Thiago Regis Longo Cesar Paixão

Subrayal Medapati Reddy Editors

Materials
for Chemical
Sensing
Materials for Chemical Sensing
Thiago Regis Longo Cesar Paixão
Subrayal Medapati Reddy
Editors

Materials for Chemical

Sensing

123
Editors
Thiago Regis Longo Cesar Paixão Subrayal Medapati Reddy
Department of Fundamental Chemistry Chemistry Division, School of Physical
Institute of Chemistry Sciences and Computing
University of São Paulo University of Central Lancashire
São Paulo, SP Preston, Lancashire
Brazil UK

ISBN 978-3-319-47833-3 ISBN 978-3-319-47835-7 (eBook)

DOI 10.1007/978-3-319-47835-7
Library of Congress Control Number: 2016954604

© Springer International Publishing AG 2017

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt from
the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, express or implied, with respect to the material contained herein or
for any errors or omissions that may have been made.

Printed on acid-free paper

This Springer imprint is published by Springer Nature

The registered company is Springer International Publishing AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Thiago Regis Longo Cesar Paixão dedicates
this book to his wife, Juliana Naozuka and
son, Thiago Akio Naozuka da Paixão—who
have inspired, encouraged and helped him in
everything he has done.

Subrayal Medapati Reddy dedicates this book

to his wife and children for their friendship,
support and understanding.
Preface

Over the years, the number of academic publications on the development of

chemical sensors has increased dramatically from 76 articles in 1983 to 7461
articles in 2015 (according to a Scopus Web search using the phrase “chemical
sensor”). Additionally, the commercial market is increasing its search for new
chemical sensors in order to monitor a plethora of analytes in real time and in an
efficient and cost-effective way. Areas of application include environmental mon-
itoring, food science and safety, and wearable technologies for point-of-care
devices tracking medical status. There is an ongoing work in the development and
application of cheap, biodegradable materials to monitor diseases in areas with poor
infrastructure. The development of new chemical sensors and the improvement of
existing ones have become a collaborative endeavor, integrating multiple disci-
plines, such as chemistry, physics, engineering, biology, materials science, math-
ematics, and bioinformatics. This book starts with a definition of the chemical
sensor in Chap. 1, followed by how this chemical sensor could extract chemical
information using optical and electrochemical techniques in Chap. 2.
The combination of chemical sensor with low-cost materials, more specifically
paper-based devices, has progressed in the last 10 years and has been stimulating
recent research activities in the development of point-of-care device to be used in
developing countries. Chapter 3 is dedicated to this type of device with the
exploitation of biomaterials, such as enzymes and antibodies, to recognize the
chemical information. The key point for the development of a chemical sensor is
related to the materials used to recognize and translate the chemical information
through a chemical interaction or reaction. However, there is not a quantitative
theory or model to be used which describes physical and chemical parameters to
obtain improved materials that will recognize the species being analyzed, and the
steps to obtain a better material are largely empirical and could be sometimes seen
more like an art than a science. However, the use of enzymes to translate the
information is common, and researchers are attempting to mimic the enzyme and
antibody environments in order to create artificial receptors and other biomimetic
compounds, such as molecularly imprinted polymers. In this approach, they are

vii
viii Preface

trying to copy nature in order to achieve the same level of recognition and affinity,
and Chap. 4 will discuss this topic.
Chapters 4–8 will show how different natural and synthetic materials influence
the development of chemical sensors. Chapter 5 will discuss how classes of carbon,
such as graphene and its derivatives, offer different nanofeatures, structures, and
dimensions, which provide transduction recognition with potentially novel sensing
properties. Chapter 6 will show how nanomaterials could be turned into a per-
sonalized monitoring platform in the form of wearable and implantable sensor
network systems, which would allow people’s activities to be monitored. Chapters
7 and 8 will demonstrate how other synthetic or natural materials, such as
self-assembled films and phthalocyanines, can be useful for enhancing sensing
performance. Chapter 9 will demonstrate how array-based chemical sensors,
combined with chemometric data processing tools, can be used to mimic the human
tongue, one of our more sophisticated sensors.
We would like to thank all the contributing authors for their enthusiasm and
participation in the preparation of this book. We also would like to express our
gratitude to the staff of Springer, in particular, Ania Levinson and Brian Halm, for
their assistance in bringing this book to print and publication.

São Paulo, Brazil Thiago Regis Longo Cesar Paixão

Lancashire, UK Subrayal Medapati Reddy
Contents

1 Introduction of Materials Used in Chemical Sensors . . . . . . . . . . . . 1

William Reis de Araujo, Subrayal Medapati Reddy
and Thiago Regis Longo Cesar Paixão
2 Information Extraction Techniques in Chemical Sensing . . . . . . . . . 7
Thiago Matheus Guimarães Selva, Tiago Luiz Ferreira
and Thiago Regis Longo Cesar Paixão
3 (Bio)Chemical Sensors Based on Paper . . . . . . . . . . . . . . . . . . . . . . . 29
Nipapan Ruecha, Kentaro Yamada, Koji Suzuki and Daniel Citterio
4 Membrane Technologies for Sensing and Biosensing . . . . . . . . . . . . 75
Subrayal Medapati Reddy
5 Interfacing Graphene for Electrochemical Biosensing . . . . . . . . . . . 105
Onur Parlak
6 Nanomaterials as Implantable Sensors . . . . . . . . . . . . . . . . . . . . . . . 123
Roger Jagdish Narayan and Nishant Verma
7 Self-assembly Thin Films for Sensing . . . . . . . . . . . . . . . . . . . . . . . . 141
Celina Massumi Miyazaki, Anerise de Barros,
Daniela Branco Tavares Mascagni, Juliana Santos Graça,
Paula Pereira Campos and Marystela Ferreira
8 Phthalocyanines as Sensitive Materials for Chemical Sensors . . . . . 165
Debdyuti Mukherjee, Revanasiddappa Manjunatha,
Srinivasan Sampath and Asim Kumar Ray
9 Materials for Electronic Tongues: Smart Sensor Combining
Different Materials and Chemometric Tools . . . . . . . . . . . . . . . . . . . 227
Manel del Valle
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267

ix
Chapter 1
Introduction of Materials Used
in Chemical Sensors

William Reis de Araujo, Subrayal Medapati Reddy

and Thiago Regis Longo Cesar Paixão

1.1 From Sensors to Chemical Sensing

Since the advent of smartphone technologies, the word “sensor” has become more
and more commonplace outside of the academic environment. Nowadays, it is easy
to find smartphones with a variety of sensors, for example, proximity, motion,
ambient light, gyroscopic, and magnetic. These sensors are devices that detect
inputs from the physical environment, in order to generate an output signal that can
be read and understood by a human and/or can be transmitted electrically by
someone or a machine. A simple example of a sensor is the mercury-based glass
thermometer that has a heat as input and as consequence of the change in tem-
perature the liquid mercury expands, or contracts, indicating a value of the tem-
perature measured in a calibrated marked gauge that can be detected by a natural
sensor, the human eye. Basically, the physical devices highlighted above translate
physical properties into a human-readable output just as some human analogues can
do through, for example, touch, vision, or hearing. However, nature has given us
sensorial systems responsible not only to translate physical quantities as an inter-
pretation of the outside world, but also the ability to sense chemicals through taste
and olfaction systems. Combinatorially, chemical information can be translated
together with physical information, to better understand the environment [1, 2].

W.R. de Araujo
T.R.L.C. Paixão

Departamento de Química Fundamental, Instituto de Química, Universidade de São Paulo,
Avenida Prof. Lineu Prestes, 748, 05508-000 São Paulo, SP, Brazil
e-mail: [email protected]
S.M. Reddy (&)
Chemistry Division, School of Physical Sciences and Computing, University of Central
Lancashire, Preston, Lancashire PR2 2HE, UK
e-mail: [email protected]

© Springer International Publishing AG 2017 1

T.R.L.C. Paixão and S.M. Reddy (eds.), Materials for Chemical Sensing,
DOI 10.1007/978-3-319-47835-7_1
2 W.R. de Araujo et al.

Natural chemical sensors, like the mammalian taste systems, perform an eval-
uation of the content of food and beverages during all the meals experiences that we
have, and depending on how well trained is this natural sensor, the better are the
chances to qualitatively detect the ingestion of toxic substances present in food and
beverages or any adulteration (deliberate or otherwise) of the food taste. This
recognition of the mammalian taste system is due to receptors in our tongue
mediating sweet, bitter, umami, and sour taste and generating a molecular tools
decoded by our brain to create the basis of the taste [1] and Fig. 1.1 shows a simple
schematic representation of how the mammalian taste system works to decode the
chemical information.
This schematic representation agrees with the common definition of chemical
sensors found in the literature, and introduced by Wolfbeis in 1990 [3], as “a device
comprising a recognition element, a transduction element, and a signal processor.”
This definition shows that artificial chemical sensors mimic the signal processing
flow reported in Fig. 1.1 and can be simplified as Fig. 1.2.
This common sense definition is incomplete as the decoding system or the
computer information processing stage needs to be better defined in order to sim-
ulate how a human goes about reading the information, then decoding the infor-
mation, and then linking some interpretive understanding to the readable output.
Wolfbeis [3] extended the definition of chemical sensors as “small-sized devices
comprising a recognition element, a transduction element, and a signal processor
capable of continuously and reversibly reporting a chemical concentration.” New
requirements appear in this new definition. One is the attribute of reversibility,
which mainly becomes important if the same chemical sensor is required to make

Fig. 1.1 Simple schematic representation of how the chemical information is decoded by the
mammalian taste system, for example, mammalian taste

Fig. 1.2 Schematic view of how the information is decoded by the artificial chemical sensor
system
1 Introduction of Materials Used in Chemical Sensors 3

multiple uninterrupted measurements in a closed-loop system. This reversibility

stops normally due to external interferences that either passivate (or foul) the sensor
or the sensor is used outside its tolerances resulting in damage and/or deactivation.
When there is no requirement for the sensor to be reversible, it becomes a single-use
device. Disposable sensor devices, such as those based on paper (will be discussed
in Chap. 3), are being used more and more to transduce chemical information due
to the raft of different transducer methods and recognition materials that can be
integrated with paper to produce devices to translate chemical information [4, 5]
Based on this definition, these devices could not be called chemical sensors, as well
as devices without a recognition element, like devices to monitor blood oxygena-
tion by reflectometry1 [6]. Referring to the latter as a chemical sensor was con-
troversial at the time as there was not an acceptable definition for the technique
under the umbrella of analytical chemistry techniques, such as for example,
spectroscopy.
Additionally, the attribute “reporting a chemical concentration” in the definition
is equally interesting and the readable output is more apparent. In this case, the
function of the output device (e.g., computer, smartphone, or tablet) is to read the
electrical information and then to process the information to report the actual
concentration of the analyzed chemical species based on a calibration curve, i.e.,
interpolating the unknown signal of the sample to find the chemical concentration
based on plot of the electrical response versus concentration of the standards. With
this in mind and returning to the parallel discussion about natural and artificial
chemical sensors, none of our natural chemical sensing systems returns the actual
concentration value of a given chemical species, for example, caffeine and other
ingredients when we drink coffee, but we know qualitatively that it is coffee and
could even allude to the type of coffee. Additionally, we think in terms of
thresholds of strength and therefore return a semiquantitative measure of that rather
than an actual value. Based on this, the computer-based artificial sensor could also
be used to compare chemical “fingerprints” extracted for the samples, like a
chemical spectrum, and compare this analytically useful information with a data-
base in order to give a yes or no answer, i.e., qualitative information through
discrimination of different types of coffee based on the extracted information or a
qualitative composition analysis. Such devices were first reported in the literature in
1982 [7] and called electronic nose, with the idea to use an array of sensors to
extract chemical information in order to discriminate qualitatively the samples. The
term electronic tongue (Chap. 9), to discriminate liquid samples instead of gas
samples (electronic nose), was introduced in 1996 in the literature [8].
In 1991, IUPAC [9] made a more general definition for chemical sensors as “a
device that transforms chemical information, ranging from concentrations of a
specific sample component to total composition analysis, into an analytically useful
signal.” This definition enhanced the possibilities of chemical sensors as well as
including some devices, reported above, not comprised in the previous definitions
reported here. Hence, a chemical sensor could be defined as a device which
responds to an analyte based on a chemical reaction (or recognition) and can be
used for qualitative or quantitative determinations of the species being analyzed (to
4 W.R. de Araujo et al.

give an analytically useful signal), and this definition could be generalized for an
array of chemical sensors (Chap. 9).
One main information behind the IUPAC definition is the signal transduction.
The chemical transduction occurs by monitoring a physicochemical property of the
analyte that is related to its concentration, like the absorbance or peak current for a
reversible system measured by spectrophotometry or cyclic voltammetry, respec-
tively, that is related with the Beer–Lambert (Eq. 1.1) and Randles–Sevcik
(Eq. 1.2) equations:

A ¼ ebc ð1:1Þ

where A is absorbance (dimensionless), e is the molar absorptivity (L mol−1 cm−1),

b is the path length of the sample (cm)—that is, the path length of the cuvette in
which the sample is contained, and c is the concentration of the analyte in solution
(mol L−1).


Ip ¼ 2:69  105 n3=2 AD1=2
o Co v
1=2
ð1:2Þ

where Ip is peak current (A), n is number of electrons, A is area of the electrode

(cm2), D is diffusion coefficient (cm2 s−1), v is scan rate (V s−1), and Co is bulk
concentration of analyte in solution (mol cm−3).
Mainly, this book will focus on optical and electrochemical measurements to
extract the chemical information by analytical techniques; such extraction methods
will be discussed in the Chap. 2.
The key point for the development of a chemical sensor is related with the
materials used to recognize and translate the chemical information through a
chemical interaction or reaction. However, there is not a quantitative theory which
will describe physical and chemical parameters to obtain the better material that will
recognize the species being analyzed, and the steps to obtain a better material are
largely empirical and could be sometimes compared more like an art.
Notwithstanding this, for biosensors (chemical sensors in which the recognition
system is based on the biochemical or biological mechanism), researchers are
attempting to mimic the enzyme and antibody environments in order to create
artificial enzymes and other biomimetic compounds, such as molecularly imprinted
polymers (Chap. 4) or biomimetic sensors [10]. In this approach, we are trying to
copy nature in order to achieve the same level of recognition and affinity.
Chapters 4–8 will demonstrate the use of new and smart materials as transducing
elements for chemical sensors, summarized in Fig. 1.3. Analytical approaches and
strategies to obtain and extract/convert the chemical information into a readable
signal will be discussed in detail in Chap. 2. Chapter 3 is dedicated to portable
devices (paper-based sensors) with low cost and easy operation mode. Finally,
Chapter 9 demonstrates the use of (electrochemical) sensor arrays combined with
chemometric data processing tools to improve overall sensor performance.
Part of the difficulty to define a chemical sensor and create chemical sensors is
because research must have multidisciplinary collaborations of researchers with
1 Introduction of Materials Used in Chemical Sensors 5

Fig. 1.3 Schematic representation of chemical sensor components focused on transducers’

materials

different complementary expertise such as chemists, electronic and mechanical

engineers, materials scientists, and specialists in chemometrics (to decode the large
amount of information, like the brain does seemingly effortlessly). Hence, the idea
of this book is to pull together all of these professionals with different expertise in
order to showcase the complementarity leading to advances in the development of
different materials for chemical sensing.

References

1. Chandrashekar J, Hoon MA, Ryba NJP, Zuker CS (2006) The receptors and cells for
mammalian taste. Nature 444(7117):288–294. doi:10.1038/nature05401
2. Firestein S (2001) How the olfactory system makes sense of scents. Nature 413(6852):211–
218. doi:10.1038/35093026
3. Wolfbeis OS (1990) Chemical sensors? Survey and trends. Fresenius’ J Anal Chem 337
(5):522–527. doi:10.1007/BF00322857
4. Cate DM, Adkins JA, Mettakoonpitak J, Henry CS (2015) Recent developments in
paper-based microfluidic devices. Anal Chem 87(1):19–41. doi:10.1021/ac503968p
5. Nery EW, Kubota LT (2013) Sensing approaches on paper-based devices: a review. Anal
Bioanal Chem 405(24):7573–7595. doi:10.1007/s00216-013-6911-4
6. Vucea V, Bernard PJ, Sauvageau P, Diaconu V (2011) Blood oxygenation measurements by
multichannel reflectometry on the venous and arterial structures of the retina. Appl Opt 50
(26):5185. doi:10.1364/AO.50.005185
7. Persaud K, Dodd G (1982) Analysis of discrimination mechanisms in the mammalian
olfactory system using a model nose. Nature 299(5881):352–355. doi:10.1038/299352a0
8. Hayashi K, Yamanaka M, Toko K. Yamafuji K (1990) Multichannel taste sensor using lipid
membranes. Sens Actuat B: Chem 2(3):205–213. doi:10.1016/0925-4005(90)85006-K
9. Hulanicki A, Glab S, Ingman F (1991) Chemical sensors: definitions and classification. Pure
Appl Chem 63(9). doi:10.1351/pac199163091247
10. Lee JH, Jin H-E, Desai MS, Ren S, Kim S, Lee S-W (2015) Biomimetic sensor design.
Nanoscale 7(44):18379–18391. doi:10.1039/c5nr05226b
Chapter 2
Information Extraction Techniques
in Chemical Sensing

Thiago Matheus Guimarães Selva, Tiago Luiz Ferreira

and Thiago Regis Longo Cesar Paixão

In order to produce an analytically useful signal, as mentioned in the previous

chapter, we need to transduce chemical information using an instrumental tech-
nique. Numerous analytical chemistry techniques exist for the extraction of
chemical information, e.g. spectrometry, separation techniques coupled with
spectroscopic detection, or electrochemical and other methods. However, mainly
due to the cost and necessity of portability, electrochemical and colorimetric
techniques are frequently used to translate chemical information into a readable
output for the analysts and users in in-field applications of chemical sensors. This
chapter will introduce the concepts involved in these techniques, which are mainly
used to extract information for fabricating chemical sensors.

2.1 Electrochemical Sensors

Essentially, the electrochemical sensors are classified as: conductimetric,

potentiometric, and amperometric or voltammetric sensors.

T.M.G. Selva
Instituto Federal de Educação, Ciência e Tecnologia de Pernambuco,
Avenida Prof. Luiz Freire, 500, 50740-540 Recife, PE, Brazil
T.M.G. Selva
T.R.L.C. Paixão (&)
Departamento de Química Fundamental, Instituto de Química,
Universidade de São Paulo, Avenida Prof. Lineu Prestes, 748,
05508-000 São Paulo, SP, Brazil
e-mail: [email protected]
T.L. Ferreira (&)
Instituto de Ciências Ambientais, Químicas e Farmacêuticas,
Universidade Federal de São Paulo, Rua Prof. Artur Riedel, 275,
09972-270 Diadema, SP, Brazil
e-mail: [email protected]

© Springer International Publishing AG 2017 7

T.R.L.C. Paixão and S.M. Reddy (eds.), Materials for Chemical Sensing,
DOI 10.1007/978-3-319-47835-7_2
8 T.M.G. Selva et al.

2.1.1 Conductimetric Sensors

Conductimetric sensors are based on measuring the ionic conductance of solutions.

This conductance results from the individual contributions of each ion in the
solution; it is therefore a property that does not depend on the specific reaction
levels of an electrode, as opposed to, for example, voltammetric sensors.
Conductimetric sensors can be employed in direct conductometry, where elec-
trolyte concentration is determined by a single conductance measurement, or rel-
ative conductometry, where conductance is monitored during titration, and the end
point is determined from the collected data.
Since these sensors measure electrical conductance arising from all ionic species
present in solution, they do not respond to specific ions. This specificity may be
achieved by a separation technique, such as chromatography or capillary elec-
trophoresis, employing conductimetric sensors for the detection of charged species
of interest.
The conductance of a solution or a solid material can be expressed by Ohm´s
law, Eq. 2.1:

I ¼GE ð2:1Þ

where I is the electric current flowing through a solution or a solid, E is the potential
difference, and G is the conductance. Generally, Ohm´s law is expressed in terms of
resistance, Eq. 2.2:

E ¼RI ð2:2Þ

which leads to the definition of G as the inverse of resistance, Eq. 2.3:

1
G¼ ð2:3Þ
R

The resistance or conductance of a specimen depends on its temperature,

chemical nature, homogeneity, size, and shape. For solutions, the conductance also
depends on the number of ions present.
For a specimen uniform over its whole length:

A
G¼j ð2:4Þ
l

where A is the cross-sectional area of the specimen, and l is its length. The pro-
portionality constant j is called conductivity (Fig. 2.1). The same relation is valid
for a solution between two electrodes (Fig. 2.1).
Experimentally, either the resistance or the conductance of a solution is mea-
sured to determine j. The basic experimentally measured parameter in both cases is
the solution resistance, but modern conductance bridges are calibrated to directly
2 Information Extraction Techniques … 9

Fig. 2.1 Schematic representation of solid and electrolytic conductors

provide conductance read-outs. The experimental set-up is based on a Wheatstone

bridge apparatus.
When studying the conductivity of a solution, it is essential that the concen-
tration remains constant throughout the measurement. Passage of an electric current
through a solution induces chemical reactions at the electrodes, resulting in changes
in the solution concentration. These reactions or their effects must be avoided in
conductivity measurements. This is achieved by changing the direction of the
current, so that the reactions are continually reversed and have no net chemical
effect. Alternating current is generally used, e.g. one having a frequency of
1000 Hz.
The current passing through a solution is created by the movement of ions
contained therein. The amount of ions present is likely to be inversely proportional
to the resistance of the solution and directly proportional to its conductivity,
j. Experiments show that j varies considerably with concentration.
10 T.M.G. Selva et al.

Further discussion and procedures can be found in textbooks on conductimetric

titrations, determination of dissociation constants of weak electrolytes, etc. [1, 2].
This notwithstanding, there are two important cases of interesting applications of
conductimetric sensors: (i) measurement of solution conductance employing elec-
trodes outside the solution (oscillometry) and (ii) measurement of conductance of
an array of modified sensors as an electronic nose.
Oscillometry is often employed for conductance measurements in corrosive
solutions, e.g. ones that could damage the cell electrodes. In order to measure the
conductance of a solution using this procedure, the electrodes are positioned outside
the solution, on the external wall of the conductance cell. To perform these mea-
surements, equipment capable of operating at high frequencies is required (ca.
106 Hz). An interesting application of oscillometry is the use of contactless conduc-
timetric cells in capillary electrophoresis to detect different charged species as they are
separated by electroosmotic flow. In these cases, the conductimetric cell is positioned
on an appropriate portion of the external wall of the capillary [3].
The second case deals with smell identification using an array of interdigitated
electrodes modified with a conducting polymer, where each electrode is modified
with a different polymer. When gaseous molecules are absorbed by the polymer, its
electrical conductivity is affected. Different gases affect the conductivity in different
ways. These signal variations can provide a “digital impression” or “chemical
fingerprint” of the studied vapour [4]. The electronic nose (E-nose) must be
“trained” to recognize different smells through electrical conductivity using
chemometric approaches described in Chap. 9.

2.1.2 Potentiometric Sensors

Potentiometric sensors work by measuring the equilibrium potential (potential of

zero current) of the sensor versus a reference electrode. These potentials are a
function of the activity of the species in solution. The equality of activity and
concentration is reasonable to assume only for dilute solutions [5].
Typical instrumentation for potentiometric measurements includes a reference
electrode and an indicator electrode (potentiometric sensor) connected to a high
output impedance voltmeter (1012 X) [6].
There are two classes of potentiometric sensors: (i) metallic electrodes, i.e.
electrodes that develop a potential determined by redox equilibria (Nernst equation)
at the electrode–solution interface (e.g. platinum electrode) and (ii) ion-selective
electrodes, where the difference of potentials across a membrane is measured,
which is influenced by the activity of the species on either side of the membrane.
The use of metallic electrodes commonly brings poor selectivity if more than one
redox couple is present in solution, since all couples contribute to the overall
equilibrium potential. On the other hand, the potential generated for ion-selective
electrodes (ISEs) is due to a selective interaction between the electrode membrane
and an ion.
2 Information Extraction Techniques … 11

Fig. 2.2 Schematic

representation of a membrane
electrode. Arrows symbolize
the exchange of ions across
the membrane between the
internal and external solutions

ISEs measure the potential difference created by the movement of ions between
an internal and an external solution phase, delimited by the membrane (Fig. 2.2).
The membrane potential, Emembrane, is given by Eq. 2.5:

RT a2
Emembrane ¼ ln ð2:5Þ
z i F a1

where R is the universal gas constant, T is the temperature in Kelvin, F is the

Faraday constant, and a is the activity of an ion i of charge zi. As the activity of the
ion i in the internal solution is constant:

RT
Emembrane ¼ c þ ln a2 ð2:6Þ
zi F

where c is a constant.
The membrane potential is measured by calculating the potential difference
between an internal reference electrode and an external reference electrode. Thus,
the membrane serves as a link between two halves of a concentration cell.
A perfect ISE responds to only one ion in a solution containing “any” ions. This
ideal situation cannot be achieved, particularly when ions with similar properties
are present in solution. The interference effects of other ions depend on their
potentiometric selectivity coefficients, Kij, according to the Nicolsky–Eisenman
Eq. 2.7:
12 T.M.G. Selva et al.

RT  X zi 
Emembrane ¼ c þ ln a2 þ Kij  azj ð2:7Þ
zi F

where j is the interfering species with charge zj.

These selectivity coefficients can be evaluated by a fixed interference method
(varying the primary ion activity at a constant level of interferent) or a separated
method (comparing the response of the electrode in primary ion solution with that
in a solution containing only the interferent ion with the same activity) [5].
Selective electrodes are divided into three classes:
(i) primary ion-selective electrodes;
(ii) compound or multiple-membrane ion-selective electrodes;
(iii) all-solid-state ion-selective electrodes.

2.1.2.1 Glass Electrodes

Glass electrodes were the first ISEs to be developed and are used mainly to measure
pH. Glass is an amorphous solid consisting predominantly of silicates and is per-
meable to H+, Na+, and K+. The composition of glass determines the permeability
to each type of ion, but some interference always occurs.
The glass membrane must be conductive to serve as a potentiometric sensor.
Conduction within the hydrated gel layer involves the movement of H+. Sodium
ions are the charge carriers in the dry interior of the membrane. This sensor
functions by exchange of solution protons with sodium ions in the surface region, to
a depth of ca. 50 nm.
þ þ þ þ
Hsolvent þ Naglass  Hglass þ Nasolvent ð2:8Þ

So, for low proton and high sodium concentrations in solution, this exchange is
not complete and the observed potential is higher (pH is lower) than expected,
according to Eq. 2.9 (Eisenman equation) for the interference of Na+.
In strongly acidic or alkaline solutions, the activity coefficients of H+ and Na+
 X 
Emembrane ¼ c þ 0:0592 log aH þ þ KNa þ  aNa þ ð2:9Þ
Hþ

can significantly depend on the environment, possibly leading to a deviating

potential. These deviations at high activities occur in all ISEs.
In pH measurement, the potential difference between two reference electrodes on
both sides of the glass membrane is monitored. The two electrodes are often
combined with the glass membrane (Fig. 2.3).
It is very important to calibrate the glass electrode prior to measurements due to
the differences between its inner and outer surfaces, which lead to differences in the
monitored potential. This potential contribution (often called asymmetry potential)
2 Information Extraction Techniques … 13

Fig. 2.3 Combined glass electrode for pH sensing

can also change with time when the electrode is used, making periodic calibration
necessary [1, 6, 7].

2.1.2.2 Crystalline Membrane Electrodes

These potentiometric sensors are based on a solid-state crystalline membrane. The

homogeneous membrane is an ionic solid with a low solubility product, and the
sensed ion corresponds to the cationic or anionic constituent of the above mem-
brane. The potential is created by ion exchange between the solution and the surface
of the ionic crystal. Migration of crystal structure defects accounts for the charge
transport through the membrane.
As these sensors respond to both the cation and anion of the solid membrane, it
is expected that they would also be the main interfering species. The electrode is
also sensitive to ions that can bind the membrane components, especially if the
binding products have lower solubility than the membrane material.
Other crystalline membranes, called heterogeneous membranes, are based on an
inert plastic matrix (e.g. PVC, silicone rubber, or conducting epoxy resin) with
incorporated small crystals of the ionic solid.
Generally, this kind of potentiometric sensor does not use an internal reference
electrode, but an ohmic contact [1].
14 T.M.G. Selva et al.

2.1.2.3 Non-crystalline Membrane Electrodes

Non-crystalline membrane electrodes are based on a polymer-supported membrane

containing solvent and an ion exchanger or neutral carrier (commonly a chelating
agent) selective for the species to be determined. Transport across the membrane is
achieved by exchange of the species of interest between adjacent chelating agents
[1].

2.1.2.4 Gas-Sensing Electrodes

These potentiometric sensors are simple ISEs with a second gas-permeable mem-
brane, which allows certain molecules to pass. Usually, a small amount of elec-
trolyte solution is placed between the selective membrane and the outer membrane.
The selective membrane is commonly a pH glass membrane, and the variation of
pH is related to the partial pressure of the gas [1].

2.1.2.5 Potentiometric Enzyme Electrodes

These sensors also have a second membrane, which contains an immobilized

enzyme. Since enzymes are highly specific catalysts, one of the products of an
enzymatic reaction can be monitored and the analyte indirectly determined [1].

2.1.2.6 Ion-Selective Field-Effect Transistors

In order to miniaturize potentiometric sensors achieving reproducible signals with a

high signal/noise ratio, ion-selective field-effect transistors (ISFETs) [8] were
developed using semiconductor transistor technology (Fig. 2.4). The function of a
conventional field-effect transistor is to respond to tiny voltage differences of a
metallic gate between the source and drain, converting them into a low-impedance
output signal (current signal).
In ISFETs, the metallic gate is replaced by an ion-selective membrane, which is
in contact with solution. The drain signal (output) is directly related to the activity
of ions in solution [8].

2.1.3 Voltammetric Sensors

Voltammetric sensors are based on measuring the relationship between the current
and the applied potential. There are two main approaches to carry out voltammetric
experiments: (i) measure the current response as a function of applied potential and
(ii) monitor the potential response as a function of applied current. Most
2 Information Extraction Techniques … 15

Fig. 2.4 ISFET for pH sensing. Reprinted from Jimenez-Jorquera et al. [8]. Copyright (2010),
with the permission from MDPI®

voltammetric sensors are based on potential control. Amperometric sensors are a

special kind of voltammetric sensors, where determination of electroactive species
is performed at constant potential [5, 6, 9, 10].
The instrumentation for voltammetric sensors is more complex than that for
conductimetric and potentiometric sensors. Three electrodes are necessary to avoid
current passage through the reference electrode, which would change its potential.
The current passes through an electrical circuit between the working electrode and
an auxiliary electrode, with the reference electrode used to control the potential of
the working electrode. A potentiostat is necessary to control the applied potential
and register the current at the working electrode. To gain information on
current-controlled experiments and monitor changes in the potential of the working
electrode, a galvanostat is required.
The current of analytical interest in voltammetry is the faradaic current, which is
generated by oxidation or reduction of the analyte at the surface of the working
electrode. Another current, called a capacitive current, interferes with each mea-
surement. For example, when the potentiostat forces the electron transfer for a
reduction process to occur on the working electrode, bringing the potential to more
negative (or less positive) values, the cations in the solution are attracted to the
electrode surface, whereas anions are repelled. This flux of ions and electrons, i.e.
the capacitive current, is not a contribution from the redox reaction and must be
minimized in order to achieve lower voltammetric detection limits.
In voltammetry, the potential excitation signal can be imposed on a working
electrode in different waveforms, with each potential waveform eliciting a char-
acteristic current response (Fig. 2.5).
A classical voltammetry excitation signal is a linear potential scan, where the DC
potential applied to the electrochemical cell varies linearly as a function of time.
16 T.M.G. Selva et al.

Fig. 2.5 Potential excitation waveforms for a linear sweep voltammetry and b pulse voltammetry
c shows the behaviour of faradaic and capacitive currents as a function of time during a potential
pulse

The current that flows in the cell is recorded as a function of time and thus as a
function of the applied potential, resulting in a voltammogram. Among the
parameters that need to be specified to record a voltammogram, the potential sweep
rate is crucial. This parameter controls the slope of the potential variation as a
function of time.
A typical response to a linear potential sweep is a peak-shaped voltammogram.
The current starts to rise when the potential values match those of an electrode
process. This creates a concentration gradient of electroactive species between the
electrode surface and bulk solution, with the lack of electroactive species on the
electrode surface making the current fall.

2.1.3.1 Cyclic Voltammetry

The potential can also be cycled multiple times between two values, e.g. first being
increased linearly and then lowered at the same rate (Fig. 2.6).

Fig. 2.6 Cyclic voltammetry

potential excitation signal
2 Information Extraction Techniques … 17

Fig. 2.7 A typical

peak-shaped cyclic
voltammogram. The
parameters t0, t1, and t2 are
shown in Fig. 2.6

In cyclic voltammetry of reversible systems (i.e. ones with fast electrode kinetics
relative to the potential sweep timescale), the product of initial oxidation or
reduction can be regenerated by reversing the scan direction (Fig. 2.7). The fol-
lowing equation relates the peak current with other parameters of linear sweep
voltammetry:

Ip ¼ 2:69  105 n2 AD2 Cv2

3 1 1
ð2:10Þ

where Ip is the current peak in A, n is the number of electrons transferred in the

electrode process, A is the electrode active area in cm2, D is the diffusion coefficient
of electroactive species in cm2 s−1, C is the concentration of electroactive species in
mol cm−3, and v is the potential scan rate in V s−1.
For reversible systems, (i) the ratio of oxidation and reduction current peak
values (anodic and cathodic current peaks, IPA and IPC, respectively) is close to one
and (ii) the separation between the cathodic and anodic potential peaks (EPC and
EPA, respectively) is equal to 59.0/n mV, or equivalently:

0:0285
EPC ¼ E12  ð2:11Þ
n
0:0285
EPA ¼ E12 þ ð2:12Þ
n
18 T.M.G. Selva et al.

For completely irreversible systems, only the oxidation or reduction process is

detected, with no peak in the reversed sweep. Most of the redox couples are
positioned between the completely reversible and irreversible systems (called
quasi-reversible systems). In these cases, the reverse peak appears, but is smaller
than the forward peak [5, 6, 10].

2.1.3.2 Hydrodynamic Voltammetry

Hydrodynamic electrodes can be employed as voltammetric sensors, subject to

controlled convection imposed by solution or electrode movement. Convection
enhances the mass transport of electroactive species to the electrode surface, so that
the diffusion layer, with a concentration gradient present therein, is thinner than in
the absence of convection. Consequently, the current response is enhanced.
Hydrodynamic electrodes are important voltammetric sensors that operate under
steady-state conditions. For analytical purposes, the sensors with highest sensitivity
are those where a potential corresponding to the limiting current region is applied. If
the rate of convective transport is constant, together with all the other control
parameters, the current response of the electrode is also constant. These electrodes
are usually operated under laminar flow conditions (in the absence of turbulence).
The best-known hydrodynamic electrode is the rotating disc electrode. The limiting
current for this electrode is given by:

IL ¼ 1:554 prnFCD3 t6 x2

2 1 1
ð2:12Þ

where r is the radius of the electrode in cm, F is the Faraday constant, C is the
concentration of electroactive species in mol cm−3, D is the diffusion coefficient in
cm2 s−1, t is the kinematic viscosity, and x is the speed of rotation of the electrode
in Hz [6, 10].
In some situations, electrodes are used in flow systems. There are many elec-
trochemical flow cell detectors based on wall-jet or channel-tube electrodes.
Generally, these cells are designed for chromatography, capillary electrophoresis,
flow injection analysis, or batch injection analysis.

2.1.3.3 Microelectrode Voltammetry

Microelectrodes are electrodes with at least one dimension in the micrometre range.
This minute dimension leads to low capacitive current contributions and the
possibility of registering steady-state currents in a short time (Fig. 2.8).
Microelectrodes have many advantages compared to conventional electrodes: (i) in-
sertion of microelectrodes in places where other electrodes are too large; (ii) high
2 Information Extraction Techniques … 19

Fig. 2.8 A typical

microelectrode cyclic
voltammogram

signal/noise ratio; (iii) possibility of registering voltammograms in highly resistive

media without the addition of an inert electrolyte; and (iv) relative insensitivity to
forced convection of the solution. For a microdisc electrode, the steady-state current
in the limiting current region is given by the following equation [10]:

IL ¼ 4nFDCr ð2:13Þ

2.1.3.4 Pulsed Voltammetric Techniques

Pulse techniques are based on the current response to a sequence of potential steps
in the forward and/or reverse directions. This response is a pulse of current that
decreases with time as the electroactive species is consumed in the region near the
electrode surface.
The registered current has a contribution from both faradaic and capacitive
processes. The capacitive current decreases faster than the faradaic current. Thus,
the current is usually sampled after the capacitive contribution becomes very low.
Pulse widths are adjusted to achieve this condition (Fig. 2.5).
The most frequently used pulse techniques are differential pulse voltammetry
and square wave voltammetry. Conceptually, the two techniques are very similar.
The detection limits are of the order of 10−7 mol L−1 for differential pulse
voltammetry and 10−8 mol L−1 for square wave voltammetry [5, 10].
20 T.M.G. Selva et al.

2.1.3.5 Membrane and Modified Electrodes

There are many situations when controlling the potential is not sufficient to gain
selectivity in voltammetric experiments. Response overlap can occur due to the
proximity of electroactive species potentials or electrode kinetic processes. In some
cases, the current response decreases with time due to blocking of the electrode
surface by strongly adsorbed species. These problems can be circumvented by
using modified electrodes as voltammetric sensors. Usually, the modification of
electrodes brings selectivity by: (i) creation of physical barriers/membranes that
block interfering species or (ii) deposition of material that reacts with the analyte
more selectively or acts as a mediator for electron transfer.
Porous membranes can be used in voltammetric sensors, covering the electrode
surface directly or having a thin layer of separating electrolyte. These membranes
can act as size exclusion separators (blocking larger species like proteins) or as a
gas-permeable membrane (as in the Clark oxygen electrode).
The use of enzymes immobilized on the electrode surface directly or within a
membrane covering the electrode also allows to achieve high specificity. These
biosensors combine electrochemical signal transduction with a biological sensing
component.
In modified electrodes, changes are promoted in the surface layers of the elec-
trode. Alternatively, a new layer at the electrode surface is formed to gain selec-
tivity. The general intention is to enhance or facilitate some electrode processes
while inhibiting other ones. There are many strategies for voltammetric sensor
modification, including adsorption, chemical modification, electrodeposition, and
surface treatment [11].

2.1.3.6 Other Techniques

Important information about the electroactive species, such as the number of

transferred electrons and the diffusion coefficient, can be gained using techniques
such as chronoamperometry (recording current as a function of time) and
coulometry or chronocoulometry (recording charge as a function of time).
In AC voltammetry, a small amplitude sine wave is superimposed on a pro-
grammed potential variation. The perturbation of the system results in current
responses that vary in amplitude and phase angle. The obtained voltammograms
can provide information on the kinetics, and the response can be useful for ana-
lytical determinations. The in-phase and out-of-phase current components are
related to faradaic currents and the separation of charging currents, respectively.
Generally, this technique allows to achieve low detection limits, but processes with
slow electrode kinetics result in the loss of sensitivity [5, 6, 10].
2 Information Extraction Techniques … 21

2.2 Colorimetric Sensors and Strategies for Extracting

Colorimetric Information

In its early years, chemical analysis, either qualitative or quantitative, was per-
formed using reagent-based colorimetric tests. After the rise of the instrumental
methods of analysis, quantitative titration was practically abandoned. On the other
hand, qualitative and/or semi-quantitative spot tests are still popular [12], with the
use of pH colour-fixed indicators being a popular example.
The most typical instrumental way to obtain information on solution colour is to
use a spectrophotometer or a photometer operating in the visible spectral range
(400–800 nm), based on transmittance/absorbance measurement. Briefly, in
absorption spectroscopy, the radiation intensity from a source of light at a specific
wavelength is attenuated by passing through a coloured solution in a cuvette that is
situated between the light source and a detector. The mathematical description of
this process is known as the Beer–Lambert law (Eqs. 2.14 and 2.15):

I
T¼ ð2:14Þ
I0

I
A ¼  log ð2:15Þ
I0

where T is the transmittance of the radiation passing through the solution, I0 is the
total intensity of the radiation source, I is the intensity of the radiation after passing
through the solution, and A is the solution absorbance. The choice of wavelength
(k) to monitor coloured species is determined from the visible absorption spectra
plot (A vs. k), which may be performed sweeping the wavelengths in the visible
spectral range. Often, the maximum absorption wavelength (kmax) is used, mainly
because it provides higher sensitivity. Absorption is proportional to the concen-
tration and length of the cuvette optical path, as can be inferred from Eq. 2.16.
Longer beam paths of the incident radiation increase the probability of the coloured
species absorbing a part of this radiation:

A ¼ ebc ð2:16Þ

where ɛ is the molar absorption in cm−1 mol−1 L, b is the optical path in cm, and c is
the concentration of the coloured species in mol L−1. Figure 2.9 shows a simple
scheme of the instrumentation used.
A classical application of spectrophotometric measurements is the indirect colori-
metric quantification of glucose in biological fluids. This method is based on the
reaction of glucose with glucose oxidase to produce hydrogen peroxide. The latter
reacts with a chromogenic oxygen acceptor in the presence of peroxidase, producing a
chromogenic species, which is spectrophotometrically monitored at 460 nm [8].
Even after the rise of portable spectrophotometers and the decrease of their price,
the search for alternative ways of analytical colorimetric measurements continued.
22 T.M.G. Selva et al.

Fig. 2.9 a Scheme of the visible radiation absorption process. b Resultant absorbance (A) versus
wavelength (k) plot

Recently, various research groups have used desktop or portable scanners [14],
digital cameras, webcams [15], cell phones, and smartphones [16] to collect ana-
lytical information based on colour measurement. In these methods, the reflection
of the system was used, instead of the traditional way of measuring transmittance/
absorbance. This is, therefore, an advantage, allowing the analysis of turbid samples
[17].
Among the colour systems used, the most common way to process analytical data
from digital images is based on the RGB colour model, which is used in computer
screens and utilizes the primary colours of light. This is an additive model, and the
name comes from the three primary colours: red (R), green (G), and blue (B), also
called channels. In computers, each of these three channels is represented by an integer
number from 0 to 255, and each combination represents a particular colour, making it
possible to represent more than 16 million combinations (2563 = 16,777,216). White
colour, for example, is obtained when all three channels have values of 255, while
black corresponds to all channels equal to zero. Some researchers use the H parameter
(hue) of the HSV colour space [18, 19], which can be correlated with the RGB system,
to monitor a specific coloured reaction. The extraction of the RGB code of a digital
image can be performed by software [20, 21] or by a smartphone app. These apps may
be home-made [22] or downloaded from popular virtual stores [23] of the operating
system, such as iOS and Android. In addition, it is possible to convert the RGB colour
model to other models, such as grayscale and CMYK (cyan, magenta, yellow, and
black/key). The CMYK colour model, used for colour printing, is a subtractive system
and uses secondary colours created by mixing two primary ones (RGB model). For
example, mixing red and blue colours gives magenta [24]. An alternative way to
record analytical colour information without the use of any sophisticated instru-
mentation was proposed by Cate et al. [25]. The authors explored the microfluidic
properties offilter paper to propose a paper-based analytical device (µPAD), shaped as
a strip and limited by a wax barrier, which was spotted with reagents giving a coloured
reaction. A simple measurement of the reaction extent distance, using a ruler, was
correlated with the concentration of the analyte [25]. The use of Google Glass to
perform diagnostic colorimetric tests has also been proposed [26].
Various ways of treating the RGB data in a digital image are described in the
literature. A widespread method is to find a channel (R, G, or B) that correlates
2 Information Extraction Techniques … 23

linearly with the analyte concentration [26, 27, 28]. However, the use of grayscale
and CMYK models has also been reported [16] in quantitative methods. Instead of
using the classical above-mentioned method of glucose quantification, it was
demonstrated that the information obtained from digital images can also be used for
the same purpose. Researchers adopted the µPAD approach using a camera phone
and a portable scanner as a detection system for the development of a colorimetric
spot test to quantify glucose and protein content in synthetic urine [16]. Focusing on
the glucose test, the method is similar to the classical one, i.e. the glucose is
oxidized by glucose oxidase to gluconic acid, generating hydrogen peroxide. The
peroxide is promptly reduced by iodide (catalysed by horseradish peroxidase) to
give iodine, and the colour of the paper spot test region changes from colourless to
brown. All reagents used in the assays were prespotted on the respective test zone
of the paper, and the colour digital image, taken after exposing the µPAD to a
synthetic urine sample, was converted to grayscale by software, with the mean
intensity taken as an analytical signal [16].
Others approaches are based on the use of colour data from digital images cou-
pled with widespread chemometrical tools [29], such as principal component anal-
ysis (PCA) [20, 22, 23, 30, 31], hierarchical cluster analysis (HCA) [22, 30, 32, 33],
principal component regression (PCR) [21], partial least squares (PLS) [17, 29], soft
independent modelling by class analogy (SIMCA) [35–38], linear discriminant
analysis (LDA) [34–37], k-nearest neighbour (k-NN) [31], and artificial neural
network (ANN) [18, 19]. These mathematical tools can replace the analysis per-
formed by naked eye, avoiding major errors, especially when the method is based on
spot tests. Illustrating the application of chemometrical tools and the use of colour
images in analysis, Huang et al. [39] reported an array of colorimetric sensors, called
an electronic nose (E-nose), based on a reversed-phase silica gel plate overprinted
with specific dyes for the evaluation of fish freshness, as shown in Fig. 2.10.

Fig. 2.10 Profile of the

colorimetric sensor array
(E-nose) for the evaluation of
fish freshness prior to being
exposed to analytes.
Reprinted from Huang et al.
[39]. Copyright (2011), with
the permission from Elsevier
24 T.M.G. Selva et al.

Fig. 2.11 Colour profiles of the E-nose after exposure to a fish sample at a day 1, b day 4, and
c day 7. Reprinted from Huang et al. [39]. Copyright (2011), with the permission from Elsevier

The dyes were chosen with respect to the main volatile organic compounds
(VOCs) released during the fish spoilage process. When these VOCs react with the
dyes during the time of exposure, a change in the colour profile results, as shown in
Fig. 2.11.
The sensor array was placed in a reaction chamber containing VOCs released by
the fish sample, and images of the sensor were taken using a desktop scanner. The
images were recorded before and after exposure to VOCs at different times. Next, a
difference map of the images was obtained by subtracting the images after exposure
to VOCs from those taken before the exposure and converting the result into RGB
values. These values were used as input data to perform principal component
analysis. The authors observed that the first-day data were dispersed on a smaller
area when compared to that for the second through fifth days. This observation is
probably due to the fact that on the first-day, the fish was still fresh, while between
the second and fifth days, the fish released new VOCs owing to the deterioration
process. It was also noted that the data group for the sixth and seventh days of the
storage is easily separable from other data, once the fish begins to release a large
amount of biogenic amines. Thus, the authors used the data obtained by PCA to
build a model with a neural network approach, obtaining an accuracy about 87.5 %
[39]. Although the authors used a desktop scanner to collect RGB data, it is easily
possible to use a smartphone, which is a ubiquitous gadget. The use of a
2 Information Extraction Techniques … 25

smartphone with an app that could readily perform a scan of the colorimetric sensor
array placed in the package of the product stored in a supermarket, for example, was
proposed by Bueno et al. [22]. This can help the consumer avoid buying a spoiled
product.
In contrast, the use of the above-mentioned devices to take digital images in
order to perform analytical inferences about the subject matter is not so simple and
shows some drawbacks. Different devices can produce different images, i.e. dif-
ferent RGB values for the same region, which is not always easy to spot with the
naked eye [40, 41]. The resolution of the device, the technology used to build it, the
shooting distance, and the imaging angle are some examples of factors that may
contribute to providing different RGB values for the same target. Besides, lumi-
nosity control at the moment a digital image is taken is of paramount importance
and is another factor of concern that influences RGB values [28, 30, 42]. However,
researchers have designed strategies to circumvent this problem. Usually, a black
chamber with controlled luminosity and a fixed position of device components is
used [22, 28, 30, 43] to capture the images before (blank) and after exposure to the
analyte, and thus, the images are subtracted directly (difference map). Approaches
that try to compensate for the influence of the illumination conditions, cell
phone/smartphone model, distance, and angle to the substrate, without the use of an
extra accessory, have been proposed [33, 44, 45].

References

1. Mendham J, Denney RC, Barnes JD, Thomas MJK (1979) Vogel’s quantitative chemical
analysis, 5th edn. Wiley, New York
2. Pungor E (1965) Oscillometry and conductometry, 1st edn. Pergamon Press, Budapest
3. Fracassi da Silva JA, do Lago CL (1998) An oscillometric detector for capillary
electrophoresis. Anal Chem 70:4339–4343. doi:10.1021/ac980185g
4. Cordeiro JR, Martinez MIV, Li RWC et al (2012) Identification of four wood species by an
electronic nose and by LIBS. Int J Electrochem 2012:1–5. doi:10.1155/2012/563939
5. Brett CMA, Brett AMO (1993) Electrochemistry: principles, methods, and applications, 1st
edn. Oxford University Press, New York
6. Wang J (2006) Analytical electrochemistry, 3rd edn. WILEY-VCH, New Jersey
7. Amemiya S (2007) Potentiometric Ion-Selective Electrodes. In: Zoski CG (ed) Handbook of
electrochemitry, 1st edn. Elsevier Science, Amsterdam, p 935
8. Jimenez-Jorquera C, Orozco J, Baldi A (2010) ISFET based microsensors for environmental
monitoring. Sensors (Basel) 10:61–83. doi:10.3390/s100100061
9. Brett CMA, BRETT AMO (1998) Electroanalysis, 1st edn. Oxford University Press, New York
10. Bard AJ, Faulkner LR (1990) Electrochemical methods: Fundamentals and applications.
Wiley, New York
26 T.M.G. Selva et al.

11. Edwards GA, Bergren AJ, Porter MD (2007) Chemically Modified Electrodes. In: Zoski CG
(ed) Handbook of electrochemistry, 1st edn. Elsevier Science, Amsterdam, p 935
12. Byrne L, Barker J, Pennarun-Thomas G et al (2000) Digital imaging as a detector for generic
analytical measurements. TrAC Trends Anal Chem. 19:517–522. doi:10.1016/S0165-9936
(00)00019-4
13. Bergmeyer H-U, Bernt E (1965) Determination with glucose oxidase and peroxidase. In:
Bergmeyer H-U (ed) Methods of enzymatic analysis. Elsevier, London, pp 123–130
14. Carey JR, Suslick KS, Hulkower KI et al (2011) Rapid identification of bacteria with a
disposable colorimetric sensing array. J Am Chem Soc 133:7571–7576. doi:10.1021/
ja201634d
15. Gaiao EN, Martins VL, Lyra W da S, et al (2006) Digital image-based titrations. Anal Chim
Acta 570:283–290. doi:10.1016/j.aca.2006.04.048
16. Martinez AW, Phillips ST, Carrilho E et al (2008) Simple telemedicine for developing
regions: camera phones and paper-based microfluidic devices for real-time, off-site diagnosis.
Anal Chem 80:3699–3707. doi:10.1021/ac800112r.Simple
17. Abbaspour A, TalebanpourBayat E, Mirahmadi E (2012) A reliable and budget-friendly,
solution-based analysis of multiple analytes of boiler water based on reflection scanometry.
Anal Methods 4:1968. doi:10.1039/c2ay05799a
18. Ariza-Avidad M, Salinas-Castillo A, Cuéllar MP et al (2014) Printed disposable colorimetric
array for metal ion discrimination. Anal Chem 86:8634–8641. doi:10.1021/ac501670f
19. Ariza-Avidad M, Cuellar MP, Salinas-Castillo A et al (2013) Feasibility of the use of
disposable optical tongue based on neural networks for heavy metal identification and
determination. Anal Chim Acta 783:56–64. doi:10.1016/j.aca.2013.04.035
20. San Park T, Baynes C, Cho S-I, Yoon J-Y (2014) Paper microfluidics for red wine tasting.
RSC Adv 4:24356–24362. doi:10.1039/C4RA01471E
21. Firdaus ML, Alwi W, Trinoveldi F et al (2014) Determination of chromium and iron using
digital image-based colorimetry. Procedia Environ Sci 20:298–304. doi:10.1016/j.proenv.
2014.03.037
22. Bueno L, Meloni GN, Reddy SM, Paixão TRLC (2015) Use of plastic-based analytical
device, smartphone and chemometric tools to discriminate amines. RSC Adv 5:20148–20154.
doi:10.1039/C5RA01822F
23. Vallejos S, Muñoz A, Ibeas S et al (2013) Solid sensory polymer substrates for the
quantification of iron in blood, wine and water by a scalable RGB technique. J Mater Chem A
1:15435. doi:10.1039/c3ta12703f
24. Galer M, Horvat L (2005) Digital imaging, 3rd edn. Elsevier, Oxford
25. Cate DM, Dungchai W, Cunningham JC et al (2013) Simple, distance-based measurement for
paper analytical devices. Lab Chip 13:2397. doi:10.1039/c3lc50072a
26. Feng S, Caire R, Cortazar B, et al (2014) Terms of use immunochromatographic diagnostic
test analysis using Google Glass. ACS Nano 3069–3079
27. Abbaspour A, Mehrgardi MA, Noori A et al (2006) Speciation of iron(II), iron(III) and
full-range pH monitoring using paptode: A simple colorimetric method as an appropriate
alternative for optodes. Sens Actuators B Chem 113:857–865. doi:10.1016/j.snb.2005.03.119
28. Benedetti LPDS, dos Santos VB, Silva TA et al (2015) A digital image-based method
employing a spot-test for quantification of ethanol in drinks. Anal Methods 7:4138–4144.
doi:10.1039/C5AY00529A
29. Diehl KL, Anslyn EV (2013) Array sensing using optical methods for detection of chemical
and biological hazards. Chem Soc Rev 42:8596. doi:10.1039/c3cs60136f
30. Salles MO, Meloni GN, de Araujo WR, Paixão TRLC (2014) Explosive colorimetric
discrimination using a smartphone, paper device and chemometrical approach. Anal Methods
6:2047–2052. doi:10.1039/C3AY41727A
31. Bueno L, Cottell A, Reddy SM, Paixão TRLC (2015) Coupling dye-integrated polymeric
membranes with smartphone detection to classify bacteria. RSC Adv 5:97962–97965. doi:10.
1039/C5RA19874G
2 Information Extraction Techniques … 27

32. Suslick KS, Rakow NA, Sen A (2004) Colorimetric sensor arrays for molecular recognition.
Tetrahedron 60:11133–11138. doi:10.1016/j.tet.2004.09.007
33. Jia M-Y, Wu Q, Li H et al (2015) The calibration of cellphone camera-based colorimetric
sensor array and its application in the determination of glucose in urine. Biosens Bioelectron
74:1029–1037. doi:10.1016/j.bios.2015.07.072
34. Garrido-Novell C, Pérez-Marin D, Amigo JM et al (2012) Grading and color evolution of
apples using RGB and hyperspectral imaging vision cameras. J. Food Eng 113:281–288.
doi:10.1016/j.jfoodeng.2012.05.038
35. Costa GB, Fernandes DDS, Almeida VE et al (2015) Digital image-based classification of
biodiesel. Talanta 139:50–55. doi:10.1016/j.talanta.2015.02.043
36. Diniz PHGD, Dantas HV, Melo KDT et al (2012) Using a simple digital camera and
SPA-LDA modeling to screen teas. Anal Methods 4:2648. doi:10.1039/c2ay25481f
37. Almeida VE, da Costa GB, de Sousa Fernandes DD et al (2014) Using color histograms and
SPA-LDA to classify bacteria. Anal Bioanal Chem. 406:5989–5995. doi:10.1007/s00216-
014-8015-1
38. Pedrosa VA, Caetano J, Machado SAS, Bertotti M (2008) Determination of parathion and
carbaryl pesticides in water and food samples using a self assembled monolayer/
acetylcholinesterase electrochemical biosensor. Sensors 8:4600–4610. doi:10.3390/
s8084600
39. Huang X, Xin J, Zhao J (2011) A novel technique for rapid evaluation of fish freshness using
colorimetric sensor array. J. Food Eng. 105:632–637. doi:10.1016/j.jfoodeng.2011.03.034
40. Dorsey J, Rushmeier H, Sillion F (2008) Digital modeling of material appearance, 1st edn.
Elsevier, London
41. Lasarte M De, Vilaseca M, Pujol J, et al (2005) Influence of technology, color architecture
and bit-depth of optoelectronic imaging sensors used as color measurement instruments. In:
Congress of the International Colour Association. pp 1203–1206
42. García A, Erenas MM, Marinetto ED et al (2011) Mobile phone platform as portable chemical
analyzer. Sensors Actuators B Chem 156:350–359. doi:10.1016/j.snb.2011.04.045
43. Steiner M-S, Meier RJ, Duerkop A, Wolfbeis OS (2010) Chromogenic sensing of biogenic
amines using a chameleon probe and the red—green—blue readout of digital camera images.
Anal Chem 82:8402–8405. doi:10.1021/ac102029j
44. Hong J Il, Chang B-Y (2014) Development of the smartphone-based colorimetry for
multi-analyte sensing arrays. Lab Chip 14:1725–1732. doi:10.1039/c3lc51451j
45. Gupta R, Reifenberger RG, Kulkarni GU (2014) Cellphone camera imaging of a periodically
patterned chip as a potential method for point-of-care diagnostics. ACS Appl Mater Interfaces
6:3923–3929. doi:10.1021/am4050426
Chapter 3
(Bio)Chemical Sensors Based on Paper

Nipapan Ruecha, Kentaro Yamada, Koji Suzuki and Daniel Citterio

3.1 What Is a Paper-Based (Bio)Chemical Sensor?

In this chapter, the authors use the term “(bio)chemical paper-based sensor” in a
rather broad meaning. After all, the literature does not provide a unique definition for
a “chemical sensor,” and different people have different understanding of this
expression, as discussed in Chap. 1. A very widely used definition among specialists
is the so-called Cambridge definition, according to which “chemical sensors are
miniaturized devices that can deliver real-time and online information on the pres-
ence of specific compounds or ions in even complex samples [1].” The terms “real
time” and “online” are often interpreted as a chemical sensor being fully reversible or
in other words, able to indicate both increasing and decreasing concentration
changes in continuous measurement mode. This situation is very rarely the case for
the (bio)chemical sensors based on paper materials discussed in this chapter.
Actually, one of the motivations, if not even the main purpose for using paper in
connection with chemical sensing, is the low cost and easy disposability of this
material. A chemical sensor made from paper is therefore primarily intended for
single use, and with few exceptions not required to and not able to work in a
reversible manner. As a consequence, the examples of paper-based chemical sensors
discussed here should be regarded as sensors under a wider definition, as for example
given by IUPAC in 1991, where “a chemical sensor is a device that transforms
chemical information, ranging from the concentration of a specific sample compo-
nent to total composition analysis, into an analytically useful signal [2].” The devices
presented in this chapter provide their users with analytically useful information,
ideally in the simplest way requiring no other step than the introduction of a sample
into the device. Selectivity, a key characteristic of any sensor, is often achieved like

N. Ruecha
K. Yamada
K. Suzuki
D. Citterio (&)

Department of Applied Chemistry, Keio University,
3-14-1 Hiyoshi, Kohoku-ku, Yokohama 223-8522, Japan
e-mail: [email protected]

© Springer International Publishing AG 2017 29

T.R.L.C. Paixão and S.M. Reddy (eds.), Materials for Chemical Sensing,
DOI 10.1007/978-3-319-47835-7_3
30 N. Ruecha et al.

in a “conventional (bio)chemical sensor” by selective molecular recognition.

Alternatively, the influence of potential interfering compounds is eliminated by
on-device separation or masking, among others, without the need of user
intervention.
The primary role of paper is limited to being a versatile matrix/substrate for the
immobilization of sensing components. However, in the case of microfluidics-based
paper sensors, liquid transport becomes another function where paper demonstrates
its strength in terms of material properties.
A large variety of signal transduction principles is known for (bio)chemical
sensors in general, including optical, electrochemical, thermal, magnetic, and
mechanical methods [3]. This is not very different from paper-based sensors,
although not all of the principles listed might be implemented on paper sensor
platforms. Similar to their non-paper counterparts, optical and electrochemical
transduction are by far the most dominant approaches for paper sensors.

3.2 Short History of Paper-Based (Bio)Chemical Sensors

Although there might be earlier examples that are unaccounted for, probably the
first confirmed chemical sensor based on paper was the litmus paper allowing to test
for acidity or basicity of a solution, assumed to have been introduced in the sev-
enteenth century by the Irish chemist Robert Boyle [4]. Surprisingly, this simple pH
sensor, together with its numerous derivatives using pH-indicator impregnated
papers, is probably still the most widely used paper-made chemical sensor today.
The mid-nineteenth century brought the first reports of urine test strips—another
example of historical paper-based sensing technology still used in modern clinical
chemistry, although with different underlying chemical principles. In 1850, a
French scientist described a merino tissue-based sensor to test the presence of sugar
in urine [5]. Approximately 30 years later, Oliver described how he found it useful
for the medical practitioner to eliminate the need of handling liquid reagents in
urine tests for albumin, sugar, and acidity by predepositing and drying the required
reagents on pieces of filter paper, linen, or other similar fabrics [6]. He continues
that these pretreated papers allow for the quantitative bedside monitoring of urine
parameters by a simple direct color read off, implicitly stating that simplicity and
on-site usability are a motivation to work with paper sensors. The first paper-based
test strips for urine glucose relying on deposited enzymes can be regarded as the
start to paper-based biosensing [7, 8]. They have been introduced to overcome the
selectivity problems encountered with the previous non-enzymatic versions.
Another important step in the historical development of paper-based biosensors
is the introduction of the lateral flow immunochromatographic strip concept in
1982, where an antigen has been directly deposited onto a nitrocellulose membrane,
a “paper-like” material, allowing for the simple visual inspection of the presence of
specific antibodies in a sample [9]. As with other paper-based technologies intro-
duced in this section, this working principle is still in widespread use nowadays,
3 (Bio)Chemical Sensors Based on Paper 31

especially in the form of over-the-counter pregnancy test kits for the detection of
human chorionic gonadotropin (hCG) in urine.
From the 1980s onward, no revolutionary development seemed to have occurred
in terms of paper-based (bio)chemical sensing approaches. Nevertheless, an
increasing number of simple colorimetric test paper assays for a variety of analytes
became commercially available, as for example represented by the Merckoquant®
series of products, clearly indicating the users’ needs for simple and disposable
sensing devices.
It was only in 2007, when Whitesides et al. presented an idea of using paper as
the basis for a quite different type of (bio)chemical sensors, by introducing
microfluidic structures on the surface of filter paper [10]. Although strictly not new
in terms of technology—Clegg and Müller in 1949 were the first to demonstrate the
usefulness of paper with a fluidic channel in the context of paper chromatography
[11]—microfluidically patterned paper substrates opened up a large new area of
sensing applications that have previously exclusively been available to microfluidic
devices made from glass or synthetic polymeric materials. The combination of
paper substrates with microfluidic structures resulted in a new type of (bio)chemical
sensors that have been named as “microfluidic paper-based analytical devices,” or
in short µPADs [12]. The development of µPADs has become the objective of a
rapidly growing new research area of analytical chemistry. Just 10 years ago,
probably only few people would have believed that paper could be rediscovered as
a valuable material stimulating research activities of an increasing number of
analytical chemists. At least in terms of research efforts, µPADs have evolved into
the dominant type of (bio)chemical sensing devices involving paper. As a conse-
quence, µPADs are the major type of paper-based sensors discussed in the fol-
lowing sections of this chapter.

3.3 Paper Basics

Paper is composed of cellulose (Fig. 3.1), a biopolymer consisting of D-glucose

units. Cellulose is the worldwide most abundant biopolymer, with wood being its
major natural source, containing about 40–50 % of cellulose besides hemicellulose,
lignin, and other compounds, many of them being removed in the papermaking
process in order to obtain high-grade materials [13]. Cotton, another important
natural source of cellulose fibers, contains about 95 % of cellulose and only minor
amounts of other compounds. Therefore, cotton is the raw material of choice, when
it comes to the fabrication of papers where chemical purity is of particular high
importance, such as in the case of laboratory filter paper, which is most widely
applied for paper-based (bio)chemical sensors. In brief, paper is fabricated by
passing an aqueous suspension of cellulose fibers through a screen allowing the
drainage of water, followed by pressing and drying [14]. This process results in
sheets consisting of porous networks of randomly interwoven fibers.
32 N. Ruecha et al.

Fig. 3.1 Chemical structure of cellulose and potential hydroxyl group activation pathways for
covalent attachment of amino-residue carrying compounds. Adapted with the permission from
[20]. Copyright© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

The chemical properties of paper are generally determined by those of cellulose,

although chemical modifications, some of them discussed in the next paragraph, are
common during papermaking. Cellulose as a polysaccharide has abundant hydroxyl
groups (–OH), making paper a hydrophilic material. In addition, the cellulose
backbone in paper also carries a low amount of carboxyl groups (–COOH) origi-
nating from the oxidation of primary alcohol groups during the papermaking pro-
cess [15]. The resulting pH-dependent slightly anionic nature of paper becomes
relevant in the case of working with positively charged sensing compounds or
analytes. While nonionic or anionic compounds are less affected, positively charged
compounds are electrostatically adsorbed to paper surfaces [16]. Proteins can
electrostatically interact with paper surfaces through their cationic regions, even if
they carry no net ionic charges.
During the manufacturing of paper, a variety of process chemicals are used,
which can be roughly divided into two classes: compounds supporting the paper-
making process itself and additives used to enhance the performance and func-
tionality of the final paper product [17]. When it comes to the use of paper for
(bio)chemical sensing, all of these chemicals have to be regarded as potentially
detrimental to the performance of a sensor. Optical brighteners added to improve
the whiteness of paper, for example, can result in interference in cases where the
sensor signal detection relies on fluorescence. Calcium carbonate added as so-called
filler to printing and office papers to enhance physical or optical properties, as well
3 (Bio)Chemical Sensors Based on Paper 33

as to facilitate printability, can result in an unwanted carbonate buffer function that

has to be accounted for when using such a paper for a (bio)chemical sensor [15].
Hydrophobic sizing agents used to induce a certain degree of hydrophobicity can
prevent the penetration of aqueous liquids into the bulk of the paper. It has also
been reported that a high degree of sizing promoted the loss of enzyme activity in
the case of horseradish peroxidase [15]. What makes the situation especially dif-
ficult is the fact that the numerous varieties of commercially available papers are
mostly “black box” systems to the researcher developing paper-based sensors, since
papermakers are generally reluctant to disclose details. It is therefore not surprising
that aside from the immunochromatographic strips relying on nitrocellulose
membranes and the exception of very few other cases, cellulosic laboratory filter
paper with maker-guaranteed purity is by far the most widely applied paper material
for (bio)chemical sensing purposes. In laboratory use filter papers, functionality-
enhancing additives are mostly absent, and the few impurities found in cellulose
from cotton sources are completely removed. Even if researchers mostly limit
themselves to the use of filter papers, this material is still available in a surprising
variety differing in characteristics such as thickness, pore size, basis weight, and
wet strength, as a simple look at a manufacturer’s catalog reveals.

3.4 Motivations for Using Paper in (Bio)Chemical Sensors

In all of the previously mentioned historical approaches, paper or similar porous

materials composed of interwoven fibers have been selected as a readily available
substrate for the temporary fixation and dry storage of reagents normally applied as
solutions. The microporous nature of paper resulting in a comparably large surface-
to-volume ratio is an important physical parameter in this context. Signal detection
has originally been exclusively relying on colorimetric methods by simple visual
inspection or comparison with a reference color chart, making the fact of paper
being white a significant advantage. Scientists involved in the early development of
paper-based devices might not necessarily have been aware of these advantageous
characteristics of paper. Probably, there was simply no other material available at
the time, let alone analytical instruments.
Nowadays, an almost incomprehensible choice of specifically engineered syn-
thetic organic, inorganic or hybrid materials is available and many of those are
applied as (bio)chemical sensor matrices. Why would one want to revert to a rather
old and presumably low-tech material such as an ordinary laboratory filter paper?
As early as in Oliver’s work published in 1883, simplicity and on-site applicability
of paper-based tests in contrast to solution-based assays have been pointed out as
positive features [6], and those still remain important driving forces in the devel-
opment of current paper-based (bio)chemical sensors. At present, however, the
primary motivation for “ignoring” other materials in favor of paper is
the requirement for a low cost, abundantly available, and disposable material, with
34 N. Ruecha et al.

the cost factor being the most important one [18]. The reason for this paradigm shift
is a growing demand driven by societal challenges for extremely low cost, reliable,
disposable, and simple-to-use analytical devices that are applicable in low-resource
settings, such as they are typically met during health care in developing countries,
but also in home healthcare situations in industrialized nations, among others. In
such places, financial resources and technical equipment are scarce, and trained
experts are unavailable, making it essential to have the most low cost, simple, and
infrastructure-independent technology at hand.
The physical–chemical properties of paper bring along a number of “positive
side effects” that are unrelated to the cost factor. The most advantageous point in the
case of application to microfluidic sensing devices is the ability of paper to spon-
taneously transport aqueous liquids by capillary flow [19]. This functionality is not
so easily achieved by more “high-tech” materials. Additionally, the
already-mentioned large surface-to-volume ratio allows for the immobilization of
various reagents not only by simple physical entrapment or adsorption, but also by
the possibility of covalent attachment. Although not sufficiently reactive under mild
conditions, there exist multiple strategies to activate cellulosic hydroxyl groups, as
for example outlined in Fig. 3.1 for the case of reacting with amino-residue car-
rying compounds.
Last but not least, many kinds of paper substrate modifications required to obtain
(bio)chemical sensing devices are achievable by printing methods, which is not
surprising given that the major non-chemistry-related use of paper is nowadays as
printing medium. Almost any thinkable kind of stamping and printing method has
been reported, in particular for the formation of microfluidic structures on paper
surfaces for µPADs. Often used technologies include screen printing [21, 22], inkjet
printing [23–27], wax printing [28–30], and flexographic printing [31, 32], among
others. Inkjet printing is the only method that enables not only the microfluidic
patterning of the paper surface, but also the deposition of all required sensing
reagents. In some cases, complete (bio)chemical sensing devices have been
obtained by inkjet printing alone [23, 25, 33, 34]. It would go beyond the scope of
this chapter to introduce and discuss all of the µPAD microfluidic patterning
methods in detail. The reader is referred to the above-listed references or some
recent review articles covering the field [20, 35–37].
The possibility to fabricate (bio)chemical sensors entirely by printing methods
that are compatible with roll-to-roll processes is a big advantage for large-scale
mass production [31], additionally contributing to cost reduction. Finally, it is
emphasized here one more time that despite many interesting and advantageous
features accompanying the use of paper in (bio)chemical sensors, cost and not
primarily performance remains the dominant factor for choosing paper over alter-
native substrate materials. Nevertheless, as the following sections with specific
examples demonstrate, paper-based (bio)chemical sensing has become an estab-
lished method in analytical chemical research activities with some achievements
that can compete with more sophisticated sensing technologies.
3 (Bio)Chemical Sensors Based on Paper 35

3.5 Optical Sensors

3.5.1 Colorimetric Signal Transduction

Colorimetric signaling has been undoubtedly one of the most familiar detection
systems in classical chemical analysis. Accordingly, chemistries for colorimetric
detection cover a wide range of target analytes including ions, small molecules,
biomolecules, and bacteria, to name just a few examples. Aside from abundant
detection reagents being available, the ease of signal detection (i.e., observable even
by the naked eye) goes well along with the philosophy of paper-based sensors (low
cost and user-friendly), and thus, a vast number of practical paper devices have
been elaborated based on colorimetry. Actually, the number of paper-based (bio)-
chemical sensors relying on colorimetric signal transduction has grown so large that
it is impossible to discuss them all within this chapter. Alternatively, a few
examples are used to illustrate some important issues related to the combination of
sensing reagents with paper matrices. These include the immobilization and storage
of sensing reagents, as well as special functions achieved by using paper as the
sensor matrix.
A variety of classical chromogenic methods established in solution, such as
those based on small-molecule indicators (e.g., pH-indicator), enzymatic activity
assays (e.g., catalytic oxidation using oxidase and peroxidase enzymes), and
nanoparticles (e.g., aggregation/dispersion of gold nanoparticles), have successfully
been transferred onto paper matrices. However, reliable quantitative sensing devi-
ces are not achieved by merely depositing relevant assay components on a paper
substrate.
For example, water-soluble colorimetric indicators simply deposited on a paper
matrix tend to move with the flow of the introduced aqueous sample. This phe-
nomenon results in inhomogeneous color appearance after sample introduction
(Fig. 3.2), making visual judgment or precise analysis difficult [35, 38, 39]. Most
small-molecule indicators are available in the form of sodium sulfonate salts and
exhibit excellent water solubility, but this desirable characteristic for solution-based
assays becomes an obstacle when it comes to adaptation to paper substrates. The
use of indicators with poor water solubility applied to the paper as solution in an
organic solvent may seem to be a straightforward solution to this issue. However,
the use of such compounds together with volatile organic solvents causes not only
challenges to human health, environment, and manufacturing equipment, but also
results in hydrophobized paper surfaces with difficult wetting properties for aqueous
samples. Additives such as humectants (e.g., polyethylene glycol [40]) or surfac-
tants (e.g., Triton X-100 [41]) can help to recover from this drawback by enhancing
the hydrophilicity of the paper surface. An exception where a paper surface
hydrophobized by a hydrophobic indicator system has a beneficial impact is the use
of ion-selective optodes (ISOs) on cellulosic substrates [42]. In the conventional
colorimetric detection with ISOs, an ionophore is embedded in a hydrophobic
polymer membrane (most commonly polyvinyl chloride) together with a
36 N. Ruecha et al.

