process in which ink

can influence viscosity and surface tension may also provide
beneficial effects through interaction with the conducting poly-
mer material, itself. For example, ethylene glycol (EG) can be
used to modify viscosity and it also influences the conductivity of
PEDOT:PSS significantly.79,80 The quality of the print will also
be affected by the choice of the solvent, along with the deposition
method and the substrate being used. To avoid issues such as
bleeding and spreading on the surface it is critical to control the
solvent interaction with the substrate and the rate of solvent
evaporation. The printed materials must be rendered a solid/
semi-solid, continuous conductive film after the removal of the
solvent. The use of organic systems provides a great deal of
control over drying temperatures and rates and often combina-
tions of solvents can be used to fine tune the system. There is less
control over the drying/evaporation process with aqueous-based
systems. However, forced air drying and/or the use of elevated
Dr. Karl Crowley is a research
fellow in Dublin City University
specialising in sensors and prin-
ted electronics. He received
a BSc in physics and chemistry
in 1998 and a PhD in analytical
and materials chemistry in 2003
from the Dublin Institute of
Technology. Previously he has
worked on wireless sensor
networks, smart packaging and
electrochemical and colorimetric
sensors. His current interests are
Karl Crowley in the development of new nano-
materials for gas sensors applied
to industrial and biomedical
analysis in addition to fabrication methodologies with inkjet, screen
and other printing technologies.
Tony Killard BA(Mod) Trinity
College Dublin (1993) and PhD
in Biotechnology at Dublin City
University (1998). Tony is
currently Principal Investigator
at the Biomedical Diagnostics
Institute and National Centre
for Sensor Research, Dublin
City University.
Tony Killard
is applied to a substrate in order to transmit information
(images, graphics, text) in a repeatable form.24 Printing tech-
nologies that require a printing plate, such as lithography
(offset), gravure, letterpress and screen printing, are known as
conventional methods of printing. Other approaches such as
inkjet printing or electrophotography fall into the category of
non-impact printing (NIP) technologies since a printing plate is
not required. Amongst the printing technologies available,
screen printing,25 microcontact printing,26–28 gravure, flexo-
graphic, offset29,30 and inkjet printing3,4,22,31 are the most
commonly used in the production and fabrication of conductive
films and structures with conducting polymer formulations.
These printed structures are being further integrated with bio-
logical materials and microfluidic systems to add even greater
functionality.32
2. Printing techniques
Screen printing
Screen printing is a process in which ink is forced through
a screen (Fig. 1). During the printing procedure, the ink is
imprinted or transferred onto the screen, and moved by the
squeegee blade to penetrate the screen through an image-specific
open mesh. The screen printing plate is a combination of screen
and stencil.24 Screen printable inks must have high viscosity when
not under sheer, but when forced through the mesh screen with
the squeegee blade undergo sheer thinning. This allows it to pass
through the stencil and to contact the substrate, whereupon it
Gordon Wallace obtained his
PhD from Deakin University,
Geelong, Australia, in 1983. He
joined the University of Wol-
longong in 1986 and is currently
Director of the Intelligent Poly-
mer Research Institute (http://
ipri.uow.edu.au/aboutipri/index.
html) and Executive Research
Director of the ARC Centre
of Excellence for Elec-
tromaterials Science (http://
www.electromaterials.edu.au/).
Gordon Wallace He is a Fellow of the Australian
Academy of Science and the
Australian Academy of Techno-
logical Sciences and Engineering. He has published over 500 papers
and numerous patents in the areas of organic conductors, nano-
materials, the development of intelligent polymer systems, and
their exploitation in medical bionics and energy production and
storage.

2780 | Analyst, 2010, 135, 2779–2789 This journal is ª The Royal Society of Chemistry 2010

Published on 31 August 2010. Downloaded by Universidade de Pernambuco on 19/02/2018 19:39:58.
Fig. 1 Screen printing (schematic diagram)24 (adapted from ref. 24).
returns to its viscous state and so forms a sharp image with well
defined boundaries.
Films obtained by screen printing typically have a thickness in
the range of ca. 20–100 mm. This is thicker than films produced
by other printing technologies (ca. 0.5–2 mm) and screen printing
is thus referred to as a thick film technology. The thickness of
screen-printed layers is determined by the thickness of the stencil
(the distance that the stencil stands above the screen) and the
thickness and dimensions of the mesh. The typical minimum line
width practically achievable is approximately 100 mm with
similar line spacing resolution.
Screen printing has been used as the basis of electrode strip
manufacture for at least 20 years. The choice of inks available for
screen printing is much larger than any of the other printing
technologies reviewed here. Inks containing metal conductors –
carbon, silver, gold, platinum – are readily available. The
complex rheological characteristics required of screen printing
inks need a combination of the conductor with appropriate
solvents and organic binders. Thus, a balance between the
conductive solids content and liquid solvent content wherein the
conductivity percolation threshold is exceeded must be obtained
while maintaining the rheological properties of the ink.
Following curing, the ink must transform from a semi-liquid to
a solid without shrinkage or cracking. As an example, Gwent
Electronic Materials (Gwent, UK) produce carbon-based pastes.
The typical solids content for such inks is from 40 to 45%
(graphite and binder) with typical viscosities in the range of
3–6 Pa$s at a sheer rate of 230 s1 and sheet resistance in the
range from 50 to 100 U ,1 with a film thicknesses of 12–24 mm.
Rotary printing
Printing methods such as gravure, flexographic and offset
printing are traditionally used for the roll-to-roll printing of
This journal is ª The Royal Society of Chemistry 2010
View Article Online
newspapers, magazines and packages. The minimum line widths
obtained using these methods are typically in a range of tens of
microns33,34 with film thicknesses of just a couple of microns. This
combination of good resolution, thin film deposition and high
throughputs of up to 100 m min1 make these extremely powerful
techniques for the manufacture of organic electronics. In terms
of feature size, they are exceeded by photolithography and laser
techniques. However, these have linear throughputs at least two
orders of magnitude lower than rotary methods. In addition, the
rotary techniques are suitable for wide area printing. However,
they are difficult to employ at a research and development scale
due to volume requirements and design constraints and so tend
to be used when production scale-up is required. A range of
functional organic and inorganic printing inks are becoming
commercially available.
The distinctive feature of gravure printing technology is the
fact that the image elements are carved into the surface of the
cylinder, while the non-image areas are at the original level.
Before printing the entire cylinder (printing and non-printing
elements) goes through the ink fountain to be inked and flooded
with ink. Ink is then removed from the non-image surface by
a wiper or doctor blade and hence ink remains only in the carved
cells before printing. When the ink makes contact with the
printing substrate, it is transferred from the carved cells to
substrate by the high printing pressure and adhesive forces
between ink and substrate. The principle of gravure printing is
illustrated in Fig. 2. As well as the high throughput and resolu-
tion achievable with gravure it can also transfer different
amounts of material to different areas of the substrate. The depth
and density of the cells determines the thickness of the printed
layers. A disadvantage, however, is that the cellular image
elements create a ‘dotted’ image which can sometimes be seen by
the naked eye.
Gravure printing requires high pressure (3 MPa) to transfer
ink from cells on the cylinder to the printing substrate. To ensure
that the ink can be loaded and transferred properly, the viscosity
is typically in the range of 0.05–0.2 Pa$s.
Flexographic or ‘flexo’ printing is the opposite of gravure in
which a compliant impression roller transfers the image from the
raised areas of the printing plate to the substrate (Fig. 3).
Resolution is similar to that of gravure at approx. 20 mm, with
Fig. 2 Gravure printing (schematic diagram)24 (adapted from ref. 24).
Analyst, 2010, 135, 2779–2789 | 2781