Fig. 3.2 Mechanism responsible for uneven colorimetric signal appearance on paper substrates:
Unreacted and reacted water-soluble indicators are schematically illustrated in green and yellow,
respectively

hydrophobic pH indicator, because the detection principle is reliant on ion

exchange between the aqueous sample phase and the hydrophobic polymer phase
[43]. Interestingly, this mechanism was found to work without the aid of a polymer
film on a filter paper matrix. It is assumed that the hydrophobic assay components
(ionophore, pH-indicator, and anionic additive) themselves create a hydrophobic
microenvironment within the porous structure of the paper substrate, and that those
components are retained on the cellulosic surface via hydrophobic interactions such
as van der Waals forces and p bonding [42].
In those cases where the hydrophilic character of the paper matrix is to be
retained, the combination with an “anchoring” component is an effective approach
for using water-soluble reagents on paper-based sensors. For this purpose, poly-
meric materials oppositely charged to the chromogenic indicator are primarily
employed, since their higher molecular weight normally reduces their mobility
within cellulosic fiber networks compared to small molecules. Poly(vinylamine)
[44, 45] and poly-L-arginine [46], both positively charged polymers due to the
presence of protonated amino groups (–NH3 þ ), have for example been used to
entrap colored anionic compounds (5-thio-2-nitrobenzoate TNB− [45, 46]; and
indophenoxide IDO− [44]) in the detection region of a paper-based sensor for
pesticide testing (Fig. 3.3). Poly(acrylic acid), a negatively charged polymer pos-
sessing carboxyl groups, has been demonstrated to be able to retain the
water-soluble metal–chelator complex [Fe(phenanthroline)3]2+ in a paper-based
sensing device for the simultaneous assaying of airborne metals (Fe, Cu, Ni) [40].
Particulate materials, which exhibit irreversible adsorption onto filter paper [47],
3 (Bio)Chemical Sensors Based on Paper 37

(a) (b) (c)

(d) (e)

Unevenly
spread TNB- Unevenly
Entrapped spread IDO-
TNB- Faint Entrapped
color IDO- Faint
color

With Without Poly(vinylamine) With Without

Fig. 3.3 Electrostatic immobilization of colorimetric indicators on a paper matrix for pesticide
sensing: structures of a poly(vinylamine), b 5-thio-2-nitrobenzoate (TNB−), and c indophenoxide
(IDO−); effect of poly(vinylamine) for the entrapment of d TNB−. Adapted with the permission
from Hossain et al. [45]. Copyright (2009) (American Chemical Society), and e IDO−. Adapted
with the permission from Hossain et al. [44]. Copyright (2009) American Chemical Society;
arrows and dotted areas indicate the direction of sample liquid flow and the position of the
colorimetric signal detection region, respectively

have also been reported to play a key role in reducing inhomogeneous color in an
on-paper enzymatic assay [39]. This strategy relies on the protein adsorption
phenomenon onto the solid surface of nanoparticles. Likewise, immobilization of
soluble indicators with a charged nanoparticle (e.g., positively charged particle for
sulfonated indicators) is feasible. The above examples rely on indirect immobi-
lization of readily transported components via electrostatic attractive forces. As
mentioned earlier (Fig. 3.1), covalent bonding is feasible by cellulose derivatization
via reactive functional groups (e.g., aldehyde group [48], epoxy group [49], car-
boxybetaine [50], and divinyl sulfone [51]). Although being a much more robust
way of immobilization, considering the effort for the multistep reaction procedure,
covalent bonding is probably only the last option for the prevention of wash away
effects by aqueous sample liquids. Finally, one approach for eliminating uneven
color distribution on paper sensors without any chemical modification has been
reported [23]. This earlier work on a microfluidic paper-based sensor describes the
achievement of homogeneous color signals in pH and protein assays by simply
modifying the geometrical arrangement of the water-soluble colorimetric indicators
free of any “anchoring” components [23]. In this approach, the very precise posi-
tioning of reagents into only 500-µm-wide microfluidic channels of a µPAD has
38 N. Ruecha et al.

been achieved by inkjet printing. In this context, the above-mentioned compatibility

of paper with printing technologies is a significant advantage, since identical results
would hardly be possible by manual pipetting.
Enzyme-supported colorimetric assays are also a powerful technique in (bio)-
chemical sensing, including on paper matrices. Acquisition of a color signal is
based on the catalytic activities of enzymes, such as oxidation (e.g., horseradish
peroxidase) or bond cleavage (e.g., alkaline phosphatase), and thus, signal ampli-
fication is achieved by converting the analyte concentration into enzymatic activity.
The ELISA (enzyme-linked immunosorbent assay) method, commonly known for
its high sensitivity and selectivity, was first performed on a pure cellulosic substrate
(i.e., not nitrocellulose) in 2010 [52], where 96 circular test zones (5 mm diameter)
mimicking a microtiter plate have been prepared in a sheet of chromatography
paper. Although the limit of detection was 10 times higher than that of the con-
ventional ELISA, paper as the substrate contributed not only to lowered costs and
small required sample volume (3 lL), but also shortened the incubation time
(10 min) thanks to the high surface-to-volume ratio compared to a microtiter plate
well [52]. In addition, a flatbed scanner has substituted the costly microplate reader
itself for quantitative signal acquisition.
Addition of reagents in a fixed and timed order is often required in conventional
enzyme-supported assays and regarded as a big shortcoming. However, automated
sequential delivery of assay components is feasible by preparing a maze-like
structure on a porous substrate. A single-step sandwich ELISA has been demon-
strated on a single piece of nitrocellulose, wherein the user is only required to
introduce the sample at the bottom of the device (Fig. 3.4) [53]. In this sensing

Fig. 3.4 Schematics of a nitrocellulose-based biosensing device for automatically operated

ELISA; relevant assay components are separately located within the maze-like channels: a control
zone with antibodies for capturing unreacted secondary antibodies, b test zone with the analyte
antigen-specific antibodies, c secondary antibodies labeled with enzyme, d substrates forming
insoluble products after the enzymatic reaction; the dotted line indicates the cutting shape of the
absorbent area. Reproduced from [53] with the permission from the Royal Society of Chemistry
3 (Bio)Chemical Sensors Based on Paper 39

device, different travel distances to the test line enable timed and ordered delivery of
each assay component. This approach achieved a strongly simplified assay proce-
dure. Since all assay components have to be deposited on the paper (or nitrocel-
lulose) matrix in advance, this sensor design is only meaningful on the condition
that those reagents are stable in dry state for a certain period. In particular, enzymes
and antibodies lose their activity (catalytic activity, antigen recognition ability)
rapidly when stored at room temperature. There exist various strategies for
enhancing the stability of proteins. Addition of sugars (most commonly trehalose) is
often used in pharmaceutical industries. Several reports describe the capability of
transition metals to stabilize enzymes [54–56]. Sol–gel materials are also long
known for their capacity to reinforce the stability of biomolecules [57–59]. Reports
on the extended storage stability of proteins (enzyme, antibody) for the develop-
ment of colorimetric paper-based sensors are listed in Table 3.1.

3.5.2 Fluorescence-Based Signal Transduction

Off–on-type fluorescence signaling is attractive for its better sensitivity compared to

that based on color transitions. Moreover, incorporation of ratiometric detection
methods is expected to eliminate the influence from various conditions such as
ambient light or sample matrix. However, the necessity of an excitation light source
is the biggest weakness over the colorimetric assays. The interference by the
excitation light potentially contributes to background noise. During fluorescence
detection with a fluorometer, the influence of excitation light is to a large degree
removed from the emission signal by changing the angle between the photodetector
and the light source and by using optical filters. In contrast, it is much more
challenging to completely eliminate excitation light reflected off the surface of a
paper substrate. What can be even more troublesome are the above-mentioned
optical brightening agents mixed into paper materials. The most common
consumer-use types of paper are impregnated with compounds such as stilbene or
coumarin derivatives for enhancing their whiteness, and those substances emit
intense blue fluorescence upon excitation by UV light, causing a low
signal-to-noise ratio in the case of fluorescence-based signal transduction on paper
sensors (Fig. 3.5) [65]. Although black paper substrates seem to be suitable for
suppressing the noise, black pigments absorb most of the excitation light and can
reduce the emission signal intensity itself. For fluorescence-based detection,
additive-free, 100 % cellulose paper (filter paper or chromatography paper) is
conclusively the best substrate for fabricating a paper-based sensor without con-
cerns about the influence of paper ingredients.
As for the reports on fluorometric analysis on paper-based sensing devices, the
detection of nucleic acid hybridization covers the highest percentage. Fluorescence
labeling is one of the most common techniques in DNA analysis, as seen in DNA
 40

Table 3.1 Summary of reports on extended storage stability of proteins on a cellulosic substrate
Protein Stabilizing compound Storage period Paper substrate References
Acetylcholine Silica sol–gel material Full activity for at least 2 month Mead brand Hossain et al.
esterase (4 °C) cardboard paper [45]
Acetylcholine Silica sol–gel material Full activity for at least 1 month Whatman 1 filter Hossain et al.
esterase (4 °C) paper [44]
b-Galactosidase Silica sol–gel material Full activity for at least 2 weeks (r. Whatman 1 filter Hossain et al.
t.), at least 2 months (4 °C) paper [60]
Tyrosinase Chitosan–alginate-layered composite 92 % activity after 260 days (r.t.) Fisherbrand filter Alkasir et al.
paper (P5 grade) [61]
HRP-labeled 0.01 M FeSO4-EDTA, 4 % trehalose, 0.1 % BSA *80 % activity after 5 months Glass fiber (grade Ramachandran
antibody (45 °C, dry state) 8964) et al. [62]
HRP ALP Pellet composed of 75 % poly(ethyleneglycol) Full activity for at least 63 days Chromatography Mitchell et al.
methyl ether and 25 % graphite powder Full activity for at least 21 days paper (grade 1 Chr) [63]
IgG antibody 5 w/v% trehalose, 10 w/v% BSA, silica beads Full activity after 5 weeks Whatman 1 filter Wu et al. [64]
(desiccant) paper
N. Ruecha et al.
3 (Bio)Chemical Sensors Based on Paper 41

Fig. 3.5 Fluorescence images of paper substrates after deposition of oligodeoxyfluoroside: (left)
looseleaf paper, (middle) black construction paper, (right) 100 % cotton paper; orange
fluorescence indicates the sensing signal emission; images captured by an epifluorescence
microscope. Adapted from [65] with the permission from the Royal Society of Chemistry

microarrays. Recent papers show the quantification of nucleic acid hybridization by

means of turn-on emission from a fluorescent dye [66], luminescence resonance
energy transfer [67–69], and fluorescence emission from a fluorescent dye (Cy5) in
competition with a quencher (Iowa Black RQ) [70]. Signal acquisition is performed
either with a fluorescence scanner for microarray analysis or two-dimensional
electrophoresis, the combination of a UV handlamp and a digital camera embedded
in a tablet, or a fluorescence microscope. The relatively low absolute number of
examples of successful nucleic acid detection on paper devices is attributed in part
to the difficulty to achieve sufficiently low detection limits (sub-nanomolar level)
[36]. In some other cases, fluorescence-based detection has been employed to
selectively detect substance(s) of interest by simple means. Explosive compounds
(trinitrotoluene, 2,4-dinitrotoluene and 8 other compounds) have been detected by
using the fluorescence quenching of pyrene deposited on a paper matrix [71].
Conventionally, those analytes need to be inspected by means of sophisticated
analytical instruments (e.g., HPLC and LC-MS) or analytical methods requiring
labor-intensive procedures (e.g., immunoassays and amperometric sensing). In
another example, lactoferrin, a biological protein in human tears, has been assayed
on a microfluidic paper device based on fluorescence sensitization of terbium
cations (Tb3+) [34]. Signal quantification has been carried out by using comparably
accessible equipment, a digital camera equipped with a cutoff filter and UV
handlamps (kex = 254 nm). Again, the conventional analytical technique was
reliant on the labor-intensive immunoassay. Although due to the requirement of an
excitation light source completely instrument-free sensing is conceptually impos-
sible, the extension of the fluorescence detection technique to paper-based sensors
opens the door for simplified analyses compared to the existing detection
technologies.
42 N. Ruecha et al.

3.5.3 Chemiluminescence-Based Signal Transduction

Chemiluminescence (CL) is a light emission phenomenon, where an electronically

excited state intermediate is produced in a chemical reaction such as oxidation. The
excited state intermediate emits luminescence during relaxation to the ground state
(direct CL), or the intermediate transfers the energy to an acceptor, which then
emits luminescence (indirect CL). An advantageous aspect of CL detection com-
pared to fluorescence, although likewise based on luminescence measurement, is
that no excitation light is needed to observe the luminescence signal. This property
eliminates the need of optical filters and results in low limits of detection (LOD). It
has for example been reported that the LOD of a DNA chip was 100 times lowered
by the use of CL in comparison with the fluorescence method [72]. In particular, no
requirement of an excitation light source becomes a big advantage for on-paper
detection, since the noise caused by both reflection and auto-fluorescence of the
paper surface is eliminated. Thanks to this merit, CL-based paper sensors are pri-
marily devoted to clinically relevant targets, where biochemical substances have
often to be detected in a very low concentration range. Table 3.2 summarizes
analytical targets, detection systems, and analytical performance of CL-based
sensors made from paper reported in the literature, so far.
As confirmed by Table 3.2, the classical luminol oxidation CL chemistry has
been dominantly applied on paper sensors in combination with catalysts such as
oxidases or metal particles. Carbon dots, a fluorescent carbon nanomaterial recently
drawing significant attention [88], have joined the field of paper-based CL detec-
tion. This material has originally been regarded as a promising fluorescent reporter,
because of its advantageous properties (ease of fabrication, low material cost,
biocompatibility, water dispersibility, photostability, and quite high quantum yield,
among others), and not surprisingly, has been utilized for developing some
fluorescence-based paper sensors [89, 90]. Although the number of literature reports
is still limited, carbon dots are increasingly incorporated into paper-based sensing
devices as an accessible signaling material, wherein detection is not limited to
fluorescence, but also includes CL [78] and electrochemiluminescence (discussed in
the following section).
As one drawback of CL-based detection on paper sensors, challenges associated
with signal acquisition have to be pointed out. Luminescence emission is initiated at
the moment of contact of all the involved reaction components (assay reagents and
sample), and its duration is limited. In particular, the latter shortcoming makes
sensitive detection by simple means a challenging task. Luminol shows relatively
prolonged emission, which probably explains why it is preferred for on-paper
detection. As for detectors used in CL measurements, sophisticated instruments
such as a microplate reader, CCD camera, or a luminescence analyzer equipped
with a photomultiplier have been employed. Since those instruments are not really
suitable for situations requiring low costs and portability (on-site analysis),
3 (Bio)Chemical Sensors Based on Paper 43

Table 3.2 Summary of reported paper sensors based on chemiluminescence (CL) detection
Analyte CL system Limit of Linear detection range References
detection
Uric acid M4NRASP, H2O2 1.9 mM 2.6–49.0 mM Yu et al. [73]
Glucose M4NRASP, H2O2 0.14 mM 0.42–50 mM Yu et al. [74]
Uric acid 0.52 mM 1.4–47 mM
CEA Luminol, H2O2, HRP, 6.5 pg mL−1 0.01–30.0 ng mL−1 Wang et al.
p-iodophenol [75]
AFP Luminol, H2O2, AgNP 1.0 ng mL−1 2.5–110 ng mL−1 Ge et al. [76]
CA153 0.4 U mL−1 1.0–100 U mL−1
CA199 0.06 U mL−1 0.5–150 U mL−1
CEA 0.02 ng mL−1 0.1–130 ng mL−1
AFP Luminol, H2O2, HRP, 0.06 ng mL−1 0.1–35.0 ng mL−1 Wang et al.
CA125 p-iodophenol 0.33 U mL−1 0.5–80.0 U mL−1 [77]
CEA 0.05 ng mL−1 0.1–70.0 ng mL−1
DNA Carbon dot, nanoporous gold, 8.56  10−19 M 10−18 to 10−14 M Wang et al.
KMnO4 [78]
Cotinine Luminol, H2O2, HRP 5 ng mL−1 0.01–1 lg mL−1 Liu et al. [79]
−1
Dichlorvos Luminol, H2O2 3.6 ng mL 10.0 ng mL−1–1.0 lg mL−1 Liu et al. [80]
L-cysteine Luminol, H2O2, AuNP 8.2  10−10 M 10−8 to 10−6 M Liu et al. [81]
DNA Luminol, H2O2, HRP, AuNP 3.35  10−17 M 10−16 to 10−14 M Wang et al.
[82]
Ofloxacin Luminol, H2O2, AgNP, 3.0  10−10 1.0  10−9 to 1.0  10−6 Liu et al. [83]
ofloxacin g mL−1 g mL−1
b-Agonists Luminol, KIO4, b-agonists 1.0  10−9 M 1.0  10−8 to 1.0  10−5 Chen et al.
M [84]
DNA Luminol, H2O2, HRP, 6.3  10−2 pM 1.94  10−1 to 1.94  104 Liu et al. [85]
p-iodophenol pM[a]
Dichlorvos Luminol, H2O2, dichlorvos 0.8 ng mL−1 3.0 ng mL−1 to Liu et al. [86]
1.0 mg mL−1
H2O2 TCPO, H2O2, rubrene, 250 nM N. A.[b] Lebiga et al.
imidazole [87]
a
log-linear range, blinear range is not available. Nonlinear response range is reported to be 250–750 nM
M4NRASP 3-p-nitrylphenyl-5-(4′-methyl-2′-sulfonophenylazo) rhodanine, CEA carcinoembryonic antigen, HRP
horseradish peroxidase, AFP a-fetoprotein, CA153 carcinoma antigen 153, CA199 carcinoma antigen 199, CA125
cancer antigen 125, AgNP silver nanoparticle, AuNP gold nanoparticle, TCPO bis(2,4,6-trichlorophenyl) oxalate

CL-based paper platform sensors will for the time being remain limited to
“high-end” laboratory applications. The use of paper matrices can nevertheless be a
significant advantage in terms of cost reduction of consumables and performance
enhancement (e.g., short incubation times and low sample volumes). The concep-
tually related signal transduction method based on electrochemiluminescence,
however, can be adapted to more simple and user-friendly detection systems. This
approach is discussed in the following section.
44 N. Ruecha et al.

3.5.4 Electrochemiluminescence-Based Signal

Transduction

Electrochemiluminescence (ECL) can be regarded as derivative configuration of

CL. In ECL detection, the acquisition of luminescence emission is initiated by
application of voltage to electrochemically generate an excited state intermediate
species. Upon relaxation of the intermediate from the high-energy state, light of
which the wavelength is dependent on the bandgap between the excited and ground
states is emitted. In addition to the luminescent agent (ECL reagent), an additional
substance (coreactant) is normally used to facilitate the production of the emissive
intermediate. The most popular ECL system is perhaps the combination of tris(2,2′-
bipyridyl)ruthenium(II) (commonly abbreviated as RuðbpyÞ3 2 þ ) as the ECL
reagent and tri-n-propylamine (TPA) as the coreactant. In the first report on
ECL-based paper platform sensing, however [91], DBAE (2-(dibutylamino)-etha-
nol) or nicotinamide adenine dinucleotide (NADH) have been employed as core-
actants and analytes to demonstrate the feasibility of ECL measurements on paper
(Fig. 3.6a). The acquisition of the luminescence signal was performed with a
smartphone camera to show the applicability of easily accessible and simple
equipment (Fig. 3.6b). In terms of detection, ECL combines the advantages of CL
and electrochemical methods. Since excitation light is not involved, ECL mea-
surements do not suffer from background noise. In contrast to CL, the required
application of voltage provides spatially and temporally controlled reaction con-
ditions, as well as selective redox activation of species depending on their redox
potentials. Actually, the latter advantage has been proved by multiplexed detection
of cancer markers in a single reservoir on a patterned paper substrate [92]. Herein,
dual detection of cancer markers has been demonstrated by one working electrode
based on two ECL systems: (i) RuðbpyÞ3 2 þ and TPA; and (ii) carbon dots and
K2S2O8. Two types of ECL-based sandwich immunoassays were performed with
two kinds of secondary antibodies for cancer antigens labeled with either
RuðbpyÞ3 2 þ or carbon dots in the same detection reservoir of the paper device. By
duplicating the detection reservoir, a total of four cancer markers could be detected
with a single sensing device (Fig. 3.7) [92]. By switching the potential applied to
the working electrode, one ECL system can be selectively addressed and activated
in each detection reservoir [a positive potential of +1.2 V drives only ECL system
(i), and a negative potential of −1.2 V drives only ECL system (ii)]. After opti-
mizing the voltages to properly activate a single ECL system only, simple switching
of the electrode connection enabled cross-reaction-free dual-mode detection in a
single region of the paper sensing device.
ECL systems reported so far are not limited to the above examples. As ECL
reagents, the aforementioned ruthenium complex [91–96] is frequently used, and
the application of carbon dots [92, 97] as a new luminescent material is increasing,
as already noted in the context of CL-based signal transduction. Other candidates
3 (Bio)Chemical Sensors Based on Paper 45

Fig. 3.6 a ECL detection principle used in Ref. [91]; the figure represents the detection of DBAE
as an example; b smartphone-based ECL signal acquisition: Right figures show the ECL signal
emission captured by a smartphone camera during DBAE detection (concentration described in
each panel). Adapted with the permission from Delaney et al. [91]. Copyright (2011) American
Chemical Society

Fig. 3.7 Configuration of an ECL paper sensing device for the determination of cancer markers
(AFP: a-fetoprotein, CEA: carcino embryonic antigen, CA199: carcinoma antigen 199, CA153:
carcinoma antigen 153) (left: photograph of the device, right: schematic illustration of the dual
sandwich immunoassay per single detection reservoir); the labeling ECL reagents (RuðbpyÞ3 2 þ
and carbon dots) are abbreviated in the right scheme as “Ru” and “CD,” respectively. Reproduced
from [92] with the permission from the Royal Society of Chemistry
46 N. Ruecha et al.

include quantum dots [98–100], carbon nanocrystals [96], and 4,4′-

(2,5-dimethoxy-1,4-phenylene)bis(ethyne-2,1-diyl)dibenzoic acid (p-acid) in com-
bination with metal materials (nanoporous silver [101] and Pt–Ag alloy nanopar-
ticles [102]). As for the coreactant, TPA [91–97, 101, 102] is by far the most
common, but occasionally replaced by other substances such as guanine [103],
DBAE [91, 97], NADH [91], or K2S2O8 [92].

3.5.5 Other Optical Transduction Schemes

Aside from the aforementioned detection principles, transmission [104] and

reflection [105] have been demonstrated to be useful for paper-based sensing
devices. Quantitative detection schemes discussed so far are based on numerical
information obtained as color values or luminescence intensities of a certain area
(detection region) of the paper device. In such a detection motif, however, the use
of a signal quantification system (e.g., the combination of a camera and color
analysis software) is inevitable to make the assay result accurate and
operator-independent. For further improving the convenience of paper-based
sensing devices, efforts have been devoted to the development of (nearly)
instrument-free, yet quantitative detection methods.
Such user-friendly detection motif was pioneered by Phillips and co-workers
[106]. They have developed a multi-layered paper-based sensor allowing (semi)
quantitative determination of H2O2 by simply counting the number of colored
detection regions appearing on the top of the device at a certain time (Fig. 3.8a).
The intermediate paper layer contains a hydrophobic compound, which is
decomposed to hydrophilic compounds upon contact with H2O2. Beneath each
detection region on the topside of the device, a paper layer with a colorant and
paper layers with varying amounts of the H2O2-degradable compound are stacked
(Fig. 3.8b). When the sample arriving from the bottom decomposes the
hydrophobic compound and passes through the intermediate layer, the eluted col-
orant appears at the topside after a sample H2O2 concentration dependent amount of
time (Fig. 3.8c). The higher the concentration of H2O2, the more colored detection
regions appear, and thus, H2O2 concentration can be estimated by counting the
number of the colored bars at a specified time. This detection motif requires only a
timer, readily accessible even in a resource-limited environment. Later on, this
decomposable hydrophobic compound supported “stop-flow” approach was
extended to “timing readout”-type paper-based (bio)chemical sensors. Enzymes
(alkaline phosphatase and b-galactosidase) [107] and toxic metals (Hg2+, Pb2+)
[108] have been assayed using the time lag between the appearance of two colored
spots. Additionally, potassium has been determined based on the wicking time per
distance unit of a colored solution, using a hemin/G-quadruplex DNAzyme as the
selective assay component [109]. Those devices allow maximally simplified (semi)
quantitative assays for non-trained users, but the challenges associated with
3 (Bio)Chemical Sensors Based on Paper 47

Fig. 3.8 (Semi)quantitative signal detection relying on counting of colored regions visible after a
fixed period of time: a pictures of the topside of the sensor 10 min after depositing H2O2 samples
(1, 35, 75, and 100 mM from top to bottom); b schematic of the multilayered device structure; and
c relation between H2O2 concentration and the flow-through time of the sample in a single
cylindrical channel; filled and open circles represent the results in the presence (1.7 lg mm−3) and
absence of the H2O2-degradable compound, respectively; the inset shows the expanded view of the
dotted area in the main figure; markers and error bars reflect the average and standard deviations
of 10 trials. Adapted with the permission from [106]. Copyright© 2012 WILEY-VCH Verlag
GmbH & Co. KGaA, Weinheim

automated fabrication have to be pointed out as a bottleneck for stacking-type paper

devices.
On the other hand, a totally instrument-free, easy-to-manufacture paper-based
sensing device has been proposed by the Henry group [110–112]. They introduced
“distance” as the quantification signal on a paper device. The device consists of a
sample inlet and a microfluidic paper channel, wherein assay reagent(s) forming
insoluble colored products is deposited. As the sample travels within the paper
channel via capillary action, the analyte is continuously consumed by reaction with
the assay component(s). Upon analyte depletion, the generation of colored products
is stopped, resulting in a distance-based color signal along the straight channel
depending on the initial analyte content (Fig. 3.9a, b). This concept has been
extended by Yamada et al. [113] by converting the distance into “concentration
scale marks.” Direct readout of the sample analyte concentration becomes feasible
even by the naked eye in a similar fashion to a classical analogue thermometer
48 N. Ruecha et al.

Fig. 3.9 a Schematic of a typical distance-based paper device; b detection chemistry for nickel
(II) (top) and typical result of a distance-based nickel assay (bottom). Reproduced from [110] with
the permission from the Royal Society of Chemistry; and c a distance-based paper device for
lactoferrin using a thermometer-like signal readout. Reprinted with the permission from Yamada
et al. [113]. Copyright (2015) American Chemical Society

(Fig. 3.9c). Distance detection-based paper sensors have been successfully

demonstrated for several sets of analytes and assay component(s), as summarized in
Table 3.3. This type of sensor devices can be manufactured relying on printing
technologies alone [112] and, thus, is compatible with mass production.
Although gaining quantitative information has not been an objective, Shen et al.
reported an innovative paper device, which directly displays blood types (ABO and
RhD) in written character(s) [116]. Blood type is determined by the species of
antigens expressed on red blood cells, and each type of blood shows hemaggluti-
nation upon contact with the corresponding antibody (e.g., blood cells of A + type
agglutinate in the presence of anti-A or anti-D antibodies). The strong retention of
hemagglutinated blood cells on the paper substrate after saline washing was utilized
for reporting the test result in text. Kleenex paper towel has been preferred as the
device substrate over filter paper or blotting paper, since the large pore size guar-
antees the smooth detachment of non-agglutinated red blood cells [116].
3 (Bio)Chemical Sensors Based on Paper 49

Table 3.3 Summary of reported distance detection-based microfluidic paper sensors

Analyte Assay component(s) Detection Detectable range References
principle
Glucose GOx, HRP, DAB Colorimetry 11–270 mg dL−1[a] Cate et al. [110]
Ni2+ Dimethylglyoxime Colorimetry 0.7–92 lg m−3[a] Cate et al. [110]
[b]
Glutathione AgNP Colorimetry 0.12–2.0 nmol Cate et al. [110]
1,4-naphtho AgNP, glutathione Colorimetry 5–25 ng, Dungchai et al.
quinone 15–30 ng[c] [111]
Fe2+ Bathophenanthroline Colorimetry 3.3–470 ppm[d] Cate et al. [112]
20–1300 ppm[e]
Ni2+ Dimethylglyoxime[f] Colorimetry 6.7–670 ppm[d], Cate et al. [112]
100–1100 ppm[e]
Cu2+ Dithiooxamide[f] Colorimetry 6.7–1100 ppm[d], Cate et al. [112]
100–1300 ppm[e]
Amphiphilic Not required Wet or dry 0–100 lM (DNAs), Chen et al. [114]
compounds 0–0.2 g mL−1
(albumin),
0–1000 mg dL−1
(LDL)
Human lactoferrin Tb3+[g] Fluorometry 0.1–4 mg mL−1 Yamada et al.
[113]
Cu2+ BSA-coated AuNC Fluorometry 0–500 lM Fang et al. [115]
a b c
Linear range, log-linear range, linear range. The range changes depending on the amount of glutathione
deposited onto the device. dDynamic response range obtained from a single-channel device. eDynamic response
range obtained from a multiple-channel device. fRemoval of interfering metal species (masking) is performed.
g
Paper surface treatment with anionic polysaccharide is necessary when using complex sample matrix
GOx glucose oxidase, HRP horseradish peroxidase, DAB 3,3′-diaminobenzidine, AgNP silver nanoparticle,
LDL low-density lipoprotein, BSA bovine serum albumin, AuNC gold nanocluster

3.6 Electrochemical Sensors

Electrochemical sensing techniques are ideally suited for the detection of chemical
and biological compounds, due to the availability of portable compact field-use
instrumentation, low operating costs, and high sensitivity and selectivity. Compared
to colorimetric techniques, electrochemical detection generally exhibits faster
response times and higher sensitivities down to the nanomolar range, in addition to
not being affected by the influence of sample color. Electrochemical techniques
measure the current or voltage generated through an electrochemical process. This
is a useful alternative analytical method for quantitative analysis, because choosing
proper detection conditions (e.g., operating potential) and electrode materials
contributes to rendering the method insensitive to interfering molecules. There are
several electroanalytical sensing systems, which are widely available for
point-of-care testing or clinical diagnosis, and these have been successfully com-
mercialized through large-scale production. These include personal glucometers
50 N. Ruecha et al.

(for monitoring diabetes) and cholesterol meters (for monitoring the risk of
cardiovascular disease).
Paper-based microfluidic sensing devices have the potential to become good
alternatives for point-of-care testing. However, they sometimes lack the sensitivity
and selectivity required for practical use, when coupled with colorimetric trans-
duction methods. Therefore, paper-based electrochemical sensors have recently
gained attraction as a promising alternative approach for clinical diagnostics and
environmental monitoring. The most common electrochemical sensing systems
consist of three electrodes: working, counter, and reference electrodes. Various types
of materials such as carbon-based materials (e.g., graphite, graphene, and carbon
nanotubes) and metal-based materials (e.g., gold) have been selected for use as
working electrodes on paper-based sensors, because they are electrochemically inert
over a wide range of working potentials. Silver/silver chloride (Ag/AgCl) is the most
widely used combination of materials for reference electrode fabrication. To fabri-
cate electrodes on paper substrates, many different types of fabrication methods have
been reported. In this context, the compatibility of paper matrices with printing
methods in general and the progress achieved in the field of printed electronics [117]
have had a tremendous impact. Screen printing is one of the most advantageous and
widely used fabrication techniques for preparing low cost and disposable
carbon-based electrodes or Ag/AgCl electrodes with flexible designs [118–120].
Inkjet printing has been introduced as a rapid, precise, and reproducible technique
for electrode preparation, because of the broad variety of materials performing as
electrodes on paper substrates, such as conductive inks and metallic nanoparticles
[121]. While these methods require some kind of printing equipment for electrode
fabrication, an even simpler and paper-specific method has been demonstrated by
drawing electrodes using graphite pencils [122, 123]. These graphite pencils are low
cost and can be found everywhere in the world. Gold or gold nanoparticles have been
applied as electrodes to paper by sputtering methods [124] or a calligraphy pen
[120]. Finally, the possibility of easily incorporating microwire-working electrodes
at any location on paper-based sensing devices has been demonstrated as another
valuable approach, since prefabricated electrodes can be introduced at any time
during device fabrication [125, 126]. This enables the independent modification of
electrodes and paper matrix, which is an advantage in cases where harsh reaction
conditions are required that are incompatible with cellulose.
The coupling of paper-based microfluidics and electrochemical signal transduc-
tion for the analysis of multiple biomarkers was demonstrated for the first time by
Dungchai et al. in 2009 [118]. Three biomarkers including glucose, uric acid, and
lactate were simultaneously measured in a biological sample (e.g., human serum)
using an enzyme-based sensing scheme with amperometric detection. In the
meantime, many kinds of paper-based electrochemical sensors have been presented
for quantitative analysis including amperometry-, voltammetry-, and
potentiometry-based systems.
3 (Bio)Chemical Sensors Based on Paper 51

3.6.1 Amperometry-Based Sensors

Amperometric measurements, where a constant potential is applied on the working

electrode and the current caused by the oxidation or reduction of an analyte and/or
an electroactive species produced in a (bio)chemical reaction is continuously
measured, are among the most important electroanalytical techniques. The resulting
current is directly proportional to the concentration of the analyte. However, the
major drawback of this technique is the lack of selectivity, because some ubiquitous
small molecules or metabolites (e.g., ascorbic acid and uric acid) are electroactive
under common assay conditions. In addition, interferents or impurities of
by-products can be adsorbed on the working electrode surface, changing its
response characteristics. To increase the selectivity and sensitivity of this technique,
enzymatic assays are selected, wherein the enzyme converts the analyte of interest
into an electroactive product species. Paper-based electrochemical sensors coupled
with enzymatic reactions for determination of glucose [118, 127–131], lactate [118,
128, 131], uric acid [118, 128], ethanol [131], and cholesterol [119, 131] have been
demonstrated by many groups. Table 3.4 provides an overview of selected exam-
ples of amperometric sensors involving paper substrates. Two different layouts
combining paper matrices and electrochemical detection have been successfully
realized: In the first type, electrodes are fabricated and integrated on the paper-based
device (“integrated type”), whereas in the second type, a paper-based device (e.g.,
microfluidically patterned paper) is combined with an external set of electrodes
(“combination type”).
The classical amperometric three-electrode system (Fig. 3.10a) was integrated
into a paper-based device at first in 2009 [118]. This integrated-type device con-
sisted of three channels and three detection zones to simultaneously amperomet-
rically detect the three biomarkers, namely glucose, uric acid, and lactate, by
enzymatic analyte conversion in human serum and urine. In this approach, no
interference from the matrix and color of samples was observed, because of the high
selectivity of the enzymes, the low detection potential (0 V vs. Ag/AgCl), and the
type of working electrode material (Prussian blue redox mediator-modified carbon).
A combination-type paper-based amperometric sensor has been reported by
Noiphung et al. [132]. In this case, a paper-based device coupled with an external
set of electrodes for one-step detection of glucose levels in whole blood samples
was successfully demonstrated. The schematic of this device is shown in
Fig. 3.10b. A disposable set of commercially available screen-printed carbon
electrodes (SPCEs) modified with Prussian blue mediator was used for glucose
detection. To perform clinical diagnosis in whole blood, separation of red blood
cells from the sample is often required to prevent contamination and malfunction of
the analytical system. The most common methods for blood separation are cen-
trifugation or sedimentation. However, they require large blood volume and special
instrumentation [133]. Alternatively, blood cell separation can be performed by
paper-like materials. A paper-based device consisting of a detection zone (Whatman
1 filter paper) located in-between two separate zones (Whatman VF blood
Table 3.4 Selected examples of paper-based electroanalytical sensors coupled with amperometry and voltammetry
52

Electrochemical technique Electrode material(s) Analyte Analytical performance References

Linear range Limit of detection
Amperometry PB-modified SPCE Glucose 1–100 mM 0.21 mM Dungchai et al. [118]
Lactate 1–50 mM 0.36 mM
Uric acid 1–35 mM 1.38 mM
Commercial PB-SPCE Glucose 0–33.1 mM N. A. Noiphung et al. [132]
SPCE (on polyester film) Glucose 0–22.2 mM 0.22 mM Nie et al. [127]
G/PVP/PANI-modified SPCE Cholesterol 0.05–10 mM 1 µM Ruecha et al. [119]
SPCE Glucose 0–20 mM 0.35 mM Chen et al. [128]
Lactate 0–25 mM 1.76 mM
Uric acid 0–10 mM 0.52 mM
Carbon (by handheld Glucose 2–20 mM 2 mM Li et al. [129]
pressure-assisted ball pen)
SPCE Glucose 0–27.75 mM 1.443 mM Nie et al. [131]
Cholesterol 0.5–5.2 mM 0.34 mM
Lactate 1–11 mM 1.1 mM
Ethanol 0.1–3 mM 0.1 mM
Commercial PB-SPCE Glucose 0.25–2 mM 0.01 mM Chandra Sekar et al. [135]
Voltammetry Commercial SPCE Dopamine 1–100 µM 0.37 µM Rattanarat et al. [143]
MWCNT-modified SPCE CA125 0.001–75 U mL−1 0.2 U mL−1 Wang et al. [144]
CEA 0.05–50 ng mL−1 0.01 ng mL−1
AgNPs-modified SPCE Ricin 34 pM–21 nM 34 pM Cunningham et al. [145]
GO/Chitosan-modified SPCE AFP 0.001–100 ng mL−1 1 pg mL−1 Wu et al. [146]
CEA 0.005–100 ng mL−1 5 pg mL−1
CA125 0.001–100 ng mL−1 1 pg mL−1
CA153 0.005–100 ng mL−1 5 pg mL−1
(continued)
N. Ruecha et al.
Table 3.4 (continued)
Electrochemical technique Electrode material(s) Analyte Analytical performance References
Linear range Limit of detection
AuNPs/G-modified SPCE DNA 0.0008–500 pM 0.2 fM Lu et al. [147]
Au particles-modified SPCE DNA N. A. 30 nM Cunningham et al. [148]
Thrombin N. A. 16 nM
CNT-modified SPCE Pb2+ 5–150 ppb 1 ppb Rattanarat et al. [149]
Cd2+ 5–150 ppb 1 ppb
SPCE Pb2+ 0–100 ppb 1 ppb Nie et al. [127]
SPCE Pb2+ 0–100 ppb 2.0 ppb Shi et al. [150]
Cd2+ 0–100 ppb 2.3 ppb
Ag foil and Ag/AgCl foil Cl− 32 µM–0.6 M 10 µM Cuartero et al. [151]
3 (Bio)Chemical Sensors Based on Paper

Br− 16 µM–0.1 M 10 µM
I− 16 µM–0.1 M 10 µM
PB Prussian blue, SPCE screen-printed carbon electrode, N. A. not available, G graphene, PVP polyvinylpyrrolidone, PANI polyaniline, MWCNT multiwall
carbon nanotubes, CA125 carcinoma antigen 125, CEA carcinoembryonic antigen, AgNPs silver nanoparticles, GO graphene oxide, AFP a-fetoprotein, CA153
carcinoma antigen 153, AuNPs gold nanoparticles, CNT carbon nanotubes
53
54 N. Ruecha et al.

Fig. 3.10 a Integrated-type paper-based amperometric sensing device with three electrodes for
simultaneous detection of three analytes. Reprinted with the permission from Dungchai et al.
[118]. Copyright (2009) American Chemical Society; b combination-type paper-based ampero-
metric sensing device for glucose detection in whole blood on commercial Prussian blue-modified
screen-printed electrodes (PB SPCEs). Reproduced from Noiphung et al. [132]. Copyright (2013),
with the permission from Elsevier

separation paper) was used to obtain blood plasma separated from the whole blood.
Glucose present in the blood plasma transported to the detection zone through the
connecting channel is converted to hydrogen peroxide by predeposited glucose
oxidase and amperometrically detected by the underlying SPCEs [132]. In several
other cases of combination-type paper-based amperometric glucose sensors [127,
134, 135], reagents or enzyme has been immobilized into the cellulose matrix to
improve the storage stability or sensitivity of the electrochemical sensor.
In contrast to the previously discussed colorimetric sensors where the naked eye
can serve as the simplest detector, the electrochemical analogues always require an
electronic device for signal detection (e.g., external potentiostat). Due to their
flexible and variable design, however, paper-based electrochemical sensors can be
applied with portable or in-house potentiostats to reduce the operating costs. The
combination of paper-based amperometric sensors with commercially available
handheld glucose meters has been introduced by Whitesides’ group in 2010, as
shown in Fig. 3.11a [131]. They have demonstrated the use for quantitative elec-
trochemical analysis of important compounds including not only, of course, glu-
cose, but also cholesterol and lactate in biological samples, as well as alcohol
monitoring in food quality control. These devices were designed to mimic the test
strips sold for the glucose meter and were adapted for cholesterol, lactate, and
ethanol detection by simply changing the enzyme. A different but conceptually
related work on paper-based amperometric sensors aiming at cost optimization used
a custom-made handheld electrochemical signal reader (potentiostat) and an array
of eight electrochemical sensors for simultaneous determination of glucose, lactate,
and uric acid [128]. This electrochemical paper-based sensor array allowed
detecting several analytes in a sample solution, while simultaneously producing
multiple test results for each analyte.
3 (Bio)Chemical Sensors Based on Paper 55

Fig. 3.11 a Paper-based electroanalytical sensor from a single layer of paper adapted to fit a
standard glucometer used as a signal reader. Reproduced from [131] with the permission from the
Royal Society of Chemistry; b paper-based electrochemical sensing device with integrated
metal/air battery as power supply for electrochemical sensing and electrochromic readout (PB
Prussian blue; PW Prussian white). Adapted with the permission from Liu et al. [130]. Copyright
(2012) American Chemical Society

In an interesting approach, a completely self-contained paper-based electroana-

lytical device without the requirement of an external potentiostat has been introduced
by Crooks’ group [130]. A metal/air battery integrated into the paper device serves as
power supply for both the electrochemical sensing reactions and the color conversion
of an electrochromic material (Prussian blue) for analyte signaling (Fig. 3.11b). The
battery is activated by the application of the sample itself. Although their device is
currently not entirely made from paper (ITO electrodes required), the inventors
predict that an analogous “all-paper-approach” will become available, making it
useful for electrochemical point-of-care testing in low-resource settings.
56 N. Ruecha et al.

In recent years, there has been an increasing interest in the development of new
systems for sensitive detection of clinical biomarkers (e.g., disease markers, proteins,
DNA/RNA, and heavy metals), because abnormal levels of these compounds may
indicate serious health conditions. Early detection of diseases is greatly desired, since
at an early stage, treatment commonly shows the highest probability of success. With
the emerging field of nanotechnology, the use of nanomaterials for various appli-
cations has concomitantly increased. The modification of working electrodes with
nanoparticles, such as gold (AuNPs) [136–138], silver (AgNPs) [139], carbon nan-
otubes (CNT) [140, 141], and graphene (G) [119], leads to substantially enhanced
electrode surface areas, as well as conductivity. Accordingly, these materials have
also found use on paper-based electrochemical sensing devices. Moreover,
nanocomposite forms such as metallic nanoparticles/polymer or nanocarbon/
polymer have attracted more attention than their pure forms, since the composites
are more compatible with fabrication and accessible to further functionalization.
A paper-based electroanalytical sensor integrating a graphene/conducting polymer
composite has been demonstrated by Ruecha et al. [119]. The prepared graphene/
polyvinylpyrrolidone/polyaniline (G/PVP/PANI) nanocomposite was used to mod-
ify a paper-based sensor for determination of cholesterol via electrospraying. This
modification resulted in high sensitivity, fast response time, a wide linear range from
0.05 to 10 mM, and a very low detection limit of 1 lM. Moreover, it might be
applied to sensitively estimate the amount of cholesterol in human serum samples.
Zinc metal oxide is another nanomaterial that has shown outstanding capacities for
paper-based electrochemical sensing [142]. ZnO nanoparticles possess not only high
surface area, but also high biocompatibility, in addition to providing a fast electron
transfer between electrode surface and analytes.
Even though amperometric techniques (current measurements under application
of a fixed potential) are ideally suited for the rapid analysis of a single analyte, they
fail for the simultaneous detection of multiple analytes with only a single working
electrode. Therefore, voltammetric measurements are of interest for the simulta-
neous determination of multiple analytes.

3.6.2 Voltammetry-Based Sensors

Voltammetry, measuring the current resulting from an electrochemical reaction as a

function of a variable applied potential, is another important electroanalytical
technique. It has been widely used for the quantitative analysis of various com-
pounds in environmental (e.g., heavy metals) and clinical samples (e.g., dopamine,
ascorbic acid, and uric acid). Recently, combinations of paper-based devices with
voltammetric detection have been comprehensively developed for many applica-
tions, especially for clinical diagnosis and environmental monitoring. A selection of
voltammetry-based paper sensors is provided in Table 3.4. Henry’s group, for
example, introduced a new approach for the preconcentration of dopamine using a
3 (Bio)Chemical Sensors Based on Paper 57

(a) m (b)
1c Paper-based device

Transparency with 2 holes

Commercial screen-printed
carbon electrode

Electrochemical measurement Paper zone with chitosan,

MWCNTs and BSA
DA 80 µM with SDS
and pre-concentration

5 µA DA 80 µM with pre-
concentration and
without SDS

DA 80 µM without SDS
and pre-concentration
−0.2 0 0.2 0.4 0.6 0.8
Potential (V) vs Ag/AgCl

(c)

Slip Incubate
down Add buffer & Run sample Run ACV
ACV or SWV or SWV

Fig. 3.12 Examples of paper-based voltammetric sensors for: a the determination of dopamine
(DA). Adapted from Rattanarat et al. [143]. Copyright (2012), with the permission from Elsevier;
b simultaneous quantification of multiple cancer markers. Adapted from [152] with the permission
from the Royal Society of Chemistry; and c DNA or protein detection (ACV alternating current
voltammetry; SWV square wave voltammetry). Adapted with the permission from Cunningham
et al. [148]. Copyright (2014) American Chemical Society

paper-based device consisting of three layers (Fig. 3.12a): (1) a filter paper layer for
the application and preconcentration of the hydrophilic sample; (2) a transparency
film layer for sample preconcentration and transfer of the preconcentrated sample to
the underlying detection layer; and (3) commercial SPCEs for voltammetric signal
detection [143]. This paper-based device shows several benefits including high
measurement reproducibility by using commercial SPCEs, and low limits of
detection due to the preconcentration of the sample using the paper. Moreover,
potentially interfering compounds relevant in biological samples such as ascorbic
acid and uric acid did not affect the sensor response, because of different redox
potentials. Furthermore, voltammetric methods are effectively applied in
paper-based immunoassays by labeling antibodies with electroactive species such
as AgNPs [145], or enzymes such as horseradish peroxidase [152, 153]. In the case
of enzyme labeling, the addition of the enzyme substrate is required to produce the
electrochemical signal.
In general, the signal intensity of paper-based electrochemical immunosensors
can be amplified by two ways. The first way is to enhance the surface area and the
conductivity of the working electrode, and the second way is to increase the
efficiency of the redox-active probe by integrating an enzyme or a nanomaterial
58 N. Ruecha et al.

[144, 146, 153]. In an example of the first approach, the Yu group modified the
porous fiber network of filter paper with multiwalled carbon nanotubes (MWCNTs)
to enhance the conductivity of the electrodes on the paper-based device (Fig. 3.12b)
[144, 152]. The presence of the MWCNTs did not only enhance the conductivity
for the electrochemical measurements, but also contributed to stabilizing the
immobilized antibodies on the paper. Their efforts resulted in a simple, sensitive,
low-cost, disposable, and portable paper-based electrochemical immunosensor for
the screening of cancer markers including a-fetoprotein (AFP), carcinoma antigen
125 (CA125), carcinoma antigen 199 (CA199), and carcinoembryonic antigen
(CEA) [152]. The sensing device comprises two stacked layers of patterned paper.
The first layer includes eight CNT working electrodes (two electrodes per cancer
marker), whereas the counter and reference electrodes were integrated into the
second layer. Wu et al. have given an example of combining an enzyme and a
nanomaterial with additional electrode modification for signal amplification on a
paper-based sensor [146]. The combination of horseradish peroxidase (HRP) and
antibody coimmobilized silica nanoparticles with graphene-modified paper elec-
trodes was used to achieve signal amplification for four kinds of cancer marker
detection. The electrochemical paper-based device exhibited good reproducibility,
accuracy, and stability for immunosensing.
DNA biosensors are another group of biosensors receiving significant attention
for their potential applications, especially in clinical diagnostics. Various analytical
techniques and instruments have been developed. However, they all require pro-
fessional training, time-consuming sample preparation, and expensive instrumental
equipment. Therefore, simple, rapid, inexpensive, and accurate methods for DNA
detection are an urgent need. In 2012, a folding paper-based electrochemical sensor
for target DNA detection was introduced for the first time by Lu et al. [147].
Folding paper 3D devices based on gold nanoparticle and graphene-modified
SPCEs possessed high stability and provided a good platform for capture DNA
immobilization. The combination of two types of conductive materials provided an
increased electrode surface area and electron transfer rates, leading to enhanced
sensitivity of the sensors. Moreover, bioconjugates of complementary ssDNA
immobilized on nanoporous gold as voltammetric signal amplification label resulted
in high sensitivity and selectivity with detection limits for target DNA as low as
2  10−16 mM. A different design of an inexpensive paper-based electroanalytical
sensing device for oligonucleotide and protein determination has been reported by
Crooks’ group based on the principle of target-induced conformational switching
[148]. Target-selective aptamers with a methylene blue (MB) electroactive reporter
undergo a conformation change in the presence of the analyte, increasing the dis-
tance of the MB residue from the electrode surface. Quantitative information is
obtained by comparison with the voltammetric signal in the absence (MB close to
working electrode surface) and presence of the target (increased MB distance from
working electrode). This novel concept makes use of two pieces of paper, a slip
layer and a base layer with a hole-punched paper flap (Fig. 3.12c). The mechanical
switch consisting of movable paper layers allows for simple user control of incu-
bation time of the sample on top of the aptamer-modified working electrode. This
3 (Bio)Chemical Sensors Based on Paper 59

function cannot be achieved with continuous flow-type sensors and is an example

demonstrating that the use of paper even allows the integration of simple
mechanical operations contributing to increased sensor performance, while main-
taining low cost, disposability, and user-friendliness.
Awareness about elevated levels of certain metal ions found in environmental or
biological samples is increasing. While some of these metals are essential for
human health, others are known to be highly toxic. Although a number of
paper-based sensors for heavy metals with colorimetric signal detection have been
developed, there are certain limitations in sensitivity [108, 149, 154, 155].
Therefore, paper-based electrochemical sensors for the detection of heavy metals
are of substantial interest. This task has been approached by designing flow
channels in filter paper with integrated SPCEs [127, 150]. Whitesides’ group first
demonstrated a paper-based electrochemical sensor relying on anodic stripping
voltammetry (ASV) for heavy metal detection as shown in Fig. 3.13a [127]. The
sensor featured excellent sensitivity for Pb2+ and improved selectivity in the
presence of Zn2+. In the context of this sensing application, the paper matrix
demonstrates a surprising advantage other than the cost and simplicity factor.
Higher sensitivity and lower detection limits for Pb2+ were achieved with paper
(1.0 ppb) compared to conventional electrochemical setups (2.5 ppb). This is
attributed to the fact that the capillary-driven sample flow in the cellulose fiber
matrix provides an ideal way of analyte transport over the electrode surface,
resulting in high sensitivity, reliability, and reproducibility of the measurements.
Another sensing device making use of this advantage has been applied to Pb2+ and
Cd2+ detection in beverages [150]. Again, a paper flow channel was designed to
cross the surface of the SPCEs, resulting in good wettability properties and contact
with the electrodes. For similar reasons, the Bakker research group referred to filter
paper as the material for sample and electrolyte chambers providing stable confined
thin aqueous layers in a coulometric sensor for the simultaneous determination of
chloride, bromide, and iodide ions (Fig. 3.13b) [151].
A vast choice of materials is available as platforms for the deposition of con-
ductive materials in the fabrication of electrodes for electrochemical sensors, and
paper is simply one of those. Only recently, the performance of paper and plastic
substrates in electrochemical Pb2+, Cd2+, and Zn2+ detection has been directly
compared (Fig. 3.13c) [156]. In this particular case, the electrodes fabricated on
filter paper showed clearly inferior analytical performance for Pb2+, Cd2+, and Zn2+
detection, because the penetration of the ink used to fabricate the SPCEs into the
porous paper network leads to thinner layers of conductive material and an increase
in the resistance. Moreover, the degree of ink penetration is influenced by external
factors including humidity and temperature. This example illustrates that paper is
not always the material of choice in terms of performance, but might simply be
considered for sensor cost reduction.
A combination of voltammetric and colorimetric detection on a paper-based
analytical device was developed by Henry’s group for the quantification of six
metals using a double detection layer outline (Fig. 3.14) [149]. Colorimetric
detection was used for Fe, Ni, Cr, and Cu, while square-wave anodic stripping
60 N. Ruecha et al.

Fig. 3.13 a Sensing device consisting of photoresist patterned paper channel and a three-electrode
system (carbon as the working and counter electrodes, and Ag/AgCl as a pseudo-reference
electrode). Reproduced from [127] with the permission from the Royal Society of Chemistry;
b sensor for the simultaneous coulometric detection of chloride, bromide, and iodide using paper
for the formation of thin sample liquid and reference electrolyte layers. Adapted with the
permission from Cuartero et al. [148]. Copyright (2015) American Chemical Society; and
c comparison of bare SPCEs on paper and plastic substrates for voltammetric Pb2+, Cd2+, and Zn2+
detection. Reproduced from Ruecha et al. [156]. Copyright (2015), with the permission from
Elsevier

voltammetry (SWASV) was applied to determine Pb and Cd. The separation of

detection layers allowed this device to incorporate unique chemistries for enhancing
the selectivity and sensitivity to quantify six metals. The sensor was successfully
applied for simultaneous determination of particulate metals collected from air
samples.

3.6.3 Potentiometric Sensors

Potentiometry is a widely used electroanalytical technique due to its simplicity,

versatility, and low cost. It requires only a two-electrode setup to measure a
potential difference between an indicator and a reference electrode in the presence
of sample solution. Selectivity to the species of interest is achieved by coating the
indicator electrode with an ion-selective membrane (ISM). There are several pos-
sibilities to make use of paper in the context of ion-selective potentiometric
3 (Bio)Chemical Sensors Based on Paper 61

Fig. 3.14 Analytical procedure for six metal assays using a combination of colorimetric and
voltammetric detection methods on a single paper-based device. Reprinted with the permission
from Rattanarat et al. [149]. Copyright (2014) American Chemical Society

systems. Table 3.5 provides an overview of selected examples. In the simplest

approach, paper has been used as a disposable microfluidic sampler and “container”
for measurements in combination with external ion-selective and reference elec-
trodes [157, 158]. This allows for a single-use sampling system (disposable paper
unit), while the ion-selective and reference electrodes are usually used for many
times.
Novell and coworkers have used a conductive paper obtained by the modifica-
tion of filter paper with MWCNTs as the substrate for the deposition of ISMs based
on the classical poly(vinyl chloride)/bis(2-ethylhexyl) sebacate (PVC/DOS) matrix
to fabricate ion-selective electrodes [159]. The paper electrodes in combination with
a standard reference electrode were applied for the sensing of potassium, ammo-
nium, and hydrogen ions. The analytical performance of the paper electrodes is
similar to the classical solid-state ion-selective electrodes (ISEs) based on glassy
carbon substrates, with detection limits in the micromolar range. Paper-based ISEs
with nanomolar-order detection limits for cadmium, silver, and potassium ions have
been realized on filter paper substrates modified by single-walled carbon nanotubes
 62

Table 3.5 Selected examples of paper-based electroanalytical sensors coupled with potentiometry and electrochemical impedance spectroscopy (EIS)
Electrochemical Electrode material(s) Analyte Analytical performance References
technique Linear range Limit of
detection
Potentiometry Commercial ISE Pb2+ 0.01 mM–0.28 M N. A. Lisak et al.
Cd2+ 1 mM–1 M 0.1 mM [158]
Cl− 1 mM–1 M 10 µM
MWCNT-modified filter paper NH4+ 0.1 mM–1 M 7.2 µM Cui et al.
K+ 0.1 mM–1 M 4.1 µM [157]
pH 4–10 N. A.
POT/AuNPs/SWCNT-filter paper Ca2+ N. A. 1.2 nM Mensah
Ag+ N. A. 25.1 nM et al. [160]
K+ N. A. 11.0 nM
SPCE on filter paper Cl− 0.01–1 M N. A. Lan et al.
K+ 0.001–1 M N. A. [162]
Na+ 0.01–1 M N. A.
CIM carbon-based reference electrodes Cl− 0.0067–1 M N. A. Hu et al.
integrated into paper-based device [163]
EIS AuNPs/rGO-modified filter paper Escherichia 1.5  102 to 1.5  102 Wang et al.
coli O157:H7 1.5  107 cfu mL−1 cfu mL−1 [164]
Gold nanorods/chitosan-modified ITO K562 cell 7.5  102 to 500 cells mL−1 Yu et al.
3.9  106 cells mL−1 [165]
ISE ion-selective electrode, N. A. not available, MWCNT multiwall carbon nanotubes, POT poly-3-octylthiophene film, AuNPs gold nanoparticles, SPCE
screen-printed carbon electrode, CIM colloid-imprinted mesoporous, CA125 carcinoma antigen 125, CEA carcinoembryonic antigen, AgNPs silver
nanoparticles, rGO reduced graphene oxide, ITO indium tin oxide electrode
N. Ruecha et al.
3 (Bio)Chemical Sensors Based on Paper 63

Fig. 3.15 Schematic of a POT/Au/SWCNT-modified paper-strip ion-selective electrode.

Reprinted with the permission from Mensah et al. [160]. Copyright (2014) American Chemical
Society

(SWCNTs), gold sputtering and the conductive polymer poly(3-octylthiophene)

(POT), with ISMs made from methylmethacrylate–decylmethacrylate copolymer
(PMMA-PDMA) instead of PVC/DOS [160]. A schematic of this device is shown
in Fig. 3.15. Although very low LODs achieved are not paper specific but attributed
to the selected combination of polymers (POT and PMMA-PDMA) preventing the
formation of water layers detrimental to electrode performance [161], this example
demonstrates that there are cases where paper can replace more costly substrate
materials without loss of analytical performance.
All of the ISE examples just mentioned rely on the use of external reference
electrodes, which prevents the miniaturization and the fabrication of fully integrated
paper-based potentiometric ion sensors. The year 2014 brought the first demon-
stration of an “all-paper-based” potentiometric ion sensing system with classical
plasticized PVC ISMs for potassium, sodium, calcium, and chloride ion detection
(Fig. 3.16a) [162]. While this potentiometric paper-based ion-sensor is a portable,
inexpensive, and disposable device for measuring ions of interest in aqueous
solutions, it still uses an Ag/AgCl reference electrode system requiring the addition
of KCl reference electrolyte solution by the user. To overcome this shortcoming,
Bühlmann’s group developed a stable and reproducible all-solid-state reference
electrode consisting of a colloid-imprinted mesoporous (CIM) carbon solid contact
and a PVC reference membrane [163]. The latter was doped with a moderately
hydrophilic ionic liquid and a hydrophobic redox couple to provide a constant
potential at the membrane/sample and membrane/solid contact interfaces
(Fig. 3.16b). This approach to a stable and user-friendly reference electrode is fully
compatible with paper substrates, and its analytical applicability has been demon-
strated by integration into a paper-based device for Cl− sensing.
64 N. Ruecha et al.

Fig. 3.16 a Schematic of an ion-selective sensing device made of paper and its response curve for
potassium ions; a PVC-based ion-selective membrane and an indicator electrode are sequentially
attached to the sample zone. Reprinted with the permission from Lan et al. [162]. Copyright (2014)
American Chemical Society; b schematic of paper-based device combining a chloride selective
ISE (Ag/AgCl) and a CIM carbon-based reference electrode with a reference membrane. Adapted
with the permission from Hu et al. [163]. Copyright (2015) American Chemical Society

3.6.4 Electrochemical Impedance Spectroscopy

Electrochemically inert molecules such as antibodies and antigens are not directly
detectable by electrochemical methods, but require labeling with electroactive
molecules. This problem was resolved by using electrochemical impedance spec-
troscopy (EIS), which measures the change of impedance without the labeling
requirement. While EIS has previously been mostly applied for the study of elec-
trochemical properties at biological interfaces, it has recently become widely used
for quantitative analysis based on immunoassays. This also includes the imple-
mentation on paper substrates (Table 3.5), resulting for example in the paper-based
3 (Bio)Chemical Sensors Based on Paper 65

Fig. 3.17 Illustration of the fabrication process of reduced graphene oxide paper (rGOP)
decorated with gold nanoparticles for an EIS immunoassay. Reproduced from Wang et al. [164].
Copyright (2015), with the permission from Elsevier

electrochemical impedance sensor for bacteria detection shown in Fig. 3.17 [164].
A reduced graphene oxide paper, decorated with gold nanoparticles to enhance the
specific surface area of the paper electrode, was prepared and used as substrate for
the immobilization of capture antibodies targeting bacteria (Escherichia coli O157:
H7). The changes of impedance after target bacteria capture linearly correlated with
bacteria concentration, allowing the quantification of E. coli O157:H7 with low
detection limit (1.5  102 cfu mL−1). This sensor has been successfully applied for
the detection of E. coli O157:H7 in complex food samples including ground beef
and cucumbers. Another example of an immunoassay device involving paper
coupled with EIS uses gold nanorod-modified indium tin oxide working electrodes
to detect cells down to a concentration of 500 cells mL−1 for the study of anticancer
drug toxicity [165]. The paper plays the role of a disposable, low-cost electro-
chemical cell closing the electrical circuit between working, counter and reference
electrodes.

3.7 Summary and Future Perspectives

There has been a long history of using paper as a material in the fabrication of
(bio)chemical sensing devices, but the starting shot to an almost exponential
increase in research activities on this topic occurred in 2007 with the introduction of
microfluidic paper-based analytical devices (µPADs). Made from cellulose, the
most abundant biopolymer, paper is ubiquitously available in a large variety of
grades and shapes. While the low cost of this material is the major driving force for
its currently expanding application in (bio)chemical sensing devices, the selected
66 N. Ruecha et al.

representative examples discussed in the previous sections have probably demon-

strated that there is often more than just economical reasons to refer to paper when
developing a (bio)chemical sensor. Some of the unique physical and chemical
properties of paper discussed in this chapter doubtlessly result in an added value
improving the analytical performance or the stability of the sensing devices built on
this material. Among those, the microporous structure of the cellulose fiber net-
work, giving rise to an entirely capillary flow-driven controlled and reproducible
liquid transport in a relatively thin layer, is probably the most important. In addi-
tion, the compatibility of paper with printing techniques is a big advantage in terms
of ease and reproducibility of sensor fabrication.
Laboratory filter paper is currently the most widely used paper material, due to
its purity and known composition. However, it can be expected that specifically
engineered new types of paper with designed new properties will join the list of
available materials. The declining use of paper in traditional printing-related
applications caused by the widespread use of electronic devices and media forces
the paper industry to look for alternatives. In this context, the development of
specialty papers made for (bio)chemical sensing applications might become an
attractive future business. It should be noted that already at present, countries with
strong paper-making industry are particularly active in research related to the
development of paper-based (bio)chemical sensing devices, which is certainly not a
coincidence.
The demand for low-cost and simple analytical devices suitable for untrained
general users can be expected to increase. On the one hand, there is a strong need
and also ethical obligation to provide at least a minimal degree of health care in
regions of the world that had up to now no access to diagnostic tools at all. This
field of application has been the original driving force for the development of
µPADs and has lead to the foundation of nonprofit enterprises such as “Diagnostics
for All,” a direct result of research achievements by the Whitesides’ group at
Harvard University. Paper as the most versatile and low-cost platform to build
sensing devices plays a key role in these efforts. On the other hand, the rapidly
aging societies in industrialized nations are another societal factor driving the
development of simple and low-cost analytical tools for health care-related appli-
cations. Finally, the general public in many countries is becoming increasingly
aware of issues such as environmental pollution and food safety, both of which call
for the possibility to perform simple analytical on-site tests. The analytical per-
formance achieved by using paper as a material in (bio)chemical sensing, the
demand for low-cost analytical tools by the general public, and the need for the
paper-making industry to expand into new markets could finally become a com-
bination of factors further driving research and development into paper-based
(bio)chemical sensors going beyond the optical and electrochemical devices
introduced in this chapter.
3 (Bio)Chemical Sensors Based on Paper 67

References

1. Camman K, Guibault EA, Hall H, Kellner R, Wolfbeis OS (1996) The Cambridge definition
of chemical sensors. In: Proceedings of the Cambridge workshop on chemical sensors and
biosensors, Cambridge, UK. Cambridge University Press, New York
2. Hulanicki A, Glab S, Ingman F (1991) Chemical sensors: definitions and classification. Pure
Appl Chem 63:1247
3. Spichiger-Keller UE (1998) Chemical sensors and biosensors for medical and biological
applications. Wiley-VCH, Weinheim
4. Carlisle R (2004) Scientific american inventions and discoveries. Wiley, Hoboken
5. Maumené M (1850) On a new reagent for ascertaining the presence of sugar in certain
liquids. Philos Mag Ser 3(36):482
6. Oliver G (1883) On bedside urinary tests. Lancet 121:139–140
7. Comer JP (1956) Semiquantitative specific test paper for glucose in urine. Anal Chem
28:1748–1750
8. Free AH, Adams EC, Kercher ML, Free HM, Cook MH (1957) Simple specific test for urine
glucose. Clin Chem 3:163–168
9. Hawkes R, Niday E, Gordon J (1982) A dot-immunobinding assay for monoclonal and other
antibodies. Anal Biochem 119:142–147
10. Martinez AW, Phillips ST, Butte MJ, Whitesides GM (2007) Patterned paper as a platform
for inexpensive, low-volume, portable bioassays. Angew Chem Int Ed 46:1318–1320
11. Müller RH, Clegg DL (1949) Automatic paper chromatography. Anal Chem 21:1123–1125
12. Martinez AW, Phillips ST, Wiley BJ, Gupta M, Whitesides GM (2008) FLASH: a rapid
method for prototyping paper-based microfluidic devices. Lab Chip 8:2146–2150
13. Klemm D, Schmauder H-P, Heinze T (2005) Cellulose. In: Biopolymers online. Wiley-VCH
Verlag GmbH & Co. KGaA. doi:10.1002/3527600035.bpol6010
14. Credou J, Berthelot T (2014) Cellulose: from biocompatible to bioactive material. J Mater
Chem B 2:4767–4788
15. Pelton R (2009) Bioactive paper provides a low-cost platform for diagnostics. Trends Anal
Chem 28:925–942
16. Kong F, Hu Y (2012) Biomolecule immobilization techniques for bioactive paper
fabrication. Anal Bioanal Chem 403:7–13
17. Thorn I, Au CO (eds) (2009) Applications of wet-end paper chemistry, 2nd edn. Springer,
Netherlands
18. Nery EW, Kubota LT (2013) Sensing approaches on paper-based devices: a review. Anal
Bioanal Chem 405:7573–7595
19. Masoodi R, Pillai KM (2010) Darcy’s law-based model for wicking in paper-like swelling
porous media. AIChE J 56:2257–2267
20. Yamada K, Henares TG, Suzuki K, Citterio D (2015) Paper-based inkjet-printed
microfluidic analytical devices. Angew Chem Int Ed 54:5294–5310
21. Dungchai W, Chailapakul O, Henry CS (2011) A low-cost, simple, and rapid fabrication
method for paper-based microfluidics using wax screen-printing. Analyst 136:77–82
22. Sameenoi Y, Nongkai PN, Nouanthavong S, Henry CS, Nacapricha D (2014) One-step
polymer screen-printing for microfluidic paper-based analytical device (µPAD) fabrication.
Analyst 139:6580–6588
23. Abe K, Suzuki K, Citterio D (2008) Inkjet-printed microfluidic multianalyte chemical
sensing paper. Anal Chem 80:6928–6934
24. Li X, Tian J, Garnier G, Shen W (2010) Fabrication of paper-based microfluidic sensors by
printing. Colloids Surf B: Biointerfaces 76:564–570
25. Maejima K, Tomikawa S, Suzuki K, Citterio D (2013) Inkjet printing: an integrated and
green chemical approach to microfluidic paper-based analytical devices. RSC Adv 3:9258–
9263
68 N. Ruecha et al.

26. Wang J, Monton MRN, Zhang X, Filipe CDM, Pelton R, Brennan JD (2014) Hydrophobic
sol-gel channel patterning strategies for paper-based microfluidics. Lab Chip 14:691–695
27. Rajendra V, Sicard C, Brennan JD, Brook MA (2014) Printing silicone-based hydrophobic
barriers on paper for microfluidic assays using low-cost ink jet printers. Analyst 139:6361–
6365
28. Lu Y, Shi W, Jiang L, Qin J, Lin B (2009) Rapid prototyping of paper-based microfluidics
with wax for low-cost, portable bioassay. Electrophoresis 30:1497–1500
29. Carrilho E, Martinez AW, Whitesides GM (2009) Understanding wax printing: a simple
micropatterning process for paper-based microfluidics. Anal Chem 81:7091–7095
30. Renault C, Koehne J, Ricco AJ, Crooks RM (2014) Three-dimensional wax patterning of
paper fluidic devices. Langmuir 30:7030–7036
31. Olkkonen J, Lehtinen K, Erho T (2010) Flexographically printed fluidic structures in paper.
Anal Chem 82:10246–10250
32. Maattanen A, Fors D, Wang S, Valtakari D, Ihalainen P, Peltonen J (2011) Paper-based
planar reaction arrays for printed diagnostics. Sens Actuat B: Chem 160:1404–1412
33. Abe K, Kotera K, Suzuki K, Citterio D (2010) Inkjet-printed paperfluidic immuno-chemical
sensing device. Anal Bioanal Chem 398:885–893
34. Yamada K, Takaki S, Komuro N, Suzuki K, Citterio D (2014) An antibody-free microfluidic
paper-based analytical device for the determination of tear fluid lactoferrin by fluorescence
sensitization of Tb3+. Analyst 139:1637–1643
35. Yetisen AK, Akram MS, Lowe CR (2013) Paper-based microfluidic point-of-care diagnostic
devices. Lab Chip 13:2210–2251
36. Cate DM, Adkins JA, Mettakoonpitak J, Henry CS (2015) Recent developments in
paper-based microfluidic devices. Anal Chem 87:19–41
37. Xia Y, Si J, Li Z (2015) Fabrication techniques for microfluidic paper-based analytical
devices and their applications for biological testing: a review. Biosens Bioelectron 77:774–
789
38. Evans E, Gabriel EFM, Coltro WKT, Garcia CD (2014) Rational selection of substrates to
improve color intensity and uniformity on microfluidic paper-based analytical devices.
Analyst 139:2127–2132
39. Evans E, Moreira Gabriel EF, Benavidez TE, Tomazelli Coltro WK, Garcia CD (2014)
Modification of microfluidic paper-based devices with silica nanoparticles. Analyst
139:5560–5567
40. Mentele MM, Cunningham J, Koehler K, Volckens J, Henry CS (2012) Microfluidic
paper-based analytical device for particulate metals. Anal Chem 84:4474–4480
41. Vella SJ, Beattie P, Cademartiri R, Laromaine A, Martinez AW, Phillips ST, Mirica KA,
Whitesides GM (2012) Measuring markers of liver function using a micropatterned paper
device designed for blood from a fingerstick. Anal Chem 84:2883–2891
42. Wang X, Qin Y, Meyerhoff ME (2015) Paper-based plasticizer-free sodium ion-selective
sensor with camera phone as a detector. Chem Commun 51:15176–15179
43. Bakker E, Bühlmann P, Pretsch E (1997) Carrier-based ion-selective electrodes and bulk
optodes. 1. General characteristics. Chem Rev 97:3083–3132
44. Hossain SM, Luckham RE, McFadden MJ, Brennan JD (2009) Reagentless bidirectional
lateral flow bioactive paper sensors for detection of pesticides in beverage and food samples.
Anal Chem 81:9055–9064
45. Hossain SM, Luckham RE, Smith AM, Lebert JM, Davies LM, Pelton RH, Filipe CD,
Brennan JD (2009) Development of a bioactive paper sensor for detection of neurotoxins
using piezoelectric inkjet printing of sol-gel-derived bioinks. Anal Chem 81:5474–5483
46. Wang J, Bowie D, Zhang X, Filipe C, Pelton R, Brennan JD (2014) Morphology and
entrapped enzyme performance in inkjet-printed sol–gel coatings on paper. Chem Mater
26:1941–1947
47. Lahr RH, Wallace GC, Vikesland PJ (2015) Raman characterization of nanoparticle
transport in microfluidic paper-based analytical devices (lPADs). ACS Appl Mater
Interfaces 7:9139–9146
3 (Bio)Chemical Sensors Based on Paper 69

48. Su S, Nutiu R, Filipe CDM, Li Y, Pelton R (2007) Adsorption and covalent coupling of
ATP-binding DNA aptamers onto cellulose. Langmuir 23:1300–1302
49. Tyagi C, Tomar LK, Singh H (2009) Surface modification of cellulose filter paper by
glycidyl methacrylate grafting for biomolecule immobilization: Influence of grafting
parameters and urease immobilization. J Appl Polym Sci 111:1381–1390
50. Zhu Y, Xu X, Brault ND, Keefe AJ, Han X, Deng Y, Xu J, Yu Q, Jiang S (2014) Cellulose
paper sensors modified with zwitterionic poly(carboxybetaine) for sensing and detection in
complex media. Anal Chem 86:2871–2875
51. Yu A, Shang J, Cheng F, Paik BA, Kaplan JM, Andrade RB, Ratner DM (2012)
Biofunctional paper via the covalent modification of cellulose. Langmuir 28:11265–11273
52. Cheng CM, Martinez AW, Gong J, Mace CR, Phillips ST, Carrilho E, Mirica KA,
Whitesides GM (2010) Paper-based ELISA. Angew Chem Int Ed 49:4771–4774
53. Apilux A, Ukita Y, Chikae M, Chailapakul O, Takamura Y (2013) Development of
automated paper-based devices for sequential multistep sandwich enzyme-linked
immunosorbent assays using inkjet printing. Lab Chip 13:126–135
54. Carpenter JF, Crowe JH (1988) Modes of stabilization of a protein by organic solutes during
desiccation. Cryobiology 25:459–470
55. Carpenter JF, Crowe LM, Crowe JH (1987) Stabilization of phosphofructokinase with
sugars during freeze-drying: characterization of enhanced protection in the presence of
divalent cations. Biochim Biophys Acta 923:109–115
56. Carpenter JF, Martin B, Crowe LM, Crowe JH (1987) Stabilization of phosphofructokinase
during air-drying with sugars and sugar/transition metal mixtures. Cryobiology 24:455–464
57. Avnir D, Braun S, Lev O, Ottolenghi M (1994) Enzymes and other proteins entrapped in
sol-gel materials. Chem Mater 6:1605–1614
58. Gill I, Ballesteros A (2000) Bioencapsulation within synthetic polymers (Part 1): sol-gel
encapsulated biologicals. Trends Biotechnol 18:282–296
59. Monton MRN, Forsberg EM, Brennan JD (2012) Tailoring sol-gel-derived silica materials
for optical biosensing. Chem Mater 24:796–811
60. Hossain SMZ, Brennan JD (2011) b-galactosidase-based colorimetric paper sensor for
determination of heavy metals. Anal Chem 83:8772–8778
61. Alkasir RSJ, Ornatska M, Andreescu S (2012) Colorimetric paper bioassay for the detection
of phenolic compounds. Anal Chem 84:9729–9737
62. Ramachandran S, Fu E, Lutz B, Yager P (2014) Long-term dry storage of an enzyme-based
reagent system for ELISA in point-of-care devices. Analyst 139:1456–1462
63. Mitchell HT, Noxon IC, Chaplan CA, Carlton SJ, Liu CH, Ganaja KA, Martinez NW,
Immoos CE, Costanzo PJ, Martinez AW (2015) Reagent pencils: a new technique for
solvent-free deposition of reagents onto paper-based microfluidic devices. Lab Chip
15:2213–2220
64. Wu G, Srivastava J, Zaman MH (2014) Stability measurements of antibodies stored on
paper. Anal Biochem 449:147–154
65. Kwon H, Samain F, Kool ET (2012) Fluorescent DNAs printed on paper: sensing food
spoilage and ripening in the vapor phase. Chem Sci 3:2542–2549
66. Rosa AM, Louro AF, Martins SA, Inacio J, Azevedo AM, Prazeres DM (2014) Capture and
detection of DNA hybrids on paper via the anchoring of antibodies with fusions of
carbohydrate binding modules and ZZ-domains. Anal Chem 86:4340–4347
67. He M, Liu Z (2013) Paper-based microfluidic device with upconversion fluorescence assay.
Anal Chem 85:11691–11694
68. Noor MO, Krull UJ (2014) Camera-based ratiometric fluorescence transduction of nucleic
acid hybridization with reagentless signal amplification on a paper-based platform using
immobilized quantum dots as donors. Anal Chem 86:10331–10339
69. Zhou F, Noor MO, Krull UJ (2014) Luminescence resonance energy transfer-based nucleic
acid hybridization assay on cellulose paper with upconverting phosphor as donors. Anal
Chem 86:2719–2726
70 N. Ruecha et al.