Published on 31 August 2010. Downloaded by Universidade de Pernambuco on 19/02/2018 19:39:58.
Fig. 3 Flexographic printing (schematic diagram)24 (adapted from ref.
24).
Fig. 4 Offset printing (schematic diagram)24 (adapted from ref. 24).
comparable print thicknesses obtained and ink viscosities
required. The applied pressures during transfer can affect the
edge quality of the deposited ink.
Fig. 5 Inkjet technology: (a) continuous inkjet printing; (b) ‘drop on demand’ inkjet printing24 (adapted from ref. 24).
2782 | Analyst, 2010, 135, 2779–2789
View Article Online
Offset printing (Fig. 4) is a type of flexographic printing with
two distinctive features. The first is that the image is not
transferred directly between the image roller to the substrate,
but is first transferred to an offset roller. Secondly, the image
roller contains a pattern of hydrophobic and hydrophilic areas
which are receptive and repulsive to a hydrophobic ink,
respectively. The hydrophobic ink is transferred to the printing
plate along with a hydrophilic water-based ink. These repel
one another and take up a pattern of the underlying substrate
before being transferred to the featureless offset cylinder and
then to the substrate. This requirement for hydrophobic/
hydrophilic materials places additional design criteria on the
inks, which must also have extremely high viscosities (up to
100 Pa$s).
Inkjet printing
In general, inkjet printing is a digital, non-contact printing
technology which does not require an intermediate carrier for the
image information deposited onto the substrate. In the inkjet
process the ink can be transferred and patterned directly onto the
substrate from a nozzle or print head. This allows a great deal of
flexibility at the research and development stage as it does not
require the costly design and fabrication of specific stencils,
screens or rollers. Printed patterns can be evaluated and changed
rapidly and at minimal cost using simple CAD software. In
addition, the availability of flat bed instruments for intermediate
research, development and low volume production scale allows
transition from the laboratory bench to the production plant.
Inkjet can also be made large area and high throughput by the
ganging of many print heads into large linear arrays. However, to
achieve patterning the substrate, print source or a combination
of both must be moved. Typically, for small-scale developmental
work, printers with a movable XY print head are used. For large-
scale production, single pass linear arrays of inkjet print
heads are employed which can achieve resolutions of 360 dpi at
This journal is ª The Royal Society of Chemistry 2010
 View Article Online

24 m min1 with print widths of up to 1.1 m. Feature sizes of as

small as 10–20 mm are now being reported. Due to the inherent
accessibility of the technique for developmental purposes, its
transferability to production scale up and the low viscosity
requirements of the inks (2–25 mPa$s), inkjet is proving
increasingly popular in the development of printed electronic
devices.23
Inkjet technologies can be classified as continuous and drop on
Published on 31 August 2010. Downloaded by Universidade de Pernambuco on 19/02/2018 19:39:58.

demand (Fig. 5). Continuous inkjet technology generates

a constant stream of small ink droplets, which are charged
according to the image and controlled electronically. The
charged droplets are deflected by a subsequent electric field, while
the uncharged ones flow onto the substrate. This means that the
imaging signal for charging the droplets corresponds to a nega-
tive print image.
Drop on demand inkjet technology, on the other hand,
produces a droplet only if it is required by the image. Thermal
inkjet and piezo inkjet printing are known as the most important
drop on demand technologies. Thermal inkjet (bubble jet)
generates the drops by heating and localized vapourization of the
liquid in a jet chamber. However, there are some limitations for
thermal inkjet printers. Usually, only water can be used as the
solvent of the ink because of its vapour point, and this imposes
restrictions on printing materials that only dissolve in organic
solvents to form non-aqueous thermal inks. With piezo inkjet the
ink drop is formed and catapulted out of the nozzle by
mechanically deforming the jet chamber, an action resulting
from an electronic signal and the piezoelectric properties of the
chamber wall. Therefore, inks based on either organic or
aqueous solvents are suitable for piezoelectric inkjet printers. For
both printer types nozzle sizes are typically 20–30 mm in diam-
eter, droplet volumes produced are 10–20 pL and final drop
diameters are in the range 30–40 mm. Smaller nozzles allow for
smaller droplets to be produced and higher print resolution, with
drop volumes of 1 pL possible.
Crucial aspects of inkjet printing technology are the ink and
its physical properties, in particular the viscosity and surface Fig. 6 Chemical structure of (a) PANi; (b) PPy; (c) PEDOT:PSS; (d)
tension. The viscosity should be low enough to allow the PMAS.
channel to be refilled in about 100 ms, and surface tension
should be such that ink is held in the nozzle without dripping.21
In general, a typical viscosity for an inkjet printable ink is
below 20 mPa$s, and in practice, surface tension ranges from 28 when doped with a salt that protonates the imine nitrogens on
to 350 mN/m.35 The particle size and their distribution in the the polymer backbone.
ink is also quite an important factor for printing. The recom- The conductivity of PANi increases with doping from the
mended particle size is dependent on the print head used but is undoped insulating base form (1010 S cm1) to the fully doped,
typically below 1 mm and ideally <200 nm. When the particle conducting acid form (>1 S cm1).36 Doping and dedoping
size becomes of the order of 1 mm, printability problems processes are typically carried out chemically with common acids
may occur. such as hydrochloric acid and bases such as ammonium
hydroxide; electrochemical processes can also be used.
PANi has been studied for numerous applications including
3. Printable conducting polymers lightweight battery electrodes,37 electromagnetic shielding
devices,38,39 and anticorrosion coatings.40,41 However, even given
Polyaniline (PANi)
the extensive range of useful electrochemical properties, PANi
The chemical structure of PANi is illustrated in Fig. 6(a).36 The has seen limited widespread use, for many reasons including the
character ‘y’ indicates the degree of oxidation. Leucoemeraldine difficulty of processing the conducting form which, once formed
(y ¼ 1) is the fully reduced state. Pernigraniline (y ¼ 0) is the fully is environmentally robust but relatively intractable. Thus a key
oxidized state with imine links instead of amine links present. objective has been to increase the ease of processing and
These two states of PANi are both insulating. PANi is only manipulation of the polymer while maintaining its conducting
electrically conductive in its emeraldine oxidation state (y ¼ 0.5) polymeric form.

This journal is ª The Royal Society of Chemistry 2010 Analyst, 2010, 135, 2779–2789 | 2783

The printed deposition of polyanilines has been achieved using
both organic solvent-based inks and aqueous nanodispersions.
The printing of polyanilines from organic solvents has been
achieved by screen printing while examples of inkjet printed
PANi have employed nanodispersions.
Inkjet printed polyaniline. Until recently, fabrication options
available for conductive polyaniline were limited principally to
Published on 31 August 2010. Downloaded by Universidade de Pernambuco on 19/02/2018 19:39:58.
simple chemical polymerisation or electrochemical polymerisa-
tion. The former could either be a bulk process or for the direct
formation of films. However, the latter was only suitable for in
situ film formation. Although electrochemical deposition
produces superior films, it is more time consuming. Bulk chem-
ical production via the formation of polyaniline powders has
been achieved. The fabrication of PANi nanoparticles has now
been used to achieve both improved properties and increased
processability. Aqueous PANi nanodispersions doped with
dodecylbenzenesulfonic acid (DBSA) have been produced42 with
a uniform particle size of approximately 80 nm in the doped
conductive form. These nanodispersions have been found to be
highly stable over long periods of time (at least several months)
(unpublished observations). A rheological study indicated that
PANi-DBSA nanodispersions could be prepared as inks with
viscosities suitable for piezoelectric inkjet printing (2–12 mPa$s).
Surface tensions of the nanodispersions (27–30 dyn cm1) were
also similar to that of a commercial inkjet printing ink. Inkjet
printing of the nanoparticles was successfully carried out using
an Epson Stylus C45 printer. High-quality images were gener-
ated on paper in just 30 s. It was shown that both electrochem-
ically deposited3 and inkjet printed films formed from the
nanoparticles were comparable with electrochemically poly-
merised films in terms of conductivity and electroactivity.22
When inkjet printed onto carbon paste electrodes using an Epson
C46 printer at room temperature, the nanoparticles coalesced
to form high-quality, nanometrically smooth, uniform films.
The cyclic voltammograms obtained for this printed film were
Fig. 7 Scan rate study in HCl (1 M) of an inkjet-printed film (24 prints)
on a screen-printed carbon-paste electrode. Inset: graph of the depen-
dence of peak current of peaks A (r2 ¼ 0.9678) and A0 (r2 ¼ 0.9584) on
scan rate.22 (Reprinted with permission from ref. 22. Copyright 2008,
Elsevier.)
2784 | Analyst, 2010, 135, 2779–2789
View Article Online
Fig. 8 Effect of operating temperature on analytical response to
ammonia (I/I0 vs. [NH3]). Inset shows (I0  I)/I0 vs. log[NH3] at 60  C,
70  C and 80  C; +0.1 V potential applied to sensor.43 (Reprinted with
permission from ref. 43. Copyright 2008, Elsevier.)
well-defined and stable, containing two sets of polyaniline
oxidation/reduction couples corresponding to the trans-
formations among leucoemeraldine (LB), to emeraldine salt (ES)
and pernigraniline (PS) (Fig. 7). This study demonstrated that
these inkjet printed films could be finely controlled in terms of
their two-dimensional pattern, thickness and conductivity,
highlighting the level of precision that can be achieved by inkjet
printing. It also demonstrated that discrete polymer nano-
particles could communicate electrically with each other in an
analogous manner to a single polymer film.
A number of chemical sensors43 and biosensors3 have been
produced using inkjet printable PANi nanodispersions.42 PANi
is well known for its ammonia sensing capabilities and an
ammonia gas sensor was produced by inkjet printing PANi
nanodispersions onto an interdigitated electrode (IDE) array.43
The silver- or carbon-based IDEs were first prepared by screen
printing. Inkjet printing was performed using an Epson C46/C48
piezoelectric printer wherein the nanodispersion ink was used to
fill the cartridge. At room temperature, the sensor responded
rapidly to ammonia (t50 ¼ 15 s). Sensor recovery time, response
linearity and sensitivity were all significantly influenced by
the number of printed layers. The sensor was found to have
a stable logarithmic response to ammonia in the range of interest
(1–100 ppm) (Fig. 8).
A sensor for the amperometric detection of aqueous ammonia
was also fabricated by inkjet printing dodecylbenzenesulfonic
acid (DBSA) doped polyaniline nanoparticles (nanoPANi), this
time onto a screen-printed carbon paste electrode using
a Dimatix Materials Printer under 16 Volts with 20 mm drop
spaces.44 Cyclic voltammograms obtained using the inkjet prin-
ted PANi electrode indicated a greater sensitivity to ammonium
in aqueous solution under neural state (pH ¼ 7.5) than at more
acidic (pH ¼ 6) and more alkaline (pH ¼ 9) pH. The printed
sensor was also tested in a flow injection analysis system and was
found to have good sensitivity with an experimental detection
This journal is ª The Royal Society of Chemistry 2010
 View Article Online