70. Scida K, Li B, Ellington AD, Crooks RM (2013) DNA detection using origami paper
analytical devices. Anal Chem 85:9713–9720
71. Taudte RV, Beavis A, Wilson-Wilde L, Roux C, Doble P, Blanes L (2013) A portable
explosive detector based on fluorescence quenching of pyrene deposited on coloured
wax-printed µPADs. Lab Chip 13:4164–4172
72. Hatakeyama K, Tanaka T, Sawaguchi M, Iwadate A, Mizutani Y, Sasaki K, Tateishi N,
Matsunaga T (2009) Microfluidic device using chemiluminescence and a DNA-arrayed thin
film transistor photosensor for single nucleotide polymorphism genotyping of PCR
amplicons from whole blood. Lab Chip 9:1052–1058
73. Yu J, Wang S, Ge L, Ge S (2011) A novel chemiluminescence paper microfluidic biosensor
based on enzymatic reaction for uric acid determination. Biosens Bioelectron 26:3284–3289
74. Yu J, Ge L, Huang J, Wang S, Ge S (2011) Microfluidic paper-based chemiluminescence
biosensor for simultaneous determination of glucose and uric acid. Lab Chip 11:1286–1291
75. Wang S, Ge L, Song X, Yan M, Ge S, Yu J, Zeng F (2012) Simple and covalent fabrication
of a paper device and its application in sensitive chemiluminescence immunoassay. Analyst
137:3821–3827
76. Ge L, Wang S, Song X, Ge S, Yu J (2012) 3D Origami-based multifunction-integrated
immunodevice: low-cost and multiplexed sandwich chemiluminescence immunoassay on
microfluidic paper-based analytical device. Lab Chip 12:3150–3158
77. Wang S, Ge L, Song X, Yu J, Ge S, Huang J, Zeng F (2012) Paper-based chemilumi-
nescence ELISA: Lab-on-paper based on chitosan modified paper device and
wax-screen-printing. Biosens Bioelectron 31:212–218
78. Wang Y, Wang S, Ge S, Wang S, Yan M, Zang D, Yu J (2013) Facile and sensitive
paper-based chemiluminescence DNA biosensor using carbon dots dotted nanoporous gold
signal amplification label. Anal Methods 5:1328–1336
79. Liu W, Cassano CL, Xu X, Fan ZH (2013) Laminated paper-based analytical devices
(LPAD) with origami-enabled chemiluminescence immunoassay for cotinine detection in
mouse serum. Anal Chem 85:10270–10276
80. Liu W, Kou J, Xing H, Li B (2014) Paper-based chromatographic chemiluminescence chip
for the detection of dichlorvos in vegetables. Biosens Bioelectron 52:76–81
81. Liu W, Luo J, Guo Y, Kou J, Li B, Zhang Z (2014) Nanoparticle coated paper-based
chemiluminescence device for the determination of L-cysteine. Talanta 120:336–341
82. Wang Y, Wang S, Ge S, Wang S, Yan M, Zang D, Yu J (2014) Ultrasensitive
chemiluminescence detection of DNA on a microfluidic paper-based analytical device.
Monatsh Chem 145:129–135
83. Liu W, Guo Y, Li H, Zhao M, Lai Z, Li B (2015) A paper-based chemiluminescence device
for the determination of ofloxacin. Spectrochim Acta, Part A 137:1298–1303
84. Chen X, Luo Y, Shi B, Liu X, Gao Z, Du Y, Zhao W, Lin B (2014) Chemiluminescence
diminishment on a paper-based analytical device: high throughput determination of
b-agonists in swine hair. Anal Methods 6:9684–9690
85. Liu F, Zhang C (2015) A novel paper-based microfluidic enhanced chemiluminescence
biosensor for facile, reliable and highly-sensitive gene detection of Listeria monocytogenes.
Sens Actuat B: Chem 209:399–406
86. Liu W, Guo Y, Luo J, Kou J, Zheng H, Li B, Zhang Z (2015) A molecularly imprinted
polymer based a lab-on-paper chemiluminescence device for the detection of dichlorvos.
Spectrochim Acta, Part A 141:51–57
87. Lebiga E, Edwin Fernandez R, Beskok A (2015) Confined chemiluminescence detection of
nanomolar levels of H2O2 in a paper-plastic disposable microfluidic device using a
smartphone. Analyst 140:5006–5011
88. Baptista FR, Belhout SA, Giordani S, Quinn SJ (2015) Recent developments in carbon
nanomaterial sensors. Chem Soc Rev 44:4433–4453
89. Yuan C, Liu B, Liu F, Han MY, Zhang Z (2014) Fluorescence “turn on” detection of
mercuric ion based on bis(dithiocarbamato)copper(II) complex functionalized carbon
nanodots. Anal Chem 86:1123–1130
3 (Bio)Chemical Sensors Based on Paper 71

90. Kim Y, Jang G, Lee TS (2015) New fluorescent metal-ion detection using a paper-based
sensor strip containing tethered rhodamine carbon nanodots. ACS Appl Mater Interfaces
7:15649–15657
91. Delaney JL, Hogan CF, Tian J, Shen W (2011) Electrogenerated chemiluminescence
detection in paper-based microfluidic sensors. Anal Chem 83:1300–1306
92. Wang S, Ge L, Zhang Y, Song X, Li N, Ge S, Yu J (2012) Battery-triggered microfluidic
paper-based multiplex electrochemiluminescence immunodevice based on
potential-resolution strategy. Lab Chip 12:4489–4498
93. Ge L, Yan J, Song X, Yan M, Ge S, Yu J (2012) Three-dimensional paper-based
electrochemiluminescence immunodevice for multiplexed measurement of biomarkers and
point-of-care testing. Biomaterials 33:1024–1031
94. Yan J, Ge L, Song X, Yan M, Ge S, Yu J (2012) Paper-based electrochemiluminescent 3D
immunodevice for lab-on-paper, specific, and sensitive point-of-care testing. Chem Eur J
18:4938–4945
95. Wang S, Dai W, Ge L, Yan M, Yu J, Song X, Ge S, Huang J (2012) Rechargeable
battery-triggered electrochemiluminescence detection on microfluidic origami immunode-
vice based on two electrodes. Chem Commun 48:9971–9973
96. Zhang M, Ge L, Ge S, Yan M, Yu J, Huang J, Liu S (2013) Three-dimensional paper-based
electrochemiluminescence device for simultaneous detection of Pb2+ and Hg2+ based on
potential-control technique. Biosens Bioelectron 41:544–550
97. Zhang L, Li L, Ma C, Ge S, Yan M, Bian C (2015) Detection of a-fetoprotein with an
ultrasensitive electrochemiluminescence paper device based on green-luminescent
nitrogen-doped graphene quantum dots. Sens Actuat B: Chem 221:799–806
98. Li W, Li L, Ge S, Song X, Ge L, Yan M, Yu J (2013) A 3D origami multiple
electrochemiluminescence immunodevice based on a porous silver-paper electrode and
multi-labeled nanoporous gold-carbon spheres. Chem Commun 49:7687–7689
99. Li W, Li L, Li S, Wang X, Li M, Wang S, Yu J (2013) 3D origami electrochemilumi-
nescence immunodevice based on porous silver-paper electrode and nanoporous silver
double-assisted signal amplification. Sens Actuat B: Chem 188:417–424
100. Shi CG, Shan X, Pan ZQ, Xu JJ, Lu C, Bao N, Gu HY (2012) Quantum dot (QD)-modified
carbon tape electrodes for reproducible electrochemiluminescence (ECL) emission on a
paper-based platform. Anal Chem 84:3033–3038
101. Li W, Li M, Ge S, Yan M, Huang J, Yu J (2013) Battery-triggered ultrasensitive
electrochemiluminescence detection on microfluidic paper-based immunodevice based on
dual-signal amplification strategy. Anal Chim Acta 767:66–74
102. Yan J, Yan M, Ge L, Yu J, Ge S, Huang J (2013) A microfluidic origami electrochemi-
luminescence aptamer-device based on a porous Au-paper electrode and a phenyle-
neethynylene derivative. Chem Commun 49:1383–1385
103. Mani V, Kadimisetty K, Malla S, Joshi AA, Rusling JF (2013) Paper-based electrochemi-
luminescent screening for genotoxic activity in the environment. Environ Sci Technol
47:1937–1944
104. Ellerbee AK, Phillips ST, Siegel AC, Mirica KA, Martinez AW, Striehl P, Jain N,
Prentiss M, Whitesides GM (2009) Quantifying colorimetric assays in paper-based
microfluidic devices by measuring the transmission of light through paper. Anal Chem
81:8447–8452
105. Li B, Fu L, Zhang W, Feng W, Chen L (2014) Portable paper-based device for quantitative
colorimetric assays relying on light reflectance principle. Electrophoresis 35:1152–1159
106. Lewis GG, DiTucci MJ, Phillips ST (2012) Quantifying analytes in paper-based microfluidic
devices without using external electronic readers. Angew Chem Int Ed 51:12707–12710
107. Lewis GG, Robbins JS, Phillips ST (2013) Point-of-care assay platform for quantifying
active enzymes to femtomolar levels using measurements of time as the readout. Anal Chem
85:10432–10439
72 N. Ruecha et al.

108. Lewis GG, Robbins JS, Phillips ST (2014) A prototype point-of-use assay for measuring
heavy metal contamination in water using time as a quantitative readout. Chem Commun
50:5352–5354
109. Zhang Y, Fan J, Nie J, Le S, Zhu W, Gao D, Yang J, Zhang S, Li J (2015) Timing readout in
paper device for quantitative point-of-use hemin/G-quadruplex DNAzyme-based bioassays.
Biosens Bioelectron 73:13–18
110. Cate DM, Dungchai W, Cunningham JC, Volckens J, Henry CS (2013) Simple,
distance-based measurement for paper analytical devices. Lab Chip 13:2397–2404
111. Dungchai W, Sameenoi Y, Chailapakul O, Volckens J, Henry CS (2013) Determination of
aerosol oxidative activity using silver nanoparticle aggregation on paper-based analytical
devices. Analyst 138:6766–6773
112. Cate DM, Noblitt SD, Volckens J, Henry CS (2015) Multiplexed paper analytical device for
quantification of metals using distance-based detection. Lab Chip 15:2808–2818
113. Yamada K, Henares TG, Suzuki K, Citterio D (2015) Distance-based tear lactoferrin assay
on microfluidic paper device using interfacial interactions on surface-modified cellulose.
ACS Appl Mater Interfaces 7:24864–24875
114. Chen Y-T, Yang J-T (2015) Detection of an amphiphilic biosample in a paper microchannel
based on length. Biomed Microdevices 17:1–8
115. Fang X, Zhao Q, Cao H, Liu J, Guan M, Kong J (2015) Rapid detection of Cu by a
paper-based microfluidic device coated with bovine serum albumin (BSA)-Au nanoclusters.
Analyst 140:7823–7826
116. Li M, Tian J, Al-Tamimi M, Shen W (2012) Paper-based blood typing device that reports
patient’s blood type “in writing”. Angew Chem Int Ed 51:5497–5501
117. Teichler A, Perelaer J, Schubert US (2013) Inkjet printing of organic electronics -
comparison of deposition techniques and state-of-the-art developments. J Mater Chem C
1:1910–1925
118. Dungchai W, Chailapakul O, Henry CS (2009) Electrochemical detection for paper-based
microfluidics. Anal Chem 81:5821–5826
119. Ruecha N, Rangkupan R, Rodthongkum N, Chailapakul O (2014) Novel paper-based
cholesterol biosensor using graphene/polyvinylpyrrolidone/polyaniline nanocomposite.
Biosens Bioelectron 52:13–19
120. Liana DD, Raguse B, Wieczorek L, Baxter GR, Chuah K, Gooding JJ, Chow E (2013)
Sintered gold nanoparticles as an electrode material for paper-based electrochemical sensors.
RSC Adv 3:8683–8691
121. Jenkins G, Wang Y, Xie Y, Wu Q, Huang W, Wang L, Yang X (2015) Printed electronics
integrated with paper-based microfluidics: new methodologies for next-generation health
care. Microfluid Nanofluidics 19:251–261
122. Santhiago M, Henry CS, Kubota LT (2014) Low cost, simple three dimensional
electrochemical paper-based analytical device for determination of p-nitrophenol.
Electrochim Acta 130:771–777
123. Dossi N, Toniolo R, Piccin E, Susmel S, Pizzariello A, Bontempelli G (2013) Pencil-drawn
dual electrode detectors to discriminate between analytes comigrating on paper-based fluidic
devices but undergoing electrochemical processes with different reversibility.
Electroanalysis 25:2515–2522
124. Shiroma LY, Santhiago M, Gobbi AL, Kubota LT (2012) Separation and electrochemical
detection of paracetamol and 4-aminophenol in a paper-based microfluidic device. Anal
Chim Acta 725:44–50
125. Fosdick SE, Anderson MJ, Renault C, DeGregory PR, Loussaert JA, Crooks RM (2014)
Wire, mesh, and fiber electrodes for paper-based electroanalytical devices. Anal Chem
86:3659–3666
126. Adkins JA, Henry CS (2015) Electrochemical detection in paper-based analytical devices
using microwire electrodes. Anal Chim Acta 891:247–254
3 (Bio)Chemical Sensors Based on Paper 73

127. Nie Z, Nijhuis CA, Gong J, Chen X, Kumachev A, Martinez AW, Narovlyansky M,
Whitesides GM (2010) Electrochemical sensing in paper-based microfluidic devices. Lab
Chip 10:477–483
128. Chen Z, Martin MT, Xinyu L (2013) A microfluidic paper-based electrochemical biosensor
array for multiplexed detection of metabolic biomarkers. Sci Technol Adv Mat 14:054402
129. Li Z, Li F, Hu J, Wee WH, Han YL, Pingguan-Murphy B, Lu TJ, Xu F (2015) Direct writing
electrodes using a ball pen for paper-based point-of-care testing. Analyst 140:5526–5535
130. Liu H, Crooks RM (2012) Paper-based electrochemical sensing platform with integral
battery and electrochromic read-out. Anal Chem 84:2528–2532
131. Nie Z, Deiss F, Liu X, Akbulut O, Whitesides GM (2010) Integration of paper-based
microfluidic devices with commercial electrochemical readers. Lab Chip 10:3163–3169
132. Noiphung J, Songjaroen T, Dungchai W, Henry CS, Chailapakul O, Laiwattanapaisal W
(2013) Electrochemical detection of glucose from whole blood using paper-based
microfluidic devices. Anal Chim Acta 788:39–45
133. Songjaroen T, Dungchai W, Chailapakul O, Henry CS, Laiwattanapaisal W (2012) Blood
separation on microfluidic paper-based analytical devices. Lab Chip 12:3392–3398
134. Kuek Lawrence CS, Tan SN, Floresca CZ (2014) A “green” cellulose paper based glucose
amperometric biosensor. Sens Actuat B: Chem 193:536–541
135. Chandra Sekar N, Mousavi Shaegh SA, Ng SH, Ge L, Tan SN (2014) A paper-based
amperometric glucose biosensor developed with prussian blue-modified screen-printed
electrodes. Sens Actuat B: Chem 204:414–420
136. Arya SK, Dey A, Bhansali S (2011) Polyaniline protected gold nanoparticles based mediator
and label free electrochemical cortisol biosensor. Biosens Bioelectron 28:166–173
137. Xian Y, Hu Y, Liu F, Xian Y, Wang H, Jin L (2006) Glucose biosensor based on Au
nanoparticles-conductive polyaniline nanocomposite. Biosens Bioelectron 21:1996–2000
138. Hu C, Bai X, Wang Y, Jin W, Zhang X, Hu S (2012) Inkjet printing of nanoporous gold
electrode arrays on cellulose membranes for high-sensitive paper-like electrochemical
oxygen sensors using ionic liquid electrolytes. Anal Chem 84:3745–3750
139. Tiwari P, Kumar A, Prakash R (2015) Electrochemical detection of azidothymidine on
modified probes based on chitosan stabilised silver nanoparticles hybrid material. RSC Adv
5:90089–90097
140. Shah B, Lafleur T, Chen A (2013) Carbon nanotube based electrochemical sensor for the
sensitive detection of valacyclovir. Faraday Discuss 164:135–146
141. da Costa TH, Song E, Tortorich RP, Choi J-W (2015) A paper-based electrochemical sensor
using inkjet-printed carbon nanotube electrodes. ECS J Solid State Sci Technol 4:S3044–
S3047
142. Li X, Zhao C, Liu X (2015) A paper-based microfluidic biosensor integrating zinc oxide
nanowires for electrochemical glucose detection. Microsystems Nanoeng 1:15014
143. Rattanarat P, Dungchai W, Siangproh W, Chailapakul O, Henry CS (2012) Sodium dodecyl
sulfate-modified electrochemical paper-based analytical device for determination of
dopamine levels in biological samples. Anal Chim Acta 744:1–7
144. Wang P, Ge L, Yan M, Song X, Ge S, Yu J (2012) Paper-based three-dimensional
electrochemical immunodevice based on multi-walled carbon nanotubes functionalized
paper for sensitive point-of-care testing. Biosens Bioelectron 32:238–243
145. Cunningham JC, Scida K, Kogan MR, Wang B, Ellington AD, Crooks RM (2015) Paper
diagnostic device for quantitative electrochemical detection of ricin at picomolar levels. Lab
Chip 15:3707–3715
146. Wu Y, Xue P, Kang Y, Hui KM (2013) Paper-based microfluidic electrochemical
immunodevice integrated with nanobioprobes onto graphene film for ultrasensitive
multiplexed detection of cancer biomarkers. Anal Chem 85:8661–8668
147. Lu J, Ge S, Ge L, Yan M, Yu J (2012) Electrochemical DNA sensor based on
three-dimensional folding paper device for specific and sensitive point-of-care testing.
Electrochim Acta 80:334–341
74 N. Ruecha et al.

148. Cunningham JC, Brenes NJ, Crooks RM (2014) Paper electrochemical device for detection
of DNA and thrombin by target-induced conformational switching. Anal Chem 86:6166–
6170
149. Rattanarat P, Dungchai W, Cate D, Volckens J, Chailapakul O, Henry CS (2014) Multilayer
paper-based device for colorimetric and electrochemical quantification of metals. Anal Chem
86:3555–3562
150. Shi J, Tang F, Xing H, Zheng H, Lianhua B, Wei W (2012) Electrochemical detection of Pb
and Cd in paper-based microfluidic devices. J Braz Chem Soc 23:1124–1130
151. Cuartero M, Crespo GA, Bakker E (2015) Paper-based thin-layer coulometric sensor for
halide determination. Anal Chem 87:1981–1990
152. Zang D, Ge L, Yan M, Song X, Yu J (2012) Electrochemical immunoassay on a 3D
microfluidic paper-based device. Chem Commun 48:4683–4685
153. Li X, Xia J, Li W, Zhang S (2010) Multianalyte electrochemical biosensor based on
aptamer- and nanoparticle-integrated bio-barcode amplification. Chem Asian J 5:294–300
154. Li M, Cao R, Nilghaz A, Guan L, Zhang X, Shen W (2015) “Periodic-table-style” paper
device for monitoring heavy metals in water. Anal Chem 87:2555–2559
155. Feng L, Li X, Li H, Yang W, Chen L, Guan Y (2013) Enhancement of sensitivity of
paper-based sensor array for the identification of heavy-metal ions. Anal Chim Acta 780:74–
80
156. Ruecha N, Rodthongkum N, Cate DM, Volckens J, Chailapakul O, Henry CS (2015)
Sensitive electrochemical sensor using a graphene–polyaniline nanocomposite for simulta-
neous detection of Zn(II), Cd(II), and Pb(II). Anal Chim Acta 874:40–48
157. Cui J, Lisak G, Strzalkowska S, Bobacka J (2014) Potentiometric sensing utilizing
paper-based microfluidic sampling. Analyst 139:2133–2136
158. Lisak G, Cui J, Bobacka J (2015) Paper-based microfluidic sampling for potentiometric
determination of ions. Sens Actuat B: Chem 207:933–939
159. Novell M, Parrilla M, Crespo GA, Rius FX, Andrade FJ (2012) Paper-based ion-selective
potentiometric sensors. Anal Chem 84:4695–4702
160. Mensah ST, Gonzalez Y, Calvo-Marzal P, Chumbimuni-Torres KY (2014) Nanomolar
detection limits of Cd2+, Ag+, and K+ using paper-strip ion-selective electrodes. Anal Chem
86:7269–7273
161. Veder JP, De Marco R, Clarke G, Chester R, Nelson A, Prince K, Pretsch E, Bakker E
(2008) Elimination of undesirable water layers in solid-contact polymeric ion-selective
electrodes. Anal Chem 80:6731–6740
162. Lan WJ, Zou XU, Hamedi MM, Hu J, Parolo C, Maxwell EJ, Bühlmann P, Whitesides GM
(2014) Paper-based potentiometric ion sensing. Anal Chem 86:9548–9553
163. Hu J, Ho KT, Zou XU, Smyrl WH, Stein A, Bühlmann P (2015) All-solid-state reference
electrodes based on colloid-imprinted mesoporous carbon and their application in disposable
paper-based potentiometric sensing devices. Anal Chem 87:2981–2987
164. Wang Y, Ping J, Ye Z, Wu J, Ying Y (2013) Impedimetric immunosensor based on gold
nanoparticles modified graphene paper for label-free detection of Escherichia coli O157:H7.
Biosens Bioelectron 49:492–498
165. Yu C, Wang Q, Li W, Li Y, Liu S, Bao N, Gu H (2015) Paper-based cell impedance sensor
and its application for cytotoxic evaluation. Nanotechnology 26:325501
Chapter 4
Membrane Technologies for Sensing
and Biosensing

Subrayal Medapati Reddy

4.1 Synthetic Polymeric Membranes

Membranes are ubiquitous in life encapsulating and regulating the content of all cells
and have pervaded society, from large-scale industrial such as seawater desalination to
medical applications such as haemodialysis [1–3]. Membranes provide an essential
barrier between two phases, in situations where the mixing between the two phases
needs to be somehow regulated. They invariably allow controlled transport of
chemical species between the two phases. While cell membranes are composed of a
metastable lipid bilayer, those used in industrial applications are generally based on
more robust synthetic polymeric materials and are inherently more pH and temperature
stable and hardy [4–6].
Chemical and biosensors are devices designed to be portable and low-cost and able
to measure chemical and biological species to inform the user of important changes,
whether it be in a biological fluid or environmental sample [7–9]. Whereas chemical
sensors can demonstrate broad-band selectivity to a range of chemicals [10, 11],
biosensors are a little more sophisticated in their specificity for a particular chemical,
by virtue of using an in-built biological recognition entity (such as an enzyme or an
antibody) to specifically recognise the target chemical of interest [12, 13]. In either
case, membranes have also been incorporated into the structure in order to improve the
performance and reliability of the sensor. As an example, ion-selective electrodes
(ISEs) have demonstrated high selectivity to the target ion of interest by virtue of the
specific composition of the ion-selective membrane [14, 15].
Biosensors have been researched for over 50 years, the most renowned appli-
cation still being the glucose biosensor which has had international commercial
success [16]. Other successes have been few and far between due to issues of sensor
signal stability, longevity and matrix effects especially when the sensor is exposed

S.M. Reddy (&)

Chemistry Division, University of Central Lancashire, Preston, Lancashire PR2 2HE, UK
e-mail: [email protected]

© Springer International Publishing AG 2017 75

T.R.L.C. Paixão and S.M. Reddy (eds.), Materials for Chemical Sensing,
DOI 10.1007/978-3-319-47835-7_4
76 S.M. Reddy

to complex biological samples such as blood and seawater for periods longer than a
few seconds [17–20].
Membranes can be produced from virtually any polymer. The two common
techniques for producing membranes from preformed polymers is solvent-casting
[21, 22] and melt extrusion [23]. The former relies on dissolving the polymer in a
suitable compatible solvent and then casting the solution on a solid surface (such as
glass). Desolvation of the solvent results in precipitation of the polymer in the form
of a thin, free-standing film [24–26]. Spin coating of polymer solutions has been
used to produce thinner films, the thickness being dependent on solution viscosity
and spin speed [27, 28]. Melt extrusion relies on heating the polymer above its glass
transition temperature. Above this temperature, the polymer enters a semi-fluidic
state where polymer chains move freely over each other as one mass. This mass of
material can then be moulded, shaped and pressed to form membranes [23].
Membranes can be produced from single polymers or copolymers as well as
polymer blends [29–31], the latter resulting in multiple phases within one mem-
brane structure. When the membrane forms one continuous phase, it is said to be
homogenous. A membrane that contains more than one phase such as a microp-
orous is said to be heterogeneous. Membranes used in sensing and biosensing need
to possess some key features. They need to be semi-permeable and also need to be
able to retain their structural integrity. In some cases, these membranes have been
developed to be stimuli-responsive [32] and change can their structure and shape as
a function of microenvironmental changes.
This chapter reviews membrane-based biosensor techniques and focuses on the
recent applications of synthetic membrane systems for sensor and biosensor
applications.

4.1.1 Nanoporous Membranes

The ‘nano’ word has percolated membrane research to describe the developments in
the applications of membranes to chemical and biosensing. Nanoporosity refers to
pore sizes ranging from 10 nm up to a few hundred nanometres. Nanoporous
membranes have been used in biosensor strategies to restrict the diffusion of
interfering large molecular weight biologicals (such as proteins), while still
allowing facile diffusion of low molecular weight solutes such as metabolites to the
lower layers of the sensor for biospecific signal transduction and small molecule
detection [25, 33, 34]. They have also been used to selectively capture proteins for
single-molecule detection strategies [35] as well as optical and electrochemical
sensing of protons, metal ions and small biomolecules within polymer
nanopores/nanochannels [36]. Nanopores can be introduced into polymer mem-
branes in a controlled fashion through (a) focused ion-beam [37], (b) nuclear track
etching [38, 39] or (c) dissolution of one soluble polymer component of a di-block
copolymer [40]. Nanopore size and density for the nuclear track-etched membranes
4 Membrane Technologies for Sensing and Biosensing 77

depend on the duration of radiation exposure as well as the duration of etching.

With di-block copolymers, the porosity is a function of the ratio of the concen-
tration of the two polymers used.
Those readers interested in the production of pores within foams are directed to a
review by Aram et al. [41], and the preparation and biomedical applications of
nanoporous anodised aluminium oxide [42] are discussed later in this chapter.

4.2 ISEs/Amperometric Sensors

Ion-selective electrodes (ISEs) rely on a thin-film membrane interposed at the

junction between a test sample solution and a reference solution. The membrane is
designed to be selective for an ion of interest through the incorporation of an
ionophore within the membrane phase. The ionophore interacts and selectively
binds with the ion of interest in the sample solution. There are various mechanisms
by which this process occurs depending on the nature of the ionophore including
ion-pairing and neutral or charged ligand binding. Ion-pairing is the result of
electrostatic interaction between the target ion and an oppositely charged ion
localised within the membrane phase. With ligand binding, selectivity is a function
of a host–guest interaction between the chelating ligand and the ion of interest.
Examples of such ligands include macrocyclic molecules such as crown ethers [43,
44] for monovalent ion detection (e.g. sodium and potassium) and more recently
calixarenes [45] for the determination of molecular anions such as perchlorate and
salicylate [46] and calixpyrroles for halides [47]. In all cases, the selective uptake of
ions into the membrane phase results in a transmembrane potential which is pro-
portional to the chemical gradient that exists for the target ion between the solutions
on either side of the membrane. This is an example of where the ion-selective
membrane gives controlled mixing of the two solutions. In the absence of the
membrane, where the sample and reference solutions are allowed to freely mix, a
gradient would essentially not exist and so a measurable potential could not develop
or reach equilibrium. In the presence of the membrane, the transmembrane potential
is measured with an internal reference electrode to give a Nernstian response which
is directly proportional to the activity of the ion in the sample solution. The
advantage of ISEs is they have a dynamic sensor range on the log scale, a classic
example being the pH electrode.
Amperometric enzyme electrodes have been extensively researched for a range
of metabolites. Membranes have been used to separate the biological sample being
studied from the lower layers of the sensor construct, primarily to protect the sensor
components from biofouling or inhibition of the biorecognition component, in this
case the enzyme. This membrane sandwich (with enzyme filling) is illustrated in
Fig. 4.1 [48].
78 S.M. Reddy

Fig. 4.1 A dual-membrane amperometric enzyme electrode

The enzyme is the biological catalyst that catalyses the conversion of the target
molecule into a product that possesses electrochemical activity and so can be
detected at a suitably polarised amperometric electrode. The current output can be
directly related to the target concentration. In addition to protecting the enzyme, the
sample-interfacing membrane serves a second important function that of controlling
the diffusion of the target molecule to the enzyme layer. This is important in
dictating the dynamic linear range of the sensor. The enzyme within the enzyme
layer is there at a finite concentration, and at target concentrations above the Km of
the enzyme, we approach enzyme saturation. The kinetics associated with this
process means that at such high target concentrations, there will be no linear
relationship between target molecule concentration and the amperometric electrode
response. The outer membrane therefore controls the diffusion of substrate to the
enzyme layer, thereby increasing the ‘apparent’ Km of the enzyme. Many enzyme
electrode sensor strategies have interposed a second membrane between the enzyme
and the underlying electrode in order to prevent electrochemically active com-
pounds (native to the test biological sample) from reaching the electrode. These
electro-active compounds (e.g. ascorbic acid, 4-acetaminophen and uric acid) are
known as interferents and if they were allowed to reach the electrode would result
in a false-positive signal, when attempting to measure the target analyte. Typical
membranes used for their anti-interference properties include cellulose acetate [48]
and Nafion® [49, 50]. Whereas cellulose acetate rejects ionic interference through
reverse osmosis, the Nafion® membrane is a perfluorosulphonated hydrocarbon
possessing a repeating sulphonate ion on the side chain of the polymeric hydro-
carbon backbone. These fixed anionic sites enable the polymer to have ion
exchange capacity, attracting cation mobility within the polymer phase but essen-
tially screening out anionic molecules such as ascorbate and urate.
4 Membrane Technologies for Sensing and Biosensing 79

4.2.1 Biocompatible and Permselective Membranes

PVC membranes plasticised with surfactants have been used to improve the bio-
compatibility of the outer membrane when exposed to whole blood. Non-ionic and
anionic surfactants have been shown to radically reduce biofouling during blood
exposure [25, 33]. However, cationic surfactants such as trioctylmethyl ammonium
chloride (Aliquat®; Al) [26] were shown to induce protein adsorption and subse-
quent biofouling of the PVC membrane surface by cells and platelets. It has been
postulated that the cationic nature of Aliquat® within the PVC phase also dictated
the surface properties of the polymer and served as a positively charged layer
capable of attracting negatively charged biologicals including proteins, cells and
platelets.
PVC membranes plasticised with the anionic surfactant bis(2-ethylhexyl)
hydrogen phosphate (BEP) were found to be more biocompatible than commer-
cially available haemodialysis membranes [33]. They exhibited much reduced
surface fouling and smooth homogeneity of any deposited material from whole
blood samples. It is thought that the negative charge nature of BEP at physiological
pH endows negative charge character to the supporting PVC membrane and sur-
face, thereby repelling the adhesion of the aforementioned biologicals which also
possess negative charge character. It is possible therefore that by combining the
charge repulsion properties of cellulose acetate with the biocompatible nature of
BEP as a plasticiser, one can envisage a permselective and biocompatible mem-
brane. Microporous polycarbonate membranes are produced by a nuclear
track-etching process of polycarbonate polymer sheets, followed by an alkaline bath
to wash away the weakened polymer sites to leave pores. These membranes with
fixed pores are said to be heterogeneous, and solute permeability across the
membrane depends on the tortuosity of the pores and the movement of solvent
through those pores. The plasticised PVC membranes are homogeneous in nature in
that they are one phase, comprising a predominantly liquid phase within the bulk
polymer support. Diffusion of a solute through such a membrane separating two
aqueous solutions where the plasticised PVC membrane represents an insoluble
liquid phase in between the two solutions therefore depends on the solubility of the
solute into the PVC membrane phase. This type of diffusion is known as parti-
tioning, and therefore, the solute will have a defined partition coefficient defining
the solubility ratio of the solute between the aqueous and plasticised PVC phases.
Permeability and, more importantly, permselectivity can be controlled by judicious
choice of the plasticiser used. Christie et al. [51, 52] extensively studied iso-
propylmyristate (IPM) as a hydrophobic plasticiser to modify PVC membrane
permeability. They showed that PVC (IPM) allowed preferential partitioning of
aromatics (e.g. paracetamol and dopamine) over hydrophilic (hydrogen peroxide) or
charged (ascorbate) organic molecules. The membrane therefore showed promise
for the direct electrochemical quantification of physiologically relevant hydropho-
bic molecules while screening out electrochemically active interference from
hydrophilic compounds native to the same test biological sample.
80 S.M. Reddy

Plasticised PVC outer membranes have also improved the signal-to-noise ratio
of oxalate biosensors. In the latter case, IPM and Aliquat® were coplasticised within
the PVC [26, 53]. Whereas oxalate was allowed to diffuse from the bulk solution
into the enzyme layer, the enzymatically produced hydrogen peroxide was unable
to escape through the outer membrane into the bulk solution and was therefore
concentrated in the direction of the underlying amperometric electrode where it was
subsequently electrochemically decomposed and detected at higher levels. The
selectivity for oxalate was possible due to the ion exchange of oxalate through the
membrane via ion-pairing with the cationic surfactant. The selectivity against the
in situ produced hydrogen peroxide was primarily due to the hydrophobic barrier
presented by the presence of IPM within the outer membrane.
It should be noted that with increasing complexity of layers to improve the
membrane function, any increase in thickness may inevitably result in a reduced
flux of desired molecules through the composite membrane. The consequences of
this are reduced sensitivity and increased response times. There is therefore a
trade-off between final membrane thickness and the need to handle the material as a
free-standing membrane. In the advancement of biosensor technologies, the move
has been away from having malleable free-standing membranes to ultra-thin-film
membranes which have either been electrochemically grown from a monomer or
spin-coated from a preformed polymer solution.

4.3 Electroconductive Hydrogels

When a polymer surface is brought into contact with a biological sample such as
blood, components therein tend to deposit on the surface. Proteins, cells and pla-
telets will interact with the surface, and depending on the surface chemical prop-
erties (e.g. charge and hydrophilic/lipophilic balance) and physical properties (e.g.
roughness), the degree of bioadhesion can vary. The pH at the sensor sample
interface will affect the surface chemistry depending on whether the surface con-
tains acidic (carboxylate) or basic (amino) functionalisation. The pH will also affect
the net charge of proteins. Attractive and repulsive intermolecular forces subse-
quently come into play in the first layers of bioadhesion. If the surface properties are
such that they encourage the deposition of biomaterial, further protein deposition on
a protein precipitated surface becomes possible, followed by the deposition of other
blood components including cells and platelets in a cascade process. Interfacial
energies, using contact angle analysis with protein solutions as probe liquids, have
been studied to better understand the surface properties that can result bioadhesion
[33]. In the biosensor world, such bioadhesion has been seen as a major bottleneck
in the development of successful functional sensors for long-term implantation in
the body or the environment.
Electroconductive hydrogels (ECHs) are polymer blends or copolymer networks
which combine conductive electro-active polymers with natural or synthetic
hydrogels [54]. This combination uniquely offers materials with the combined
4 Membrane Technologies for Sensing and Biosensing 81

functions of electrical responsivity and biocompatibility [55]. The materials lend

themselves to sensor applications [56]. Also, the materials can be made to swell or
contract under electrical stimulation, thereby controlling movement as in the case of
artificial muscle [57] and nerve systems [58]. Alternatively, the swell change due to
an environmental stimulus (such as selective solute diffusion in or out of the
material) can be interrogated electrochemically to give a qualitative or quantitative
measure of the stimulus being studied [59]. Typical electroconductive polymers
include polyaniline, polypyrrole (PPy) and polythiophene, and typical hydrogel
polymers which have been integrated include polyvinyl pyrrolidone (PVP), poly-
acrylamide (PA) and poly(2-hydroxethylmethacrylate), abbreviated to pHEMA.
Moschou et al. [60] produced an artificial muscle material based on an
acrylamide/acrylic acid copolymer hydrogel doped with a polypyrrole/carbon black
composite. Lira et al. [61] developed polyaniline–polyacrylamide by electropoly-
merising the conducting polymer within the pores of a microporous PA gel. This
doping allowed an otherwise insulating hydrogel material to become conducting.
Electrochemically induced swelling and contraction of such materials potentially
allow them to be used for controlled drug delivery [62]. Redox enzymes such as
glucose oxidase and cholesterol oxidase have been physically entrapped within
electroconductive hydrogels comprising a pHEMA/PPy hydrogel membrane to
produce biocompatible biosensors for glucose and cholesterol, respectively [63].
Whereas the polymerisation of the HEMA (and subsequent entrapment of enzyme)
at a platinum electrode surface was by UV activation of the monomer solution in
the presence of a cross-linker (tetra-ethylene glycol-diacrylate, TEGDA), the PPy
was incorporated by electropolymerisation of pyrrole monomer present in the gel.
Polymerisation occurred under galvanostatic (fixed current) conditions at 1 mA
over 50 s. The platinum electrode with integrated electroconductive gel and
immobilised enzyme was then used in the amperometric mode to interrogate
solutions containing either cholesterol or glucose. The electroconductive nature of
the conducting polymer allows the amperometric electrode to be extended into the
hydrogel phase. The biosensor minimised physiological interference, such as from
acetaminophen, uric acid and ascorbic acid [64].
The major disadvantage with chemically combining these two materials is to
control the integration of the two components. Polymer blends can be better con-
trolled than the covalent copolymerisation of the two monomer components. The
latter can lead to copolymer systems with unpredictable final material properties
leading to a lack of confidence in reproducibly synthesising the desired electro-
conductive hydrogel material. This has largely been addressed by either chemical
oxidative polymerisation or electrochemical polymerisation of electroconductive
component into a preformed hydrogel [55] in order to form a hydrogel-conducting
polymer co-network (or polymer blend) (Fig. 4.2).
Whether a biosensor is used in vitro or in vivo, it is desirable that the biosensor–
sample interface is biocompatible. Biocompatibility is the study of how materials
behave in the presence of a biological sample. The biological sample, for example,
can be blood, urine or tissue fluid. Depending on a host of properties of the material
such as surface tension, hydrophobicity, hydrophilicity, surface roughness and
82 S.M. Reddy

Fig. 4.2 Schematic illustration of the generalised synthetic routes to electroconductive hydrogels
[55]

elastic modulus (material stiffness) to name a few, the spectrum of physical

chemical reactions that can be induced within a biological sample or host body at
the material interface will vary. The body responds to any foreign object by
launching a series of reactions triggered by the denaturation of absorbed proteins,
receptor-mediated attachment of cells, the production of a range of cytokines, and
then the eventual fibrous biofouling of the material [65]. Hydrogels offer bio-
compatible hydrated surfaces that potentially reduce surface tension and offer a
hydrophilic/hydrophobic balance at the interface conducive to low shearing stress,
thereby inhibiting protein adsorption and circumventing the host immune response
that often leads to rapid device failure. Changing the cross-linker concentration
within the hydrogel can determine the water-carrying capacity and mechanical
strength of the gel. Hydrogels can therefore be designed to mimic biological
materials such as tissue and the inner linings of veins and arteries. Hydrogels
incorporating phosphorylcholine moieties have been extensively exploited in this
regard [66–69]. Such artificial biocompatible surfaces which are resistant to bio-
fouling are crucial to the development of successful implantable biosensors as well
as drug delivery systems. The biological host has to recognise the implant material
as non-hostile. The current understanding is that if protein adsorption and subse-
quent denaturation are avoided, then it is likely that the aforementioned inflam-
matory cascade leading to fibrous encapsulation will be inhibited. If such fibrous
deposits were allowed to occur at the sensor sample interface, then it would present
itself as an additional diffusion barrier to the target analyte to be measured, resulting
in a significant reduction in the sensor response.
4 Membrane Technologies for Sensing and Biosensing 83

For example, polyethylene glycol (PEG) has been shown to be a non-cytotoxic

material attributed to its renowned ability to inhibit protein adsorption. Implantable
biosensors offer the potential for long-term and real-time monitoring of the meta-
bolic and physiological status of the patient in an intensive care setting. For
example, theophylline is a highly effective bronchodilator which is often used in
critically ill patients. It has a narrow therapeutic range, below which it is ineffective
and above which is detrimental to health. Therefore, an implantable sensor that can
continuously measure the concentration of such drugs in the body will ensure that
the drug remained in the body in the desired therapeutic range. A hydrogel polymer
based on HEMA, PEG and MPC [70, 71] was shown to be resistant to protein
adsorption, and sensor lifetime was increased from 24 h to 5 days. The fluidity of
the PEG and PC arms in the hydrated polymer was shown to be instrumental in
extending the useful lifetime of the sensor. By incorporating an electroconductive
polymer, an otherwise high impedance insulating hydrogel is converted to a low
impedance conductive gel material, while still retaining the desirable biocompatible
properties. Electrochemical biosensors to date have relied on the redox-active
properties of enzymes to catalyse the chemical conversion of a target molecule to an
electrochemically active compound which can be measured with amperometry.
Oxidase and dehydrogenase enzymes have been used extensively for this purpose
to measure metabolites such as lactate, cholesterol, glucose and galactose. The
oxidase enzymes typically rely on the presence of molecular oxygen to oxidise the
metabolite of interest, thus producing hydrogen peroxide as a by-product. Hydrogen
peroxide is electrochemically oxidised at around 650 mV versus Ag/AgCl. The
measurement of hydrogen peroxide electrochemical decomposition can be directly
related to the presence of the target metabolite. The need for oxygen as a
co-substrate in this enzymatic process can be mitigated by using either an artificial
electron mediator such as ferrocene or by incorporating an electroconductive
polymer such as polypyrrole. Whereas ferrocene can shuttle electrons from the
metabolite to the electrode via the enzyme, the electroconductive polymer can
theoretically be hard-wired into the enzyme redox site obviating the need for a
mediator and the electrons move directly from the enzyme to the electrode.
Therefore, the preparation of an electroconductive hydrogel offers the advantages of
integrating a biocompatible material that can immobilise an enzyme [72] of interest
with a conductive polymer material that can efficiently reconcile the transfer of
electrons from a target analyte to the electrode [73].

4.4 Molecularly Imprinted Polymers

Molecular imprinting is an established molecular recognition technique. It relies on

the polymerisation of functional monomers around a target template molecule, in
the presence of a cross-linker. When the template molecule is subsequently
removed from the polymer, cavities or holes are left in the polymer, which contain
both a chemical and physical architecture to recognise and rebind the original
84 S.M. Reddy

template molecule. The molecularly imprinted polymer (MIP) is essentially

behaving as a plastic antibody. MIPs have been produced for small molecule
recognition using acrylate-based monomers and are being used in solid-phase
extraction cartridges and are commercially available. However, recent excitement
has been in the imprinting of biomolecules. Various strategies have been studied for
the preparation of MIPs, namely 3D bulk polymerisation resulting in MIP particles
and 2D thin-film polymerisation resulting in MIP membranes. Whereas the
particle-based MIPs possess multiple cavities around each particle surface, the
cavities within membrane-based MIPs are essentially confined to the membrane
surface layer. Therefore, given the surface area, the MIP particles contain more
cavities than the MIP membrane. Whereas MIP particles lend themselves to bulk
solution-type immuno-assay formats, MIP membranes can be integrated with var-
ious electrode surfaces to develop sensors and biosensors.

4.4.1 3D MIPs

Whereas 3D MIP particles do not readily lend themselves to forming controlled,

homogeneous thin films, it has been possible to confine the particles between
microporous membranes. Reddy et al. [74] demonstrated that polyacrylamide-based
MIPs for haemoglobin could be immobilised between a polycarbonate membrane
and a glassy carbon electrode. The group demonstrated that haemoglobin was able
to diffuse from bulk solution through the outer polycarbonate membrane and
selectively bind to the haemoglobin at the electrode. They were able to interrogate
the binding event using cyclic voltammetry and to probe the change in current due
to electro catalytic reduction in oxygen at −0.4 V versus Ag/AgCl. The electro-
catalysis was mediated by the iron centres within the protoporphyrin of the selec-
tively bound haemoglobin. By contrast, when a control polymer (NIP) was
interfaced with the electrode, the electrocatalytic response was much reduced due to
the lack of selective protein binding on the control polymer. This strategy was
extended to multiprotein determinations using MIPs interfaced with glassy carbon
electrodes [75].
The selectivity of MIP binding has been improved by tailoring the rebinding
conditions. Parameters such as buffer composition and pH have been found to be
important in radically increasing the MIP:NIP selective binding ratio. Kd values
have also been determined [76] for a range of MIPs and their target protein with
values in the range 10−6–10−8 M. The Kd value for antibody–antigen interactions is
typically 10−9 M and that for biotin–avidin, the strongest biological interaction
known to man, is typically 10−15 M. There is therefore scope for more research to
improve 3D MIP binding.
Despite these notable successes, the big challenge has been to control the con-
centration of MIP particles at the electrode surface. This is possible by shifting to
thin-film (2D) MIPs formation.
4 Membrane Technologies for Sensing and Biosensing 85

4.4.2 2D MIPs

MIP membranes have been formed either by chemical polymerisation under con-
trolled pressure to form thin films on sensor substrates (such as QCM [77, 78]) or
grown on electrode sensor surfaces using electrochemically induced polymerisation
(ECIP) [79].
With the controlled pressure method, a polymerising solution of an acrylamide
monomer/bis-acrylamide cross-linker (containing the target protein for imprinting)
was first introduced to the sensor surface in the form of a droplet, followed by
overlaying a glass slide and applying small (e.g. 20 g) weights on top of the glass
side in order to produce a thin liquid film off polymerising solution. After poly-
merisation, the glass slide was prised away leaving a thin polymer hydrogel film
(typically 40 nm thick) with MIP characteristics on the sensor surface. The resulting
surface was then washed with a 10 % solution of acetic acid and sodium dodecyl-
sulphate (SDS) in order to denature and strip away protein imprinted at the surface.
This resulted in selective cavities being exposed just at the surface. It was shown that
when such MIPs in films were coupled with a QCM sensor device, the sensor
responded selectively for the rebinding of the target protein, and depending on the
nature of chemical functionalisation of the acrylamide, the MIPs thin film was also
non-responsive to interfering proteins. For example, N-hydroxyacrylamide-based
MIPs was able to distinguish between selective haemoglobin binding and the
non-target protein such as bovine serum albumin (BSA). Acrylamide-only-based
MIPs were unable to discriminate between the two proteins and allowed binding of
both. N-isopropylacrylamide (NiPAM) where did not allowed binding of either
protein. Reddy et al. [78] demonstrated that such films could also be prepared on
silicon oxynitride surfaces. They used dual-polarisation interferometry (DPI) to
interrogate MIP-coupled silicon oxynitride chips and were able to show selective
binding as a function of changing thickness of the film. Interestingly, the micro-
fluidics of the DPI system allowed for the study and characterisation of 3D MIP
particles also. In the latter case, the target protein was allowed to pre-adsorb to the
silicon oxynitride bare surface, and then, the MIP or NIP was allowed to flow over
the adsorbed protein layer. Whereas the MIP stripped away the protein from the chip
surface, the NIP had no net effect on the pre-adsorbed protein layer.
Electrochemically induced polymerisation (ECIP) obviates the need to apply
pressure during polymerisation to form thin films. Cycling the voltage under
negative potentials, in the presence of an acrylamide monomer solution and
potassium persulphate, results in controlled layering of an electropolymerised thin
film. The thickness of the film becomes self-limiting due to the growth of the
insulating polymer film. Typical thicknesses range between 10 and 50 nm. Once
the film is grown, the surface is again treated with acid and surfactant in order to
remove the surface protein and to leave the protein-selective cavities at the surface.
This technique has been used for both small molecule imprinting as well as large
biomolecule imprinting. The mode of interrogation of selective binding is to
measure the signal to a chosen electrochemical marker before and after target
86 S.M. Reddy

molecule binding. The ferricyanide/ferrocyanide redox couple has been used to

good effect as an electrochemical marker. This strategy relies on the permeability of
the MIP thin film to the redox marker being attenuated when the target molecule is
selectively bound within the cavities. Therefore, an apparent reduction in the
electrochemical signal is indicative of selective binding, because the redox marker
is unable to reach the electrode [80].
Xu et al. [81] have reported on stimuli-responsive MIPs. Stimuli-responsive
polymers [32] have been prepared to respond to physical parameters such as
temperature, electric field, magnetic field, ultrasound and light. The preparation of a
photo-responsive MIP requires the presence of a photosensitive group which is
chemically bound to the functional monomer. Azobenzene derivatives have been
typically used for this purpose [82]. The trans to cis photo-isomerisation of the
azobenzene-based functional monomer brings about significant changes in the
structure of the chromophore, resulting in a change in receptor binding geometry.
Therefore, photo-modulation can result in binding or release of the template
molecule [82].
Magnetically responsive MIPs have been prepared using a core shell technique
where the MIP is grown on Fe3O4 nanoparticles. Such magnetic tagging of MIPs
offers the ability to separate MIP particles from a sample solution under the
influence of a magnetic field [81].

4.5 Thermoresponsive Membranes

Thermosensitive polymers can change their volume depending on the temperature.

The polymers are typically water-based and comprise hydrophilic and hydrophobic
moieties. Below a critical temperature, extensive hydrogen-bonding interactions
occur between the hydrophilic groups in the polymer and water, resulting in
increased water solubility and swelling of the polymer. The temperature at which this
occurs is known as the lower critical solution temperature (LCST). Above the LCST,
the hydrogen-bonding network is broken and the hydrophobic groups of the polymer
dominate, resulting in expulsion of water and polymer gel contraction. When such a
polymer is developed into a MIP, the selective binding of the template molecule can
be modulated by altering the temperature of the gel. PolyNiPAM is the most widely
studied thermosensitive hydrogel polymer possessing a LCST of around 32 °C.
Thermoresponsive PolyNiPAM MIPs (T-MIPs) have been prepared for proteins. Qin
et al. [83] were able to show that the latter gel as a copolymer with methacrylate could
be used to form a MIP for lysozyme. It was thought the presence of a
copper-chelating agent improved the binding affinity of protein to the MIP through
dative bond formation [84]. The optimum temperature for maximum selective
binding of lysozyme in the presence of other proteins was found to be 28 °C.
Whereas the PNiPAM MIP was relied upon for its thermal responsiveness, the
copolymer provided an essential support matrix for the MIP.
4 Membrane Technologies for Sensing and Biosensing 87

T-MIPs have also been used in conjunction with quantum dots (QDs). QDs are
nanosized semiconductor particles (ranging 5–50 nm in diameter) with fluorescence
capability. The HOMO-LUMO bandgap is inversely proportional to the particle
diameter. PNiPAM imprinted for bovine haemoglobin was coated around CdSe
QDs. The QD fluorescence intensity was modulated depending on the temperature
and the selective binding of the template molecule over other interfering proteins
including lysozyme, ovalbumin and BSA. The change in fluorescence was directly
related to the volume transition of the outer T-MIP during selective binding only
[85]. It could be envisaged that an array of quantum dots of various particle size and
MIP configurations could be used for screening of a range of protein-based cancer
and cardiac disease markers, offering opportunities for personalised blood screening
and rapid diagnosis in the convenience of a general practitioner’s surgery or a
routine biochemistry laboratory setting. An extra dimension could be added for
protein discrimination by researching novel thermal-responsive polymers. Recently,
chemically cross-linked chitosan hydrogels have demonstrated thermal-responsive
behaviour at temperatures not exceeding 50 °C [86]. Such T-MIPs with higher
LCSTs would be more suitable for imprinting of small molecules (e.g. for dopa-
mine [87]) since they are unlikely to be adversely affected at such temperatures
compared with metastable biologicals such as proteins and DNA which could
change their conformation based on the increased temperatures. Additionally, the
use of organic solvents in the preparation of T-MIPs makes them more suitable
again for small organic molecules. For example, Suedee et al. [87] developed a
T-MIP using methacrylic acid and acrylamide as functional monomer is with
bisacrylamide as a cross-linker. Synthesis was carried out in 80 % aqueous
methanol solution. The resulting T-MIP exhibited the best rebinding of dopamine
under the same solution conditions at an optimum operating temperature of 35 °C.
The latter temperature corresponded to the optimum swelling required for the
hydrogel to ensure the highest selective binding. Thermal-responsive polymers
have also been used as a gating membrane to coat and an underlying MIP particle.
Initially, MIP particles are produced by living polymerisation, and then following
template removal, PNiPAm bristles are grafted onto the outer surface of the par-
ticles. At temperatures above the LCST, the PNiPAm bristles exude water and
collapse to shut down permeability of solids to and from the MIP particles. By
decreasing the temperature, the PNiPAm absorbs water and forms an open structure
to allow the facile movement of solutes across the PNiPAm semi-permeable barrier.

4.6 pH-Responsive Membranes

pH-sensitive functional groups such as carboxylic, sulphonic and amino groups are
ionisable depending on pH and ionic strength. Their ability to donate or accept
protons can have significant impact on the formation or disruption of
hydrogen-bonding interactions. Integration of such functional groups into MIPs
offers the opportunity for target molecule binding to be a function of environmental
88 S.M. Reddy

pH. The polymers can undergo large volume transitions. Polymer swelling can
occur due to charge repulsive forces following either mass deprotonation of acidic
functional groups (e.g. when using a methacrylic acid functional monomer) or mass
protonation of basic functional groups (e.g. with acrylamide or 4-vinylpyridine). Of
course, if both types of functional group exist within the same polymer system (i.e.
if both acidic and basic monomers are copolymerised), then the net polymer
swelling (and conformation) will be a function of the ratio of acidic and basic
groups within the polymer, just like for a protein. However, for maximum
swelling/deswelling effects, the two types are generally not introduced within the
same MIP formulation. Also, in the interest of observing maximum pH respon-
siveness, the concentration of cross-linking agent used during MIPs formation is
also critical. At high cross-linking densities, the rigidity within the MIP required to
confirm cavity formation makes it less conducive to respond through volume
transitions upon pH modulation. The latter has been a major drawback in the
development of pH-responsive MIPs [81]. In an alternative approach,
pH-responsive polymers have been used to coat MIP particles. The outer layer will
open and close depending on the environmental pH. Such systems offer interesting
opportunities for drug delivery and drug therapy. For example, a MIP with a drug
payload may be introduced into the body and the drug only released under a
predetermined pH condition.

4.7 Electrospinning of Membranes

Electrospinning is an electrically driven process for the production and deposition

of polymer fibres from charged threads of polymer solutions onto a solid substrate.
Such fibres can also be produced from polymer melts, obviating the need for an
organic solvent. The spinning process is much like rayon spinning in the textile
industry, but electrospinning happens on the nanoscale and the fibres can be
deposited on the surface to produce matted structures resembling membranes. In
order for a continuous thread to be formed, the molecular cohesive forces within the
polymer solution needs to be sufficiently high. Otherwise, the liquid stream is
disrupted and an electrogenerated spray is produced instead [88]. The basic con-
stituents of the laboratory-based set-up comprises a spinneret (such as a stainless
steel hypodermic needle) connected to high-voltage DC power source (5–50 kV), a
syringe pump and a grounded surface for fibre collection [88] (Fig. 4.3).
With this level of simplicity in the process, it is possible to prepare electrospun
fibre membranes for a vast range of polymers with submicron to nanometre
diameters. In comparison to their microscale counterparts, the nanoscale fibre
materials possess a very large surface area-to-volume ratio (by a factor of 103) as
well as superior stiffness and tensile strength and enhanced chemical functionality.
Applications have included the preparation of surfaces and structures for
re-enforced nanocomposites [88] and biomedical applications such as enzyme
4 Membrane Technologies for Sensing and Biosensing 89

Fig. 4.3 Schematic diagram to show polymer nanofibres by electrospinning [88]

immobilisation [89] for biosensor applications, tissue scaffolds in tissue engineering

[90], and drug delivery modalities [91, 92].
Li et al. [93] prepared electrospun polylactic acid (PLA) nanofibre membranes
for the immobilisation of biotin to develop biosensors based on biotin–streptavidin
biospecific DNA probe binding. Although capture of DNA was demonstrated, the
workers highlighted issues regarding non-specific binding between streptavidin and
the PLA as well as the uneven distribution (aggregation) of immobilised biotin
within the PLA.
Mahmoudifard et al. [94] have demonstrated antibody immobilisation via
covalent coupling within polyethersulphone (PES) electrospun fibres using ethyl-3-
(3-dimethylaminopropyl)-carbodiimide/N-hydroxysuccinimide (EDC/NHS) cou-
pling chemistry. The PES surface was activated with O2 plasma to produce carboxyl
functional groups for subsequent antibody coupling. The antibody-functionalised
material was specific for staphylococcus enterotoxin B (SEB) and demonstrated
improved sensitivity as a biosensor when compared to conventional ELISA-
based immunoassays. Glucose oxidase has also been studied extensively for
immobilisation into electrospun biocomposite membranes. Wu et al. [95] prepared a
PVA/GOx/graphene in the development of glucose nanobiosensors. Whereas gra-
phene has been widely studied and integrated with redox enzymes to improve the
electrochemical response characteristics (through facilitated electron transfer
mechanisms), this research group also demonstrated that graphene aided in stabil-
ising the enzyme’s conformation as well as its lifetime. Su et al. [96] encapsulated
GOx in PVA/chitosan electrospun nanofibres onto a platinum electrode surface
through a chemical cross-linking process in the presence of glutaraldehyde vapour.
The composite was completed with a thin Nafion film. The sensor was found to
exhibit a wide calibration range with a low apparent Km, which is surprising in that
an extended linear range normally gives an increase in apparent Km. The workers
90 S.M. Reddy

also demonstrated biosensor application within human serum samples, the chitosan
component offering not only an apparent improvement in biocompatibility but also
anti-interference capability. Wang et al. [97] developed fibro-porous polyurethane
(PU) coatings. Membrane thickness was controlled by varying the PU solution
concentration (8–12 % w/v) as well as the electrospinning times (2.5–10 min).
Electrospinning was allowed to occur on at platinum–iridium electrode with
preadsorbed GOx. A direct comparison was made between electrospun
membrane-based biosensors and solvent-cast membrane-based biosensors. The
electrospun system offered some key advantages such as robustness and speed of
depositing and a better controlled interconnecting porous structure. Zhou et al. [98]
demonstrated that a polystyrene-based and indium complex-doped fibrous optical
membrane could be used as a glucose biosensor, in the presence of the enzyme
glucose oxidase. With a wide dynamic and low concentration range of glucose
(10−10–10−4 M), the luminescence response at 562 nm was reported to be less than
1 s. Ji et al. [99] produced polyurethane nanofibre-based membrane substrates for
dual-enzyme immobilisation through coaxial electrospinning for the development of
glucose testing strips.
The enzymes such as GOx and horseradish peroxidase were co-immobilised in
the presence of a chromogenic agent such as o-dianisidine. Hollow nanofibres were
produced due to the coaxial electrospinning method used. The latter required two
solutions to be electrospun together resulting in a shell layer (comprising poly-
urethane) and a core layer comprising a mixture of the enzymes and the chro-
mogenic agent (see Fig. 4.4).
The resulting nanofibre membrane, when dipped in a solution of glucose,
resulted in a glucose-specific colour change within the solution which could be

Fig. 4.4 Schematic illustrations of the bienzyme reaction for glucose measurement and the set-up
for coaxial electrospinning to prepare hollow nanofibre membrane-based glucose testing strips.
During coaxial electrospinning, GOD, HRP and chromogenic agent (ABTS or odianisidine) were
simultaneously immobilised in situ in the hollow nanofibre membrane [99]
4 Membrane Technologies for Sensing and Biosensing 91

measured spectrophotometrically at 440 nm. The membranes allowed for reliable

glucose measurement in the range 0–20 mM. The colour change could also be
quantified directly on the membrane patch with diffusive reflectance measurement
at 440 nm using a spectrodensitometer. In order to achieve the in situ colour
change, a 10 µL droplet of glucose solution was place on the patch and measure-
ments taken at 30 s after addition, when there is no further colour development.
This method allowed for an extension in linear range of glucose measurement to
50 mM. When testing real biological samples such as human serum, the workers
demonstrated excellent correlation (within 1 %) with a commercially available
glucose test kit. When the unused strips were stored at 4 °C, they retained near
100 % activity for up to 140 days, whereas at 25 °C storage, the strips only retained
100 % activity for approximately 40 days and activity rapidly declined thereafter.

4.8 Layer-by-Layer Membrane Assemblies

Layer-by-layer assembly refers to a surface modification technique whereby

sequential spontaneous adsorption of at least two distinct materials is allowed to
occur on a planar substrate. Layer-by-Layer (LbL) deposition is a facile and low-cost
technique, which involves construction of polyelectrolyte multilayer (PEM) films by
simple alternate dip-coating of a substrate into polyelectrolyte solutions of alter-
nating charge [100]. The film thickness can be easily controlled at the nanometre
scale. The resulting layers can be further functionalised with biomolecules or
nanoparticles. The common polyelectrolytes used for LbL deposition are the anionic
polymers such as polyacrylic acid (PAA) and polystyrene sulphonate (PSS) and the
cationic polymers such as polyallylamine hydrochloride (PAH) and polydial-
lyldimethylammonium chloride (PDADMAC). Proteins can also be layered in the
same fashion due to their polyelectrolyte nature, although this does depend on the
protein isoelectric point (pI) and the solution pH. LbL membranes can be robust and
multifunctional and can be easily prepared, without the need to use costly instru-
mentation. In order to improve the adhesive nature during LbL film adhesion,
polyelectrolytes functionalised with either 3,4-dihydroxyphenylalanine (DOPA) or
catechol, or combinations of the two have been employed. The latter aromatic
hydroxyl compounds, when cross-linked to produce poly(phenol) sections to the
bulk polymer [such as cellulose, chitosan or poly(ethyleneimine)], can promote
adhesion to a range of even non-adhesive and inert surfaces including polyte-
trafluoroethylene (PTFE) and gold [101]. In addition to adhesive properties, the
latter method has offered a means to fine-tune the mechanical strength of the
deposited polyelectrolyte layers. Cell adhesion has also been improved by using
such hydroxyl function-rich surfaces [102].
One disadvantage of LbL systems is the time taken to form each layer (typically
several minutes). This can be improved by using automated systems working in
parallel to simultaneously produce multiple surfaces.
92 S.M. Reddy

Stimuli-responsive LbL membranes have been produced by integrating LbL

technologies with pH-sensitive or thermo-responsive polymers. These more intel-
ligent materials have seen application in selective cell seeding and harvesting. For
example, pNIPAM-modified alginate/chitosan multilayers were used to selectively
produce sheets of cells which could be lifted off by dropping the temperature from
37 to 4 °C; the phase transition experienced by the pNIPAM component of LbL
system induced the lifting-off of the cell sheet which could be used for follow-on
seeding in a different culture support [103]. Due to their versatility, LbL systems
have been developed for as coatings for implantable devices [104], intracellular
drug delivery [105, 106] and in vivo tissue healing [107] and with anti-bacterial
properties to promote wound healing. Applications in tissue engineering have also
been demonstrated, where a three-dimensional porous substrate approach is
required to produce the LbL rather than on a flat substrate. The latter is important
for cell seeding and tissue growth in an environment closely mimicking tissues and
organs [108, 109].
LbLs have been integrated with pH sensitive, magnetic nanoparticles, fluor-
ophores and enzymes leading to the development of optical biosensors. For
example, PSS-PAH multilayers in encapsulated semi-naphtho-rhodafluor-1-dye
(SNARF-1) with polysaccharide conjugate [110]. The LbL microcapsules changed
colour due to the change in pH of the microenvironment during cellular uptake from
alkaline medium to the acidic intracellular environment, resulting in a shift from red
to green. Breast cancer cells were studied in this way, and it was found that
microcapsule incorporation did not affect the cells’ ability to divide and proliferate,
thereby demonstrating their ability to be biocompatible.
LbL microcapsules have been used to encapsulate enzymes such as glucose
oxidase and peroxidase and, when coupled with a redox-active dye such as dihy-
drorhodamine, resulted in a sensor that signals the presence of glucose. In this
well-studied enzyme reaction, glucose present is catalytically oxidised to gluconic
acid and hydrogen peroxide. The hydrogen peroxide produced is in turn reduced in
the presence of peroxidase and the dye becomes oxidised to emit green fluorescence
[111]. The key advantage here of using LbL methodologies has been the less harsh
form of enzyme immobilisation compared with either chemical cross-linking of
enzyme or enzyme physical entrapment during in situ polymerisation.

4.9 Anodised Aluminium Oxide Membranes

Anodised aluminium oxide (AAO) [112] has attracted much interest due to its
unique nanoporous morphology. AAO films can be produced via room temperature
electrochemical anodisation of prevapour deposited aluminium films on silicon
substrate (anodisation typically occurring between 10 and 200 V) in acidic media
such as sulphuric, phosphoric or oxalic acid, to give hexagonal arrays of regular
pores. The anodising voltage and subsequent pore size depends on the acid used,
4 Membrane Technologies for Sensing and Biosensing 93

Fig. 4.5 a Schematic drawing of AAO structure prepared by electrochemical anodisation of Al.
b Summary of self-ordering voltage and corresponding interpore distance of AAO produced
within three well-known regimes of electrolytes (suphuric, oxalic and phosphoric). c Top SEM
cross-sectional view of AAO membrane formed by MA (0.3 M H2C2O4, 1 °C, 40 V) and bottom
by HA (at 140 V) for 2 h (insets SEM top view of pore structures) [121]

and the pore size increases from sulphuric acid to oxalic acid and to phosphoric acid
as shown in Fig. 4.5.
These nano-ordered porous membrane structures are finding applications in
molecular separation [113, 114], drug delivery [115, 116], opto-electronics [117],
energy storage [113], catalysis [118] and biosensors [119, 120], and these wider
field of applications has been extensively reviewed by Mutalib Md Jani et al. [121].
AAOs readily lend themselves to optical sensing applications due to their moderate
transparency and transmission mode measurements in both the UV and IR spectra
during biointeraction analysis of proteins, enzymes, antibodies and even DNA
[122].
94 S.M. Reddy

4.9.1 Surface Treatment

The AAO structures are typically insulating but more importantly, acid labile which
presents limitations to their utilisation in chemically changing environments (pH
and ionic strength). However, they do readily lend themselves to chemical func-
tionalisation, which has significantly expanded the potential applications of
AAO-based materials. The AAO surfaces, being rich in hydroxyl functional [123]
groups allows for facile functionalisation using wet chemical approaches. Using
this method, AAOs have been modified via silanisation of the AAO hydroxyl
groups using organosilanes such as (3-aminopropyl)triethoxysilane (or APTES) as
shown in Fig. 4.6. APTES attachment is the precursor to grafting polymer brushes
using surface-initiated polymerisation on the AAO membrane surfaces. Polymers
grafted include PNIPAM [124] and PHEMA [125]. Following silane coating with
isocyanatopropyl triethoxysilane, it is possible to graft N-hydroxysuccinimidyl
carbonate polyethylene glycol (NHS-PEG) [126] (Fig. 4.7). This pendant iso-
cyanate group can then react readily with amino groups as found in enzymes and
amino-terminated DNA in order to chemically link biorecognition molecules within
the cavities of the AAO membranes.
Using an LbL approach, polyelectrolyte multilayers of PAA and PAH have been
deposited within AAOs. Subsequent carbodiimide coupling (using NHS/EDC) has
allowed the chemical attachment of antibodies with the carboxylate groups of PAA
within the nanopores [127].
Proteins have also been immobilised using conducting polymers (CP) as sup-
ports for attachment within AAO [128]. Glutaraldehyde was used to chemically
cross-link enzymes onto an AAO-polyaniline composite.

Fig. 4.6 Common schematic route of silanisation used for surface modification on AAO
membranes [5]
4 Membrane Technologies for Sensing and Biosensing 95

Fig. 4.7 Silanisation of hydroxylated AAO surface with isocyanatopropyl triethoxysilane and
subsequent immobilisation of amino-terminated DNA [5]

4.9.2 Biosensors Applications

In the chemical immobilisation of biorecognition molecules (such as enzymes,

DNA or antibodies) within the nanopores of AAO, enhanced bioactivity was
reported when compared with simple chemisorption or physisorption onto, for
example, flat membrane surfaces. This is a direct result of the increased surface area
available within the pores compared with a non-porous surface. In one study [129],
biotin chemically immobilised within AAO pores showed a higher density of
biomolecule (up to seven times) per unit area than a flat glass surface when
interrogated fluorescently following selective binding by fluorescently labelled
streptavidin. The coverage was also shown to be stable for extended periods with
minimal leaching of biotin. Similarly, Oliveira [128] demonstrated that AAO-CP
composites with subsequent enzyme immobilisation demonstrated enhanced
activities compared with when flat membranes of CP were used to retain the
enzyme. This 3D versus 2D approach to enzyme retention also imparts superior
reusability of the immobilised enzyme with much reduced loss of activity.
Due to their inherent transparency, AAO substrates lend themselves to the
development of optical sensors and biosensors. In one example [130], AAO sur-
faces were functionalised with cationic APTES derivative (namely 3-aminopropyl
dimethylethoxysilane). The latter layer was then modified in an LbL fashion to
deposit alternating layers of anionic-charged quantum dots and cationic dendrimers.
The surface was then primed for probe DNA attachment using carbodiimide
chemistry. Photoluminescence (PL) depends on Forster Resonance Energy Transfer
(FRET) effect between the QD and a (Cy5) fluorescently labelled target comple-
mentary 15-mer DNA. In the presence of target DNA in a sample, the Cy5-labelled
96 S.M. Reddy

DNA strand is preferentially displaced, resulting in measurable changes in PL

spectra.
Due to the regular spacing of the nanoporous structure within AAO substrates
and with the ability to control pore size and depth [131], the surfaces can be made
to produce interference patterns in the presence of white light which can result in
red shifts in spectroreflectance wavelength with biomolecule interaction on the
surface. Such interferometric measurements have allowed the monitoring of avidin
and biotin interaction as well as the detection of circulating tumour cells [132].
AAOs can be metallised with gold and platinum using thermal and chemical
vapour deposition techniques to produce electrode surfaces for the development of
electrochemical redox [133] and impedance-based [134] biosensors. Redox
enzymes such as glucose oxidase and choline oxidase as well as hydrolytic
enzymes such as urease have been physically or chemically immobilised onto the
metal-coated AAO surfaces in the development of amperometric biosensors [135].
Non-electrochemically active biologicals such as antibodies and DNA have also
been immobilised, and by using redox enzyme labels or redox-active intercalating
agents, respectively, AAO-based electrochemical immunosensors [136] and
genosensors [137] have been developed. With physical immobilisation of enzyme,
the issue of enzyme leaching remains (resulting in loss of activity). Polymeric
membranes have been either dip-coated or spin-coated from polymer solution in
order to physically retain the enzyme, with the added benefit of extended sensor
linear range. As with free-standing diffusion-limiting polymeric membranes dis-
cussed earlier in this chapter, the response time of membrane-coated AAO-based
electrochemical biosensors can be dependent on pore size as well as membrane
thickness. Reduced thickness and increased pore sizes can improve the response
time.
Nanopore analysis is an emerging technique that involves using a voltage to
drive molecules through a nanoscale pore in a membrane between two electrolytes,
and monitoring how the ionic current/electrical impedance through the nanopore
changes as molecules pass through it. This approach has allowed charged polymers
(including single-stranded DNA, double-stranded DNA and RNA) to be analysed
with sub-nanometre resolution and without the need for labels or amplification
[138]. Electrochemical impedance spectroscopy has been used to detect ion chan-
nelling through the AAO pores as a function of AAO top surface being modified
with lipid bilayers. Membrane protein incorporation allowed for highly sensitive
and selective measurement of ion diffusion [134] due to changes in membrane
resistance. In a similar fashion, DNA entrapped inside AAO pores could be tracked
using EIS, leading to a significant decrease in resistance due to the current-carrying
capacity of charged DNA. Hybridisation with complimentary strand led to an
increase in resistance [139] with a limit of detection of 0.5 nM for the compli-
mentary target. EIS has also been used for antibody-based microbial detection of,
for example, E. coli. Anti-E. coli antibody was chemically immobilised within
AAO nanopores within a microwell array format. A decrease in resistance was
observed when the target organism was selectively captured by the antibody, with a
4 Membrane Technologies for Sensing and Biosensing 97

reported sensitivity of 100 CFU/mL. This was a significant improvement on

conventional microelectrode arrays by 2–5 orders of magnitude.

4.10 Future Direction

Although many new technologies are coming online for sensing of molecular and
biomolecular interactions, there is still a need to interface these technologies with an
appropriate physical and chemical support. In this regard, this chapter has served to
emphasise the continued need for synthetic polymeric and some inorganic mem-
brane development to not only confine the biorecognition machinery but also to
stabilise the chemical environment and enhance the activity of the biorecognition
component. There are also opportunities to extend the AAO technology to mass
sensing platforms including microcantilevers [140, 141], the quartz crystal
microbalance (QCM) [142] and optical sensors such as dual-polarisation interfer-
ometry (DPI). Vapour deposition of aluminium would be feasible on an underlying
gold electrode on a QCM surface. Anodisation would leave a rigidly coupled
nanoporous inorganic membrane layer of AAO on the QCM. The QCM normally
responds to mass and viscoelastic changes due to a viscously entrained liquid. The
advantage of using AAO in combination with QCM would mean any liquid would
be confined within the nanoporous region resulting in liquid being treated as a rigid
mass and not a Newtonian liquid. The advantage of this would be a reduced
dampening of the QCM resonance during sensing operation, leading to a more
sensitive sensor system that could operate in a liquid environment.