Table 1 PANi characteristics (by Aldrich) and inkjet parameters used to

make sensing lines of this conductive polymer on alumina substrate48

Ink characteristics Inkjet parameters

Ink ¼ polyaniline (PANi) Nozzle diameter ¼ 70 mm

Concentration ¼ 2–3 wt%
(dispersion on xylene)
Conductivity ¼ 10–20 S/cm (film)
Viscosity ¼ 3 cP
Density ¼ 0.90–0.95 g/mL at 25  C
Published on 31 August 2010. Downloaded by Universidade de Pernambuco on 19/02/2018 19:39:58.
Voltage ¼ 50 Volts
Pulse length ¼ 80 ms
Frequency ¼ 15 Hz
Substrate speed ¼ 750 mm/s

Substrate ¼ alumina (1 cm 
1.5 cm)

and copper(II) chloride dispersion deposited onto a silver paste,

Fig. 9 Relationship between the amperometric response of the inkjet-
printed PANi modified electrode and ammonium concentration (as
ammonium chloride) at pH 7.5 (n ¼ 3): (a) 0–80 mM (0–1.44 ppm); (b)
0.25–10 mM (1.8–450 ppm). Injection volume: 200 mL; flow rate: 400 mL
min1; injection interval: 200 s; applied potential: 0.3 V vs. AgCl
pseudo-reference electrode; electrode area: 7.07 mm2. Inset: (i) real-time
amperometric responses for calibration plot, and (ii) calibration plot
before logarithmic treatment.44 (Reprinted with permission from ref. 44.)
limit of 20 mM and a theoretical detection limit of 3.17 mM
(0.54 ppm) (Fig. 9). The printing approach allowed for rapid
production of low cost, solid state, polyaniline based aqueous
ammonia sensors.
Ammonia sensors have also been prepared by printing poly-
aniline dispersions on various substrates and using a radio
frequency detector to monitor conductance.45 The inkjet printed
PANi-DBSA dispersion worked as an ammonia sensor but
cycling time on exposure to volatile acid or heat was slow,
probably because of deep penetration of the low viscosity PANi-
DBSA inkjet formulation through the sheet. A screen print paste
prepared by evaporating the PANi-DBSA inkjet formulation to
a higher viscosity was more successful. Together, the PANi-
DBSA and the radio frequency detector form a non-contact
sensor for the detection of ammonia within sealed packages, and
is therefore a low cost device that might be suitable for smart
packaging applications.
Virji et al. had shown that the selectivity of PANi nanofibres
could be modified by combining it with copper(II) chloride,
showing irreversible sensitivity to hydrogen sulfide.46 Recently,
Crowley et al. demonstrated that a hybrid organic/inorganic inkjet
printed electrode based on a combination of PANi nanoparticles
screen-printed interdigitated electrode could achieve reversible
detection of hydrogen sulfide in the range of 1–100 ppm.47
Commercial PANi dispersions have also been inkjet printed to
fabricate sensors.48 Loffredo et al. printed a PANi suspension
obtained from Sigma Aldrich (Table 1) onto alumina substrates
to from gas sensing films and the response to ammonia was tested
and compared with sensors obtained by spin coating. Both
fabrication methods produced PANi films that were sensitive to
NH3, and the response of devices obtained by inkjet was greater
than those observed with films prepared by spin coating.
PANi–poly(4-styrenesulfonate) (PSS) nanoparticles (average
diameter ¼ 28 nm) have also been used to form aqueous
dispersions with physical properties that make them amenable to
inkjet printing.5 The viscosity and surface tension were found to
be ca. 4 mPa$s and ca. 63 mN m1 respectively, making them
suitable for inkjet printing. A commercial Canon printer (Pixma
ip 1000) with a resolution of 4800  1200 dpi (dpi ¼ dots per
inch; 1 inch z 2.54 cm) was used to print PANi HCl and NH3
gas sensors on photopaper. The printed sensors showed superior
sensitivity and more rapid response time relative to conventional
photolithography printed PANi-based chemical sensors.
Inkjet printing has recently been combined with vapour
deposition polymerization to achieve the patterned deposition of
PANi.49 Here, the oxidant (ammonium persulfate) was first
patterned using inkjet printing and a PANi film was formed by
the introduction of a chemical vapour of aniline monomer which
polymerized when in contact with the oxidant. Films of less
than 0.5 mm were formed with line widths down to 80 mm,
although with very poor edge definition. Below a line width
of 500 mm, sheet resistance began to increase significantly to
1.9  104 U ,1.
Roll-to-roll printing of polyanilines. A custom designed roll-to-
roll system in which a flexo printing unit, a gravure printing unit
and a nanoimprinting unit were combined has been used to
produce novel polyaniline based structures in a single pass.29 The
width of the web is 50 mm, making it possible to run experiments
using only a few millilitres of printing ink in a run. The thickness
of the resulting PANi-DBSA film as determined by AFM was
160 nm and the conductivity of the PANi-DBSA (3 S/cm) was
not affected by the printing processes.
Screen-printed polyaniline sensors. PANi thermocouples have
been fabricated by screen printing a Ni/PANipol composite ink
formulated from PANipol M (PANi in N-methylpyrroldinone)

This journal is ª The Royal Society of Chemistry 2010 Analyst, 2010, 135, 2779–2789 | 2785

onto flexible polyester substrates. The optimum device configu-
ration made use of a Ni/PANi couple combined with a printed
NiCr couple, and performance of these printed temperature
sensors was comparable to conventional thermocouples.50 PANi
thick film (35–120 mm) pH sensors have also been fabricated by
screen printing a PANi composite ink onto an alumina substrate
with screen-printed gold IDA electrodes. In this case PANi
powder was combined with a binder (polyvinyl butyral),
Published on 31 August 2010. Downloaded by Universidade de Pernambuco on 19/02/2018 19:39:58.
surfactant (PS3) and solvent (ethylene glycol monobutyl ether) to
form a screen printable paste. The sensors showed excellent pH
sensitivity with up to three orders of magnitude change in
conductance over a pH range 2–11, and responses were stable for
over 96 h.51
Polypyrrole (PPy)
PPy can be used in a broad range of applications, including
rechargeable batteries,9,52 capacitors,53 sensors54 and actuators.6,7
The chemical structure of PPy is shown in Fig. 6(b). The heter-
oaromatic nature and extended p-conjugated backbone struc-
ture of PPy provide it with chemical stability and electrical
conductivity, respectively. However, the p-conjugated backbone
structure is not sufficient to produce appreciable conductivity on
its own. Partial charge extraction from the PPy chain is also
required, and this is achieved by a chemical or an electrochemical
process referred to as doping. The conductivity of the neutral
PPy is remarkably changed from insulating to a metallic by
doping.55–57
Functionalisation of PPys to render the polymer soluble has
significant deleterious effects on electronic properties. The
formation of stable nanodispersions of PPy has also proven more
challenging than for PANi. Both of these factors have limited
printing applications.
Inkjet printed PPy sensor. Inkjet printed films of the conduc-
tive polymer PPy have been used for vapour sensing at room
temperature by Mabrook et al.58 A commercial PPy dispersion
from Sigma Aldrich (Product Number 482552) was printed using
an HP thermal printer with a resolution of 600  600 dots per
inch. Ethylene glycol was added (10%) to this polypyrrole
dispersion to reduce the viscosity to less than 100 cP and obtain
a surface tension of approximately 35 mN/m. A significant
increase in conductivity was observed upon exposure of the films
to the vapours of simple alcohols. The value of the fractional
resistance change, DR/R, of the films increased linearly with the
concentration of either ethanol or methanol. A relatively high
DR/R value, approximately 90%, was obtained on exposure to
5000 parts per million of methanol. The response time of the
inkjet printed sensors to polar vapours was generally shorter
than that for non-polar compounds.
Screen-printed PPy carbon dioxide gas sensor. PPy sensors for
CO2 monitoring were prepared by screen printing.25 PPy powder
was prepared by chemical oxidation using FeCl3 as oxidant. The
PPy paste for screen printing was prepared by mixing 60 wt%
PPy powder with 40 wt% binder (8 wt% butyl carbitol mixed with
92 wt% ethyl cellulose in an agate mortar). The rheological
properties of the final paste were not discussed. Screen-printed
PPy sensors showed a linear variation in response with an
2786 | Analyst, 2010, 135, 2779–2789
View Article Online
Fig. 10 Dynamic response of PPy-I and PPy-II sensors to 100, 400 and
700 ppm CO2 at room temperature.25 (Reprinted with permission from
ref. 25. Copyright 2008, Elsevier.)
increase in CO2 concentration but a saturation effect at higher
concentrations (Fig. 10). SEM images revealed that the
morphology of the PPy obtained was dependent on the Py/FeCl3
weight ratio used in the chemical polymerization.
Polythiophene (PTh)
The structure of polythiophenes is analogous to the polypyrroles
discussed above. The chemistries available to polythiophenes has
meant they are more amenable to functionalisation and solubi-
lisation in organic solvents. However, attaching functional
groups (e.g. alkyl sulfonates) to render them water soluble does
compromise electronic properties. The synthesis of polymers of
the bicyclic 3,4-ethylenedioxythiophene (EDOT) and its deriva-
tives59 provided a significant breakthrough in processable poly-
thiophenes. PEDOT has a very stable and highly conductive
cationic ‘doped’ state. An industrially useful form of oxidized
PEDOT has been prepared by aqueous oxidative polymerization
of the EDOT monomer in the presence of a template polymer,
usually polystyrene sulfonic acid (PSS). PSS is a commercially
available water-soluble polymer which serves as a good disper-
sant for aqueous PEDOT. Polymerization with the oxidant
sodium peroxodisulfate yields a processable PEDOT:PSS-
complex in the conductive, cationic form (Fig. 6(c)).
Poly(3-hexylthiophene) (P3HT) (Fig. 6(d)) is also widely
employed as an organic semi-conductor and is being employed
extensively in the development of organic and printed transistors
and solar cells. The main challenge with organic semi-conductors
is to meet the charge carrier mobilities achievable with current
forms of silicon. Traditionally, amorphous silicon could reach
mobilities of approx. 1 cm2/(Vs). However, crystalline silicon can
reach mobilities of 10–100 cm2/(Vs), resulting in extremely high
switching speeds. However, P3HT remains several orders of
magnitude away with carrier mobilities typically of the order of
0.01–0.1 cm2/(Vs). Nonetheless, such materials still have many
useful applications.60
P3HT is being investigated for its application in printed
organic semi-conductors and has recently been studied using
several printing methods including gravure and inkjet. P3HT
possesses the typical conducting polythiophene backbone and is
This journal is ª The Royal Society of Chemistry 2010