References

1. Akther N et al (2015) Recent advancements in forward osmosis desalination: a review.

Chem Eng J 281:502–522
2. Xu G-R et al (2015) Layer-by-layer (LBL) assembly technology as promising strategy for
tailoring pressure-driven desalination membranes. J Membr Sci 493:428–443
3. Zhou D et al (2015) Development of lower cost seawater desalination processes using
nanofiltration technologies—a review. Desalination 376:109–116
4. van Rijn P et al (2013) Challenges and advances in the field of self-assembled membranes.
Chem Soc Rev 42(16):6578–6592
5. Wang Y et al (2013) When worlds collide: interactions at the interface between biological
systems and synthetic cationic conjugated polyelectrolytes and oligomers. Langmuir 29
(34):10635–10647
6. Weingart J, Vabbilisetty P, Sun X-L (2013) Membrane mimetic surface functionalization of
nanoparticles: methods and applications. Adv Colloid Interface Sci 197:68–84
7. Bratov A, Abramova N (2010) Chemical sensors and biosensors based on impedimetric
interdigitated electrode array transducers. In: Harrison RV (ed) Chemical sensors: properties,
performance and applications. Nova Science Publisher Inc, New York, pp 93–117
8. Davies ML et al (1992) Polymer membranes in clinical sensor applications 1. An overview
of membrane-function. Biomaterials 13(14):971–978
98 S.M. Reddy

9. Zajoncova L, Pospiskova K (2009) Membranes for amperometric biosensors. Chem Listy

103(4):291–301
10. Major KJ et al (2014) Filter-based chemical sensors for hazardous materials. In: Bishop SS,
Isaacs JC (eds) Detection and sensing of mines, explosive objects, and obscured targets, XIX
11. Sierra C et al (2014) Multicolour PbSe sensors for analytical applications. Sens Actuators
B-Chem 190:464–471
12. Ansari AA et al (2010) Prospects of nanotechnology in clinical immunodiagnostics. Sensors
10(7):6535–6581
13. Soldatkin AP et al (2013) Biosensors. A quarter of a century of R&D experience. Biopolym.
Cell 29(3):188–206
14. Allan JTS, Prest LE, Easton EB (2015) The sulfonation of polyvinyl chloride: synthesis and
characterization for proton conducting membrane applications. J Membr Sci 489:175–182
15. Pankratova N et al (2015) Potentiometric sensing array for monitoring aquatic systems.
Environ Sci Process Impacts 17(5):906–914
16. Vashist SK et al (2011) Technology behind commercial devices for blood glucose
monitoring in diabetes management: a review. Anal Chim Acta 703(2):124–136
17. Alegret S et al (1999) The strategy of biosensor surface renewal: past, present and future.
Quim Anal (Barcelona) 18:23–29
18. Arwin H (2014) Adsorption of proteins at solid surfaces. In: Hinrichs K, Eichhorn KJ
(eds) Ellipsometry of functional organic surfaces and films. Springer, Berlin, pp 29–46
19. Cho WK, Kang SM, Lee JK (2014) Non-biofouling polymeric thin films on solid substrates.
J Nanosci Nanotechnol 14(2):1231–1252
20. Gavalas VG, Berrocal MJ, Bachas LG (2006) Enhancing the blood compatibility of
ion-selective electrodes. Anal Bioanal Chem 384(1):65–72
21. Karyakin AA et al (2002) Optimal environment for glucose oxidase in perfluorosulfonated
ionomer membranes: improvement of first-generation biosensors. Anal Chem 74(7):
1597–1603
22. Treloar PH, Christie IM, Vadgama PM (1995) Engineering the right membranes for
electrodes at the biological interface—solvent-cast and electropolymerized. Biosens
Bioelectron 10(1–2):195–201
23. Ciobanu M et al (2009) Aromatic polysulfones used in sensor applications. Rev Adv Mater
Sci 22(1–2):89–96
24. Reddy SM, Gangadharan B, Manini J (2004) Amperometric sensor study of the selectivity
of poly(vinyl chloride) membranes plasticised with surfactants and liquid crystals. Anal
Chim Acta 508(1):71–78
25. Reddy SM, Vadgama PM (1997) A study of the permeability properties of surfactant
modified poly(vinyl chloride) membranes. Anal Chim Acta 350(1–2):67–76
26. Reddy SM, Vadgama PM (1997) Ion exchanger modified PVC membranes selectivity
studies and response amplification of oxalate and lactate enzyme electrodes. Biosens
Bioelectron 12(9–10):1003–1012
27. Brglez P et al (2014) Spin-coating for optical-oxygen-sensor preparation. Mater Technol 48
(2):181–188
28. Karpitschka S, Weber CM, Riegler H (2015) Spin casting of dilute solutions: vertical
composition profile during hydrodynamic-evaporative film thinning. Chem Eng Sci
129:243–248
29. Han JH et al (2007) Glucose biosensor with a hydrophilic poly-urethane (HPU) blended with
polyvinyl alcohol/vinyl butyral copolymer (PVAB) outer membrane. Sens Actuators
B-Chem 123(1):384–390
30. Julkapli NM, Akil HM, Ahmad Z (2011) Preparation, properties and applications of
chitosan-based biocomposites/blend materials: a review. Compos Interf 18(6):449–507
31. Naghib SM et al (2012) Investigation of a biosensor based on phenylalanine dehydrogenase
immobilized on a polymer-blend film for phenylketonuria diagnosis. Electroanalysis 24
(2):407–417
4 Membrane Technologies for Sensing and Biosensing 99

32. Zapotoczny S (2012) Stimuli responsive polymers for nanoengineering of biointerfaces.

Nanotechnol Regenerative Med Methods Protoc 811:51–78
33. Reddy SM, Vadgama PM (1997) Surfactant-modified poly(vinyl chloride) membranes as
biocompatible interfaces for amperometric enzyme electrodes. Anal Chim Acta 350
(1–2):77–89
34. Kyrolainen M, Reddy SM, Vadgama PM (1997) Blood compatibility and extended linearity
of lactate enzyme electrode using poly(vinyl chloride) outer membranes. Anal Chim Acta
353(2–3):281–289
35. Movileanu L (2014) Watching single proteins using engineered nanopores. Protein Pept Lett
21(3):235–246
36. Wang Q et al (2015) Biological recognition with bio-inspired smart nanopores and
nanochannels based on polymer membrane. Sci Adv Mater 7(10):2147–2167
37. Guo Z et al (2011) Application of biomimetic nanopore fabricated in self-supported
membrane in analytical chemistry. Prog Chem 23(10):2103–2112
38. Fink D et al (2014) Coupled chemical reactions in dynamic nanometric confinement: V. The
influence of Li+ and F− ions on etching of nuclear tracks in polymers. Radiat Eff Defects
Solids 169(5):396–417
39. Fink D et al (2012) Optimization of transport processes in etched track-based biosensors.
Radiat Eff Defects Solids 167(8):548–568
40. Zhang Y et al (2015) Nanoporous membranes generated from self-assembled block polymer
precursors: Quo Vadis? J Appl Polym Sci 132(21)
41. Aram E, Mehdipour-Ataei S (2016) A review on the micro- and nanoporous polymeric
foams: preparation and properties. Int J Polym Mater Polym Biomater 65(7):358–375
42. Toccafondi C et al (2015) Biomedical applications of anodic porous alumina. Curr Nanosci
11(5):572–580
43. Jarolimova Z et al (2016) Potassium ion-selective fluorescent and pH independent
nanosensors based on functionalized polyether macrocycles. Chem Sci 7(1):525–533
44. Olsen G, Ulstrup J, Chi Q (2016) Crown-ether derived graphene hybrid composite for
membrane free potentiometric sensing of alkali metal ions. ACS Appl Mater Interf 8
(1):37–41
45. Saiapina OY et al (2016) Development of conductometric sensor based on 25,27-Di-
(5-thio-octyloxy)calix 4 arene-crown-6 for determination of ammonium. Nanoscale Res Lett
11(1):105
46. Ayanoglu MN et al (2015) Salicylate ion-selective electrode based on a calix 4 arene as
ionophore. Electroanalysis 27(7):1676–1684
47. Aydogan A, Akar A (2012) Tri- and pentacalix 4 pyrroles: synthesis, characterization and
their use in the extraction of halide salts. Chem Eur J 18(7):1999–2005
48. Reddy SM, Higson SP, Vadgama PM (1997) Amperometric enzyme electrode for the
determination of urine oxalate. Anal Chim Acta 343(1–2):59–68
49. Liu M et al (2011) A stable sandwich-type amperometric biosensor based on poly
(3,4-ethylenedioxythiophene)-single walled carbon nanotubes/ascorbate oxidase/nafion films
for detection of L-ascorbic acid. Sens Actuators B-Chem 159(1):277–285
50. Tang H et al (2004) Amperometric glucose biosensor based on adsorption of glucose
oxidase at platinum nanoparticle-modified carbon nanotube electrode. Anal Biochem 331
(1):89–97
51. Christie IM et al (1997) The use of chemical sensor systems for sampling, selectivity and
speciation. In: Nikolelis DP et al (eds) Biosensors for direct monitoring of environmental
pollutants in field, vol 38. Springer, Berlin, pp 41–56
52. Christie IM et al (1997) Plasticized PVC as a high efficiency selective barrier for the
amperometric detection of phenolics. Electroanalysis 9(14):1078–1082
53. Reddy SM, Vadgama PM (1997) Membranes to improve amperometric sensor character-
istics. In: Kress-Rogers E (ed) Handbook of biosensors and electronic noses: medicine, food,
and the environment. CRC Press, Boca Raton, pp 111–135
100 S.M. Reddy

54. Gong JP et al (1991) Electroconductive organogel. 3. Preparation and properties of a

charge-transfer complex gel in an organic-solvent. Macromolecules 24(19):5246–5250
55. Guiseppi-Elie A (2010) Electroconductive hydrogels: synthesis, characterization and
biomedical applications. Biomaterials 31(10):2701–2716
56. Bai H et al (2011) Graphene oxide/conducting polymer composite hydrogels. J Mater Chem
21(46):18653–18658
57. Greco F et al (2013) Liquid single crystal elastomer/conducting polymer bilayer composite
actuator: modelling and experiments. Soft Matter 9(47):11405–11416
58. Kang D-W et al (2015) Enhancement of primary neuronal cell proliferation using
printing-transferred carbon nanotube sheets. J Biomed Mater Res, Part A 103(5):1746–1754
59. Wei Q-B et al (2008) Assembly of Cu nanoparticles in a polyacrylamide grafted poly(vinyl
alcohol) copolymer matrix and vapor-induced response. Sens Actuators B-Chem 134(1):49–
56
60. Moschou EA et al (2006) Voltage-switchable artificial muscles actuating at near neutral pH.
Sens Actuators B-Chem 115(1):379–383
61. Lira LM, de Torresi SIC (2005) Conducting polymer-hydrogel composites for electrochem-
ical release devices: synthesis and characterization of semi-interpenetrating
polyaniline-polyacrylamide networks. Electrochem Commun 7(7):717–723
62. Nikpour M et al (1999) Porous conducting membranes based on polypyrrole-PMMA
composites. Synth Met 99(2):121–126
63. Brahim S, Narinesingh D, Guiseppi-Elie A (2001) Amperometric determination of
cholesterol in serum using a biosensor of cholesterol oxidase contained within a
polypyrrole-hydrogel membrane. Anal Chim Acta 448(1–2):27–36
64. Brahim SI et al (2002) Design and characterization of a galactose biosensor using a novel
polypyrrole-hydrogel composite membrane. Anal Lett 35(5):797–812
65. Gerritsen M et al (2000) Biocompatibility evaluation of sol-gel coatings for subcutaneously
implantable glucose sensors. Biomaterials 21(1):71–78
66. Nederberg F et al (2006) Biocompatible and biodegradable phosphorylcholine ionomers
with reduced protein adsorption and cell adhesion. J Biomater Sci Polym Ed 17(6):605–614
67. Snyder TA et al (2007) Preclinical biocompatibility assessment of the EVAHEART
ventricular assist device: coating comparison and platelet activation. J Biomed Mater Res,
Part A 81A(1):85–92
68. Sun F et al (2011) Improving hydrophilicity and protein resistance of silicone hydrogel by
plasma induced graft polymerization of 2-methacryloyloxyethyl phosphorylcholine.
e-Polymer 11(1):463–473
69. Zhang T et al (2008) Synthesis and characterization of phosphoryl-choline-capped poly
(epsilon-caprolactone)-poly(ethylene oxide) di-block co-polymers and its surface modifica-
tion on polyurethanes. J Biomater Sci Polym Ed 19(4):509–524
70. Abraham S et al (2005) Molecularly engineered p(HEMA)-based hydrogels for implant
biochip biocompatibility. Biomaterials 26(23):4767–4778
71. Brahim S, Narinesingh D, Guiseppi-Elie A (2003) Synthesis and hydration properties of
pH-sensitive p(HEMA)-based hydrogels containing 3-(trimethoxysilyl)propyl methacrylate.
Biomacromolecules 4(3):497–503
72. Boztas AO, Guiseppi-Elie A (2009) Immobilization and release of the redox mediator
ferrocene monocarboxylic acid from within cross-linked p(HEMA-co-PEGMA-co-HMMA)
hydrogels. Biomacromolecules 10(8):2135–2143
73. Ramanavicius A et al (1999) Polypyrrole entrapped quinohemoprotein alcohol dehydroge-
nase. Evidence for direct electron transfer via conducting-polymer chains. Anal Chem 71
(16):3581–3586
74. Reddy SM, Sette G, Phan Q (2011) Electrochemical probing of selective haemoglobin
binding in hydrogel-based molecularly imprinted polymers. Electrochim Acta 56(25):
9203–9208
75. Bueno L et al (2014) MIP-based electrochemical protein profiling. Sens Actuators B-Chem
204:88–95
4 Membrane Technologies for Sensing and Biosensing 101

76. El-Sharif HF, Phan QT, Reddy SM (2014) Enhanced selectivity of hydrogel-based
molecularly imprinted polymers (HydroMIPs) following buffer conditioning. Anal Chim
Acta 809:155–161
77. El-Sharif HF, Aizawa H, Reddy SM (2015) Spectroscopic and quartz crystal microbalance
(QCM) characterisation of protein-based MIPs. Sens Actuators B-Chem 206:239–245
78. Reddy SM et al (2013) Protein detection using hydrogel-based molecularly imprinted
polymers integrated with dual polarisation interferometry. Sens Actuators B-Chem
176:190–197
79. Reuber J, Reinhardt H, Johannsmann D (2006) Formation of surface-attached responsive gel
layers via electrochemically induced free-radical polymerization. Langmuir 22
(7):3362–3367
80. Wu S, Tan W, Xu H (2010) Protein molecularly imprinted polyacrylamide membrane: for
hemoglobin sensing. Analyst 135(10):2523–2527
81. Xu S et al (2013) Stimuli-responsive molecularly imprinted polymers: versatile functional
materials. J Mater Chem C 1(29):4406–4422
82. Gong C, Lam MH-W, Yu H (2006) The fabrication of a photoresponsive molecularly
imprinted polymer for the photoregulated uptake and release of caffeine. Adv Funct Mater
16(13):1759–1767
83. Qin L et al (2009) Macroporous thermosensitive imprinted hydrogel for recognition of
protein by metal coordinate interaction. Anal Chem 81(17):7206–7216
84. El-Sharif HF et al (2015) Highly selective BSA imprinted polyacrylamide hydrogels
facilitated by a metal-coding MIP approach. Acta Biomater 28:121–127
85. Zhang W et al (2012) Thermo-sensitive imprinted polymer coating CdTe quantum dots for
target protein specific recognition. Chem Commun 48(12):1757–1759
86. Aburto J, Le Borgne S (2004) Selective adsorption of dibenzothiophene sulfone by an
imprinted and stimuli-responsive chitosan hydrogel. Macromolecules 37(8):2938–2943
87. Suedee R et al (2006) Temperature sensitive dopamine-imprinted (N, N-
methylene-bis-acrylamide cross-linked) polymer and its potential application to the selective
extraction of adrenergic drugs from urine. J Chromatogr A 1114(2):239–249
88. Huang ZM et al (2003) A review on polymer nanofibers by electrospinning and their
applications in nanocomposites. Compos Sci Technol 63(15):2223–2253
89. De Cesare F, Macagnano A (2015) Electrospinning-based nanobiosensors. In:
Macagnano A, Zampetti E, Kny E (eds) Electrospinning for high performance sensors.
Springer, Berlin, pp 225–279
90. Jiang T et al (2015) Electrospinning of polymer nanofibers for tissue regeneration. Prog
Polym Sci 46:1–24
91. Pelipenko J, Kocbek P, Kristl J (2015) Critical attributes of nanofibers: preparation, drug
loading, and tissue regeneration. Int J Pharm 484(1–2):57–74
92. Sridhar R et al (2015) Electrosprayed nanoparticles and electrospun nanofibers based on
natural materials: applications in tissue regeneration, drug delivery and pharmaceuticals.
Chem Soc Rev 44(3):790–814
93. Li DP, Frey MW, Baeumner AJ (2006) Electrospun polylactic acid nanofiber membranes as
substrates for biosensor assemblies. J Membr Sci 279(1–2):354–363
94. Mahmoudifard M et al (2016) Efficient protein immobilization on polyethersolfone
electrospun nanofibrous membrane via covalent binding for biosensing applications.
Mater Sci Eng C-Mater Bio Appl 58:586–594
95. Wu CM, Yu SA, Lin SL (2014) Graphene modified electrospun poly(vinyl alcohol)
nanofibrous membranes for glucose oxidase immobilization. Express Polym Lett 8
(8):565–573
96. Su X et al (2013) A new amperometric glucose biosensor based on one-step electrospun poly
(vinyl alcohol)/chitosan nanofibers. J Biomed Nanotechnol 9(10):1776–1783
97. Wang N et al (2013) Electrospun fibro-porous polyurethane coatings for implantable glucose
biosensors. Biomaterials 34(4):888–901
102 S.M. Reddy

98. Zhou C et al (2013) Development of a fast and sensitive glucose biosensor using iridium
complex-doped electrospun optical fibrous membrane. Anal Chem 85(2):1171–1176
99. Ji X et al (2014) “Ready-to-use” hollow nanofiber membrane-based glucose testing strips.
Analyst 139(24):6467–6473
100. Costa RR, Mano JF (2014) Polyelectrolyte multilayered assemblies in biomedical
technologies. Chem Soc Rev 43(10):3453–3479
101. Karabulut E et al (2012) Adhesive layer-by-layer films of carboxymethylated cellulose
nanofibril dopamine covalent bioconjugates inspired by marine mussel threads. ACS Nano 6
(6):4731–4739
102. Donot F et al (2012) Microbial exopolysaccharides: main examples of synthesis, excretion,
genetics and extraction. Carbohydr Polym 87(2):951–962
103. Martins GV, Mano JF, Alves NM (2011) Dual responsive nanostructured surfaces for
biomedical applications. Langmuir 27(13):8415–8423
104. Zankovych S et al (2013) The effect of polyelectrolyte multilayer coated titanium alloy
surfaces on implant anchorage in rats. Acta Biomater 9(1):4926–4934
105. Rivera-Gil P et al (2009) Intracellular processing of proteins mediated by biodegradable
polyelectrolyte capsules. Nano Lett 9(12):4398–4402
106. Szarpak A et al (2010) Designing hyaluronic acid-based layer-by-layer capsules as a carrier
for intracellular drug delivery. Biomacromolecules 11(3):713–720
107. Fujie T et al (2010) Dual therapeutic action of antibiotic-loaded nanosheets for the treatment
of gastrointestinal tissue defects. Biomaterials 31(24):6269–6278
108. Kim Y, Rajagopalan P (2010) 3D hepatic cultures simultaneously maintain primary
hepatocyte and liver sinusoidal endothelial cell phenotypes. Plos One 5(11):e15456
109. Sher P, Custodio CA, Mano JF (2010) Layer-by-layer technique for producing porous
nanostructured 3D constructs using moldable freeform assembly of spherical templates.
Small 6(23):2644–2648
110. Kreft O et al (2007) Polymer microcapsules as mobile local pH-sensors. J Mater Chem 17
(42):4471–4476
111. Kazakova LI et al (2013) Chemosensors and biosensors based on polyelectrolyte
microcapsules containing fluorescent dyes and enzymes. Anal Bioanal Chem 405
(5):1559–1568
112. Abd-Elnaiem AM et al (2016) Tailoring the porous nanostructure of porous anodic alumina
membrane with the impurity control. J Alloy Compd 659:270–278
113. Das C, Jain B, Krishnamoorthy K (2015) Phenols from green tea as a dual functional coating
to prepare devices for energy storage and molecular separation. Chem Commun 51
(58):11662–11664
114. Wei T-H et al (2013) Porous polymer nanostructures fabricated by the surface-induced phase
separation of polymer solutions in anodic aluminum oxide templates. Langmuir 29
(32):9972–9978
115. Saji VS et al (2015) Localized drug delivery of selenium (Se) using nanoporous anodic
aluminium oxide for bone implants. J Mater Chem B 3(35):7090–7098
116. Zhang J et al (2015) Preparation and size control of sub-100 nm pure nanodrugs. Nano Lett
15(1):313–318
117. Zhang D et al (2011) Polypyrrole/ZnS Core/shell coaxial nanowires prepared by anodic
aluminum oxide template methods. J Phys Chem C 115(5):2360–2365
118. Raj V, Silambarasan J, Rajakumar P (2014) Electrocatalytic reduction of ortho nitroben-
zaldehyde using modified aluminum electrode and its determination. J Environ Sci 26
(7):1531–1539
119. Santos A, Kumeria T, Losic D (2013) Nanoporous anodic aluminum oxide for chemical
sensing and biosensors. TrAC, Trends Anal Chem 44:25–38
120. Sola L et al (2015) Characterization of porous alumina membranes for efficient, real-time,
flow through biosensing. J Membr Sci 476:128–135
121. Mutalib Md Jani A et al (2009) Nanoporous anodic aluminium oxide membranes with
layered surface chemistry. Chem Commun (Cambridge, England) 21:3062–3064
4 Membrane Technologies for Sensing and Biosensing 103

122. Vlassiouk I, Takmakov P, Smirnov S (2005) Sensing DNA hybridization via ionic
conductance through a nanoporous electrode. Langmuir 21(11):4776–4778
123. Szczepanski V, Vlassiouk I, Smirnov S (2006) Stability of silane modifiers on alumina
nanoporous membranes. J Membr Sci 281(1–2):587–591
124. Li P-F et al (2009) Thermo-responsive gating membranes with controllable length and
density of poly(N-isopropylacrylamide) chains grafted by ATRP method. J Membr Sci 337
(1–2):310–317
125. Jain P et al (2007) High-capacity purification of his-tagged proteins by affinity membranes
containing functionalized polymer brushes. Biomacromolecules 8(10):3102–3107
126. Deme B, Marchal D (2005) Polymer-cushioned lipid bilayers in porous alumina. Eur
Biophys J Biophys Lett 34(2):170–179
127. Dotzauer DM et al (2006) Catalytic membranes prepared using layer-by-layer adsorption of
polyelectrolyte/metal nanoparticle films in porous supports. Nano Lett 6(10):2268–2272
128. Oliveira GB et al (2008) Enzyme immobilization on anodic aluminum
oxide/polyethyleneimine or polyaniline composites. React Funct Polym 68(1):27–32
129. Takmakov P, Vlassiouk I, Smirnov S (2006) Application of anodized aluminum in
fluorescence detection of biological species. Anal Bioanal Chem 385(5):954–958
130. Feng C-L et al (2007) Graded-bandgap quantum-dot-modified nanotubes: a sensitive
biosensor for enhanced detection of DNA hybridization. Adv Mater 19(15):1933–1936
131. Kumeria T, Losic D (2012) Controlling interferometric properties of nanoporous anodic
aluminium oxide. Nanoscale Res Lett 7(1):1
132. Kumeria T et al (2012) Label-free reflectometric interference microchip biosensor based on
nanoporous alumina for detection of circulating tumour cells. Biosens Bioelectron 35
(1):167–173
133. Shimomura T et al (2009) Amperometric determination of choline with enzyme immobilized
in a hybrid mesoporous membrane. Talanta 78(1):217–220
134. Schmitt EK et al (2008) Electrically insulating pore-suspending membranes on highly
ordered porous alumina obtained from vesicle spreading. Soft Matter 4(2):250–253
135. Grieshaber D et al (2008) Electrochemical biosensors—sensor principles and architectures.
Sensors 8(3):1400–1458
136. Grieshaber D et al (2008) Vesicles for signal amplification in a biosensor for the detection of
low antigen concentrations. Sensors 8(12):7894–7903
137. Venkatesan BM, Bashir R (2011) Nanopore sensors for nucleic acid analysis. Nat
Nanotechnol 6(10):615–624
138. Venkatesan BM et al (2011) Lipid bilayer coated Al2O3 nanopore sensors: towards a hybrid
biological solid-state nanopore. Biomed Microdevices 13(4):671–682
139. Wang L et al (2009) A novel electrochemical biosensor based on dynamic
polymerase-extending hybridization for E. coli O157:H7 DNA detection. Talanta 78
(3):647–652
140. Chae I et al (2015) Electronic nose for recognition of volatile vapor mixtures using a
nanopore-enhanced opto-calorimetric spectroscopy. Anal Chem 87(14):7125–7132
141. Yim C et al (2015) CO2-selective nanoporous metal-organic framework microcantilevers.
Sci Rep 5:10674
142. Lee M et al (2015) Surface wetting of superhydrophobic aluminum oxide nanostructures
investigated using the fiber-optic spectrometer and quartz crystal microbalance. Sens
Actuators B-Chem 220:799–804
Chapter 5
Interfacing Graphene for Electrochemical
Biosensing

Onur Parlak

5.1 Introduction

The integration of carbon-based materials to bridge the biological and electronic

worlds has fundamentally changed the understanding of how to generate functional
bioelectronic devices, including biosensors, biofuel cells and bioactuators, and also
opened up a new window for the future of bioelectronics [1, 2]. The use of
carbon-based nanomaterials as a functional interface provides many practical
solutions and has recently emerged as a highly promising route to overcome
technical problems in the regulation of communication between biological and
electronics systems (Fig. 5.1) [3]. Moreover, the interfacing of these materials is
yielding a broad platform of functional units for bioelectronic interfaces and is
beginning to have significant impact on many fields within the life sciences [4].
In general sense, the first condition for the successful interfacing between bio-
logical and electronic systems is to find the right interface. The interface should be
first a “good host” for the biomolecules [5]. For example, if the biomolecule is a
protein, the activity should remain as close to the native activity as possible, or if
the biomolecule is DNA, the structure and composition should remain more or less
same after each interaction [6]. A second priority is that the interface should not
interfere with the signal moving between biomolecules and the electronic compo-
nents [7]. For these reasons, the investigation of the interface between nanostruc-
tured carbon-based materials and biomolecules, in order to understand the basics of
the dynamic physical and chemical interactions, and kinetic and thermodynamic
exchanges, is crucial. In order to yield successful bioelectronic devices, whether at
the laboratory scale or commercial, we must understand the dynamic forces and
molecular components that shape these interactions. Even though it is not easy to

O. Parlak (&)
Department of Materials Science and Engineering, Stanford University,
Stanford, CA 94305, USA
e-mail: [email protected]

© Springer International Publishing AG 2017 105

T.R.L.C. Paixão and S.M. Reddy (eds.), Materials for Chemical Sensing,
DOI 10.1007/978-3-319-47835-7_5
106 O. Parlak

Fig. 5.1 Illustrations of merging the area of biotechnology, electronics and materials science.
Copyright © 2015 Elsevier B.V

describe all physical and chemical interactions with high certainty, we need to give
at least conceptual remarks to guide this investigation.
Many important advances have been achieved in the field of material science
bioelectronics in last two decades; however, there are still many technical chal-
lenges waiting for real solutions for both the laboratory scale and for commercial
applications [8]. Researchers have continuously tried to overcome technical diffi-
culties to create more precise platforms not only for fundamental studies but also for
applied studies. One of the obstacles for the realisation of this concept is the lack of
functional interfaces to bring practical solutions to bioelectronics [9]. Progress in
materials science and specifically, carbon-based nanomaterial technology, adds new
dimensions to the area of bioelectronics. Different classes of carbon ranging from
carbon nanotubes to graphene offering different nano-features, structure and
dimensions provide nano-biointerfaces with potentially novel electronic properties.
In the following section, we look at the recently emerging field of carbon-based
interfaces and their implications for bioelectronics focusing on graphene and related
materials. We seek to piece together early breakthroughs and key developments,
highlight and discuss the future of graphene-based bioelectronics by concentrating
on recent studies on electrochemical biosensing. But first, we must ponder the
question: “What makes graphene an important material in the area of electro-
chemical biosensing?”
5 Interfacing Graphene for Electrochemical Biosensing 107

5.2 Fundamentals of Graphene and Graphene Derivatives

Graphene consists of single layer of sp2-bonded carbon atoms arranged in

two-dimensional structure [10]. It has “exploded” into many different research areas
as a “promising” material, which brings exclusive chemical and physical properties
[11]. However, early discussions and theoretical explanations of graphene date back
to the early 1940s and the proof of its existence as a single-layer material since the
1960s [12]. Recent interest on graphene has resulted mainly from the award of the
2010 Nobel Physics Prize given to Andre Geim and Konstantin Novoselov for their
“ground-breaking experiments regarding the two-dimensional material graphene”
related to their very simple methodology with the so-called scotch-tape experiment
in 2004 [10]. Even though graphene had been introduced before 2004 by other
scientists, they generally failed to explain any of the “extraordinary” properties of
graphene. Since 2004, and especially after the 2010 Nobel Prize, studies related to
graphene and its derivatives and their applications have increased immensely.
Graphene is basically considered as a fundamental building block of similar
types of carbon materials, including three-dimensional graphite, one-dimensional
carbon nanotubes (CNTs) and zero-dimensional fullerene [13]. In addition to
structural similarity, there are other types of atomically thick, layered structure that
are extensively used such as graphene oxide. Before going through the details about
graphene, it is useful to understand the structural and chemical limits of the defi-
nition of graphene. Figure 5.2 shows a classification of different graphene types
according to three fundamental graphene-based material properties: number of
layers, lateral dimensions and atomic carbon/oxygen ratio [13].
Several synthetic methods have been developed recently for the synthesis of
graphene. These methods are divided into two main categories in a similar fashion
to the general methodologies for most of the nanostructure materials: top-down and
bottom-up approaches as listed in Fig. 5.3 [14].
Graphene and graphene-based materials provide functional interface in the area
of bioelectronics due to their various physicochemical properties, which allow
control of their interfacial electrochemical properties [15]. Graphene and
graphene-based materials are used mainly as voltammetric, potentiometric and
impedimetric bio-sensors, and as field-effect transistor-based (bio)sensors due not
only to their “unique” electrochemical properties, but also because of their ease of
processing and high-loading capacity of biomolecules together with biocompati-
bility [1, 16].
Since graphene is the key building block for graphitic structures, the study of
highly ordered pyrolytic graphite (HOPG) which has been conducted earlier gives
very valuable information for today’s graphene research [17]. As described before,
graphitic materials show anisotropic behaviour over different parts of same surface.
The graphite surface possesses two distinctive parts having different electrochem-
ical properties [18]. These are called basal and edge planes. As depicted in Fig. 5.4
and related studies, the electrochemical reactions on the edge plane are enormously
faster than basal plane side, which is mainly due to structural defects on the former,
108 O. Parlak

Fig. 5.2 Classification of graphene-based materials. Copyright © 2014 WILEY-VCH Verlag

GmbH & Co. KGaA, Weinheim

even though the basal plane shows time-dependent chemical activity in some
studies [18]. Another important parameter for an electrode material and their
electrical properties is the density of electronic state (DOS) which varies in different
electrode materials. The DOS value of gold extends to 0.28 states atom−1 eV−1, and
the minimum reported DOS value of HOPG is 0.0022 states atom−1 eV−1, which is
about 0.8 % of gold. Hover, the DOS value can be increased by introducing defects
into the structure [19].
One general way of using graphene as an electrode material is direct immobil-
isation on a macro-electrode surface such as gold or glassy carbon [20]. However, it
is important to note that immobilisation of graphene sheets on macro-electrodes
generates a heterogeneous electrode interface [21]. This creates complications in
revealing the genuine properties of the graphene and usually results in misinter-
pretations. Under these circumstances, the underlying electrode material makes the
main contribution or even dominates with respect to the general electrochemical
activity of interface [1]. The direct use of graphene without an underlying electrode
surface, however, gives clearer information about the electrochemistry of graphene.
Such results demonstrate that the incorporation of individual monolayer crystals as
an electrode material produces an ultra-microelectrode response. In this study, it
was observed that the standard electrochemical rate constant was reduced for a
ferrocene methanol (FcMeOH) probe to *0.5 cm s−1 [18]. This value shows that
5 Interfacing Graphene for Electrochemical Biosensing 109

Fig. 5.3 Schematic illustrations of synthetic approaches for graphene synthesis. Copyright ©
2015 American Chemical Society

the electrode material possesses faster electron transfer kinetics. The reason for this
improvement in electron transfer kinetics is explained by edge plane-like defects
across the basal plane of the graphene and exposed edges, which behave like
ultra-microelectrodes to produce a sigmoidal response. All these properties make
graphene an ideal interface material for electrochemical studies especially in the
area of bioelectronics, where electron transfer and charge transport are relatively
difficult to achieve compared to other systems. Hence, the future of graphene in the
area of bioelectronics is rather promising and offers considerable potential for the
realisation of cutting-edge technologies, providing it is used in the correct way.

5.3 Interactions at the Nano-Biointerface

The interfacing of graphene with biomolecules, such as proteins, DNA, cells and
membranes, brings into focus a wide range of interactions at the interface that
depend mainly on colloidal forces and dynamic physical and chemical interactions.
110 O. Parlak

Fig. 5.4 Schematic representation of a graphitic surface showing basal and edge planes.
Copyright © Royal Society of Chemistry 2015

These interactions play an important role in bio-catalytic processes particularly for

the sensing of physiological analytes. The investigation of various nanostructured
graphene as an interface allows us to develop a strong relationship between gra-
phene and biomaterials and their structures and activities, which are mainly
determined by size, shape and the surface characteristics of the materials. We
believe that this knowledge is crucial from the perspective of efficient use of gra-
phene for applications in bioelectronics [22, 23].
The different aspects of nano-biointerfaces and related interactions are sum-
marised in this section. Generally, we can categorise main types of interaction at the
nano-biointerface as below:
1. surfaces of graphene which are defined by physicochemical composition;
2. solid–liquid interfaces where graphene meet their surroundings; and
3. graphene in contact with the biological substrate.

Physicochemical composition of a graphene: The chemical composition, surface

functionalisation, shape, porosity, crystallinity and defects, and wettability of
nanomaterials play a key role in determining their interfacial properties. The other
important parameters such as surface charge, degree of stability and solubility,
which are determined by the physical and chemical properties of the suspending
media, have a strong impact on the interaction of graphene with the surrounding
medium and biological elements. These parameters affect mainly:
5 Interfacing Graphene for Electrochemical Biosensing 111

Fig. 5.5 Schematic representation of the interface between graphene and biomolecules

1. adsorption of natural or synthetic organic molecules and biomaterials;

2. formation of the double layer, which is crucial for electrochemical applications;
and
3. reduction in the surface energy (Fig. 5.5).

Forces at a solid–liquid interface: In addition to surface composition and prop-

erties of graphene, understanding of the interfacial forces at a solid–liquid interface
is also crucial. To determine the bulk properties of suspensions such as net-charge
and isoelectronic point, steady-state behaviours are usually considered, even though
the interface is not steady state. But the fact is that nano-biointerfaces continuously
change as a result of environmental influences. Even though interactions at the
interface involve large numbers of forces, the successful use of nanomaterials to
achieve measurable outcomes indicates that it is possible to probe the
nano-biointerfaces experimentally.
Forces at the nano-biointerface: The interaction between graphene and biomole-
cules follows some of the same principles as those between colloidal particles. The
van der Waals, electrostatic, hydrophilic/hydrophobic and many others are still
applicable, but they require special attention because the cases occur at the
nanoscale. Since graphene is an atom-thick material, it possesses relatively few
atoms and so forces related to them are highly dependent on the position of their
surface atoms and their standard bulk-permittivity functions [24].

5.4 Graphene in Electrochemical Biosensing

The dimensions of various graphene and graphene-based materials are comparable

to those of biomolecules such as enzymes, antigens/antibodies or deoxyribonucleic
acid (DNA) (Fig. 5.6) [25]. Not surprisingly, the conjugation of biomolecules with
112 O. Parlak

Fig. 5.6 Schematic

representation of the graphene
as a chemical and biochemical
sensing platform

graphene-based nanostructures often yields hybrid systems with new electronic

properties. Indeed, tremendous progress has been accomplished in the realisation of
biomolecule–graphene hybrid systems for various bioelectronic applications [3].
The electrical contacting of redox enzymes with electrodes by using graphene, the
use of graphene–nucleic acid conjugates for the catalytic deposition and inducing
electrical conductivity between electrodes, the electrochemical analysis of metal
ions originating from the chemical reaction of graphene labels associated with
DNA, and the photoelectrochemical assay of enzyme reactions by means of gra-
phene represent a few examples that highlight the potential of biomolecule–gra-
phene hybrid systems in bioelectronic design. Moreover, recently developed
graphene-based nanomaterials are also highly desirable for application in bioelec-
tronics research, due to their biocompatible and flexible nature. Applications
include wearable electronics (e.g. sensors and actuators) and more recently
implantable electronics. Today, for instance, it is almost impossible to imagine
flexible electrode materials without carbon-based interfaces, especially the
two-dimensional form (e.g. graphene, graphene oxide). The integration of many
different sizes and shapes of graphene with biomolecules has enormous potential to
yield new functional systems that may help to miniaturise biosensors, mechanical
devices and electronic circuitry [1, 16, 26, 27]. As a result, it is reasonable to
believe that these graphenes start to play even more important roles in our daily life,
from safety, disease diagnostics, and even life-sustaining technologies.
The miniaturisation of electronic devices and advances in graphene-based
technologies and the production with the application of functional nanostructures is
at the forefront of scientific and industrial attention [28]. The use of graphene as an
5 Interfacing Graphene for Electrochemical Biosensing 113

interface element on their own or as part of a hybrid structure allows new properties
to be exploited in the area of bioelectronics [22]. The different types of graphene-
related materials, from graphene or graphene oxide with different size and shapes
based on 2D structure, have an important ability to provide suitable platforms for
the interfacing of biomolecules for bioelectronic applications. The unique physical
and chemical properties of graphene provide an ideal microenvironment for bio-
molecule immobilisation while retaining their biological activity, and to facilitate
electron transfer between the immobilised biomaterials and electrode surfaces. This
has led to intensive use of graphene for the construction of electrochemical
bio-devices with enhanced analytical performance. Advances in these applications
require a fundamental understanding of the complex interactions between graphene
and bio-systems. Using this insight, the tools of chemical synthesis can be used to
create nanostructured graphene that interact efficiently and predictably with
biosystems including proteins, nucleic acids, cells and tissues.
In the area of nanobioelectronics, biological systems have been employed as a
biochemical transducer of biochemical signals to electronic information. There
have, of course, been many important challenges for the integration of biological
and electronic systems whether at laboratory scale or in commercial applications.
The barriers to charge transport in biological matrices and electron transfer between
redox proteins hinder the construction of efficient interaction between abiotic–biotic
interfaces, to name just two [29]. Most microorganisms are naturally able to affect
external electron transfer to and/or from an electrode surface [6, 30]. Similarly to
the well-established area of electrochemical biosensors, there are three main elec-
tron transfer mechanisms available between microorganisms and electrode surface.
The electron transfer may occur through either direct contact with an electrode or
transfer through conductive wiring between active side of microorganisms and
electrode or mediated transport via redox-active shuttles [31]. However, the pres-
ence of such microorganisms that are able to affect extracellular electron transfer is
limited. Therefore, there is a special interest in the area of bioelectronics to find
general and easy methods that can facilitate electron transfer at biotic–abiotic
interfaces [16]. One way is to interface biomolecules and/or microorganisms with
novel nanomaterials to achieve efficient electron and charge transport. Among
different materials including small molecules in various forms, such as
self-assembled monolayer (SAM) or supramolecular structures, semiconducting or
redox-active polymers, inorganic materials and nanoparticles, and carbon-based
materials but especially graphene stands out as an ideal material. In the following
section, we highlight some important examples from literature which use graphene
successfully as an interface material for various bioelectronic applications.
In one study, researchers developed a very useful bioelectronic platform using a
graphene–lipid bilayer interface to detect bactericidal activity of antimicrobial
peptides (Fig. 5.7) [32]. Here, they achieved to modulate electronic properties of
graphene by charged lipid bilayer adsorbing on the surface. In this way, changes in
membrane integrity led by biorecognition events could be monitored electrically
using an electrolyte-gated biomimetic membrane–graphene transistor.
114 O. Parlak

Fig. 5.7 Schematic representations of biomimetic membrane graphene field-effect transistor and
current voltage measurements with antimicrobial peptides. Copyright © 2010, American Chemical
Society

This study is important in the area of graphene-based bioelectronics due to the

clear demonstration of the effect of charged lipid membranes and the interplay of
charged impurity and ionic screening mechanism in the control of electrical
response in a biomimetic membrane-coated graphene biosensor. Here, it can be
seen that the charged biomimetic membrane imposes an impurity potential which
causes voltage shifts in the charge neutrality point and increases minimum con-
ductivity. The impurity potential is screened by mobile ions upon thinning or
disruption of the membrane by a membrane-disrupting agent, leading to a recovery
of the voltage shift initially induced by the uncoated graphene plane. The sensing
concept developed based on the dynamics of membrane diffusion and changes in
ionic screening on graphene may be extended to other biorecognition systems such
as ligand-receptor binding and gated control of ion channels embedded in
membranes.
In another interesting study, researchers demonstrated a very novel approach of
interfacing graphene and used it as a nanosensor of biomaterials with silk biore-
sorption [33]. Figure 5.8 shows the concept of graphene on interdigitated electrodes
and binding/sensing of pathogenic bacteria by peptides self-assembled on the
graphene transducer. In this study, researchers firstly printed the graphene onto

Fig. 5.8 Schematic illustration of biotransferrable graphene wireless nanosensor. Copyright ©

2012, Rights Managed by Nature Publishing Group
5 Interfacing Graphene for Electrochemical Biosensing 115

water-soluble silk film and then contacted it by an interdigitated electrode together

with an inductive coil antenna. Then, the whole structure was transferred to bio-
materials (e.g. tooth enamel), followed by functionalisation with bifunctional gra-
phene–biorecognition moieties. The resulting device exhibited sensitive
characteristics to chemical and biochemical sensing with a very low detection limit.
One of the emerging fields of bioelectronics research is flexible and/or wearable
biosensing [4, 34]. In particular, wearable biosensors are gaining endless interest
and promise to be one of the great developments in the sector of wearable health
technology. Wearable biosensors, a rapidly evolving category of biosensors
research, are being developed for healthcare applications as well as sports and
military. These devices promise to provide advantages such as ease of use, low cost
and providing real-time information [35]. The use of wearable monitoring devices
or wearable biosensors that allow constant monitoring of physiological signals is
essential for the advancement of both the diagnosis and treatment of
diseases. Wearable systems, in general, are devices that allow physicians to over-
come the limitations of technology and provide a response to the need for remote
monitoring of individuals over weeks or months. There are obvious advantages for
the supporting of patients outside the hospital and thereby freeing up much needed
bed and ward space for critically ill patients. Wearable biosensors typically rely on
wireless sensors enclosed in bandages or patches or in items that can be worn. The
data sets recorded using these systems are then transmitted and processed to detect
events predictive of possible worsening of the patient’s clinical situations and they
are informative of the need for clinical interventions.
In the following study, the authors have demonstrated for the first time the
development of a flexible graphene bio-nanosensor for detecting lactate, based on
the transformation of graphene film from a rigid substrate to a flexible substrate and
immobilizing a bioreceptor on graphene [36]. The bio-nanosensor shows highly
sensitive and rapid detection of lactate, and excellent flexibility under a variety of
mechanical conditions. Researchers designed a flexible nanosensor to detect lactate
using graphene. Recently, wearable and flexible bioelectronics on plastics have
attracted great interest for healthcare, sports and defence applications due to their
advantages of being light weight, bendable, or stretchable. The results show that
flexible lactate biosensors can be fabricated on a variety of plastic substrates. The
developed sensor can also detect lactate under different mechanical bending con-
ditions, the sensor response decreased as the bending angle and number of bending
repetitions increased. In this early example, it is anticipated that these results could
open exciting opportunities for fundamental studies of flexible graphene bioelec-
tronics by using other bioreceptors, as well as a variety of wearable, implantable,
real-time or on-site applications in fields ranging from clinical analysis to defence
(Fig. 5.9).
116 O. Parlak

Fig. 5.9 Schematic illustration of the enzyme-functionalised graphene on a flexible substrate.

Camera and optical images of graphene on a flexible plastic substrate (PET). Copyright © 2012
Elsevier B.V

5.4.1 Switchable Graphene Interface for Biosensing

Applications

Another emerging field of research in bioelectronics is switchable biosensing sys-

tems [9]. There has been growing interest in switchable bio-catalysis in response to
real-life chemical and physical stimuli as a new platform to understand control and
regulation mechanisms underlying physiological processes [37]. According to this
approach, the understanding of natural biochemical interactions and electron
transfer phenomena can be furthered by mimicking biochemical reactions and
controlling the environment and operations in these models by using external
physical stimuli. Modelling of physical interactions of biomolecules in a confined
volume has, for example, had significant impact on efficient bio-catalysis and
functional control by external physical and chemical stimuli using light, tempera-
ture or pH [38]. It is believed that bio-molecular interactions involving
non-covalent bonding, intermolecular forces and van der Waals interactions play an
important role in bio-catalysis. Hence, the control and regulation of these interac-
tions dominate their function.
In recent studies of switchable bio-catalysis on graphene interface, researchers
have reported the fabrication of temperature, light and pH switchable bio-interfaces
5 Interfacing Graphene for Electrochemical Biosensing 117

using graphene-stimuli-responsive polymer hybrids to control and regulate

enzyme-based biomolecular reactions. In each study, they demonstrated that
interfacial bioelectrochemical properties can be tuned with a modest change in the
surroundings of the biomolecules. It is believed that this responds to a major
challenge in nanoscale materials research by regulating the behaviour of switchable
bio-interfaces.
In one study, researchers studied the fabrication of temperature switchable
bio-interfaces using graphene-stimuli-responsive polymer hybrids to control and
regulate enzyme-based biomolecular reactions [39]. Here, they fabricated an on-off
switchable graphene–cholesterol oxidase-poly(N-isopropylacrylamide) polymer
(PNIPAAM) based bio-interface that is capable of positively responding to external
temperature change. At relatively low temperature conditions such as 20 °C,
hydrogen bonds formed an interaction between surface-modified graphene and
polymer, which creates a coalescence of the graphene interface, thereby resulting in
considerable shrinkage at the interface. The shrinking conditions restrict the access
of enzyme to its substrate, resulting in a decrease in the diffusion of reactants and
the consequent activity of the system. In contrast, under high temperature condi-
tions such as at 40 °C, hydrogen bonding is broken. The disappearance of hydrogen
bond at the interface allows access of the biosubstrate to the enzyme, and in this
way biocatalytic interaction is facilitated (Fig. 5.10).

Fig. 5.10 Representations of an on-off switchable bioelectrocatalytic graphene interface.

Copyright © 2013 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim
118 O. Parlak

More importantly, they have elaborated the morphological changes in polymer

using in situ scanning electrochemical microscopy. This study provides the first
example of responsive bioelectronics being achieved on a two-dimensional gra-
phene interface by controlling the external temperature as an on-off switchable
model.
In another study, researchers fabricated chemically stimulated bio-interfaces
comprising stimuli-responsive elements together with biomolecules, which are able
to deliver functionally reversible reactivity with their corresponding analytes [40].
This allows an event to positively respond to the activity of another biological
event, such as an enzyme-based electrocatalytic reaction modulated via pH change
in the medium. In this study, they demonstrated the development of a pH-encoded
bio-catalysis by employing pH-responsive poly(4-vinyl pyridine, P4VP) graphene
oxide bio-interfaces to control and regulate enzyme-based molecular interactions.
Using electrochemical methods, they show that tunable interfacial electrochemical
properties can be generated by relatively modest change in pH of the medium. The
resulting switchable interface facilitated highly specific, on-demand operation of
biosensors and biofuel cells, which has significant potential in a wide range of
analytical and energy-harvesting applications (Fig. 5.11).
In a similar study, temperature and pH switchable interfaces were designed as
well as a light switchable interface consisting of a light-responsive polymer with
two-dimensional graphene [41]. The bio-interface consists of a pyrroloquino-
line quinone (PQQ)-dependent glucose dehydrogenase, immobilised on
two-dimensional graphene and polyacrylamide copolymerised with spiropyran

Fig. 5.11 Representations of pH-encoded switchable graphene interfaces under two different pH
conditions. Copyright © Royal Society of Chemistry 2015
5 Interfacing Graphene for Electrochemical Biosensing 119

Fig. 5.12 Schematic of the light switchable bioelectrocatalytic graphene interface. Copyright ©
2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

methacrylate Poly(Aam-co-SPMA), which is electrostatically assembled together

(Fig. 5.12).
Light-responsive poly(Aam-co-SPMA) is obtained by free radical copolymeri-
sation of spiropyran methacrylate (SP) and acrylamide monomers. In the polymer
chain, SP units are converted to charge-separated merocyanine (MC) upon exposure
to UV light and return back to neutral SP under visible light or dark conditions. To
show the light-sensitive character of the bio-interface, cyclic voltammetry, impe-
dance spectroscopy, amperometry, charge-discharge tests, contact angle measure-
ments and optical characterisation were performed by irradiating the system at
different wavelengths of light to produce an “ON” state (k  385 nm, UV) and an
“OFF” state (k  420 nm, daylight).
120 O. Parlak

5.5 Concluding Remarks

In conclusion, we have sought to reflect the emerging field of graphene-based

bioelectronics by highlighting early breakthroughs and key developments in order
to show the possible advantages and discuss the future of the area. Recently, a
significant number of graphene bio-interfaces based on self-assembled monolayers
and polymers have been described, which provide in-depth understanding of
biomolecular interaction between graphene and biomolecules including proteins,
DNA and cells. A variety of synthesis and assembling techniques have been used to
achieve this target. However, since the current area of interest is “bioelectronics,”
one should ask “how all these graphene-based bio-interfaces will be used in real
devices” or “which features of graphene-based bioelectronics will be beneficial for
the general realisation of bioelectronics”? Answers to these questions still remain
unclear. In many papers, authors have attempted to respond to these conundrums,
but all too often the discussions descend into elaborate rhetoric and persuasive
phraseology that lacks strong underpinning arguments. We believe that some
fundamental challenges remain to be solved for the evolution of graphene-based
bioelectronics. First, biological organisms use signal transduction at a single
molecule level, so optimisation of electronic platforms is required in order to scale
up/down the system, depending on the application or device, to achieve comparable
spatial resolution. Secondly, improvement in the sensitivity of these systems is
crucial. Most of the publications mentioned in this review did not discuss the
stability or selectivity, or biocompatibility of their designs especially wearable
technologies, and yet this should be one of the first concerns of an author seeking to
generate a practical graphene bioelectronic system. Even though there are many
different works available, therefore, we believe the area is still in its infancy and
there is a long way to go before we can clearly see the route to manufacture robust
graphene-based devices that can deliver the promise offered by harnessing bio-
logical interactions and material property. It is believed that there is considerable
potential for future progress in this field, and the concept of interfacing graphene
with bioelectronics is very promising and even though the field does not have a
long track record and we also await further creative experimentation and perspi-
cacity to reveal the true efficacy of this interdisciplinary interface. There is no doubt
that future research will progress the successful interfacing of graphene and related
materials with real applications with expanding sets of complex combinations to
generate a variety of new biosensor instrumentation.

References

1. Wu SX, He QY, Tan CL, Wang YD, Zhang H (2013) Graphene-based electrochemical
sensors. Small 9:1160–1172
2. Huang X, Zeng ZY, Fan ZX, Liu JQ, Zhang H (2012) Graphene-based electrodes. Adv Mater
24:5979–6004
5 Interfacing Graphene for Electrochemical Biosensing 121

3. Wang Y, Li ZH, Wang J, Li JH, Lin YH (2011) Graphene and graphene oxide:
biofunctionalization and applications in biotechnology. Trends Biotechnol 29:205–212
4. Turner APF (2013) Biosensors: sense and sensibility. Chem Soc Rev 42:3184–3196
5. Noy A (2011) Bionanoelectronics. Adv Mater 23:807–820
6. Duan XJ, Lieber CM (2015) Nanoscience and the nano-bioelectronics frontier. Nano Res 8:1–
22
7. Katz E (2014) Implantable bioelectronics. Wiley, Hoboken
8. Rivnay J, Owens RM, Malliaras GG (2014) The rise of organic bioelectronics. Chem Mater
26:679–685
9. Parlak O, Turner APF (2016) Switchable bioelectronics. Biosens Bioelectron 76:251–265
10. Novoselov KS, Geim AK, Morozov SV, Jiang D, Zhang Y, Dubonos SV, Grigorieva IV,
Firsov AA (2004) Electric field effect in atomically thin carbon films. Science 306:666–669
11. Geim AK, Novoselov KS (2007) The rise of graphene. Nat Mater 6:183–191
12. Castro Neto AH, Guinea F, Peres NMR, Novoselov KS, Geim AK (2009) The electronic
properties of graphene. Rev Mod Phys 81:109–162
13. Wick P, Louw-Gaume AE, Kucki M, Krug HF, Kostarelos K, Fadeel B, Dawson KA,
Salvati A, Vazquez E, Ballerini L, Tretiach M, Benfenati F, Flahaut E, Gauthier L, Prato M,
Bianco A (2014) Classification framework for graphene-based materials. Angew Chem Int Ed
53:7714–7718
14. Ambrosi A, Chua CK, Bonanni A, Pumera M (2014) Electrochemistry of graphene and
related materials. Chem Rev 114:7150–7188
15. Song H, Ni Y, Kokot S (2014) Investigations of an electrochemical platform based on the
layered MoS2–graphene and horseradish peroxidase nanocomposite for direct electrochem-
istry and electrocatalysis. Biosens Bioelectron 56:137–143
16. Du J, Catania C, Bazan GC (2014) Modification of abiotic-biotic interfaces with small
molecules and nanomaterials for improved bioelectronics. Chem Mater 26:686–697
17. Compton RG, Banks CE (2007) Understanding voltammetry. World Scientific, Singapore
18. Brownson DAC, Banks CE (2010) Graphene electrochemistry: an overview of potential
applications. Analyst 135:2768–2778
19. Brownson DAC, Kampouris DK, Banks CE (2012) Graphene electrochemistry: fundamental
concepts through to prominent applications. Chem Soc Rev 41:6944–6976
20. Xu C, Zhao CQ, Li M, Wu L, Ren JS, Qu XG (2014) Artificial evolution of graphene oxide
chemzyme with enantioselectivity and near-infrared photothermal effect for cascade
biocatalysis reactions. Small 10:1841–1847
21. Parlak O, Tiwari A, Turner APF, Tiwari A (2013) Template-directed hierarchical
self-assembly of graphene based hybrid structure for electrochemical biosensing. Biosens
Bioelectron 49:53–62
22. Morales-Narvaez E, Merkoci A (2012) Graphene oxide as an optical biosensing platform.
Adv Mater 24:3298–3308
23. Konkena B, Vasudevan S (2012) Understanding aqueous dispersibility of graphene oxide and
reduced graphene oxide through pK(a) measurements. J Phys Chem Lett 3:867–872
24. Kim J, Cote LJ, Kim F, Yuan W, Shull KR, Huang JX (2010) Graphene oxide sheets at
interfaces. J Am Chem Soc 132:8180–8186
25. Song YJ, Qu KG, Zhao C, Ren JS, Qu XG (2010) Graphene oxide: intrinsic peroxidase
catalytic activity and its application to glucose detection. Adv Mater 22:2206–2210
26. Unnikrishnan B, Palanisamy S, Chen SM (2013) A simple electrochemical approach to
fabricate a glucose biosensor based on graphene-glucose oxidase biocomposite. Biosens
Bioelectron 39:70–75
27. Yang K, Feng LZ, Shi XZ, Liu Z (2013) Nano-graphene in biomedicine: theranostic
applications. Chem Soc Rev 42:530–547
28. Xu M, Liang T, Shi M, Chen H (2013) Graphene-like two-dimensional materials. Chem Rev
113:3766–3798
29. Zhirnov VV, Cavin RK (2015) Microsystems for bioelectronics: scaling and performance
limits. Elsevier Science, Amsterdam
122 O. Parlak

30. Min K, Yoo YJ (2014) Recent progress in nanobiocatalysis for enzyme immobilization and its
application. Biotechnol Bioprocess Eng 19:553–567
31. Wooten M, Karra S, Zhang M, Gorski W (2014) On the direct electron transfer, sensing, and
enzyme activity in the glucose oxidase/carbon nanotubes system. Anal Chem 86:752–757
32. Ang PK, Jaiswal M, Lim CHYX, Wang Y, Sankaran J, Li A, Lim CT, Wohland T,
Oezyilmaz B, Loh KP (2010) A bioelectronic platform using a graphene-lipid bilayer
interface. ACS Nano 4:7387–7394
33. Mannoor MS, Tao H, Clayton JD, Sengupta A, Kaplan DL, Naik RR, Verma N,
Omenetto FG, McAlpine MC (2012) Graphene-based wireless bacteria detection on tooth
enamel. Nat Commun 3:763
34. Axisa F, Schmitt PM, Gehin C, Delhomme G, McAdams E, Dittmar A (2005) Flexible
technologies and smart clothing for citizen medicine, home healthcare, and disease
prevention. IEEE Trans Inf Technol Biomed 9:325–336
35. Urban G, Jobst G, Keplinger F, Aschauer E, Fasching R, Svasek P (1994) Miniaturized
integrated biosensors. Technol Health Care: Official J Eur Soc Eng Med 1:215–218
36. Labroo P, Cui Y (2013) Flexible graphene bio-nanosensor for lactate. Biosens Bioelectron
41:852–856
37. Bocharova V, Katz E (2012) Switchable electrode interfaces controlled by physical, chemical
and biological signals. Chem Rec 12:114–130
38. Privman M, Tam TK, Pita M, Katz E (2009) Switchable electrode controlled by enzyme logic
network system: approaching physiologically regulated bioelectronics. J Am Chem Soc
131:1314–1321
39. Parlak O, Turner APF, Tiwari A (2014) On/off-switchable zipper-like bioelectronics on a
graphene interface. Adv Mater 26:482–486
40. Parlak O, Turner APF, Tiwari A (2015) pH-induced on/off-switchable graphene bioelectron-
ics. J Mater Chem B 3:7434–7439
41. Parlak O, Beyazit S, Tse Sum Bui B, Haupt K, Tiwari A, Turner APF (2015) Light-triggered
switchable graphene–polymer hybrid bioelectronics. Adv Mater Interf 3(1500353)
Chapter 6
Nanomaterials as Implantable Sensors

Roger Jagdish Narayan and Nishant Verma

6.1 Introduction

One of the biggest challenges the world is currently facing is a rapid increase in the
population in both developing and developed countries. With increasing popula-
tion, an effort to provide adequate healthcare service while minimizing healthcare
cost is an important issue particularly to “at risk” population group, such as the
elderly. To improve the survivability and quality of life, it requires a continuous
monitoring of their physical and mental conditions in order to diagnose any disease
in early stages; otherwise, it may impart a heavy monetary and administrative
burden. There are a number of diseases such as diabetes, hypertension, asthma,
renal failure, and infectious disease that can easily be diagnosed by monitoring
specific parameters which includes blood glucose level, blood pressure, partial
pressure of oxygen, urea, and inflammatory markers, respectively. Therefore to
achieve this objective, setting a personalized monitoring platform in the form of
wearable and implantable body sensor network systems can be a very effective tool
which allows people to be monitored during their everyday activities [17]. Sensors,
including biosensors, are the exceptional analytical system characterized by their
higher specificity and sensitivity toward the analyte [7]. Biosensors are different
from sensors in that their recognition element is biological in nature. As compared
to in vitro sensors, in vivo sensors are rapidly gaining interest as in vitro sensors
often fail to provide the exact complexity of the intact organ systems and are unable
to monitor biological events in continuous mode [20]. There have been a number of

R.J. Narayan (&)

UNC/NCSU Joint Department of Biomedical Engineering, The University of North
Carolina/North Carolina State University (UNC-NCSU), Raleigh, USA
e-mail: [email protected]
N. Verma
Biotechnology Department, National Institute of Technology, Raipur, Chhattisgarh, India
e-mail: [email protected]

© Springer International Publishing AG 2017 123

T.R.L.C. Paixão and S.M. Reddy (eds.), Materials for Chemical Sensing,
DOI 10.1007/978-3-319-47835-7_6
124 R.J. Narayan and N. Verma

trials to develop implantable sensors for monitoring the physiological and bio-
chemical parameters of patients [15, 24, 42, 59, 67, 91]. For example, GlySens,
California has developed a fully implanted sensor that can provide continuous
monitoring of glucose levels in the body for more than a year; however, the sensor
is still in clinical trials [26]. The development of miniaturized and implantable
sensors that help in the continuous monitoring of metabolites is an emerging area of
scientific and technological interests [85]. Moreover, progress in key areas such as
sensor manufacturing, microelectromechanical system (MEMS) technology, and
nanotechnology offers the prospect of producing miniaturized sensor devices for
sophisticated sensing. The domain of implanted sensors has greatly been benefited
by the advances in nanotechnology leading to new levels of sensitivity, precision,
and rapid response. Different technical issues, such as biocompatibility, miniatur-
ization, biofouling, high power consumption, wireless transmission, and integration
with therapeutic system, are the major focus of current medical research which is
currently being scrutinized to be resolved by the incorporation of nanotechnology.
With the advancement in the technology related to the development and charac-
terization, nanomaterials with novel properties are being tested out for the fabri-
cation of implanted sensors. They are used to provide sustainability in the body in
terms of wear, toxicity, and energy supply [68]. Implantable sensors have limita-
tions and require engineering advances in the form of nanotechnology to match the
accuracy of already established monitoring systems [41]. This initiative will not
only enable us in the early detection of disease but also helps in understanding the
mechanism of disease progression.

6.2 Overview of Nanotechnology

Currently, the field of nanotechnology is being incorporated into all aspects of life.
Nanotechnology deals with the processes that take place on the nanometer scale of
1–100 nm, and nanoscale materials are those materials which have at least one
critical dimension less than 100 nm [8]. In principle, the exact dimension for grain
size below which the materials can be classified as “nano” cannot be defined. The
reason is that it is subjective and depends on the application or end property of
interest [54]. Therefore, nanomaterials may be classified as the materials having at
least one of their dimensions in nanoscale, below which the property of interest
varies significantly. At this smaller size, they exhibit some novel properties, such as
quantum size effect, nonlinear optical properties, electrical and thermal properties,
which are entirely different from the bulk materials [103]. As a result of reduced
dimension, the surface area per unit mass increases which alters the physical and
chemical properties of the material. These novel properties of materials at nanoscale
could further be exploited for the fabrication of implantable sensors with enhanced
performance in terms of biocompatibility, sensitivity, miniaturization, and accuracy
[85]. The fabrication techniques of nanomaterials may be classified on the basis of
phase of the reaction medium, such as gas phase, liquid phase, and solid phase;
6 Nanomaterials as Implantable Sensors 125

however, the most popular approach of nanomaterial synthesis is classified as

top-down and bottom-up methods [12]. A top-down approach corresponds to
reduction of large dimension structure to nanostructure through various size
reduction techniques, while in bottom-up approach, individual atoms or molecules
are used as a building block to produce nanomaterials of desired shape and size.
This latter approach has the advantage of fabricating nanomaterials with homoge-
nous chemical composition and fewer imperfections which may further enhance the
stability of implants [82].

6.3 Biocompatible Material

The appropriate selection of any material to be used as implant is a key factor for
the long-term success of implantable sensors [66]. The major prerequisite for any
implantable device is that the device must be “biocompatible.” However, there is a
serious concern over the use of this term, biocompatibility. Traditionally, it has
been used for any device which is implanted in the body for a long time, and
satisfying the criteria of non-toxic and chemically non/least reactive. It is certain
that a good biocompatible material should be non-toxic to the body, but all the
non-toxic materials are not necessarily found to be biocompatible. With advance-
ment in the technology, and the development of new materials with enhanced
properties, the selection criterion of implanted devices is continuously changing. It
now includes different properties, such as non-toxicity, non-carcinogenicity, and
non-immunogenicity that a material must possess. So, biocompatibility can be
defined in different ways on the basis of a single, or a combination of properties it
possesses. It can be said that biocompatibility is a property of material, but it cannot
be taken as an absolute term without considering its application [55]. Materials of
any types undergo tissue responses when implanted into living tissue as biomate-
rials or medical devices [4, 5]. This biological response may vary from tissue to
tissue; organ to organ; and even individual to individual. It is accepted that the
biological environment will react to any foreign objects that it interacts, but
the point is whether the response is acceptable or not. The interaction between the
material and tissue should be so minimal that material is not affected by the tissue
and the tissue is not affected by the material [57]. The biological environment does
not accept the material completely, but a minimum of negative biological response
is often a desirable characteristic of a biocompatible material. Therefore, the term
“biocompatibility” is not the property of a material but the characteristic of a
material–biological host system and can be defined as ability of a material to
perform with an appropriate host response in a specific situation [87, 88]. Although
biocompatibility is largely dependent on the exact application conditions, in some
cases, different properties of the material such as heterogeneity, chemical compo-
sition, elasticity, surface morphology, surface tension, and surface energy also
affect the biocompatibility of an implant [66]. Therefore, choice of an appropriate
material must be done from two important perspectives: the properties of selected
126 R.J. Narayan and N. Verma

materials to be used as an implant and the information on the biological environ-

ment that will receive the implant.

6.4 Host Response to Implantable Sensors

The entrance to interaction of materials in the human body has generated intense
scientific curiosity and concern from the industrial and academic field [78]. The
host response to any implant may be classified as inflammatory response and
foreign body response (FBR) [56]. Initially, the tissue injury caused by sensor
implantation triggers the inflammatory response which further initiates the wound
healing process of body. This healing process is hampered by the continual pres-
ence of sensor inside the body which leads to the initiation of FBR [30]. In FBR, as
soon as a sensor is implanted in the body, its surface is covered by a thick layer
(10–100 µm) of proteins, cells, and other biological materials, a process called as
sensor “fouling” [90, 100]. Fouling occurs due to a natural response of the body to
prevent any interaction of implant with the surrounding tissues. This encasing of
sensor builds a mass transfer barrier for the diffusion of analyte to the sensing
element. In order to be detected by the sensor, the analyte must overcome this
diffusional barrier of sensor surface fouling. Consequently, the in vivo sensor
performance and its stability are impeded and thus, limited up to few hours to days
[39, 56]. For instance, despite considerable research in the development of glucose
sensor, implantable glucose sensors are still unable to provide in vivo monitoring of
glucose level for a long duration. In addition to this sensor fouling, there are some
other failure modes which affect the biocompatibility and thus, the lifetime of an
implantable sensor. It includes electrode passivation, fibrous encapsulation, and
membrane degradation as shown in Fig. 6.1 [90].
Different approaches have been tried to improve the sensor biocompatibility by
characterizing and preventing the fouling of sensor membrane. It includes sensor
modification by hydrogels, phospholipid-based biomimicry, flow-based system,
Nafion, surfactants, naturally derived materials, covalent attachment, diamond-like
carbons, and topology [89, 90]. A simple strategy to improve the sensor fouling is
to avert the protein adsorption at the sensor surface which can be achieved by the
use of antifouling coating. The coating provides a desired and continuous flow of
analyte toward the sensor by preventing the protein adsorption and by providing
better integration of the sensor within the tissue. Furthermore, consideration must
be taken to ensure that the membrane coating should be sufficiently thin and porous
so that any variation in the analyte concentration can easily be sensed by the
implanted sensor [96]. Another approach to reduce the sensor fouling is to incor-
porate a material that stimulates angiogenesis and prevent cells of immune system
to come at sensor/tissue interface [83].
6 Nanomaterials as Implantable Sensors 127

Fig. 6.1 Different failure modes which affect the biocompatibility of implanted sensor. Reprinted
from Wisniewski and Reichert [90] with permission from Elsevier

6.5 Other Technical Challenges

In addition to the problems wherein the host system plays a crucial role, such as
biocompatibility and biofouling, some other challenges lie ahead that need to be
considered while designing an implantable sensor. It includes a need for better
sensor design that allows minimally invasibe and reliable patient monitoring. It is
always desirable that the implant should be so small so that it can easily be
implanted and explanted (for example, needle assisted) from the body without any
complicated surgery [84]. A reduction in the needle size also reduced the extent to
which the inflammation can occur [43]. Therefore, miniaturization technology
could be exploited in the fabrication of miniaturized sensor devices that would be
helpful in minimizing the effects of foreign body response after implantation. The
area of implantable sensing could further be benefited by incorporating an auto-
mated drug delivery system where the diagnosis and therapeutics can be performed
side by side. Implanted sensors with an automated closed-loop system have been
investigated for the delivery of insulin, also known as “artificial pancreas”, in
diabetes treatment [65, 86]. However, these developments have been hindered by
different factors such as reliability, sub-optimal accuracy, individual variability, and
food intake timings [31]. The power source is another key challenge while
designing a sensor as it determines the size and lifetime of sensor in the host
128 R.J. Narayan and N. Verma

system. Due to limited power supply a battery can provide, a search for constant
source of energy supply is currently being investigated. To provide a continuous
power supply, integration of implanted sensors with enzymatic biofuel cell to
develop a self-powered sensor system could be explored [19, 102]. Some other
power scavenging sources such as motion, vibration, temperature variation, and
ambient electromagnetic fields have also been proposed as the continuous power
source [49]. Since most of the power is generally required for the wireless com-
munication, the development of low-power receivers is in great demand. For
wireless sensing, radio frequency (RF) receiver is the most popular, and extensive
research has been focused on the development of low-power RF transceiver.
Although, a number of frequency bands have been approved for medical implants
which are non-lethal to the body, the efficiency of the receiver is largely influenced
by different factors such as location of implant, power required to run the implanted
device, data upload, and download bandwidth from the implant device [35].
Context awareness is another key factor affecting the efficiency and pervasiveness
of implantable sensors [96]. Since the presence and concentration of different
analytes (for example, metabolites and biomarkers) temporarily changes according
to the physiological situations, such as sleeping, walking, diet, and medications, the
sensor output may deviate from the optimum. This is particularly important when a
therapeutic system is integrated with an implanted sensing system. Therefore, it is
very important to consider the context in which a person is being monitored.
A number of context awareness sensors such as accelerometers, magnetometers,
and gyroscope which are able to define the physical status of body can be used
along with the implanted sensors to develop a multi-sensor array, but in that case,
cost effectiveness will be a major issue.

6.6 Nanomaterials for Implantable Sensing

In recent years, research on implantable sensor for in vivo monitoring is greatly

influenced by the advancement in nanotechnology. This technology is currently
being used to address several key challenges that a conventional sensor (both,
implantable and non-implantable) is facing, by providing a large surface area,
improved catalytic properties, and nanoscale sensors [14]. Various biocompatible
nanomaterials may either be incorporated as a component in the sensing system to
improve its efficiency, or may act as sensors themselves to form nanosensors.
Sensor biocompatibility can be enhanced through use of nanostructured mem-
branes, which have been noted to possess anti-inflammatory properties [32, 60] and
antifouling properties [23, 74]; hence, these materials can be used to suppress the
foreign body response during sensor implantation. The exact mechanism is not
properly understood, but it has been theorized that it interferes with the adsorption
of proteins on the surface of implant by providing a different contact angle and
hydrophobicity [85]. In another instance, ZnO nanowires have been incorporated in
developing self-powered nanosystems to improvize the power harvesting in
6 Nanomaterials as Implantable Sensors 129

implanted sensor [94, 98]. Different forms of nanostructured materials to be

incorporated in a sensor may be classified on the basis of dimensionality such as 0D
(quantum dots, nanoparticles), 1D (nanowires, nanotubes), and 2D (nanosheet,
nanoplates) nanostructures [50]. These nanomaterials can be prepared from a large
variety of source materials such as metals, metals oxides, polymers, ionic com-
pounds, composites, and semiconductors [80]. Here, we will discuss some nano-
materials, such as carbon nanotube (single- and multi-walled), graphene sheets,
graphene foam, dendrimers, and nanowires, which have the potential to be used for
the fabrication of implantable sensors (Fig. 6.2).
For sensor fabrication, nanostructured carbon allotropes, such as carbon nan-
otubes and graphene, have been widely used due to their unique properties such as
high surface to volume ratio, chemical reactivity, and electronic properties. Carbon
nanotubes are the hollow cylinder formed by a unique carbon sheet to form
single-walled carbon nanotubes (SWCNTs), or concentric carbon sheet of different
diameters to form multi-walled carbon nanotubes (MWCNTs). A number of

Fig. 6.2 a Different forms of nanostructured materials such as graphene sheet, single-walled
carbon nanotube, and multi-walled carbon nanotube. Reprinted from Han et al. [28] with
permission from Elsevier. b Helium ion microscopy images of carbon nanofoam with 75 µm field
of view. Reprinted from Mitchell et al. [52] with permission from Elsevier. c Scanning electron
microscopy images of indium phosphide nanowires. Reprinted with permission from Cui et al.
[16]. Copyright (2013) American Chemical Society
130 R.J. Narayan and N. Verma

techniques such as chemical vapor deposition, laser ablation, and arc synthesis can
be used for the synthesis of both types of carbon nanotubes in bulk quantities [63].
In recent years, both SWCNTs and MWCNTs have been widely explored for its
application in sensor development. As compared to MWCNTs, SWCNTs are more
suitable for biomedical applications as they exhibit the three essential figures of
merit (FOM) which are required to develop fluorescence-based sensors: quantum
yield, photostability, and tissue transparency in its emission range [13]. Due to the
semiconducting nature of SWCNTs, where any change in the charged state can
change the device characteristics via field effect, extremely sensitive sensors, i.e.,
carbon nanotube field-effect sensors, can be developed [69]. Since
non-functionalized SWCNTs possess fluorescence instability and biocompatibility,
it is required to chemically alter nanotubes for biosensing purposes [53].
Figure 6.3a shows the schematic representation of modified SWCNTs used for
fluorescent sensing of biomolecules where a shift in the emission wavelength or a
change in the fluorescence intensity occurs after the binding of analyte. Therefore,
the characterizing property of band gap fluorescence of SWCNTs in near-infrared
range could be further explored to develop long-term continuous glucose moni-
toring system [9, 10]. An optical sensor has been developed using single-walled
carbon nanotube for in vivo fluorescence detection of glucose. It is based on the
competitive binding of glucose to dextran (glucose analogue)-coated nanotube
maintained with glucose-specific protein, such as concanavalin A. This protein
attenuates SWCNTs fluorescence which is then reversed by the addition of glucose
[10].
Detection of some important neurotransmitters such as dopamine and serotonin
can also be very important as they are involved in many physiological processes in
the mammalian central nervous system. In vivo detection of these neurotransmitters

Fig. 6.3 a Schematic representation of principle of fluorescent sensing of modified single-walled

carbon nanotube in which binding of analyte results in the fluorescence intensity. Reprinted from
Mundra et al. [53] with permission from Elsevier. b Model diagram of an optical sensor based on
nanotube for glucose detection. Reprinted from Barone et al. [10] with permission from American
Chemical Society
6 Nanomaterials as Implantable Sensors 131

has already been demonstrated in the striatum of anesthetized rat utilizing carbon
fiber microelectrodes modified with single-walled carbon nanotubes [75].
Incorporation of carbon nanotube-coated carbon electrodes in the sensor has been
found to increase the sensitivity and decrease the fouling of electrode. This
approach opens the possibility of its incorporation in human body for the detection
of neurological disorders. Further extension of this approach includes the devel-
opment of sensor for single molecule detection, such as hydrogen peroxide [95] and
nitric oxide [34]. Hydrogen peroxide is relatively important as a marker for those
enzymatic reactions in which it is released as a by-product, such as glucose
detection through glucose oxidase reaction, while nitric oxide (NO) is an important
signaling molecule in inflammation, neurotransmission, smooth muscle relaxation,
and neurodegeneration. A rapid and selective fluorescence detection of NO in the
near-infrared region in a mouse model has been demonstrated by using
single-walled carbon nanotube wrapped with 3,4-diaminophenyl-functionalized
dextran [40]. The near-infrared fluorescence region of SWCNTs has been shown to
be directly bleached by NO in a reversible manner. It has been reported that
alginate-encapsulated SWCNT can be used as an implantable inflammation sensor
for in vivo detection of nitric oxide without any immune response for more than
100 days [33]. Recently, it has been proposed that the excellent electronic prop-
erties of single-walled carbon nanotubes may also be used for the development of
stable implantable pH sensors [25]. In addition to single-walled carbon nanotube,
multi-walled carbon nanotube has also been reported for in vivo determination of
dopamine [37, 46] and epinephrine [11, 62]. Despite their extensive applications,
the use of carbon nanotube as implantable sensors is still debatable due to the
observance of carcinogenic effects in vivo [58, 73, 79].
Graphene, a two-dimensional single atom carbon sheet, has recently attracted
much attention in the area of biosensing due to its unusual properties such as high
catalytic properties, thermal conductivity, large surface area, and biocompatibility.
It is usually synthesized by exploitation of graphite [48] or reduced graphite oxide
[21]. However, its broad application is generally limited by the property of
aggregation during bulk synthesis [36]. Due to a high electrocatalytic activity
toward hydrogen peroxide, a product of glucose oxidase enzymatic reaction, it has
the potential to be used in continuous glucose monitoring system as implanted
sensor [36, 38, 109]. Recently, researchers have developed a graphene-modified
acupuncture needle that can detect dopamine, a neurotransmitter, at a detection limit
of 0.24 µM in human serum [76]. This nanomaterial-based acupuncture needle has
shown the possibility of direct detection of biologically active molecules in vivo.
Graphene oxide, a structurally different form of graphene, has been found appli-
cations in the development of novel electrode system for electrochemical sensing
platform. Different biologically active compounds, such as neurotransmitters,
nucleic acids, uric acids, acetaminophen, and hydrogen peroxide, can be determined
by employing graphene oxide-modified electrodes [107]. Graphene foam, another
novel form of nanostructured carbon with a 3D microporous network, has also
found application in implantable sensing applications due to its high charge transfer
rate, surface area, efficient mass transport of redox species, and non-covalent
132 R.J. Narayan and N. Verma

interaction with biological molecules [92]. It has been utilized to prepare sensors for
various biological analytes, including dopamine [18, 45, 99], glucose [72, 101], uric
acid [45, 99], and tumor biomarker [47].
Dendrimers, a term originates from the word “dendron” means tree in Greek, are
the hyperbranched, radially symmetric nanosized polymeric molecules with
well-defined and homogenous structure (Fig. 6.4a). They are generally synthesized
in a multi-step process through up to ten generations (5–50 nm) in which each
generation represents a layer of branching groups. Different unique properties of
dendrimers, such as structural uniformity, globular shape, monodispersity, high
functional group density, hydrophilicity, and versatility to synthesize dendrimers of
specific nanometric size, can be exploited in the development of highly sensitive
sensor [29]. Several designs of dendrimers have been developed and widely used
for drug and gene delivery, with a few approaches in sensor development. Some
glucose sensors have been fabricated employing dendrimers layer modified elec-
trode [6, 70, 71], but none has yet been investigated under in vivo conditions. To
reduce the foreign body response and increase tissue integration, nitric
oxide-releasing dendrimers have been investigated to develop a needle-type
implantable glucose sensor [42]. Nitric oxide-releasing interfaces have been shown
to reduce the adhesion of inflammatory cells and collagen capsule thickness while
increasing the blood vessel formation near the implants, thus preventing the implant
associated infections. To further improve the biocompatibility and sensitivity of
implanted sensors, metallic nanoparticles, such as platinum nanoparticles, can also
be incorporated in encapsulated form in dendrimers (Fig. 6.4b), as reported by
some researchers [93, 97, 108]. Fluorescence, in most cases, is the detection method
of dendrimer-based sensor, particularly in pH sensing. A dendrimer-based sensor
has been developed by the conjugation of fluorescent pH indicators to the den-
drimer’s scaffold capable of providing in vivo measurement of pH in living HeLa
cells and mouse brain [1, 2]. These dendrimers also have the ability to diagnose a
diseased state related to overproduction of acidic by-products. A pH-sensing
biodegradable near-infrared nanoprobe has been reported that can differentiate
between the healthy tissue and diseased tissue based on their pH [3]. With the
development of biocompatible dendritic polymer, these nanoprobes could further be
explored as implanted sensor for diagnostic purposes.
Recently, nanowire technology is being applied to fabricate miniaturized devices
for chemical and biological sensing [51]. A nanowire is a wire with nanometer scale
length, which may be metallic, semiconducting, or insulating. These nanostructures
may also be considered as one-dimensional (1-D) structure due to their increased
length as compared to width. These nanowires can be synthesized from metals,
metals oxides, silicon/indium/gallium semiconductors, and silicon/titanium oxide
insulators which further control the conductance of synthesized nanowire [77].
These wires, either in the form of nanoelectrodes or nanoelectrodes array, can be
coated with a desired receptor, and any interaction with the analyte, such as nucleic
acids, proteins, and ions, can be sensed through various electrochemical modes with
a high sensitivity. Due to a reduced diameter and the high surface to volume ratio,
the nanowires offer extremely high sensitivity for the detection of several
6 Nanomaterials as Implantable Sensors 133

Fig. 6.4 a A simple architecture of a dendrimer. Reprinted from Twibanire and Grindley [81]
under Creative Commons Attribution license, 2012. b HR-TEM image of a dendrimers
encapsulated platinum nanoparticles used for the development of glucose sensor. Reprinted from
Zhu et al. [108] with permission from Wiley

biomolecular targets such as nucleic acids and proteins. Several nanowire-based

detection devices have been reported, but their use as implanted sensors is yet to be
investigated. For example, silicon nanowires incorporated into arrays have been
successfully used for the detection of cancer markers such as carcinoembryonic
antigen and prostate-specific antigen [106]. In a similar approach, a real-time
voltammetric detection of cancer biomarkers, such as interleukin-10 and osteo-
pontin, has been reported using localized silica nanowires as a template [64]. For
non-enzymatic continuous glucose monitoring, copper nanowires [22, 105] and
copper oxide nanowires [44, 104] have been shown to possess high selectivity and
sensitivity toward glucose without any interference from oxygen and chloride.
Nanowire has also found application in the detection of neurotransmitters that
interact in brain. It has been used in the development of microneedle sensor plat-
form with integrated silicon nanowire tip for biochemical detection, such as neu-
rotransmitter activities during the synaptic communication between neuron cells
[61]. The use of nanowire technology to create arrays which are very small to probe
a single cell has great prospects in vascular implantable and catheter-based sensors
[27].

6.7 Conclusions

Personalized medicine in the form of implantable sensors will have a significant

impact on improving the quality of life by providing a continuous monitoring
system. This will be not only helpful in the diagnosis of several diseases in early
stages, but also empowers the therapeutics technology too. However, there are
134 R.J. Narayan and N. Verma

several issues, such as biocompatibility and biofouling, with implanted sensors

which need to be tackled before it can be implemented at commercial level.
Advancement in miniaturization technology, such as nanotechnology, has largely
benefited the sensor domain by providing different and novel properties at the
nanolevel to develop new generation of implantable sensors. Undoubtedly in future,
the fabrication of implanted sensor with improved sensitivity and biocompatibility
will rely on the development and integration of new materials with nanosized
dimensions.

References

1. Albertazzi L, Brondi M, Pavan GM, Sato SS, Signore G, Storti B, Ratto GM, Beltram F
(2011) Dendrimer-based fluorescent indicators: in vitro and in vivo applications. PLoS ONE
6(12):e28450
2. Albertazzi L, Storti B, Brondi M, Sato SS, Ratto GM, Signore G, Beltram F (2013)
Synthesis, cellular delivery and in vivo application of dendrimer-based pH sensors.
J Visualized Exp (JoVE) 79
3. Almutairi A, Guillaudeu SJ, Berezin MY, Achilefu S, Fréchet JM (2008) Biodegradable
pH-sensing dendritic nanoprobes for near-infrared fluorescence lifetime and intensity
imaging. J Am Chem Soc 130(2):444–445
4. Anderson JM (2001) Biological responses to materials. Annu Rev Mater Res 31(1):81–110
5. Anderson JM, Rodriguez A, Chang DT (2008) Foreign body reaction to biomaterials. In:
Seminars in immunology, vol 2. Elsevier, London, pp 86–100
6. Armada MPG, Losada J, Zamora M, Alonso B, Cuadrado I, Casado CM (2006)
Electrocatalytical properties of polymethylferrocenyl dendrimers and their applications in
biosensing. Bioelectrochemistry 69(1):65–73
7. Bachan Upadhyay LS, Verma N (2013) Enzyme inhibition based biosensors: a review. Anal
Lett 46(2):225–241
8. Bai C, Liu M (2012) Implantation of nanomaterials and nanostructures on surface and their
applications. Nano Today 7(4):258–281
9. Barone PW, Baik S, Heller DA, Strano MS (2005) Near-infrared optical sensors based on
single-walled carbon nanotubes. Nat Mater 4(1):86–92
10. Barone PW, Parker RS, Strano MS (2005) In vivo fluorescence detection of glucose using a
single-walled carbon nanotube optical sensor: design, fluorophore properties, advantages,
and disadvantages. Anal Chem 77(23):7556–7562
11. Beitollahi H, Sheikhshoaie I (2011) Electrocatalytic oxidation and determination of
epinephrine in the presence of uric acid and folic acid at multiwalled carbon
nanotubes/molybdenum (VI) complex modified carbon paste electrode. Anal Methods 3
(8):1810–1814
12. Biswas A, Bayer IS, Biris AS, Wang T, Dervishi E, Faupel F (2012) Advances in top–down
and bottom–up surface nanofabrication: techniques, applications & future prospects. Adv
Colloid Interface Sci 170(1):2–27
13. Boghossian AA, Zhang J, Barone PW, Reuel NF, Kim JH, Heller DA, Ahn JH, Hilmer AJ,
Rwei A, Arkalgud JR (2011) Near-infrared fluorescent sensors based on single-walled
carbon nanotubes for life sciences applications. ChemSusChem 4(7):848–863
14. Cash KJ, Clark HA (2010) Nanosensors and nanomaterials for monitoring glucose in
diabetes. Trends Mol Med 16(12):584–593
15. Chiu N-F, Wang J-M, Liao C-W, Chen C-H, Chen H-C, Yang L-J, Lu S-S, Lin C-W (2005)
An implantable multifunctional needle type biosensor with integrated RF capability. In:
6 Nanomaterials as Implantable Sensors 135

Annual international conference of the engineering in Medicine and Biology Society.

IEEE-EMBS 2005. 27th , 2006. IEEE, pp 1933–1936
16. Cui Y, Wang J, Plissard SR, Cavalli A, Vu TT, van Veldhoven RP, Gao L, Trainor M,
Verheijen MA, Haverkort JE (2013) Efficiency enhancement of InP nanowire solar cells by
surface cleaning. Nano Lett 13(9):4113–4117
17. Darwish A, Hassanien AE (2011) Wearable and implantable wireless sensor network
solutions for healthcare monitoring. Sensors 11(6):5561–5595
18. Dong X, Wang X, Wang L, Song H, Zhang H, Huang W, Chen P (2012) 3D graphene foam
as a monolithic and macroporous carbon electrode for electrochemical sensing. ACS Appl
Mater Interfaces 4(6):3129–3133
19. Du Toit H, Di Lorenzo M (2014) Glucose oxidase directly immobilized onto highly porous
gold electrodes for sensing and fuel cell applications. Electrochim Acta 138:86–92
20. Eckert MA, Vu PQ, Zhang K, Kang D, Ali MM, Xu C, Zhao W (2013) Novel molecular and
nanosensors for in vivo sensing. Theranostics 3(8)
21. Fan Z-J, Kai W, Yan J, Wei T, Zhi L-J, Feng J, Y-m Ren, Song L-P, Wei F (2010) Facile
synthesis of graphene nanosheets via Fe reduction of exfoliated graphite oxide. ACS Nano 5
(1):191–198
22. Fan Z, Liu B, Liu X, Li Z, Wang H, Yang S, Wang J (2013) A flexible and disposable hybrid
electrode based on Cu nanowires modified graphene transparent electrode for non-enzymatic
glucose sensor. Electrochim Acta 109:602–608
23. Feng Y, Wang K, Davies CH, Wang H (2015) Carbon nanotube/alumina/polyethersulfone
hybrid hollow fiber membranes with enhanced mechanical and anti-fouling properties.
Nanomaterials 5(3):1366–1378
24. Fiala J, Bingger P, Ruh D, Foerster K, Heilmann C, Beyersdorf F, Zappe H, Seifert A (2013)
An implantable optical blood pressure sensor based on pulse transit time. Biomed
Microdevices 15(1):73–81
25. Gou P, Kraut ND, Feigel IM, Bai H, Morgan GJ, Chen Y, Tang Y, Bocan K, Stachel J,
Berger L (2014) Carbon nanotube chemiresistor for wireless pH sensing. Sci Rep 4
26. Gough DA, Kumosa LS, Routh TL, Lin JT, Lucisano JY (2010) Function of an implanted
tissue glucose sensor for more than 1 year in animals. Sci Transl Med 2(42):42ra53–42ra53
27. Greco RS, Prinz FB, Smith RL (2004) Nanoscale technology in biological systems. CRC
Press, Boca Raton
28. Han ZJ, Rider AE, Fisher C, van der Laan T, Kumar S, Levchenko I, Ostrikov K (2014)
Biological application of carbon nanotubes and graphene. In: Carbon nanotubes and
graphene, 2nd edn., pp 279–312
29. Hasanzadeh M, Shadjou N, Eskandani M, Soleymani J, Jafari F, de la Guardia M (2014)
Dendrimer-encapsulated and cored metal nanoparticles for electrochemical nanobiosensing.
TrAC Trends Anal Chem 53:137–149
30. Helton KL, Ratner BD, Wisniewski NA (2011) Biomechanics of the sensor-tissue interface
—effects of motion, pressure, and design on sensor performance and the foreign body
response—part I: theoretical framework. J Diabetes Sci Technol 5(3):632–646
31. Hovorka R (2011) Closed-loop insulin delivery: from bench to clinical practice. Nat Rev
Endocrinol 7(7):385–395
32. Ilinskaya A, Dobrovolskaia M (2014) Immunosuppressive and anti-inflammatory properties
of engineered nanomaterials. Br J Pharmacol 171(17):3988–4000
33. Iverson NM, Barone PW, Shandell M, Trudel LJ, Sen S, Sen F, Ivanov V, Atolia E, Farias E,
McNicholas TP (2013) In vivo biosensing via tissue-localizable near-infrared-fluorescent
single-walled carbon nanotubes. Nat Nanotechnol 8(11):873–880
34. Iverson NM, Strano MS, Wogan GN (2014) In vivo delivery of nitric oxide‐sensing, single‐
walled carbon nanotubes. Curr Protoc Chem Biol 93–102
35. Jiang G (2010) Design challenges of implantable pressure monitoring system. Front
Neurosci 4:2
136 R.J. Narayan and N. Verma

36. Jiang Y, Zhang Q, Li F, Niu L (2012) Glucose oxidase and graphene bionanocomposite
bridged by ionic liquid unit for glucose biosensing application. Sens Actuators B: Chem 161
(1):728–733
37. Kan X, Zhao Y, Geng Z, Wang Z, Zhu J-J (2008) Composites of multiwalled carbon
nanotubes and molecularly imprinted polymers for dopamine recognition. J Phys Chem C
112(13):4849–4854
38. Kang X, Wang J, Wu H, Aksay IA, Liu J, Lin Y (2009) Glucose oxidase–graphene–chitosan
modified electrode for direct electrochemistry and glucose sensing. Biosens Bioelectron 25
(4):901–905
39. Khanna VK (2015) Implantable medical electronics: prosthetics, drug delivery, and health
monitoring. Springer, Berlin
40. Kim J-H, Heller DA, Jin H, Barone PW, Song C, Zhang J, Trudel LJ, Wogan GN,
Tannenbaum SR, Strano MS (2009) The rational design of nitric oxide selectivity in
single-walled carbon nanotube near-infrared fluorescence sensors for biological detection.
Nat Chem 1(6):473–481
41. Klonoff DC (2007) The benefits of implanted glucose sensors. J Diab Sci Tech 1(6):797–800
42. Koh A, Lu Y, Schoenfisch MH (2013) Fabrication of nitric oxide-releasing porous
polyurethane membranes-coated needle-type implantable glucose biosensors. Anal Chem 85
(21):10488–10494
43. Kvist PH, Iburg T, Aalbaek B, Gerstenberg M, Schoier C, Kaastrup P, Buch-Rasmussen T,
Hasselager E, Jensen HE (2006) Biocompatibility of an enzyme-based, electrochemical
glucose sensor for short-term implantation in the subcutis. Diabetes Technol Ther 8(5):546–
559
44. Li C, Yamahara H, Lee Y, Tabata H, Delaunay J-J (2015) CuO
nanowire/microflower/nanowire modified Cu electrode with enhanced electrochemical
performance for non-enzymatic glucose sensing. Nanotechnology 26(30):305503
45. Li H, Wang Y, Ye D, Luo J, Su B, Zhang S, Kong J (2014) An electrochemical sensor for
simultaneous determination of ascorbic acid, dopamine, uric acid and tryptophan based on
MWNTs bridged mesocellular graphene foam nanocomposite. Talanta 127:255–261
46. Lin L, Cai Y, Lin R, Yu L, Song C, Gao H, Li X (2011) New integrated in vivo
microdialysis-electrochemical device for determination of the neurotransmitter dopamine in
rat striatum of freely moving rats. Microchim Acta 172(1–2):217–223
47. Liu J, Wang J, Wang T, Li D, Xi F, Wang J, Wang E (2015) Three-dimensional
electrochemical immunosensor for sensitive detection of carcinoembryonic antigen based on
monolithic and macroporous graphene foam. Biosens Bioelectron 65:281–286
48. Liu N, Luo F, Wu H, Liu Y, Zhang C, Chen J (2008) One-step ionic-liquid-assisted
electrochemical synthesis of ionic-liquid-functionalized graphene sheets directly from
graphite. Adv Funct Mater 18(10):1518–1525
49. Lo B, Yang G-Z (2005) Key technical challenges and current implementations of body
sensor networks. In: Proceeding 2nd international workshop on body sensor networks (BSN
2005)
50. Luz RA, Iost RM, Crespilho FN (2013) Nanomaterials for biosensors and implantable
biodevices. In: Nanobioelectrochemistry. Springer, Berlin, pp 27–48
51. McAlpine MC, Ahmad H, Wang D, Heath JR (2007) Highly ordered nanowire arrays on
plastic substrates for ultrasensitive flexible chemical sensors. Nat Mater 6(5):379–384
52. Mitchell ST, Frese N, Gölzhäuser A, Bowers A, Sattler K (2015) Ultralight carbon
nanofoam from naphtalene-mediated hydrothermal sucrose carbonization. Carbon 95:
434–441
53. Mundra RV, Wu X, Sauer J, Dordick JS, Kane RS (2014) Nanotubes in biological
applications. Curr Opin Biotechnol 28:25–32
54. Murty BS, Shankar P, Raj B, Rath B, Murday J (2013) Textbook of nanoscience and
nanotechnology. Springer Science & Business Media, Berlin
55. Neves H (2013) Materials for implantable systems. Implantable Sens Syst Med Appl 1
6 Nanomaterials as Implantable Sensors 137

56. Onuki Y, Bhardwaj U, Papadimitrakopoulos F, Burgess DJ (2008) A review of the

biocompatibility of implantable devices: current challenges to overcome foreign body
response. J Diabetes Sci Technol 2(6):1003–1015
57. Oshida Y, Guven Y (2015) Biocompatible coatings for metallic biomaterials. Surf Coat
Modif Metallic Biomater 287
58. Pacurari M, Castranova V, Vallyathan V (2010) Single-and multi-wall carbon nanotubes
versus asbestos: are the carbon nanotubes a new health risk to humans? J Toxicol Environ
Health Part A 73(5–6):378–395
59. Paek S-H, Cho I-H, Kim D-H, Jeon J-W, Lim G-S, Paek S-H (2013) Label-free, needle-type
biosensor for continuous glucose monitoring based on competitive binding. Biosens
Bioelectron 40(1):38–44
60. Park H, Kim S, Kim S, Song Y, Seung K, Hong D, Khang G, Lee D (2010) Antioxidant and
anti-inflammatory activities of hydroxybenzyl alcohol releasing biodegradable polyoxalate
nanoparticles. Biomacromolecules 11(8):2103–2108
61. Park I, Li Z, Li X, Pisano AP, Williams RS (2007) Towards the silicon nanowire-based
sensor for intracellular biochemical detection. Biosens Bioelectron 22(9):2065–2070
62. Prasad BB, Prasad A, Tiwari MP, Madhuri R (2013) Multiwalled carbon nanotubes bearing
‘terminal monomeric unit’for the fabrication of epinephrine imprinted polymer-based
electrochemical sensor. Biosens Bioelectron 45:114–122
63. Rafique MMA, Iqbal J (2011) Production of carbon nanotubes by different routes-a review.
J Encapsulation Adsorp Sci 1(02):29
64. Ramgir NS, Zajac A, Sekhar PK, Lee L, Zhukov TA, Bhansali S (2007) Voltammetric
detection of cancer biomarkers exemplified by interleukin-10 and osteopontin with silica
nanowires. J Phys Chem C 111(37):13981–13987
65. Renard E, Place J, Cantwell M, Chevassus H, Palerm CC (2010) Closed-loop insulin
delivery using a subcutaneous glucose sensor and intraperitoneal insulin delivery feasibility
study testing a new model for the artificial pancreas. Diabetes Care 33(1):121–127
66. Saini M, Singh Y, Arora P, Arora V, Jain K (2015) Implant biomaterials: a comprehensive
review. World J Clin (WJCC) 3(1):52
67. Schachtrupp A, Wetter O, Höer J (2015) An implantable sensor device measuring suture
tension dynamics: results of developmental and experimental work. Hernia 1–6
68. Schulz MJ, Shanov VN, Yun Y (2009) Nanomedicine design of particles, sensors, motors,
implants, robots, and devices. Artech House
69. Schwalke U, Rispal L (2008) Fabrication of Ultra-Sensitive Carbon Nanotube Field-Effect
Sensors (CNTFES) for Biomedical Applications. ECS Trans 13(22):39–45
70. Şenel M, Nergiz C (2012) Development of a novel amperometric glucose biosensor based on
copolymer of pyrrole-PAMAM dendrimers. Synth Met 162(7):688–694
71. Şenel M, Nergiz C, Çevik E (2013) Novel reagentless glucose biosensor based on ferrocene
cored asymmetric PAMAM dendrimers. Sens Actuators B: Chem 176:299–306
72. Si P, Dong X-C, Chen P, Kim D-H (2013) A hierarchically structured composite of Mn 3 O
4/3D graphene foam for flexible nonenzymatic biosensors. J Mater Chem B 1(1):110–115
73. Stella GM (2011) Carbon nanotubes and pleural damage: perspectives of nanosafety in the
light of asbestos experience. Biointerphases 6(2):P1–P17
74. Sun X, Wu J, Chen Z, Su X, Hinds BJ (2013) Fouling characteristics and electrochemical
recovery of carbon nanotube membranes. Adv Funct Mater 23(12):1500–1506
75. Swamy BK, Venton BJ (2007) Carbon nanotube-modified microelectrodes for simultaneous
detection of dopamine and serotonin in vivo. Analyst 132(9):876–884
76. Tang L, Du D, Yang F, Liang Z, Ning Y, Wang H, Zhang G-J (2015) Preparation of
graphene-modified acupuncture needle and its application in detecting neurotransmitters. Sci
Rep 5
77. Tiwari A, Turner AP (2015) Advanced bioelectronics materials. Wiley, New York
78. Tjong SC (2009) Advances in biomedical sciences and engineering. Bentham Science
Publishers, Sharjah
138 R.J. Narayan and N. Verma

79. Toyokuni S (2013) Genotoxicity and carcinogenicity risk of carbon nanotubes. Adv Drug
Deliv Rev 65(15):2098–2110
80. Tuantranont A (2013) Applications of Nanomaterials in Sensors and Diagnostics. Springer
Ser Chem Sens Biosens 14
81. Twibanire JdAK, Grindley TB (2012) Polyester dendrimers. Polymers 4(1):794–879
82. Upadhyay LSB, Verma N (2015) Recent developments and applications in plant-extract
mediated synthesis of silver nanoparticles. Anal Lett 48(17):2676–2692
83. Updike SJ, Shults MC, Gilligan BJ, Rhodes RK (2000) A subcutaneous glucose sensor with
improved longevity, dynamic range, and stability of calibration. Diabetes Care 23(2):208–
214
84. Vaddiraju S, Burgess DJ, Tomazos I, Jain FC, Papadimitrakopoulos F (2010) Technologies
for continuous glucose monitoring: current problems and future promises. J Diabetes Sci
Technol 4(6):1540–1562
85. Vaddiraju S, Tomazos I, Burgess DJ, Jain FC, Papadimitrakopoulos F (2010) Emerging
synergy between nanotechnology and implantable biosensors: a review. Biosens Bioelectron
25(7):1553–1565
86. Weinzimer SA, Steil GM, Swan KL, Dziura J, Kurtz N, Tamborlane WV (2008) Fully
automated closed-loop insulin delivery versus semiautomated hybrid control in pediatric
patients with type 1 diabetes using an artificial pancreas. Diabetes Care 31(5):934–939
87. Williams DF (2008) On the mechanisms of biocompatibility. Biomaterials 29(20):2941–
2953
88. Williams DF (2014) There is no such thing as a biocompatible material. Biomaterials 35
(38):10009–10014
89. Wisniewski N, Moussy F, Reichert W (2000) Characterization of implantable biosensor
membrane biofouling. Fresenius’ J Anal Chem 366(6–7):611–621
90. Wisniewski N, Reichert M (2000) Methods for reducing biosensor membrane biofouling.
Colloids Surf B 18(3):197–219
91. Wu Y, Meyerhoff ME (2008) Nitric oxide-releasing/generating polymers for the develop-
ment of implantable chemical sensors with enhanced biocompatibility. Talanta 75(3):642–
650
92. Wujcik EK, Monty CN (2013) Nanotechnology for implantable sensors: carbon nanotubes
and graphene in medicine. Wiley Interdisc Rev: Nanomed Nanobiotechnol 5(3):233–249
93. Xu L, Zhu Y, Tang L, Yang X, Li C (2007) Biosensor based on self-assembling glucose
oxidase and dendrimer-encapsulated Pt nanoparticles on carbon nanotubes for glucose
detection. Electroanalysis 19(6):717–722
94. Xu S, Qin Y, Xu C, Wei Y, Yang R, Wang ZL (2010) Self-powered nanowire devices. Nat
Nanotechnol 5(5):366–373
95. Xu Y, Pehrsson PE, Chen L, Zhang R, Zhao W (2007) Double-stranded DNA single-walled
carbon nanotube hybrids for optical hydrogen peroxide and glucose sensing. J Phys Chem C
111(24):8638–8643
96. Yang G-Z, Yacoub M (2006) Body sensor networks
97. Yao K, Zhu Y, Yang X, Li C (2008) ENFET glucose biosensor produced with dendrimer
encapsulated Pt nanoparticles. Mater Sci Eng C 28(8):1236–1241
98. Yu R, Pan C, Chen J, Zhu G, Wang ZL (2013) Enhanced performance of a ZnO
nanowire-based self-powered glucose sensor by piezotronic effect. Adv Funct Mater 23
(47):5868–5874
99. Yue HY, Huang S, Chang J, Heo C, Yao F, Adhikari S, Gunes F, Liu LC, Lee TH, Oh ES
(2014) ZnO nanowire arrays on 3D hierachical graphene foam: biomarker detection of
Parkinson’s disease. ACS Nano 8(2):1639–1646
100. Yun Y-H, Eteshola E, Bhattacharya A, Dong Z, Shim J-S, Conforti L, Kim D, Schulz MJ,
Ahn CH, Watts N (2009) Tiny medicine: nanomaterial-based biosensors. Sensors 9
(11):9275–9299
6 Nanomaterials as Implantable Sensors 139

101. Zhan B, Liu C, Chen H, Shi H, Wang L, Chen P, Huang W, Dong X (2014) Free-standing
electrochemical electrode based on Ni (OH) 2/3D graphene foam for nonenzymatic glucose
detection. Nanoscale 6(13):7424–7429
102. Zhang L, Zhou M, Dong S (2012) A self-powered acetaldehyde sensor based on biofuel cell.
Anal Chem 84(23):10345–10349
103. Zhang W, Qiao X, Chen J (2007) Synthesis of silver nanoparticles—effects of concerned
parameters in water/oil microemulsion. Mater Sci Eng B 142(1):1–15
104. Zhang Y, Liu Y, Su L, Zhang Z, Huo D, Hou C, Lei Y (2014) CuO nanowires based
sensitive and selective non-enzymatic glucose detection. Sens Actuators B: Chem 191:86–93
105. Zhang Y, Su L, Manuzzi D, de los Monteros HVE, Jia W, Huo D, Hou C, Lei Y (2012)
Ultrasensitive and selective non-enzymatic glucose detection using copper nanowires.
Biosens Bioelectron 31(1):426–432
106. Zheng G, Patolsky F, Cui Y, Wang WU, Lieber CM (2005) Multiplexed electrical detection
of cancer markers with nanowire sensor arrays. Nat Biotechnol 23(10):1294–1301
107. Zhou M, Zhai Y, Dong S (2009) Electrochemical sensing and biosensing platform based on
chemically reduced graphene oxide. Anal Chem 81(14):5603–5613
108. Zhu Y, Zhu H, Yang X, Xu L, Li C (2007) Sensitive biosensors based on (dendrimer
encapsulated Pt nanoparticles)/enzyme multilayers. Electroanalysis 19(6):698–703
109. Zhu Z, Garcia-Gancedo L, Flewitt AJ, Xie H, Moussy F, Milne WI (2012) A critical review
of glucose biosensors based on carbon nanomaterials: carbon nanotubes and graphene.
Sensors 12(5):5996–6022
Chapter 7
Self-assembly Thin Films for Sensing

Celina Massumi Miyazaki, Anerise de Barros,

Daniela Branco Tavares Mascagni, Juliana Santos Graça,
Paula Pereira Campos and Marystela Ferreira

7.1 Introduction

The need for increasingly sensitive devices with faster and cheaper detection for
various analytes has caused considerable efforts in the last years. Identification and
quantification of several chemicals is of utmost importance for environmental,
medical, biological, and clinical applications. The challenge is developing devices
capable of detecting very low concentrations, as environmental contaminants in ppb
level or even cancer biomarkes at pmol L−1, for example, always considering
parameters such as selectivity and manufacturing cost. Searching for signal
amplification, nanomaterials have received extended attention due to the excellent
catalytic and reactive properties.
Figure 7.1 illustrates typical components in a sensor. Different types of analytes
interact with a receptor (or bioreceptor) element immobilized on a substrate, which
is responsible for the electrical interface with an analytical technique. The assembly
formed by receptor + substrate + analytical technique results in the transducer of
the system. The recognition element can be composed just by nanomaterials or with
an assembly of nano- and biomaterials. Last but not least, an electronic system is
necessary to collect, amplify, and display the measurements.

C.M. Miyazaki
J.S. Graça
M. Ferreira (&)

Universidade Federal de São Carlos, UFSCar, CCTS, Sorocaba, SP, Brazil
e-mail: [email protected]
A. de Barros
Universidade Estadual de Campinas, UNICAMP,
Instituto de Química – IQ, Campinas, SP, Brazil
D.B.T. Mascagni
P.P. Campos
Universidade Estadual Paulista, POSMAT, UNESP, Sorocaba, SP, Brazil

© Springer International Publishing AG 2017 141

T.R.L.C. Paixão and S.M. Reddy (eds.), Materials for Chemical Sensing,
DOI 10.1007/978-3-319-47835-7_7
142 C.M. Miyazaki et al.

ELECTRONIC
SAMPLES TRANSDUCERS SYSTEM

Air, Soil, Water, Carbon nanotube

Antibody
Vegetation
Amplifier

DNAs
Graphene

Food
Enzymes
Electrodes
Signal processor

Blood, Urine,
Salive Nanoparticles FET Devices

Cells Optical Measured signal

Electric Magnetic

Fig. 7.1 Scheme representing components in a typical sensor

In this chapter, we describe the recent advances in the development of

self-assembled films for sensing, emphasizing the use of some important nano-
materials such as metal nanoparticles, carbon nanotubes, and graphene nanosheets.
We have focused on how the nanostructuring process by Langmuir–Blodgett and
layer-by-layer techniques contributes to the high-level organization and proper
immobilization of materials aiming at enhanced analytical performance in the
sensing devices.

7.2 Langmuir–Blodgett and Layer-by-Layer Films

The Langmuir–Blodgett (LB) technique is an important tool to high organized

nanostructured film fabrication. It is based in spreading materials on the water
surface followed by compression until the formation of the Langmuir monolayer on
the air–water interface. This thin film formed with high organization can be
transferred to a solid substrate as a Langmuir–Blodgett (LB) film. The different
stages of the film organization can be monitored by a surface pressure sensor. This
technique is well known, and further details can be found in [1, 2]. Initially, this
technique was limited only to amphiphilic molecules. However, it was expanded to
7 Self-assembly Thin Films for Sensing 143

several materials to find new technological applications. In the literature, a range of

applications of LB films with different materials were found, such as metallic
nanoparticles, conducting polymers, carbon nanotubes, graphene, enzymes, lipids,
proteins, and others [3–5].
Different approaches can be followed for LB film growth with mixed materials.
In a traditional methodology, an amphiphilic material solubilized in a volatile
solvent (usually chloroform) is spread in the surface, while aqueous solubilized
materials/biomaterials are mixed in the subphase [2, 3, 6]. Figure 7.2 shows an
idealized scheme of a Langmuir trough after spreading of nanoparticles and carbon
nanotubes, forming a single layer on the air/water interface after compression of the
trough barriers followed by enzyme adsorption from the aqueous subphase.
Sometimes, the biomolecules are spread onto [7, 8] or are injected under [9, 10] the
previously compressed monolayer.
Important advantages of LB methodology are associated with the high control
fabrication and deposition of monolayers; for example, the molecular ordering can
be monitored by Langmuir pressure isotherm. On the other hand, the deposition
film is evaluated by the transfer ratio (TR) of monolayers onto the substrate. More
details about Langmuir pressure isotherm and TR deposition were described by

Fig. 7.2 Idealized scheme illustrating the construction of a LB film composed by nanoparticles,
carbon nanotubes, and enzyme
144 C.M. Miyazaki et al.

petty [1] and Ulman [2]. Other current methodology for the deposition of Langmuir
monolayers was introduced by Irving Langmuir and Vicent Schaefer [2], where the
substrate is immersed in the horizontal position. This methodology is commonly
applied to rigid monolayers [11], and it is called the Langmuir–Schaefer
(LS) method. One disadvantage for the LS method is associated with low control of
deposition [2]. Figure 7.3 shows the LS deposition method exemplifying the for-
mation of an idealized structure of rigid monolayers.
The possibility of high organization by LB technique is very attractive, but it
requires special equipment for film fabrication [1, 2].
A very simple and versatile approach for nanostructured thin films was proposed
by Decher in 1992 [12]. The so-called Layer-by-Layer (LBL) technique produces
multilayered films formed by sequential and spontaneous adsorption of materials that
interact themselves through Columbic interactions, van der Waals forces, and
hydrogen bonds [13]. It allows the easy control of thickness in molecular level
through experimental parameters, and the versatile way to functionalize surfaces with
unlimited size/shape makes the LbL assembly interesting to obtain new materials
with synergistic combinations of properties. Even firstly proposed for charged

Fig. 7.3 Idealized scheme

illustrating the deposition of
LS film composed by rigid
monolayer
7 Self-assembly Thin Films for Sensing 145

polyelectrolytes, a range of materials can be applied via LbL technique, such as metal
nanoparticles, carbon-based materials, enzymes, antibodies, DNAs, and many oth-
ers. LbL nanostructured films present good stability and unique properties enabling
the wide use in the development of several electronic and optical devices, such as
energy generation and storage devices [14–16], sensors, and biosensors [17, 18].
The traditional immersive LbL assembly is based on the spontaneous adsorption
of the material on a solid substrate submerged in a reservoir containing a stable
aqueous suspension (see Fig. 7.4). After the first immersion, the substrate is washed
and dried before the next immersion (to remove the excess of material weakly
adsorbed) to avoid cross-contamination. Therefore, thickness can be easily con-
trolled by the number of layers deposited. The possibility of controlled conditions,
such as pH and temperature, and the entrapment of water molecules between the
layers [19] are attractive when proteins and other biological molecules are needed to
be immobilized on solid substrates for biosensor applications.
Currently, other LbL categories have been used, as spraying and fluidic
assembly (see Fig. 7.4). In the spraying methodology, materials suspensions are
aerosolized and sprayed onto the surface. Standard spray deposition is faster (about
6 s per layer) than immersive approach, it allows industrial-scale application and
additionally, it is not limited to the planar substrates [20]. Other important con-
siderations about spraying LbL are the possibility of film assembly without the need
of a binder, thereby preventing the properties of interest being changed by the
unnecessary interlaying of binder [21]. Different parameters can be controlled for
the optimization of spray assembly of the film, such as sweep speed, target to
nozzle distance, air pressure, and nozzle flow rate [21]. In the fluidic approach,
nanostructured films can be deposited in the fluidic channels through movement of
the liquids (material suspension and washing solutions) caused by pressure or
vacuum force. Parameters such as the concentration of material, time of contact,
and flow rate will influence film characteristics and properties [20]. Fluidic
assembly is easily integrated, leads to a fast response, and allows control of the
coverage of specific regions of interest, saving reagents and avoiding waste [22].

Immersive Spraying Fluidic

Fig. 7.4 Different Layer-by-Layer assembly categories: immersive, spraying, and fluidic
assembly
146 C.M. Miyazaki et al.

7.3 Advanced Materials Assembled in LB and LbL Films

Nanostructured materials have received special attention in the last years due to
their unique physical and chemical properties compared to bulk material. Several
nanostructures have been used for ultrathin film fabrication, even through LB or
LbL techniques, focusing on the possibility of enhancement of analytical perfor-
mance in sensor applications. Here, we discuss some of the most used materials due
to its undeniably interesting properties, which purport to the easy control of
experimental parameters by LbL technique and highly controlled organization by
LB technique, leading to significant improvements in signal response. In the fol-
lowing sections, we will briefly discuss advanced materials such as metal
nanoparticles, carbon nanotubes, and graphene sheets of composite LB and LbL
films.

7.3.1 Metal Nanoparticle-Based Materials

Metal nanoparticles have fascinated scientists in several areas of research. The main
focus of interest for this material is its huge potential for nanotechnology, opening a
wide range of potential applications in optical sensors [23, 24], electrochemical
sensors [25–27], biosensors [28, 29], electrical sensors [30], fuel and solar cells
[31–33], and others devices [34–36]. Nanoparticles have diameters below 100 nm
[37], with a large surface area to volume ratio and quantum size effects [38, 39].
The optical properties of some metal nanoparticles (such as Au and Ag) also depend
on their size and shape, which are related to the excitation of surface plasmons [40].
This special characteristic is the basis of the surface plasmon resonance (SPR), and
it is also important in surface-enhanced spectroscopies. Focusing on electrochem-
ical sensor application, the combination of large amounts of atoms with large
surface area enables the attachment of a greater amount of molecules on the
nanoparticles surface, allowing them to be conjugated with ligands, antibodies,
vesicles, and other molecules [34–36], thus creating the possibility of increasing the
loading of electroactive species and thus an increase in the catalysis of electro-
chemical processes.
To exemplify the application of nanoparticles to explore the high surface area for
biomolecule immobilization and consequent signal amplification, Samanman et al.
[41] used the LbL technique to deposit alternating AuNPs with thiourea (TU) onto
poly-tyramine (Pty) electropolymerized gold electrode ({AuNPsn/Tu(n−1)}/Pty/Ge).
Different numbers of layers of AuNPs (1–5 layers) were tested for further immo-
bilization of anti-human serum albumin (anti-HSA) detection. Low limit of
detection and high sensitivity were achieved for the electrode composed by 2 layers
of AuNPs, which was related to the greater surface coverage and therefore the
highest percentage of immobilization of anti-HSA (90.2 % ± 0.5).
7 Self-assembly Thin Films for Sensing 147

High sensitivity and low limit of detection were achieved by the combinations of
AuNPs with other materials for Escherichia coli detection [42]. The electrochem-
ical immunosensor for E. coli O157:H7 (E. coli O157:H7) was prepared as a
composite (CHIT-MWNTs-SiO2 @ THI) containing chitosan–multiwalled nan-
otubes (MWNTs) and nanoparticles of SiO2/thionine (THI).
11-amino-1-undecanethiol hydrochloride (AUT)-modified gold electrode was used
as substrate for LbL deposition producing multilayered nanocomposite of
(CHIT-MWNTs-SiO2 @ THI)/AuNPs. Finally, the anti-E. coli O157:H7 was
immobilized on AuNPs film layer via cross-linking using glutaraldehyde. AuNPs
help in increasing the amount of antibody immobilized on the electrode surface,
enabling greater sensitivity. The bacterium E. coli O157:H7 was detected by cyclic
voltammetry in milk samples, and the immunosensor presented a linear range of
4.12  102–4.12  105 colony-forming units (CFU)/ml and limit of detection of
250 CFU/ml of E. coli O157:H7. The total assay time was less than 45 min.
Ou et al. fabricated an electrochemical label-free immunosensor to detect car-
cinoembryonic antigen (CEA), a marker of colorectal cancer tumors. A solution
(AuNPs-MWNTs-THI-CHIT) composite of AuNPs, multiwalled carbon nanotube
with thionine (MWNTs-THI) and chitosan (CHIT) was prepared and assembled by
LbL technique in alternated layers with PSS on 3-mercaptopropanesulfonic (MPS)-
modified gold electrode [43]. Subsequently, anti-CEA antibodies were immobilized
via covalent bonding. The nanoparticles in the self-assembled multilayers
(anti-CEA/(AuNPs-MWNTs-THI-Chit)8/MPS-Au-electrode) led to an increase in
the electron transfer. Therefore, the detection of CEA was based on the variation of
current before and after the antigen binding on the immobilized antibody. The
formation of the antigen–antibody complex on the electrode surface inhibits the
electron transfer, causing the decrease in electrochemical signal with the increase of
concentration CEA on the surface. They also studied the same system in the
absence of AuNPs, and the results indicated sensitivity of 3.87 and
0.48 lA mL ng−1, respectively, for systems with and without AuNPs [43].
Sophisticated systems as molecular imprinting can also be applied using the LbL
technique, which enables sensors with high specificity. Regarding environmental
monitoring applications, Xu et al. [44] developed an LbL-based sensor for
p-nitrophenol (p-NPh), an organic pollutant with high environmental impact in ppb
levels. They combined the LbL and molecular imprinting technique for a macro-
porous molecular imprint polymer (MMIP) electrode. Figure 7.5 shows the
preparation procedure of the electrode used for electrochemical detection: (a) Au
substrate was immersed in the silica microspheres functionalized with thiol groups;
(b) the modified electrode was immersed in the AuNP colloid solution (steps a and
b were performed 5 times); (c) the assembled electrode was immersed in a mixture
containing a mixture of pyrrole and p-NPh. After an electrochemical polymeriza-
tion, a polymer modified assembled electrode was obtained; (d) the electrode was
treated with hydrofluoric acid (HF) solution to etch off silica microspheres com-
pletely; (e) embedded p-NPh was extracted through an electrochemical procedure
creating specific templates for p-NPh detection. The differential pulse voltammetry
148 C.M. Miyazaki et al.
 7 Self-assembly Thin Films for Sensing 149

b Fig. 7.5 A Preparation procedure of a molecularly imprint polymer sensor for p-NPh detection
using the LbL technique. B DPV responses to different concentrations of p-NPh in PBS solution.
a–n Indicates 3.0  10−7–1.4  10−3 mol L−1. Inset calibration curve showing the linear relation
between peak current and p-NPh concentration. Reproduced with the permission from [44]
(adapted)

Fig. 7.6 Voltammetric response of bare ITO and LB film composed by AuNPs functionalized
with n-dodecanethiol electrodes to a caffeic acid and b gallic acid. Reproduced with the permission
from [27] (adapted)

(DPV) showed a linear range between 0.1 µmol L−1 and 1.4 mmol L−1 and limit of
detection of 0.1 µmol L−1 (S/N = 3) [44].
Gold nanoparticles (AuNPs) functionalized with n-dodecanethiol were obtained
by the LB technique and investigated as a voltammetric sensor for organic and
phenolic compounds mainly used in the wine industry. The LB films of function-
alized AuNPs were capable of detecting the main organic acids present in grape and
wines, when compared to the bare ITO electrode, with a shift in the reduction
potential at a less positive potential. The increase in the catalytic properties allowed
discrimination of different dissociated protons of polyprotic acids. The drastic
increase in the sensitivity toward organic acids and phenolic acids, reaching limits
of detection in the order of lmol L−1, was attributed to the presence of AuNPs [27].
Figure 7.6 shows a schematic deposition of LB films composed of AuNPs
150 C.M. Miyazaki et al.

functionalized with n-dodecanethiol and their comparison with bare ITO by

voltammetric analysis in the presence of different phenols compounds.
Significant improvements in its properties are achieved when these nanostruc-
tures are combined with methods of high control organization, which makes the LB
method quite interesting as a nanomaterial immobilization technique. For example,
Charllie et al. produced LB films for H2O2 detection. Electrocatalysis of H2O2
oxidation is one of the most important reactions in the biosensor area because it is
an intermediate product generated by enzymatic reactions between oxidase-type
enzyme and a specific analyte. They observed that LB films composed by poly
(methylmetacrilatacid)-grafted platinum nanoparticles (PtNPs) are very good can-
didates to be applied as high-sensitivity enzymatic biosensors. The authors verified
that the H2O2 oxidation current increases when the PMMA-PtNPs surface density
decreases. The catalysis of H2O2 may be significantly enhanced by the
PMMA-PtNPs nanostructure due to the higher accessibility of the active site for the
nanoparticles located at the edges of the domains in the nanostructures [45].
The direct electron transfer process in enzymes avoids the use of electron mediators
in biosensor applications. The high surface, biocompatibility, and conductivity of
gold nanoparticles have attracted researchers due to the adsorption of redox enzymes
without loss of bioactivity [46–48], giving the protein more freedom of orientation,
which reduces the insulating property of the protein shell, thus facilitating the electron
transfer through the conducting tunnels formed by gold nanoparticles [46]. Matharu
et al. reported high catalytic activity and sensitivity of an electron mediator-free
enzymatic sensor composed of modified electrodes with gold nanoparticles (AuNPs)
functionalized with 11-mercaptoundecanoic (11-MUA) and octadecylamine (ODA).
The enzyme sensor was fabricated by covalent immobilization of cholesterol oxidase
(ChOx) onto MUA-AuNPs/ODA from the LB technique. The cyclic voltammetric
and electrochemical impedance studies reveal that MUA-AuNP/ODA LB film has
good affinity for ChOx and provides favorable microenvironment for direct electron
transfer between enzyme and electrode. The detection limit for this sensor was
23.38 mg dL−1, and sensitivity was 1.085 lA mmol L−1 [28].
The influence of AuNPs was investigated by Barros et al. for the LB films com-
posed of polyaniline (PAni) and organophilic montmorillonite clay (OMt). Different
architectures to the films, such as PAni/AuNPs, OMt/Au, and OMt/AuNPs/PAni,
were obtained and investigated as electrochemical sensor for the detection of metal
ions such as cadmium (Cd2+), lead (Pb2+), and copper (Cu2+). Due to the presence of
AuNPs, high catalytic activity, better distinction between characteristic peaks of each
metal ion, and increase in the sensitivity toward Cd2+, Pb2+, and Cu2+ were observed,
with limits of detection in the order of lg L−1. In a direct comparison of the per-
formance of the three different architectures as sensors, the OMt/AuNPs/PAni-es LB
film presented superior results. The limit of detection was of 0.0049, 0.0059, and
0.089 lg L−1 for the Cd2+, Pb2+, and Cu2+, respectively [49, 50].
Other examples of use of metal nanoparticles can be cited, such as arrays of
magnetic nanoparticles, which are promising structures to new devices for data
storage and spintronic device, reported by Dochter et al. [24]. The authors used the
iron oxide nanoparticle assembled by LB technique and other methodologies
7 Self-assembly Thin Films for Sensing 151

demonstrating the final properties depending essentially on the spatial arrangement of

nanoparticles [24]. A dense film of AuNPs was fabricated by LB assembly and
positioned between the ITO electrodes and the anode modification with PEDOT:PSS
for solar cell application. The enhanced performance in polymer-based solar cell by
the incorporation of AuNPs is attributed to the localized surface plasmon resonance
(SPR). The principal conditions that govern this SPR effect have been optimized by
selecting different sizes of AuNPs and controlling the LB assembly [31]. The hybrid
films composed of polythiophenes and AuNPs fabricated by LB technique are quite
promising for sensing photovoltaic and memory devices. The incorporation of AuNPs
into the polythiophenes matrix favors higher conductivity for this nanostructure,
which is strongly dependent on the molecular arrangement of the matrix [51].

7.3.2 Graphene Derivative-Based Materials

Graphene is composed of a two-dimensional network of sp2-hybridized carbon

atoms in a honeycomb lattice [52]. One s-orbital and two in-plane p-orbitals (px and
py) in each atom hybridize to form strong covalent sp2 bonds (in a 120° C–C–C
angle), which are responsible for the mechanical stability of the graphene sheet. The
remaining pz orbital (perpendicularly oriented to the plane) on each carbon atom
overlaps with other pz orbitals on its neighbor carbons to form the valence (p) and
conduction (p*) bands, which dominate the planar conduction [52, 53], one of the
most interesting and highly explored properties of graphene. The oxidized form of
graphene—the graphene oxide (GO)—obtained by the oxidation of graphite fol-
lowed by an exfoliation process has also received attention concerning sensor and
biosensor development. Even though GO has insulating properties, its high density
of oxygenated functional groups on the basal plane allows it to interact with a wide
range of organic and inorganic materials [54], including biological molecules such
as enzymes [55, 56], antibodies [57], and DNAs [58], among others. Focusing on
biosensor applications, the ionic groups and aromatic domains in GO allow to
interact with biomolecules in different ways, such as by electrostatic interaction
between oxygen groups of GO and charged amine groups of proteins and DNAs, or
by hydrophobic domains of GO providing p–p stacking and quenching with aro-
matic biomolecules or dyes [54]. Additionally, since it is relatively easy to obtain
and has inexpensive synthesis, and also because of its good solubility and stability
in aqueous medium, GO has been used as a precursor for reduced and conductive
form, thus being named reduced graphene oxide (RGO). Chemical, thermal, and
electrochemical methods are used to obtain the reduced form. This reduction can be
performed before the film assembly or after immobilization in a substrate.
Based on electronic transfer in graphene sheets, a nonenzymatic detection of
H2O2 with high sensitivity in a wide linear range was acquired by Liu et al. [59].
LbL film composite of poly (diallyldimethylammonium chloride) (PDDA) func-
tionalized graphene (positively charged suspension) with negatively charged Fe3O4
nanoparticles fabricated onto ITO substrate demonstrated efficient electrocatalytic
152 C.M. Miyazaki et al.

response toward H2O2, even in an air saturated environment. The amperometric

sensor response was in a linear range from 20 lmol L−1 to 6.25 m mol L−1; a low
detection limit (2.5 l mol L−1) with good stability was attributed to the synergistic
effect of the catalytic ability of Fe3O4 nanoparticles and fast electron transfer of
graphene sheets, according to the authors [59].
Graphene sheets also present high surface area, guaranteeing high catalytical
properties. Also, this has been used as substrate to anchor nanoparticles and bio-
molecules toward biosensor applications. Cao et al. [60] produced a graphene–gold
nanoparticle (GR-AuNP) composite as a matrix for glucose oxidase immobilization
toward an electrochemical biosensor for glucose. For the construction of GR-AuNP
composite biosensor, glass carbon electrode (GCE) was coated with alternated
layers of GR-AuNPs and GOx, as shown in Fig. 7.7a. The amperometric responses

Fig. 7.7 a Schematic diagram of LbL assembly of AuNPs-graphene and GOx. b Amperometric
response of 5 bilayers (AuNPs-graphene/GOx) on GCE at 0.6 V upon successive additions of
glucose in PBS (pH 7.0), inset calibration curve. c Lineweaver–Burk plot for 5 bilayers
(GOx/AuNPs-graphene). Reproduced with the permission from [60] (adapted)
7 Self-assembly Thin Films for Sensing 153

at 0.6 V is shown in Fig. 7.7b with the calibration curve in the inset. Using this
electrode, the apparent Michaelis–Menten value was 0.038 mmol L−1 (obtained by
Lineweaver–Burk plot shown in Fig. 7.7c), which is lower than values found in the
literature, indicating higher affinity and enzymatic activity for glucose. Because of
the electronic properties and the maintenance of the enzyme bioactivity and the
direct electron transfer ability, this glucose sensor exhibited a detection limit of
4.1 µmol L−1 and sensitivity of 3.84 lA mmol L−1 cm−2.
The high catalytic property of graphene has been explored for the detection of
very low concentrations of analytes. Dopamine (DA) is an important neurotrans-
mitter, and it is found in very low concentrations (about 0.1 l mol L−1) in the
extracellular fluid of the central nervous system [61]. Electrochemically, its
detection is very difficult due to the low concentration and the presence of other
interferents (such as uric acid and ascorbic acid), which coexist in the extracellular
fluids of the central nervous system [62] and exhibit similar oxidation potentials.
Weng et al. developed a graphene/chitosan LbL film onto polished GCE electrode.
Five bilayers of negatively charged graphene dispersed in DMF and positively
charged chitosan suspension were deposited and tested by DPV for simultaneous
detection of DA and UA. The results showed a linear range between 0.1 and
140 l mol L−1 and LOD of 0.05 l mol L−1 for DA detection and linear range
between 1.0 and 125 l mol L−1 and LOD of 0.1 l mol L−1 for UA. The authors
attributed the high sensitivity and reproducibility of the sensors to two factors:
(i) unique electrocatalytic properties of graphene and (ii) LbL assembly of
graphene/chitosan, giving good thermal and mechanical stability [61]. The synergy
of graphene properties and metal nanoparticles has been investigated by authors for
the achievement of sensors with high catalytic properties. Liu et al. fabricated a LbL
film onto a polished GCE electrode using a first layer of PDDA and varying
depositions of PSS-functionalized reduced graphene oxide (negatively charged) and
PAMAM stabilized AuNPs (positively charged) solutions. Twenty bilayers were
assembled and used for electrochemical determination via differential pulse
voltammetry (DPV). For electrochemical detection, the oxidation potential of uric
acid, ascorbic acid, and DA is very close, making necessary the study of materials
and methodologies that produces responses with high sensitivity and selectivity for
DA. The synergistic effect of AuNPs and reduced graphene oxide contributed to a
significant electrocatalytic activity, allowing simultaneous determination of DA and
UA. A linear response of DA concentration from 1 to 60 lmol L−1, a sensitivity of
0.3857 lA lmol L−1, and a LOD of 0.02 lmol L−1 (S/N = 3) were achieved.
Meanwhile, UA was detected in the range of 10–120 lmol L−1, with a sensitivity
of 0.0317 lA lmol L−1 and with a LOD of 0.27 lmol L−1 [62].
Immunosensors have also received attention in research on graphene-based
sensors. Before antibody immobilization, a primary layer of electron transfer cap-
able structures was deposited onto the electrode [63]. A sandwich-type electro-
chemical immunosensor was developed using IgG as a model of ligand for hIgG
detection of human serum samples. The authors used the LbL technique to modify
the surface of the glassy carbon electrode with MWCNTs and reduced graphene
oxide (RGO), whose adsorption of nanomaterials was given by electrostatic
154 C.M. Miyazaki et al.

interactions between the positive charged poly (diallyldimethylammonium chlo-

ride) (PDDA) and the negative charge of MWCNT and RGO. Subsequently, the
primary IgG antibody (Ab1) was covalently immobilized on the electrode surface
and subsequently linked to the hIgG antigen (Ag) and the secondary IgG-HRP
antibody (Ab2-HRP), containing peroxidase. The electrochemical immunoassay
was performed in PBS containing hydroquinone and H2O2. The presence of the
composite on the electrode surface enabled to explore both the properties of 1D
structure of MWCNT and the 2D structure of graphene, demonstrating an
improvement in the electron transfer at the electrode interface. In addition, the
presence of the composite increased the electrode surface area favouring the anti-
bodies immobilization. The immunosensor (Ab2-HRP/Ag/Ab1/GR–MWCT/GC)
has great potential for clinical applications since it showed excellent selectivity,
stability, reproducibility, and low detection limit (0.2 ng mL−1).
The possibility of organizing and controlling the microstructure of graphene
nanosheets is very interesting because it affects its properties directly. The propa-
gation of charge in graphene is 103 times higher in the in-plane direction compared
to the out-of-plane direction [53] making the sheets’ organization important to
control the conductivity of the formed nanostructured film. Single layer and mul-
tilayer of graphene can be obtained by chemical vapor deposition (CVD); however,
their high cost of production and expensive equipments limit large-scale produc-
tion. In this way, the LB technique can allow high organization of the nanosheets
and also control the density of the package of the monolayer transferred to the
substrate [64, 65]. The influence of the chemical structure and deposition
methodology for the graphene oxide films was investigated by Hidalgo et al. [11].
They verified that the LB methodology renders the highest solid coverage and no
significant dependence on chemical structure of graphene sheets. On the other hand,
the coverage found for the films produced by the Langmuir–Schaefer
(LS) methodology increases when the amount of C–O groups attached to the basal
plane of sheets is higher. The XPS results proved the possibility to modulate the
graphene oxide sheet coverage and deposition methodology. This strategy presents
an easy way to obtain reproducible graphene oxide films of different morphology
and coverage [11].
Graphene oxide (GO) was deposited on a substrate through the LB method,
which provided a tunable and ordered GO arrangement. Subsequently, the GO LB
films were reduced to RGO following thermal treatment. The conducting polymer
(PEDOT) was directly coated on RGO from a vapor phase polymerization process.
The electrochemical activity results revealed that the RGO/PEDOT LB films
exhibit 213 F/g high specific capacitance at a 0.5 A/g current density and show
better capacitance retention rate than pure PEDOT. More detailed studies indicated
that the arrangement of RGO obtained by LB technique shows distinct influence on
the electrical and electrochemical properties. Due to the high conductivity and
electrochemical activity, this LB film nanocomposite is seen as quite promising as
an organic and flexible electrode for sustainable energy storage [66]. Jia and Zou
successfully assembled LB films of sulphonated graphene nanosheets with the
assistance of ethanol. The electrochemical properties present to multilayered
7 Self-assembly Thin Films for Sensing 155

Fig. 7.8 Schematic illustration of field effect transistor device based on PtNPs/RGO LB film and
real-time recording of the hybridization between target DNA and probe DNA immobilized on
PtNPs/RGO channels in PBS buffer, Vds = 400 mV (black line), and the control experiment result
with addition of noncomplementary ssDNA (red line). Reproduced with the permission from [68]
(adapted)

sulphonated graphene are promising potential for applications as electrodes for

supercapacitors and capacitive deionization of saline water. The sulphonated gra-
phene films showed excellent cyclic stability in a long-term charge/discharge test,
and even after 50 cycles, the capacitance value decreases only 0.93 % exhibiting a
very high degree of reversibility over all cycles [65].
The LB technique also allows the construction of hybrid materials containing
nanoparticles and graphene nanostructures promoting the synergy of the individual
properties and consequently enhancing catalytic power. Kumar et al. described
graphene oxide (GO) sheets decorated with AuNPs obtained by the LB technique
by electrostatic interactions. The GO-AuNP composite favors visible light-driven
plasmonic photocatalysis with enhanced charge separation and transport properties.
The results demonstrated the augmentation of the photocatalytic activity and con-
ductivity with high transmittance offering an alternative to replace the ITO substrate
in organic solar cells [67]. Yin et al. fabricated few-layer reduced graphene oxide
(RGO) on a Si/SiO2 wafer from the LB method followed by thermal reduction.
The RGO reduction was made in the presence of platinum nanoparticles (PtNPs),
and PtNPs/RGO composite was obtained. These LB film composite is employed as
the conductive channel in a solution-gated field effect transistor (FET). The tran-
sistors were used for real-time detection of hybridization of single-stranded DNA
with high sensitivity of 2.4 nmol L−1 indicating a great potential for the production
of graphene-based electronic biosensors [68]. Figure 7.8 shows the scheme of the
transistor composed of PtNPs/RGO LB films and their response at different con-
centrations of DNA.

7.3.3 Carbon Nanotube-Based Materials

Carbon nanotubes (CNTs) are graphene sheets rolled in cylindrical form, composed
by carbon atoms with sp2 bounds, arranged in a hexagonal network in two
156 C.M. Miyazaki et al.

dimensions. CNTs present tubular structure with diameter close to 1 nm and length
in micrometers. Two structural categories are found: single-walled carbon nan-
otubes (SWCNTs) and multi-walled carbon nanotubes (MWCNTs) [69–71]. The
sheets can be wrapped in three different geometries, and the integer indices (n,
m) represent the form in which the graphene sheet is wrapped. n and m represent the
vector in two directions in the crystal lattice of graphene. For m = 0, the CNT is
called zigzag; for n = m, the CNT is armchair, and in other cases, the CNT is called
chiral [72]. The bandgap is different for the types of CNTs, and the electrical
conductivity can show semiconducting or metallic behavior [72, 73]. The structure
and chemistry of CNTs are widely discussed and can be found in Refs. [71, 73].
The electronic transport on CNTs is ballistic (without losses by dispersion) due to
unidimensional structure, allowing the transport of high current at room temperature
[73]. Research in sensor development has achieved high sensitivity, fast response,
and good reversibility applying CNTs [74].
Based on the increased electron transfer in electrochemical reactions and easy
immobilization of biomolecules [74], a glucose sensor was developed by Wu et al.
[75] through LbL assembly of MWCNTs, AuNPs, and glucose oxidase (GOx) onto
Pt electrode, forming a GOx/AuNP/MWCNT/Pt electrode. This composite elec-
trode presented high catalytic activity for H2O2 detection by the enzymatic oxi-
dation of glucose. Due to the good electron transfer ability of MWCNT and AuNP,
the sensor response occurred in low potential in amperometric measurements with
high sensitivity (2.5 lA/mmol L−1), low detection limit (6.7 lM), and fast
response (7 s).
Because of its large surface area, CNTs have also been applied on gas sensors,
enabling the adsorption of large amount of molecules. Ping et al. developed an
ammonia gas sensor, by deposition of carboxyl (COOH)-modified SWCNT-COOH
dispersed on poly (sodium-p-styrenesulfonate) (PSS) in LbL films alternately with
poly (diallyldimethylammonium chloride) PDDA. They studied the gas sensor
sensitivity in quartz crystal microbalance (QCM) based on the fact that when the gas
reaches the QCM sensor, the adsorbed molecules on the film cause a mass variation.
Therefore, carbon nanotubes contribute to higher amount of gas molecules being
adsorbed on the film, which increases the sensitivity of the sensor. The SWCNT-
based mass-sensitive QCM gas sensor showed fast response to ammonia gas and
good sensitivity [76]. Similar studies were performed by Jing et al. using LbL films
of PDDA and SWCNT on QCM toward humidity sensor applications. They com-
pared the efficiency of two types of LbL films, the SWNT(PDDA/SWNT)n and
carboxyl (COOH)-modified SWCNT-COOH(PDDA/SWCNT-COOH)n. The sec-
ond film showed best sensitivity because of the water molecules adsorbed on main
groups on the surface of functionalized nanotubes [77].
Based on the electrocatalytic activity, high conductivity, and large surface area,
Sun et al. [78] developed a strategy to LbL film preparation by alternating layers of
carboxylated multi-walled carbon nanotubes (CMWCNTs) and amino multi-walled
carbon nanotubes (AMWCNTs) to produce covalent amide bond formed film, as
shown in Fig. 7.9a. The multilayered carbon nanotubes film was assembled on
PAA-modified GCE, varying the number of bilayers from 2 to 4. The
7 Self-assembly Thin Films for Sensing 157

(a)

(b) (c) 25
20
15

I / µA
10
200 µM
5
0
0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 150 µM
CNADH / mM
100 µM

70 µM
50 µM
10 µM 30 µM

Fig. 7.9 a Schematic illustration of CMWCNT and AMWCNT multilayered film assembly.
b Cyclic voltammograms of (I) bare GCE in PBS without NADH; (II), (III), (IV),
(V) PAA-modified GCE with 2, 3, and 4 bilayers of CMWCNT/AMWCNT, respectively, in
PBS containing 1 mM NADH. c Amperometric response for 4-bilayer CMWCNT/AMWCNT at
+0.15 V for successive additions of NADH. Inset Calibration curve for 4-bilayer
CMWCNT/AMWCNT. Reproduced with the permission from [78] (adapted)

electrocatalytic performance for NADH was investigated by cyclic voltammetry

(Fig. 7.9b), indicating a negative shift in the oxidation potential and an increase in
the peak current with the increase in the number of layers of nanotubes. This fact is
related to the higher electroactive surface area and also to the presence of oxy-
genated groups (such as quinones) on the MWCNT surface, which promote NADH
oxidation. As shown in Fig. 7.9c, the amperometric response of the sensor was fast
(less than 3 s) indicating fast charge transport, with high sensitivity
(223.8 lA mmol L−1 cm−2), low limit of detection (1.5 lmol L−1), and good
selectivity and stability.
CNTs can be used as electrical supporting layers and electron transfer mediators
between an electrode and a redox species. When carbon nanotubes are aligned,
different properties can be found, such as higher conductivity when compared to the
uniformed structure, which lends itself to applications in electrochemical sensors and
biological sensors [79–81]. The immobilized CNT layer has high active surface area
for adsorption of electroactive molecules as well as strong stability. In this context,
several approaches to produce thin films of precisely controlled CNTs have been
reported [82, 83]. The LB technique is one of the most attractive to build thin films
controlling the nanostructure and the film thickness. The alignment can be per-
formed during or post-growth. For alignment during carbon nanotube growth,
158 C.M. Miyazaki et al.

different metal catalysts can be used, such as Fe, Ni, Co, and Mo [82]. Takezawa
et al. [81] used Fe nanoparticles (FeNPs) to induce a vertical alignment of carbon
nanotubes. The FeNPs were mixed with the arachidic acid, deposited as LB film and
used as filler molecules for carbon nanotube growth. This allowed to reduce thermal
aggregation of FeNPs through the control of the density of the number of FeNPs.
The catalysis occurred at the water interface during compression of both materials
from the LB technique. These structures have been attracting attention as electrode
materials for polymer electrolyte fuel cell application. In the post-growth alignment,
previously produced carbon nanotubes can be submitted to external forces or fields
to induce the organization [82]. Exemplifying this, Lee et al. dispersed SWCNTs in
the air/water interface and transferred to a solid substrate. After aligned nanotubes
were deposited on the solid substrate, the palladium nanoparticles were electro-
chemically deposited, producing a hybrid palladium nanoparticle-decorated carbon
nanotubes toward hydrogen gas sensing. The sensor demonstrated the specific and
fast detection of hydrogen under a N2 atmosphere, presenting reversibility. The
detection range was about 0.025–2.25 % (v/v) hydrogen in N2, considering that the
lower explosive limit of hydrogen is 4 % in air [84].
Poonia et al. investigated LB films where SWCNTs are parallel and perpen-
dicular aligned to the direction of the applied electric field were fabricated to
produce a methane (CH4) gas sensor. These films were compared with a randomly
oriented SWCNT-composed film in a drop cast film, and the results confirmed that
the aligned SWCNTs show a steplike response due to a change in the concentration
of the CH4 gas molecules. The enhanced capability for sensing CH4 gas using the
LB films is attributed to the aligned SWCNTs, which provides also an alignment of
adsorption sites for the gas molecules [79]. The same sensing performance
improvement was achieved for a voltammetric sensor for methylparaben using a
glassy carbon electrode modified with a LB film composed with MWCNTs per-
pendicularly aligned. The results demonstrated that the perpendicularly deposited
MWCNT exhibits excellent conduction improving the sensing performance of
electrochemical sensors, with a limit of detection of the 0.4 lmol L−1 [80].
Figure 7.10 shows the linear voltammetric response of LB films composed by
MWCNT in the presence of methylparaben.
Caseli et al. investigated the use of carbon nanotubes in hybrid LB films of lipids
and urease to improve the enzyme immobilization and the catalytic performance.
The enzyme was adsorbed from the aqueous subphase on a Langmuir monolayer of
dimyristoyl phosphatidic acid (DMPA) followed by CNT incorporation within a
hybrid film. The analysis by colorimetric method indicated that the presence of
CNTs preserved and enhanced the enzyme activity of the film, even after one
month. These results reveal that the hybrid film is very promising for biosensor
development [9]. Langmuir monolayers of amphiphilic viologens were electro-
statically adsorbed on MWCNT. The surface pressure isotherms indicated good
stability for hybrid monolayers. The hybrid LB films were characterized by cyclic
voltammetry indicating one or two couples of one electron transfer processes
corresponding to the viologen-MWCNT hybrid films. The voltammograms also
showed cathodic and anodic potentials closely related to the alkyl chains of the
7 Self-assembly Thin Films for Sensing 159

Fig. 7.10 a Linear voltammetric response of LB films of MWCNT with different concentrations
of methylparaben in 0.1 mol L−1 PBS (pH = 3.0), from curve a to i: 0, 2  10−6, 4  10−6,
6  10−6, 8  10−6, 2  10−5, 4  10−5, 6  10−5, 8  10−5 mol L−1. b The linear relationship
between ip and Cmethylparaben. Reproduced with the permission from [80] (adapted)

viologens [85]. The effect of MWCNT concentration in mixes with polythiophene

(P3HT) was investigated by Lo et al. [86]. The monolayers were transferred by
Langmuir–Blodgett and Langmuir–Schaefer methodology. The authors verified by
UV–Vis spectroscopy that the MWCNT/P3HT monolayers were successfully
obtained by horizontal deposition (LB method). The current–voltage characteristic
of the LB films showed that current increases linearly with the increase in the
voltage, indicating that the MWCNT/P3HT LB film forms an ohmic contact with
gold and the increase in electric current was mainly contributed by MWCNTs [86].
The LB films composed of pyridylthio and MWCNTs formed stable floating
monolayers at the air/water interface and transferred onto substrate. These LB films
were used as support to immobilize hydrogenase (H2ase) to form bionanocom-
posites of pythio-MWCNT-H2ase. Cyclic voltammograms reveal the high catalytic
activity and strong stability in the LB films of pythio-MWCNT-H2ase, suggesting
that the LB films could be used as heterogeneous biocatalysts to catalyze a
reversible reaction between protons and H2, resulting in potential applications in
biohydrogen evolution and H2 biofuel cells [87].

7.4 Final Remarks

In this chapter, we have discussed the growing importance of nanostructured and

organized films in coating surfaces where molecular level control can be exploited
using the LB and LbL techniques. After a brief discussion about these techniques,
we concentrated on summarizing how to use such films for functionalized surfaces
160 C.M. Miyazaki et al.

using advanced materials, especially carbon nanotubes, graphene nanosheets, and

metal nanoparticles, focusing on electrochemical sensing and biosensing. The
unique properties of these materials allied to the nanostructured organization have
enabled us to apply it in various areas such as food, environmental and clinical
analysis, as discussed in this chapter. Although many other applications have not
been included here, lots of new ones will appear in the future. Thus, this represents
an updated idea of the main applications covering these advanced materials in the
form of nanostructured LB and LbL films demonstrating the huge potential of this
combination.

Acknowledgments The authors acknowledge the financial support of FAPESP, CNPq, CAPES,
and nBioNet network (Brazil).

References

1. Petty MC (1996) Langmuir-Blodgett films: an introduction, 1st edn. Cambridge University

Press, New York
2. Ulman A (1991) An introduction to ultrathin organic films: from Langmuir-Blodgett to
self-assembly, 1st edn. Academic Press, San Diego
3. Ariga K, Yamauchi Y, Mori T, Hill JP (2013) 25th anniversary article: What can be done with
the Langmuir-Blodgett method? Recent developments and its critical role in materials science.
Adv Mater 25:6477–6512
4. Acharya S, Hill JP, Ariga K (2009) Soft Langmuir-Blodgett technique for hard nanomaterials.
Adv Mater 21:2959–2981
5. Mandal S, Lee MV, Hill JP et al (2010) Recent developments in supramolecular approach for
nanocomposites. J Nanosci Nanotechnol 10:21–33
6. de Oliveira RF, Xavier ACF, de Barros A, Ferreira M (2015) Filmes Nanoestruturados:
Técnicas de Langmuir-Blodgett (LB) e Layer-by-Layer (LbL). In: Da Róz AL, Leite FL,
Ferreira M, Oliveira ON Jr (eds) Nanoestruturas Princípios E Apl, 1st edn. Elsevier, Rio de
Janeiro, pp 121–140
7. Kato N, Sasaki T, Mukai Y (2015) Partially induced transition from horizontal to vertical
orientation of helical peptides at the air–water interface and the structure of their monolayers
transferred on the solid substrates. Biochim Biophys Acta BBA—Biomembr 1848:967–975
8. Togashi K, Iizuka T, Kato N et al (2012) Fabrication of Langmuir-Blodgett films using
amphiphilic peptides. J Nanosci Nanotechnol 12:568–572
9. Caseli L, Siqueira JR (2012) High enzymatic activity preservation with carbon nanotubes
incorporated in urease-lipid hybrid Langmuir-Blodgett films. Langmuir 28:5398–5403
10. Goto TE, Lopez RF, Oliveira ON, Caseli L (2010) Enzyme activity of catalase immobilized in
Langmuir-Blodgett films of phospholipids. Langmuir 26:11135–11139
11. Hidalgo RS, López-Díaz D, Velázquez MM (2015) Graphene oxide thin films: influence of
chemical structure and deposition methodology. Langmuir 31:2697–2705
12. Decher G, Hong JD, Schmitt J (1992) Buildup of ultrathin multilayer films by a self-assembly
process: III. Consecutively alternating adsorption of anionic and cationic polyelectrolytes on
charged surfaces. Thin Solid Films 210–211(Part 2):831–835
13. Schönhoff M (2003) Layered polyelectrolyte complexes: physics of formation and molecular
properties. J Phys: Condens Matter 15:R1781–R1808
14. Wang D, Wang X (2011) Self-assembled graphene/azo polyelectrolyte multilayer film and its
application in electrochemical energy storage device. Langmuir 27:2007–2013
7 Self-assembly Thin Films for Sensing 161

15. Almeida TP, Miyazaki CM, Paganin VA et al (2014) PEDOT:PSS self-assembled films to
methanol crossover reduction in Nafion® membranes. Appl Surf Sci 323:7–12
16. Jiang SP, Liu Z, Tian ZQ (2006) Layer-by-layer self-assembly of composite
polyelectrolyte-Nafion membranes for direct methanol fuel cells. Adv Mater 18:1068–1072
17. Riul Jr A, Miyazaki CM, Dantas CAR, Oliveira Jr ON (2013) The use of nanostructured films
in sensing applications. In: Multisens Syst Chem Anal Mater Sens. CRC Press, Boca Raton,
p 408
18. Crespilho FN, Zucolotto V, Oliveira ON Jr, Nart FC (2006) Electrochemistry of
layer-by-layer films: a review. Int J Electrochem Sci 1:194–214
19. Ariga K, Yamauchi Y, Rydzek G et al (2014) Layer-by-layer nanoarchitectonics: invention,
innovation, and evolution. Chem Lett 43:36–68
20. Richardson JJ, Bjornmalm M, Caruso F (2015) Technology-driven layer-by-layer assembly of
nanofilms. Science 348:2491–2491
21. Saetia K, Schnorr JM, Mannarino MM et al (2014) Spray-layer-by-layer carbon
nanotube/electrospun fiber electrodes for flexible chemiresistive sensor applications. Adv
Funct Mater 24:492–502
22. Daikuzono CM, Dantas CAR, Volpati D et al (2015) Microfluidic electronic tongue. Sens
Actuators B Chem 207:1129–1135
23. Lavenn C, Albrieux F, Tuel A, Demessence A (2014) Synthesis, characterization and optical
properties of an amino-functionalized gold thiolate cluster: Au10(SPh-pNH2)10. J Colloid
Interf Sci 418:234–239
24. Dochter A, Pichon BP, Fleutot S et al (2013) A multiscale structural study of nanoparticle
films prepared by the Langmuir-Blodgett technique. Solid State Sci 16:81–89
25. Zou L, Li Y, Cao S, Ye B (2013) Gold nanoparticles/polyaniline Langmuir-Blodgett film
modified glassy carbon electrode as voltammetric sensor for detection of epinephrine and uric
acid. Talanta 117:333–337
26. Medina-Plaza C, Furini LN, Constantino CJL et al (2014) Synergistic electrocatalytic effect of
nanostructured mixed films formed by functionalised gold nanoparticles and bisphthalocya-
nines. Anal Chim Acta 851:95–102
27. Medina-Plaza C, García-Cabezón C, García-Hernández C et al (2015) Analysis of organic
acids and phenols of interest in the wine industry using Langmuir-Blodgett films based on
functionalized nanoparticles. Anal Chim Acta 853:572–578
28. Matharu Z, Pandey P, Pandey MK et al (2009) Functionalized gold nanoparticles—
octadecylamine hybrid Langmuir-Blodgett film for enzyme sensor. Electroanalysis
21:1587–1596
29. Chen S, Yuan R, Chai Y, Hu F (2012) Electrochemical sensing of hydrogen peroxide using
metal nanoparticles: a review. Microchim Acta 180:15–32
30. Raguse B, Chow E, Barton CS, Wieczorek L (2007) Gold nanoparticle chemiresistor sensors:
direct sensing of organics in aqueous electrolyte solution. Anal Chem 79:7333–7339
31. Chen X, Yang X, Fu W et al (2013) Enhanced performance of polymer solar cells with a
monolayer of assembled gold nanoparticle films fabricated by Langmuir-Blodgett technique.
Mater Sci Eng, B 178:53–59
32. Zhavnerko G, Marletta G (2010) Developing Langmuir-Blodgett strategies towards practical
devices. Mater Sci Eng, B 169:43–48
33. Guo W, Pi Y, Song H et al (2012) Layer-by-layer assembled gold nanoparticles modified
anode and its application in microbial fuel cells. Colloids Surf Physicochem Eng Asp
415:105–111
34. Mody VV, Siwale R, Singh A, Mody HR (2010) Introduction to metallic nanoparticles.
J Pharm Bioallied Sci 2:282–289
35. Sardar R, Funston AM, Mulvaney P, Murray RW (2009) Gold nanoparticles: past, present,
and future. Langmuir 25:13840–13851
36. Ghosh P, Han G, De M et al (2008) Gold nanoparticles in delivery applications. Adv Drug
Deliv Rev 60:1307–1315
162 C.M. Miyazaki et al.