made processable by the addition of alkyl side chains which make
it soluble in organic solvents. The presence of such solvents puts
additional rigours on the printing process and materials
compatibilities. Notwithstanding this, recent reports have shown
that it can form the basis of a fully gravure-printed field-effect
transistor with mobilities of 0.04 cm2/(Vs).61 In organic solar
cells, it has been used as the electron donor in combination with
the electron acceptor fullerene derivative [6,6]-phenyl C61 butyric
Published on 31 August 2010. Downloaded by Universidade de Pernambuco on 19/02/2018 19:39:58.
acid methyl ester (PCBM). This blend has been deposited using
inkjet printing and has achieved efficiencies of 3.5%.62
Inkjet printing. Printed polythiophenes such as PEDOT:PSS
have also been utilised as chemical sensors. PEDOT:PSS has
been used for the detection of organic vapours.63 The electrical
resistance of inkjet printed films was monitored when exposed to
atmospheres containing alcohol. The resistance of thin films (one
or two printed layers) increased sharply and non-reversibly as the
concentration of alcohol vapour in a carrier gas increased. In
contrast, the resistance of thick films (four to five printed layers),
showed a decrease in resistance with increasing vapour concen-
tration. An intended application for this inkjet printed device is
a handheld instrument (either a single shot or multi-use device,
depending on the film thickness) to monitor the presence of
organic vapours. For example, the PEDOT:PSS inkjet printed
chemiresistors could be used as the sensing elements in a cheap,
disposable, handheld personal breathalyser to monitor the
concentration of alcohol in the blood or as the sensing elements
in ALCOLOCK devices.
Gas sensors have also been inkjet printed using a range of
regioregular polythiophenes with different functional groups.64
Sensors were printed using a custom inkjet deposition system
with a Microfab drop-on-demand single nozzle print head
delivering 50 pL drops through 30 mm nozzles with 4 mm drop
placement accuracy. Each sensor film had a diameter of 200 mm
and the overall size of the 24-element array was 3 mm  3 mm.
The resulting PTh chemical sensor array was used to detect
a range of volatile organic compound vapours in the low ppm
range and to discriminate between them using principal
component analysis.
Polythiophene biosensors have also been fabricated by
thermal inkjet printing of PEDOT:PSS and glucose oxidase
(GOD) in sequence onto ITO glass.65 Films of PEDOT and
GOD were printed using a prototype system based on an Olivetti
I-Jet print head with 208 nozzles and drop volume of 10–12 pL.
The resulting amperometric biosensors were finally encapsulated
with a cellulose acetate membrane by dip coating to prevent
dissolution of the active layers. The biosensor did successfully
detect glucose but sensitivity was relatively low (6.43 mA M1
cm2) compared to other materials based on GOD entrapped in
electropolymerised PEDOT films (5 mA M1 cm2). This was
attributed to the relatively thick printed PEDOT films used
(230 nm) and the effect of this on diffusion processes.
Conducting polymers can also be used as strain gauges and
show a much more significant change in resistivity with applied
strain compared to metals. Controlled inkjet printing of
PEDOT:PSS onto cotton fabric resulted in selective conduc-
tivity, which increased with curing of the printed sample.66,67
Silver connecting leads (0.5 mm width) were applied to the fabric
by both inkjet printing and electroless plating. SEM and EDX
This journal is ª The Royal Society of Chemistry 2010
View Article Online
observations revealed thick deposition of silver on the Nylon 6,6
fabrics, even after washing and rubbing which indicates reason-
able adhesion. PEDOT:PSS sensing lines were inkjet printed
onto the fabrics and showed reversible resistance changes in
response to applied external strain. Electrical analysis proved
that human motions can be sensed by the printed assembly of
sensors and connectors.
PEDOT:PSS is extensively used as a transparent organic
conductor to replace inorganic materials such as ITO and as the
basis of conductive interconnects for use in flexible and printed
electronics. The main focus of development has been on
increasing its conductivity and formulation for a range of
deposition techniques.68 Currently, conductivities of 1000 S cm1
are available from commercial sources (Clevios from H.C. Starck
GmbH, Germany). The rheological properties and conductivities
of the ink are thus affected by the requirements of the print
process. For example, Clevios PH1000 has conductivities in
excess of 900 S cm1 with a high solids content which yields
a viscosity of 50 mPa$s, making it ideal for flexo and gravure. For
inkjet printing, a lower viscosity formulation of 5–18 mPa$s with
conductivities of >200 S cm1 is available (Clevios P Jet V2).
Parallel, merging and future developments
Other advances in materials science and engineering will impact
on our ability to print sensing materials based on conducting
polymers. The advance of printable electronics should eventually
lead to the ability to print integrated sensing systems. Some
advances of relevance are reviewed below.
Carbon nanotubes (CNTs). Recently, the addition of CNTs to
conducting polymer inks to produce printable formulations with
improved properties has attracted attention. The application of
these printable organic conductor based materials as sensors is
yet to be fully explored.
Transparent and conductive patterns of carboxyl functional-
ized single-walled carbon nanotubes (SWCNT-COOHs) and the
composites of those with PEDOT:PSS have been deposited on
various substrates by inkjet printing.69 For low print repetitions,
the PEDOT:PSS/SWCNT-COOH composite patterns show
enhanced conductance as compared to the corresponding
PEDOT:PSS conductors. Patterns with sheet resistivities as low
as 1 kU/, were achieved, and while there is a trade-off between
transparency and conductivity – highly transparent patterns
(90%) with a reasonably low resistivity of 10 kU/, were
realised.
Denneulin et al. mixed PEDOT:PSS (Baytron P) with PEG
functionalized single-walled carbon nanotubes to produce
a much more conductive film compared to the original
PEDOT:PSS film70 Conductive patterns were deposited on
polymer films using a Dimtix 2831 inkjet printer. Performances
of several CNTs were evaluated (single-walled, multi-walled and
functionalized CNTs) and sheet resistances ranged from 10 537
to 225 U/,. The later achieved using CNTs with poly(ethylene
glycol) (PEG-SWCNT) appeared to be the best candidates for
printed electronics with sheet resistances as low as 225 U/,,
which is one of the lowest resistances obtained by inkjet printing.
This work also discusses the CNT network performances and
emphasizes the degradation of electrical properties linked to the
Analyst, 2010, 135, 2779–2789 | 2787