37. Ding L, Bond AM, Zhai J, Zhang J (2013) Utilization of nanoparticle labels for signal
amplification in ultrasensitive electrochemical affinity biosensors: a review. Anal Chim Acta
797:1–12
38. Daniel MC, Astruc D (2004) Gold nanoparticles: assembly, supramolecular chemistry,
quantum-size-related properties, and applications toward biology, catalysis, and nanotech-
nology. Chem Rev 104:293–346
39. Cao G (2004) Zero-dimensional nanostructures: nanoparticles. Nanostructures and nanoma-
terials: synthesis, properties and applications. College Press, London, pp 51–109
40. Klabunde KJ (2001) Introduction to nanotechnology. In: Klabunde KJ (ed) Nanoscale
materials in chemistry. John Wiley & Sons Inc, Hoboken, pp 1–13
41. Samanman S, Kanatharana P, Asawatreratanakul P, Thavarungkul P (2012) Characterization
and application of self-assembled layer by layer gold nanoparticles for highly sensitive
label-free capacitive immunosensing. Electrochim Acta 80:202–212
42. Li Y, Cheng P, Gong J et al (2012) Amperometric immunosensor for the detection of
Escherichia coli O157:H7 in food specimens. Anal Biochem 421:227–233. doi:10.1016/j.ab.
2011.10.049
43. Ou C, Yuan R, Chai Y et al (2007) A novel amperometric immunosensor based on
layer-by-layer assembly of gold nanoparticles–multi-walled carbon nanotubes-thionine
multilayer films on polyelectrolyte surface. Anal Chim Acta 603:205–213
44. Xu G, Yang L, Zhong M et al (2013) Selective recognition and electrochemical detection of
p-nitrophenol based on a macroporous imprinted polymer containing gold nanoparticles.
Microchim Acta 180:1461–1469
45. Challier L, Gal F, Carrot G et al (2013) Hybrid platinum nanoparticle ensemble for the
electrocatalytic oxidation of H2O2: toward nanostructured biosensor design. Electrochem
Commun 28:118–121
46. Liu S, Ju H (2003) Electrocatalysis via direct electrochemistry of myoglobin immobilized on
colloidal gold nanoparticles. Electroanalysis 15:1488–1493
47. Zhao S, Zhang K, Bai Y et al (2006) Glucose oxidase/colloidal gold nanoparticles
immobilized in Nafion film on glassy carbon electrode: direct electron transfer and
electrocatalysis. Bioelectrochemistry 69:158–163
48. Yi X, Huang-Xian J, Hong-Yuan C (2000) Direct electrochemistry of horseradish peroxidase
immobilized on a colloid/cysteamine-modified gold electrode. Anal Biochem 278:22–28
49. de Barros A, Ferreira M, Constantino CJL et al (2015) Synergy between polyaniline and OMt
clay mineral in Langmuir-Blodgett films for the simultaneous detection of traces of metal
ions. ACS Appl Mater Interfaces 7:6828–6834
50. Barros A (2015) Filmes de Nanocompósitos de Polímero Condutor, Nanopartículas de Argila
e Nanopartículas de Ouro para Aplicação em Sensores Ambientais. Tese de doutorado,
Universidade Estadual Paulista
51. Jayaraman S, Yu LT, Srinivasan MP (2013) Polythiophene–gold nanoparticle hybrid systems:
Langmuir-Blodgett assembly of nanostructured films. Nanoscale 5:2974–2982
52. Soldano C, Mahmood A, Dujardin E (2010) Production, properties and potential of graphene.
Carbon 48:2127–2150
53. Allen MJ, Tung VC, Kaner RB (2010) Honeycomb carbon: a review of graphene. Chem Rev
110:132–145
54. Loh KP, Bao Q, Eda G, Chhowalla M (2010) Graphene oxide as a chemically tunable
platform for optical applications. Nat Chem 2:1015–1024
55. Zhang J, Zhang F, Yang H et al (2010) Graphene oxide as a matrix for enzyme
immobilization. Langmuir 26:6083–6085
56. Liu Y, Yu D, Zeng C et al (2010) Biocompatible graphene oxide-based glucose biosensors.
Langmuir 26:6158–6160
57. Du D, Wang L, Shao Y et al (2011) Functionalized graphene oxide as a nanocarrier in a
multienzyme labeling amplification strategy for ultrasensitive electrochemical immunoassay
of phosphorylated p53 (S392). Anal Chem 83:746–752
7 Self-assembly Thin Films for Sensing 163

58. Wu M, Kempaiah R, Huang P-JJ et al (2011) Adsorption and desorption of dna on graphene
oxide studied by fluorescently labeled oligonucleotides. Langmuir 27:2731–2738
59. Liu X, Zhu H, Yang X (2011) An amperometric hydrogen peroxide chemical sensor based on
graphene-Fe3O4 multilayer films modified ITO electrode. Talanta 87:243–248
60. Cao X, Ye Y, Li Y et al (2013) Self-assembled glucose oxidase/graphene/gold ternary
nanocomposites for direct electrochemistry and electrocatalysis. J Electroanal Chem
697:10–14
61. Weng X, Cao Q, Liang L et al (2013) Simultaneous determination of dopamine and uric acid
using layer-by-layer graphene and chitosan assembled multilayer films. Talanta 117:359–365
62. Liu S, Yan J, He G et al (2012) Layer-by-layer assembled multilayer films of reduced
graphene oxide/gold nanoparticles for the electrochemical detection of dopamine.
J Electroanal Chem 672:40–44
63. Liu Y, Liu Y, Feng H et al (2012) Layer-by-layer assembly of chemical reduced graphene and
carbon nanotubes for sensitive electrochemical immunoassay. Biosens Bioelectron 35:63–68
64. Cote LJ, Kim F, Huang J (2009) Langmuir-Blodgett assembly of graphite oxide single layers.
J Am Chem Soc 131:1043–1049
65. Jia B, Zou L (2013) Langmuir-Blodgett assembly of sulphonated graphene nanosheets into
single- and multi-layered thin films. Chem Phys Lett 568–569:101–105
66. Wen J, Jiang Y, Yang Y, Li S (2013) Conducting polymer and reduced graphene oxide
Langmuir-Blodgett films: a hybrid nanostructure for high performance electrode applications.
J Mater Sci: Mater Electron 25:1063–1071
67. Kumar V, Bahadur N, Sachdev D et al (2014) Restructural confirmation and photocatalytic
applications of graphene oxide–gold composites synthesized by Langmuir-Blodgett method.
Carbon 80:290–304
68. Yin Z, He Q, Huang X et al (2011) Real-time DNA detection using Pt nanoparticle-decorated
reduced graphene oxide field-effect transistors. Nanoscale 4:293–297
69. Thostenson ET, Ren Z, Chou T-W (2001) Advances in the science and technology of carbon
nanotubes and their composites: a review. Compos Sci Technol 61:1899–1912
70. Rivas GA, Rubianes MD, Rodríguez MC et al (2007) Carbon nanotubes for electrochemical
biosensing. Talanta 74:291–307
71. Ajayan PM (1999) Nanotubes from Carbon. Chem Rev 99:1787–1800
72. Gharbavi K, Badehian H (2014) Structural and electronic properties of armchair (7, 7) carbon
nanotubes using DFT. Comput Mater Sci 82:159–164
73. Baughman RH, Zakhidov AA, de Heer WA (2002) Carbon nanotubes–the route toward
applications. Science 297:787–792
74. Zhao Q, Gan Z, Zhuang Q (2002) Electrochemical sensors based on carbon nanotubes.
Electroanalysis 14:1609–1613
75. Wu B-Y, Hou S-H, Yin F et al (2007) Amperometric glucose biosensor based on multilayer
films via layer-by-layer self-assembly of multi-wall carbon nanotubes, gold nanoparticles and
glucose oxidase on the Pt electrode. Biosens Bioelectron 22:2854–2860
76. Sun P, Jiang Y, Xie G et al (2011) Layer-by-layer assembly carbon nanotubes thin film based
gas sensors for ammonia detection. Sci China Inf Sci 54:2680–2686
77. Jing H, Jiang Y, Du X et al (2012) Humidity sensing properties of different single-walled
carbon nanotube composite films fabricated by layer-by-layer self-assembly technique. Appl
Phys A 109:111–118
78. Sun Y, Ren Q, Liu X et al (2013) A simple route to fabricate controllable and stable
multilayered all-MWNTs films and their applications for the detection of NADH at low
potentials. Biosens Bioelectron 39:289–295
79. Poonia M, Manjuladevi V, Gupta RK et al (2015) Ultrathin films of single-walled carbon
nanotubes: a potential methane gas sensor. Sci Adv Mater 7:455–462
80. Wang L, Li Y, Li G, Ye B (2015) A new strategy for enhancing electrochemical sensing from
MWCNTs modified electrode with Langmuir-Blodgett film and used in determination of
methylparaben. Sens Actuators B Chem 211:332–338
164 C.M. Miyazaki et al.

81. Takezawa A, Nara R, Kushida M (2014) Control of Fe catalyst nanoparticle density for
carbon nanotube growth using Langmuir-Blodgett deposition of mixtures of Fe nanoparticles
and arachidic acid. Thin Solid Films 554:69–73
82. Iakoubovskii K (2009) Techniques of aligning carbon nanotubes. Cent Eur J Phys 7:645–653
83. Venet C, Pearson C, Jombert AS et al (2010) The morphology and electrical conductivity of
single-wall carbon nanotube thin films prepared by the Langmuir-Blodgett technique.
Colloids Surf Physicochem Eng Asp 354:113–117
84. Lee J-H, Kang W-S, Najeeb CK et al (2013) A hydrogen gas sensor using single-walled
carbon nanotube Langmuir-Blodgett films decorated with palladium nanoparticles. Sens
Actuators B Chem 188:169–175
85. Fu Y-R, Zhang S, Chen M, Qian D-J (2012) Morphology and electrochemical properties of
amphiphilic viologen functionalized multiwalled carbon nanotube hybrids in
Langmuir-Blodgett films. Thin Solid Films 520:6994–7001
86. Lo C-L, Lee Y-L, Hsu W-P (2010) Behavior of mixed multi-walled carbon nanotube/P3HT
monolayer at the air/water interface. Synth Met 160:2219–2223
87. Sun Q, Zorin NA, Chen D et al (2010) Langmuir-Blodgett films of pyridyldithio-modified
multiwalled carbon nanotubes as a support to immobilize hydrogenase. Langmuir 26:
10259–10265
Chapter 8
Phthalocyanines as Sensitive Materials
for Chemical Sensors

Debdyuti Mukherjee, Revanasiddappa Manjunatha,

Srinivasan Sampath and Asim Kumar Ray

8.1 Introduction

Metal phthalocyanine (MPc) is an intensely blue-green-coloured, non-toxic aromatic

macrocyclic (18 p-electron conjugated ring system) compound with four nitrogen
atoms in the central cavity (Fig. 8.1a). This man-made compound was discovered in
1927 as a side product of the conversion of o-dibromobenzene to phthalonitrile [1].
The structures of phthalocyanines are related to porphyrins, where the four –CH
groups in the inner ring are replaced by four meso nitrogen atoms. Phthalocyanine
(Pc) molecules are characterised by high thermal and chemical stability, optical
absorption and chemical functionality [2]. The central atom can be seventy different
elemental ions, including H+ ion or metalloid elements. There are sixteen possible
substitution sites on the benzenoid rings and axial ligation sites at the central metal or
metalloid. Suitable substitutions on the Pc rings, however, confer solubility in dif-
ferent solvents. For example, sulphonated phthalocyanine compounds with SO3H on
ring are soluble in water, while solubility in organic solvents can be achieved with Pc
molecules substituted with aliphatic chains. As shown in Fig. 8.1b, the physical
properties can be tuned by careful substitutions. The unique structural and physical

D. Mukherjee
R. Manjunatha
S. Sampath

Department of Inorganic and Physical Chemistry, Indian Institute of Science,
Bangalore 560012, India
e-mail: [email protected]; [email protected]
R. Manjunatha
e-mail: [email protected]
S. Sampath
e-mail: [email protected]
A.K. Ray (&)
Institute of Materials and Manufacturing, Brunel University London,
Uxbridge, London UB8 3PH, UK
e-mail: [email protected]

© Springer International Publishing AG 2017 165

T.R.L.C. Paixão and S.M. Reddy (eds.), Materials for Chemical Sensing,
DOI 10.1007/978-3-319-47835-7_8
166 D. Mukherjee et al.

(a)

(b)

C6H13S SC6H13

C6H13S N N N SC6H13

N M N

N N N
C6H13S SC6H13

C6H13S SC6H13

Fig. 8.1 a Chemical structure of unsubstituted phthalocyanine, b optical absorption spectra for
four different metal phthalocyanines with the hexyl substitutions with their absorption peak at
779 nm (brown), 782 nm (purple), 818 nm (red) and 836 nm (mauve) for Zn, Cu, Pb and InCl,
respectively

properties have made these molecules important in various areas of material science
such as semiconductors, molecular electronics, optoelectronics, photonics, photo-
voltaics, electrochromism, liquid crystal applications and catalysis (such as elec-
trochemical oxygen reduction reaction and carbon dioxide reduction reaction) [3–
13]. These compounds are also used in the active sites of enzymes, which are
responsible for catalytic reactions such as aerobic oxidation, reduction and transport
of dioxygen and destruction of peroxides [14]. Certain derivatives of phthalocya-
nines are used as second-generation photosensitisers for photodynamic therapy
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 167

(PDT) of cancer [15] as they show strong absorption in visible to near-IR region
(600–850 nm), possessing greater tissue penetration properties [16] and satisfactory
photosensitisation of singlet oxygen [17].
Several studies have been carried out based on chemical sensing by varying the
central metal atom in the phthalocyanine core or by changing the peripheral sub-
stituents [18–20]. The presence of p-electrons in the core-shell promotes interaction
with various analytes including gases and liquids [21], which results in detectable
changes in the physical properties of the material such as mass [22], conductivity
[23, 24] and optical properties [25, 26]. The interactions between the Pc films and
the gases may be classified in terms of irreversible chemical affinity, reversible
(usually charge-transfer) chemical reaction or bulk sorption. The properties of
phthalocyanines, mainly the ionisation potentials, can be varied by the insertion of
metal ions or attachment of additional atoms or groups, and it is found that the
detection limits, sensitivity and selectivity can be tuned for different environmen-
tally relevant gases by the introduction of substituents [27, 28].
This chapter is organised into eight sections describing six different transduction
methods for detecting environmental pollutants, biological cofactors and neuro-
transmitting agents. The detectionof pollutant gases and volatile organic compounds
(VOCs) whether in the gaseous phase or dissolved in water is of great environmental
importance due to the extreme hazards posed by their presence in small amounts in
the ambient temperature. Air quality standards for the UK, USA and World Health
Organization are summarised for principal pollutants in Table 8.1. The next section
presents a summary of amperometric sensors used primarily for environmental
pollutants such as nitrogen dioxide (NO2), carbon monoxide (CO) and volatile
organic compounds (VOCs). Organic field-effect transistor (OFET) sensors have
attracted enormous interest in chemical sensing and have been reported for various
analytes [29]. The third section is devoted to the application of OFET for a range of
pollutants in gaseous phases and volatile and non-volatile solvents. Adverse effects
of toluene exposure on human beings have been well established by the World Health
Organization (WHO). Continuous exposure to it causes impairment of the nervous
system, hormonal imbalance in males and spontaneous abortion in females.
Organophosphonates and organic amines such as dimethyl methyl phosphonate
(DMMP) and triethylamine are target analytes of much interest for applications such
as security or food analysis, respectively. DMMP is a common test gas acting as a
simulant for G class CW agents and amines are markers for chemical processes
occurring during food degradation. Two major classes of electrochemical sensors are
amperometric and potentiometric. The amperometric sensor is based on the mea-
surement of current signal, which is related to the concentration of the analyte [30],
while the ion-selective electrodes belong to the class of potentiometric sensors. The
deployment of electrochemical sensors and their applications are described in the
fourth section of this chapter. Quartz crystal microbalances (QCMs) have been
extensively used as both bio- and environmental sensors. The measurement of
analytes with different concentrations depends upon the decrease of the resonant
frequency of an oscillating, specially cut quartz crystal due to the binding mass on a
phthalocyanine-coated surface. The performance of Pc-based QCMs is explained in
 168

Table 8.1 Air quality standards for the UK, USA and World Health Organization (WHO)
Pollutant UK WHO USA
Concentration Standard measured Concentration Standard measured Concentration Standard measured
Benzene 5 ppb Annual mean Data not available 10 ppm 8 h
1,3-butadiene 1 ppb Annual mean Data not available
Carbon monoxide (CO) 8.6 ppm 8 h mean 8.6 ppm 8 h mean 9 ppm 8 h mean
Lead (Pb) 0.25 lg/m3 Annual mean 0.5 lg/m3 Annual mean 1.5 lg/m3 Quarterly mean
Nitrogen dioxide (NO2) 21 ppb Annual mean 105 ppb 1 h mean 0.05 ppm Annual mean
Ozone (O3) 50 ppb 8 h mean 60 ppb 8 h mean 0.08 ppm 8 h mean
Particles (PM10) 50 lg/m3 24 h mean 70 lg/m3 24 h mean 50 lg/m3 Annual mean
Sulphur dioxide (SO2) 100 ppb 15 min mean 188 ppb 10 min mean 0.14 ppm 24 h mean
Toluene 200 ppm 8h
D. Mukherjee et al.
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 169

the fifth section with an emphasis of selectivity, sensitivity and detection limits. The
sixth section introduces piezoelectric surface acoustic wave (SAW) sensors with an
emphasis on the applications of phthalocyanine films as a delay line. Pcs are highly
absorbing materials with the Soret and Q-bands in the regions of 250–350 and
600–700 nm, respectively. These properties have been exploited in the development
of UV, fibre-optic, ellipsometric and surface plasma sensors for use in various fields.
Optical sensors are often used in the food industry, medicine and environmen-
tal control for the analytical measurements of gas or dissolved oxygen.
A comprehensive description is given in the seventh section, pointing out the
importance of different substituents on the ring and core metal ions with regard to
highly selective, sensitive and reversible sensing mechanisms. The chapter concludes
with summarising the main points of up-to-date achievements in the investigation of
macrocyclic phthalocyanine molecules as sensing materials. The scope of future
developments has also been highlighted as a concluding remark.

8.2 Amperometric Sensors

Detection of nitrogen dioxide (NO2) gas under ambient atmosphere is highly

important in industrial environment such as engine exhaust and in medical diag-
nostics such as human breath and various other places, due to its poisonous and
hazardous properties. Langmuir–Blodgett (LB) films of metal-free amphiphilic
phthalocyanine molecules in Fig. 8.2a have been used in the development of
amperometric sensors in the configuration of platinum-interdigitated electrode
patterns on glass substrates (Fig. 8.2b) for selective detection of NO2 gas in the
occupational hygiene range. This interdigitated electrode structure has been chosen
in order to facilitate the direct analyte exposure on an enlarged active area. The
films are exposed to alternate 2-min pulses of air and NO2 with increasing con-
centration from 1 to 5 ppm. It is shown in Fig. 8.2c that the sensors exhibit a useful
room temperature, reproducible response on exposure to NO2 gas. The magnitude
of the response is proportional to the NO2 gas concentration and is also dependent
upon the LB dipping direction relative to the axes of the electrodes. The response is
larger by a factor of 7 for a dipping direction that is perpendicular than parallel to
the electrode fingers. This behaviour is ascribed to the alignment of columns of
molecules along the dipping direction. The sensors have been tested by exposing to
five cycles of 0.625 ppm chlorine (Cl2) and 50 ppm carbon monoxide (CO).
Figure 8.2d shows the complete lack of response to both Cl2 and CO gases, but
devices work well on subsequent exposure to NO2 gas. The device can withstand at
least 65% of relative humidity without detrimental effects [31]. The growth of
current through a five-layer-thick tert-butyl silicon-[bis ethyloxy]-phthalocyanine
(Pc(OC2H5)2), (SiPc) LB film on interdigitated gold electrode has been monitored
at a fixed bias voltage of 5 V over nearly two hours under a N2 environment
containing 5 ppm NO2 at room temperature. As expected, NO2 gas molecules
170 D. Mukherjee et al.

(a) (b)
C10H21 C10H21

C10H21 N N C10H21
N

N-H H-N

N N N
C10H21 C10H21

(c)
OH HO

(d)

Fig. 8.2 a Chemical structure of substituted, amphiphilic metal-free phthalocyanine, b interdig-

itated platinum electrodes on glass substrate, c electrical response to NO2 of different ppm levels
and d selective response to exposure of Cl2 and H2S

accept electrons from Pc films, increasing hole-dominated conduction. The initial

response is fast, presumably due to easy displacement of oxygen species from the
receptive adsorption sites. The sensor performance has been analysed in terms of
sensitivity and half-life. The sensitivity which is defined as normalised ratio of
current growth until the NO2 switch-off time is estimated to be 706 %. The response
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 171

time for the conductivity to attain 50 % of its final value is found to be 457 s. The
measurements were repeated for 2- and 10-layer-thick SiPcLB films, and both
sensitivity and response time depended upon the thickness of the SiPc sensing film
[32]. Using the experimental configurations similar to the one described earlier for
SiPc, the sensing properties of 20-layer-thick LB films of amphiphilically substi-
tuted gadolinium bisphthalocyanine molecules have been investigated under peri-
odic exposure to varying concentrations of NO2 from 0.25 to 1.0 ppm. The
response time is typically 90 s with slower recovery under dry air. The apparent
linearity between relative response and NO2 provides an easy-to-implement cali-
bration tool [33]. Ytterbium bisphthalocyanine 25-layer-thick Langmuir–Blodgett
(LB) films were found to exhibit an increase of conductivity by 200 % in N2
atmosphere containing 5 ppm Cl2 from pyrovoltage measurements at 120 °C. This
effect is not observed in the sublimed film of these same molecules, implying that
the organised molecular patterns in LB films are responsible for this effect [34].
Significant changes in the room temperature conductivity of LB films of europium
bisphthalocyanine (EuPc2) LB films are observed when exposed to Cl2 in N2
atmosphere with a low detection limit of 2 ppm. The increase in conductivity is
linearly dependent upon the Cl2 concentrations up to 80 ppm. The response is
reversible when cycled at 1-min exposure to Cl2 followed by the 1-min N2 cleaning.
The EuPc2-based amperometric sensor may be considered to be selective to Cl2 in
light of the observation that there is no change in conductivity on exposure to
ammonia (NH3), nitric oxide (NO) and NO2 [35]. The 5–100 ppm NO2 sensing
response of evaporated cobalt phthalocyanine films (35–80 nm thick) on
Au-interdigitated electrodes has been investigated by recording current at a constant
bias of 1 V. Like the LB phthalocyanine films, the kinetic responses over 100 min
are characterised by initially rapid surface adsorption, and subsequently, slow
diffusion into the bulk and the complete recovery is, therefore, not achieved [36].
Organic molecular beam deposited 80-nm-thick films of tetrafluoro-substituted
cobalt phthalocyanine (CoPcF4) on interdigitated substrates are exposed to NH3 of
the concentrations 10–50 ppm and the resistance of the CoPcF4 film is found to
decrease because of charge transfer between the CoPcF4 and NH3 molecules. The
recovery time becomes shorter from 10 and 47 s to 8 and 41 s of exposure of NH3
concentrations of 10 and 50 ppm, respectively, to CoPcF4 films annealed at 300 °C.
The detection limit is also lowered from 2.5 to 0.7 ppm on annealing; the response
time is, however, 4–5 s in both cases. Fluorinated CoPcF4 sensors are found to be
more sensitive than unsubstituted CoPc [37]. The electrical conductivity of
single-component water-soluble copper phthalocyanine (CuPc-4SO3Na) films
increases by a factor of 400 under an ammonia environment with pressure ranging
from 5  10−1 to 3  103 Pa. This is better than the value of 150 for composite
films containing CuPc-4SO3Na and tin dioxide (SnO2) nanoparticles of 20 nm in
size. This superior sensing behaviour may be attributed to possible increase in
carrier near the junction surface between the CuPc-4SO3Na film and the electrode.
The recovery time is 1 s for both films [38]. Cast-coated films of dinuclear copper,
zinc and cobalt phthalocyanines have been employed in the fabrication of
172 D. Mukherjee et al.

amperometric sensors for volatile organic compounds mixed with high-purity

nitrogen as a carrier gas. Films show good room temperature stability to exposure of
toluene (250–1000 ppm), chloroform, acetone and methanol (1500–6000 ppm).
Cobalt derivatives give the best sensitivity of 1.3  10−1 ppm−1, 1.6  10−2 ppm−1
and 2.1  10−2 ppm−1 for toluene (250 ppm), acetone (150 ppm) and chloroform
(1500 ppm), respectively. Response time varies from 30 to 110 s relative to the
concentration and type of volatile organic compounds (VOCs) [39]. Normalised
changes in electrical resistances of drop-cast hybrid films of single-walled carbon
nanotubes (SWCNTs) and symmetrically octa-substituted zinc phthalocyanine
(ZnPc) bearing eight polyoxyethylene groups have been studied as a function of
concentration of ammonia (NH3) vapour in the range of 1–200 ppm. The detection
limit of hybrid sensors is found to be 1 ppm which is ten times smaller than sensors
using SWCNT alone. Similar investigations have been carried out on NH3 sensing
of hybrid films using asymmetrically substituted zinc phthalocyanine bearing one
pyrene and six polyoxyethylene groups as side chains, and the response is found to
be smaller than that of ZnPc derivatives without pyrene substitutions by a factor of
nearly 3 [40]. 8–10-lm-thick porous silicon (PS) layers have been coated with
freshly prepared soluble cadmium (Cd), cobalt (Co) and aluminium (Al) phthalo-
cyanine films to fabricate amperometric NO2 sensors. The dynamic responses of
NO2 have been investigated by measuring the sheet resistance for different con-
centrations of NO2 varying between 100 and 500 ppm. All samples show increases
in conductivity on NO2 exposure because of the overall enhancement in hole
concentrations. In the case of the detection of 100 ppm concentration, CdPc and
AlPc coatings improve the sensitivity of detection to over 92 from 70 % for the
virgin PS sensor. The response time is found to be 4 min for PS/CdPc and PS/AlPc,
and the corresponding recovery time is 6–7 min. The photoluminescence spectra of
optically luminescent PS sample are found to be sensitive to the NO environment,
and thus, it provides an optical NO sensor [41].

8.3 Organic Field-Effect Transistor (OFET) Sensors

OFET sensors have been extensively studied for monitoring the analytes that may
be divided into three groups. The first group consists of gases such as ammonia
(NH3), hydrogen sulphide (H2S), nitrogen dioxide (NO2), ozone (O3) and ethylene
that are highly toxic and affecting the environment and health [42]. Volatile sol-
vents such as acetonitrile, THF, toluene, acetone, hexane and isopropyl amine and
less volatile liquids such as hydrogen peroxide (H2O2) and dimethyl methyl
phosphonate (DMMP) belong to the second group [43]. Pc-based OFETs for
sensing bions such as phosphate [44] and biomarkers in physiological solutions,
such as protein and glucose [45] and lactic acid [46], have also been included.
Detection of explosive solids, such as trinitrotoluene (TNT) and dinitrotoluene
(DNT), [47] have also attracted attention in recent years.
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 173

Fig. 8.3 K probe (KP) signal

for NO2 sensing based on
CuPc sensing layer at a
temperature of 25 °C, 50 %
of relative humidity and
carrier gas: 200 sccm
synthetic air. Adapted from
Ref. [49]

8.3.1 Gas-Phase Sensing of Environmental Pollutants

Copper phthalocyanine (CuPc)-based FET sensors are reported to show high sen-
sitivity towards the detection of NO2 gas [48]. Flip-chip suspended gate (SG) FETs
using thermally evaporated 150–200-nm-thick CuPc films as sensing layers may be
operated at room temperature for low-power sensors of NO2 under 1 mW dissi-
pation where a short recovery is not required [49]. The sensitivity of the Pc layers
towards NO2 gas has been estimated in terms of the contact potential difference
(CPD) between CuPc films and the reference gold electrode from Kelvin probe
(KP) measurements. Figure 8.3 shows a detection limit of <50 ppb coupled with
good sensitivity (20–70 mV/concentration decade) and high selectivity. It is
important to note that the FET transducers are suited to read CPD signals induced
by gases. These sensors survive many months under operation without visible
degradation. A p-channel OFET using a Langmuir film of symmetrically substituted
CuPc molecules is reported to have achieved the detection limit of *100 ppb with
high stability [50]. The change in drain-to-source current is observed to be less than
10 % on exposure of the active layer to NH3, H2, H2S, CO and alcohol up to
500 ppm, showing the excellent selectivity for NO2 gas. It is found that the
source-to-drain current IDS values are directly proportional to both the film thick-
ness and NO2 concentration. Thicker films exhibit stronger response (Fig. 8.4a, b).
It is also observed that the sensing characteristics of the devices are reversible.
Figure 8.4c gives the response and recovery times for various film thicknesses
when the sensor is exposed to 2.5 ppm NO2. Thus, thin LB films exhibit fast
response and recovery times less than 10 min under vacuum of 10 kPa or at a high
temperature of 70 °C. The simple preparation method is easy to scale up for
commercial exploitation. Physical vapour-deposited CuPc films are used as active
layer in floating gate hybrid FET transducers to detect NO2 in fire according to the
European standard. Figure 8.5 presents the results of response sensor characteristics
of two different test fires. The detection limit of the low-powered fire detector is
about 0.1 ppm. The early sensor response is found to be adequate for open fires
174 D. Mukherjee et al.

Fig. 8.4 a Change of IDS of the LB-OFET gas sensor as a function of NO2 concentration for
different film thicknesses of the sensing material, b output characteristics of the sensor (35-layer
LB film), c response and recovery characteristics of the sensor for two film thicknesses
(CNO2 = 2.5 ppm). Adapted from Ref. [50]

(TF6) involving the production of NO2 gas up to 5 ppm. The sensor performance is
not adequate for early warning of smouldering fire (TF2) test since the concentration
of NO2 is normally lower than the detection limit [51]. NO2 concentrations as low
as 2 ppm have been detected at room temperature by using hybrid silicon transistors
in which the gate electrodes are replaced by the charge flow capacitors of
150-nm-thick spin-coated film of metal-free octahexyl-substituted phthalocyanine
between two metal electrodes. The sensitivity obtained from the slope of kinetic
measurements of the drain current at a fixed gate voltage of *5 V is found to show
a linear dependence on NO2 concentrations between 2 and 12 ppm. The device is
reversible with 80 % recovery within 1 h, followed by a slower recovery to the
baseline within 4 h. No demonstrable response to exposure of SO2 and H2S up to
25 ppm is recorded [52].
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 175

Fig. 8.5 Response of a

CuPc-based OFET sensor to
different test fires. Adapted
from Ref. [51]

Sulphur dioxide (SO2) is a toxic gas that is released during the burning of fossil
fuels in air. These molecules dissolve easily in water droplets in cloud, forming acid
rain which is extremely dangerous to the environment. In human beings, repeated
exposure to low levels of SO2 blocks nerve signals from pulmonary stretch
receptors and causes permanent pulmonary impairment, and therefore, attempts
have been made to the design and fabrication of room temperature-reliable and
highly sensitive SO2 gas sensors with a very low detection limit, satisfying the
environmental agency requirement of 0.5 ppm [53, 54]. As shown in Fig. 8.6a, a
bottom gate FET is designed with CuPc nanowires as an active semiconducting
layer. As the CuPc nanowires are exposed to pulses of SO2 analyte varying in
concentration (CSO2) between 0.5 and 20 ppm, typical room temperature sensor
characteristics in Fig. 8.6c are obtained in terms of the time response of IDS. The
detection limit is 0.5 ppm with corresponding fast response and recovery times of 3
and 8 min, respectively. The sensitivity in percentage is estimated as the ratio of IDS
to the base current in a N2 environment, and it is shown in Fig. 8.6d that the device
exhibits a high sensitivity of 119 % for 0.5 ppm of SO2 [55].
Ozone is a strong oxidant and is used in many industrial and consumer applica-
tions related to oxidation processes. The presence of an ozone layer in the upper
atmosphere is beneficial, preventing ultraviolet light from reaching the earth’s sur-
face. Ozone is, on the other hand, very harmful and damages mucous and respiratory
tissues in animals and plants, if the concentration exceeds 100 ppb. Thus, ozone is a
potent respiratory hazard and pollutant near the ground level and a key component of
photochemical smog. It is, therefore, necessary to be able to detect the ozone gas at
ppb levels. Ozone sensors based on different metal phthalocyanines have been
reported [56]. Figure 8.7a shows a molecular OFET with a thin layer (30–60 nm) of
vacuum sublimed nickel phthalocyanine (NiPc) as active material for sensing ozone
up to 300 ppb concentration. In this design, the source and drain electrodes remain
buried in the dielectric SiO2 layer and the NiPc sensing film can easily be deposited
on the flat surface [57]. Figure 8.7b presents the sensitivity as a function of ozone
176 D. Mukherjee et al.

Fig. 8.6 a Schematic illustration of CuPc nanowire-based FET sensor for SO2 analyte.
b Scanning electron micrograph of the device, nanowire is placed on PMMA. c Response of
the device towards SO2 concentration from 0.5 to 20 ppm at gate voltage VG = −10 V and
drain-to-source voltage VSD = −15 V. d Sensitivity of the device at VG = −10 V and
VSD = −15 V. Adapted from Ref. [55]

Fig. 8.7 a Schematic representation of molecular field-effect transistor for ozone; b difference in
exposure – Irecovery) as a function of ozone concentration in charcoal-filtered air
current (DI = Imax min

(response, 0.079 nA ppb−1) for texposure = 2 min, trecovery = 8 min. The first two points (blue
colour) are for the measurements in ambient air. Adapted from Ref. [57]

concentration where DI is the difference between the drain currents for 2-min
exposure and 8-min recovery cycles. A linear relationship is observed in the range of
0–100 ppb of ozone concentration compatible with the range encountered in urban
atmospheres. The sensitivity is estimated to be 0.079 nAppb−1 from the slope, giving
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 177

a promising possibility of calibration over concentration ranges naturally occurring in

ambient environment. The use of a dynamic rest (flow of ozone-free air) instead of a
static rest reduces the duration of the recovery period. Values of <1 % and
>200 ppm h reported for the noise-to-signal ratio and the lifetime, respectively, for
the present device compared very well with previously published data. It is estab-
lished from X-ray photoelectron spectroscopy and UV–Visible absorption mea-
surements that complete degradation of the phthalocyanine layer takes approximately
one year in the presence of 1 ppm O3, which is much larger than the ambient
concentration, usually in 10–50 ppb range.
Hydrogen sulphide (H2S) is heavier than air, highly poisonous, corrosive,
flammable and explosive. In addition, it is essential to detect H2S in the hydrogen
feed stream of fuel cells to prevent the catalyst poisoning and to measure the quality
of guard beds used to remove sulphur from hydrocarbon fuels. Portable H2S sensors
are also used in personal protective applications. The current-voltage characteristics
of bottom-contact FETs using vacuum-deposited CuPc film have studied to monitor
H2S gases of varying concentration (CH2S) between 100 and 500 ppm in dry air
[58]. With increase in H2S, oxygen adsorbed on the surface of the CuPc film is
displaced by the H2S molecules. The reducing gas injects electrons into the p-type

Fig. 8.8 a Real-time response curve, b sensing characteristics of CuPc-OFET to different H2S
concentrations, c selectivity of CuPc-OFET towards different gases (100 ppm). Adapted from
Ref. [58]
178 D. Mukherjee et al.

CuPc material, decreasing the hole concentration, and consequently, IDS values tend
to decrease over time (Fig. 8.8a). The recovery is slower than response because of
relatively gradual desorption at strong adsorption sites compared to weak adsorp-
tion sites. Figure 8.8b reveals a linear relationship of the gas sensing response R as
IdsðH SÞI
a function of CH2 S where R ¼ 2 dsðairÞ
IdsðairÞ corresponding to the saturation regime.
Figure 8.8c compares the response of 100 ppm H2S to that of four other analytes
SO2, CO2, CH4 and H2, giving a good degree of selectivity for H2S detection. The
relatively high response may be attributed to the polar nature of H2S.

8.3.2 Sensing of Volatile and Non-volatile Solvents

Metal phthalocyanines have also been used to detect various volatile and
non-volatile solvents such as acetonitrile, tetrahydrofuran (THF), toluene, acetone,
hexane, isopropyl amine (IPAm), water, hydrogen peroxide (H2O2) and dimethyl
methyl phosphonate (DMMP) [59]. When MPc is used in chemical sensors, the
analyte molecules typically bind to the central metal ion of the MPcs and chemical
selectivity may be achieved by varying the central metal. Figure 8.9a summarises
the characteristic fingerprint of MPcs with four different metal centres (Cu2+, Ni2+,
Co2+ and Zn2+) for three different volatile solvents, acetonitrile, THF and toluene.
The central metal atom is believed to be responsible for determining the capacity for
binding the solvent molecules to different MPcs [60]. It is shown in Fig. 8.9b that
the ratio of association rate constant (kON) to dissociation rate constant (kOFF)
indicates the strength of solvent-channel interactions kON is expected to be much
larger than kOFF when the solvent is bound tightly to the channel. It is observed that
all the four MPc channels strongly bind to acetonitrile and loosely bind to THF. In
the case of toluene, CoPc and NiPc are observed to bind strongly, while the binding

Fig. 8.9 a Steady-state current modulation data (DIDS/IDS) and b ratio of transient rates (kON/
kOFF) upon exposure to 100 ppm of the specified solvent. Transistor bias conditions used are
VDS = −20 V, VGS = −20 V. Adapted from Ref. [60]
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 179

Fig. 8.10 Sensing performance of a CuPc-, b 8-NTCDI- and c DMP-NTCDI-based OFETs in

terms of mobility (square) and threshold voltage (triangle) by comparing pristine values (solid
lines) with those obtained after exposure to eight analytes (broken lines)

with CuPc and ZnPc is not very effective. OFET sensors are becoming efficient for
analytes with binding energies that are greater than the intrinsic energetic disorder
of the organic semiconductor.
OFET sensors were fabricated using evaporated films of three organic semi-
conductors, namely CuPc and naphthalenetetracarboxylic diimide derivatives
(8-NTCDI) and dimethylaminopropyl (DMP-NTCDI). The sensor array of these
180 D. Mukherjee et al.

OFETs was found to be capable of discriminating between analytes when exposed

to acetic acid (1500 ppm), acetone (11,700 ppm), IPAm (1000 ppm), DMMP
(800 ppm), water (4700 ppm), H2O2 (3800 ppm), hexane (2600 ppm) and toluene
(2400 ppm). Figure 8.10a shows the effect of the gas exposure on transistor
parameters such as mobility and threshold voltage of CuPc-based OFETs in terms
of change in mobility and threshold voltage. It can be seen that a limited selectivity
of H2O2 over other analytes has been achieved 8-NTCDIs are regarded as being n-
type semiconductors, while DMP-NTCDI, because of its ionisable side chains,
shows response directions and magnitudes that are not correlated with those of the
other two compounds. Responses in terms of mobility and threshold voltage
changes are depicted in Fig. 8.10b, c, showing the differences in both magnitude
and direction. The mobility of 8-NTCDI is greater than that of DMP-NTCDI by
three orders of magnitude, presumably due to the film crystallinity and charge
carrier trapping by protonated amine groups in the sensing layer. This indicates that
the array is capable of qualitative discrimination among all these analytes. The
analyte properties contribute to the sensor performance. For example, IPAm is a
strongly electron-donating agent, while H2O2 will accept electrons because of its
oxidative capability. Analytes may, therefore, be classified into two groups: IPAm,
DMMP and acetone belonging to the first group and water, acetic acid, hexane and
toluene forming the second group [61].

8.3.3 Anions Sensing

Phosphorus compounds are an essential nutrient for all plants and are also an
important ingredient in fertilisers used in modern agriculture which are generally
manufactured from phosphate rock. In order to precisely control the amount of
phosphorus supplied, a fast, simple, accurate and sensitive method of measuring
phosphate in soil is required. Clinical diagnosis is another area where phosphate
measurements are very important.
Suitable devices can be fabricated by connecting a coated platinum wire elec-
trode to the gate of a conventional field-effect transistor using cobalt phthalocyanine
as ion exchange electrode substance and polyvinyl chloride as the membrane matrix
[62]. The saturation current IDS at a fixed drain-to-source bias VDS and
gate-to-source VGS is recorded in Fig. 8.11a as a function of the ion activity (ai) of
H2PO4− on a logarithmic-linear scale. Figure 8.11b shows a similar dependence of
VGS on ion activity at fixed VDS and IDS in the saturation region. A linear response
to H2PO4− is achieved in the range 10−5–10−1 M with the response slope of 45 mV
per decade change of ion activity. Figure 8.11c shows the rise in potential with
increasing response time as the concentration of H2PO4− ions is decreased from
10−1 to 10−5 M. The selectivity coefficients are found to be within the range 10−3–
10−2 for fluorine (F−), sulphate (SO42−), nitric oxide (NO3−) and acetate ions, while
this value lies between 10−2 and 10−1 for chlorine (Cl−), bromine (Br−), iodine (I−),
perchlorate (ClO4−) and phosphate (H2PO22−) ions.
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 181

Fig. 8.11 a (IDS)1/2 (@ VDS = 3 V, VGS = 0 V) and b VGS (@ VDS = 3 V, IDS = 10 nA) as a
function of phosphate ion concentration, c response potential time curve for CW/FET with
different concentration of phosphate ion. Adapted from Ref. [62]

8.3.4 Biological Analytes

Lactic acid is a chemical that plays a substantial role in various biochemical pro-
cesses. Its level in body fluids is very important in detecting bacterial and fungal
infection inside the body. Elevated lactic acid levels are often observed in infected
patients [63]. The presence of excess lactic acid in food products has impacts on
their stability and storage lifetime. Hence, it is necessary to detect lactic acid in
clinical diagnosis, biotechnology and food industries [64]. Figure 8.12 shows the
decrease of IDS of CuPc-based OFET as the time of exposure to lactic acid is
prolonged [65]. The degree of reduction of IDS is correlated with the analyte
concentration. A clear response is observed for 10 µM solutions, demonstrating
high sensitivity of the device. It is also reported that CuPc is sensitive to pyruvic
acid, but the threshold concentration of pyruvic acid detection is 5 times greater
than that observed for lactic acid, implying that the device is less sensitive towards
182 D. Mukherjee et al.

Fig. 8.12 IDS of CuPc-OFET

device at a constant VGS as a
function of time for different
concentrations of lactic acid
(10 µM–2 mM). Adapted
from Ref. [65]

pyruvic acid than lactic acid. Water-gated OFETs of 50-nm-thick thermally sub-
limed films in bottom-contact top-gate configurations have been used in the
transduction of the biospecific interaction between reduced glutathione (GSH) and
glutathione S-transferase (GST) in aqueous Tris buffer solution (50 mM, pH 7.4) by
functionalising Au gate electrode with GSH and GST. These interactions are
important for immobilising GST on solid surfaces for biotechnological uses [66].

8.4 Electrochemical Sensors

Potentiometric sensors are most often based on the measurement of changes in the
interfacial potential at an electrode surface due to a selective ion exchange reaction.
The following gives an account of electrochemical sensors based on phthalocya-
nines as active elements.

8.4.1 Catecholamines

Catecholamines are a group of naturally occurring amines that act as neurotrans-

mitters and hormones. Dopamine (DA), one of the important catecholamines, is a
neurotransmitter in the mammalian central nervous system. In addition, it plays a
significant role in the control of cardiovascular, renal and hormonal functions [67].
Therefore, electrochemical detection of dopamine attains paramount importance. In
this direction, metallophthalocyanines have been used as electrocatalysts for the
detection of catecholamines. The negatively charged sulphonate group of func-
tionalised metallophthalocyanines have been exploited to develop multilayers by
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 183

means of layer-by-layer (LbL) techniques for the detection of different cate-

cholamines. Ultrathin films of negatively charged cobalt phthalocyanine and posi-
tively charged non-electroactive Mg-Al-layered double hydroxide are developed
using indium tin oxide (ITO) substrates and efficiently used for DA detection [68].
The sensor has been shown to possess high sensitivity, low detection limit and
excellent anti-interference activity in the presence of ascorbic acid (AA). Further, the
electrode modification is robust in terms of reproducibility and long-term stability as
compared to organic multilayers of poly(diallyldimethylammonium chloride)
(PDDA) and cobalt phthalocyanine. Chitosan is a naturally available polymer and is
positively charged. Multilayers of chitosan and nickel, copper and iron containing
phthalocyanines are proposed for the detection of DA [69]. The chitosan/copper- and
chitosan/iron-based multilayers show good selectivity for DA in the presence of AA.
The detection limit of DA is quite comparable with polyaniline/iron phthalocyanine
multilayer-based systems [70]. Ultrathin films based on naturally occurring cationic
antimicrobial peptide and nickel phthalocyanine (NiPc) [71], cashew gum
(Anacardium occidentale L.) and NiPc [72] are fabricated by LbL technique for the
detection of DA. Synthetic polymers such as poly(O-methoxyaniline) [73], poly
(allylamine hydrochloride) [74] and polyamidoamine [75] that consist of amine
groups have been used with metallophthalocyanines to fabricate thin films by
layer-by-layer (LbL) deposition technique for the detection of DA. Even though
cobalt phthalocyanine (CoPc) is found to be suitable as an electrocatalyst, major
concern arises from the detachment of electroactive material from the electrode
material under physiological conditions [76]. It is, therefore, essential to select good
composite material which will provide the microenvironment required to the elec-
troactive species as well as enhance electron transfer rate between electroactive
species and the electrode. Phosphovanadomolybdate fulfils these criteria and has
been used to modify the electrode along with CoPc. The modified electrode shows
superior electrocatalytic activity for the detection of DA. Mesoporous silica [77] and
carbon ceramic composites are also used as supporting matrices for CoPc modifi-
cation on the electrode surface.
Electrochemical detection of more than one analyte poses difficulties in terms of
electrode modification. Chernyshov et al. [78] fabricated screen-printed electrodes
(SPEs) modified with ionic liquid/graphite powder composite (CoPc is added as an
additive) and have shown good electrocatalytic activity towards a series of cate-
cholamines such as dopamine, adrenaline and dobutamine (a drug used in the
treatment of heart failure). The sensitivity and selectivity of the modified surface
have been explained on the basis of preconcentration effects at open-circuit
potentials. On the other hand, selective and simultaneous determinations of
dobutamine/morphine and dobutamine/paracetamol have been achieved by means
of gold nanoparticles/cobalt phthalocyanine-modified carbon paste electrodes [79].
Visible LED photoelectrochemical sensors have been developed for the detection of
L-dopa using oxygen reduction on TiO2-sensitised iron phthalocyanine (FePc)-
modified surface [80].
184 D. Mukherjee et al.

8.4.2 Ascorbic Acid and Uric Acid

Ascorbic acid (AA) is a water-soluble vitamin and is an essential nutrient for humans.
Cobalt phthalocyanine (CoPc) and multiwalled carbon nanotube composites have
been used to detect AA. The composite is prepared using p-p stacking interactions to
form a uniform network—like structure assisted by the four peripheral isoheptyl
substituents of CoPc. As a result, facile mass transport of analyte and a continuous
conducting pathway for the movement of electrons are achieved. Electrochemical
sensors fabricated using the composite show high currents and low potentials for the
oxidation of AA [81]. Similarly, copper phthalocyanine (CuPc) and graphene com-
posite have been used for the selective determination of AA, in the presence of DA
and uric acid. During the modification of the electrode, polyaniline is used as the
supporting matrix and it effectively reduces the leaching of graphene/CuPc from the
electrode surface, while simultaneously enhancing the composite conductivity [82].
Polyaniline is a highly conductive polymer in acidic medium. Aniline, when elec-
tropolymerised in the presence of CoPc, it produces a composite that shows good
conductivity at basic pH (pH 9.0). This composite has been used for the
sub-millimolar detection of AA under physiological pH, 7.4 [83]. Korkut et al. [84]
have reported the homogeneous and heterogeneous electrochemical detection of AA
using (2,2,6,6-tetramethylpiperidin-1-yl)oxyl functionalised zinc phthalocyanine
(ZnPc)-modified glassy carbon electrodes. A highly selective ionophore has been
developed based on CoPc nanoparticles for the fabrication of solid-contact poten-
tiometric sensors for AA detection. The electrochemical sensor is found to be active
for AA detection in real samples such as pharmaceutical tablets, injections and orange
juice with satisfactory results [85]. Uric acid (UA) is the end product of purine
derivatives formed during human metabolism. Analysis of UA in body fluids such as
serum and urine yields information on human health, and hence, it is essential to
develop sensors for UA. UA has been detected using metallophthalocyanines as
electrocatalysts in both enzymatic [86] and non-enzymatic [87] methods.

8.4.3 Hydrogen Peroxide (H2O2)

Hydrogen peroxide is an important analyte that plays a significant role in biomedical,

environmental science and industrial environments. The unique pentamer structure
based on CoPc and cobalt tetraphenylporphyrin (Fig. 8.13) immobilised on GCE is
found to be very active for the detection of H2O2 [88]. It is found that the electro-
chemical and electrocatalytic activities of the pentamer depend on the central unit
(CoPc) coupled with the synergistic effects of cobalt phthalocyanine and four cobalt
tetraphenylporphyrin units. The sensitivity of modified electrodes is generally
enhanced by incorporating conducting materials such as CNTs [89], graphene [90]
and gold nanoparticles [91]. The miniaturised device has been reported to be very
useful for the detection of H2O2/glucose. The apparatus consists of centrifuge-based
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 185

Fig. 8.13 Pentamer structure of CoPc-cobalt(II) tetraphenylporphyrin. Figure adapted from

Ref. [88]

microfluidic device coupled with an electrochemical detector. Carbon paste elec-

trodes modified with CoPc and graphene/polyaniline have also been reported [92].
Since H2O2 is a by-product of many enzymatic reactions such as glucose oxidase and
cholesterol oxidase, electrochemical biosensors have been developed based on the
detection of H2O2. Ye and coworkers [93] developed glucose biosensors by coim-
mobilising glucose oxidase with poly-o-aminophenol on an iron phthalocyanine/
CNT composite. Interestingly, the biosensor shows long-term stability up to
120 days with high reproducible response, attributed to the catalytic activity of the
composite and permselectivity of poly-o-aminophenol film. Covalent bonding is
another effective method to attach the enzyme to the electrode surface. Glucose
oxidase is covalently immobilised onto self-assembled monolayers of CoPc-
modified gold electrode using coupling agents such as dimethylaminopropyl car-
bodiimide and N,N′-dicyclohexylcarbodiimide. The biosensor showed good activity
towards the detection of glucose with the low Michaelis–Menten constant of 4.8 mM
[94]. An enzyme-less biosensor is a good alternative for overcoming the problem of
enzyme stability. Ozcan et al. [95] have developed a non-enzymatic glucose sensor
using overoxidised polypyrrole nanofiber electrode modified with CoPc. The
as-prepared electrode showed very good electrocatalytic activity for glucose under
alkaline conditions.
186 D. Mukherjee et al.

8.4.4 Nitric Oxide (NO)

Nitric oxide is an important molecule biosynthesised from L-arginine by a family of

enzymes known as NO synthases, and it acts as a physiological messenger as well
as a cytotoxic agent. Unlike other biomolecules, electrochemical detection of NO is
mostly carried out using nickel phthalocyanine (NiPc)-modified microelectrodes.
The sensitivity of the electrodes greatly depends upon the method of electrode
modification, thickness of the film, pH, hydrophobicity, etc. Modification via
electrodeposition is found to yield better response than the film formed by
dip-coating or drop-coating techniques [96]. Similarly, thin films of NiPc deposited
on carbon microfiber electrode by ac voltammetry reveal better sensitivity for NO
than that of the film prepared by controlled potential electrolysis [97]. Furthermore,
the hydrophobic nature and thin uniform morphology of the coating also contribute
to improve the sensitivity [98]. Inexpensive, disposable and biocompatible
NiPc/Nafion-modified screen-printed electrodes are prepared and used for the
selective and sensitive detection of NO. The screen-printed electrodes show better
tolerance to the more aggressive conditions required for the electrodeposition of
phthalocyanines without affecting their native properties [99]. Electrochemical
in vivo measurements have been carried out for the production and estimation of
NO in human blood platelets and endothelial cells of umbilical cord vein [100,
101], and the results demonstrate the great potential of phthalocyanine-based
electrodes for biological studies.

8.4.5 Amino Acids

Low molecular weight thiols such as cysteine, homocysteine and glutathione play an
important role in metabolism as well as in cellular homeostasis. A comparative study
of various substituted CoPc complexes towards the electrocatalytic oxidation of
cysteine has been reported. Electron-withdrawing functional groups attached to the
CoPc ring are easy to reduce, and CoPc-containing electron-donating groups are easy
to oxidise (Fig. 8.14). It is revealed that the oxidation of cysteine in acidic medium is
closely related to the CoIII/CoII redox couple, whereas in basic medium, the redox
couple CoII/CoI predominates. Furthermore, at pH 8.3, the carboxylic group-
substituted CoPc shows high catalytic activity [102]. The general mechanism of the
oxidation of cysteine on phthalocyanine-modified electrodes is based on the for-
mation of an extra coordinated complex between cysteine sulphhydryl group and the
central metal of the phthalocyanine molecule followed by electron transfer and
oxidation of cysteine [103, 104]. The cysteine biomolecule contains carboxylic, thiol
and amino groups, and their pKa values are 1.7, 8.3 and 10.8, respectively. At neutral
pH, it exists as a zwitterion. Lead phthalocyanine (PbPc) is used as an ionophore for
the development of cysteine potentiometric sensors [105]. The potentiometric
response of the modified electrode is based on the axial coordination of cysteine anion
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 187

Fig. 8.14 Cyclic

voltammograms of CoPc
complex containing different
functional groups such as
a SO3− (sodium salt), b H,
c COOH, d C(CH3)3, e SO3−
(acid salt), f NO2 and g NH2.
The figure is adapted from
Ref. [102]

with the Pb central metal atom. The metallophthalocyanine-modified self-assembled

monolayer electrodes show high selectivity, sensitivity and long-term stability of up
to one month [106–108]. Glutathione exists in both oxidised and reduced forms in
many physiological fluids that play a major role in biological processes such as
catabolism, metabolism and transportation. Screen-printed electrodes with micro-
metric working electrode areas have advantages over macroelectrodes. Improved
signal-to-noise ratio, low Ohmic resistance and reduction of forced convection are
some of the important aspects that are responsible for good performance. Figure 8.15
shows images of macro- and microband working screen-printed electrodes based on
CoPc. The modified microband screen-printed electrodes show about fivefold
enhanced current density as compared to the macroelectrodes [109]. Electrode
fouling is one of the serious issues during electrochemical sensing. This problem can
be effectively addressed using carbon-epoxy composites containing CoPc. The
modified electrodes have been shown to result in superior activity for glutathione
oxidation to that of the graphite paste-modified electrodes [110]. The determination
188 D. Mukherjee et al.

Fig. 8.15 Images of macro- and microband screen-printed electrodes. The images are adapted
from Ref. [109]

of endogenous levels of reduced glutathione has been carried out using bovine ery-
throcyte glutathione peroxidase immobilised on a cobalt phthalocyanine-modified
screen-printed electrode [111].

8.5 Quartz Crystal Sensors

Since the discovery of piezoelectric properties of quartz in 1880, there have been
continuous efforts in using quartz crystal microbalance (QCM) as transducers in a
variety of areas including the detection of volatile organic compounds (VOCs) in a
real-time situation. [112]. This shear mode device consists of thin AT-cut quartz
crystal sandwiched between two metal electrodes. When an electric field is applied
to this device, the frequency shift Df from the nominal resonance frequency f0 of the
oscillating crystal due to the change in the mass Dm arising from the adsorption of
the vapours is described by the Sauerbrey equation in the form:

 Df ðHzÞ
Dm mg=cm2 ¼ ð8:1Þ
2:26  106 f02

0:4 lm-thick thermally evaporated films of tetra-tert-butyl copper phthalocyanine

(ttb-CuPc) on 5MHz AT-cut quartz crystals, 25:4 mm in diameter, have been
employed with gold/chromium to produce room temperature highly sensitive,
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 189

reversible, selective toluene sensors with a view to meeting the specification required
by the World Health Organization (WHO) and European Committee for
Standardization (CEN). The response and recovery characteristics are determined by
successive exposure to 500 ppm toluene for 2 h followed by 3-h-long recovery. As
shown in Fig. 8.16a, the frequency shift remains almost constant at 60 Hz although
the baseline response decreases with time. The high desorption rates implies weak p-p
stacking interactions between phthalocyanine and toluene molecules. Easy cleaning
of the sensing membrane surface between successive exposures under pure air
ensures the repeatable and reversible operation of the sensor. Figure 8.16b shows that
the room temperature sensor response in terms of the frequency shift is found to be
almost linear as the toluene concentration is varied after 3 min of exposure from 35 to
600 ppm in both increasing and decreasing orders. This behaviour is very suitable for
sensor calibration for determining the unknown concentration. It is also to be noted
that the detection limit is found to be 35 ppm which is much lower than the WHO
guideline of 88 ppm. The sensor shows partial selectivity to toluene and xylene
[113]. No significant change in the sensing characteristic is observed for different
central atoms such as zinc, iron and cobalt. However, tert-butyl-substituted

Fig. 8.16 a Ttb-CuPc QCM

sensor signal for consecutive
exposures to 500 ppm of
toluene at room temperature
with recovery under clean air.
b Responses r versus toluene
concentration. Adapted from
Ref. [113]
190 D. Mukherjee et al.

metallophthalocyanine-based sensors exhibit higher sensitivity than unsubstituted

and fluorinated phthalocyanine because of the amorphous and porous nature of
the sensing layer leading to fast diffusion of toluene within the layer volume. QCM
sensor using ttb-CuPc films are reported to remain stable for at least two weeks under
continuous toluene exposure [114]. Tetra- and octa-substituted oxo-titanium
phthalocyanine (TiOPcs)-based QCM have been developed to detect a range of
volatile compounds including triethylamine (Et3N) and DMMP. The sensing mem-
brane is prepared by an air-brush coating system and a repeated exposure to analyte
gas for 20 min and subsequent purging with pure air to recover the baseline. The
responses of these QCM-Et3N and DMMP are as high as 230 and 120 Hz, respec-
tively. The sensitivities for Et3N and DMMP are around 0.05 and 6.9 Hz/ppm,
respectively [115]. The performance of solution-processed thin films of low-melting
ionic copper phthalocyanine derivative [P66614]4[CuPcS4]-based QCM sensors is
monitored for exposure of eleven different organic vapours as shown in Fig. 8.17. The
Sauerbrey equation is found to be valid for [P66614]4[CuPcS4]/organic vapour
interaction within the studied range, giving the linear dependence of the sensor
response on the analyte concentrations. The sensor showed the highest sensitivity and
lowest detection limit towards 3-methy-1-butanol vapour sensing. These phthalo-
cyanines belong to a group of uniform materials based upon organic salts
(GUMBOS), and similar investigations were carried out on the sensing performance
of a porphyrin-based GUMBOS copper (II) meso-tetra (4-carboxyphenyl) porphyrin
[P66614]4[CuTCPP] for the same eleven organic vapours, and the response is not
strictly linear with the vapour concentrations. However, the cross-reactive response
patterns with copper (II) meso-tetra (4-carboxyphenyl) porphyrin are believed to be
very promising for array-based vapour sensing applications. The response and
recovery times are estimated to be 35 and 55 s, respectively [116].
Langmuir–Schäfer (LS) films of octa-substituted copper phthalocyanine
(CuPcR8) as a QCM membrane are studied for determining the total content of four
types of phenols in aqueous solutions. Phenols are highly toxic, carcinogenic and
mutagenic agents; they can induce severe injuries to respiratory tract and dermatitis
and are very dangerous for the environment. Figure 8.18a shows the linear rise in
response as the concentration of 4-bromophenol (4Brph) is increased from 1 to
8 mM irrespective of the thickness of the sensing membrane. The response also
becomes larger with thinner films. This behaviour indicates that surface adsorption
is involved at relatively low concentrations, while diffusion into the active layer
takes place at higher concentrations. The response of the 40-layer-thick CuPcR8
film in Fig. 8.18b remains almost invariant over one hundred days under exposure
to 4 mM 4Brph. Similar measurements have been performed for sensing
4-chlorophenol, 4-bromophenol, 4-iodophenol, 3-nitrophenol and 4-nitrophenol of
different concentrations. Sensors show a high response with satisfactory repeata-
bility and good stability for a wide temporal range (over 100 days). The active
CuPcR8 layer is highly sensitive to para-substituted iodophenol. The 42-min-long
response of the QCM coated with 20 LS layers to a mixture of 2 mM 4-iodophenol,
4.8 mM phenol, 2 mM pentafluorophenol, and 10 mM 4-nitrophenol is shown in
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 191

Fig. 8.17
[P66614]4[CuPcS4]-coated
QCM sensor for eleven
organic vapours with the
range studied in mg/l given in
parenthesis: acetone (0.955–
47.8), acetonitrile (0.380–
39.1), methanol (0.191–38.2),
toluene (0.209–20.9),
nitromethane (0.275–27.5),
chloroform (0.357–35.7),
ethanol (0.191–38.1),
2-propanol (0.190–38.0),
1-propanol (0.184–16.7),
1-butanol (0.196–5.09) and
3-methy-1-butanol (0.196–
2.15)

Fig. 8.18c. The sensing mechanism is not greatly influenced by the acidity of the
analytes [117]. Peripherally tetra- or octa-substituted nickel, zinc, copper and cobalt
phthalocyanines have also been studied as sensitive materials for the detection and
identification of pesticides such as deltamethrin, fenthion, methiocarb, and triadi-
menol in aqueous media using QCM sensors. Pesticide pollution in natural waters
as a result of intensive agriculture and horticulture has a detrimental impact on both
humans and the environment. The performance of the sensors is found to be very
satisfactory in terms of highly stable baselines, short response and recovery times
and complete reversibility. The Pcs are generally found to have strong pesticide
sorption capabilities leading to QCM sensitivities as high as 113 Hz/mg l−1 for
pesticides such as fenthion and triadimenol [118]. The reliable identification and
192 D. Mukherjee et al.

Fig. 8.18 a Effect of the thickness of CuPcR8 LS films on the response of the QCM device to
4Brph injections, b time evolution of the response to 5 mM 4Brph of the QCM crystal covered
with CuPcR8 by 40 LS runs and c QCM response of a sensor coated with 20 LS layers to a mixture
consisting of 5 mM phenol, 2.5 mM 4-chlorophenol, and 8 mM 4-nitrophenol. Adapted from
Ref. [117]

assessment of soil and water contamination with explosive substances such as

2,4,6-trinitrotoluene (TNT) and 1,3,5-trinitro-1,3,5-triazinane (RDX) are necessary
to prepare and support decontamination and remediation activities. Octa-substituted
cobalt phthalocyanine-based QCM sensors are able to register TNT with high
sensitivity of 100 Hz/ppm and detection limit of 40 ppb. The central metal atoms
have an effect on explosive identification. For example, cobalt phthalocyanine
shows relatively strong affinity for the nitro aromatics including TNT. Similarly,
octa-substituted copper and zinc phthalocyanines show preference towards RDX.
Similar difference in activity is observed between octa and thio substitutions of zinc
phthalocyanine [119]. A group of peripherally substituted Pcs with polyalkoxy and
fluorinated alkyl and polyalkoxy substituents and with four different central ions
has been employed in the fabrication of 5 MHz QCM-based sensors for low con-
centrations of pentachlorophenol (PCP) and phenol in water. An accurate detection
of PCP in the drinking water is required according to the European Union policy
because of high toxicity and slow biodegradation. The response time is less than
90 s for intermittent exposure of PCP with concentration varying between 0.02 and
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 193

2 mg/l to an aqueous solution alternating with PCP-free water. The baseline drift
remains constant at 0.5 Hz/h over 80-min-long measurements. A value of
200 Hz/mg/l for sensitivity is obtained for PCP with QCMs coated with
polyalkoxy-substituted CuPcs with the detection limit of 0.01 mg/l. Phenol (0.02–
2 mg/l), tetrachloroethylene (0.3–32 mg/l) and p-xylene (0.2–17 mg/l) have also
been studied as representatives of chlorocarbons and aromatic hydrocarbons, and
the selectivity is found to be significantly dependent upon the substituents and
central ions from the principal component analysis [120].

8.6 Surface Acoustic Wave Sensors

The acoustic waves propagating along the surface of a piezoelectric substrate such
as lithium niobate (LiNbO3) and lithium tantalate (LiTaO3) crystals are generated
due to mechanical deformation under the application of an electric field, and the
surface modification will cause changes in the velocity and the amplitude of the
propagating acoustic waves. Two types of device configurations have commonly
been used for practical implementation to record sensing parameters. In the delay
line design of surface acoustic wave (SAW) sensors in Fig. 8.19a, the crystal
surface space between the input and the output interdigitated transducers (IDTs) is
coated with a sensing layer such as phthalocyanine film. Thin silicon films at the
crystal edges absorb the acoustic waves to avoid the subsequent interference. The
IDTs of two-port SAW resonators in Fig. 8.19b are coated with sensing layers, and
the gratings are used at the edges to reflect the advancing acoustic wave. The SAW
devices operating at relatively high frequencies are considered to be suitable for
continuous monitoring of VOC in terms of high sensitivity and low detection limits
and selectivity [121]. A 3.8-mm-wide LiNbO3 interspace between two IDTs in the
delay line SAW devices has been covered with 160-nm-thick sublimed lead
phthalocyanine films (PbPc) for sensing 10 ppm NO2 gas in the flow of N2 at a rate
of 1 l/min. The choice of LiNbO3 as a piezoelectric substrate is made because of its
high coupling factor with a view to attain large attenuation. Figure 8.19c shows
response of a differential SAW sensor at 80 °C over 1400 s in terms of the
time-dependent ppm fractional shift (Δf/f0) as the gas flow over the PbPc sensing
layer is changed from 10 ppm NO2 in the N2 flow and then back to pure N2. f0 is the
baseline frequency of 110 MHz of LiNbO3 crystal. The changes in the wave
velocity occur due to the mass loading and the conductivity increase of the PbPc
overlayer due to oxidation by NO2 interaction. This differential SAW system is able
to distinguish the effect of the conductivity changes from that due to the mass
loading. The temperature fluctuations are found to have produced a very small
frequency shift of 23 Hz. Also, interactions of the inherent electric potential wave
with mobile holes in the PbPc films cause both a propagation velocity shift and
attenuation increase. This SAW device is found to be 1000 times more sensitive
than a device based upon mass loading in terms of high detectability of 1016
molecules/cm3 on the PbPc film [122]. However, this oscillator circuit suffers from
194 D. Mukherjee et al.

Fig. 8.19 a Delay line and b resonator SAW devices: typical designs (after Ref. [121]).
c Transient ppm fractional shift (Δf/f0) after Ref. [122]. d SAW sensor response (solid line) to
atmospheric NO2 together with NO2 concentration determined simultaneously with a white cell
(broken line). Adapted from Ref. [124]
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 195

additional 6 dB attenuation of the SAW energy launched in the opposite direction.

To overcome this problem, a separate IDT is added to determine the frequency of
this oscillator component. Using 160-nm-thick sublimed PbPc films as previously
reported, transient responses of this modified SAW devices are recorded at the
working frequency of 35.5 MHz to periodic exposure of 5, 10 and 15 ppm NO2
nearly every 250 s with in-between N2 surge. The sensitivity and response are
similar to earlier reported values [123]. 15-nm-thick sublimed copper phthalocya-
nine films have been used as a NO2 sensing layer in LiNbO3 SAW sensors oper-
ating with a centre frequency of 360 MHz and a delay time of 2.2 µs. Figure 8.19d
compares the atmospheric NO2 sensing performance of these SAW devices with
that of long-path UV–Visible absorption spectroscopy over 24 h. The detection
limit of phthalocyanine-based SAW sensors is found to be 1 ppb of NO2 gas in
atmosphere implying their usefulness of the air quality surveillance. There exists a
good agreement between the SAW and optical results except for the end of
observation period due to comparatively large optical signal-to-noise ratio and
simultaneous reduction in SAW sensitivity [124]. The effect of operating temper-
ature on NO2 detection has been investigated in dry air using 83–310-nm-thick
sublimed PbPc films as sensing layers in SAW devices in the medium temperature
between 27 and 95 °C. The differential frequency Δf is recorded for an 83-nm-thick
film after 480-s-long exposure of NO2 gas at room temperature. The concentration
dependence of Δf is found to be linear between 30 and 90 ppm, giving a value of
1.73 kHz/ppm for the sensitivity. The increase in NO2 sensitivity was observed for
the concentration range between 1 and 4 ppm with the rise of temperature from
1.19 kHz/ppm at 27 °C to 1.99 kHz/ppm at 95 °C. The temperature cycling
between 70 and 80 °C produces limited reversibility for the concentration as low as
640 ppb. 310-nm-thick PbPc-based SAW sensors operating at four temperatures
between 75 and 95 °C were also exposed to 2, 4 and 40 ppm NO2 gases for the
similar time period, and the linear frequency-dependent sensitivity is also observed.
83-nm-thick PbPc layers show a frequency shift of 3 kHz for a concentration as low
as 640 ppb within about 300 s at 70 °C. A fast regeneration of the 83-nm-thick
sensing layer may be achieved by switching on and off the system during annealing
at a stable temperature of 80 °C in air [125].
Films of newly synthesised 2,9,16,23-tetrakis [4-(hexafluoroisopropanol)-anili-
nomethyl]phthalocyaninato zinc (ZnPcF24N4O4) molecules were deposited on the
delay line of a 300 MHz SAW device, and kinetic response of the sensor was
monitored as it was exposed to sarin (GB) concentrations varying from 0.6 to 43.4
mgm−3. The frequency shift increases linearly with the rise of GB concentrations,
producing a high value of 205.8 Hz/mg m−3 for the detection sensitivity. The time
required for the frequency to reach 80 % of the maximum is 60 s since the GB
vapour being switched on after the exposure of fresh air. This SAW sensor is,
196 D. Mukherjee et al.

therefore, fairly fast and reversible response to GB presence with a very small
baseline drift. The selectivity of this ZnPcF24N4O4-based SAW sensor was
examined by recording the real-time normalised frequency response of different
VOCs such as acetone, ethanol, chloroform, toluene, hexane and benzene, and the
sensitivity of GB detection is at least five times larger than tested VOCs [126].
Using tandem CoPc nanopillars as a sensing layer, the dual-line SAW devices are
employed to measure the phase shift on exposure of DA and AA. It is found to be
selective towards the DA detection with high sensitivity of 1.6°/nM and low
detection limit of 0.1 nM. The detection limit of 0.1 nM for AA is considered to be
very high. The dative bonding of DA and ionic interaction of AA with the CoPc
films are found from the density function theatrical (DFT) calculations to be
responsible for the difference in sensing detection limits. Also, the calculations
show that the liquid-phase interaction energy of 81 kJ mol−1 for DA/CoPc com-
plexes is much larger than that of 38 kJ mol−1 for AA/CoPc, accounting for more
selective response of CoPc nanopillars towards DA than AA [127].
Hydrogen sensors are of great importance for health and safety since the con-
centrations of hydrogen (H2) at more than 4 % can cause explosion. Flammable H2
is produced during normal charging of domestic batteries, and their overcharging is
potentially dangerous causing fire and explosions. Faulty oil-filled gas power plants
generate H2, early detection of which is critical for their safe maintenance. Cars and
buses can be run using hydrogen fuel. Hydrogen is also used as therapeutic medical
gas. Palladium (Pd) thin films are known to be very sensitive to the hydrogen
presence, and therefore, Pd/Pc bilayers have been explored as a potential H2-
sensitive delay line coatings in SAW sensors with well-resolved interaction speed
(Δf/Δt) as an indicator. The thickness of H2Pc layer is found to be important for
achieving the desired degree of resolution in the interaction speed. The speed (Δf/
Δt) of the bilayer containing 20-nm-thick top Pd layer is found to vary from 7.5
Hz s−1 for 2.5–42.8 % for 42.5 % H2 in air as the thickness of the top H2Pc layer is
decreased from 160 to 80 nm. As the bilayer is exposed to hydrogen gas, the
dissociation of H2 molecules to atomic occurs on the exposed Pd surface of the
bilayer. Gradual diffusion of atomic hydrogen deep into the Pd thin films becomes
adsorbed at the interface between Pd and Pc layers, causing the change in the
surface conductivity of Pc films. This results in considerable fluctuations of the
SAW velocity leading to a detectable modification of the measuring frequency
[128]. As shown in Fig. 8.20a, the bilayer consisting of a 20-nm-thick evaporated
Pd film on the top surface of a 110-nm-thick evaporated H2Pc film was periodically
exposed at room temperature to 0.5–4 % H2 gas in synthetic air over 6 h. The
maximum value of Δf is found to be −1.75 Hz independent of concentrations
although the baseline tends to decrease over the time. However, the rate of inter-
action is shown in Fig. 8.20b to be linear with increasing concentration of H2 gas,
giving the values of 2.1 and 21.2 Hz s−1 for 1.5 and 4 % H2 gas concentrations,
respectively. The measurements were repeated on the bilayer with the same con-
figurations and dimensions in N2 flow at the same rate of 1000 ml/min, and it is
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 197

evident from Fig. 8.20b that the interaction is slow in nitrogen environment. In
general, the speed (Δf/Δt) increases with H2 concentrations, giving the values of 6.0
and 7.6 Hz s−1 % for the increase in speed relative to concentrations in air and N2,
respectively [129]. Figure 8.20c illustrates the impact of thickness of PVD copper
phthalocyanine (CuPc) films forming bilayers with 18-nm-thick Pd films for H2
sensing in dry air. The most pronounced effect was observed for 200-nm-thick
CuPc films, but the variation of the frequency shift with the H2 concentration is
nonlinear between 0.5 and 4 % at 37–38 °C. However, the acoustoelectric
parameter for the bilayer with 100-nm-thick CuPc film is estimated to 5.5  10−2
from the knowledge of surface conductivity, mean dielectric constant and unper-
turbed SAW velocity, implying relatively strong H2 interactions with this film. The
response is found to be fast nearly 4 s [130].

8.7 Optical Sensors

Optical transduction methods offer many advantages over other competing tech-
niques in terms of fast speed, high sensitivity (often ppb level), immunity to
interference, precision discrimination between very similar analytes and nearly
perfect fingerprinting of extremely similar mixtures over a wide range of both liquid
and gaseous analytes. Metal phthalocyanine (MPc) molecules are optically active
organic materials because their optical spectra as mentioned before are tunable by
changing the central metal atoms and peripheral substituents. An ordinary UV–
visible spectrophotometer can be easily adapted for optical electronic nose-type
measurement for the films on transparent substrates [131].

8.7.1 Fibre-Optic Sensors

Standard single-mode telecommunication fibres are used in the development of

optical VOC sensors excited at the standard wavelength of 1310 nm. As shown in
Fig. 8.21a, the end of the test fibre is coated with LB films of specially synthesised
lutetium bisphthalocyanine (LuPc2). The optical power from port 1 is split into
ports 3 and 4 for guiding power reflected at the fibre/LuPc2 interface and through
matching gel with no reflection, respectively. The attenuation in the optical prop-
agation is recorded by a photodetector as a consequence of interaction between
LuPc2 film and the analytic VOC agent in the test chamber. Figure 8.21b presents
the attenuation of the reflected power as the vaporisation of liquid acetic acid in the
test chamber begins to take place. The attenuation becomes stable to a decrease of
9.5 dB of reflected optical power within 5 h, indicating that the acetic acid may be
198 D. Mukherjee et al.
 8 Phthalocyanines as Sensitive Materials for Chemical Sensors 199

b Fig. 8.20 a Interaction of H2 with synthetic air at six hydrogen concentrations from 0.5 to 4 % in
air at room temperature, b comparison of interaction speeds [S = (Δf⁄Δt)] of H2 with Pd/H2Pc
bilayer in air and nitrogen. Adapted from Ref. [129], c comparison of the frequency shifts for
Pd/CuPc bilayer of the investigated nanostructures at the similar operating temperatures  37–
38 °C. Adapted from Ref. [130]

completely vaporised. The recovery time is 20 min, reaching the initial optical
power without the presence of acetic acid. The acid takes longer time to reach
stability with higher concentration. Similar experiments are performed for ethanol,
hexanal, and n-butyl acetate, and the effect is more pronounced for acetic acid and
ethanol than hexanal and n-butyl acetate gases [132].
Chlorophenol compounds are essential industrial ingredients for the manufac-
turing of agricultural herbicides and pesticides and wood preservatives. However,
their uncontrolled use is a major concern because of their propensity to accumu-
lation in animal bodies and plants. A fibre-optic oxygen sensor has been designed
by attaching its sensing end to a cellulose acetate (CA) membrane with immobilised
fluorescent ruthenium complexes in order to monitor oxygen consumption in the
catalytic oxidation reaction between 2-chlorophenol (2-CP) and tetranitro iron
(II) phthalocyanine (TNFe(II) Pc) in water. The phase modulation technique has
been employed to record the phase shift of weak optical signal corresponding to the
aerobic oxidation. The ruthenium complex is excited at 417 nm wavelength, and
the dynamic oxygen quenching of the luminescence is recorded by a photodiode.
The peak of fluorescence spectra of the membrane appears at 660 nm implying the
successful immobilisation of ruthenium complex in the CA matrix. The UV–Visible
spectra show the formation of pink dye within five minutes of the catalytic oxi-
dation process involving 30 mg of TNFe(II)Pc. These results suggest that the
response time can be shortened by using an increased amount of TNFe(II) Pc
catalyst. Figure 8.21c shows the linear dependence of the phase delay on 2-CP
concentration varying between 3.0  10−7 M and 7.0  10−6 M, and this beha-
viour may be interpreted in terms of a modified Stern–Volmer equation. The sen-
sitivity of 37100 M−1 is estimated from the slope. The value of 8.0  10−7 M for
the detection limit of 2-CP in water is regarded to be encouraging in light of
regulatory limits of volatile phenol in agricultural water in China [133].
Multimode fibres are coated with thermally evaporated 140-nm-thick copper,
lead and samarium (SmPc) phthalocyanine films over the uncladded portion with a
40 cm exposure length. When the sensor is interrogated at 670 nm, a
time-dependent decrease in the output power from the fibre is observed as a con-
sequence of increased evanescent wave absorption under different NO2 concen-
trations. The rate of the output decay increases with a rise in NO2 concentrations,
but it is observed that the responses of CuPc-, PbPc- and SmPc-based sensors
follow a similar pattern on exposure to 2 mbar of NO2 [134].
200 D. Mukherjee et al.

Fig. 8.21 a LuPc2-based fibre-optic sensor interrogation arrangement. b Response of the sensor
to the presence of liquid acetic acid of 4, 11, 22, 44, 66 and 88 mmol/l. Aadapted from Ref. [132].
c The calibration curve obtained from the phase delay responding to 2-CP concentration in the
range of 3.0  10−6 to 7.0  10−5 M. Adapted from Ref. [133]
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 201

8.7.2 UV–Visible Absorption Sensors

As mentioned earlier, an ordinary UV–Visible spectrophotometer can be easily

adapted for optical electronic nose-type measurement for the films on transparent
substrates [135]. The disappearance of Davydov splitting from the UV–Vis
absorption spectra of 20-layer-thick LB films of substituted amphiphilic copper
molecules was observed when the films were exposed to 100 ppm NO2. The local
interaction between the central metal atoms and NO2 distorted the herringbone
structure. Response and recovery times are estimated to be 200 and 405 s,
respectively, from an exponential time dependence of film absorbance. Because of
the slow recovery, these molecules are suitable for fabricating a disposable
single-shot NO2 sensor [136]. An electronic optical nose consists of four quartz
substrates covered with spin-coated films of four peripherally substituted MPc
molecules, and its performance has been investigated for selectively sensing five
VOCs of interest for food analysis, namely diethylamine, dibutylamine,
tert-butylamine and 2-butanone, and acetic acid. Two different data matrices con-
taining the normalised sensor responses to all the VOCs have been built both for the
data related to two Q-bands at 640 and 680 nm. The principal component analysis
technique is employed for the synthesis of the data obtained from the 5 cycle of
VOC sensing tests. 98 and 100 % of total variance within the data collected in the
first two and three principal components for the band at 640 nm is a good indicator
for discriminating between four VOCs [137]. Lead (Pb), zinc (Zn), copper (Cu) and
cadmium (Cd) can often seriously deteriorate water quality. A bilayer sensing
membrane of spun acetylcholine esterase (AChE) immobilised in calcium alginate
beads and 20-nm-thick drop-cast film of sulphonated copper (II) phthalocyanine
(CuTsPc) on a quartz substrate was exposed for 15 min to water contaminated with
cadmium (Cd) with four different concentrations ranging from 1 to 100 ppm. The
inhibition of catalytic reaction between acetylcholine chloride and enzyme due to
Cd contaminants is monitored by recording changes in spectra of drop-cast CuTsPc
as an indicator. Values of wavelength shifts of Q-bands due to Cd2+ ion exposure
were found to be linearly dependent upon the concentrations, and the sensitivity is
estimated to be 1.6 %/ppm from the slope of linear variation normalised by the
peak positions of unexposed samples. The detection limit of cadmium content in
water was found to be 1 ppm. The membrane is insoluble in the test solution, and
the immobilisation process is also irreversible. The bilayer sensing membrane is,
therefore, suitable for use in a disposable sensor. The method is cost-effective,
easily adaptable to practical implementation and less disruptive to the enzymatic
protein than other chemical attachments [138].
Bovine serum albumin (BSA), the most abundant protein in serum, has several
functions, for example regulation of colloidal osmotic pressure of blood and
transportation of fatty acids. The binding between BSA and phthalocyanine
derivatives has already been studied by the measurements of fluorescence emission.
The interaction between bovine serum (BSA) and octa-hexylthio-substituted Cu(II)
phthalocyanine derivative [(C6S)8PcCu] molecules has recently been reported, and
202 D. Mukherjee et al.

it is found that the mechanism involved a range of non-covalent interactions

including hydrogen bonding especially with the nitrogen in the phthalocyanine ring,
ionic interactions between negatively charged residues and the core metal ion as
well as van der Waals interactions. Heat treatment produced changes in the ori-
entation of phthalocyanine stacks, making favourable contribution to the protein
adsorption and surface coverage profile [139]. Fluorescent emission spectra
between 290 and 500 nm have been recorded for the aqueous solution containing
newly synthesised water-soluble zwitterionic and cationic zinc phthalocyanine
derivatives and BSA for the PDT applications. A systematic decrease in the
emission intensity is observed because of fluorescence quenching with a large
constant in the order of 1013 M−1s−1. The near unity value for number of the
binding sites on BSA implies that Pc-BSA adducts are formed in the ratio of 1:1.
These Pc molecules are suitable for use as photosensitisers for PDT [140].

8.7.2.1 Chemichromic Sensors for Health Care

Approximately 100-nm-thick spin-coated films of a newly synthesised bis[octakis

(octyl)phthalocyaninato] lutetium(III) complex (C8LuPc2) on ultrasonically cleaned
glass substrates exhibit pronounced chemichromic behaviour with potential appli-
cations in health care. Electron transfer takes place through the oxidation of reduced
nicotinamide adenine dinucleotide (NADH) with the regeneration of the corre-
sponding oxidised form (NAD+) producing sufficient adenosine triphosphate
enzymes for cellular energy supply to sustain normal brain functions. NAD+/
NADH ratio in the range of 1–0.1 determines the normal metabolic state of the
body, and the increase in free NAD+/NADH ratio arising from calorie restriction
leads to the reduction of several age-related diseases, such as cancer, neurode-
generative diseases and diabetes. A deficiency of L-ascorbic acid (vitamin C) may
lead to scurvy, osteoporosis, bone homeostasis, high blood pressure and gall
bladder disease. A simple optical method can be developed to detect NADH and
vitamin C by studying the recovery kinetics of the absorption spectra as the oxi-
dised films are neutralised by the biological cofactors in an aqueous solution
wetting the surface of an ITO-coated glass substrate modified with spin-coated
electrochromic films. As shown in Fig. 8.22i, the UV–Vis spectrum of oxidised
C8LuPc2 displays bathochromic shift of Q-band from 618 to 778, 60 nm with
respect to the neutral film due to the removal of the electrons in the antibonding
HOMO resulting in relatively short Pc-Pc inter-ring distance. The broadbands of the
oxidised films in the range 540–640 nm are indicative of radical cations, possibly
owing to the presence of charge-transfer Br2-C8LuPc2 adducts. The spectra have
been recorded for C8LuPc2 samples which had been left in air for 3 months to
examine the effect of ageing. As shown in Fig. 8.22ii, the two Q-bands of the
3-month-old neutral spin-coated film spectrum exhibited two different types of
wavelength shifts: a small redshift from 718 to 736 nm and a blueshift from 653 to
643 nm. However, the solution of C8LuPc2 in chloroform appears not to suffer
from any aging effect with the Q-band peaks at almost the same positions of 714
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 203

and 634 nm [141]. It is possible that some relaxation of the molecular packing
distortion occurs over time. When this film is oxidised by exposure to bromine
vapour, the main Q-band peak of the oxidised film underwent a blueshift to 772
from 736 nm. However, this is smaller than that exhibited by a fresh film upon
oxidation. The bromine-oxidised C8LuPc2+ film remained stable for at least 3 h,
long enough to examine its interaction with freshly prepared biological cofactor
NADH in 1.5 M LiClO4 aqueous solution. Figure 8.23i shows the spectral changes
of C8LuPc2 film in the presence of 3 mM NADH in the 1.5 M LiClO4 aqueous
solution as the reduction takes place. The results of similar kinetic measurements of
the oxidised film in the presence of 3.5 mM vitamin C in 1.5 M LiClO4 aqueous
solution are presented in Fig. 8.23ii. The presence of an isosbestic point at 763 nm
is observed during reduction over 90 min for both cases. The Q-band at 778 nm

Fig. 8.22 Electronic absorption spectra of i freshly prepared and ii three-month-old C8LuPc2
films: (a) neutral film (broken line), (b) Br2-oxidised film (dotted line) and (c) neutral species in
chloroform solution (solid line). Inset In Figure (i) oxidation for 3 s (solid line) and oxidation for
7 min (dashed line)
204 D. Mukherjee et al.

progressively decreased in intensity accompanied by an increase intensity of the

718 nm band. The logarithmic-linear plot of ln ðAt  A1 Þ against time in Fig. 8.24a
shows a first-order kinetic reaction and exponential dependence of the instanta-
neous value At of absorbance at 778 nm on time t in the form of

ðAt  A1 Þ
¼ expðktÞ ð8:2Þ
ðA0  A1 Þ

where A0 and A1 are the absorbance of the completely oxidised and reduced films,
respectively. k is the rate constant. The Lewis basicity of NADH and vitamin C

Fig. 8.23 Electronic absorption spectral changes as a function of time (indicated by arrows) of
C8LuPc2 in i 3 mM NADH and ii 3.5 mM vitamin C in a 1.5 M LiClO4 aqueous solution. Data
recorded after (a) 2, (b) 7, (c) 10 and (d) 90 min of adsorption
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 205

Fig. 8.24 a Plot of ln(At–

A∞) versus time for 3 mM
NADH (dashed line) and
3.5 mM vitamin C (solid line)
in a 1.5 M LiClO4 aqueous
solution. b Dependence of
oxidation half-life on NADH
concentration for freshly
prepared (solid line) and
three-month-old films (broken
line)

became dominant in reducing the C8LuPc2 film to the neutral state over the ability
of LiClO4 to counteract the neutralising effect of water molecules. Values of 4.3 and
4.8 min for the half-life t1=2 are determined from the slope corresponding to the
neutralisation by NADH and vitamin C, respectively, using the formula:

0:693
t1=2 ¼ ð8:3Þ
k obs

As shown in Fig. 8.24b, values of the half-life of the oxidation of C8LuPc+2 are
found to decrease with the concentration of NADH. The NADH reduction of the
3-month-old oxidised film is consistently slower than a similarly oxidised fresh
film, giving a percentage change from 94 % for 0.05 M to 6 % for 1 mM con-
centration of NADH. The detection limit was found to be 0.05 mM for NADH and
vitamin C both [142]. The study showed satisfactory response times and linear
concentration ranges up to two orders of magnitude: 0.05–3 mM for NADH and
0.03–3.48 mM for vitamin C. The method of sensing NADH and vitamin C works
well up to a concentration of 10−5 M, and the complete recovery of the neutral
bisphthalocyanine from the oxidised film via NADH reduction leads to the reali-
sation of developing reusable membranes. The reduction rate was greatly increased
when redox biomolecules were added to the aqueous medium of water and LiClO4.
The reduction of the oxidised film in the presence of NADH and vitamin C was
206 D. Mukherjee et al.

described in terms of first-order kinetics. The detection of NADH and vitamin C

was found to be reproducible. NADH reduction activity toward functionalised
spin-coated films of electrochromic differently substituted bis[octakis(hexylthio)
phthalocyaninato] lutetium(III) [[(C6H13S)8Pc]2Lu] also shows the first-order
reduction kinetics, giving a value of 10.4 min for half-life. A lower NADH
detection limit of 1  10−6 M is obtained from in situ UV–Vis absorption mea-
surements [143]. The optical density at 778 nm peak of the reversible UV spectra of
oxidised bis[octakis(hexylthio)phthalocyaninato] dysprosium(III) [(C6H13S)8Pc]2
Dy] on glass substrates monotonically increases from 0.110 to 0.144 as it is reduced
by increasing the NADH concentration from 10−4 to 5  10−3 M [144]. Neutral
and oxidised forms of the films of all three molecules C8LuPc2, (C6H13S)8Pc]2Lu
and [(C6H13S)8Pc]2Dy have also been identified using Raman spectra.
Electrochromic bisphthalocyanine films are found to be good sensing materials for
reusable biological cofactor monitors at the point of care.

8.7.3 Ellipsometry—An Optical Gas Sensor

Ellipsometry as shown in Fig. 8.25a is regarded as being a powerful tool for

monitoring changes in the thickness, the refractive index or the microstructure of a
sensing layer by measuring ratio of polarisation states after and before reflection.
Sub-nanometre resolution can be achieved in bioaffinity-based sensing and ppm
sensitivity in gas sensing. The ratio of the Fresnel reflection coefficients Rp and RS
corresponding to the light polarised in the p- and s-directions can be described in
terms of measured ellipsometric parameters w and D

Rp
¼ tan(wÞtanðiDÞ ð8:4Þ
Rs

where tanðwÞ is the ratio of p- to s-components and D is the phase shift between the
p- and s-components. Equation (8.4) depends upon the thickness and the refractive
index of the sensing film [145].
Spectroscopic ellipsometric studies on 9-nm-thick spin-coated thin films of
tetrasulphonated copper phthalocyanine (CuTSPc) on gold substrates have been
made at 256 equally spaced photon energies in the wavelength range between 275
and 825 nm in order to demonstrate the NO2/N204 (NOx) sensitivity. The angle of
incidence used was 68°. The kinetic of W was recorded for repetitive 2700 ppm
NOx exposure with an interval of 30 min, and slow decay over 6 min is believed to

Fig. 8.25 a Experimental set-up for ellipsometric measurements. b W(k) and D(k) spectra of c
ZnPcR8-coated Cr/Au films in water (1) fresh, injection of pesticide solution of (2) 0.8 lg/l and (3)
5 lg/l; after flushing (4). The enlarged section of D(k) spectrum shown above. c Changes in the
layer thickness caused by the adsorption of the pentachlorophenol of concentration varying from
0.8 to 50 lg/l. Adapted from Ref. [147]
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 207
208 D. Mukherjee et al.

be caused by the presence of residual gas in the gas exposing system. The detection
limit was found to be approximately 100 ppm [146].
In situ optical response of octa-substituted zinc phthalocyanine (ZnPcR8) thin
films on gold-chromium-coated glass substrates to the presence of pen-
tachlorophenol (PCP) in water has been investigated using the total internal
reflection ellipsometric technique within the wavelength ðkÞ range of 370–100 nm.
Figure 8.25b presents experimentally measured wðkÞ and DðkÞ, showing a sharp
change in phase angle D from +270° to −90° near the plasmon frequency. This
implies that the phase angle is comparatively more sensitive to the changes in optical
parameters of the reflection system. Using the dedicated software, experimental data
have been to the four-layer model and change in the thickness of the sensing ph-
thalocyanine layer is shown in Fig. 8.25c. Fast changes in thickness are observed for
the PCP concentration range between 0.8 and 8 lg/l. However, the film response
becomes saturated at higher concentrations because of the non-availability of suf-
ficient binding sites. A clear recovery occurs for concentrations up to 6 lg/l when
the measurement cell containing PCP solution is flushed with deionised water [147].
In recent years, hybrid materials of phthalocyanine and acidified single-walled
carbon nanotubes (SWCNTs) are produced as sensing materials for use in ellipso-
metric measurements to detect 0.5–10 lg/l PCP. The phase shift DðkÞ spectrum is
two times larger than the one for pristine phthalocyanine. This may be attributed to
the formation of MPc/SWCNT networks through the attachment of Pc molecules to
the surface of SWCNT as revealed by the atomic force images. The QCM sensitivity
using hybrid materials was found to be 1.52  10−2 RIU/(lg/l) from the normalised
changes in refractive index. This value is at least one order of magnitude higher than
similarly estimated value of 2.18  10−3 for pristine phthalocyanine. The detection
limit was estimated to be 0.5 lg/l from the sensor characteristics of hybrid materials.
The interaction between oxygen-containing aromatic PCP molecules and
high-density delocalised p-electrons on SWCNT surfaces is responsible for
enhanced performance [148]. Further work has been carried out on these hybrid
layers for the detection of 2-chlorophenol, diuron and simazine with concentrations
ranging from 1 to 25 lg/l in water. The ellipsometric sensors using hybrid films
exhibit sensitivity in the order of 4.0  10−4 µg/l for PCP which is higher than other
analytes by a factor of 4. The lower detection limit is estimated to be 690 ng/L, while
this value is 1.34 µg/l for simazine. Complete recovery may be achieved for PCP on
flushing the system with deionised water. However, similar recovery is not possible
for simazine and diuron contamination [149].

8.7.4 Surface Plasmon Resonance (SPR) Technique

The surface plasmon resonance (SPR) technique was first introduced for optical
characterisation of organised thin films on a metal surface. However, considerable
efforts have been spent at the same time on commercial exploitation as a label-free,
highly sensitive, high-throughput and non-invasive method for real-time monitoring
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 209

of biointeractions at molecular levels [150]. The technique is based on the exci-

tation of surface electromagnetic waves of transverse magnetic (TM) modes trav-
elling along the interface between a metal and a dielectric medium. Using a
semicylindrical prism of high refractive index np in Kretschmannʼs configuration, as
shown in Fig. 8.26i, these waves are excited by a p-polarised light of wavelength k
via the evanescent field. The propagation constant ksp of surface plasmons is
dependent upon the effective dielectric constants 2m and 2i of the metal and the
interfacing medium in contact with the metal, respectively, in the form of:
 rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2p 2m 2i
ksp ¼ ð8:5Þ
k 2m þ 2i

where k is the interrogating incident wavelength. The propagation constant kx of

polarised incident monochromatic beam parallel to the interface between the metal
and the glass can be written in the form of:
 
2p
kx ¼ np sin h ð8:6Þ
k

For a sufficiently thin metal film, the light incident on the glass/metal interface
will also penetrate into the metal/surrounding medium. Under these conditions, the
resonance occurs for a given angle h, when Eqs. (8.5) and (8.6) match each other
[151].
The SPR response of 22-nm-thick physical vapour-deposited (PVD) films of
copper hexadecafluorophthalocyanine (CuPcF16) on exposure to ammonia NH3 is
shown in Fig. 8.26ii. Adsorption of the ammonia gas results in decreases of both
refractive index n and extinction coefficient k of the CuPcF16 overlayer by 4 and
13 %, respectively. Figure 8.26iii shows the transient SPR response on exposure of
the same CuPcF16 overlayer to 100 ppm NH3 for 2 min followed by the injection of
dry air for a further 2-min period. The film is then exposed to 200 ppm NH3 for the
next cycle. The angle of incidence remains fixed at 44.2° during the whole mea-
surements. The SPR response is completely reversible with response and recovery
times of 2 min. Unsubstituted CuPc is practically insensitive to NH3 because of
strong electron-withdrawing fluorine substituents. Substituted CuPcF16 compounds
are sensitive to reducing gases such as NH3 [152]. Spun films of nickel (II) tetrakis
(4-cumylphenoxy) phthalocyanine molecules are introduced to a sequence of
exposure at 20-min interval to ozone (O3) of 2 ppm concentration. The resonance
minimum moves to an increasingly larger angle after each exposure to ozone. The
half-width became narrower at the same time, and this is believed to be the conse-
quence of oxidation of the phthalocyanine ring by ozone, which results in bleaching
of the dye [153]. Reported results of preliminary SPR experiments indicate that single
LB overlayers of substituted copper tetrakis- (3,3-dimethyl-1-neopentoxycarbonyl)
phthalocyanine (CuPcNC) are sensitive to the presence of 50 ppm toluene vapour in
N2 environment. The maximum increase of photodetector signal is found to be about
4.2 % as the vapour concentration is periodically varied every 20 min from 50 to
210 D. Mukherjee et al.

Fig. 8.26 i Kretschmann-configured optical coupling set-up, ii SPR curves for (a) gold,
(b) gold/CuPcF16 overlayer and (c) gold/CuPcF16 overlayer exposed to NH3, iii kinetics of the
SPR response of CuPcF16 film to NH3 gas at 100 and 200 ppm. Measurements are made at a fixed
angle of h* = 44.2°. Adapted from Ref. [152]
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 211

200 pm over 160 min. The recovery is partial at low concentrations up to 100 ppm
when the toluene is turned off. The effect is, however, irreversible at high concen-
trations, possibly the interaction with the 50-nm-thick Ag layers through the pene-
tration of the CuPcNC overlayer. The toluene sensing properties of tetrakis-
(3,3-dimethyl-l-butoxy-carbonyl) substituents are similar, but the responses of these
derivatives with nickel as central atoms are relatively weaker [154].
Films of different phthalocyanine derivatives have been employed in the SPR
experiments as active add-on layers in order to investigate the interaction mecha-
nism with NO2 of varying concentrations. Figure 8.27a, b presents the resonance
depth and shift in incident angles as spun films of 21-crown-7 and 15-crown-5
metal-free phthalocyanines remain exposed to 100 ppm NO2 gas over 15 min. The
rise in the resonance depth is recorded to be 6.6 and 4.2 % for 21-crown-7 and
15-crown-5 H2Pc, respectively, presumably associated with the decrease in imag-
inary component of optical permittivity. The shift in resonance angles to smaller
values for both molecules may be attributed to slow diffusion of NO2 into the bulk
of the film. A series of approximately 10-min exposure to 1 ppm NO2, with
reversals in clean air for approximately 1 h carried out both in dry conditions and in
50 % relative humidity. It is evident from Fig. 8.27c that the response of
15-crown-5 H2Pc is significantly larger and more reversible in humid environment
than that in dry air [155]. Adsorption of NO2 is reported to have taken place both on
the surface of the film in the first instance and subsequently into the bulk of LB
films of tetra-4-tert-butyl-phthalocyanitosilicon dichloride (ttb-PcSiCl2/Ag). The
surface adsorption increases the film thickness, while the change in the real part of
the film refractive index occurs due to the diffusion of NO2 into the LB film. These
processes are reversible. The SPR response of the bare Ag film to the exposure of
NOx (the equilibrium mixture of nitrogen dioxide and dinitrogen tetroxide) was
slightly slower than the LB films of tetra-4-tert-butyl-phthalocyanitosilicon
dichloride (ttb-PcSiCl2/Ag) system. This effect was associated with the high
affinity of the phthalocyanine to NOx gas. The NO2 gas was also able to reach the
Ag/Pc interface film, and the reaction with the metal film resulted in the growth of a
surface layer. This irreversible effect was found to be predominant during the
recovery cycle and considerably reduced by the presence of a buffer LB layer of x-
tricosanoic (x-TA) acid [156]. An improvement in sensing NO2 is achieved for LB
film of copper phthalocyanine-based SPR using ultrathin nickel film to protect the
underlying Ag layer [157]. SPR studies show the decrease of the absorbance of
spin-coated thin films of 18-crown-6 metal-free phthalocyanine molecules on
ten-minute exposure to NO2 of 100 ppm concentration. There is an increase in the
phthalocyanine overlayer thickness by nearly 8 % because of swelling caused by
both surface adsorption and bulk diffusion of NO2. The optical permittivity of the
NO2-treated Pc overlayer is regarded as being an average of the optical permit-
tivities of both components since the optical fields associated with surface plasmon
polarities sample both the Pc and the NO2 molecules. The disruption in the charge
conjugation of the phthalocyanine molecules was observed because of the gener-
ation of organic radical cations as end products of the charge transfer during the
physisorption process [158]. Ordered self-assembled monolayers of
212 D. Mukherjee et al.

Fig. 8.27 Effect of 100 ppm NO2 in dry air on SPR of a 21-crown-7 H2Pc and b 15-crown-5 c
H2Pc (solid curve, before exposure; dot dash curve, after 2-min exposure; dotted curve, after 4-min
exposure; dashed curve, after 15-min exposure), c successive exposure and reversal cycles for a
15-crown-5 H2Pc film in 1 ppm NO*. (The labels ON and OFF refer to 1 ppm NO in air).
Adapted from Ref. [155]

diphthalocyanine disulfide molecules on gold-coated substrates exhibit satisfactory

response to NO2 with concentration-dependent sensitivity [159]. 300-nm-thick
films of cobalt phthalocyanine (CoPc) trapped in polyvinyl chloride have been
investigated for NO2 sensing using the SPR technique. NO2 interactions with CoPc
films increase the resonance angle from 43.05° to 44.06°, indicating an increase in
the real part of propagation constant ksp in Eq. (8.5) to 1.4  107m−1. 7 % dip in
resonance and half-width increase to nearly 0.009 radians are caused by the axial
adsorption of NO2 by Co central atoms [160].
The sensing mechanism of liquid crystalline-substituted Pc is interesting because
of easy solution-based formulation of supramolecular structures with the ability to
design the desired mesomorphic properties and to control optoelectronic properties
along a preferential direction. The kinetic SPR responses of about 22-nm-thick
spin-coated films of mesogenic nickel phthalocyanine with three hexylthio sub-
stituents are presented in Fig. 8.28i, ii to periodic exposure of non-aromatic chlo-
roform and aromatic benzene vapours. The sensor responses are reversible within
2 min as the vapour is switched off. The effect of the substituent type 3 on chlo-
roform sensing appears to be large relative to benzene as the corresponding increase
in the film thickness is about 10 times higher. It is observed from the Raman spectra
that the hydrogen bonds with alkyl chains of the substituents are formed through the
interaction saturated C–C bonds of chloroform. The bond with
4,7,10-trioxaundecan-1-sulphanyl substituents is believed to be much stronger
possibly because of greater contacts. The substituents have no apparent influence on
benzene detection because of their p-p interaction with the conjugated Pc rings
[161]. A comparative study of SPR sensing response of hexadecafluorinated 3d
metal phthalocyanine (MPcF16) with four different central metal ions (Zn, Co, Cu
and Ni) has recently been reported for exposures of 100 and 200 ppm NH3 [162],
and the results of kinetic measurements are presented in Fig. 8.29i corresponding to
a fixed incident angle of h = 44.2° close to their resonance angle. The film swelling
due to adsorption of NH3 becomes evident from the corresponding increase in
thickness from 4 % for NiPcF16 to 18 % for ZnPcF16 films. The most pronounced
spectral shift of 5–6 cm−1 upon exposure to ammonia vapour is observed in the
case of ZnPcF16 for the IR bands at 603, 654, 937, 960, 1148 and 1266 cm−1. These
shifts are mainly due to axial coordination of NH3 to the metal ion leading to
out-of-plane distortion of the MPc core. Figure 8.29ii shows a good correlation
between the DFT-calculated binding energies (DE), experimentally measured shift
of the selected IR bands (Dm) and the optical sensor response in terms of relative
refractive index change (Dn). The sensor responses are reversible, and the films are
found to be stable and exhibit no significant loss in sensitivity after a few tens of
repeated cycles of measurements. The sensor properties of thin films of
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 213
214 D. Mukherjee et al.

Fig. 8.28 Kinetics of SPR response of spin-coated octa-substituted phthalocyanine nickel films
with three different substituents to i chloroform and ii benzene vapour of two concentrations:
(a) ps/10 and (b) ps/5 where ps is the saturation vapour pressure, [peripheral substituent type (1)–S
(CH2)11CH3, (2)–SCH(CH2OC12H25)2 and (3)–S(CH2CH2O)3CH3. Adapted from Ref. [161]

octa-substituted copper phthalocyanine derivative with bulky substituents to ben-

zene and chloroform have been investigated by both SPR and ellipsometric tech-
niques. The SPR measurements are limited response in terms of small changes of
refractive index of 0.04 and 0.02 for chloroform and benzene, respectively.
However, the ellipsometry method is found to be more suitable because of its
ability to measure two ellipsometric parameters (W and D). Shifts in D of 4.8 and
1.6 nm are observed for chloroform and benzene, respectively [163].
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 215

Fig. 8.29 i Fluorinated 3d

metal phthalocyanines to 100
and 200 ppm ammonia and ii
correlation between SPR data
and DFT calculations.
Adapted from Ref. [162]

Standard SPR biosensors with 3–4 flow cells on a single sensor chip are not
suitable for applications in high-throughput screening [164]. The weakly bound
biointeractions in the presence of excess solution species can be studied with a high
degree of sensitivity, and therefore, SPR imaging based upon rapid optical arrays
has been developed in recent years to implement high-throughput screening of
drugs and biomarkers as well as clinical diagnosis. Label-free detection is partic-
ularly important in the determination of protein-protein interactions without inter-
ference of any labelling agent in the protein binding. Protein arrays have been used
216 D. Mukherjee et al.

Fig. 8.30 SPR images of adsorbed BSA on a as-deposited and b heated [(C6S)(8)PcCu] film at
617 nm and reflected intensity versus wavelength obtained at the incident angle of 46° for
adsorbed BSA on c as-deposited and d heated [(C6S)(8)PcCu] film. Adapted from Ref. [166]

for multiplexed kinetic information in the development of HIV inhibitors [165].

Using a CCD camera, the surface plasmon resonance imaging method has been
employed to investigate the BSA protein adsorption on spin-coated films of liquid
crystalline copper octakishexylthio phthalocyanines [(C6S)(8)PcM]. The
self-reorganisation of [(C6S)(8)PcCu] on postdeposition heat treatment at 338 K for
3 h in vacuum is believed to influence the BSA adsorption from the intensity
changes in the SPR images (Figs. 8.30a, b). The mean intensity of the reflected
light is estimated to be 73 % higher for thermally treated than as-deposited film.
The blueshift of 7 nm is observed in the variation of the reflected light intensity
with the wavelength spectra in Fig. 8.30c after 5-min-long washing with phosphate
buffer saline (PBS). The resonance occurs at 635 nm for BSA adsorbed on the
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 217

as-deposited [(C6S)(8)PcCu] film. In contrast, a shift from 622 to 663 nm is

observed in Fig. 8.30d for the [(C6S)(8)PcCu] annealed film, similarly washed with
PBS. Similar measurements have been carried out on [(C6S)(8)PcM] films, and it is
found that the amount of BSA adsorption on [(C6S)(8)PcNi] is more limited than
that on [(C6S)(8)PcCu] and this observation is consistent with the results on the
surface wettability obtained from the contact angle measurements. SPR sensitivity
compares well with that of other commonly used methods such as fluorescence
spectroscopy. Phthalocyanine surface modified with protein coating may be useful
for photodynamic therapy in order to increase the biocompatibility [166].

8.8 Concluding Remarks

Metallophthalocyanine (MPc) are found to be easily solution processable at room

temperature because of chemically suitable substitutions, and this leads to the
exploitation of low-cost manufacturing of practical sensors. Theamperometric
approach generally suffers from long response times, and their use is restricted due
to possible fire and explosion hazards caused by electrical power. Organic systems
are also prone to slow degradation arising from the chemical reactivity of charge
carriers. The drain current of a p-channel OFET increases on exposure of the
electron-accepting analytes, while an n-channel OFETs show the decrease in cur-
rent on similar treatments. The opposite effect is observed for electron-donating
analytes. A typical OFET is therefore limited in selection between electron-
denoting and electron-accepting analytes. Future work is, therefore, likely to
involve the design of complementary circuit using p- and n-channel OFETs which
show good discriminatory response with high selectivity. The sensors can be made
selective to a particular analyte by making suitable choice of substitutions or layer
deposition techniques. Potentiometric sensors are most often based on the mea-
surement of interfacial potential at an electrode surface due to a selective ion
exchange reaction. The use of a dynamic rest instead of a static rest reduces the
duration of the recovery period of ozone sensors, but is not necessary to achieve
good reproducibility, in contrast to the behaviour for more stable gases, for example
NO2. The performances of SAW and QCM sensors are found to be sensitive to
operating temperature. Optical transduction techniques, on the other hand, offer
several advantages of low energy consumption, high speed, high precision and
immunity to interference. Future work also entails the development of fibre-optic
SPR probe with remote sensing capabilities for the detection of trace gas species.

Acknowledgments The authors are grateful to the UK India Education and Research Initiative
for the award of grant (UKIERI-DST-2012/13-73), which brings them together for collaboration.
Sincere gratitude is due to Professors Michael J Cook and Andrew N Cammidge of University of
East Anglia (UK), Professor Tamara V Basova of Nikolaev Institute of Inorganic Chemistry
(Russia) and Dr Paul Harris and Dr Rachel Burch of Brunel University London for fruitful col-
laboration and support. AKR, in particular, appreciates thankfully encouragement and support he
has received from his two grandsons Rishabh and Mayan Ray.
218 D. Mukherjee et al.

References

1. de Diesbach H, von der Weid E (1927) Quelques sels complexes des o-dinitriles avec le
cuivre et la pyridine. Helv Chim Acta 10:886–888
2. Guillaud G, Simon J, Germain JP (1998) Metallophthalocyanines: gas sensors, resistors and
field effect transistors. Coord Chem Rev 178–180: 1433–1484; c) Armstrong NR (2000)
Phthalocyanines and porphyrins as materials. J. Porphyrins Phthalocyanines 4:414–417
3. de la Torre G, Vazquez P, Torres T (2004) Role of structural factors in the nonlinear optical
properties of phthalocyanines and related componds. Chem Rev 104:3723–3750
4. Cid J-J, Yum J-H, Jang S-R, Nazeeruddin MK, Martinez- Ferrero E, Palomares E, Ko J,
Graetzel M, Torres T (2007) Molecular cosensitization for efficient panchromatic
dye-sensitized solar cells. Angew Chem Int Ed 46:8358–8362
5. Kadish KM, Smith KM, Guilard R (2003) The porphyrin handbook, vol 15–20. Academic
Press, San Diego, CA, USA
6. Pinzon JR, Plonska-Brzezinska ME, Cadona CM, Athans AJ, Gayathri SS, Guldi DM,
Herranz MA, Martin N, Torres T, Echegoyen L (2008) Sc3N@C80-ferrocene
electrondonor/acceptor conjugates as promising materials for photovoltalic applications.
Angew Chem Int Ed 47:4173–4176
7. Campidelli S, Ballesteros B, Filoramo A, Diaz D, de la Torre G, Torres T, Rahman GMA,
Aminur EC, Kiessling D, Werner F, Sgobba V, Guldi DM, Cioffi C, Prato M, Bourgoin J-P
(2008) Facile decoration of functionalized single-wall carbon nanotubes with phthalocya-
nines via “Click Chemistry”. J Am Chem Soc 130:11503–11509
8. de la Torre G, Claessens CG, Torres T (2007) Phthalocyanines: old dyes, new materials,
putting colour in nanotechnology. Chem Commun 20:2000–2015
9. D’Souza F, Ito O (2009) Supramolecular donor-acceptor hybrids of porphyrins/
phthalocyanines with fullerenes/carbon nanotubes: electron transfer, sensing, switching
and catalytic applications. Chem Commun 33:4913–4928
10. Martinez-Diaz MV, de la Torre G, Torres T (2010) Lighting porphyrins and phthalocyanines
for molecular photovoltaics. Chem Commun 46:7090–7108
11. Bottari G, de la Torre G, Guldi DM, Torres T (2010) Covalent and noncovalent
phthalocyanine-carbon nanostructure systems: synthesis, photoinduced electron transfer and
application to molecular photovoltaics. Chem Rev 110(11):6768–6816
12. Bottari G, Suanzes JA, Trukhina O, Torres T (2011) Phthalocyanine-carbon nanostructure
materials assembled through supramolecular interactions. J Phys Chem Lett 2(8):905–913
13. Choi H-J, Kumar NA, Baek J-B (2015) Graphene supported non-precious metal-macrocycle
catalysts for oxygen reduction reaction in fuel cells. Nanoscale 7:6991–6998
14. Ortiz de Montellano PR (2004) Cytochrome P-450: structure, mechanism and biochemistry,
3rd edn. New York, KluwerAcademic/Plenum
15. Dougherty TJ (1993) Photodynamic therapy. Phtotochem Phthobiol 58(6):895–900
16. Moan J (1990) Properties for optimal PDT sensitizers. J Photochem Photobiol B Biol
5:521–524
17. a) Camerin M, Moreno M, Marin MJ, Schofield CL, Chambrier I, Cook MJ, Coppellotti O,
Jori G, Russell DA (2016) Delivery of a hydrophobic phthalocyanine photosensitizer using
PEGylated gold nanoparticle conjugates for the in vivo photodynamic therapy of amelanotic
melanoma. Photochem Photobiol Sci 15(5):618–625; b) Stuchinskaya T, Moreno M,
Cook MJ, Edwards DR, Russell DA (2011) Targeted photodynamic therapy of breast cancer
cells using antibody-phthalocyanine-gold nanoparticle conjugates. Photochem Photobiol Sci
10(5):822–831
18. a) Bohrer FI, Colesniuc CN, Park J, Schuller IK, Kummel AC, Trogler WC (2008) Selective
detection of vapor phase hydrogen peroxide with phthalocyanine chemiresistors. J Am
Chem Soc 130:3712–3713; b) Rakow NA, Sen A, Janzen MC, Ponder JB, Suslick KS
(2005) Molecular recognitionand discrimination of amines with a colorimetric array. Angew
Chemie Int Ed 44:4528–4532. c) Wang B, Zuo X, Wu YQ, Chen ZM, He CY, Duan WB
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 219

(2011) Comparative gas sensing in copper porphyrin and copper phthalocyanine

spin-coating films. Sens Actuators B Chem 152:191–195
19. Harbeck S, Gocmen S, Emirik OF, Ozturk ZZ, Ahsen V, Gurek AG (2016) Synthesis of
branched alkoxy side chains containing phthalocyanine derivates and their application in
mass sensitive QCM sensors. Sens Actuator B-Chem 233:55–62
20. Guillaud G, Simon J, Germain JP (1998) Metallophthalocyanines gas sensor, resistors and
field effect transistors. Coord Chem Rev 178:1433–1484
21. Muzikante I, Parra V, Dobulans R, Fonavs E, Latvels J, Bouvet M (2007) A novel gas sensor
transducer based on phthalocyanine heterojunction devices. Sensors 7:2984–2996
22. Harbeck S, Gocmen S, Emirik OF, Ozturk Z, Ahsen V’Gurek AG (2016) Synthesis of
branched alkoxy side chains containing phthalocyanine derivates and their application in
mass sensitive QCM sensors. Sens Actuator B-Chem 233:55–62
23. Duruk EG, Yenilmez HY, Altindal A, Bayir ZA (2015) Microwave-assisted synthesis of
novel non-peripherally substituted metallophthalocyanines and their sensing behaviour for a
broad range of Lewis bases. Dalton Trans 44(21):10060–10068
24. Saini R, Mahajan A, Bedi RK, Aswal DK, Debnath K (2014) Solution processed films and
nanobelts of substituted zinc phthalocyanine as room temperature ppb level Cl-2 sensors.
Sens Actuator B-Chem 198:164–172
25. Kumawat KL, Mergu N, Singh AK, Gupta VK (2015) A novel optical sensor for copper ions
based on phthalocyanine tetrasulfonic acid. Sens Actuator B-Chem 212:389–394
26. Spadavecchia J, Ciccarella G, Siciliano P, Capone S, Rella R (2004) Spin-coated thin films
of metal porphyrin-phthalocyanine blend for an optochemical sensor of alcohol vapours.
Sens Actuators B 100:88–93
27. Yang WY, Xu JH, Mao YW, Yang YJ, Jiang YD (2016) Detection of volatile organic
compounds using langmuir-blodgett films of zinc-porphyrin and zinc-phthalocyanine. Synth
React Inorg Met-Org Nano-Metal Chem 46(5):735–740
28. Bohrer FI, Colesniuc CN, Park J, Ruidiaz ME, Schuller IK, Kummel AC, Trogler WC
(2009) Comparative gas sensing in cobalt, nickel, copper, zinc and metal free phthalocyanine
chemiresistors. 131:478–485
29. a) Torsi L, Dodabalapur A, Sabbatini L, Zambonin PC (2000) Multi-parameter gas sensors
based on organic thin-film-transistors. Sens Actuators B 67:312–316; b) Crone B,
Dodabalapur A, Gelperin A, Torsi L, Katz HE, Lovinger AJ, Bao Z (2001) Electronic
sensing of vapors with organic transistors. Appl Phys Lett 78:2229–2231
30. Stetter JR, Penrose WR, Yao S (2003) Sensors, chemical sensors, electrochemical sensors,
and ECS. J Electrochem Soc 150:S11–S16
31. Crouch D, Thorpe SC, Cook MJ, Chambrier I, Ray AK (1994) Langmuir-Blodgett-films of
an asymmetrically substituted phthalocyanine—improved gas-sensing properties. Sens
Actuator B-Chem 19(1–3):411–414
32. Fernandes AN, Richardson TH (2008) Conductometric gas sensing studies of tert-butyl
silicon-[bis ethyloxy]-phthalocyanine LB films. J Mater Sci 43(4):1305–1310
33. Dooling CM, Richardson TH, Valli L, Rella R, Jiang J (2002) NO2 sensitivity of gadolinium
bis-phthalocyanine assemblies prepared by ultra-fast LB deposition. Colloid Surf
A-Physicochem Eng Asp 198(SI):791–796
34. Rickwood KR, Lovett DR, Lukas B, Silver J (1995) Semiconducting Pyro-electric and
chlorine sensing properties of ytterbium bisphthalocyanine langmuir boldgett thin films.
J Mater Chem 5:725–729
35. Liang BJ, Yuan CW, Yu W, Yin Z (1998) Europium bisphthalocyanine LB films for gas
sensibility and selectivity. Supramol Sci 5(5–6):661–664
36. Gould RD, Ibrahim NA (2001) The electrical response of evaporated cobalt phthalocyanine
thin films on exposure to NO2. Thin Solid Films 398:432–437
37. Klyamer DD, Sukhikh AS, Krasnov PO, Gromilov SA, Morozova NB, Basova TV (2016)
Thin films of tetrafluorosubstituted cobalt phthalocyanine: Structure and sensor properties.
Appl Surf Sci 372:79–86
220 D. Mukherjee et al.

38. Komolov AS, Lazneva EF, Gerasimova NB, Zimina MV, Si P, Panina YA (2015)
Water-soluble copper phthalocyanine for optimization of gas-sensor characteristics of tin
dioxide upon adsorption of ammonia. Phys Solid State 57(12):2550–2554
39. Cimen Y, Ermis E, Dumludag F, Ozkaya AR, Salih B, Bekaroglu O (2014) Synthesis,
characterization, electrochemistry and VOC sensing properties of novel ball-type dinuclear
metallophthalocyanines. Sens Actuator B-Chem 202:1137–1147
40. Kaya EN, Tuncel S, Basova TV, Banimuslem H, Hassan A, Gurek AG, Ahsen V,
Durmus M (2014) Effect of pyrene substitution on the formation and sensor properties of
phthalocyanine-single walled carbon nanotube hybrids. Sens Actuator B-Chem
199:277–283
41. Chakane S, Gokarna A, Bhoraskar SV (2003) Metallophthalocyanine coated porous silicon
gas sensor selective to NO2. Sens Actuator B-Chem 92(1–2):1–5
42. a) Huang WG, Besar K, LeCover R, Rule AM, Breysse PN, Katz HE (2012) Highly
sensitive NH3 detection based on organic field-effect transistors with Tris(pentafluorophenyl)
borane as receptor. J Am Chem Soc 134(36):14650–14653; b) Bouvet M, Xiong H,
Parra V (2010) Molecular semiconductor-doped insulator (MSDI) heterojunctions:
oligothiophene/bisphtalocyanine (LuPc2) and perylene/bisphthalocyanine as new structures
for gas sensing. Sens Actuators B 145(1):501–506
43. Royer JE, Kappe ED, Zhang CY, Martin DT, Troglerm WC, Kummel AC (2012) Organic
thin-film transistors for selective hydrogen peroxide and organic peroxide vapor detection.
J Phys Chem C 116(46):24566–24572
44. Mapare SV, Yu PL, Sarkar A, Mukhopadhyay SC (2013) A review of sensor technology for
in-field phosphate monitoring. I Conf Sens Technol 411–418
45. Elkington D, Cooling N, Belcher W, Dastoor PC, Zhou X (2014) Organic thin-film transistor
(OTFT)-based sensors. Electronics 3:234–254
46. a) Khan HU, Roberts ME, Knoll W, Bao ZN (2011) Pentacene based organic thin film
transistors as the transducer for biochemical sensing in aqueous media. Chem Mater
23:1946–1953; b) Roberts ME, LeMieux MC, Bao ZN (2009) Sorted and aligned
single-walled carbon nanotube networks for transistor-based aqueous chemical sensors. ACS
Nano 3:3287–3293
47. Kong H, Jung BJ, Sinha J, Katz HE (2012) Electrical “turn-on” response of poly(3,3‴-
didodecylquaterthiophene) and electron donor blend transistors to 2,4,6-trinitrotoluene.
Chem Mater 24:2621–2623
48. Andringa AM, Meijboom JR, Smits ECP, Mathijssen SGJ, Blom PWM, de Leeuw DM
(2011) Gate-bias controlled charge trapping as a mechanism for NO2 detection with
field-effect transistors. Adv Funct Mater 21:100–107
49. Oprea A, Weimar U, Simon E, Fleischer M, Frerichs H-P, Wilbertz Ch, Lehmann M (2006)
Copper phthalocyanine suspended gate field effect transistors for NO-detection. Sens
Actuators B 118:249–254
50. Gu C, Sun L, Zhang T, Li T, Zhang X (1998) High-sensitivity phthalocyanine LB film gas
sensor based on field effect transistors. Thin Solid Films 327–329:383–386
51. Pohle R, Simon E, Schneider R, Fleischer M, Sollacher R, Gao H, Müller K, Jauch P,
Loepfe M, Frerichs HP, Wilbertz C (2007) Fire detection with low power FET gas sensors.
Sens Actuators B 120:669–672
52. Barker PS, Petty MC, Monkman AP, McMurdo J, Cook MJ, Pride R (1996) A hybrid
phthalocyanine/silicon field-effect transistor sensor for NO2. Thin Solid Films 284:94–97
53. Lee SC, Hwang BW, Lee SJ, Choi HY, Kim SY, Jung SY, Ragupathy D, Lee DD, Kim JC
(2011) A novel tin oxide-based recoverable thick film SO2 gas sensor promoted with
magnesium and vanadium oxides. Sens Actuators B 160:1328–1334
54. Cid CC, Jimenez-Cadena G, Riu J, Maroto A, Xavier Rius F, Batema GD, van Koten G
(2009) Selective detection of SO2 at room temperature based on organoplatinum
functionalized single-walled carbon nanotube field effect transistors. Sens Actuators B
Chem 141:97–103
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 221

55. Shaymurat T, Tang Q, Tong Y, Dong L, Liu Y (2013) Gas dielectric transistor of CuPc
single crystalline nanowire for SO2 detection down to sub-ppm levels at room temperature.
Adv Mater 25:2269–2273
56. Schütze A, Pieper N, Zacheja J (1995) Quantitative ozone measurement using a
phthalocyanine thin film sensor and dynamic signal evaluation. Sens Actuators B
23:215–217
57. Bouvet M, Leroy A, Simon J, Tournilhac F, Guillaud G, Lessnick P, Maillard A,
Spirkovitch S, Debliquy M, de Hann A, Decroly A (2001) Detection and titration of ozone
using metallophthalocyanine based field effect transistors. Sens Actuators B 72:86–93
58. Li X, Jiang Y, Xie G, Tai H, Sun P, Zhang B (2013) Copper phthalocyanine thin film
transistors for hydrogen sulfide detection. Sens Actuators B 176:1191–1196
59. Trogler WC (2012) Chemical sensing with semiconducting metal phthalocyanines,
molecular electronic structures of transition metal complexes. Book Ser Struc Bond
142:91–117
60. Bora M, Schut D, Baldo MA (2007) Combinatorial detection of volatile organic compounds
using metal-phthalocyanine field effect transistors. Anal Chem 79:3298–3303
61. Huang W, Sinha J, Yeh M-L, Hardigree JFM, LeCover R, Besar K, Rule AM, Breysse PN,
Katz HE (2013) Diverse organic field effect transistor sensor responses from two
functionalized naphthalenetetracarboxylic diimides and copper phthalocyanine semiconduc-
tors distinguishable over a wide analyte range. Adv Funct Mater 23:4094–4104
62. Liu J, Masuda Y, Sekido E, Wakida S-I, Hiiro K (1989) Phosphate ion-sensitive
coated-wire/field effect transistor electrode based on cobalt phthalocyanine with poly(viny
chloride) as the membrane matrix. Anal Chim Acta 224:145–151
63. a) Brook I (1981) The importance of lactic acid levels in body fluids in the detection of
bacterial infections. Clin Infect Dis 3(3):470–478; b) Smith SM, Eng RH, Campos JM,
Chmel H (1989) D-lactic acid measurements in the diagnosis of bacterial infections. J Clin
Microbiol 27(3):385–388
64. Nikolaus N, Strehlitz B (2008) Amperometric lactate biosensors and their application in
(sports) medicine, for life quality and wellbeing. Microchim Acta 160:15–55
65. Someya T, Dodabalapur A, Gelperin A, Katz HE, Bao Z (2002) Integration and response of
organic electronics with aqueous microfluidics. Langmuir 18:5299–5302
66. de Oliveira RF, Merces L, Vello TP, Bufon CCB (2016) Water-gated phthalocyanine
transistors: operation and transduction of the peptide-enzyme interaction. Org Electron
31:217–226
67. Stradiotto NR, Yamanaka H, Zanoni MVB (2003) Electrochemical sensors: a powerful tool
in analytical chemistry. J Braz Chem Soc 14:159–173
68. Han J, Xu X, Rao X, Wei M, Evans DG, Duan XJ (2011) Layer-by-layer assembly of
layered double hydroxide/cobalt phthalocyanine ultrathin film and its application for sensors.
Mater Chem 21:2126–2130
69. Siqueira JR, Gasparotto LHS Jr, Crespilho FN, Carvalho AJF, Zucolotto V, Oliveira ON
(2006) Physicochemical properties and sensing ability of metallophthalocyanines/chitosan
nanocomposites. J Phys Chem B 110:22690–22694
70. Zucolotto V, Ferreira M, Cordeiro MR, Constantino CJL, Moreira WC, Oliveira ON Jr
(2006) Nanoscale processing of polyaniline and phthalocyanines for sensing applications.
Sens Actuator B 113:809–815
71. Zampa MF, Araujo IMS, Junior JRS, Zucolotto V, Leite JRSA, Eiras C (2012) Development
of a novel biosensor using cationic antimicrobial peptide and nickel phthalocyanine ultrathin
films for electrochemical detection of dopamine. Inter J Anal Chem 1–7
72. Araujo IMS, Zampa MF, Moura JB, dos Santos Jr JR, Eaton P, Zucolotto V, Veras LMC, de
Paula RCM, Feitosa JPA, Leite JRSA, Eiras C (2012) Contribution of the cashew gum
(Anacardium occidentale L.) for development of layer-by-layer films with potential
application in nanobiomedical devices. Mater Sci Eng C 32:1588–1593
222 D. Mukherjee et al.

73. Santos AC, Zucolotto V, Constantino CJL, Cunha HN, dos Santos Jr JR, Eiras C (2007)
Electroactive LbL films of metallic phthalocyanines and poly(0-methoxyaniline) for sensing.
J Solid State Electrochem 11:1505–1510
74. Siqueira JR, Gasparotto LHS, Oliveira ON, Zucolotto V (2008) Processing of electroactive
nanostructured films incorporating carbon nanotubes and phthalocyanines for sensing.
J Phys Chem C 112:9050–9055
75. Zhu W, Zhang W, Li S, Ma H, Chen W, Pang H (2013) Fabrication and electrochemical
sensing performance of a composite film containing a phosphovanadomolybdate and cobalt
(II) tetrasulfonate phthalocyanine. Sens Actuator B 181:773–781
76. Deon M, Caldas EM, da Rosa DS, de Menezes EW, Dias SLP, Pereira MB, Costa TMH,
Arenas LT, Benvenutti EV (2015) Mesoporous silica xerogel modified with bridged ionic
silsesquioxane used to immobilize copper tetrasulfonated phthalocyanine applied to
electrochemical determination of dopamine. J Solid State Electrochem 19:2095–2105
77. Rahim A, Barros SBA, Kubota LT, Gushikem Y (2011) SiO2/C/Cu(II)phthalocyanine as a
biomimetic catalyst for dopamine monooxygenase in the development of an amperometric
sensor. Electrochim Acta 56:10116–10121
78. Chernyshov DV, Shvedene NV, Antipova ER, Pletnev IV (2008) Ionic liquid-based
miniature electrochemical sensors for the voltammetric determination of catecholamines.
Anal Chim Acta 621:178–184
79. Atta NF, Galal A, Abdel-Gawad FM, Mohamed EF (2015) Electrochemistry and detection
of dobutamine at gold nanoparticles cobalt-phthalocyanine modified carbon paste electrode.
J Electrochem Soc 162:B304–B311
80. Neto SY, de Luz RCS, Damos FS (2016) Visible LED light photoelectrochemical sensor for
detection of L-Dopa based on oxygen reduction on TiO2 sensitized with iron phthalocya-
nine. Electrochem Commun 62:1–4
81. Shi M, Chen Z, Guo L, Liang X, Zhang J, He C, Wang B, Wu Y (2014) A multiwalled
carbon nanotube/tetra-bisoheptyloxyphthalocyanine cobalt(II) composite with high dis-
persibility for electrochemical detection of ascorbic acid. J Mater Chem B 2:4876–4882
82. Pakapongpan S, Mensing JP, Phokharatkul D, Lomas T, Tuantranont A (2014) Highly
selective electrochemical sensor for ascorbic acid based on anovel hybrid graphene-copper
phthalocyanine-polyanilinenanocomposites. Electrochim Acta 133:294–301
83. Milczarek G (2009) Self-doped polyaniline films prepared by electropolymerization in the
presence of sulfonated nickel phthalocyanine. Thin Solid Films 517:6100–6104
84. Korkut SE, Akyuz D, Ozdogan K, Yerli Y, Koca A, Sener MK (2016)
TEMPO-functionalized zinc phthalocyanine: synthesis, magnetic properties, and its utility
for electrochemical sensing of ascorbic acid. Dalton Trans 45:3086–3092
85. Wang K, Xu J-J, Tang K-S, Chen H-Y (2005) Solid-contact potentiometric sensor for
ascorbic acid based on cobalt phthalocyanine nanoparticles as ionophore. Talanta
67:798–805
86. Gilmartin MAT, Hart JP (1994) Novel, reagentless, amperometric biosensor for uric acid
based on a chemically modified screen-printed carbon electrode coated with cellulose acetate
and uricase. Analyst 119:833–840
87. Giarola J, Pereira A (2016) Development and application of a sensor based on carbonaceous
materials and cobalt phthalocyanine composite for electrochemical determination of uric
acid. Electroanalysis 28:1–9
88. Ozoemena KI, Zhao Z, Nyokong T (2005) Immobilized cobalt(II) phthalocyanine–cobalt(II)
porphyrin pentamer at a glassy carbon electrode: Applications to efficient amperometric
sensing of hydrogen peroxide in neutral and basic media. Electrochem Commun 7:679–684
89. Pillay J, Ozoemena KI (2009) Layer-by-layer self-assembled nanostructured phthalocyan-
inatoiron(II)/SWCNT-poly(m-aminobenzenesulfonic acid) hybrid system on gold surface:
Electron transfer dynamics and amplification of H2O2 response. Electrochim Acta
54:5053–5059
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 223

90. Wang H, Bu Y, Dai W, Li K, Wang H, Zuo X (2015) Well-dispersed cobalt phthalocyanine

nanorods on graphene for theelectrochemical detection of hydrogen peroxide and glucose
sensing. Sens Actuator B 216:298–306
91. Shumba M, Nyokong T (2016) Characterization and electrocatalytic activity of nanocom-
posites consisting of nanosized cobalt tetraaminophenoxy phthalocyanine, multi-walled
carbon nanotubes and gold nanoparticles. Electroanalysis 28:1–12
92. Rattanarat P, Teengam P, Siangproh W, Ishimatsu R, Nakano K, Chailapakul O, Imato T
(2015) An electrochemical compact disk-type microfluidics platform for use as an enzymatic
biosensor. Electroanalysis 27:703–712
93. Ye J-S, Wen Y, Zhang WD, Cui HF, Xu GQ, Sheu F-S (2005) Electrochemical biosensing
platforms using phthalocyanine-functionalized carbon nanotube electrode. Electroanalysis
17:89–96
94. Mashazi PN, Ozoemena KI, Nyokong T (2006) Tetracarboxylic acid cobalt phthalocyanine
SAM on gold: potential applications as amperometric sensor for H2O2 and fabrication of
glucose biosensor. Electrochim Acta 52:177–186
95. Ozcan L, Sahin Y, Turk H (2008) Non-enzymatic glucose biosensor based on overoxidized
polypyrrole nanofiber electrode modified with cobalt(II) phthalocyanine tetrasulfonate.
Biosens Bioelectron 24:512–517
96. Oni J, Diab N, Reiter S, Schuhmann W (2005) Metallophthalocyanine-modified glassy
carbon electrodes: effects of film formation conditions on electrocatalytic activity towards
the oxidation of nitric oxide. Sens Actuator B 105:208–213
97. Pontie M, Lecture H, Bedioui F (1999) Improvement in the performance of a nickel
complex-based electrochemical sensor for the detection of nitric oxide in solution. Sens
Actuator B 56:1–5
98. Pontie M, Gobin C, Pauporte T, Bedioui F, Devynck J (2000) Electrochemical nitric oxide
microsensors: sensitivity and selectivity characterization. Anal Chim Acta 411:175–185
99. Miserere S, Ledru S, Ruille N, Griveau S, Boujtita M, Bedioui F (2006) Biocompatible
carbon-based screen-printed electrodes for the electrochemical detection of nitric oxide.
Electrochem Commun 8:238–244
100. Bedioui F, Trevin S, Devynck J, Lantoine F, Brunet A, Devynck M-A (1997) Elaboration
and use of nickel planar macrocyclic complex-based sensors for the direct electrochemical
measurement of nitric oxide in biological media. Biosen Bioelectron 12:225–212
101. Pailleret A, Oni J, Reiter S, Isik S, Etienne M, Bedioui F, Schuhman W (2003) In situ
formation and scanning electrochemical microscopy assisted positioning of NO-sensors
above human umbilical vein endothelial cells for the detection of nitric oxide release.
Electrochem Commun 5:847–852
102. Maree S, Nyokong T (2000) Electrocatalytic behavior of substituted cobalt phthalocyanines
towards the oxidation of cysteine. J Electroanal Chem 492(2):120–127
103. Shaidarova LG, Ziganshina SA, Gedmina AV, Chelnokova IA, Budnikov GK (2011)
Electrochemical behavior and voltammetric determination of cysteine and cystine at
carbon-paste electrodes modified with metal phthalocyanines. J Anal Chem 66:633–641
104. Abbas MN, Saeed AA, Singh B, Radowana AA, Dempsey E (2015) A cysteine sensor based
on a gold nanoparticle–iron phthalocyanine modified graphite paste electrode. Anal Methods
7:2529–2536
105. Shahrokhian S (2001) Lead phthalocyanine as a selective carrier for preparation of a
cysteine-selective electrode. Anal Chem 73:5972–5978
106. Ozoemena K, Westbroek P, Nyokong T (2001) Long-term stability of a gold electrode
modified with a self-assembled monolayer of octabutylthiophthalocyaninato-cobalt(II)
towards L-cysteine detection. Electrochem Commun 3:529–531
107. Kalimuthu P, Sivanesan A, John SA (2012) Fabrication of optochemical and electrochemical
sensors using thin films of porphyrin and phthalocyanine derivatives. J Chem Sci
124:1315–1325
108. Ozoemena KI, Nyokong T, Westbroek P (2003) Self-assembled monolayers of cobalt and
iron phthalocyanine complexes on gold electrodes: comparative surface electrochemistry and
electrocatalytic interaction with thiols and thiocyanate. Electroanalysis 14:1762–1770
224 D. Mukherjee et al.

109. Honeychurch KC, Hart JP (2012) The chronoamperometric and voltammetric behavior of
glutathione at screen printed carbon micro-band electrodes with cobalt phthalocyanine. Adv
Anal Chem 2:46–52
110. Wring SA, Har JP, Birch BJ (1989) Development of an improved carbon electrode
chemically modified with cobalt phthalocyanine as a re-usable sensor for glutathione.
Analyst 114:1563–1570
111. Wring SA, Har JP, Birch BJ (1992) Development of an amperometric assay for the
determination of reduced glutathione, using glutathione peroxidase and screen-printed
carbon electrodes chemically modified with cobalt phthalocyanine. Electroanalysis
4:299–309
112. Vashist SK, Vashist P (2011) Recent advances in quartz crystal microbalance-based sensors.
J Sens Article number 571405
113. Kumar A, Brunet J, Varenne C, Ndiaye A, Pauly A, Penza M, Alvisi M (2015)
Tetra-tert-butyl copper phthalocyanine-based detection in air at room temperature, Sens
Actuator B-Chem 210:398–407
114. Kumar A, Brunet J, Varenne C, Ndiaye A, Pauly A (2016) Phthalocyanines based QCM
sensors for aromatic hydrocarbons monitoring: role of metal atoms and substituents on
response to toluene. Sens Actuator B-Chem 230:320–329
115. Sen Z, Tarakci DK, Gurol I, Ahsen V, Harbeck M (2016) Governing the sorption and
sensing properties of titanium phthalocyanines by means of axial ligands. Sens Actuator
B-Chem 229:581–586
116. Regmi BP, Galpothdeniya WIS, Siraj N, Webb MH, Speller NC, Warner IM (2015)
Phthalocyanine- and porphyrin-based GUMBOS for rapid and sensitive detection of organic
vapors. Sens Actuator B-Chem 209:172–179
117. Giancane G, Basova T, Hassan A, Gumus G, Gurek AG, Ahsen V, Valli L (2012)
Investigations and application in piezoelectric phenol sensor of Langmuir-Schafer films of a
copper phthalocyanine derivative functionalized with bulky substituents. J Colloid Interf Sci
377:176–183
118. Erbahar DD, Gurol I, Gumus G, Musluoglu E, Ozturk ZZ, Ahsen V, Harbeck M (2012)
Pesticide sensing in water with phthalocyanine based QCM sensors. Sens Actuator B-Chem
173:562–568
119. Erbahar DD, Gurol I, Ahsen V, Ozturk ZZ, Musluoglu E, Harbeck M (2011) Explosives
detection in sea water with phthalocyanine quartz crystal microbalance sensors. Sens Lett 9
(2):745–748
120. Erbahar DD, Gumus G, Pamir O, Musluoglu E, Ahsen V, Gurol I, Harbeck M (2016)
Polyalkoxy substituted phthalocyanines sensitive to phenolic compounds in water. Sens
Actuator B-Chem 227:277–282
121. Afzal A, Iqbal N, Mujahid A, Schirhagl R (2013) Advanced vapour recognition materials for
selective and fast responsive surface acoustic wave sensors: a review. Anal Chim Acta
787:36–49
122. Ricco AJ, Martin SJ, Zipperian TE (1985) Surface wave acoustic gas sensor based on film
conductivity changes. Sens Actuator 8(4):319–333
123. Newton MI, Starke TKH, McHale G, Willis MR, Krier A (1998) Surface acoustic wave
device design for gas sensing applications. Electron Lett 34(17):1706–1707
124. Rugemer A, Reiss S, Geyer A, von Schickfus M, Hunklinger S (1999) Surface acoustic
wave NO2 sensing using attenuation as the measured quantity. Sens Actuator B-Chem 56(1–
2):45–49
125. Jakubik W, Urbanczyk M, Opilski A (2001) Sensor properties of lead phthalocyanine in a
surface acoustic wave system. Ultrasonics 39(3):227–232
126. Wang PH, Yu JH, Li ZJ, Ding ZJ, Guo L, Du B (2013) Synthesis and evaluation of a new
phthalocyanine as sensor material for sarin detection. Sens Actuator B-Chem
188:1306–1311
127. Fourati N, Seydou M, Zerrouki C, Singh A, Samanta S, Maurel F, Aswav DK, Chehimi M
(2014) Ultrasensitive and selective detection of dopamine using cobalt-phihalocyanine
nanopillar-based surface acoustic wave sensor. ACS Appl Mater Interf 6(24):22378–22386
8 Phthalocyanines as Sensitive Materials for Chemical Sensors 225

128. Jakubik WP, Urbanczyk MW, Maciak E (2006) Palladium and metal-free phthalocyanine
bilayer structures for hydrogen detection in the SAW sensor system based on interaction
speed. IEEE Sens J 6(5):1178–1185
129. Jakubik WP, Urbanczyk MW (2005) SAW hydrogen sensor with a bilayer structure based
on interaction speed. Sens Actuator B-Chem 106(2):602–608
130. Jakubik W, Urbanczyk M, Maciak E, Krzywiecki M (2012) Bi-layer nanostructures of CuPc
and Pd for resistance-type and SAW-type hydrogen gas sensors. Sens Actuator B-Chem 175
(SI):255–262
131. Askim JR, Mahmoud M, Suslick KS (2013) Optical sensor arrays for chemical sensing: the
optoelectronic nose. Chem Soc Rev 42(22):8649–8682
132. Bariain C, Matias IR, Fernandez-Valdivielso C, Arregui FJ, Rodriguez-Mendez ML, de
Saja JA (2003) Optical fiber sensor based on lutetium bisphthalocyanine for the detection of
gases using standard telecommunication wavelengths. Sens Actuator B-Chem 93(1–
3):153–158
133. Li DP, Zeng D, Huang J, Gong JJ, Yan P, Ge SX, Peng WJ, Ding LY, Yuan YQ (2014)
Development of 2-chlorophenol sensor based on a fiber optic oxygen transducer via
oxidation reaction catalyzed by tetranitro iron (II) phthalocyanine. IEEE Sens J 14
(10):3693–3700
134. John MS, Thomas J, Unnikrishnan KP, Radhakrishnan P, Nampoori VPN, Vallabhan CPG
(1999) NO2 detection with a fibre optic evanescent wave sensor. P Soc Photo-Opt Ins
3897:173–178
135. Kerdcharoen T, Kladsomboon S (2013) Optical chemical sensor and electronic nose based
on porphyrin and phthalocyanine. In: Applications of nanomaterials in sensors and
diagnostics. Springer series on chemical sensors and biosensors, vol 14, pp 237–255
136. Hassan AK, Ray AK, Travis JR, Ghassemlooy Z, Cook MJ, Abass A, Collins RA (1998)
The effect of NO2 on optical absorption in Langmuir-Blodgett films of octa-substituted
amphiphilic copper phthalocyanine molecules. Sens Actuator B-Chem 49(3):235–239
137. Spadavecchia J, Ciccarella G, Rella R, Capone S, Siciliano P (2003) Metallophthalocyanines
thin films in array configuration for electronic optical nose applications. Sens Actuator
B-Chem 96(3):489–497
138. Chaure S, Paul D, Vadagma P, Ray AK (2010) Spectroscopic investigation of sulfonate
phthalocyanine to probe enzyme reactions for heavy metals detection. J Hazard Mater 173
(1–3):253–257
139. Basova T, Paul S, Paul D, Vadgama P, Gürek AG, Ahsen V, Ray AK (2008) Liquid
crystalline phthalocyanine thin films as nanoscale substrates for protein adsorption.
J Bionanosci 2:114–118
140. Colak S, Durmus M, Yildiz SZ (2016) The water-soluble zwitterionic and cationic
tetra-substituted zinc(II) phthalocyanines: Synthesis, photophysical, photochemical and
protein binding properties. Polyhedron 113:115–122
141. Pal C, Cammidge AN, Cook MJ, Sosa-Sanchez JL, Sharma AK, Ray AK (2012) In situ
chemichromic studies of interactions between a lutetium bis-octaalkyl-substituted phthalo-
cyanine and selected biological cofactors. J R Soc Interf 9(66):183–189
142. Pal C, Sharma AK, Cammidge AN, Cook MJ, Ray AK (2013) Octaoctyl-Substituted
Lutetium Bisphthalocyanine for NADH Biosensing. J Phys Chem B 117(48):15033–15040
143. Basova T, Jushina I, Guerek AG, Ahsen V, Ray AK (2013) Electrochromic lutetium
phthalocyanine films for in situ detection of NADH. Opt Mater 35(3):634–637
144. Basova T, Jushina I, Guerek AG, Ahsen V, Ray AK (2008) Use of the electrochromic
behaviour of lanthanide phthalocyanine films for nicotinamide adenine dinucleotide
detection. J R Soc Interf 5(24):801–806
145. Arwin H (2001) Is ellipsometry suitable for sensor applications? Sens Actuator A-Phys 92
(1–3):43–51
146. Martensson J, Arwin H, Lundstrom I (1990) Thin-films of phthalocyanines studied with
spectroscopic ellipsometry—an optical gas sensor. Sens Actuator B-Chem 1(1–6):134–137
226 D. Mukherjee et al.

147. Basova T, Plyashkevich V, Hassan A, Gurek AG, Gumus G, Ahsen V (2009)

Phthalocyanine films as active layers of optical sensors for pentachlorophenol detection.
Sens Actuator B-Chem 139(2):557–562
148. Banimuslem H, Hassan A, Basova T, Gulmez AD, Tuncel S, Durmus M, Gurek AG,
Ahsen V (2014) Copper phthalocyanine/single walled carbon nanotubes hybrid thin films for
pentachlorophenol detection. Sens Actuator B-Chem 190:990–998
149. Hassan A, Banimuslem H, Basova T, Gulmez AD, Durmus M, Gurek AG, Ahsen V (2016)
Surface interaction of copper phthalocyanine modified single walled carbon nanotubes with
pesticides. Sens Actuator B-Chem 224:780–788
150. Liedberg B, Nylander C, Lundstrom I (1995) Biosensing with surface-plasmon resonance—
how it all started. Biosens Bioelectron 10(8):R1–R9
151. Liedberg B, Nylander C, Lundstrom I (1983) Surface-plasmon resonance for gas-detection
and biosensing. Sens Actuator 4(2):299–304
152. Basova T, Kol’tsov E, Hassan A, Tsargorodskaya A, Ray AK, Igumenov I (2005) Thin films
of copper hexadecafluorophthalocyanine CuPcF16 . Phys Stat Sol B 242(4):822–827
153. Ray AK Organic materials for chemical sensing in springer handbook of electronic and
photonic materials. In: Kasap S, Capper P (eds) pp 1241–1266
154. Granito C, Wilde JN, Petty MC, Houghton S, Iredale PJ (1996) Toluene vapour sensing
using copper and nickel phthalocyanine Langmuir-Blodgett films. Thin Solid Films
284:98–101
155. Wright JD, Cado A, Peacock SJ, Rivalle V, Smith AM (1995) Effects of nitrogen-dioxide on
surface-plasmon resonance of substituted phthalocyanine films. Sens Actuator B 29(1–
3):108–114
156. Lloyd JP, Pearson C, Petty MC (1988) Surface-plasmon resonance studies of gas effects in
phthalocyanine langmuir-blodgett films. Thin Solid Films 160(1–2):431–443
157. Zhu DG, Petty MC, Harris M (1990) An optical sensor for nitrogen-dioxide based on a
copper phthalocyanine langmuir-blodgett-film. Sens Actuator B-Chem 2(4):265–269
158. Jory MJ, Cann PS, Sambles JR (1994) Surface-plasmon polariton studies of 18-crown-6
metal-free phthalocyanine. J Phys D Appl Phys 27(1):169–174
159. Simpson TRE, Cook MJ, Petty MC, Thorpe SC, Russell DA (1996) Surface plasmon
resonance of self-assembled phthalocyanine monolayers: Possibilities for optical gas
sensing. Analyst 121(10):1501–1505
160. El-Basaty AS, El-Brolossy TA, Abdalla S, Negm S, Abdella RA, Talaat H (2008) Surface
plasmon sensor for NO2 gas. Surf Interf Anal 40(13):1623–1626
161. Basova T, Kol’stov E, Ray AK, Hassan AK, Gurek AG, Ahsen V (2006) Liquid crystalline
phthalocyanine spun films for organic vapour sensing. Sens Actuator B-Chem 113
(1):127–134
162. Basova TV, Mikhaleva NS, Hassan AK, Kiselev VG (2016) Thin films of fluorinated
3d-metal phthalocyanines as chemical sensors of ammonia: an optical spectroscopy study.
Sens Actuator B-Chem 227:634–642
163. Basova T, Tsargorodskaya A, Nabok A, Hassan AK, Gurek AG, Gumus G, Ahsen V (2009)
Investigation of gas-sensing properties of copper phthalocyanine films. Mater Sci Eng
C-Biomimetic Supramol Syst 29(3):814–818
164. Nguyen HH, Park J, Kang S, Kim M (2015) Surface plasmon resonance: a versatile
technique for biosensor applications. Sensors 15(5):10481–10510
165. Paul S, Vadgama P, Ray AK (2009) Surface plasmon resonance imaging for biosensing. IET
Nanobiotechnol 3(3):71–80
166. Paul S, Paul D, Basova T, Ray AK (2010) Characterisation of protein adsorption n different
liquid crystal phthalocyanine thin films. IET Nanobiotechnol 4(1):1–9
Chapter 9
Materials for Electronic Tongues: Smart
Sensor Combining Different Materials
and Chemometric Tools

Manel del Valle

9.1 Introduction

In recent years, we have been promulgating the concept of electronic tongues in the
field of chemical sensors. This paradigm has proved to be one of the most striking
benefits of the combination of chemical sensors with data processing tools [1].
Electronic tongues have been defined as an instrument devised for chemical anal-
ysis formed by an array of sensors having non-specific or partial overlapped
responses in combination with advanced mathematical procedures for signal pro-
cessing based on the pattern recognition (PARC) and/or multivariate analysis, viz.
artificial neural networks (ANNs), principal component analysis (PCA), etc. [2].
Variants can be a classical quantitative determination, where a number of sub-
stances may be determined in a single, direct operation, or the determination of a
substance in the presence of its interferents, provided their separation is not needed
[3–5]. Also, the concept can provide more interesting applications, such as iden-
tification of varieties or establishing the membership to a group or class, when
pattern recognition procedures are applied. The latter qualitative applications are of
great interest, as they can be the basis of automated detection principles in fields
such as food, beverages, or pharmaceutics, where alternatives to the human expert
are in demand [6]. It is also possible to correlate a characteristic or property, for
example, a perception, with the contribution of the different species sensed. In our
studies, the preferred chemometric tool for the processing of data has been ANN.
These are known to be powerful nonlinear modelers, applicable for quantitative and
also qualitative applications. In this sense, our approach is doubly biomimetic;
firstly, the use of groups of sensors with cross-response is the sensing scheme in the
taste buds of animals and, secondly, employing ANNs which are parallel infor-

M. del Valle (&)

Sensors and Biosensors Group, Chemistry Department,
Universitat Autònoma de Barcelona, Bellaterra, 08193 Barcelona, Spain
e-mail: [email protected]

© Springer International Publishing AG 2017 227

T.R.L.C. Paixão and S.M. Reddy (eds.), Materials for Chemical Sensing,
DOI 10.1007/978-3-319-47835-7_9
228 M. del Valle

mation processing algorithms inspired in the animal nervous system, whose max-
imum expression is the human brain [7].
There is no doubt that main stream research in the sensor field is the improve-
ment of selectivity, the foremost analytical property if one needs to integrate sensor
use in a stand-alone operation, without any sample pretreatment or elimination of
interfering species. The novel complementary, bioinspired way to do this is by the
use of an array of non-specific sensors, responding to primary ions and interferents,
coupled with the multivariate chemometric treatment of the complex data to extract
the different components present [8]. The purpose of the processing tool differs
depending on the application; it can be to identify a chemical species or to deter-
mine its concentration without having to eliminate interferences or to quantify them
at all. Key points are that the multidimensional generated data have to comprise the
required information about the system and that the high-order data have no
co-linearity (thanks to the cross-sensitivity assumption). Since all sensors used in
the array may respond to all analytes, a great amount of complex data are generated,
which must be processed using a multivariate calibration approach. Chemometrics
is in charge then for the extraction of significant features and for their transfor-
mation into the sought information. This coupling of chemometrics and electro-
chemical sensors, already identified as one of the recommended ways to improve
the sensor performance and presently within the “big data” trend, it is representing
a consolidated line of research in the electroanalysis field [9, 10].
If a literature search is performed with the terms “electronic tongue,” it can be
observed that the number of articles has been growing steadily since the initial works
dated in 1997, although it seems to stabilize in the last years. These trends are
summarized in Fig. 9.1. The number of articles in the most recent years is ca. 46 per
year—an additional observation is that nearly all the records documented per year
correspond to the use of electrochemical sensors; apart from those, very few elec-
tronic tongues have been proposed employing optical sensors or piezoelectric mass
sensors. Lastly, it can be commented that at the very beginning, most of the works
related to electronic tongues employed potentiometric sensors, while more recently,
this variant accounts only for around the half of the contributions. That is, as more
research groups have entered the field, the different variants are more equally dis-
tributed than at the beginning and the more diverse applications are covered. In brief,
one may conclude that the research field of electronic tongues is now mature.
In practice, then, there are essentially two main types of electronic tongue
systems, distinguishing the nature of the electrochemical sensors mostly used.
These may be potentiometric, when an ion-selective electrode (ISE) array is used or
voltammetric, using, for example, a number of metal electrodes of different nature
or a number of modified electrodes. There is also the possibility of using both types
of sensors, and then, a hybrid electronic tongue is the proper name.
A second interesting insight from the literature is to visualize how the research
on electronic tongues is distributed geographically. This can be easily grasped by
using Springer’s author mapper tool, an Internet application that creates a world
map according to the incidence of certain keywords on each country (see Fig. 9.2).
After this database search, the tool visualizes where the groups are with
9 Materials for Electronic Tongues: Smart Sensor Combining … 229

2015

2010
Year

2005

2000
Potentiometric
Voltammetric
Impedimetric
Non electrochemical

0 10 20 30 40 50 60
N. research papers

Fig. 9.1 Share of the transduction technique used for the sensors forming ET systems. Data
obtained from the literature search for the period 1996–2014 using SCOPUS database (Elsevier B.
V.), October 2015

Fig. 9.2 Visualization of the research groups doing contribution to the topic “electronic tongue”
according to Springer’s author mapper tool (http://www.authormapper.com), October 2015
230 M. del Valle

publications containing that keyword, and also indicates with color intensity their
research activity, the most intense the hottest color. Although coverage is limited to
publications appearing in Springer’s products, it serves as a first attempt visual-
ization tool on how the authors are distributed. From the generated map, it is
curious how the topic has been cultivated mainly from Europe and Asia, with scarce
intervention from North and Latin America, and only occasional appearance by
Africa and Oceania. The hottest groups, the ones in orange, correspond to Spain
(accounting for active groups in Valladolid, Valencia, and Barcelona) or the
Russian federation (pointing to the founder group from St. Petersburg). Other
regions especially active are Sweden (with pioneering work in voltammetric elec-
tronic tongues in Linköping), France (that accounts for interesting groups plus the
company Alpha MOS, manufacturer of one of the commercial electronic tongues),
China, and Japan (in this case with the works from Kyushu University related to
artificial taste, i.e., human taste mimicking).
On one additional inspection, checking the journals that concentrate the publi-
cations on electronic tongues, the major one is Sensors and Actuators B: Chemical
having a larger share of papers related to voltammetric, acoustic or impedimetric
sensors; a second place is headed by Analytica Chimica Acta, showing high number
on works related to potentiometry, but also other sensor types; a third placement can be
shared by Talanta, Analytical Chemistry, and Electroanalysis, also devoted to mainly
electrochemical sensors; as observed, mainly journals are from the field Analytical
Chemistry, whereas also other journals from the food field or the nanoscience field are
also participants on the share of publications, but to a lesser extent.
An arrangement of appropriate sensors, which provides the information related
to the problem to be addressed, is the first requirement for the development of an
electronic tongue. This chapter will focus on how the different technologies and
materials existing for the development of electrochemical sensors have been uti-
lized in the development of electronic tongues. On one side, for the potentiometric
subtype, the different variants in obtaining potentiometric transduction, with the
usage of different materials for developing the ion-selective membranes, will be
commented. On the other side, the different materials and options to achieve dif-
ferentiated current response when experimenting electrode configurations in a redox
system (i.e., metallic electrodes and modified electrodes) will be also reviewed.
The information generated by the sensors must be related to the chemical species
involved in the problem, either to the exact species considered or to the range of the
significant concentrations in which they may appear. This information will be the
starting point to create an intelligent system. Thus, the assembly can correctly
interpolate or classify unknown samples and identify them similarly to our brain
does. The transformation required needs to perform the correspondence between 1
and k sensors, n-dimensional vectors to another m-dimensional vector, which
describes the characteristics of interest of the sample. An ANN is an example of
chemometric tool that easily processes situations where the size of the input
information is given by k  n; here, it is also contained the simpler case of the
potentiometric electronic tongue, in which k sensors that provide the instant
information of complexity n = 1 are used. However, this departing scenario can be
9 Materials for Electronic Tongues: Smart Sensor Combining … 231

much more complex (i.e., high-dimensional), if vector info is supplied by each

sensor, for example, the case in voltammetric electronic tongues (where n for each
sensor can be on the order of several hundred measures).