Published on 31 August 2010. Downloaded by Universidade de Pernambuco on 19/02/2018 19:39:58.
Fig. 11 Normalized current responses of inkjet printed PANi nano-
particle (black) and PANi-MWCNT (red) electrodes to ammonia in the
range 0.25–1000 ppb (previously unpublished data).
use of surfactants. This study represents an important step for
the integration of CNTs in printed electronics applications and
offers new opportunities to produce cost-effective electronics.
Recent unpublished collaborative work between our two
groups (Wallace and Killard) has shown the compositing of
PANi nanoparticles with multi-walled carbon nanotubes
(MWCNTs). The PANi nanoparticle dispersion was found to be
an excellent solvent for the distribution of 1% (w/v) MWCNTs
into a stable formulation suitable for inkjet printing. Both PANi
and CNT-modified PANi electrodes were reasonably compa-
rable in their sensitivity to ammonia gas, both being capable of
detecting 250 ppb. However, the CNT-modified formulation
showed more rapid response and recovery times. At 1 ppm, t50
responses were 16 s for PANi nanoparticles, and 7 s for PANi/
MWCNTS. Recoveries (t50) were greater than 300 s for the PANi
nanoparticle based system and 52 s for PANi/MWCNTS
(Fig. 11).
The synthesis and inkjet processing of a water-dispersible poly-
aniline composite ink comprising poly(methoxyanilinesulfonic
acid) (PMAS)71 with MWNT loadings of up to 32% has been
described72,73 Dispersions with nanotubes (concentration 10 mg
mL1, temperature 25  C, viscosity 5.5 cP, and surface tension
72 mN m1) were inkjet printed using a Dimatix materials depo-
sition system on PET. Cyclic voltammograms obtained using films
printed onto Au-PVDF exhibited three redox couples attributed
previously to the following interconversions: PANi/PMAS leu-
coemeraldine–emeraldine, PMAS emeraldine–pernigraniline, and
PANi emeraldine–pernigraniline. The sheet resistance and
conductivity of free-standing films at the highest loading fractions
were 5 U/, and 51 S/cm respectively. The unique combination
of conducting electroactive polymers with conducting carbon
nanotubes has been proven to be an ideal formulation with all
of the demanding characteristics needed for inkjet printing.
These materials could be readily deposited onto substrates such
as photopaper, PET, Pt-ITO and Au-PVDF. A sheet resistance
of 500 U/, was attainable for a single printed layer on photo-
paper.
2788 | Analyst, 2010, 135, 2779–2789
View Article Online
Printed electronics. A range of electronic components
including transistors, resistors, capacitors, electrochromic
display cells, solar cells and interface circuits have been printed
with PEDOT:PSS. Such devices are important in the realization
of all-printed conducting polymer sensing systems.
The printed transistor circuit was first studied by Sirringhaus
et al.74 in 2000. Since then, a series of all printed transistors and
devices were fabricated and characterized.
Kawase et al. fabricated all-polymer inkjet printed
PEDOT:PSS thin film transistors (TFTs) with a desktop inkjet
printer.75 The transfer characteristics of the device exhibited
a high on–off ratio of more than 105, which was found to be the
same level as reference devices with gold electrodes.
A study by Mannerbro et al. demonstrated that it is possible to
use a simple desktop inkjet printer to manufacture organic
electrochemical circuitry on rough flexible carrier substrates, e.g.
glossy paper.76 Due to the ease with which a circuit designer can
go from design idea to a real prototype device using a flexible
inkjet printing system, this technique can be anticipated to be
used for the production of a series of small devices based on the
electrochemical technology discussed.34
Printing has also been used to produce all-polymer
PEDOT:PSS capacitors. A commercial Epson Stylus color 480
SXU printer with resolution of 720  720 dpi was used to inkjet
print a commercial PEDOT:PSS dispersion (Baytron P) as the
conductive layers of the capacitor. Poly(biphenyltetracarboxylic
dianhydride-co-phenylenediamine) (PBPDA-PD) was chosen as
the insulating material.77
RC filter circuits based on these capacitors have also been
successfully fabricated using PEDOT:PSS (Baytron P) and PANi
(Sigma) as the conductive layers.78
Conclusion
Given the inherent properties of conducting polymers they have
become attractive for use as sensing elements in a diverse range of
application areas. However, many practical applications have
been thwarted by a lack of convenient processability and device
fabrication options.
The development of appropriate chemistries to produce solu-
tions or stable nanodispersions has overcome this limitation. The
advent of convenient printing technologies coupled with appro-
priate conducting polymer ink formulations has helped address
this limitation.
Other parallel advances in ink formulations containing addi-
tional conductors (e.g. conducting polymers, carbon nanotubes)
as well as in printing electronic circuits are merging with the areas
discussed above. No doubt the ability to print complete sensing
systems including power supply, simple electronics and the
conducting polymer transducers will emerge – maybe it already
has?
Acknowledgements
The authors K.C. and A.J.K. would like to acknowledge the
assistance of Enterprise Ireland in this work under Grant No.
PC/2007/138. G.G.W., R.L.S. and B.W. acknowledge the
continued support of the Australian Research Council.
This journal is ª The Royal Society of Chemistry 2010