9.1.1 Strategies for Signal Processing

In applications of electronic tongues, it is common to use sensor arrays; however,

due to the poor selectivity of these, raw data is a complex signal containing
information about the item of interest, but also information from other interfering
elements. This scenario proves more complex when voltammetric techniques are
employed; the signals obtained are of a higher dimension, being constituted of a
large number of points.
The second component of an electronic tongue is the stage of signal processing.
For this purpose, multivariate analysis techniques are normally used to determine the
contribution of various factors to an event or outcome. These factors are related to the
explanatory variables of the system; in electronic tongues, usually they are deter-
mined experimentally and are related to the signals from each sensor. The response
variable, in turn, is related to the concentration value determined, the identification of
a chemical or a specific event, or alternatively the perception modeling.
In any case, there are usually high levels of interference that result in a poor
signal-to-noise ratio, in addition to nonlinear behavior character data relating to the
analytes to be determined. Extracting relevant information from these situations
where it is not easy to extract and interpret the data, so that they can be used in
useful models for prediction, has become a complex task that makes use of mul-
tivariate processing tools. Figure 9.3 shows the most frequent chemometric tools
employed in conjunction with electronic tongues.

PCA/LDA

ANN

PLS

PCR

k-NN

MCR

Other

0 10 20 30 40 50
% of publications

Fig. 9.3 Main chemometric tools applied in the studies with electronic tongues. PCA principal
component analysis; LDA linear discrimination analysis; ANN artificial neural network; PLS partial
least squares; PCR principal component regression; k-NN k-nearest neighbor; MCR multiple
component regression. Data obtained from the literature search on the period 1996–2015 using
SCOPUS database (Elsevier)
232 M. del Valle

The chemometric knowledge area has been proposing and developing various
methods based on mathematical, statistical, and formal logic calculations, aiming to
establish or select procedures that may perform tasks to discriminate, measure,
classify, and model systems and use the maximum relevant information from the
available analytical data [1, 11]. For this purpose, approaches mainly used are PCA,
partial least squares (PLS) regression, or ANNs. The choice of the processing tool
for a particular case depends on the particular task to be addressed, but also the
structure of the input data (nonlinear components, autocorrelation, etc.).
PCA is a matrix transformation technique, in which multivariate departure data
are linearly reorganized through a change in the visualization axis; the new access is
determined in a way so that they are orthogonal and they contain as much infor-
mation as possible, i.e., they are calculated to contain a maximum variance from the
original data. In this way, highly multivariate data can be concentrated in a few axes
containing the relevant (differentiated from noise) information, and just from the
first axis components, one can predict the pertinence to a class. In order to perform
this pattern recognition task, one must provide a classifier rule, that is, an additional
chemometric tool that will accomplish the identification/classification.
Another important chemometric tool used in electronic tongues are ANNs; these
are attractive alternatives as they are logical operations that reproduce the human
brain. The ANNs get this functionality through the use of an entity, artificial
neuron, or perceptron, which process the input stimuli in a bioinspired form, i.e., as
our neural system does. ANNs are especially interesting because of its adaptability
to almost any mathematical transform. The models obtained can be of linear or
nonlinear nature, being suitable to different types of sensors, and making them
useful not only for multivariate calibration of electronic tongues, but for many other
tasks in Analytical Chemistry [12].
The other important chemometric tool is PLS. PLS is a linear statistical model in
which a similar processing to PCA is applied both to responses (data set) of the
sensors, and to the values of the concentrations of the chemical elements that are
being predicted. Then, a linear regression is applied to each main component to
obtain a regression model between data and concentrations. Its application is
usually related to problems where there are a large number of variables and rela-
tively few samples, a situation that causes the models obtained unfit for predicting
new, too different samples although good results are usually obtained when the
relationship between measurements and concentrations is linear.
Even so, in some cases, the initial information is too complex to be handled with
the tools mentioned, for example, when a full (or various) voltammogram needs to
be processed using ANNs. Normally, the starting signal is so complex that the
network performance is not adequate. For example, this occurs for network
architectures with hundreds of input neurons, as it would be required for a regular
voltammogram, and therefore, it becomes not practical. In these situations, it is
necessary somehow to preprocess the starting signal in order to extract the mean-
ingful information components and allow for the processing.
One of the most common methods is using data reduction, capable of decreasing
the volume of information so that this amount is contained, without significant
9 Materials for Electronic Tongues: Smart Sensor Combining … 233

losses, in fewer variables. Techniques such as PCA can be used to reduce the
complexity of initial data trying to preserve the significant information; the
methodology is to observe the variance in the experimental data in the lesser
number of new coordinate axes grouping information [13, 14]. Other techniques
used are Legendre polynomial fitting where the responses of the sensors are
approximated to a power series [15, 16], the Fourier transform used to decompose
the signal into a series of coefficients corresponding to the frequency content of the
original signal [17], and recently the wavelet decomposition [18, 19], consisting of
decomposing the signal into different frequency components, so that each has a
resolution according to a scale [20]. But whatever the approach made, it must meet
two objectives, reducing the complexity of the departure information and retaining
its meaningful content.
In short, from the techniques listed, it is possible to obtain the relevant infor-
mation and enable us to build suitable calibration models. To obtain a model from a
set of reduced calibration data, we may use different tools, but the preferred ones
will be PLS regression and ANNs.
Next, we will present the two main types of electronic tongues in relation to the
type of sensors that form the array: potentiometric electronic tongues and
voltammetric electronic tongues.

9.2 Potentiometric Electronic Tongues

There are two main variants in the development of electronic tongues when the type
of sensors used is the main feature to distinguish them [7]. These may be poten-
tiometric, when the sensor array is formed by ISEs (or related sensors), or they may
be voltammetric, using, for example, a number of different metal electrodes or a
number of catalyst-modified carbon electrodes. A third variant, of minor use, is to
employ equivalent recognition elements as above, but adopting the electrochemical
impedance spectroscopy as the transduction technique.
Just with the use of commercial ISEs, sufficiently interesting seminal works can
be found in the literature. For example, there is the founding work of Van der
Linden [21] in which an array of commercial ISE responses were modeled
employing neural network software. Two cases were described, a simultaneous
determination of calcium and copper(II) ions and the simultaneous determination of
potassium, calcium, nitrate, and chloride. The measurements for the Ca2+/Cu2+
determinations were done with a pH glass electrode plus calcium and copper
ion-selective electrodes equipped with PVC membranes; for the K+/Ca2+/NO3−/Cl−
determinations, the measurements were made with the relevant four ion-selective
electrodes (made of commercial PVC membrane type) plus a glass electrode; mean
relative errors were in the ±6 and ±8 % range. In a second example [22], a mixture
of halide ions (Cl−, Br−, F−, and OH−) was estimated simultaneously using an
experimental design and multivariate calibration methods. The sensor array
employed commercial ISEs, particularly two different chloride selective, one
234 M. del Valle

bromide selective and one thiocyanate selective, of the polycrystalline type, a

fluoride ISE of the monocrystalline type, plus a glass pH electrode. PLS regression
and principal component analysis did slightly worse than multivariate linear
regression. In the case of very slight or no interference on the ISE, each ion can be
determined using the corresponding ISE and univariate calibration methods, but the
use of multivariate methods did not degrade the overall results. In a third case study
worth mentioning [23], an expert system strategy (case-based reasoning) was
applied for multicomponent analysis in water from the data obtained by an ISE
sensor array. The latter was comprised of six commercial ISEs, a glass pH electrode
plus calcium, sodium, potassium, nitrate, and chloride PVC membrane electrodes.
One of the groups more active in this research, and normally employing
potentiometric electronic tongues, is that of Andrey Legin, in St. Petersburg
(Russia). With their pioneering work in collaboration with the group of d’Amico in
Rome, they practically initiated the field [24], also coining the terminology elec-
tronic tongue. In their studies, they normally use an array of PVC membrane ISEs
[25], with large number of sensors and with membranes custom-formulated, more
with components from the ligand and ion-exchange chemistry than with commer-
cial ionophores [26].
Also, a significant research group doing research with electronic tongues is that
of Ciosek and Wróblewski, in Warsaw [3]. They essentially use custom-formulated
potentiometric PVC membranes, with the goal of optimizing the variability of the
response in a custom way for the application studied. For this purpose, they adapt
the classical formulations of the PVC membranes; for example, in order to get
generic response to alkaline ions, they can prepare ISEs equipped with membranes
having mixture of ionophores in order to get sensors with partial selectivity. For
example, to get mixed response to alkaline ions, they report the use of the mixture
of valinomycin plus commercial sodium ionophore X, or to get generic response to
anions, they formulate a membrane with a mixture of hydrogen phosphate plus
fluoride ionophore of the uranyl salophen family [27]. Their preferred configuration
has been to glue a previously casted PVC membrane on a Philips body, achieving a
final symmetric membrane configuration design, in which the potentiometric
membrane separates the sample solution and an internal reference solution [28].
Although of a bit more complex design, it is recognized that this is the most stable
and reproducible setup for potentiometric sensing.
In our laboratory, we started research with electronic tongues taking profit of our
experience in the development of solid-contact ISEs, equipped with
custom-purpose PVC liquid membranes. Just opting with different formulations
from the ion-selective membranes’ literature, rather specific response ISEs or other
with more generic response to groups of substances could be arranged in arrays, in
order to explore the specific applications. The goal pursued in these preliminary
works was the multicomponent determination of certain species in the presence of
usual secondary ions that normally act as interferents. This type of application
showed an important advantage, in determining simultaneously the ions of interest
and the different interferents present at no extra cost. As an example, the work
described in [29] performed the determination of ammonium using the well-known
9 Materials for Electronic Tongues: Smart Sensor Combining … 235

ISE based on the nonactin electrode. As it is known, this sensor is affected by the
significant interference of sodium and potassium, and the aim was to obtain a
procedure for NH4+ determination without the need of eliminating alkaline ions.
The sensor array incorporated sensors with response to ammonium, potassium, and
sodium, plus others with generic response to alkaline ions. Multivariate calibration
was implemented using an ANN model, and the final application was successful in
the analysis of water samples from river waters and wastewaters. Figure 9.4 depicts
the construction procedure of one of these ISEs, in which a solid contact made from
an epoxy–graphite composite formed the base for the deposition of the PVC
membrane by solvent casting. These membranes were of standard use in ISE
methodology and were prepared using tetrahydrofuran as the volatile solvent.
Figure 9.5 shows the general formulation of these membranes, where the coupling

Fig. 9.4 Fabrication process of a potentiometric sensor by depositing the liquid membrane onto a
solid contact based on epoxy–graphite composite. Reprinted from [29], with permission from
Taylor & Francis

Fig. 9.5 Typical

composition of a
potentiometric PVC
membrane
236 M. del Valle

of ionophore and plasticizer is important for the exact pattern of selectivity response
displayed by the device; the 33 % PVC share constitutes the inert polymer matrix
for support of the recognition chemistry. Table 9.1 summarizes the formulation of
the sensors used for a more elaborated example, in this case to monitor the nutrients
and interferents for the nutrient solution in a hydroponic cultivation facility. Ions
determined were cations including ammonium, sodium, potassium, plus
anions nitrate and chloride; to complement the response provided by the rather
specific ISEs used, three sensors with generic responses to cations (two of them),
and a third to anions were included as departure information [30]. The system used
an ANN model to predict the concentrations of the considered nutrients (ammo-
nium, potassium, and nitrate) plus the undesired species sodium and chloride that
accumulate in the recirculated solution. The scheme of the proposed procedure is
shown in Fig. 9.6. The new monitoring system represented a clear improvement in
front of the classical replenishing systems in hydroponic cultivations that just
operate from a pH plus conductivity signals.
Multichannel sensor measurement combined with advanced treatment is the
departure point for a new concept in sensorics, the electronic tongue, a sensor
monitoring principle that has shown special suitability in the environmental field
[31]. In a further research work from our laboratory, an enlarged setup worked with
an array of 20 ISEs plus an ANN used as a pattern recognition method applied to
soil analysis [32]. Formulated membranes included sensors sensitive to cations
(alkaline and alkaline earth), anions, heavy metals, and sensors with generic
response. With this design, we got a versatile tool which was able to perform
qualitative and quantitative determinations. Figure 9.7 displays the experimental
setup during measurement for one of this soil extract.
As the first application, the qualitative discrimination between six distinct soil
types based on their extractable components was attempted. The procedure was

Table 9.1 Formulation of different potentiometric PVC membranes used in the environmental
monitoring of ammonia with correction of the presence of alkaline interferent ions
Sensor PVC (%) Plasticizer (%) Recognition element (%)
NH4+ 33 BPA (66) Nonactin (1)
K+ 30 DOS (66) Valinomycin (3)a
Na+ 22 NPOE (70) CMDMM (6)a
+
H 32.8 DOS (65.6) TDDA (1)a
−
NO3 30 DBP (67) TOAN (3)
Generic 1 29 DOS (67) Dibenzo (18-crown-6) (4)
Generic 2 27 DBS (70) Lasalocide (3)
Generic 3 29 DBP (65) TOAB (4)
a
Indicated sensors incorporate the lipophilic salt potassium tetrakis-p-chlorophenyl borate
BPA bis(1-butylpentyl) adipate; NPOE o-nitrophenyloctyl ether; DOS dioctylsebacate; DBS
dibutyl sebacate; DBP dibutyl phthalate; CMDMM bis[(12-crown-4) methyl]-
2-dodecyl-2-methylmalonate; TDDA tridodecylamine; TOAN tetraoctylammonium nitrate; TOAB
tetraoctylammonium bromide
9 Materials for Electronic Tongues: Smart Sensor Combining … 237

Fig. 9.6 Artificial neural network used as a quantitative model for predicting a number of
concentrations from the readings of a potentiometric sensor array

Fig. 9.7 Conventional setup

of a potentiometric electronic
tongue made from an array of
ISEs, measuring one sample
with all sensors in parallel

simplified to a single extraction step before measurements. Water, a BaCl2 saline

solution, and an acetic acid extract were evaluated as extracting agents. The best
performance was reached with the acetic acid extraction method with a correct
classification rate and sensitivity both of 94 %, and a specificity of 100 %, as it is
presented in Fig. 9.8. In addition, a quantitative determination of several physic-
ochemical properties of agricultural interest, such as organic carbon content and
selected cations (like K+ or Mg2+) and anions (like NO3− or Cl−), was also
demonstrated, showing satisfactory agreement with the reference methods.
238 M. del Valle

Fig. 9.8 3D PCA score plot in the identification/analysis of soils, in this case just from a simple
extraction in acetic acid, and a sensor array formed by 20 ISEs, selective to specific cations,
anions, and also of generic response. The grouping regions correspond to Vallgorguina (I),
Bellmunt (II), Montesquiu (III), Taradell (IV), Bellaterra (V), and Delta de l’Ebre (VI) soils. Each
class formed by six separate subsamples. Adapted from [32], with permission from Wiley

Miniaturized electrochemical cell integrated on epoxy–glass laminate support is

an alternative reported sensor design [33, 34], essentially with the goal of inte-
gration of the sensors and miniaturization of the setup. Potassium-selective
microelectrodes based on valinomycin and two polymeric matrices [plasticized poly
(vinyl chloride) and polyurethane] were fabricated on planar Au or Ag/AgCl
transducers. Polymeric layers containing ionic liquid were cast on the surface of
Ag/AgCl microelectrodes to form the reference half cells. Performances of the
ion-sensitive and reference microelectrodes (i.e., ion selectivity, slopes of calibra-
tion curves, signal stability and repeatability, and long-term stability) were studied.
The influence of the method of membrane deposition on sensor characteristics was
not significant; however, the microelectrodes with polyurethane membranes
exhibited better durability. Theoretical calibration curves were obtained when K+
ISEs and reference electrodes were integrated on a single substrate. The designed
electrochemical cell was subsequently applied in further studies, in this case
devoted to the development of miniaturized electrode arrays for multicomponent
analysis or electronic tongue purpose (Fig. 9.9); as the noticeable case [35], for the
classification of milk originating from various producers.
In a similar attempt, in this case by our laboratory in Barcelona, we attempted
the integration of the potentiometric sensor array in a polymeric substrate and with
the use of the screen-printing technology [36]. The conductive and insulating
materials needed for the development of the devices are available from specific
suppliers centered on the electronic industry, and the potentiometric membranes are
essentially the same as the ones described up to this point, only that they are
deposited onto solid-contact microwells, previously defined by screen-printing.
9 Materials for Electronic Tongues: Smart Sensor Combining … 239

Fig. 9.9 Integrated array of solid-state microelectrodes. Array of integrated ion-selective

microelectrodes (a), schematic top view (b), cross section of a single sensor transducer through
the A–A0 line (c), and cross section of the coated wire-type microelectrode (d). 1 Au or Ag/AgCl
electrode, 2 Au or Ag contact pad, 3 sealing epoxy layer, 4 epoxy–glass substrate, and 5
ion-selective polymeric membrane. Reprinted from [33], with permission from Elsevier
240 M. del Valle

Figure 9.10 shows a scheme of the different stages in the preparation of a

disposable all-solid-state planar-type potentiometric electronic tongue, developed
with a five-sensor array screen-printed on a polymeric substrate. The fabrication
method enabled the simple and reproducible mass production of low-cost electronic
tongue systems, since the sensors showed good reproducibility (0.96 % RSD),
repeatability (0.58 % RSD), and lifetime (8 % decrease of sensitivity after 37 days
in solution). The system was used for the simultaneous determination of NH4+, K+,
and Na+ ions in synthetic and natural surface water samples. The signals were
processed by using a multilayer ANN which was optimized for that purpose.
Correct results were obtained for the three ions in synthetic samples. In natural
surface waters, ammonium and potassium ions could be determined, while sodium
ions showed some biased results due to the effect of the complex matrix.
Figure 9.11 illustrates the performance of operation, in this case displaying the
calculated versus expected concentrations of the three species considered in the
response model. Fitted comparison lines can be contrasted against the ideal
expected behavior, with highly precise results for ammonium, satisfactory for
potassium, and slightly worse for sodium. Slopes of the fitted lines were
1.02 ± 0.27 for ammonium, 1.05 ± 0.26 for potassium, and 0.90 ± 0.28 for
sodium (theoretical value, 1.00).
Different attempts have been recorded in the effort to obtain integrated electronic
tongue devices by microfabrication technologies. The group of Prof. Schöning in
Jülich, Germany, produced a microfabricated potentiometric thin-film sensor array
for the simultaneous detection of Pb2+, Cd2+, and Cu2+ [37]. The sensitive layers
used were formed on the basis of chalcogenide glass materials. These thin-film
chalcogenide glass materials consisted of mixtures of Pb–Ag–As–I–S, Cd–Ag–As–
I–S, or Cu–Ag–As–Se and were prepared by pulsed laser deposition technique,
with advantages as the short preparation time and the compatibility with silicon
planar technology microfabrication procedures. The developed sensor array, as
observed in Fig. 9.12, was physically characterized by microscopy and spectrom-
etry and next further evaluated through its potentiometric measurements. The
developed devices were employed in the simultaneous multicomponent analysis of
complex liquid media based on the principles of the electronic tongue.
Multicomponent analysis of heavy metal ion species Pb2+, Cd2+, and Zn2+ were
determined simultaneously [38] by direct potentiometric measurements using a
sensor array formed by the different chalcogenide microsensors, overcoming the
problem of the insufficient selectivity by use of single sensors.
Verrelli in the University of Rome “Tor Vergata” [39] described a miniaturized
potentiometric electronic tongue system used standard PVC membranes placed in
this case onto microfabricated silicon structures, as it is shown in Fig. 9.13.
Membranes used as ionophores different metal complexes of porphyrins and cor-
roles, therefore originating response mainly to different anions. The developed
system was applied for the qualitative and quantitative analyses of different artificial
wine samples, characterized by the presence of polluting agents such as SO2, H2S,
and acetate in a wide concentration range. The developed system was able to
9 Materials for Electronic Tongues: Smart Sensor Combining … 241

Fig. 9.10 The fabrication

process for screen-printed
five-sensor potentiometric
sensor array: a polycarbonate
substrate (50  75 mm),
b silver ink printing,
c graphite ink printing,
d insulator layer printing,
e cavities for the sensors, and
f deposition of different PVC
membranes. Reprinted from
[36], with permission from
Springer
242 M. del Valle

Fig. 9.11 Modeling performance achieved for the optimized ANN with the samples of the
external test set: a ammonium, b potassium, and c sodium. The dashed line corresponds to
ideality, and the solid one is the regression of the comparison data. Reprinted from [36] with
permission from Springer

Fig. 9.12 Thin-film sensor array after wire bonding and encapsulation of thin-film sensor array
with three different chalcogenide glass thin films. Reproduced from [38], with permission from
Springer

distinguish wines with different defects and undesired agents and also to predict the
amount of each one, also at low concentrations.
In another variant to prepare electronic tongue devices from materials and
techniques used in the electronics industry, there are the works by Gil et al., which
were conceived in thick-film technology, i.e., with the use of conductive and
resistive pastes [40]. Different conducting surfaces deposited using thick-film
technology were used as potentiometric electrodes, and the emf of each electrode in
contact with a certain aqueous solution was used as input signal for a PCA. An
array containing the electrodes RuO2 (with resistivities of 10 X/sq and 1 M X/sq),
C, Ag, Ni, Cu, Au, Pt, Al, Sn, Pb, and C (graphite) was used in a first approach as it
is shown in the photograph in Fig. 9.14. To test this “electronic tongue” design, a
family of different natural waters was studied. These include seven mineral waters,
tap water, and osmotized (reverse osmosis) water. Figure 9.15 displays the obtained
9 Materials for Electronic Tongues: Smart Sensor Combining … 243

Fig. 9.13 Sensor miniaturization can be reached by the integration of potentiometric sensors with
silicon technology which can allow the construction of microdimensioned sensors with different
geometries for deposition area. Silicon fabrication technique will allow the implementation of
different structural layer and metal thin films, to investigate the influence of different materials.
Reprinted from [39], with permission from Elsevier

Fig. 9.14 Electronic tongue developed in thick-film technology. Reproduced from [40], with
permission from Elsevier B.V.

PCA plot, where the power to differentiate the waters assayed is clearly demon-
strated. Additionally, a qualitative analysis of the different waters was performed
using fuzzy ARTMAP neural networks. The final system was successfully
employed for the differentiation of the above-mentioned waters with a success rate
higher than 93 %.
244 M. del Valle

Fig. 9.15 Principal

component analysis
(PCA) score plot for different
waters (different commercial
brands plus tap waters and
osmotized distilled water).
Data show from five different
trials. PC axes are calculated
to lie along the lines of
diminishing levels of variance
in the data set. Reproduced
from [40], with permission
from Elsevier B.V.

A special type of potentiometric sensors that needs to be included in this section

is that of ISFETs, or ion-selective field-effect transistors. These devices are
microfabricated transistors, in which the control region (the gate) has been devised
as a potentiometric membrane, in this case inorganic polycrystalline nature. The
presence/absence of ions in this zone can be translated, by the effect of the electric
field, to the intrinsic operation features of the transistor, thereby modulating the
transistor current, which is the generic signal monitored. Given that the interaction
of the ions in the gate area is of potentiometric nature, these sensors are taken of the
potentiometric type. Research groups with the availability of these devices have
prepared then ISFET sensor arrays [41] and have used them with the electronic
tongue principles. Additional potentiometric membranes can be deposited on top of
the initially inorganic prepared gate areas, exchanging the sensitivity from pH (the
usual chemical primary chemical sensitivity provided to these devices) to any
standard cation or anion, depending on the ionophores used. As the adherence of
standard PVC membranes to silicon substrates is poor, the polymeric matrix of
these modified devices is changed to polysiloxane or to acrylate photocured nature.
Figure 9.16 shows the “electronic tongue” device based on the multisensor
ion-selective field-effect transistor (ISFET) array, which is subsequently inserted in
a sequential injection analysis (SIA) flow system for automated operation. The
system was used for the analysis of mineral waters and their components (sodium,
potassium, and chloride). The precision of the ion determination in samples was
typical for potentiometric method with a standard deviation of about 3–5 %. The
proposed approach could be used for automatic analysis of waters.
To finish this section, it is important to mention the latest technology suggested
for the development of disposable electronic tongues, in this case using paper as the
cheapest substrate. These devices define contact elements with base of carbon in the
paper substrate, onto which standard potentiometric membranes (e.g., based on
PVC matrix) can be deposited, and the proper potentiometric reading can be
9 Materials for Electronic Tongues: Smart Sensor Combining … 245

Fig. 9.16 ISFET sensor array chip with different ion-selective membranes. Reproduced from
[41], with permission from Elsevier B.V.

Fig. 9.17 Paper-based ion-selective electrodes, anticipating what shortly will be a paper
electronic tongue. Reproduced from [42], with permission from Wiley

developed [42]. The main goal of this variant is the low cost of the disposable
paper-based device. This case study is comprised of 4 paper-based potentiometric
sensors, sensitive to chloride, alkaline ions (Na+/K+), alkaline earth ions (Ca2+/
Mg2+), plus NO3−, with a similar design to that shown in Fig. 9.17, displaying one
of the potentiometric channels. The developed electronic tongue was able to dis-
tinguish tap and lake waters from mineral water samples using chemometric
treatments PCA and KNN (k-nearest neighbor) pattern recognition.
246 M. del Valle

9.3 Voltammetric Electronic Tongues

Voltammetric electronic tongues are the second variant of importance in this novel
concept in sensory analysis, in which computer data processing is required to
improve the overall performance of the sensing approach. In this section, definitions
and a general review of the different subvariants reported up to this moment will be
provided; these will include the nature of sensors used, methods more widely used
for the data processing, and the more recurrent application fields.

9.3.1 Using a Single Voltammetric Sensor

From a conceptual point of view, a voltammetric system with a single electrode

may be also considered an electronic tongue, provided that the voltammogram, the
multicomponent departure information as the key point, is also characterized by a
high dimensionality of the signal supplied to the processing tool. Depending on the
particular electrochemical technique, the information may be slightly different,
because we have, for example, simple linear sweep, cyclic voltammetry, differential
pulse voltammetry, chronoamperometry, etc.
Some of the worked examples have been resolving mixtures of compounds with
electrochemical activity from a voltammogram with superimposed signals, for
example, oxidizable amino acids, phenols, or heavy metals. The conventional
identification/classification application of different varieties or classes, especially of
foods or beverages, has been also reported.
As representative application of these simple voltammetric electronic tongues,
we may cite the resolution of complex signals, superimposed or overlapping type,
where it is possible to identify which compounds are present and even its con-
centration. This software approach is typically described for signals from a single
voltammetric sensor, where the vector of currents is processed to correlate the
information sought, as illustrated in Fig. 9.18. This case has been studied for sit-
uations mixtures of heavy metals [43], oxidizable amino acid [19], or phenols [44,
45].

9.3.2 Using Voltammetric Sensor Arrays

The first voltammetric electronic tongue described as such was that reported by the
Winquist laboratory in Linköping (Sweden) [46]. This used just two metal elec-
trodes, platinum and gold, and the voltammetric technique applied was normal
pulse voltammetry. With it, it was possible to discriminate different drinks such as
fruit juices, milk, and buffers (pH), in an application of purely qualitative identi-
fication. After this assembly, an improved style multielectrode, consisting of a
9 Materials for Electronic Tongues: Smart Sensor Combining … 247

Fig. 9.18 Electronic tongue

developed in thick-film
technology. Reproduced from
[45], with permission from
Elsevier B.V.

number of different metal electrodes (or conductors) on one body fixed by an

insulating resin, was designed, see Fig. 9.19. Some element materials considered
were gold, platinum, palladium, iridium, rhodium, vitreous carbon, etc. [47]; the
recorded redox signal is dependent on the metallic electrode used. With a setup like
this, it is possible to increase the information available, provided that this is not
correlated with any of the other channels, because in that case it would be
superfluous.
A second way to achieve a similar effect can be to use a large number of
identical electrodes, such as platinum, and biasing each at a different voltage; thus,
it is possible to obtain information equivalent to a potential sweep without the need
to perform it, thus eliminating downtime. With this simplification of the voltam-
metric electronic tongue, the array can be incorporated in a flow system, and the
equivalent voltammogram can be acquired with the help of a multichannel mea-
suring device, eliminating therefore the potential scan and facilitating the use.
Precisely with these principles, a system to evaluate astringency in tea was
described, which used six platinum electrodes inserted in a flow injection
(FIA) system [48].
A very interesting application of a voltammetric electronic tongue was made in
the environmental field, where it could correctly predict the nature, burden organic
(expressed as chemical oxygen demand) conductivity, and pH of the wastewater
from a paper mill [49]. A metallic multielectrode array—formed by platinum, gold,
and rhodium electrodes—was employed as the detection system, while the mea-
surements were based on large amplitude pulse voltammetry (LAPV). LAPV
consisted of scans of pulses from 0 to 1.8 V at 0.2 V steps. Five current values were
recorded for each pulse, so a set of 300 current values (three electrodes  20
248 M. del Valle

Fig. 9.19 Configuration of

the voltammetric electronic
tongue, consisting of five
metal working electrodes, a
reference electrode, and an
auxiliary electrode of stainless
steel. This setup realizes the
sequential measuring through
the use of a relay box to
multiplex the used working
electrode. Reproduced from
[47], with permission from
Springer

pulses  five values) was recorded for each sample. Samples were first discrimi-
nated using PCA, while ANNs were used for the characterization and prediction of
chemical parameters. One potential problem of such devices, which is fouling or
degradation of electrode response by deposition of insoluble oxidized residues, or
even growth of biofilm, has been corrected recently by the use of automated
periodic polishing of the sensing surfaces, always exposing a more reproducible,
clean surface for each analysis [50].
In a similar approach to this typical arrangement in the Swedish laboratory, we
devised in Barcelona an array comprising of three conductive elements, platinum,
gold, and graphite–epoxy (the multielectrode device can be seen in the photograph
in Fig. 9.20), which was assayed for the multidetermination of three oxidizable
species in the pharmaceutical field: ascorbic acid, uric acid, and acetaminophen
[51]. Signals, merged into a single vector, were processed directly by ANNs, as
outlined in Fig. 9.21. Figure 9.22 shows the final performance of the developed
system, where the predicted concentration values can be seen versus the expected
ones, both for the training subset and for the external test subset, the one that does
not intervene at all in the building of the response model.
More recently, research in the field of electronic tongues has focused not only on
the development of new measurement techniques and the application of new pro-
cessing methods, but also on the finding of new chemosensitive materials in order
to develop greater variability response sensor arrays.
Aside from using pure metallic elements, various alternative materials have been
used to modify the surfaces of the electrode, especially when this departs from
9 Materials for Electronic Tongues: Smart Sensor Combining … 249

Fig. 9.20 Photograph of the working 4-electrode array. Au, Pt, and epoxy–graphite disks act as
working electrodes. Setup is derived from the enclosure of the four preconnected electrode
elements, inserted as 1-mm-diameter wires in epoxy resin. The Ag electrode is previously oxidized
into an Ag/AgCl disk acting as pseudoreference electrode, once in contact with a predefined
chloride concentration (added to each measured sample)

Fig. 9.21 Fusion of the signals from an array of three voltammetric sensors to develop a
multicomponent quantification application using an ANN model. The components determined are
ascorbic acid, uric acid, and paracetamol

carbon materials; for example, conductive polymers or certain metalloporphyrins

that may act as catalysts or also specific metal phthalocyanines that may be
incorporated with the carbon conductive part. Basic technology in our laboratories
departs from a traditional graphite–epoxy configuration, useful to prepare base
electrode platforms [52]. This basic design was later modified to many variants in
chemical sensing and biosensing, by incorporating specific catalysts and modifiers.
In perspective, this variant has meant a quick and convenient way to generate an
array formed by a number of sensors with differentiated response, as an ideal
250 M. del Valle

Fig. 9.22 Prediction of the developed voltammetric electronic tongue, for the determination of the
three oxidizable compounds, a ascorbic acid, b paracetamol, and c uric acid. Filled circles
correspond to the training subset, and white circles correspond to the external test subset of
samples

departure point to develop electronic tongue applications. In essence, the chemistry

of these carbon electrodes, bulk-modified with additional electroactive elements,
follows the know-how of the previously developed carbon–paste configuration, in
which fine graphite was simply agglutinated in a confined space with the use of a
mineral oil (or wax), to obtain an easily reconfigurable, repolishable, and regen-
erable electrode material, that has been in use for decades [53]. Figure 9.23 shows
the preparation of one of these sensing devices. From a plastic cylinder, in which an
electrical connector and a copper disk are placed to assure the electrical contact, a
portion of the polymer mixture (resin + hardener) plus conductive material (gra-
phite or other carbon allotropes) and modifiers are placed and then cured at proper
temperature to obtain a conducting solid, with the catalytic modifiers embedded in
its body. A proper polishing, with emery paper of increasing grade, permits to end
9 Materials for Electronic Tongues: Smart Sensor Combining … 251

Fig. 9.23 Graphite–epoxy

composite electrode
construction scheme.
a Copper disk soldered to the
connector and assembled into
the PVC tube. b Preparation
of the graphite–epoxy mixture
and incorporation of the
modifier. c Final look after
hardening and polishing.
Electronic tongue developed
in thick-film technology.
Reproduced from [54], with
permission from Wiley

in a mirror finish conductive surface, where many modifiers can be tested as

potential elements for the voltammetric electronic tongues.
Table 9.2 summarizes some of the modifiers that have been historically used in
our laboratories and the formulation details for the constructed electrodes. Apart,
currently in the nanotechnology era, specific forms of carbon like nanotubes or
graphene might be employed, also in search of improved or modified response. An
application study of such an electronic tongue system is the work in [54], where a
voltammetric sensor array formed by five modified graphite–epoxy electrodes was
applied in the qualitative and quantitative analyses of cava wines. The electrodes
used were those modified with copper nanoparticles, with platinum nanoparticles,
or with conducting polymer in powder forms like polyaniline and polypyrrole, plus
a control, non-modified graphite–epoxy electrode. The different samples were
analyzed using cyclic voltammetry without any sample pretreatment. Recorded data
were evaluated by PCA and discrete wavelet transform in order to compress and
extract significant features from the voltammetric signals. Preliminary visualization
from the PCA treatment permitted to observe the clustering of the different cava
types, according to the added sugar elaboration (see Fig. 9.24): brut nature, brut,
demi-dry, etc. Two outliers, a French champagne wine and a cava wine from a

Table 9.2 Some formulations used to obtain a number of voltammetric sensors with differentiated
response, in order to develop electronic tongue applications
Resin (%) Hardener (%) Graphite (%) Modifier (%)
81.7 1.23 17 –
81.7 1.23 15 Cu nanoparticles (2 %)
81.7 1.23 15 Pt nanoparticles (2 %)
81.7 1.23 15 Polypyrrole (2 %)
81.7 1.23 15 Polyaniline (2 %)
81.7 1.23 15 Cobalt phthalocyanine (2 %)
252 M. del Valle

Fig. 9.24 Principal component analysis plot of the first three components. A total of 21 samples
were analyzed. As can be observed, clear discrimination is obtained for the different types of cava
wines: (1) brut nature, (2) brut, (3) medium dry from Penedés region, (4) medium dry from
Extremadura region, and (5) French champagne wine (the outliers). Reproduced from [54], with
permission from Wiley

different Spanish region, were incorporated in the study and were separated (i.e.,
were not confounded) with the trained specimens. To accomplish the pattern
recognition, the preprocessed information was evaluated by an ANN that did the
qualitative classification. Moreover, a preliminary study related to the quantification
of sugar amount present was assessed by second-order standard addition method;
for this, the score from the first PC and each sample was correlated with sugar
content, analyzed by conventional method. Figure 9.25 illustrates the experimental
setup adopted in this study, accounting for the 6-channel multipotentiostat con-
trolled by computer and a voltammetric cell formed by the epoxy–graphite working
electrodes, plus a platinum auxiliary electrode and a double-junction Ag/AgCl
reference electrode.
Other modifiers that have been incorporated with success in the bulk of carbon
composite electrodes with the goal of altering their sensitivity to target compounds
are, among others: molecularly imprinted enrichment agents, either in elec-
tropolymerized form, or incorporated as microbeads, cyclodextrins as complexing
agents for specific organics, complexing ligands (essentially for metals), and
semiconducting or metallic oxides incorporated as nanopowder. Surely, these
modifiers may be postulated to be used in voltammetric electronic tongues, and
soon, we will notice reporting electronic tongue systems using these variants,
especially for specific applications with specific target chemicals involved.
In a final reflection mode, it is noteworthy that the right choice of sensitive
materials, i.e., the selection of sensors that will form the array, is of fundamental
importance in the results. Without taking into account the chemical nature of the
samples or analytes sought, and if the sensors used do not complement them, it is
9 Materials for Electronic Tongues: Smart Sensor Combining … 253

Fig. 9.25 Photograph of the six-sensor composite epoxy–graphite electrode array, together with
the multichannel potentiostat used to perform the voltammetric electronic tongue experimentation

clear that not in any case, the electronic tongue biomimetic approach will achieve
the transmutation of junk in precious gems—only a sensible and correct choice of
sensors and measurement techniques will be the determining factor in the success of
any application of this nature. Much of the voltammetric electronic tongue key
work is to use an array of sensors and to check whether their voltammetric signals
correlate with the information needed, for the type of samples under study. For
example, Winquist’s group used a sensor array formed by noble metal electrodes
[46, 55] for a variety of applications. Some of these were to identify and classify
water and beverages (fruit juices, teas, etc.), to identify industrial washing solu-
tions [56], correlating aging of dairy products and juices [57], identifying the
growth of microorganisms or assisting in the control of wastewater treatment plants.
Also, there are more elaborate applications, like that describing the identification of
microbiologic samples [58], i.e., in differentiating between various types of
microorganisms, particularly the identification of yeasts, fungi, and bacteria, or the
recognition of two fungal species growing in three nutrient media (glucose, sugar,
and malt extract). Detection by the sensor array of electroactive metabolites in the
stage of growth of these microorganisms is the basis of the developed application,
254 M. del Valle

allowing differentiating between species, type of the culture medium, or the incu-
bation period.
The examples above correspond mainly to the use of metallic electrodes or
conductors. There have also been described electronic tongues with the use of
modified sensors, i.e., with the electroactive surface coated with selective films.
Thus, the information generated is much more rich and varied, for example, by
interference effect minimization or catalysis, manifesting in increased signal, or
shifts its characteristic potential. This significantly improves selectivity and
detection power of the substances that provide weaker signals. In this mode, the
work group of Mari Luz Rodriguez Mendez in Valladolid [59] used a sensor array
consisting of carbon paste electrodes modified with phthalocyanines, polypyrrole
doped with different agents, or perylene derivatives. Many of his works have been
applied to the world of wine, for example, to detect adulterants, or determine certain
organoleptic characteristics of a wine, such as alcohol, total acidity, astringency
(tannic acid), preservatives (sulfites), sugar, and flavorings (ethanal) [60, 61]. In
Fig. 9.26, the preparation of an integrated sensor array card, with thick-film tech-
nology using an alumina substrate, can be visualized.
Analogous to the case with potentiometric sensors, also voltammetric devices
have been microfabricated in silicon substrates, with the goal of obtaining a
voltammetric electronic tongue device [62]. Direct translation of principles men-
tioned up to now is to produce a number of thin-film metal electrodes, with upper
finishing from different materials, such as gold, platinum, and palladium, in order to
get differentiated response. Figure 9.27 displays one of such devices, equipped with
four different metal working electrodes (on the corners, made of gold, platinum,
rhodium, and iridium), plus a gold central electrode with larger area.

Fig. 9.26 Lithographed gold electrodes onto an alumina substrate. Polymeric sensors have been
obtained by electrodeposition of pyrrole in the presence of different doping agents. Phthalocyanine
and perylene-based electrodes have been prepared using the carbon paste. Reprinted from [59],
with permission from Elsevier
9 Materials for Electronic Tongues: Smart Sensor Combining … 255

Fig. 9.27 Microfabricated miniaturized planar voltammetric sensor array for use in an electronic
tongue. Reprinted from [62], with permission from Elsevier

A new approach to fabricate a disposable voltammetric electronic tongue is

reported [63]. The fabrication of the disposable sensor is aimed to integrate all
electrodes necessary for measurement in the same device. The disposable device
was constructed with gold CD-R and copper sheet substrates, and the sensing
elements were gold, copper, and a gold surface modified with a layer of Prussian
blue. Figure 9.28 depicts the procedure, departing from a gold-finished CD-R
surface. Modification of the gold working electrode surface with a Prussian blue
(PB) film was accomplished by electrodepositing the insoluble layer by cyclic
voltammetry. Copper working electrodes were obtained from a similar procedure
but departing from printed circuit board and etching it. Reproducibility of the
disposable devices showed a relative standard deviation for signals obtained from
20 different disposable gold and 10 different disposable copper electrodes below
3.5 %. The performance, electrode materials, and the capability of the device to
differentiate samples were evaluated for taste substances model, milk with different
pasteurization processes (homogenized/pasteurized, ultrahigh temperature
(UHT) pasteurized, and UHT pasteurized with low-fat content), and that adulterated
with hydrogen peroxide. In all analyzed cases, a good separation between different
samples was noticed in the score plots obtained from the PCA. The performance,
low cost, reproducibility, and simplicity of the fabricated electronic tongue devices
demonstrated the advantages of modifying electrode surfaces with specific films
toward a more selective recognition strategy.
Finally, we should mention the efforts being made to improve the coupling of the
complex signals of the voltammetric electronic tongues with processing tools. In
these cases, extracting significant information is sought in order to enable the
efficient use of computational methods. Examples are the use of PCA for com-
pacting the starting information [13], the signal modeling using a polynomial fit or
using wavelet transform [43]. In a classic study [19], the principles of voltammetric
electronic tongue to resolve a mixture of three components from direct voltam-
metric signal are applied. The overlapping signal is obtained from the differential
256 M. del Valle

Fig. 9.28 Obtention of

disposable voltammetric
electronic tongues by using
Prussian blue films
electrodeposited onto CD-R
gold surfaces. Adapted from
[63], with permission from
Elsevier

pulse voltammetry response to mixtures of the three oxidizable amino acids tryp-
tophan, cysteine, and tyrosine. The signal was first compressed by wavelet trans-
form, removing both of meaningful information. Extensive comparison studies on
the choice of the compression tool were later performed by Cetó et al. [64], also in a
contribution from our laboratories.
9 Materials for Electronic Tongues: Smart Sensor Combining … 257

This operation was necessary to optimize, for that mother function type and level
of compression and its effect on the degree of reconstruction of the original signal
are studied. Then, the compressed information was used as input for a calibration
model based on ANNs, as shown as a successful combination of the chemometric
techniques with the use of complex voltammetric signals.

9.3.3 Use of Biosensors

A progress field just beginning to be investigated is the use of arrays of biosensors

for generating starting information of the electronic tongue. Here, the requirements
are known, and one needs to generate cross-selective biosensor information used for
the different species present in a sample; therefore, not very high selectivity of
biosensors is chosen. The topic has received the name bioelectronic tongue by
different researchers in the field [6, 65].
A worked example in our laboratory was to an electronic tongue formed by three
enzymatic amperometric biosensors using the enzyme glucose oxidase (GOx) with
three different formulations [66], and the target was the simultaneous determination
of glucose in the presence of two typically interfering species, ascorbic acid and
uric acid. Figure 9.29 illustrates the three-electrode system, each of them incor-
porating enzyme plus different catalysts, as Pt microparticles, or mixture of Pt/Au
microparticles, as the way to induce a differentiated response. The different
catalysts/modifiers will introduce the differentiated response of each electrode; in
this case, all of them will present signal for the primary substrate, only with dif-
ferent shapes or appearing at different potentials, depending on the catalyst or
modification of the electroanalytical signal. The device was tested in the joint
determination of glucose and its common interferents in biologic fluids, namely
ascorbic acid and uric acid, with the usage integrated in an automated SIA system
for proper automation of operation. This idea with GOx was retaken recently by
Al-Issa et al. [67]; in their work, they prepared a bioelectronic tongue immobilizing
the enzyme over Au or Pt disk electrodes. The determination of glucose in the
presence of their common interferents ascorbic acid and uric acid could be

Fig. 9.29 Photograph of the

three-biosensor integrated
array. In this case,
modifiers/catalysts are the
agents causing the
differentiation of the response.
Reproduced from [66], with
permission from Wiley
258 M. del Valle

successfully determined at the 1 mM level, while interfering species were quanti-

fied at no extra effort.
A second example worth mentioning is the inhibition electronic tongue, in which
three amperometric biosensors were used with three varieties of the acetyl-
cholinesterase enzyme, wild electric eel, and genetically engineered B1 and B394
[68]. As the degree of inhibition of dichlorvos and carbofuran insecticides is
slightly different for each enzyme, this cross-information allowed the resolution of
mixtures containing both substances, at a level below 1 nmol L−1. Table 9.3
summarizes how the two considered pesticides are resolved by the three enzymes
considered; given the response profile (inhibition constant) is different for each
pesticide and each enzyme variant, this demonstrates the correct departure point for
attempting the inhibition electronic tongue study. One point to comment is that
given the nature of the inhibition is irreversible, the application had to be developed
with disposable screen-printed electrodes, where each inhibition assay used a new,
unused electrode.
In the more recent work in our laboratories dealing with bioelectronic tongues,
that is, in voltammetric electronic tongues equipped with enzyme-modified elec-
trodes, we have been using phenol-degrading enzymes (tyrosinase and laccase) and
applying their differentiated response to the wine field [69]. In these works, the
proposed bioelectronic tongue has been used for the estimation of global
polyphenol content in wine and alternatively to differentiate and to quantify indi-
vidual (and majoritarian) polyphenolic compounds present in wine [70]; later,
equivalent work was repeated, in this case with the more specific polyphenols
present in beer [71]. In the more advanced cases, the proposed bioelectronic tongue
was formed by an array of four voltammetric enzymatic biosensors based on
epoxy–graphite composites, one blank electrode and the other three bulk-modified
with tyrosinase and laccase on one side and copper nanoparticles on the other; these
modifiers were used in order to incorporate differentiated or catalytic response to
different polyphenols present in wine and aimed to the determination of its total
polyphenol content value. The obtained voltammetric responses were preprocessed
employing the fast Fourier transform (FFT); this was used to compress the relevant
information, whereas the obtained coefficients fed an ANN model that accom-
plished the quantification of total polyphenol content. This strategy is shown in
Fig. 9.30. For comparison purposes, the obtained polyphenol content was com-
pared against the ones assessed by two different reference methods: Folin–Ciocalteu

Table 9.3 Degree of inhibition shown by the three acetylcholinesterases employed in the
bioelectronic tongue for the determination of pesticides
Ki (µM−1 min−1)/acetylcholinesterase subtype
Wild (electric eel) B1 B394
Dichlorvos 0.026 1.9 224
Carbofuran 4 6 1.08
Adapted from the data shown in [68], with permission from Wiley
9 Materials for Electronic Tongues: Smart Sensor Combining … 259

Fig. 9.30 Strategy used with the high-dimensionality voltammetric electronic tongue. Reprinted
from [69], with permission from Elsevier

Fig. 9.31 Results in the prediction of the global index indicating the total polyphenolic
compounds, Folin–Ciocalteu index, from the voltammetric bioelectronic tongue; the figure
distinguishes results from the training subset of data (left) and from the external validation subset
(right). Reprinted from [69], with permission from Elsevier

and UV polyphenol index (I280); good prediction ability was attained with cor-
relation coefficients higher than 0.949 when compared against the reference
methods; for example, Fig. 9.31 displays the satisfactory agreement obtained versus
the Folin–Ciocalteu reference method.
The modified electronic tongue has also been used for the determination of the
vintage year or origin/variety of grapes used in winemaking. Some of these works
have also been extended to the field of vegetable oils, to estimate their quality and
polyphenol content. The possibility of using enzymatic amperometric biosensors
has been previously discussed [66, 68]; again, using a greater amount of
260 M. del Valle

Fig. 9.32 Photograph of the

screen-printed eight-sensor
composite epoxy–graphite
electrode array

information, or of better quality, as provided use of an array of biosensors, can

improve the performance of the classical concepts, improving discrimination
against interference or resolution of mixtures that otherwise would provide only a
global content.
In the purpose of integration, the use of commercial multielectrode platforms has
been described that once modified might be usable for (bio)electronic tongue
applications; the firm DropSens supplies, for example, the multielectrode device
shown in Fig. 9.32. On the other hand, as the disposable paper-based potentio-
metric electronic tongue has been already described [42], the voltammetric elec-
tronic tongue, also using the paper substrate, is surely very close to appear and
offers an interesting alternative to low-cost, disposable devices.

9.4 Concluding Remarks

This chapter has provided a review of technological aspects of (bio)electronic

tongue development, especially from the point of view of more chemical nature,
i.e., from the involved materials in their construction. Potentiometric electronic
tongues were the first to be proposed and quickly progressed from existing tech-
nology related to conventional potentiometric sensors, i.e., ISEs. Later, these could
be made more complex, when specific technologies were called for, as in the
development of ISFET-based electronic tongues.
Closely in time, voltammetric electronic tongues also started and developed, and
here the possibilities offered to researchers to prepare different electrodes, and to
modify them, either in bulk or in surface variants has enlarged considerably the
type of systems described up to now. Although scientific communications with the
involvement of voltammetric electronic tongues do not exceed in number those
using potentiometric sensors, they have advantages in which sensors of very dif-
ferent nature can be used, to which the variants of chemical and biologic modifi-
cation must be added and also the possibility of using various modes of
9 Materials for Electronic Tongues: Smart Sensor Combining … 261

electrochemical technique particularly used (i.e., cyclic voltammetry, linear sweep,

differential pulse, and square wave). Therefore, it is obvious they represent a very
interesting field of work in order to develop tools for control in industrial, food,
clinical, or environmental fields.
Voltammetric electronic tongues have shown to be much more complex in use,
given the high-dimensionality signals involved, or given their complexity, for
example with overlapping shapes or interaction or cross-response. These signals
are, in turn, chemically rich in useful information and therefore represent very
interesting alternatives for classification tasks, identification, or discrimination
component. It becomes obvious only with powerful chemometric treatment that
successful applications can be developed. Even, for the more complex situations,
there are no clearly established processing protocols able to develop application
systems with bilinear or trilinear data; for that reason, this area is identified as one
of the crossroads of time where the interdisciplinary collaboration of mathemati-
cians, engineers, physicists, and/or chemists is needed.
About what can be the limits of electronic tongues of one type, or the other, very
little comparison studies have been made. One of the few appeared very recently in
which different electronic tongue technologies were compared when applied to the
pharmaceutical field [72].
Apart from the variants sketched in this chapter, it has to be mentioned that it is
also possible to mix the ideas, and hybrid electronic tongues already exist in which
the sensor array used sensors from the potentiometric and from the voltammetric
type. In this case, special efforts have to be placed in making compatible the
different dimensionality of the information provided by the two sensor types.
Feature extraction from data and aspects of data fusion are then necessary to cope
with the demanding data processing aspects in the case [73].

Acknowledgments Financial support from the Spanish Ministry of Economy and Innovation,
MINECO (Madrid), through project CTQ2013-41577-P is gratefully acknowledged. M. del Valle
thanks the support from program ICREA Academia. Many thanks are also debt to PhD students
that completed their formation in our laboratories in the research line of (bio)electronic tongues:
Jordi Gallardo, Albert Gutés, Montserrat Cortina, Manuel Gutiérrez Capitán, Juan Manuel
Gutiérrez, Xavier Cetó, Andrea Cipri, Laura Moreno-Barón, and Andreu González-Calabuig.

References

1. Pravdová V, Pravda M, Guilbault GG (2002) Role of chemometrics for electrochemical

sensors. Anal Lett 35:2389–2399
2. Vlasov Y, Legin A, Rudnitskaya A, Di Natale C, D’Amico A (2005) Nonspecific sensor
arrays (“electronic tongue”) for chemical analysis of liquids (IUPAC Technical Report). Pure
Appl Chem 77(11):1965–1983
3. Ciosek P, Wroblewski W (2007) Sensor arrays for liquid sensing—electronic tongue systems.
Analyst 132(10):963–978
4. Escuder-Gilabert L, Peris M (2010) Review: highlights in recent applications of electronic
tongues in food analysis. Anal Chim Acta 665(1):15–25
262 M. del Valle

5. Riul A Jr, Dantas CAR, Miyazaki CM, Oliveira ON Jr (2010) Recent advances in electronic
tongues. Analyst 135(10):2481–2495
6. Zeravik J, Hlavacek A, Lacina K, Skladal P (2009) State of the art in the field of electronic
and bioelectronic tongues—towards the analysis of wines. Electroanalysis 21(23):2509–2520
7. del Valle M (2010) Electronic tongues employing electrochemical sensors. Electroanalysis 22
(14):1539–1555
8. del Valle M (2011) Bioinspired sensor systems. Sensors 11(11):10180–10186
9. Esteban M, Ariño C, Díaz-Cruz JM (2006) Chemometrics for the analysis of voltammetric
data. TrAC, Trends Anal Chem 25(1):86–92
10. Ni Y, Kokot S (2008) Does chemometrics enhance the performance of electroanalysis? Anal
Chim Acta 626(2):130–146
11. Seiter JC, DeGrandpre MD (2001) Redundant chemical sensors for calibration-impossible
applications. Talanta 54(1):99–106
12. Despagne F, Massart DL (1998) Neural networks in multivariate calibration. Analyst 123
(11):157R–178R
13. de Carvalho RM, Mello C, Kubota LT (2000) Simultaneous determination of phenol isomers
in binary mixtures by differential pulse voltammetry using carbon fibre electrode and neural
network with pruning as a multivariate calibration tool. Anal Chim Acta 420(1):109–121
14. Ciosek P, Wróblewski W (2006) The analysis of sensor array data with various pattern
recognition techniques. Sens Actuators, B: Chem 114(1):85–93
15. Simons J, Bos M, Van der Linden WE (1995) Data processing for amperometric signals.
Analyst 120:1009–1012
16. Cortina M, Duran A, Alegret S, del Valle M (2006) A sequential injection electronic tongue
employing the transient response from potentiometric sensors for anion multidetermination.
Anal Bioanal Chem 385(7):1186–1194
17. Gemperline PJ (1997) Rugged spectroscopic calibration for process control. Chemometr Intell
Lab Syst 39:29–40
18. Moreno-Barón L, Cartas R, Merkoçi A, Alegret S, Gutiérrez JM, Leija L, Hernandez PR,
Muñoz R, del Valle M (2005) Data compression for a voltammetric electronic tongue
modelled with artificial neural networks. Anal Lett 38(13):2189–2206
19. Moreno-Barón L, Cartas R, Merkoçi A, Alegret S, del Valle M, Leija L, Hernandez PR,
Muñoz R (2006) Application of the wavelet transform coupled with artificial neural networks
for quantification purposes in a voltammetric electronic tongue. Sens Actuators, B: Chem 113
(1):487–499
20. Artursson T, Holmberg M (2002) Wavelet transform of electronic tongue data. Sens
Actuators, B: Chem 87(2):379–391
21. Bos M, Bos A, van der Linden WE (1990) Processing of signals from an ion-elective
electrode array by a neural network. Anal Chim Acta 233(C):31–39
22. Baret M, Massart DL, Fabry P, Menardo C, Conesa F (1999) Halide ion-selective electrode
array calibration. Talanta 50(3):541–558
23. Colilla M, Fernández CJ, Ruiz-Hitzky E (2002) Case-based reasoning (CBR) for multicom-
ponent analysis using sensor arrays: application to water quality evaluation. Analyst
127:1580–1582
24. Legin A, Rudnitskaya A, Vlasov YG, Di Natale C, Davide F, D’Amico A (1997) Tasting of
beverages using an electronic tongue. Sens Actuators, B: Chem 44(1–3):291–296
25. Legin A, Rudnitskaya A, Vlasov Y, Di Natale C, Mazzone E, D’Amico A (1999) Application
of electronic tongue for quantitative analysis of mineral water and wine. Electroanalysis 11
(10–11):814–820
26. Rudnitskaya A, Legin A, Seleznev B, Kirsanov D, Vlasov Y (2008) Detection of ultra-low
activities of heavy metal ions by an array of potentiometric chemical sensors. Microchim Acta
163(1–2):71–80
27. Ciosek P, Augustyniak E, Wróblewski W (2004) Polymeric membrane ion-selective and
cross-sensitive electrode-based electronic tongue for qualitative analysis of beverages.
Analyst 129(7):639–644
9 Materials for Electronic Tongues: Smart Sensor Combining … 263

28. Ciosek P, Brzózka Z, Wróblewski W (2004) Classification of beverages using a reduced

sensor array. Sens Actuators B: Chem 103(1–2):76–83
29. Gallardo J, Alegret S, De Román MA, Muñoz R, Hernández PR, Leija L, del Valle M (2003)
Determination of ammonium ion employing an electronic tongue based on potentiometric
sensors. Anal Lett 36(14):2893–2908
30. Gutiérrez M, Alegret S, Cáceres R, Casadesús J, Marfà O, del Valle M (2007) Application of
a potentiometric electronic tongue to fertigation strategy in greenhouse cultivation. Comput
Electron Agric 57(1):12–22
31. Gutiérrez M, Gutiérrez JM, Alegret S, Leija L, Hernández PR, Favari L, Muñoz R, Valle MD
(2008) Remote environmental monitoring employing a potentiometric electronic tongue. Int J
Environ Anal Chem 88(2):103–117
32. Mimendia A, Gutiérrez JM, Alcañiz JM, del Valle M (2014) Discrimination of soils and
assessment of soil fertility using information from an ion selective electrodes array and
artificial neural networks. CLEAN—Soil, Air Water 42(12):1808–1815
33. Ciosek P, Mamińska R, Dybko A, Wróblewski W (2007) Potentiometric electronic tongue
based on integrated array of microelectrodes. Sens Actuators B: Chem 127(1):8–14
34. Toczyłowska-Mamińska R, Ciosek P, Ciok K, Wróblewski W (2008) Development of a
miniaturised electrochemical cell integrated on epoxy-glass laminate. Microchim Acta 163(1–
2):89–95
35. Ciosek P, Wróblewski W (2008) Miniaturized electronic tongue with an integrated reference
microelectrode for the recognition of milk samples. Talanta 76(3):548–556
36. Gutiérrez M, Moo VM, Alegret S, Leija L, Hernández PR, Muñoz R, del Valle M (2008)
Electronic tongue for the determination of alkaline ions using a screen-printed potentiometric
sensor array. Microchim Acta 163(1–2):81–88
37. Kloock J, Mourzina Y, Schubert J, Schöning M (2002) A first step towards a microfabricated
thin-film sensor array on the basis of chalcogenide glass materials. Sensors 2(9):356
38. Mourzina YG, Schubert J, Zander W, Legin A, Vlasov YG, Lüth H, Schöning MJ (2001)
Development of multisensor systems based on chalcogenide thin film chemical sensors for the
simultaneous multicomponent analysis of metal ions in complex solutions. Electrochim Acta
47(1):251–258
39. Verrelli G, Francioso L, Paolesse R, Siciliano P, Di Natale C, D’Amico A, Logrieco A (2007)
Development of silicon-based potentiometric sensors: towards a miniaturized electronic
tongue. Sens Actuators B: Chem 123(1):191–197
40. Martı́nez-Máñez R, Soto J, Garcia-Breijo E, Gil L, Ibáñez J, Llobet E (2005) An “electronic
tongue” design for the qualitative analysis of natural waters. Sens Actuators B: Chem 104
(2):302–307
41. Ipatov A, Abramova N, Bratov A, Domínguez C (2008) Integrated multisensor chip with
sequential injection technique as a base for “electronic tongue” devices. Sens Actuators B:
Chem 131(1):48–52
42. Witkowska Nery E, Guimarães JA, Kubota LT (2015) Paper-based electronic tongue.
Electroanalysis 27(10):2357–2362
43. Cocchi M, Hidalgo-Hidalgo-De-Cisneros JL, Naranjo-Rodríguez I, Palacios-Santander JM,
Seeber R, Ulrici A (2003) Multicomponent analysis of electrochemical signals in the wavelet
domain. Talanta 59(4):735–749
44. Gutés A, Ibáñez AB, Céspedes F, Alegret S, del Valle M (2005) Simultaneous determination
of phenolic compounds by means of an automated voltammetric “electronic tongue”. Anal
Bioanal Chem 382(2):471–476
45. Gutes A, Cespedes F, Alegret S, del Valle M (2005) Determination of phenolic compounds by
a polyphenol oxidase amperometric biosensor and artificial neural network analysis. Biosens
Bioelectron 20(8):1668–1673
46. Winquist F, Wide P, Lundström I (1997) An electronic tongue based on voltammetry. Anal
Chim Acta 357(1–2):21–31
47. Winquist F (2008) Voltammetric electronic tongues—basic principles and applications.
Microchim Acta 163(1–2):3–10
264 M. del Valle

48. Scampicchio M, Benedetti S, Brunetti B, Mannino S (2006) Amperometric electronic tongue

for the evaluation of the tea astringency. Electroanalysis 18(17):1643–1648
49. Gutés A, Cespedes F, del Valle M, Louthander D, Krantz-Rülcker C, Winquist F (2006) A
flow injection voltammetric electronic tongue applied to paper mill industrial waters. Sens
Actuators, B: Chem 115(1):390–395
50. Cavanillas S, Winquist F, Eriksson M (2015) A self-polishing platinum ring voltammetric
sensor and its application to complex media. Anal Chim Acta 859:29–36
51. Gutés A, Calvo D, Céspedes F, del Valle M (2007) Automatic sequential injection analysis
electronic tongue with integrated reference electrode for the determination of ascorbic acid,
uric acid and paracetamol. Microchim Acta 157(1–2):1–6
52. Alegret S, Alonso J, Bartroli J, Cespedes F, MartinezFabregas E, del Valle M (1996)
Amperometric biosensors based on bulk-modified epoxy graphite biocomposites. Sens Mater
8(3):147–153
53. Kalcher K (1990) Electroanalysis 2:419
54. Cetó X, Gutiérrez JM, Moreno-Barón L, Alegret S, del Valle M (2011) Voltammetric
electronic tongue in the analysis of cava wines. Electroanalysis 23(1):72–78
55. Holmin S, Björefors F, Eriksson M, Krantz-Rülcker C, Winquist F (2002) Investigation of
electrode materials as sensors in a voltammetric electronic tongue. Electroanalysis 14
(12):839–847
56. Ivarsson P, Johansson M, Höjer NE, Krantz-Rülcker C, Winquist F, Lundström I (2005)
Supervision of rinses in a washing machine by a voltammetric electronic tongue. Sens
Actuators, B: Chem 108(1–2 SPEC. ISS.):851–857
57. Winquist F, Krantz-Rülcker C, Wide P, Lundström I (1998) Monitoring of freshness of milk
by an electronic tongue on the basis of voltammetry. Meas Sci Technol 9(12):1937–1946
58. Söderström C, Winquist F, Krantz-Rülcker C (2003) Recognition of six microbial species
with an electronic tongue. Sens Actuators, B: Chem 89(3):248–255
59. Parra V, Arrieta ÁA, Fernández-Escudero JA, García H, Apetrei C, Rodríguez-Méndez ML,
Saja JAd (2006) E-tongue based on a hybrid array of voltammetric sensors based on
phthalocyanines, perylene derivatives and conducting polymers: discrimination capability
towards red wines elaborated with different varieties of grapes. Sens Actuators B: Chem 115
(1):54–61
60. Parra V, Arrieta AA, Fernández-Escudero JA, Rodríguez-Méndez ML, De Saja JA (2006)
Electronic tongue based on chemically modified electrodes and voltammetry for the detection
of adulterations in wines. Sens Actuators, B: Chem 118(1–2):448–453
61. Rodríguez-Méndez ML, Parra V, Apetrei C, Villanueva S, Gay M, Prieto N, Martinez J, De
Saja JA (2008) Electronic tongue based on voltammetric electrodes modified with materials
showing complementary electroactive properties. Appl Microchim Acta 163(1–2):23–31
62. Twomey K, de Eulate EA, Alderman J, Arrigan DWM (2009) Fabrication and character-
ization of a miniaturized planar voltammetric sensor array for use in an electronic tongue.
Sens Actuators, B: Chem 140(2):532–541
63. Paixão TRLC, Bertotti M (2009) Fabrication of disposable voltammetric electronic tongues
by using Prussian Blue films electrodeposited onto CD-R gold surfaces and recognition of
milk adulteration. Sens Actuators B: Chem 137(1):266–273
64. Cetó X, Céspedes F, del Valle M (2013) Comparison of Methods for the processing of
voltammetric electronic tongues data. Microchim Acta 180(5–6):319–330 (in press)
65. Gutiérrez M, Alegret S, del Valle M (2008) Bioelectronic tongue for the simultaneous
determination of urea, creatinine and alkaline ions in clinical samples. Biosens Bioelectron 23
(6):795–802
66. Gutés A, Ibáñez AB, del Valle M, Céspedes F (2006) Automated SIA e-tongue employing a
voltammetric biosensor array for the simultaneous determination of glucose and ascorbic acid.
Electroanalysis 18(1):82–88
67. Al-Issa Y, Njagi J, Schuckers SC, Suni II (2015) Amperometric bioelectronic tongue for
glucose determination. Sens Bio-Sens Res 3:31–37
9 Materials for Electronic Tongues: Smart Sensor Combining … 265

68. Cortina M, del Valle M, Marty JL (2008) Electronic tongue using an enzyme inhibition
biosensor array for the resolution of pesticide mixtures. Electroanalysis 20(1):54–60
69. Cetó X, Céspedes F, del Valle M (2012) BioElectronic tongue for the quantification of total
polyphenol content in wine. Talanta 99:544–551
70. Ceto X, Cespedes F, Isabel Pividori M, Manuel Gutierrez J, del Valle M (2012) Resolution of
phenolic antioxidant mixtures employing a voltammetric bio-electronic tongue. Analyst 137
(2):349–356
71. Ceto X, Cespedes F, del Valle M (2013) Assessment of individual polyphenol content in beer
by means of a voltammetric bioelectronic tongue. Electroanalysis 25(1):68–76
72. Pein M, Kirsanov D, Ciosek P, del Valle M, Yaroshenko I, Wesoły M, Zabadaj M,
Gonzalez-Calabuig A, Wróblewski W, Legin A (2015) Independent comparison study of six
different electronic tongues applied for pharmaceutical analysis. J Pharm Biomed Anal
114:321–329
73. Gutiérrez JM, Haddi Z, Amari A, Bouchikhi B, Mimendia A, Cetó X, del Valle M (2013)
Hybrid electronic tongue based on multisensor data fusion for discrimination of beers. Sens
Actuators, B: Chem 177:989–996
Index

A D
Amino acids, 186 Data extraction, 4
Amperometric interdigitated electrodes, 165, Definitions, 3
167, 169, 171, 172
Amperometric sensors, 257–259 E
Anion sensing, 180 Electroanalysis, 228, 230
Anodised aluminium oxide (AAO) membranes, Electrochemical biosensing, 111
92 Electrochemical paper sensors, 50, 51, 54, 58, 59
Ascorbic acid, 184 Electrochemical sensor
amperometry, 7, 51, 182
B conductimetric, 8
Bioapplications, 76, 77, 89, 90, 92, 95 electrochemical impedance spectroscopy,
Biochemical sensors, paper based, 29 64
history of, 30 potentiometric, 10, 60
motivations, 33 voltammetric/amperometric, 14
Biocompatibility, 124–128, 130–132, 134 voltammetry, 56
Biocompatible, 79, 81–83, 92 Electrochemistry, 7, 15
Biocompatible material, 125 Electroconductive hydrogels (ECHs), 80
Biocompatible membranes, 79, 81 Electrodes
Biological analytes, 181 crystalline membrane, 13
Biomimetics, 4 gas sensing, 14
Biosensing, 106, 115, 116 glass, 12
Biosensors, 4, 257, 258, 260 non-crystalline membrane, 14
Biosensors applications, 95 potentiometric enzyme, 14
Biosensors uses, 257 Electronic tongues, 227, 228, 230–234, 246,
248, 254, 256, 258, 260, 261
C Electrospinning of membranes, 88
Carbon-based nanomaterials, 105, 106 Ellipsometry, 206
Carbon nanotubes, 142, 143, 146, 147, Environmental monitoring, 165
155–158, 160
Chemical changes, 2 F
Chemical information, 7 Fabrication strategies, 31, 42, 47, 49, 56, 59,
Chemical sensor, 1–4 63, 65, 66
Chemichromic sensors, 202 Fibre-optic sensors, 197
Colorimetric, 7, 21–23, 25
Colorimetric sensor, 23 G
Conductivity dependent surface acoustic wave Graphene, 106–109, 111, 113–116, 118–120,
(SAW) sensors, 165, 169, 193–195, 217 142, 143, 146, 151–155, 160
Graphene modification, 117

© Springer International Publishing AG 2017 267

T.R.L.C. Paixão and S.M. Reddy (eds.), Materials for Chemical Sensing,
DOI 10.1007/978-3-319-47835-7
268 Index

H P
Health care, 165 lPADs, 34, 65, 66
Host response, 125, 126 Paper-based devices, 33, 56
Hydrogen Peroxide (H2O2), 184 Paper basics, 31
Permselective membranes, 79
I pH responsive membranes, 87
Impedimetric sensors, 230 Phthalocyanines, 165, 167, 169, 171, 173–175,
Implantable materials, 123, 124, 126–129, 178, 180, 182–184, 186, 189, 191–193,
131–133 195, 197, 199, 201, 202, 208, 209, 211,
Implantable sensors, 126, 128 214, 215, 217
Ion-selective electrodes (ISEs), 77 Point-of-care, 49, 50, 55
Ion-selective field-effect transistors (ISFETs), Pollutant gas
14 detection, 167
Potentiometry, 230
L
Langmuir–Blodgett (LB), 142, 159 Q
Layer-by-layer (LbL), 142, 144, 145 Quartz crystal sensors, 188

M R
Mass-sensitive quartz crystal monitors Recognition element, 2, 3
(QCM) sensors, 165, 190–192, 217
Membrane technologies, 75, 76, 79, 84, 86–90, S
92, 94 Self-assembly, 141
Metal-free phthalocyanine, 211 Sensor arrays, 231, 244, 246
Metal nanoparticles, 142, 145, 146, 150, 153, Surface acoustic wave (SAW) Sensors, 193
160 Surface plasmon resonance (SPR) Technique,
Molecularly imprinted polymer (MIPs), 84–87 208
2D, 85 Surface treatment, 94
3D, 84 Synthetic membranes, 76
Synthetic polymeric membranes, 75
N
Nano-biointerfaces, 109, 111 T
Nanomaterials, 124, 128, 129, 131 Thermoresponsive membranes, 86
Nanoporous membranes, 76 Thin films, 142, 144, 157
Nanotechnology, 124, 128, 134 Transduction, 2, 4
Nitric oxide (NO), 186
U
O Uric acid, 184
Optical paper sensors, 30 UV–Visible Absorption Sensors, 201
Optical sensor, 197, 217
chemiluminescence-based signal V
transduction, 42 Volatile/Non-volatile solvents sensing, 178
colorimetric signal transduction, 35 Voltammetry, 251, 256
electrochemiluminescence-based signal cyclic, 16
transduction, 44 hydrodynamic, 18
fluorescence-based signal transduction, 39 microelectrode, 18
Optical transduction schemes, 46 step and pulse, 19
Organic field-effect transistors (OFET) sensors,
165, 167, 172