References
1 C. Chiang, et al., Phys. Rev. Lett., 1977, 39, 1098–1101.
2 H. Shirakawa, et al., J. Chem. Soc., Chem. Commun., 1977, 578–580.
3 A. Morrin, et al., Electrochem. Commun., 2005, 7, 317–322.
4 C. M. Hangarter, et al., J. Mater. Chem., 2010, 20, 3131–3140.
5 J. Jang, J. Ha and J. Cho, Adv. Mater., 2007, 19, 1772–1772.
6 Y. Berdichevsky and Y. Lo, Adv. Mater., 2006, 18, 122–125.
7 E. Smela, Adv. Mater., 2003, 15, 481–495.
8 S. Kuwabata, S. Masui and H. Yoneyama, Electrochim. Acta, 1999,
44, 4593–4600.
Published on 31 August 2010. Downloaded by Universidade de Pernambuco on 19/02/2018 19:39:58.
9 K. Park, S. Schougaard and J. Goodenough, Adv. Mater., 2007, 19,
848–848.
10 N. Gomez and C. Schmidt, J. Biomed. Mater. Res., Part A, 2007, 81,
135.
11 A. F. Quigley, et al., Adv. Mater., 2009, 21, 4393–4397.
12 J. Lee, J. Lee and C. Schmidt, J. R. Soc. Interface, 2009, 6, 801–810.
13 J. M. Razal, et al., Adv. Funct. Mater., 2009, 19, 3381–3388.
14 G. G. Wallace, et al., Conductive Electroactive Polymers: Intelligent
Polymer Systems, CRC Press, Taylor & Francis Group, Boca
Raton, 3rd edn, 2008.
15 Y. Lee, J. Lee and D. Lee, Synth. Met., 2000, 114, 347–353.
16 I. Chronakis, S. Grapenson and A. Jakob, Polymer, 2006, 47,
1597–1603.
17 J. Yu, S. Fridrikh and G. Rutledge, Adv. Mater., 2004, 16, 1562–1566.
18 J. Foroughi, et al., Synth. Met., 2008, 158, 104–107.
19 S. Pomfret, et al., Polymer, 2000, 41, 2265–2269.
20 D. Verma and V. Dutta, Sens. Lett., 2009, 7, 143–147.
21 L. Pettersson, S. Ghosh and O. Ingan€as, Org. Electron., 2002, 3,
143–148.
22 A. Morrin, et al., Electrochim. Acta, 2008, 53, 5092–5099.
23 P. Calvert, Chem. Mater., 2001, 13, 3299–3305.
24 H. Kipphan, Handbook of print media, Springer, Heidelberg, 2001, ch.
1, pp. 40–65.
25 S. Waghuley, et al., Sens. Actuators, B, 2008, 128, 366–373.
26 F. Guan, et al., Appl. Surf. Sci., 2004, 230, 131–137.
27 X. Wang, et al., Thin Solid Films, 2004, 449, 125–132.
28 R. Parashkov, et al., Adv. Mater., 2005, 17, 1523–1526.
29 T. Makela, et al., Microelectron. Eng., 2007, 84, 877–879.
30 M. Tuomikoski, et al., Microelectronics and Packaging Conference,
European, 2009.
31 J. Choi, K. Cho and J. Yim, Eur. Polym. J., 2010, 46, 389–396.
32 L. Gonzalez-Macia, et al., Analyst, 2010, 135, 845–867.
33 J. Gullichsen and H. Paulapuro, Papermaking science and technology,
Finnish Paper Engineers’ Association, Jyv€askyl€a, Finland, 1999.
34 Printed organic and molecular electronics, ed. D. Gamota et al.,
Springer, New York, 2004, ch. 3, pp. 290–293.
35 B. de Gans, P. Duineveld and U. Schubert, Adv. Mater., 2004, 16,
203–213.
36 W. Huang, B. Humphrey and A. MacDiarmid, J. Chem. Soc.,
Faraday Trans. 1, 1986, 82, 2385–2400.
37 J. Desilvestro, W. Scheifele and O. Haas, J. Electrochem. Soc., 1992,
139, 2727–2736.
38 J. Joo and A. Epstein, Appl. Phys. Lett., 1994, 65, 2278–2280.
This journal is ª The Royal Society of Chemistry 2010
View Article Online
39 D. Trivedi and S. Dhawan, Synth. Met., 1993, 59, 267–272.
40 W. Lu, R. Elsenbaumer and B. Wessling, Synth. Met., 1995, 71,
2163–2166.
41 M. Fahlman, S. Jasty and A. Epstein, Synth. Met., 1997, 85, 1323–
1326.
42 O. Ngamna, et al., Langmuir, 2007, 23, 8569–8574.
43 K. Crowley, et al., Talanta, 2008, 77, 710–717.
44 K. Crowley, et al., Analyst, 2008, 133, 391–399.
45 N. Clark and L. Maher, React. Funct. Polym., 2009, 69, 594–600.
46 S. Virji, et al., Nano Lett., 2004, 4, 491–496.
47 K. Crowley, A. Morrin, R. Shepherd, M. I. H. Panhuis,
G. G. Wallace, M. R. Smyth and A. J. Killard, IEEE Sens. J.,
2010, 10, 1419–1426.
48 F. Loffredo, et al., Macromol. Symp., 2007, 247, 357–363.
49 J. Cho, K.-H. Shin and J. Jang, Thin Solid Films, 2010, 518,
5066–5070.
50 S. Duby, B. Ramsey and D. Harrison, Electron. Lett., 2005, 41,
312–314.
51 E. Gill, et al., Sensors, 2007, 7, 3329–3346.
52 H. Song and G. Palmore, Adv. Mater., 2006, 18, 1764–1768.
53 M. Tsai, P. Chen and J. Do, J. Power Sources, 2004, 133, 302–311.
54 K. An, et al., Adv. Mater., 2004, 16, 1005–1009.
55 J. Bredas, et al., Phys. Rev. B: Condens. Matter, 1984, 30, 1023–1025.
56 J. Bredas, et al., Synth. Met., 1984, 9, 265–274.
57 Handbook of conducting polymers, ed. T. Skotheim, R. Elsenbaumer
and J. Reynolds, CRC Press, New York, 1998.
58 M. Mabrook, C. Pearson and M. Petty, Sens. Actuators, B, 2006, 115,
547–551.
59 G. Heywang and F. Jonas, Adv. Mater., 1992, 4, 116–118.
60 Organic and Printed Electronics, ed. K. Hecker, VDMA Verlag
GmbH, Germany, 3rd edn, 2009.
61 M. Voigt, et al., Adv. Funct. Mater., 2010, 20, 239–246.
62 C. Hoth, et al., J. Mater. Chem., 2009, 19, 5398–5404.
63 M. Mabrook, C. Pearson and M. Petty, Appl. Phys. Lett., 2005, 86,
013507.
64 B. Li, et al., Sens. Actuators, B, 2007, 123, 651–660.
65 L. Setti, et al., Biosens. Bioelectron., 2005, 20, 2019–2026.
66 A. Sawhney, et al., AATCC Rev., 2007, 7, 42–45.
67 P. Calvert, et al., Mol. Cryst. Liq. Cryst., 2008, 484, 291–302.
68 M. Lopez, J. Sanchez and M. Estrada, Int. Caribbean Conf. Devices,
Circuits Syst., 8th, ICCDCS, 2008.
69 T. Mustonen, et al., Phys. Status Solidi B, 2007, 244, 4336–4340.
70 A. Denneulin, et al., Nanotechnology, 2009, 20, 385701.
71 D. Zhou, et al., Synth. Met., 2000, 114, 287–293.
72 W. Small, et al., J. Mater. Chem., 2007, 17, 4359–4361.
73 W. Small, Small, 2007, 3, 1500–1503.
74 H. Sirringhaus, et al., Science, 2000, 290, 2123.
75 T. Kawase, et al., Thin Solid Films, 2003, 438–439, 279–287.
76 R. Mannerbro, et al., Synth. Met., 2008, 158, 556–560.
77 Y. Liu, T. Cui and K. Varahramyan, Solid-State Electron., 2003, 47,
1543–1548.
78 B. Chen, et al., Solid-State Electron., 2003, 47, 841–847.
79 J. A. Lim, et al., Appl. Phys. Lett., 2006, 88, 082102.
80 J. Ouyang, et al., Polymer, 2004, 45, 8443–8450.
Analyst, 2010, 135, 2779–2789 | 2789